KWAME NKRUMAH UNIVERSITY OF SCIENCE & TECHNOLOGY, KUMASI

DEPARTMENT OF BIOCHEMISTRY AND BIOTECHNOLOGY: M. Phil. BTC 561

FERMENTATION AND ENZYME TECHNOLOGY 2020/2021 (Handout 4)

SCALE-UP, UPSTREAM AND DOWNSTREAM PROCESSING

Scale Up
Scale up is the process of taking a biotechnological production from a laboratory scale to a scale
at which it is commercially useful. Fermentation processes are normally developed in three stages
or scales. In the initial stage basic screening procedures are carried out using relatively simple
microbial techniques, such as Petri dishes, Erlenmeyer flasks, etc. This is followed by a pilot plant
investigation to determine the optimal operating conditions in a volume capacity of 5 to 200 litres.
The final stage is the transfer of study to plant or production scale and final economic realisation.

Throughout these stages of development, the biotechnologist aims to maintain the optimal
environmental conditions for the process at all levels of development. These environmental
conditions involve both chemical factors (substrate concentration, etc.) and physical factors (mass
transfer ability, mixing ability, power dissipation, etc.). In particular, the physical factors create the
greatest problems when the process is moving from one scale to another. It is in this area that
the skills of the chemical or process engineer find greatest expression.

A few biotechnological processes can be run on laboratory-scale systems (for example, the
production of reagents for research use, such as monoclonal antibodies). All others have to be
done in much larger installations than a research laboratory can handle.

The difficulty with scale up is that a tonne of fermenting bacteria seldom behaves in the same way
as a gram of the same bacteria unless it is divided into a million separate tubes. In general, it is
not possible to take the conditions that have worked well in the laboratory and apply them to an
industrial process. Instead, they are gradually adapted to ever larger scales of production, each
step usually being between a four- and a 10-fold increase is size over the previous one. At each
stage, the optimum amount of various chemical parameters and mechanical parameters (such as
stirring rate, air-supply method and rate) must be determined, based on the biotechnologist’s
experience with previous production systems and a general knowledge of scale-up procedures.
There is some mathematical modeling available to help, but even so experimentation is essential.

The problems of scale up were not well understood by the early genetic engineers, and so in the
mid-1980s there was a severe shortage of scientists skilled in the field. However, it is now
understood that a marvelous laboratory result does not automatically translate into money in the
bank, as scale up might be prohibitively expensive.

Due to the larger size and scale of industrial fermentation setups it is vital to succeed beforehand
studies of possible complications, due to increase in scale, and to sort out ways to tackle them.
Scale up studies are executed at the laboratory or pilot plant scale to generate data that can be
used to anticipate and construct the large scale industrial fermentors with ample confidence that
it would function according to all its expected behaviors. Hence, scale up studies not only help
understanding the technical components of a large-scale fermentation setup but also reduce the
economic risk involved in the case of direct investment on large-scale production.
The process of fermentation is first designed at laboratory level but the actual industrial scale
production means quantitative increase of several thousand liters as compare to laboratory
fermentation. So, it is necessary to perform scale up experiments before proceeding to full scale

1
plant installation. However, this increase take place in different phases i.e. laboratory scale
experiments lead to pilot scale experiments that in turn leads to industrial production which is
called scale up [8].

Fermentation techniques are formulated using flasks and small lab scale fermentors. However,
there use to be a significant difference between designs and efficiency of small and large-scale
fermentors which calls for determination of proper incubation conditions that are needed to
employed at large scale-production tanks as based on information obtained from experiments
done with various small fermentors.

The fermentation is started from laboratory scale where pure and analytic grade chemicals are
used, conditions are controlled carefully and fermentation parameters are maintained
appropriately. However, at the industrial level managing all these expects effectively becomes
nearly impossible which calls for fine tuning of these parameters. A direct changeover from a
laboratory scale experiment to industrial production scale may give rise to results against the
expectation which means a huge financial loss. For instance, most of industries use media that is
crude and unprocessed with variable chemical composition so production flops if fermenting
organism fails to use such media and grow. Scale up means to increase the volume. This
procedure aims to increase the scale of fermentation (volume) without compromising yield or if
reduction in yield occurs then fermentation technologist should able to identify the factors that
contribute to decrease and to rectify it.

SCALE UP CONSIDERATIONS
The speculations that must be considered while developing the process of higher fermentation
productions are discussed below:

A. Inoculum development
An inoculum used for fermentation should be in its active state so that length of lag phase remains
short in the subsequent fermentations. Inoculums must have been available in sufficiently large
volume within its suitable morphological form. It must be contamination free and with product
forming capabilities retained. The process that can be adopted to produce such inoculums, that
meets forth mentioned qualities, is known as inoculum development.

Inoculum development is equally important regardless of its scale as shown by a quotation of

Hockenhulll “Once fermentation has been started it can be made worse but not better.” At
industrial level relatively, large inoculum volume is used to keep length of lag phase short so that
maximum biomass can be produced in shorter time period. So, it means that inoculum has to be
developed in a number of stages where two or three steps are performed in flasks then next two
or three in seeding fermentors before inoculum is finally added to pilot fermentor or industrial
fermentor. Number of stages or steps in between depends on size of fermentor. However, the
more the number of stages between master culture and production, fermentor will involve the
greater risk of contamination and stain degeneration will be persisting.

B. Sterilization
Sterilization in a process of fermentation means using contamination free inoculum, sterilization
of media, vessels and materials used in process and maintaining aseptic conditions during
fermentation process. However, at large scale achieving sterilization becomes difficult due to
larger size of vessels and bulk amount of media. It becomes important to select suitable technique
for process within economical limits. For example, for media sterilization there are different

2
treatments available like heat treatment, Ultrasonic treatment or chemical treatment etc. However,
besides use of steam, as it is used for sterilization of most of the media, filtration is used for
sterilization of media for animal cell cultures and other heat labile media. Filtration adds extra
expenses in the overall cost of production that must kept in mind while using filters at higher
scales.

C. Cleaning
The cleaning of fermentation vessels is important for prevention of blockage and continuous
functioning of fermentor which becomes more extensive due to increase in size of fermentor.
Mostly the distilled water is used to remove loose residual cultures and other remains. After
washing vessels are immediately dried. However, during cleaning examination of fermentors can
be done by checking presence of chips, cracks, rust or any other damage.

D. Environmental Parameters
The surrounding environment of an organism may change with the change in scale of production.
Significant gradients of different parameters such as dissolved oxygen, pH etc. can be observed
in number of industrial scale fermentation setups [9]. It means that
the cells that are fluctuating in a large volume reactor may therefore experience noticeable
changes in micro-environment especially in the case of aerobic fermentation. The parameters
which may bring such changes are summarized as follows:
✓ Nutrient availability
✓ pH
✓ Temperature
✓ Dissolved oxygen concentration
✓ Dissolved carbon dioxide concentration
✓ Shear conditions
✓ Foam production

There are two basic factors that affect all parameters mentioned. These two parameters are (in
term of bulk mixing) and aeration (in term of oxygen provision). The parameters such as Nutrient
availability, pH, Temperature and shear conditions are dependent on bulk mixing (agitation) while
parameters such as dissolved oxygen concentration, dissolved carbon dioxide concentration and
foam production depends on air flow or oxygen transfer rate (aeration). It means that agitation
and aeration tend to be dominating in discussions about scale up procedures. However, decrease
in yield due to difficulties in sterilization and inoculums development can also not be ignored.

E. Scale-Up of Aeration and Agitation Regimes in Stirred Tank

Reactors
Fox (1978) explained aeration and agitation problem during scale up in a detailed manner. He
purposed “Scale up window” to illustrate issues of aeration and agitation. The Scale up window
can be used to represent respective boundaries that are imposed by the environmental
parameters and cost of the aeration, agitation regime as shown in figure A. A range of aeration
and agitation combinations can be used to obtain suitable conditions for mixing and oxygen
transfer. The two axes shown in the figure are agitation (y-axis) and aeration (x-axis) while the
zone lying within hexagon represents acceptable aeration and agitation regimes.

3
 Figure 1: Scale Up Window

The boundaries of hexagon define limits of oxygen supply, carbon dioxide accumulation, and
damage of cells due to shear forces, foam formation, cost and bulk mixing. The rate of agitation
and aeration should lie within its maximum and minimum values. For instance, the lower limits of
aeration are determined by oxygen limitation and carbon dioxide accumulation while upper by
foam formation which may become hard and block fermentation vessels in later stages of
fermentation. Due to transport limitations in large scale fermentors the oxygen consumption rate
may exceed its transportation rate, which might result in depleted oxygen supply. Such issues are
most common in fed batch cultivations during aerobic fermentation. Some bacteria may recognise
such gradients within few seconds only. For example; E.coli recognizes oxygen depletion in the
time period as short as 13 seconds and start responding by producing undesirable by products.

Effect of verified stirrer speed, time of cultivation and type of organism on oxygen gradient was
demonstrated by Larsson et al. in 1996. The pH gradient may also lead to reduced cell viability.
For example, the pH gradients may cause the ferm entation acids produced to be driven back in
to the microbial cells. Inside cells these acids produce protons in the intracellular expanse so that
it hinders normal cellular functions hence disturb viability of microbial cells.

Mass Transfer of Oxygen (Mass Transfer, Morphology, and

Rheology)
The transfer of oxygen into a fermentation broth has been studied since 1940. The volumetric
mass transfer coefficient of oxygen (kLa) is used as a critical parameter for scaling purposes
(Reuss, 1993) and maintaining kLa constant across scales is primordial, however it is not possible
to conserve all parameters across scales, and a suitable compromise must be found.

Other important scaling parameters include the oxygen transfer rate (OTR),
the volumetric power input (P/V) and the superficial gas velocity (vg). A maximum oxygen transfer
rate must be achievable, and this rate is ultimately dependent on the mass transfer coefficient,
KLa (1/s), and the driving force for mass transfer, ∆C, since OUR= KLa × ∆C. For oxygen transfer,
the driving force ∆C is the difference between the oxygen concentration in the air bubbles and

4
that in the broth, which must always be held above the critical dO2 value throughout the fermenter
for the duration of the process.

The overall oxygen demand of the cells throughout the batch or fed‐batch fermentation must be
met by the oxygen transfer rate and the demand increases as long as the number of cells is
increasing. Accordingly, for every mole of O2 utilized, 1 mole of CO2 is produced, and the
respiratory quotient, RQ = 1 (Nienow, 2006).

Usually, the power requirement for agitation in large vessels is lower than that in small vessels,
which is principally due to the effect of the higher pressure and the increased pressure from the
liquid column. This will facilitate oxygen mass transfer. In order to achieve a high volumetric power
input in small bioreactors, the agitation speed has to be very high, which resulted in higher shear
rates in the smaller vessels. Accordingly, fermentation broths exhibited various flow behaviours
(non‐Newtonian, Newtonian, Pseudoplastic, a shear ticking, a shear thinning).

For example, the viscosity of non‐Newtonian broths is dependent on the shear rate and the
difference in shear rate across scales will lead to differences in apparent viscosity (Meerman et
al., 2004). This in turn affects kLa and explains why the mass transfer of filamentous broths is
limited in large scale. Thus, mass transfer is strongly dependent on the viscosity of the
fermentation broth (Magdouli et al., 2017). The rheology of the broth is thereby at the same time
dependent on the biomass concentration and the morphology of the cells used as
expression system (Cascaval et al., 2003). It is the complex relation between all these variables
that makes scaling up/scaling down difficult.

In order to have proper oxygen supply, stirring and aeration are a prerequisite for almost all types
of cells and even for cell cultures (Ma et al., 2006). However, agitation may also lead to changes
in morphology which in turn will affect product formation (Smith et al., 1990, Jüsten et al., 1996).
Therefore, with an aerated STR setup the question of shear sensitivity of the employed organisms
arises, and there has been an ongoing discussion around the topic.

“Shear Damage”
Bacteria and yeast on the other hand appear to be organisms that are more resistant to
mechanical stresses, and it has been suggested that their small size is the reason for their
resistance to shear stress: they are smaller than the Kolmogorov micro‐scale of turbulence
(Nienow et al., 2010). Therefore, it is believed that the morphology is not affected by mechanical
mixing. Nienow (Nienow, 2009) reported no cell damage by agitation based on flow cytometry.
Nevertheless, they reported a reduction in biomass concentration, but increase in viability when
scaling up a process; this was attributed to the poor homogeneities in industrial scale reactors
(Nienow, 2009).

“Shear damages” are frequently made to explain poor process performance when mechanical
agitation and aeration are introduced into a bioreactor as compared to the non‐agitated and non‐
sparged conditions in a shake flask or microtitre plate (Thomas, 1990). This latter has suggested
that cells might be considered to be unaffected by fluid dynamic stresses if they were of a size
smaller than the Kolmogoroff microscale of turbulence, ƛK. The microscale of turbulence is related
to the local specific energy dissipation rate ƐT.

Generally, fermentation process optimization studies are focused on filamentous fungi since
bacteria and yeast cultivation exhibit a Newtonian behavior due to the spherical shape of their
morphology, which facilitates to work with.

5
However, the most challenging expression systems are the filamentous microorganisms in terms
of mass transfer and rheology because of their morphology that impacts directly the mass
transfer. Generally, the morphology for a given bioprocess varies, cannot be generalized, and
relies on the desired product (Gibbs et al., 2000). Moreover, the phenomena of pellet formation
can occur and varied depending on the fermentation and the employed strain. Accordingly, this
latter resulted from the aggregation of spores after inoculating the media (Grimm et al., 2004).
Latter, the requirements for analysis methods to study the morphological changes is crucial, thus,
particle size analysis like laser diffraction and focused beam reflectance (FBR) can be
investigated in this aim to study the pelleted biomass (Grimm et al., 2004, Rønnest et al., 2012).

The study of pellet‐type morphology is especially important, because it contributes to the

simplification of downstream process. For instance, the Newtonian fluid behavior due to the
biomass pellet formation required low aeration and agitation power input, meanwhile the pelleted
morphology led to nutrient concentration gradients within the pellet. Other to macroscopic level,
the morphology influenced the production kinetics at the microscopic level, for example in the
case of higher enzyme secretion by A. oryzae (Spohr et al., 1998).

Consequently, the morphology of filamentous fungi had an impact on the productivity, and it is
affected by the outer appearance of the fungus, this finding was approved by Ratul et al. (2014),
who found that the morphology of Rhizopus oryzae affected the fumaric acid production (Das
and Brar, 2014). These observations agreed with Magdouli et al. (2017) who found that the
morphology of Yarrowia liyolytica is ultimately correlated with lipid production (Magdouli et al.,
2017). Usually, the shift in morphology and the pellet formation is due to the variation of
disturbance of the system/equilibrium as well as operating factors (i.e., pH changes and salt
composition) and in some case N‐starvation led to the transition from pellet to free mycelial
growth.

For non‐Newtonian fluids behaviours (e.g., fungal fermentation) the study of morphological
behaviour and at which shear rate the viscosity should be evaluated, might cause problems during
scale up. Moreover, there is no reliable shear rate that can be estimated in the fermentation tank
itself. In the case of stirred tank reactors, the shear rate varied along the fermenter, thus, it
increased at the agitator tip and decreased when approaching the vessels walls. Hence, the
simple method of shear rate calculation is to take the maximum or the average shear rate.
Meanwhile, it is not clear which shear rate is governing the mass transfer processes. The
calculation of shear rate is usually done empirically by the method of Metzner and Otto (1957)
based principally on the Reynolds numbers in the laminar and transitional regime (Metzner and
Otto, 1957). This method has been employed in both laboratory and full-scale fermenters (Stocks,
2013), although, however, it remains limited. Finally, the determination of shear rate and viscosity
across scales is particularly challenging.

Other Aspects
To obtain high product yields on a large scale, larger culture broth volumes and longer inoculation
time were needed. However, in the case of recombinant hosts, the maintenance of the stability of
the propagation of plasmids to daughter cells, is more difficult to control and less easy to be
maintained on a larger scale.

Accordingly, plasmid stability is governed by the plasmid properties (i.e., size and nucleotide
sequence) and, by the genetic background of the host (Friehs, 2004) as well as by process
parameters such as growth rates, substrate concentrations and temperature. Neubauer et al.

6
(2003) have shown that high glucose concentrations diminish plasmid stability (Neubauer et al.,
2003), meanwhile, Lin and Neubauer (2000) have mentioned that the rapid glucose oscillations
favour plasmid stability and recombinant protein production rate (Lin and Neubauer, 2000).

For this attempt, the application of runaway plasmids was investigated (Ansorge and Kula, 2000).
This approach aimed to separate the replication and the expression of plasmids from the growth
phase. More than 1000 copies of plasmids can be obtained. Other issues can be investigated
such as the reduction of the generation time and generation numbers throughout the process, the
deceleration of the metabolic speed, e.g., through a reduction in temperature, and finally the
development of more stress‐resistant microbial strains. Thus, the critical issue to be considered
during the scale up is to develop a proper process design that improves the physiological
conditions and greater enhances the metabolic accuracy by minimising microbial stress exposure.

MICROBIAL TOLERANCE
During the optimisation of fermentation process, it is crucial to increase tolerance of the strain to
the desired product, this task can be performed at any stage of strain development (Ling et al.,
2014). This will allow the direction of metabolic flux to the formation of a desired product.

Generally, the improvement of microbial product tolerance is recommended to generate the

product at a requisite level. Usually, the development of a product‐tolerant strain can be done by
a serial of cell subculture with the increase of the concentrations of product with or without
mutagen treatment, followed by the identification of cells that grow faster (Utrilla et al., 2012).

By repeating the subculture process, the tolerance level can be increased. Improvement of
product tolerance has also been achieved by using rational engineering approaches, such as the
use of an efflux pump (Dunlop et al., 2011) or the manipulation of ionic membrane gradients of S.
cerevisiae (Lam et al., 2014). The overexpression of genes encoding the L‐valine exporter led to
an increase in L‐valine production titer by more than 40% in E. coli (Lam et al., 2014, Park et al.,
2007). The engineering of efflux pumps to improve product tolerance and/or titer are a strategy
that deserves investigation. However, the producer cells selected from serial sub culturing or
rational engineering do not necessarily overproduce the product. This is mainly because
enhanced product tolerance at high concentrations does not necessarily correlate with an ability
to synthesise product at increased specific productivity and yield.

Phage Invasion
Many biotechnologically‐important by‐products are produced microbial cells which cover a wide
range of applications in various sectors of environmental biotechnology. However, serious
problems occurred in laboratories and factories when optimising the process on a large scale.
Among potential problems that face the effectiveness of the process and cause hurdles for the
by‐product production, the prevalence of natural parasites such as bacteriophages during the
fermentation process is quite common (Callanan and Klaenhammer, 2002).

Bacteriophages like all viruses are obligate intracellular parasites that need a host cell to multiply.
They require a susceptible host cell that is actively growing to multiply (Lu et al., 2003). The
bacteriophage infection resulted in detrimental problems (e.g., loss of bacterial culture and
bioproduct formation, the spread of bacteriophage contamination throughout the laboratory or
factory (Ogata, 1980).

7
This problem is critical, and many efforts should be made to avoid such operational problem that
contributes to cell lysis of cultures even after extensive cleaning process. Accordingly,
decontamination and disinfection process remain difficult and even impossible in large factories.
Consequently, phase invasion is a serious problem and if phage is present during the process
fermentation, it can spread rapidly throughout the plant and resist to decontamination process.
This causes the persistence of phage for long periods and the reappearance of phage infections
in the factory even after treatment (Primrose, 1990). Not only contamination but also phages affect
the genetic material of the host, for instance upon infection, phages can insert their genetic
material into the host chromosome, and lead to the formation of prophages, instead of lysing the
cell (Los et al., 2004). Consequently, the growth rate is lowered and the efficiency of synthesis of
bioproducts is decreased (Jones et al., 2000).

To avoid cited problems, different issues need to be taken into consideration (e.g. the location of
a bioprocessing factory, plant design) which have a crucial impact on the occurrence of infections.
In the case of bioprocesses involving genetically modified E. coli, the factory should be far from
sewage disposal plants to minimize the risk of coliphages propagation and dispersal, known to
be present at high number in sewage plants.

Besides this, the sterilisation and the control of sampling ports should be done carefully since
these places may increase the propagation of unwanted phages (Los et al., 2004). In this regard,
the decontamination process should be performed carefully with high precautions to prevent
phage infection during fermentation process. A better knowledge of the fundamental properties,
genomic taxa, and the mechanism of infection should be provided. Generally, infection occurs
when the phage is adsorbed or attached to the bacterial surface, having a complementary site.
Then, viral nucleic acid enters the cells where transcription and transduction of viral genes
occurred. Then, viral lysozymes are synthesised and attack the peptidoglycan layer of bacterial
cell walls to induce cell lysis and new phage particles are released into the environment (Holder
and Bull, 2001; Budzik, 2003).

Recent debates focused on the development of effective methods of decontamination since many
phages was found to be resistant to the common processes of sterilisation such as drying,
freezing and thawing and phages can survive for long periods (Atmar et al., 1993). Furthermore,
if infection occurred during the bioprocess, the bioreactor should be shut down immediately since
continued aeration of the culture medium will dispense large numbers of phages into the
environment of the factory which make the clean‐up process difficult.

Moreover, culture having excess foam is more sensitive to phage infection, so the control of foam
addition during the fermentation is particularly important. Besides, if the inlet and the outlet of air
are very close, their risks of contamination are high and different aeration system, e.g. filters, or
compressors will be later contaminated (Primrose, 1990; Wunsche, 1989). The sterilization of the
bioreactor with high temperatures is suggested and the bioreactor must be held at a high
temperature for as long as possible. Besides, the phage assay should be carried out after each
step of sterilization to ensure the absence of viable phages.

Moreover, different air filters and compressors, air inlets and outlets must be extensively sterilized,
because bacteriophages can pass through most of the filters, if not all. Besides, all exterior
surfaces of the pilot line must be disinfected.

All these issues depend on the size and design of the plant. The washing of whole plant with hot
water is suggested, although however, this seems to be not practicable. Furthermore, samples of
the washing liquid should be subjected to a phage assay to determine the extent of infection and

8
the efficiency of the washing procedure in phage reduction. To sum up, urgent measures must be
taken during the bioprocess to prevent the phage invasion. Most of the proposed methods
include:

Good laboratory hygiene;

(a) Employment of phage resistant mutants during the bioprocess; and;
(b) Use to potential chemicals designed specifically to halt the multiplication and
dispersal of phages.

i) Good laboratory hygiene is a crucial issue in preventing phage invasion. All the equipment and
apparatus should be washed and sterilized. The medium of the bioprocess (i.e., raw material)
should be sterilized and the assay of phage infection should be done regularly. Another key factor
in preventing phage evasion is the verification of the purity of the seed culture in the glycerol
stocks and the get rid of the spoiled cultures by autoclaving (Los et al., 2004; Ogata, 1980);

1. Employment of phage‐resistant mutants involve the isolation of a potential culture

with high resistance to phages destruction, this latter will be used after the process in
the fermentation process. The choice of potential mutants should be carried out carefully
and the selected strain should exhibit the same production capabilities as does the original
strain. However, the examination of phage infection should be done herein regularly since
some resistant mutants alone cannot cope with phage infections. Although mutants
contribute to the stabilization of the fermentation process, it is still important to ensure that
all phages are eliminated (Los et al., 2004; Jones et al., 2000);

2. Employment of chemicals that alter phage multiplication and they do not show a
side effect on the by‐product formation. For instance, the employment of divalent
cations, chelating agents or DNA injection is quite common. These chemicals will inhibit
the phage multiplication without interfering with byproduct formation. Another option can
be investigated in this aim which consists of the employment of specific antibiotics (Ogata,
1980) as well as some vitamins such as ascorbic acid that are reported to cause virus
inactivation (Wang and van Ness, 1989). There is overwhelming literature that suggested
that certain vitamins were capable of hindering bacteriophage infection (German and
Dillard, 2006).

Higher temperatures are required since phages are sensitive to heat; however, some phages can
resist to temperatures (Callanan and Klaenhammer, 2002). Generally, the sensitivity of
bacteriophages to environmental factors correlates with the corresponding properties of the host.
The employment of chemicals detergents and disinfectants is also common, for example sodium
dodecyl sulfonate showed a potential phagocidal effect (Wunsche, 1989). UV treatment can be
extremely helpful in keeping production rooms clean and disinfected (Los et al., 2004; Wunsche,
1989). UV rays are known to disrupt DNA phage and inactivate the phage. However, some
phages have the ability to repair their own DNA damage (Kokjohn et al., 2005) which make this
process not practicable.

PROCESS FAILURES
The identification and genetic engineering of the potential organism followed by prudent piloting
certainly is very important to success with scale up fermentation. Therefore, the success of any
biotechnological process lies in its capacity to be scaled up to the industrial level. So, the effective
technology transfer from the lab scale to the large scale, required the development of more efforts
and attention to several issues. Thus, a scale‐up strategy that combines integrated teamwork with

9
solid engineering efforts can go a long way despite its complexity and difficulty. This led to
minimize the cost of the whole process on time and on budget. However, to achieve this goal,
however, there are many bottlenecks and limitations in asepsis that’s needs to be resolved.

To success the scale‐up and the commercialization of any process technology; various points
must be considered and managed in such as business, market, and technology risks. The risk
identification, risk assessment and prioritization, and development and execution of risk mitigation
plans are of importance to achieve a successful track in scale‐up.

Among strategies, piloting is used for reducing the risk of failure at larger scale owing to many
factors:
1. Operational issues (fouling, plugging, corrosion, etc.);
2. Recycle problems (build‐up of impurities); and
3. Process capability issues (low yield and capacity, poor product quality, etc.).

At an early stage of scale up, the process failure is due to the inability to produce large quantities
of a viable inoculum, overheating of the bio piles. Besides, the lack of understanding of the
potential markets and process scale up can cause an early failure that generates latter negative
impacts on the development of the technology.

Other technical problems such as technical complexity, high installation and operating costs are
again a major cause of failure. In this regard, the optimization of the fermentation process is
required to avoid process failure. Once the feasibility of the production has been demonstrated,
the fermentation can be succeeded. Other points have to be taken carefully during the large‐scale
fermentation such as the stability of the strain. Accordingly, most of expressions systems based
on plasmids are sometimes unstable and depended with the host characteristics (Posno et al.,
1991), culture conditions (Koizumi et al., 1985; Reinikainen and Virkajärvi, 1989). Thus, a better
documentation of the strain and the cloning procedure of the strain is required and approved by
a competent quality assurance service. These considerations impact strongly on the scale up as
well as the downstream process. Moreover, the quality of the communication between developers
of the fermentation and the downstream process is crucial and contributes to the success of the
scaling up the process.

One of the most important factors of success is to define the variables that could affect the
productivity or the quality of the product. Statistical data analysis enables the visualisation of the
interactions among experimental variables, leading to predictions of the experimentation (Murphy,
1977). The factors that affected the scale up process are the number of generations, the mutation
probability, and medium sterilization, the quality of temperature and pH regulations, agitation,
aeration and pressure. Consequently, it is pivotal to the preparation for the scaling‐up of a process
to first scale‐down to the pilot scale of the conditions of culture (Kwanmin, 1989).

The process should be characterized and validated by some controls to evaluate the robustness
of the procedure as well as its reliability and reproducibility. All the equipment must be validated
for good manufacturing practices like the qualification of the installation, the validation of cleaning
in place and sterilizing in place, maintenance and calibration plan (Naglak et al., 1994).

Lack of technological knowledge, insufficient management capacity, high construction costs and
the absence of maintenance services contributed to failure process. The lack of success can be
explained by inadequate technology and know‐how, and a strong focus on quantitative
implementation targets.

10
Furthermore, the scale up of fermentation depends on power input or airflow rate per unit volume,
as well as the type of impeller. The difficulty in scaling up a fermentation process is that it is
unlikely to simulate simultaneously in geometry, fluid kinetics and fluid dynamics. It is imperative
to develop a feasible approach that considers various parameters, which are useful to the scale
up of a process. Added to these aspects, a well‐ planned pilot plant design and operation is
essential to minimize the risk of failure. A laboratory development program that combines both
the core technology and the necessary data for a pilot plant to handle separation steps, waste
streams, and material handling operations is required to meet different pilot goals. These latter
must be clearly stated to prevent commercial failure.

SCALE DOWN
As the conditions achievable at the laboratory level are impractical at large scale that is why scale
down operations, which means are laboratory or pilot scale experiments conducted under
conditions, that virtually mimic the conditions occurring at industrial level, are used to study yield
objectives at industrial level. It means that scale down operation is the part of scale up studies.
This approach not only helps development of a new product but also to improve existing
fermentation product at its full scale.

Aspects that are contemplated while designing laboratory or pilot plant experiments in the context
of scale down operations are discussed as follows:

A. Media Design
The media used in experiment should be relevant to media suitable for industrial level. However,
most of the time media from cheaper recourses are preferred to keep cost low. However, proper
optimisations of such media are compulsory before application at large scale levels. Different
fermentation media have different objective for example if our desire product is primary metabolite
or biomass then target is maximum growth of microorganism but in case of secondary metabolite
is not related to growth. Industrial fermentation media generally require a carbon source, nitrogen,
phosphorus, sulphur and vitamins.

B. Media Sterilization
Considering bulk amount of media that have to be sterilized at large scale demands for longer
exposure time and higher temperature as compare to lab scale specially for batch sterilization. It
means that time for sterilization at smaller scale must have to be increased to mimic requirements
of industrial scale. Otherwise, media taken from industrial unit which are already functioning can
be used for scale-down studies. This approach can further add an advantage of highlighting
situation under continues sterilization where loss of media quality could occur. However,
continuous sterilization can itself be used for laboratory or pilot scale experiments.

C. Inoculation Procedure
Inoculate of every fermentation is not necessary be under optimum conditions due to range of
possible circumstances. That’s why scale down operations can help to predict respective
consciences of such events by mimicking them i.e., by using stored inoculums or inoculum of old
ages.

11
D. Number of Generations
Higher number of generations is required for completion of fermentation process at industrial level
which may rises question on stability of strains used as inoculum. To mimic industrial conditions
at down scale operation serial sub-cultures can be used to ensure that strain remains suitably
stable or not. It becomes more important when dealing with recombinant strains.

E. Mixing
Degree of mixing decreases with the increase in scale. Laboratory experiments can be modified
by using pulse medium feeds or fluctuating process conditions such as oxygen concentration, pH
and temperature which may allow predicting suitability of new strains for industrial exploitation.

F. Oxygen Transfer Rate

High oxygen transfer rates at laboratory level are easy to maintain as compare to industrial level
that’s why a scale down operation should reflect oxygen transfer rates achievable at full scale
fermentation. In suspension of aerobic microorganisms, the concentration of dissolved oxygen
depends on the rate of oxygen transfer from gas to liquid phase [6]. It is necessary as it will be
unrealistic if fermentation is designed at high oxygen transfer rate that is difficult to achieve at
industrial scale.

12
 Figure: Flow Sheet Depicting Layout of a Commercial Fermentation Plant.

PILOT PLANT STUDIES

A pilot plant can be referred as a model that mimics its commercial prototype fermentor. There
are number of definitions suggested for pilot plants. Ogorzaly defined pilot plant as “An assembly
of equipment devoted to studying the critical features of process operation”.

An even general definition of pilot plant was given by Paluzzi “A pilot plant can be referred as
a tool that intends to allow investigation of a process or process related problems on a
manageable scale in a more realistic manner and within manage able time”. It is also defined
as a part of the pharmaceutical industry where it is used to produce viable product from a lab
process.

The definitions clearly demonstrate that the pilot plant is not the end itself; however, it is a way to
achieve a goal of installing full scale, commercial and productive fermentation unit. Furthermore,

13
a manageable scale means limited time utilisation, less risk of capital loss and security of other
resources.

A. IMPORTANT ASPECTS OF PILOT PLANTS

1. Size
Pilot plants can be divided into different classes on the bases of their size as demonstrated
below.

1a. Bench Tip Pilot Plants (Micro Units)

It includes pilot plants that may fit to bench top or inside a small laboratory. Generally, volume of
fermentation under study uses to be smaller than 1 L in a total ground area of 0.5 to 1.0 m2.

1b. Research Scale Pilot Plants

Such plants seem like the workhouse of the industry. However, the research plants may vary in
size from several frames to a unit that may occupy the small size building. The tanks used in
plants have volume around 4 m2 occupying area of 2-15 m2.

1c. Prototype Units

These plants work at the scale close to actual industrial unit having exceptionally large tanks of 4
to 40 m2 occupying area as much as 900 m2.

2. Costs
The pilot plant uses to be generally expensive as same number of instruments is required for pilot
plant as for a full-scale industrial unit. However, for full scale industrial process some extra
instruments and process controllers must be installed. By using reduced scale pilot plants i.e.,
mini units the cost of a pilot plant can be cut off up to 40%. An important expects to be considered
using reduced scale is the fact that the stability of process decreases due to difficulty in designing
it by mean of process control and management. For instance, certain materials use to start
accumulating in recycling stream, where such accumulations are not easily detectable at bench
scale, which decreases quality of study done.

3. Time Scheduling
The stage of industrial plan at which the pilot plant should be built remains an important question.
Due to pressure of commercial competition the industrialists might look to reach production phase
as soon as possible even sometimes by jumping to full scale and skipping pilot plant. A possible
solution for this problem is to start pilot and full-scale plant at same time where pilot plant can be
constructed in shorter time period so that the process can be studied before completing of full-
scale unit. However, it involves risk that if pilot plant tests prove selection of instruments, that
would have been installed in full scale production plant also, wrong then a huge economical loss
must be expected. The equipment with short installation and adjustment time may also help to
start production in manageable time after testing pilot plant.

4. Flexibility
The pilot plant must be designed along with suitable flexibility. It means that there must be enough
possibilities to take frequent samples and its analysis without disturbing basic process.

14
Furthermore, a continuous supply of raw material and easy collection of products should be
possible.

5. Recycle Streams.
To reduce the amount of waste and cost of a process the recycle stream should be part of design.
However, recycling may have some adverse effects on process, i.e. fouling of equipment’s, which
means it should be intelligently managed.

6. Safety
Possible risks should be mapped before starting any pilot plants that are dealing with any harmful
raw materials or products. An effective safety plan should available while working on such pilot
plants.

7. Use of Computers
There is an essential role played by computers in the operation of pilot plant. Most of the process
controlling online analysis and data processing use to be computerised. For example, use of
computers to control feed stream increase the speed of process so it makes feed stream more
productive besides better control and safety. Computers can also be used to stimulate process
by using model study. It means that if certain data relative to process is already available then it
can be further used to make simple experiment sufficient to obtain the data on required
parameters. It decreases the cost of product due to decrease in labour which may add a further
advantage to list. However, a good computer model is that which provides a quick and simple
stimulation of a process within least number of required experiments. Somehow, risk of premature
and over enthusiastic use of models for scaling up can lead to wrong assessment of certain
parameter. For example, complex process of large-scale aerobic fermentation cannot be simply
scaled up by computers directly.

CONCLUSION
Scale up studies must execute before constructing a large-scale fermentation unit to understand
the technical components of a large-scale fermentation setup and to reduce the economic risks.
To keep scale up studies within defined economical limits scale-down operations are important
as they mimic the challenges of large-scale production displayed at the scale that is reduced and
easier to manage. However, conducting such studies use to be a challenge over their own.

Microbial Experimentation in the Fermentation Industry: The Place of

the Pilot Plant
When the microorganism used in a fermentation is new, experimentation must be carried out to
determine conditions for its maximum productivity. It is usual to initiate the studies in a series of
conical flasks of increasing size and to progress through a 10 to 20 litre fermentor to a pilot plant
(100-500 litres) and finally to a production plant (10,000- 200,000 litres). The processes involved
in the increasing scale of operation culminating in the production plant are known as scaling up.

On the other hand, in a well-established fermentation procedure, any change to be introduced

must be experimented on and tested out in a pilot plant whose function is to simulate the
conditions and structures of the production plant. This procedure is often referred to as scaling

15
down. The processes of scaling up and scaling down are essentially in the domain of the chemical
engineer who depends on data supplied by the microbiologist.

Information gathered at the shake flask stage is used to predict requirements in the pilot plant
which itself serves a similar purpose for the production plant. The optimum requirements of
medium composition, aeration, temperature, redox potential, pH, foaming, etc., are determined
and extrapolated for the next higher scale. The pilot fermentor is also used for training new recruits
in the fermentation industry; it may also be used for continuous fermentation where a large enough
number of them exist.

One approach which helps facilitate translation of information from the pilot plant to
production is to reproduce the production plant as a geometrical replica of the pilot plant.
Baffles, agitators, etc., are increased exactly according to a predetermined scale. This, however,
does not entirely solve the problem because the mere increase in volume immediately poses its
own problems. If the same level of productivity as encountered in the pilot study is to be
maintained, then agitation and aeration may be applied at a level higher than that expected in a
proportional increase in the production fermentor.

Fermenter Design and Construction

The main function of a fermenter is to provide a suitable environment in which an organism can
efficiently produce a target product that may be cell biomass, a metabolite or bioconversion
product. Most are designed to maintain high biomass concentrations, which are essential for
many fermentation processes, whereas control strategies largely depend on the particular
process and its specific objectives. The performance of any fermenter depends on many factors,
but the key physical and chemical parameters that must be controlled are agitation rate, oxygen
transfer, pH, temperature and foam production.

Laboratory fermentations may be performed in bottles or conical flasks that can be shaken to
provide aeration where necessary. These vessels are normally plugged with cotton wool or a
styrofoam bung to prevent microbial contamination, but this can lead to evaporation losses and
restricted exchange of gases. Consequently, even on a laboratory scale, vessels specifically
constructed for fermentations are usually preferred. For industrial processes, fermenters with
capacities up to several hundred thousand litres are used. These are mostly purpose-built and
designed for a specific process, although some flexibility may be necessary in certain instances.

Their design, quality of construction, mode of operation and the level of sophistication largely
depend upon the production organism, the optimal operating conditions required for target product
formation, product value and the scale of production. Overriding considerations reliability and the
need to minimise capital investment and running costs.

Large volume, low value products that include many of the traditional fermentation products, such
as alcoholic beverages, are usually produced using relatively simple fermenters and may not
operate under aseptic conditions. There are often fewer risks when operating such non aseptic
fermentations at extreme pH or high temperatures, or where protected substrates are used. These
are substrates that few microorganisms will utilise. Nevertheless, strict adherence to good
manufacturing practices reduces the risk of microbial contamination of pure culture (axenic)
fermentations. Other fermentations do not involve pure culture inoculum and actively encourage
the development of indigenous microorganisms, e.g. some food fermentations and wastewater
treatment. Conversely, fermentations producing high value, relatively low volume products,

16
especially pharmaceuticals, invariably demand more elaborate systems and operate under strict
aseptic conditions.

Traditionally, fermenters have been open cylindrical or rectangular vessels made from wood or
stone. Some of these are still used, particularly for certain food and beverage fermentations.
However, most fermentations are now performed in closed vessels designed to exclude microbial
contamination. These fermenters must with stand repeated sterilisation and cleaning, and should
be constructed from non-toxic, corrosion-resistant materials. Small fermentation vessels of a few
litres capacity are constructed from glass and/or stainless steel. Pilot-scale and many production
vessels are normally made of stainless steel with polished internal surfaces, whereas
exceptionally large fermenters are often constructed from mild steel lined with glass or plastic, in
order to reduce the cost. If aseptic operation is required, all associated pipelines transporting air,
inoculum and nutrients for the fermentation need to be sterilisable, usually by steam.

Normally, fermenters up to 1000 L capacity have an external jacket, and larger vessels have
internal coils. Both provide a mechanism for vessel sterilisation and temperature control during
the fermentation. Fermentations can be carried out under non-aseptic conditions where the risk
of contamination is not a major concern. However, many fermenters must be designed for
prolonged aseptic operation (Table below).

Table: Examples of Aseptic and Non-Aseptic Fermentations

The design rules for an aseptic bioreactor demand that there is no direct contact between the
sterile and non-sterile sections to eliminate microbial contamination. Any connections to the
fermenter should be suitable for steam treatment to kill any resident microorganisms and systems
must be designed to allow aseptic inoculation, sampling and harvesting.

Every individual part should be easily maintained, cleaned and independently steam sterilizable,
particularly valves. Most vessel cleaning operations are now automated using spray jets, which

17
are located within the vessels. They efficiently disperse cleaning fluids and this cleaning
mechanism is referred to as cleaning-in-place (CIP).

Associated pipework must also be designed to reduce the risk of microbial contamination. There
should be no horizontal pipes or unnecessary joints and dead stagnant spaces where material
can accumulate, otherwise this may lead to ineffective sterilization. Overlapping joints are
unacceptable and flanged connections should be avoided as vibration and thermal expansion can
result in loosening of the joints to allow ingress of microbial contaminants. Butt-welded joints with
polished inner surfaces are preferred.

Other features that must be incorporated are pressure gauges and safety pressure valves, which
are required during sterilisation and operation. The safety valves prevent excess pressurisation,
thus reducing potential safety risks. They are usually in the form of a metal foil disc held in a
holder set into the wall of the fermenter. These discs burst at a specified pressure and present a
much lower contamination risk than spring-loaded valves. Pumps should be avoided if aseptic
operation is required, as they can be a major source of contamination.

Centrifugal pumps may be used, but their seals are potential routes for contamination. These
pumps generate high shear forces and are not suitable for pumping suspensions of shear-
sensitive cells. Other pumps used include magnetically coupled, jet and peristaltic pumps.
Alternate methods of liquid transfer are gravity feeding or vessel pressurization. In fermentations
operating at high temperatures or containing volatile compounds, a sterilisable condenser may
be required to prevent evaporation loss. For safety reasons, it is particularly important to contain
any aerosols generated within the fermenter by filter-sterilising the exhaust gases. Also,
fermenters are often operated under positive pressure to prevent entry of contaminants. However,
this may not be applicable if pathogens or certain recombinant DNA microorganisms are being
used.

The type of fermentation will always dictate the choice of equipment. Medium composition
(chemically defined or complex), and organism morphology (filamentous or unicellular), set
restrictions on the type of equipment that can be used.

1. VESSEL

18
In selecting equipment, one must determine what scale of operation is most suitable. The data in
the Table show medium consumption rate as a function of dilution rate and fermentor working
volume. Thus, a bench top-scale fermentor is more convenient to operate. In addition, investment
and space requirements are modest, and peristaltic pumps of adequate reliability are available.
A benchtop vessel is also easily dismantled for cleaning. This is especially advantageous after
long mycelial fermentations.

Table: Medium Requirements (litres per day) at Different Dilution Rates

Fermentor 0.01 h-1 0.05 h-1 0.5 h-1 1.0 h-1
Working Vol.

250 ml 0.06 0.3 3 6

1,000 ml 0.24 1.2 12 24
5 litres 1.2 6.0 60 120
10 litres 2.4 12.0 120 240

However, small-scale continuous culture offers some unique problems. When operating at
volumes of less than 500 ml, sampling can reduce the volume, affecting the dilution rate and thus
the metabolism of the organism.
Feed rates that are required to obtain low dilution rates can be very difficult to control, and feeding
a complex medium at low flow rates can result in settling out of insoluble medium components in
feed lines, causing plugging.
It is often difficult to get a sample for off-gas analysis due to low back-pressures and low aeration
rates. Also, wall growth can represent a larger fraction of the total population in small vessels.
Thus, wall growth may become a significant variable in studies of population dynamics.
Teflon vessels and silicone coating reagents can be used which may reduce or prevent wall
growth in particular fermentations.

19
Figure: A Simple Laboratory Fermentor Operating on a Continuous-Cultivation Basis.

Sterilise in Place Bioreactors

In the perfect fermentation laboratory SIP bioreactors (Figure 5.2) should dominate. The
bioreactors, usually 10 L working volume and above, are usually jacketed and thus require an
external steam supply to the vessel jacket. This may imply a considerable extra cost should steam
not be available to the building. In addition, extra floor space is needed in order to incorporate
additional pipe work and frame.

State of the art reactors have excellent control and monitoring of the sterilisation cycle and
relatively few operator intervention steps are required. Sterilisation protocols, supplied by the
manufacturer, must be strictly adhered to obtain the best performance and to work within safety
regulations. The main advantage here is that the temperature of the medium is always controlled.
Heating times are usually fast, as are cooling times, and this means that the medium is not
‘overcooked’, as it often is when sterilised in a standard autoclave. The medium is normally stirred
slowly throughout the sterilisation process, thus ensuring good heat distribution and avoidance of
local hot spots. Thus, medium quality is assured.

SIPs often have the additional benefit that, with the correct independently sterilisable port
installed, medium addition points can be independently sterilised, a distinct advantage when
operating in fed-batch or continuous mode.

SIP sterilisation has the additional benefits that the air supply can be turned on to the vessel
during cooling, so generating a slight positive pressure inside the vessel and its associated lines.
Hence, any potential weakness in the sterile envelope may be overcome.

Of course, any significant compromise of the sterile envelope implies that fermentor contents will
be aerosoled into the laboratory. It is therefore essential to check physically the condition of all
lines/addition points and tubing leading to and from the vessel pre- and post-sterilisation.

20
Culture Medium
Media requirements depend on the type of microorganism being used in the fermentation
process, but the basic requirements remain the same; source of energy, water, carbon source,
nitrogen source, vitamins, and minerals. Designing the media for small scale laboratory
purpose is relatively easy, but media for industrial purpose are difficult to prepare.
The culture medium should:
• Allow high yield of the desired product and at fast rate,
• Allow low yield of undesired products,
• Be sterilised easily,
• Yield consistent products i.e., minimum batch variation,
• Be cheap and readily available,
• Be compatible with the fermentation process, and,
• Not pose environmental problems before, during, or after the fermentation process.

The culture medium will affect the design of the fermentor. For example, hydrocarbons in the
media require high oxygen content so an air-lift fermentor should be used. Natural media
ingredients are cheap, but they have high batch variation. On the other hand, pure ingredients
(defined media or formulated media) have very little batch variation but are expensive. The media
should support the metabolic process of the microorganisms and allow biosynthesis of the desired
products.

Carbon & Energy source + Nitrogen source + Nutrients ―→ Product(s) + Carbon

Dioxide + Water + Heat + Biomass

Media are designed based on the above equation using minimum components required to
produce maximum product yield.
Important components of the medium are carbon sources, nitrogen sources, minerals, growth
factors, chelating agents, buffers, antifoaming agents, air, steam, and fermentations vessels.

2. NUTRIENT FEED RESERVOIR

If carboys are used for feed reservoirs, they should be calibrated by volume or placed on a
weighing device so that integrated nutrient feed rates can be determined accurately over several
hours or days.
An additional port on the reservoir should be included so that heat-labile nutrients can be added
separately after sterilisation. A vent through a sterilising filter must be attached and a nutrient
feed-line should be installed. The feed-line should be a stainless-steel tube going to the bottom
of the vessel.
A magnetic stirring bar must be included to suspend any particulates and mix nutrients which are
added after sterilisation of the bulk medium.
Silicone rubber tubing should be used where tubing is required. Silicone rubber tubing can be
sterilised repeatedly, without loss of resilience, and has excellent chemical resistance. The
method of connecting tubing should be aseptic, fast and reliable. These connectors are made of
stainless steel and can be sterilised by flaming immediately before connecting.

A pipette attached to each reservoir bottle as shown in Fig.1 (lecture notes), will allow
measurement of the instantaneous medium flow rate and adjustment of the pump rate. To obtain
the medium flow rate, the clamp that isolates the pipette from the system is removed, allowing
medium to fill the pipette by gravity to the same height as in the reservoir. The medium reservoir
is then clamped off. The time required to pump a known volume of medium from the pipette is

21
used to calculate the liquid flow rate. The pump is adjusted, and the procedure is repeated until
the desired flow rate is achieved. Over the course of a long fermentation the pipette can clog and
may need to be replaced. Therefore, it should be connected in such a way that it can be removed
if necessary.

One should be aware that the flow rate through a peristaltic pump is dependent on the resilience
of the tubing. Because silicone rubber tubing loses it resilience with use, a long piece should be
used so that the pump can be removed every other day to a new section. For extended operation,
replacement of the tubing may be necessary.

The medium inlet should be designed with a drip-tube in the line immediately before the vessel.
A drip-tube provides a break in the medium flow path that prevents microorganisms from growing
into the feed-line. A drip-tube can be purchased or constructed.

Figure: A medium break for continuous cultures.

A: medium feed; B: air feed; C air/medium feed to culture vessel (through head plate)

Some fermentor manufacturers recommend that the medium feed go into the vessel with the air.
In certain cases, this may suffice, but where insoluble medium components make up even a small
percentage of the feed stream, the air inlet can plug after prolonged use.
In some cases, very low feed rates may be required. It is difficult to find pumps that provide
reliable, low flow rates. In such cases, a timer may be used to turn the pump on only a certain
percentage of the time. In this way a flow rate can be attained which is a fraction of the minimum
flow rate of the pump. This is particularly helpful when non-limiting substrates are to be fed, such
as product precursors or antifoaming agents. The use of this system for feeding the limiting
nutrient, however, will probably result in metabolic cycling that may not be desirable.

22
Preparation of the Inoculum
Microbial inoculum must be prepared from the preservation culture so that it can be used for the
fermentation process. The aim of inoculum preparation is to select microorganisms with high
productivity and to minimise low productive, mutant strains.
The process involves several steps.
• First generation culture is prepared from the preservation culture on agar slants which is
then sub-cultured to prepare “working culture”. At this stage the microorganisms start
growing. In small scale fermentation processes working culture is used as inoculum, but
for large scale fermentation inoculum preparation involves additional steps.

• Second, sterile saline water or liquid nutrient medium containing glass beads is added to
the agar slants and shaken so that microbial suspension is prepared. This suspension is
transferred to a flatbed bottle which contains sterile agar medium. The microorganisms
are allowed to grow by incubating the bottle.

• Third, the microbial cells from the flat bed bottles are transferred to a shaker flask
containing sterile liquid nutrient medium and is placed on a rotary shaker bed in an
incubator. Microorganisms grow at a rapid rate due to aeration.

• Fourth, microbial cells from the shaker flask can be used as seed culture which are then
added to a small fermentor and allowed to grow for 1-2 days. This simulates conditions
that exist in the larger fermentor to be used for production of metabolites.

• Finally, the microorganisms are transferred to the main fermentation vessel containing
essential media and nutrients.

3. BROTH REMOVAL AND LEVEL CONTROL

To maintain accurate control of the dilution rate, it is especially important that the ungassed
volume be determined and controlled.
There are two parameters used to control the ungassed vessel volume:
(i) Gasses liquid level (assuming gas hold-up is constant), and
(ii) Mass of vessel contents.

Because control of gassed volume is more practicable at the bench scale, it is the most widely
used. However, constant gas hold-up is difficult to achieve.
Alternatively, mass control can be more accurate if properly installed, but it is much more
expensive to implement on the bench-scale of operation.

There are two methods of liquid level control. In the technique most often recommended by
vendors of continuous-culture vessels, the air outlet is placed at the broth surface. The exhaust
air is then directed into a reservoir that is vented aseptically to the atmosphere. When the culture
volume increases, broth is forced out into the product reservoir with the air. Such a system suffers
from the fact that cells can sometimes be at a different concentration at the air/liquid interface
than in the bulk liquid.

This can cause a non-representative sample to be pumped from the vessel. Such partitioning can
be affected by pH, ionic strength, and surface-active agents. In some cases, the effect is so
dramatic that foam fractionation has been used to harvest bacterial cells.

23
Figure: A Pilot-Scale Fermentor Operating on a Continuous-Cultivation Basis.
In contrast, a liquid level controller can be used. Although this technique requires more equipment,
it allows withdrawal of the whole broth from below the liquid surface. Several liquid level controllers
are available. Most are based on a sensor within the vessel that makes physical contact with the
culture fluid. Because the sensor is located within the tank, failure of the probe will end the run.
Depending on the type of sensor used, inaccuracies in volume can be caused by variations in
agitation, aeration, foam generation, and wall growth. One also loses the flexibility of being able
to change the level in the vessel without contaminating the culture.

Ultrasonic level controllers are available which can be attached to the outside of a glass or steel
vessel. These sensors can be attached after autoclaving, and if a sensor malfunctions it can be
replaced without violating sterility. Sensitivity can be adjusted so that effects due to foam can be
minimised. Problems caused by gas-holdup and foam cannot be fully eliminated, however.

The use of mass as a control parameter eliminates the problems of variable gas hold-up and
foaming but may introduce other problems. Because of high equipment weight relative to the
weight of the broth being measured, the use of a load cell to control volume is difficult at volumes
less than 5 litres.

Tubing connections must be carefully supported to avoid inadvertent changes in weight, and on
larger vessels, piping must be designed with flexible coupling to avoid supporting the vessel.
Differential pressure sensors can also be used to measure the mass in a fermentor. These
devices measure the hydrostatic pressure between the top and the bottom of the vessel. The
pressure is then used to calculate the fermentor volume. The practical lower limit of these sensors
is about 500 to 1,000 litres.

A stable continuous culture requires an accurate and reliable pump. The pump should be
cleanable and steam sterilisable. Most importantly, it should be capable of long-term aseptic
operation. On the bench-top scale, peristaltic pumps meet these criteria. As fermentation scale
increases, peristaltic pumps become impractical.

24
• Peristaltic: They provide constant flow rate and mild pumping. Liquids do not back-flow, so
check valves are not required. Peristaltic pumps are also used to control flow rate. This
controls by squeezing and releasing pulse flow. The tubing through which the liquid flows is
housed on roller. The circular motion (speed) pinches and released the tubing. More the rollers
smoother will the flow. This pump can be controlled by process controller with inputs of
respiratory quotient and optical density. Mostly used for addition of acid and base.

Figure: Peristaltic Pump

• Mini or Delta: They are used for adding pH control agents like acid, alkali, and adding
anti foam agents. Mini pumps are fixed-speed pumps and can pump against the pressure.

• Syringe pump: Useful in case of fed batch for controlling the liquid flow rate. A syringe
filled with liquid and secured onto the pump. Syringe plunger linker to movable piston and
secured firmly to the main body of pump. Regulated piston movement controls the flow
rate. This is ideal for low flow rates.

• Larger pump: They are used to add nutrients to the medium. Speed can be varied by
altering the bore size of the tubes.

• Diaphragm pumps can be used which provide reliable, adjustable flow rates. These
pumps have several advantages over centrifugal and gear pumps. Perhaps the greatest
advantage is the lack of a mechanical seal. Over the course of time mechanical seals
wear and can be a source of contamination. The diaphragm pump has no such seal and
contamination problems are minimised. Second, the pump head and the lines leading to
and from the head can be sterilised by steam. Finally, the flow rate can be maintained
over a wide range of pressures. This means that changes in the back-pressure of the
vessel required to obtain maximal oxygen transfer can be accomplished without
substantially changing the pump flow rate.

25
 Figure: Diaphragm pump

Silicone tubing should be replaced by piping. Generally, in situ steam sterilisation is required at
greater than the 10-litre scale.

26
4. SAMPLING
When complex media and mycelial organisms are studied, the culture broth can attain high
viscosity, and removing a sample aseptically can be difficult. One technique requires the air inlet
to be connected to the sample port as shown in Fig.4. To purge the sample line, the accessory
airline is opened and sterile air forces a portion of the purge volume from the sample line. The
accessory line is then closed. By unplugging a portion of the sample line in this way, the rest of
the sample line is more easily cleaned.

Figure: Sampling system suitable for small laboratory fermentors.

A: connection to sample line (through head plate); B: metal fitting holding sample
container; C: sample container; D: syringe (50 mL)

Caution must be exercised because this procedure can generate aerosols. The use of large-
diameter tubing in the sample line can also ease plugging problems but requiring too great a
purge volume. In many cases, the total sample and purge volume may become so large that the
dilution rate will be significantly affected. In general, the sample volume should not exceed 10%
of the vessel volume.

27
 Figure: Sampling

Figure: Sampling systems. (a) Sample system for an autoclavable bioreactor. (b) Sample system
for an autoclavable bioreactor.
1, Connector; 2, 0.22 µm filter; 3, O-ring; 4, grub screw; 5, sampling system insert; 6, bottle; 7,
sample pipe lid; 8, clamp; 9, tubing.

28
On-Line Sensing
Real-time sensing of bioreactor conditions, involving spatially resolved measurements of fluid
velocities and reaction components, is essential for both the experimental validation of bioreactor
models and for monitoring of ongoing performance. Even the most perfectly-designed and
thoroughly-modeled bioreactor is expected to experience unforeseen conditions occasionally,
particularly given the presence of mutable microorganisms, with the result that real-time or on-
line sensing of bioreactor conditions during operation is essential. On-line sensing is also
important, of course, in validating models during development. Especially important is the
potential of on-line sensing to allow precise, automated, feedback control devices to maintain
reactor homeostasis.

Conventional reactor sensors are often not ideal for bioprocess measurements, however,
due to for example:
• Their vulnerabilities to interference by biofilm growth,
• Inabilities to resolve overlapping signals generated in complex culture media, and,
• Inabilities to be sterilised to avoid contaminating culture media; hence, ongoing research
into sensor design is important to the future of bioprocessing.

At the same time, sufficient progress has been made in several key areas that a number of
important variables may be monitored adequately. Dissolved oxygen and pH, for example, are
frequently monitored on-line using electrochemical sensors contained within steam-sterilisable
glass electrodes. Dissolved oxygen may also be measured by a recently-commercialised method
based on fluorescence quenching. On-line measurements of cell mass are also desirable and
can now be made indirectly by electrical or optical means: capacitance and permittivity
measurements, for example, provide electrical quantitation, while light absorbance, light
scattering, or a combination of the two provides optical indications of cell mass.

Carbon dioxide, a waste product of cell respiration, is an important indicator of bioreactor status
and can be measured on-line by means of sterilisable electrode sensors, optical sensors, and
sensors based on gas permeation through selective polymer membranes, serving as the basis
for process control loops.

Several spectroscopic techniques are also available to quantify the presence of numerous
organic and inorganic species simultaneously, due to recent advances in optics and computing.

An example is shown in Figure 3, in which a noninvasive sensor has measured the whole-cell
biotransformation of L-serine and indole to tryptophan. This approach is also applicable to the
processing of sugar beet molasses at an industrial scale. Glucose, fructose, glutamine,
ammonia, CO2, and phosphate are among the many compounds that can be measured by either
near or mid-infrared spectroscopy. Recent developments of improved, low-cost optical sensors
are also promising. Further development for miniaturisation, improved robustness, and sensitivity
are expected.

A separate approach is based on the attachment of the gene for green fluorescent protein onto
a protein of interest present during manufacture. This is particularly attractive if the tagged
molecule is the product of interest, such that product concentration can be measured directly. A
number of other colour probes are also currently available, opening a promising avenue for real-
time monitoring of the expression of multiple genes simultaneously.

29
Improved real-time sensing of bioreactor conditions is essential to model validation and process
control during operations, and future bioreactors are expected to be massively instrumented to
provide detailed real-time information of vital interest. Advances in sensing technologies are
urgently needed, especially in the design of bioprocesses for commodity products where
efficiency is paramount, and this area should be considered a top priority within bioengineering
for pollution prevention.

5. ENVIRONMENTAL CONTROL
Temperature, pH and dissolved oxygen (or DO2) measurement and control are similar in batch
and continuous operation. Dissolved oxygen measurement may present problems that are not
usually encountered in short fermentations.
Over the course of time, a biofilm may begin to cover the membrane of some types of
polarographic probes. Microorganisms adhere to the steel screen protecting the probe’s
membrane. A partial solution to the problem is to cover the screen with 2.5 x 10-3 mm Teflon film.
The response of the probe is greatly reduced, however.
In small vessels there may not be room in the head-plate for a commercially available
polarographic probe, necessitation the use of a galvanic probe. If the galvanic probe is used, it
must be calibrated with nitrogen and air after autoclaving. In contrast, polarographic probes have
excellent stability after autoclaving, and electronic zero feature can be used to zero them.

6. AIR INLET AND EXIT

An example of an air supply and exhaust system is shown in Fig.3. Humidification of the air
prevents evapouration of medium from the vessel and improves the estimation of dilution rate.
This particularly important, when operating at temperatures of 30°C or more, when using
compressed air which is very dry, or when operating at dilution rates of less than 0.1 h-1.

Sterilisation of the inlet air can be provided by an absolute membrane filter or a packed glass-
wool filter. Two air filters connected in a tee are recommended as a precaution, especially when
using humidifiers. If one filter becomes plugged, the other filter can be unclamped and used
immediately. A regulated air supply is important for safety purposes. If the air exit becomes
plugged, the air pressure should be sufficiently low that the vessel does not explode. Wrapping
the vessel with tape is an added precaution to prevent flying glass.

The exhaust gas should exit to the overflow reservoir. In this way foam-out can be contained. The
overflow reservoir is then connected to a graduated cylinder filled with 0.5% bleach. The exit air
is sparged through the liquid and is vented through a sterilising filter. This additional piece of
equipment serves several functions.
(i) First, it provides a constant back-pressure on the vessel which can be changed
by changing the liquid level in the cylinder. This back-pressure is necessary if
an exit gas sample is to be sent to an analyser.

(ii) It also facilitates sampling by providing enough pressure to force the sample
out of the vessel. One also has an instantaneous visual verification of airflow
through the vessel. Absence of bubbling may indicate loss of containment, a
particular concern if pathogenic or recombinant organisms are being cultured.

30
 (iii) The use of bleach reduces the potential of viable microorganisms escaping to
the environment. This includes the microbe under investigation as well as those
which may potentially invade the vessel: bacteriophage, spore-formers etc.

(iv) Finally, it acts as a scrubber to reduce unpleasant odours in the laboratory.

Table: Sensors Used in Fermentor Monitoring and Control

OPERATION
Successful design and operation of bioreactors requires optimisation of numerous quantities,
including such conditions as pH, substrate and product concentrations, cell density, oxygen
concentration, and temperature. Understanding the roles of each of these in a particular
bioprocess, as well as predicting and controlling their spatial and temporal variation within narrow
limits throughout the bioreactor volume, is of central importance to bioreactor design.

Consequently, primary goals of current engineering efforts include accurate modeling of

bioprocesses, real-time sensing of internal reactor conditions, and the design of bioreactors in
which conditions can be maintained as nearly uniform as possible.

31
 1. EQUIPMENT STERILISATION
Proper sterilisation of culture vessels and nutrient reservoirs in excess of 1 litre requires more
time than one might expect. The Table gives some helpful data in this regard, although the time
may depend on the autoclave used and the position of the items in the autoclave.

Table: Effect of Medium Volume on Sterilisation Time

Liquid Vol. (litres) Number of Initial Liquid Time (min) for
per Container Container per load Temperature (°C) Liquid to reach
121°C
2.0 10 27 37
4.0 5 26 52
6.0 4 26 62
10.0 2 25 105

Also, sterilisation of medium containing insoluble components requires more time than that of
synthetic medium, due to the suspended solids.
It should be emphasised that glass bottles are dangerous when hot; a cool draft can easily break
a hot bottle. Sterilisable polycarbonate and polypropylene carboys are much safer but may
release low levels of plasticizer and monomer into the medium.

32
 2. START-UP
The chemostat is usually started as batch fermentation. When the culture is in the exponential
growth phase, the nutrient feed and level controller are turned on. If the start-up medium has the
same composition as the feed medium, the transition from batch to continuous fermentation may
be accompanied by an oscillation in certain fermentation parameters such as cell density or the
concentration of a secondary metabolite. If the substrate is toxic to the cells, the oscillation may
become so severe that washout occurs.

This problem occurs because the exponentially growing culture abruptly becomes substrate
limited. The metabolism of the organism must shift to accommodate the nutritional limitation, and
oscillations ensue. An excellent discussion of the transition from batch to chemostat operation is
provided by Dunn and Mor (14).

A smooth transition from batch to continuous fermentation can be obtained by using half-
strength medium for the initial batch growth. The nutrient feed is started when the cell
concentration is about one-half of that expected at steady state, and the culture becomes
a nutrient-limited fed batch.
The specific growth rate declines as the cell concentration increases until the steady-state
concentration is achieved. At this time the level controller is turned on and the continuous
fermentation begins.

3. DETERMINATION OF STEADY STATE

The assumption of steady state is fundamental to the development of the kinetics of continuous
culture. Steady state is achieved when there is no net change in the contents of the
bioreactor: the biomass, substrate, and product concentrations are unchanging, and the
metabolic state of the organism is constant,
Theoretically, a bioreactor will reach 95% of steady state after three residence times. A
residence time is the time it takes for one reactor volume to pass through the vessel and is equal
to the reciprocal of the dilution rate. Although cell concentration generally follows these kinetics,
the biological processes which accompany a cell growth may not reach steady state so quickly
(15). Thus, using cell density as a measure of steady state can often be misleading. The best
indication of steady state is a constant value of the cell product being studied for a period
of three residence times. Mutation and selection within a chemostat can affect the attainment
of steady state.

4. MEDIUM FORMULATION AND OPTIMISATION

Before operating a nutrient-limited fermentation it is useful to approximate the yield of cells on
substrate. For many fermentations, a chemically defined medium is available. Cell density and
substrate concentration are easily determined. However, for some fermentations, most notably
antibiotics, insoluble complex nutrients are used. These present the investigator with potential
problems.

How is biomass to be measured, if at all? Assuming the techniques of biomass

estimation are translated from batch fermentation to continuous fermentation, one still has the
problem of insoluble nutrients and their slow release.

Phosphates, amino acids, sulfates, and carbohydrates can all be slowly librated by enzymatic or
chemical hydrolysis of complex medium components such as soy flour, cotton seed meal etc.
Particle size and residence time in the fermentor then become additional process variables. An
alternative that can eliminate this problem is to formulate a synthetic medium.

33
The choice of a growth-limiting nutrient depends on a basic understanding of the processes one
wishes to study. In cases where the overproduction of a particular biomolecule is desired a rational
approach to the selection of a nutrient limitation may be possible.

Nutrient limitation can cause physiological and morphological changes which can be
exploited. The most direct of these changes occur in fermentations where nutrient repress the
synthesis of certain enzymes.
• Many antibiotics are produced poorly in media containing excess glucose or
phosphate.
• In some cases protease production is reduced in the presence of excess nitrogen
or sulfur.
• The production of eukaryotic proteins by genetically engineered strains of
Escherichia coli is frequently designed to be repressed by tryptophan (2). In such
systems a nutritional limitation of the repressing compound can be advantageous.
Nutrient limitation can sometimes be used to uncouple growth from other metabolic
pathways (3).
• For example, nitrogen limitation can cause excess carbon to be redirected from cell
mass into the overproduction of biopolymers such as xanthan gums and
polyhydroxybutyrate.
• Simple oxidation of products of glucose, such as gluconic acid and citric acid, can
also be overproduced when a nutrient is limiting.

Finally, nutritional deficiencies can alter the structure of microorganisms. Phosphate and
magnesium limitations are known to alter cell membranes in some bacteria to such an extent that
permeability characteristics are changed. Such microorganisms may become more susceptible
to antibiotics, and some lose pools of intracellular metabolites to the surrounding medium. The
limitation of biotin in biotin auxotrophs of Corynebacterium glutamicum has been shown to cause
leakage of glutamic acid by changing membrane permeability.

The yield of cells on substrate, YX/S, must be determined to calculate the concentration of
limiting substrate to be used in the feed medium. The yield can be approximated in shaker-
flask culture if cell lysis and sporulation do not occur. One should be aware that growth yield
derived from batch experiments can only give a rough approximation (4).

Several flasks are prepared with a medium containing all essential nutrients except
the limiting nutrient. This substrate is then added back to several flasks in
incremental concentrations.
An inoculum is centrifuged and washed twice in the basal medium to prevent carry-
over of the substrate. The flasks are the inoculated, and cell mass is determined
after growth ceases. A linear relationship between cell density and substrate added
will be obtained up to certain substrate concentration. After this point other factors
begin to limit growth such as pH and dissolved oxygen.
A linear regression of cell mass against substrate concentration in the linear region
gives and approximation of yield which can be used to formulate the chemostat
medium. A sub-inhibitory excess of all other nutrients should be provided. A five-
fold excess is recommended as a starting point.

34
 1. FERMENTATION SUBSTRATES
Many materials are used as food for growing microorganisms. These are referred to as the
substrate. The substrate and the trace materials needed, together with chemicals added to make
the fermentation easier (such as anti-foam agents to stop froth forming) make the culture medium.
The substrates can be divided according to the different essentials for life that they provide:
Usually carbon, nitrogen, and (in the case of aerobic fermentation) oxygen. Usually carbon
substrates cost the most because they needed the most.

A. Carbon Source
A carbon source is necessary to provide the cell with energy as well as the material with which to
grow and synthesise a range of primary and secondary metabolites. The best energy source
depends on the type of organism utilised, e.g., autotroph, chemotroph, etc.
There is obviously a wide range of carbon sources and the one chosen should be appropriate to
the organism but also to the economics of the process. At research scale the latter tends to be
less important, but it should be borne in mind if the objective of the programme is to develop an
industrially relevant process. This section will look at several carbon sources and indicate the
advantages and disadvantages of each. A typical microorganism is approximately 50% carbon
(Table), making carbon the most significant substrate and care should be taken to ensure that the
carbon concentration does not become limiting. The yield of biomass on carbon is approximately
0.5 which means that if a biomass concentration of 50 gL−1is required, 100 gL−1of the carbon
source must be supplied, albeit not necessarily all at once if fed-batch culture is being utilised.

The great majority of laboratory and industrial fermentations tend to use an extremely limited
range of easily utilisable substrates to supply the energy and C requirements of cultures.
This does not imply that other C and energy sources could not be used, just that this limited range
is generally available and methods for preparing and analysing the consumption rates of such
substrates are well understood. The production of pharmaceuticals, especially
biopharmaceuticals by fermentation, takes place in the context of an industry that is, except for
the nuclear industry, perhaps the most regulated in the world. This fact accounts for a degree of
‘conservatism’ in terms of nutrient supply.

Among the common carbon substrates for large-scale fermentation are:

(i) Glucose
Glucose is universally acceptable for growth of most cell lines, be they animal, plant or microbial.
Supplied as a powder in the pure form, the substrate is readily available, reliable, easily stored,
easily handled, and has no significant implications for health and safety.
These qualities make glucose a popular choice of carbon source.

There are some drawbacks to using glucose, notably:

(i) The possibility of the organism suffering from the ‘Crabtree effect’ if glucose is over
supplied in the initial stages of growth;
(ii) The loss of available substrate to the Maillard reaction, which can occur if glucose is
sterilised with a nitrogen source present.

Both situations can be overcome readily, by careful monitoring and control of glucose feed in the
first case and by either sterilising the substrates separately or by using filtration sterilisation
methods in the second.

35
A popular method of supplying glucose on a larger scale is to use glucose syrups, manufactured
by the hydrolysis of starch (a substrate which itself tends to be insoluble and consequently
unavailable to the cells, difficult to handle, but is readily hydrolysed to glucose). The drawbacks
remain the same, with the additional complications of transportation, storage and handling of a
liquid. Formulation of the medium and delivery to the bioreactor may also be considerations that
have to be taken into account. The main advantage to using glucose syrups is cost, as it is
significantly cheaper than powdered glucose.

(iii) Sucrose
Often the sugar of choice in research laboratories, sucrose, is utilised by many but not all cell
lines. Commercially available sucrose comes in many forms and grades, from pure granulated
forms to complex molasses solutions (above). Sucrose does not tend to suffer from the same
drawbacks as glucose and, although it can be subject to caramelisation when over sterilised,
generally it can be autoclaved/sterilised with nitrogen compounds without the same problems as
glucose. If using molasses, sufficient cold storage must be available to handle the needs of the
laboratory as the range of different substrate available means that the molasses can be readily
contaminated, and although sterilisation can destroy any contaminating organisms, the amount
of available sucrose to the organism of choice will decrease. Again, mixing of molasses from
several batches can help iron out some variability contributed by this source.

Molasses is a side product of sugar refining that contains most of the materials from sugar beet
or sugar cane that is not sugar; molasses is one of the cheapest substrates available. There are
several varieties with different properties, components, and costs:
• Beet molasses: the molasses left from processing beet.
• Black strap molasses: from an early stage in cane sugar processing.
• Cane refinery molasses: from a later (more sugar rich) stage in processing
• High Test Molasses: actually, concentrated cane juice treated with

(iv) Lactose
Lactose can only be utlilised by a few cell types, e.g., Escherichia coli, and is usually only
metabolised very slowly. The sugar is only useful for some commercial processes that use the
complex substrate whey, a by-product of the dairy industry, which contains both lactose
(approximately 50 g lactose per litre of whey and 4% w/w nitrogen). Available as a liquid or as
whey powder there are choices available regarding storage and handling. Some fermentation
media in the past typically batched in both glucose, to achieve rapid culture growth, and lactose,
which was only utilised after glucose exhaustion, and the slow rate of usage fueled, e.g., antibiotic
synthesis. Controlled feeding of glucose usually obviates the need for this nowadays.

Other Sugars
Other sugars can be used as substrates but tend to be extremely specific to the cell line chosen,
e.g., melobiose is sometimes used for the cultivation of yeast cells. Sugars falling into this
category tend to be too expensive to use on a routine basis. If choosing an unusual sugar, the
questions to be posed are ‘Why is this sugar required?’ and ‘What advantage will the organism
gain?’

(v) Malt Extract. Made from malted barley by soaking malt in water.

(vi) Starch and Dextrins. Polysaccharides often made from cheap crops like potatoes.

36
 (vii) Cellulose. The world produces about 100 billion tonnes of cellulose a year, so it is a
potential raw material for large-scale fermentation. But only a few organisms can degrade
cellulose.

(viii) Sulfite Liquor. A by-product of paper pulp production, which contains much of
the fermentable sugars from wood without substantial cellulose.

(ix) Whey. A side product of dairy processes, whey is cheap but expensive to store or
transport.

(x) Methanol. A cheap chemical form the oil industry, but it contains no nitrogen. Only a
restricted range of organisms can grow on methanol. Similarly, ethanol (‘alcohol’) can be
used, but more usually ethanol is the product of fermentation.

(xi) Oil, Gas. Some organisms can use natural gas or some components of crude or refined
oil as carbon substrates. However, their commercial use depends critically on the price of
oil.

B. Nitrogen Sources
The amount of nitrogen in any medium really determines the amount of biomass that will be
achieved for the particular cell line, given that plenty of carbon is available and all required
nutrients are present in the initial medium. Nitrogen is required for growth and synthesis of, for
instance, proteins and nucleic acids. The type and sources are variable.

Points to consider before selecting the nitrogen source are as follows:

(i) What do we want the process to do: manufacture cells (i.e., biomass), intracellular
or extracellular product, or both? In many fermentation processes growth alone is not
the target, the synthesis of a specific product is required.

(ii) Do we want the nitrogen to run out at any point in the process? For example, if dealing
with a true secondary metabolite, such as an antibiotic, it is necessary to design the medium so
that one substrate becomes limiting in order that growth ceases or slows down considerably to
allow production of the product. This is often achieved by allowing nitrogen to become limiting
once sufficient biomass has been achieved; this cannot be the case if a protein is the required
product for obvious reasons.

(ii) How will pH control be achieved? In a well-controlled bioreactor this is not an issue,
but if a buffered system is being employed it is worthwhile considering what contributions salts in
the medium may make to the overall pH of the broth. In some instances, the culture may be fed
N source in the form of NH3 or NH4OH at a rate just matching its metabolic activity. This also
achieves pH control in acid-forming fermentations.

A traditional rule-of-thumb guide to a balanced growth medium suggests an approximate ratio of

nutrients required of carbon: nitrogen: phosphorus of 100: 10: 1.

So, for every 10 gL−1 carbon source, 1 gL−1 nitrogen is required. A common error here is to assume
that 1 gL−1 (NH4)2SO4 contains 1 gL−1 N, which it obviously does not, so remember to consider
the other elements that may be in the chosen nitrogen source.

37
Likewise, this is important when assessing the effects of different N providing salts: make sure
the absolute amounts of N being provided are equal otherwise what is observed will be a function
of N availability not necessarily the identity of the N source.

(i) Ammonia. This a very smelly gas produced as a bulk commodity for the chemical
industry. Most organisms can use ammonia. Sometimes it is converted into ammonium salts or
into urea for ease of handling. Often used in industrial processes, ammonia is not a common
source of nitrogen used in research laboratories because of the volatility of the liquid and the
associated handling problems. Health and safety issues are a real risk with this nitrogen source,
and considerable thought is needed if this is to be the substrate of choice. Special handling
equipment, storage reservoirs and breathing equipment are all required, as are additional safety
precautions should there be a leak.
The advantages of using ammonia are that it readily dissolves in the medium, it is immediately
available to the cell, and it is cheap (although storage and handling equipment may not be).

Nitrogen-Based Salts
Ammonium sulfate, (NH4)2SO4, and ammonium chloride, NH4Cl, are common nitrogen sources
found in the literature. Ammonium is very utilisable by many organisms and use of a salt is a
cheap and convenient method of supplying the cell with what it requires. In addition, it is easy to
calculate the exact amount of nitrogen being supplied to the cells.

Storage of the substrate is easy, requires no special storage areas or equipment, the salts just
needing to be stored in a cool dry place. Analysis of the nitrogen content of the resultant
fermentation broth is usually quite simple and can be achieved using a simple assay and an
ammonia probe.
It is useful to consider other requirements the organism may have, for example if the cell has a
requirement for sulfur, then ammonium sulfate would be the salt to choose, so that one salt
satisfies more than one requirement. This is not only economical it also makes formulation of the
medium much simpler. As a rule, try to keep the medium formula as simple as possible to
minimise the chance of errors and to reduce medium generated variability.

A word of warning when dealing with any salts required in bioprocess media is to watch the
combination of salts used, and the order in which they are added. An incorrect combination can
lead to a batching vessel full of salt crystals. Be systematic and make sure the order of addition
of ingredients is consistently adhered to.

Complex Nitrogen Sources

There is a number of excellent complex nitrogen sources available, e.g., yeast extract, soya bean
meal (8% w/w nitrogen), and corn steep liquor (4% w/w). All supply a significant amount of
nitrogen, in addition to other essential nutrients such as carbon, vitamins, and minerals.

As with the complex carbon sources, there is significant batch-to-batch variability and every batch
of the substrate has to be analysed for component composition before use.
Analysis of the fermentation broth is also more complicated and often involves acid hydrolysis of
the broth in order to get the nitrogen into a readily assayable form. This involves considerable
time, labour and extra cost.

(ii) Corn Steep Liquor. This is the liquid generated in the early stages of wet-milling maize
to produce starch. Maize grains are submerged in water to soften them before milling. Low
molecular weights sugars and peptides accumulate in the water.

38
(iii) Soy Proteins. This is the protein left after the extraction of oil out of soybeans.

(iv) Yeast Extracts. Made from waste yeast from industrial fermentations and contains
everything necessary for microbial growth.

(v) Peptones, Casein Hydrolysates. These are partially digested meat or milk proteins,
respectively. The proteins used are usually waste material from the food industry: nevertheless,
this can still be very expensive source of nitrogen.

C. Precursors
Some fermentations must be supplemented with specific precursors, notably for secondary
metabolite production. When required, they are often added in controlled quantities and in a
relatively pure form. Examples include phenylacetic acid or phenylacetamide added as side chain
precursors in penicillin production. D-threonine is used as a precursor in l-isoleucine production
by Serratia marsescens, and anthranillic acid additions are made to fermentations of the yeast
Hansenula anomala during l-tryptophan production.

D. Inducers and Elicitors

If product formation is dependent upon the presence of a specific inducer compound or a
structural analogue, it must be incorporated into the culture medium or added at a specific point
during the fermentation. In plant cell culture the production of secondary metabolites, such as
flavonoids and terpenoids, can be triggered by adding elicitors. These may be isolated from
various microorganisms, particularly plant pathogens.

Inducers are often necessary in fermentations of genetically modified microorganisms (GMMs).

This is because the growth of GMMs can be impaired when the cloned genes are ‘switched on’,
due to the very high levels of their transcription and translation. Consequently, inducible systems
for the cloned genes are incorporated that allow initial maximization of growth to establish high
biomass density, whereupon the cloned gene can then be ‘switched on’ by the addition of the
specific chemical inducer.

E. Inhibitors
Inhibitors are used to redirect metabolism towards the target product and reduce formation of
other metabolic intermediates; others halt a pathway at a certain point to prevent further
metabolism of the target product. An example of an inhibitor specifically employed to redirect
metabolism is sodium bisulphite, which is used in the production of glycerol by S. cerevisiae.

Some GMMs contain plasmids bearing an antibiotic resistance gene, as well as the heterologous
gene(s). The incorporation of this antibiotic into the medium used for the production of the
heterologous product selectively inhibits any plasmid-free cells that may arise.

F. Cell Permeability Modifiers

These compounds increase cell permeability by modifying cell walls and/or membranes,
promoting the release of intracellular products into the fermentation medium. Compounds used

39
for this purpose include penicillins and surfactants. They are frequently added to amino acid
fermentations, including processes for producing L-glutamic acid using members of the genera
Corynebacterium and Brevibacterium.

G. Oxygen
Depending on the amount of oxygen required by the organism, it may be supplied in the form of
air containing about 21% (v/v) oxygen, or occasionally as pure oxygen when requirements are
particularly high. The organism’s oxygen requirements may vary widely depending upon the
carbon source. For most fermentations, the air or oxygen supply is filter sterilized prior to being
injected into the fermenter.

H. Antifoams
Antifoams are necessary to reduce foam formation during fermentation. Foaming is largely due
to media proteins that become attached to the air-broth interface where they denature to form a
stable foam. If uncontrolled the foam may block air filters, resulting in the loss of aseptic
conditions; the fermenter becomes contaminated and microorganisms are released into the
environment. Of possibly most importance is the need to allow ‘freeboard’ in fermenters to
provide space for the foam generated. If foaming is minimised, then throughputs can be
increased.

There are three possible approaches to controlling foam production: modification of medium
composition, use of mechanical foam breakers and addition of chemical antifoams. Chemical
antifoams are surface-active agents which reduce the surface tension that binds the foam
together.

The ideal antifoam should have the following properties:

1. Readily and rapidly dispersed with rapid action;
2. High activity at low concentrations;
3. Prolonged action;
4. Non-toxic to fermentation microorganisms, humans, or animals;
5. Low cost;
6. Thermostability; and
7. Compatibility with other media components and the process, i.e., having no effect
on oxygen transfer rates or downstream processing operations.

Natural antifoams include plant oils (e.g., from soya, sunflower and rapeseed), deodorized fish
oil, mineral oils and tallow. The synthetic antifoams are mostly silicon oils, poly alcohols and
alkylated glycols. Some of these may adversely affect downstream processing steps, especially
membrane filtration.

Animal Cell Culture Media

Animal cell culture media are normally based on complex basal media, such as Eagle’s cell culture
medium, which contains glucose, mineral salts, vitamins, and amino acids. For mammalian cells
a serum is usually added, such as fetal calf serum, calf serum, newborn calf serum or horse
serum. Sera provide a source of essential growth factors, including initiation and attachment
factors, and binding proteins. They also supply hormones, trace elements and protease inhibitors.

40
The highly complex composition of sera makes substitution with lower cost ingredients very
difficult. Sterilization of formulated animal culture media and media constituents is also more
problematic as many components are thermolabile, requiring filter sterilization. Normally, sera
constitute 5 to10% (v/v) of the medium, but attempts have been made to reduce and ultimately
eliminate its use. This is necessary due to its high cost and the fact that it is a potential source of
prions and viruses. In some circumstances levels have now been lowered to 1 to 2% (v/v) and
some cell lines have been developed that grow in serum-free media.

Plant Cell Culture Media

In contrast to most animal cell culture media, those used for plant cell culture are usually
chemically defined. They contain an organic carbon source (as most plant cells are grown
heterotrophically), a nitrogen source, mineral salts, and growth hormones. Sucrose is frequently
incorporated as the carbon source, particularly for secondary metabolite production, but glucose,
fructose, maltose and even lactose have been used. Nitrate is the usual nitrogen source, often
supplemented with ammonium salts. However, some species may require organic nitrogen,
normally in the form of amino acids. The combination and concentration of plant hormones
provided depend upon the specific fermentation.

Auxins are usually supplied, along with cytokinins to promote cell division. A two-phase culture
has often proved to be useful in increasing productivity, particularly for producing secondary
metabolites such as shikonin. The first phase uses a medium optimized for growth, the second
promotes product formation.

Culture Maintenance Media

These media are used for the storage and subculturing of key industrial strains. They are designed
to retain good cell viability and minimize the possible development of genetic variation. In
particular, they must reduce the production of toxic metabolites that can have strain destabilizing
effects. If strains are naturally unstable, they should be maintained on media selective for the
specific characteristic that must be retained.

Criteria for the Choice of Raw Materials Used in Industrial Media

In deciding the raw materials to be used in the production of given products using designated
microorganism(s) the following factors should be taken into account:

(A) Cost of the Material

The cheaper the raw materials the more competitive the selling price of the final product will be.
No matter, therefore, how suitable a nutrient raw materials is, it will not usually be employed in an
industrial process if its cost is so high that the selling price of the final product is not economic.
Thus, although lactose is more suitable than glucose in some processes (e.g. penicillin
production) because of the slow rate of its utilisation, it is usually replaced by the cheaper glucose.
When used, glucose is added only in small quantities intermittently in order to decelerate acid
production. Due to these economic considerations the raw materials used in many industrial
media are usually waste products from other processes. Corn steep liquor and molasses are, for
example, waste products from the starch and sugar industries, respectively.

41
(B) Ready Availability of the Raw Material
The raw material must be readily available in order not to halt production. If it is seasonal or
imported, then it must be possible to store it for a reasonable period. Many industrial
establishments keep large stocks of their raw materials for this purpose. Large stocks help beat
the ever-rising cost of raw materials; nevertheless large stocks mean that money which could
have found use elsewhere is spent in constructing large warehouses or storage depots and in
ensuring that the raw materials are not attacked during storage by microorganisms, rodents,
insects, etc. There is also the important implication, which is not always easy to realise, that the
material being used must be capable of long-term storage without concomitant deterioration in
quality.

(C) Transportation Costs

Proximity of the user-industry to the site of production of the raw materials is a factor of great
importance, because the cost of the raw materials and of the finished material and hence its
competitiveness on the market can all be affected by the transportation costs. The closer the
source of the raw material to the point of use the more suitable it is for use, if all other conditions
are satisfactory.

(D) Ease of Disposal of Wastes Resulting from the Raw Materials

The disposal of industrial waste is rigidly controlled in many countries. Waste materials often find
use as raw materials for other industries. Thus, spent grains from breweries can be used as
animal feed. But in some cases, no further use may be found for the waste from an industry. Its
disposal especially where government regulatory intervention is rigid could be expensive. When
choosing a raw material therefore the cost, if any, of treating its waste must be considered.

(E) Uniformity in the Quality of the Raw Material and Ease of Standardisation
The quality of the raw material in terms of its composition must be reasonably constant to ensure
uniformity of quality in the final product and the satisfaction of the customer and his/her
expectations. In cases where producers are plentiful, they usually compete to ensure the
maintenance of the constant quality requirement demanded by the user. Thus, in the beer industry
information is available on the quality of the barley malt before it is purchased. This is because a
large number of barley malt producers exist, and the producers attempt to meet the special needs
of the brewery industry, their main customer.
On the other hand, molasses, which is a major source of nutrient for industrial microorganisms,
is a byproduct of the sugar industry, where it is regarded as a waste product. The sugar industry
is not as concerned with the constancy of the quality of molasses, as it is with that of sugar. Each
batch of molasses must therefore be chemically analysed before being used in a fermentation
industry to ascertain how much of the various nutrients must be added.
A raw material with extremes of variability in quality is clearly undesirable as extra costs are
needed, not only for the analysis of the raw material, but for the nutrients which may need to be
added to attain the usual and expected quality in the medium.

(F) Adequate Chemical Composition of Medium

As has been discussed already, the medium must have adequate amounts of carbon, nitrogen,
minerals and vitamins in the appropriate quantities and proportions necessary for the optimum
production of the commodity in question. The demands of the microorganisms must also be met
in terms of the compounds they can utilise. Thus, most yeasts utilise hexose sugars, whereas
only a few will utilise lactose; cellulose is not easily attacked and is utilised only by a limited
number of organisms. Some organisms grow better in one or the other substrate. Fungi will for

42
instance readily grow in corn steep liquor while actinomycetes will grow more readily on soya
bean cake.

(G) Presence of Relevant Precursors

The raw material must contain the precursors necessary for the synthesis of the finished product.
Precursors often stimulate production of secondary metabolites either by increasing the amount
of a limiting metabolite, by inducing a biosynthetic enzyme or both. These are usually amino acids,
but other small molecules also function as inducers.
The nature of the finished product in many cases depends to some extent on the components of
the medium. Thus, dark beers such as stout are produced by caramelized (or over-roasted) barley
malt which introduces the dark color into these beers.
Similarly, for penicillin G to be produced the medium must contain a phenyl compound. Corn
steep liquor which is the standard component of the penicillin medium contains phenyl precursors
needed for penicillin G. Other precursors are cobalt in media for Vitamin B12 production and
chlorine for the chlorine containing antibiotics, chlortetracycline, and griseofulvin.

(H) Satisfaction of Growth and Production Requirements of the Microorganisms

Many industrial organisms have two phases of growth in batch cultivation: the phase of growth,
or the trophophase, and the phase of production, or the idiophase. In the first phase cell
multiplication takes place rapidly, with little or no production of the desired material. It is in the
second phase that production of the material takes place, usually with no cell multiplication and
following the elaboration of new enzymes. Often these two phases require different nutrients or
different proportions of the same nutrients. The medium must be complete and be able to cater
for these requirements. For example, high levels of glucose and phosphate inhibit the onset of
the idiophase in the production of a number of secondary metabolites of industrial importance.
The levels of the components added must be such that they do not adversely affect production.

CLEANING-IN-PLACE
Cleaning-in-place is the cleaning (and sterilisation) of a manufacturing system without dismantling
the system, so that the parts are cleaned as a whole: it is also called “in situ sterilisation” or
sterilisation-in-place (CIP and SIP). This is a much easier operation to perform than cleaning
and sterilising all the components separately and then trying to assemble them under sterile
conditions or having separate cleaning and sterilising operations. However, it requires some
specialist techniques and equipment.
In particular, bioreactors machinery must be designed so that there are no ‘dead legs’ (i.e., areas
that are blocked at one end), crevices, or ‘shadowed’ areas (i.e., areas where the bulk of some
other piece of the apparatus prevents fluid from flowing) into which cleaning fluid could not flow.
It is also useful for the equipment to be designed so that bits of it can be cleaned while the rest
remains in operation.
CIP systems are a combination of the equipment used and the processed used in it. Thus, a full
CIP system will include protocols for washing, usually with industrial strength detergents, and
rinsing out the wash agents.

43
 EXPERIMENTAL DESIGN
When designing an experimental program in which continuous fermentation is to be used, one
must be aware of the selective pressure in which the culture is subjected.

Spontaneous mutations arising at a rate of about 1 in 106 cell divisions may sooner or later
generate an organism which has a competitive advantage over the original strain in the vessel. In
some cases, this selection may be desired. However, if the investigator is interested in studying
a homogeneous cell population a subpopulation of mutants can invalidate the data.
Experiments should be designed to separate the metabolic response of the culture to
experimental manipulation from genetic changes which may occur in response to the same
manipulations.

One way to detect changes in the genetic makeup of the culture is to follow the productivity
of the fermentation in satellite shake flasks. Samples are taken periodically from the
continuous culture and re-inoculated into shake flasks. These parallel batch fermentations can be
assayed for growth rate and product formation. Increasing growth rate or decreasing product
formation in shake flasks over the course of the fermentation may indicate a changing cell
population.
This approach suffers from the fact that whole broth is used as an inoculum. A sub-population of
mutants may not be seen until it becomes a significant portion of the culture.

An even more rigorous approach involves dilution plating of samples from the culture as
the fermentation progresses. Individual colonies can then be re-inoculated into shake-flask
cultures. In this way a statistically significant number of clones can be evaluated and small
changes in the population can be identified.

Transient Phenomena
Cellular adaptation is the equilibration of the myriad reactions a cell embodies to a new
environment. The changes of metabolic activity required to establish the new equilibrium are
known as transient phenomena.
The study of transients in chemostat culture led to successful strategies for on-line process
control, and greater understanding of the regulation of growth (4), and product formation (5,6,7).

DOWNSTREAM PROCESSING
Industrial fermentations comprise both upstream processing (USP) and downstream
processing (DSP) stages (Figure below). USP involves all factors and processes leading to, and
including, the fermentation, and consists of three main areas. The first relates to aspects
associated with the producer microorganism. They include the strategy for initially obtaining a
suitable microorganism, industrial strain improvement to enhance productivity and yield,
maintenance of strain purity, preparation of a suitable inoculum and the continuing development
of selected strains to increase the economic efficiency of the process. The second aspect of USP
involves fermentation media, especially the selection of suitable cost-effective carbon and energy
sources, along with other essential nutrients. This media optimisation is a vital aspect of process
development to ensure maximisation of yield and profit. The third component of USP relates to
the fermentation, which is usually performed under rigorously controlled conditions, developed to
optimise the growth of the organism or the production of a target microbial product.

44
DSP encompasses all processes following the fermentation. It has the primary aim of efficiently,
reproducibly, and safely recovering the target product to the required specifications (biological
activity, purity, etc.), while maximising recovery yield and minimizing costs. The target product
may be recovered by processing the cells or the spent medium depending upon whether it is an
intracellular or extracellular product. The level of purity that must be achieved is usually
determined by the specific use of the product. Often, a product’s purity will be defined by what is
not present rather than what is. Purity of an enzyme, for example, is expressed as units of enzyme
activity per unit of total protein. Not only is it important to reduce losses of product mass, but in
many cases retention of the product’s biological activity is vitally important.

Each stage in the overall recovery procedure is strongly dependent on the protocol of the
preceding fermentation. Fermentation factors affecting DSP include the properties of
microorganisms, particularly morphology, flocculation characteristics, size and cell wall rigidity.
These factors have major influences on the filterability, sedimentation and homogenisation
efficiency. The presence of fermentation byproducts, media impurities and fermentation additives,
such as antifoams, may interfere with DSP steps and accompanying product analysis.

Consequently, a holistic approach is required when developing a new industrial purification

strategy. The whole process, both upstream and downstream factors, needs to be considered.
For example, the choice of fermentation substrate influences subsequent DSP. A cheap carbon
and energy source containing many impurities may provide initial cost savings but may
necessitate increased DSP costs. Hence overall cost savings may be achieved with a more
expensive but purer substrate. Also, adopting methods that use existing available equipment may
be more cost-effective than introducing more efficient techniques necessitating investment in new
facilities.

The physical and chemical properties of the product, along with its concentration and location,
are obviously key factors as they determine the initial separation steps and overall purification
strategy. It may be the whole cells themselves that are the target product or an intracellular
product, possibly located within an organelle or in the form of inclusion bodies. Alternatively, the
target product may have been secreted into the periplasmic space of the producer cells or the
fermentation medium.

Stability of the product also influences the requirement for any pretreatment necessary to prevent
product inactivation and/or degradation. DSP can be divided into a series of distinct unit
processes linked together to achieve product purification (Figure). Examples of the purification
strategies for two products are shown in the Figure.

Usually, the number of steps is kept to a minimum. This is not only because of cost, but because
even though individual steps may obtain high yields, the overall losses of multistage purification
processes may be prohibitive (Figure). The specific unit steps chosen will be influenced by the
economics of the process, the required purity of the product, the yield attainable at each step and
safety aspects. In many cases, integration of fermentation and DSP is now preferred. Integration
can often increase productivity, decrease the number of unit operations, and reduce both the
overall process time and costs.

Physical process integration may be achieved by placing separation units inside the fermenter or
by directly linking the two systems together. Where product formation is coupled with growth, i.e.,
primary metabolites, higher productivity may be achieved by using integrated systems that
maintain high cell densities through cell retention or recycling. Where products are inhibitory, a
wide range of methods have been employed to partition bioreactors to allow the rapid in situ

45
removal of products by extraction, adsorption, or stripping, often increasing yield and productivity.
Alternatively, the processing can be ex situ, where the product is removed outside the fermenter
and the processed medium is returned to the fermentation. Such processes have been particularly
successful for removing alcohols and solvents, but it is now also possible to extract some proteins.
This mode of product removal can often improve raw material conversion, particularly where there
are unfavourable thermodynamic equilibria, i.e., product removal ‘pulls’ the reaction in the
direction of further product formation. Early product extraction can also enhance the yield for those
products sensitive to prolonged exposure to the fermentation environment, where shear,
proteolytic enzymes, oxidation, etc., may be destructive.

46
Figure: An Outline of Upstream and Downstream Processing Operations.

47
Downstream processing is a general term for all the things that happen in a biotechnological
process after the biology, be it fermentation of a microorganism or growth of a plant. It is
particularly relevant to fermentation processes, which produce a large amount of a dilute mixture
of substrates, products, and microorganisms. These must be separated and the product
concentrated and purified and converted into a product that is useful.

Table: Examples of Bioprocessing Products and their Typical Concentrations

TYPES PRODUCTS CONCENTRATION
g l-1

Cell Itself Baker’s Yeast, Single Cell Protein (SCP) 30

Extracellular Alcohols, Organic Acids, Amino Acids 100

Extracellular Enzymes, Antibiotics 20

Intracellular Recombinant DNA Proteins 10

48
It is not enough to grow the required cells in a bioreactor: extraction and purification of the desired
end-product from the bioprocess is equally, if not more important, calling on the skills of chemists
and chemical engineers as well as those of bioscientists and process engineers. The design and
efficient operation of downstream processing operations are vital elements in getting the required
products into commercial use and should reflect the need not to lose more of the desired product
than is necessary.

The bioproducts considered here, primarily including materials either made by living organisms
or derived from biomass, typically require extraction from either a whole organism, such as a
plant; from aqueous bioreactor media (also referred to as fermentation broth or culture media), or
from downstream processing solutions. A prevalent challenge presented by bioreactor-based
processes, in particular, is the generation of products within relatively dilute aqueous solutions.
Bioreactor media usually must remain dilute, however, to prevent inhibition of enzyme activity by
accumulated products and to prevent cell mortality due to accumulated wastes. Nevertheless, the
design of bioreactors to allow higher solute concentrations while maintaining cell health and
activity is worthy of considerable effort.

Table: Downstream Processing Operations

Separation
• Filtration
• Centrifugation
• Flotation
• Disruption

Concentration
• Solubilisation
• Extraction
• Thermal processing
• Membrane filtration
• Precipitation

Purification
• Crystallisation
• Chromatography

Modification

Drying

49
Figure: Major Process Steps in Downstream Processing.

Figure: Examples of Unit Downstream Processing.

50
Downstream processing will be primarily concerned with initial separation of the
bioreactor broth into a liquid phase and a solids phase and subsequent concentration and
purification of the product. Processing will normally involve more than one stage. Methods in
use or proposed range from conventional to the arcane, including distillation, centrifuging,
filtration, ultrafiltration, solvent extraction, adsorption, selective membrane technology, reverse
osmosis, molecular sieves, electrophoresis, and affinity chromatography. It is in this area that
several potential industrial applications of modern biotechnology have come to grief either
because the extraction has defeated the ingenuity of the designers or, more probably, because
the extraction process has required so much energy input as to render it uneconomic.

Unique characteristics of bioseparation products includes:

• Products are in dilute concentration in an aqueous medium.
• Products are usually temperature sensitive.
• There is a variety of products to be separated.
• Products can be intracellular; often insoluble inclusion bodies.
• Physical and chemical properties of products are similar to contaminants.
• High purity and homogeneity may be needed for human health care use.

Final products of the downstream purification stages should have some degree of stability for
commercial distribution. Stability is best achieved for most products by using some form of drying.
In practice this is achieved by spray-drying, fluidised-bed drying or freeze-drying. The method of
choice is product and cost dependent. Products sold in the dry form include organic acids, amino
acids, antibiotics, polysaccharides, enzymes, single cell protein and many others. Many products
cannot be supplied easily in a dried form and must be sold in liquid preparations. Care must be
taken to avoid microbial contamination and deterioration and, when the product is proteinaceous,
to avoid denaturation.

Figure: The Effect of Yield in Multistage Purification.

51
There are three general steps to downstream processing:
• Separation
• Concentration
• Purification

1. Separation
The first step separates the crude product from the microbial mass and other solid lumps, the
second removes most of the water (and hence is often called dewatering), and the third takes the
concentrated product and purifies it. The order of events can be different, but generally falls into
this scheme.

The first step in the downstream processing of suspended cultures is a solid–liquid separation to
remove the cells from the spent medium. Each fraction can then undergo further processing,
depending on whether the product is intracellularly located, or has been secreted into the
periplasmic space or the medium. Choice of solid–liquid separation method is influenced by the
size and morphology of the microorganism (single cells, aggregates, or mycelia), and the specific
gravity, viscosity and rheology of the spent fermentation medium. These factors can also have
major influences on the transfer of the liquid through pumps and pipes.

Separating the microbial mass is necessary whether the product is inside the microorganism or
outside it; the difference is that in the first case the microbial mass is kept, whiles in the second,
the same mass is thrown away.

This can be done by:

(i) Centrifugation (expensive but guaranteed efficient),
(ii) Filtration methods, especially cross-flow filtration, or by
(iii) Flocculation (adding something to the microbes so that they clump together and
settle out on their own). If the product is inside the microorganism then separation also
concentrates the product: however, the microorganism must be first broken open to
get to the product (cell disruption).

Broth Conditioning
Broth conditioning techniques are mostly used in association with sedimentation and
centrifugation for the separation of cells from liquid media. They alter or exploit some property of
a microorganism, or other suspended material, such that it flocculates and usually precipitates.
However, in certain cases it may be used to promote flotation. This uses the ability of some cells
to adsorb to the gas–liquid interfaces of gas bubbles and float to the surface for collection, which
occurs naturally in traditional ale and baker’s yeast fermentations. Certain floc precipitation
methods are also used at the end of many traditional beer and wine fermentation processes,
where the addition of finings (egg albumen, isinglass, etc.) may be employed to precipitate yeast
cells. Major advantages of these techniques are their low cost and ability to separate microbial
cells from large volumes of medium.

Some organisms naturally flocculate, which can be enhanced by chemical, physical, and
biological treatments. Such treatments can also be effective with cells that would not otherwise
form flocs. Coagulation, the formation of small flocs from dispersed colloids, cells, or other
suspended material, can be promoted using coagulating agents (simple electrolytes, acids,
bases, salts, multivalent ions and polyelectrolytes). Subsequent flocculation, the agglomeration
of these smaller flocs into larger settleable particles, is often aided by inorganic salts (e.g., calcium

52
chloride) or polyelectrolytes. These are high molecular weight, water soluble, anionic,
cationic or non-ionic organic compounds, such as polyacrylamide and polystyrene sulphate.

Sedimentation
Sedimentation is extensively used for primary yeast separation in the production of alcoholic
beverages, and in wastewater treatment. This low-cost technology is relatively slow and is
suitable only for large flocs (greater than 100µm diameter). The rate of particle sedimentation is
a function of both size and density. Hence, the larger the particle and the greater its density, the
faster the rate of sedimentation. The basis of this method of separation is sedimentation under
gravity, which for a spherical particle can be represented by Stokes’ Law:

Where;
Vg = rate of particle sedimentation (m/s);
dp = diameter of the particle (m);
ps – pl = difference in density between the particle and surrounding medium (kg/m3);
g = gravitational acceleration (m/s2); and,
η = viscosity (Pascal seconds (Pa s)).

Therefore, for rapid sedimentation the difference in density between the particle and the
medium needs to be large, and the medium viscosity must be low.

Centrifugation
If instead of simply using gravitational force to separate suspended particles, a centrifugal field is
applied, the rate of solid-liquid separation is significantly increased, and much smaller particles
can be separated. Centrifugation may be used to separate particles as small as 0.1µm diameter
and is also suitable for some liquid-liquid separations. Its effectiveness, too, depends on particle
size, density difference between the cells and the medium, and medium viscosity.

In a centrifuge, the terminal velocity of a particle is:

Where;
Vc = centrifugal sedimentation rate or particle velocity (m/s);
ω = angular velocity of the centrifuge (rad/s); and,
r = distance of the particle from the centre of rotation (m)
(for η, ps – pl and dp, see Stokes law above).

Hence, the faster the operating speed (ω) and the greater the distance from the centre of rotation,
the faster the sedimentation rate (Vc).

53
Centrifuges can be compared using the relative centrifugal force (RCF) or g number (the ratio of
the velocity in a centrifuge to the velocity under gravity = ω2r/g).

The choice of centrifuge depends on the particle size and density, and the viscosity of the medium.
Higher-speed centrifuges are required for the separation of smaller microorganisms, such as
bacteria, compared with yeasts. For example, relatively slow centrifugation effectively recovers
residual yeast cells remaining in beer after the bulk has sedimented out. Conversely, an RCF of
20000xg may be required to recover suspended bacterial cells, cell debris and protein precipitates
from liquid media.

Figure: Diagram of a Tubular Bowl Centrifuge.

Advantages of centrifugation include the availability of fully continuous systems that can rapidly
process large volumes in small volume centrifuges. Centrifuges are steam sterilizable, allowing
aseptic processing, and there are no consumable costs for membranes, chemicals, or filter aids.
However, the disadvantages of centrifugation are the high initial capital costs, the noise generated
during operation and the cost of electricity. Also, physical rupture of cells may occur due to high
shear and the temperature may not be closely controllable, which can affect temperature-sensitive
products.

54
Bioaerosol generation is a further major disadvantage, particularly when centrifuges are used for
certain recombinant DNA organisms or pathogens. Under these circumstances the equipment
must be contained.

Industrial Centrifuges
Centrifuges can be divided into small-scale laboratory units and larger pilot-scale and industrial-
scale centrifuges.
Laboratory batch centrifuges include, in ascending order of speed attainable: bench-top, high-
speed and ultracentrifuges, capable of applying RCFs of 5000 to 500000xg. Although industrial
batch centrifuges are available, for most industrial purposes semi-continuous and continuous
centrifuges are required to process the large volumes involved. However, the RCFs achieved
are relatively low.

Four main types of industrial centrifuge are commonly used.

1 Tubular centrifuges
Tubular Centrifuges usually produce the highest centrifugal force of 13000 to17000xg. They
have hollow tubular rotor bowls providing a long flow path for the suspension, which is pumped
in at the bottom and flows up through the rotor (Figure). Particulate material is thrown to the side
of the bowl, and clarified liquid passes out at the top for continuous collection. As the particulate
material accumulates on the inside of the bowl, the operating diameter becomes reduced.
Consequently, there must be periodic removal of solids.

2 Multi-chamber Bowl Centrifuges

Figure: Diagram of a Multi-chamber Bowl Centrifuge

A multi-chamber bowl centrifuge consists of a bowl that is divided by vertically mounted
interconnecting cylinders and are capable of operating at 5000 to 10000xg (Figure). The liquid

55
feed passes from the centre through each chamber in turn, and the smaller particles collect in the
outer chambers.

Figure: A tubular bowl centrifuge.

(a) Schematic diagram of a tubular bowl centrifuge. Fermentor broth enters the rotor at the bottom
via a pump. The cells are centrifuged onto an internal Teflon sheet, whilst the supernatant passes
from the top of the system through a series of nozzles. (b) Photograph of cells sedimented onto
the Teflon lining of the rotor following cell harvesting and subsequent removal from the centrifuge.

56
3 Disc stack centrifuges
Disc stack centrifuges can operate at 5000–13000xg. The centrifuge bowl contains a stack of
conical discs whose close packing aids separation (Figure). As liquid enters the centrifuge
particulate material is thrown outwards, impinging on the underside of the cone discs. Particles
then travel outwards to the bowl wall where they accumulate. These centrifuges usually have the
facility to discharge the collected material periodically during operation.

Figure: Diagram of a Disc Stack Centrifuge

4 Screw-Decanter Centrifuges
A Screw-decanter centrifuge operate continuously at 1500–5000xg and are suitable for
dewatering coarse solid materials at high solids concentrations. They are used in sewage systems
for the separation of sludge, and for harvesting yeasts and fungal mycelium.

Filtration
Conventional filtration of liquids containing suspended solids involves depth filters composed of
porous media (cloth, glass wool or cellulose) that retain the solids and allow the clarified liquid
filtrate to pass through. As filtration proceeds collected solids accumulate above the filter medium,
resistance to filtration increases and flow through the filter decreases. These techniques are
generally useful for harvesting filamentous fungi but are less effective for collecting bacteria. The
two main types of conventional filtration commonly used in industry are as follows:

1 Plate and Frame Filters (Filter Presses)

These are industrial batch filtration systems (Figure). They are normally in the form of an
alternating horizontal stack of porous plates and hollow frames. The stack is mounted in a support
structure where it is held together with a hydraulic or screw ram. Filter cloths are held in place

57
between the plates that contain flow channels for the feed and permeate streams. This essentially
forms a series of cloth-lined chambers into which the cell suspension is forced under pressure.
Following batch filtration, the apparatus must be dismantled to remove the collected filter cake.
These systems are used for harvesting microorganisms from fermentations, including the
preparation of blocks of baker’s yeast, the recovery of protein precipitates and the dewatering of
sewage sludge. Similar horizontal and vertical pressure leaf filters are also available.

Figure: A Plate and Frame Filter

Figure: Cross-flow filtration. Schematic of a cross-flow (tangential-flow) filter. The fermentor broth
passes over the surface of the filter via a pump. Pressure applied by the pump forces a proportion
of the liquid through the filter, with the remainder passing over the filter surface along with the
retained cells. The retentate is then recycled around the system as a closed loop. Cell debris
cannot ‘blind’ the filter since the tangential fl ow washes the filter clear of solids.

2 Rotary Vacuum Filters

Rotary vacuum filters are simple continuous filtration systems that are used in several industrial
processes, particularly for harvesting fungal mycelium during antibiotic manufacture, for baker’s
yeast production and in dewatering sludge during waste-water treatment. The device comprises
a hollow perforated drum that supports the filter medium (Figure). This drum slowly rotates in a
continuously agitated tank containing the suspension to be filtered. Solids accumulate on the filter

58
medium as liquid filtrate is drawn, under vacuum, through the filter medium into the hollow drum
to a receiving vessel. As the drum rotates, collected solids held on the filter medium are removed
by a knife that cuts/sloughs them off into a collection vessel. Filter media may be pre-coated with
a filter aid, e.g., Kieselguhr (diatomaceous earth), which can be continuously replenished. The
rate of filtration (flow of filtrate), for a constant-pressure (vacuum) and incompressible cake, is
determined by the resistance of the cake and the filtration medium.

In terms of biosafety, neither filtration system is suitable for processing toxic products, pathogens
or certain recombinant DNA microorganisms and their products.

Figure: A Rotary Vacuum Filter Showing the Removal of Filter Cake Using a
Knife.

Membrane Filtration
Modern methods of filtration involve absolute filters rather than depth filters. These consist of
supported membranes with specified pore sizes that can be divided into three main categories.

They are, in decreasing order of pore size, microfiltration, ultrafiltration and reverse osmosis
membranes. The suspension to be filtered is pumped across the membrane (cross/tangential-
flow) rather than at a right angle to it, as occurs with conventional filtration methods. This retards
fouling of the membrane by particulate materials. Particles whose size is below the membrane
‘cut-off’ will pass through the membrane to become the ultrafiltrate or permeate, whereas the
remainder is retained as the retentate. As filtration progresses, the flux across the membrane
can slow due to membrane fouling. This may be caused by the accumulation of a layer of solute
molecules on the surface of the membrane, referred to as concentration polarization. The
presence of silicon antifoams may have a similar negative effect.

Microfiltration
Microfiltration is used to separate particles of 10 to 2 µm to 10 µm, including removal of microbial
cells from the fermentation medium. This method is relatively expensive due to the high cost of
membranes, but it has several advantages compared with centrifugation. They include quiet
operation, lower energy requirements, the product can be easily washed, good temperature

59
control is possible, containment is readily achieved and no bioaerosols are produced.
Consequently, it is suitable for handling pathogens and recombinant microorganisms.

Ultrafiltration
Ultrafiltration is similar to microfiltration except that the membranes have smaller pore sizes, and
are used to fractionate solutions according to molecular weight, normally within the range 2000
to 500000 Da. The membranes have anisotropic structure, composed of a thin membrane with
pores of specified diameter providing selectivity, lying on top of a thick, highly porous, support
structure.

Figure: A Diagrammatic Representation of a Hollow-Fibre Ultrafiltration Unit

A membrane manufactured with an exclusion size of 100000 Da, for example, should produce a
retentate of proteins and other molecules over 100000 Da and an ultrafiltrate of all molecules
below 100000 Da. However, non-spherical proteins may exhibit different exclusion reactions to
the membrane. Flat membranes are available, but for larger-scale operations hollow-fibre
systems are usually preferred (Figure).

Several of these ultrafiltration units can be linked together to produce a sophisticated purification
system. These methods are extensively employed for the purification of proteins, and for
separating and concentrating materials. Ultrafiltration is also effective in removing pyrogens

60
(bacterial cell wall lipopolysaccharides), cell debris and viruses from media, and for whey
processing.

Another variation on the ultrafiltration system is diafiltration, where water or other liquid is filtered
to remove unwanted low molecular weight contaminants. This can be used as an alternative to
gel filtration or dialysis for removing ammonium sulphate from a protein preparation precipitated
by this salt for changing a buffer or in water purification.

Reverse Osmosis
Reverse osmosis is used for dewatering or concentration steps and has been employed to
desalinate sea water for drinking. In osmosis water will cross a semipermeable membrane if the
concentration of osmotically active solutes, such as salt, is higher on the opposite side of the
membrane. However, if pressure is applied to the ‘salt side’ then reverse osmosis will occur, and
water will be driven across the membrane from the salt side. This reversal of osmosis requires a
high pressure, e.g., a pressure of 30 to 40 bar is needed to dewater a 0.6mmol/L salt solution
(note: 1 bar=100 kPa=0.987atm). Consequently, a strong metal casing is required to house this
equipment. As the membranes have pore sizes of only 10 to 2 to 10 to 4µm diameter, solute
molecules can deposit on the surface, causing a large resistance to solvent flow. However, this
fouling can be overcome by increasing the turbulence at the surface of the membrane.

Cell Disruption
Some target products are intracellular, including many enzymes and recombinant proteins,
several of which form inclusion bodies. Therefore, methods are required to disrupt the
microorganisms and release these products. The breaching of the cell wall/envelope and
cytoplasmic membrane can pose problems, particularly where cells possess strong cell walls. For
example, a pressure of 650 bar is needed to disrupt yeast cells, although this may vary somewhat
at different times during the growth cycle and depending upon the specific growth conditions.

General problems associated with cell disruption include the liberation of DNA, which can
increase the viscosity of the suspension. This may also affect further processing, such as pumping
the suspension on to the next unit process and flow through chromatography columns. A nucleic
acid precipitation step or the addition of DNase can help to prevent this problem. If mechanical
disruption is used then heat is invariably generated, which denatures proteins unless appropriate
cooling measures are implemented. Products released from eukaryotic cells are often subject to
degradation by hydrolytic enzymes (proteases, lipases, etc.) liberated from disrupted lysosomes.
This damage can be reduced by the addition of enzyme inhibitors, cooling the cell extract and
rapid processing. Alternatively, attempts may be made to produce mutant strains of the producer
microorganism lacking the damaging enzymes.

Cell disruption can be achieved by both mechanical and non-mechanical methods. The disruption
process is often quantified by monitoring changes in absorbance, particle size, total protein
concentration or the activity of a specific intracellular enzyme released into the disrupted
suspension.

61
Mechanical Cell Disruption Methods
High Pressure Homogenisators
Several mechanical methods are available for the disruption of cells. Those based on solid shear
involve extrusion of frozen cell preparations through a narrow orifice at high pressure. This
approach has been used at the laboratory scale, but not for large-scale operations.

Methods utilising liquid shear are generally more effective.

Among the liquid shear disruption devices, the high-pressure Manton-Gaulin APV type
homogenizer is probably the most widely used. The high-pressure homogenizer consists of a
positive displacement piston pump with one or more (usually three) plungers. The cell suspension
is drawn through a check valve into the pump cylinder and, on the pressure stroke (extreme high
pressure, 200 - 600 - 1000 bar), is forced through an adjustable discharge valve with restricted
orifice.

Figure: High-Pressure Homogenizer Discharge Valve Unit.

62
The French press (pressure cell) is often used in the laboratory and the high-pressure
homogenizers, such as the Manton and Gaulin homogenizer (APV-type mill), are employed for
pilot- and production-scale cell disruption. They may be used for bacterial and yeast cells, and
fungal mycelium. In these devices the cell suspension is drawn through a check valve into a pump
cylinder. At this point, it is forced under pressure (up to 1500 bar) through a very narrow annulus
or discharge valve, over which the pressure drops to atmospheric. Cell disruption is primarily
achieved by high liquid shear in the orifice and the sudden pressure drop upon discharge causes
explosion of the cells.

The rate of protein release (efficiency of disruption) is independent of the cell concentration, but
is a function of the pressure exerted, the number of cycles through the homogenizer and the
temperature. Disruption of yeast cell preparations, for example, typically requires three passes
through the pressure cell at 650bar, whereas wild-type Escherichia coli generally needs 1100 to
1500 bar. During processing the temperature rises by about 2.2 to 2.4°C per 100bar, i.e., by
approximately 20°C over one pass at 800bar. Consequently, precooling of the cell preparation is
usually essential. The energy input necessary is approximately 0.35 kW per 100bar and
the throughput is up to 6000 L/h. A problem with this method of cell disruption is that all
intracellular materials are released. As a result, the product of interest must be separated from a
complex mixture of proteins, nucleic acids and cell wall fragments. Some fragments of cell debris
are not readily separated, making the solution difficult to clarify. In addition, proteins may be
denatured if the equipment is not sufficiently cooled and filamentous microorganisms may block
the discharge valve. When used for certain categories of microorganisms, the homogenizers have
to be contained to prevent the escape of aerosols.

On a small scale, manual grinding of cells with abrasives, usually alumina, glass beads,
kieselguhr or silica, can be an effective means of disruption, but results may not be reproducible.

Bead Mills
Cell disruption in bead mills is regarded as one of the most efficient techniques for physical cell
disruption. Various designs of bead mills have been used for microbial cell disruption. These mills
consist of either a vertical or a horizontal cylindrical chamber with a motor-driven central shaft
supporting a collection of off-centred discs or other agitating element. The chamber is filled to the
desired level with steel or 0,1 to 2 mm abrasion-resistant glass beads (ballotini) which provide the
grinding action. The charge of grinding beads is retained in the chamber by a sieve plate. The
horizontal configuration of the mill is known to give a better efficiency of disruption relative to the
vertical one.

63
Advantages:
• Continuous operation possible.
• Scale up possible (up to 600 to 1000 litre grinding space)

Disadvantages:
• Large energy consumption (needs cooling)

In industry, high-speed bead mills, equipped with cooling jackets, are often used to agitate a
cell suspension with small beads (0.5 to 0.9µm diameter) of glass, zirconium oxide or titanium
carbide. Cell breakage results from shear forces, grinding between beads and collisions with
beads. The efficiency of cell breakage is a function of agitation speed, concentration of beads,
bead density and diameter, broth density, flow rate and temperature. Cell concentration is also a
major factor (optimum 30 to 60% dry weight), which is an important difference from the liquid
shear homogenizers described. Maximum throughput in these systems is about 2000 L/h.

Ultrasonic Disruption
Ultrasonic Disruption of cells involves cavitation, microscopic bubbles or cavities generated by
pressure waves. It is performed by ultrasonic vibrators that produce a high frequency sound with
a wave density of approximately 20 kilohertz/s. A transducer converts the waves into mechanical
oscillations via a titanium probe immersed in the concentrated cell suspension. However, this
technique also generates heat, which can denature thermolabile proteins. Rod-shaped bacteria
are often easier to break than cocci, and Gram-negative organisms are more easily disrupted
than Gram-positive cells.

Sonication
A common laboratory-scale method for cell disruption applies ultrasound (typically 20 to 50 kHz)
to the sample (sonication). In principle, the high-frequency is generated electronically and the
mechanical energy is transmitted to the sample via a metal probe that oscillates with high
frequency. The probe is placed into the cell-containing sample and the high-frequency

64
oscillation causes a localised low pressure region resulting in cavitation and impaction,
ultimately breaking open the cells. Acoustic cavitation is the formation, growth, and implosive
collapse of bubbles in a liquid. Although the basic technology was developed over 50 years ago,
newer systems permit cell disruption in smaller samples (including multiple samples under 200
µL in microplate wells) and with an increased ability to control ultrasonication parameters.

Disadvantages:
• Heat generated by the ultrasound process must be dissipated (cooling).
• High noise levels (most systems require hearing protection and sonic enclosures)
• Yield variability.
• Free radicals are generated that can react with other molecules.
• Labour size only

Sonication is effective on a small scale, but is not routinely used in large-scale operations, due
to problems with the transmission of power and heat dissipation.
Cell disruption is a somewhat neglected area of bioprocessing, as there has been relatively little
innovation and progress. Even the routinely used established mechanical methods were originally
devised for other purposes. However, some newer disruption systems are being developed to
give improved large-scale disruption, often with integral containment. They include a newly
designed ball mill, the CoBall Mill; the Constant Systems high-pressure disrupter, which operates
at up to 2700 bar; and two systems with no moving parts, the Microfluidics impingement jet system
and the Glass-col nebulizer. The Parr Instruments cell disruption bomb is designed for disrupting
mammalian cells. This is a relatively gentle method that works on the principle of nitrogen
decompression and does not generate heat. Nitrogen is dissolved in cells under high pressure,
and sudden pressure release then causes the cells to burst.

X-Press (Freeze-Press)
Freeze-pressing of microbial cell suspensions can be used to disrupt the cells. Examples of the
freeze-pressing equipment include the Hughes press in which a frozen paste of cells is forced
through a narrow slit or orifice, at temperatures of about -25°C. The frozen cell suspension is
pressed trough an orifice.

65
How is it possible?
In this process phase and consequent volume changes of ice contribute to disruption.
If the pressure is high enough → 2000 to 6000 bar → the ice gets compressible =
deformable.

The first triple point: -22 ˚C, 211,5 MPa

Relative density of crystal forms:

Ice-I → Ice-III
0,92 → 1,14
volume reduction: -19%
Ice-III → Ice-V
1,14 → 1,23
volume reduction: -7%

In addition, solid shear due to crystalline ice is important. Cell breakage in the press yields cell
wall membrane preparations that are relatively intact and may be a good method for isolation of
membrane-associated enzymes.

Advantages:
• High efficiency
• No denaturation, decay
• Very concentrated cell cake can be disrupted.

Disadvantages:
• Batch operation only
• No scale up
• Heavy construction

66
 Figure:

Freezing-Thawing
The freeze-thaw method is commonly used to lyse bacterial and mammalian cells. The technique
involves freezing a cell suspension in a dry ice/ethanol bath or freezer and then thawing the
material at room temperature or 37 °C. This method of lysis causes cells to swell and ultimately
break as ice crystals form during the freezing process and then contract during thawing. Multiple
cycles are necessary for efficient lysis, and the process can be quite lengthy. However,
freeze/thaw has been shown to effectively release recombinant proteins located in the cytoplasm
of bacteria and is recommended for the lysis of mammalian cells in some protocols.

Change of Physical Conditions

Heat shock – in aqueous medium (0,5 -1 hour at 60 to 80-100°C, mind the heat
denaturation!).

Osmotic shock – with neutral compounds (sugars, sugar alcohols, and glycerol) not with
salts. “Saturate” the cells with sugar, then transfer them into pure water.

Solvent treatment
• Drying with acetone than dissolve the cell membranes with ether
Autolysis of yeasts with toluene.

Detergent treatment
• They penetrate into the cell membrane and destroy its structure.
Both cationic and anionic detergents
Bile acids.

Decompression

Henry’s law:

67
At high pressure, a lot of gas is dissolved in the liquid (even inside the cells). With a sudden
pressure drop the solubility drops, too- the gas forms bubbles everywhere (like in sodas) and
explodes the cells.

Enzymatic Methods
Enzymatic cell lysis, which is attractive in terms of its delicacy and specificity for just the cell wall
structure, is restricted by the cost of the enzyme which is lost into the extract. Enzymatic cell
lysis of bacteria using lysozyme has been successful for the isolation of enzymes.

Sensitivity of microorganisms to various lytic enzymes varies greatly with the type, the phase of
growth and fermentation conditions.

Specific Enzymes Hydrolyzing the Cell Wall:

In some cases, autolysis of microbial cells without any foreign enzyme may be possible. Autolysis
of yeasts can be promoted with toluene. Although autolysis is, in general, slower than other
disruption methods, it is volume independent and so 100 m3 of ceils may be lysed as fast as 100
cm3.

Non-Mechanical Cell Disruption Methods

An alternative to mechanical methods of cell disruption is to cause their permeabilisation. This
can be accomplished by autolysis, osmotic shock, rupture with ice crystals (freezing/thawing) or
heat shock. Autolysis, for example, has been used for the production of yeast extract and other
yeast products. It has the advantages of lower cost and uses the microbes’ own enzymes, so
that no foreign substances are introduced into the product.

68
Osmotic Shock
Osmotic shock is often useful for releasing products from the periplasmic space. This may be
achieved by equilibrating the cells in 20% (w/v) buffered sucrose, then rapidly harvesting, and
resuspending in water at 4°C.

A wide range of other techniques have been developed for small-scale microbial disruption using
various chemicals and enzymes. However, some of these can lead to problems with subsequent
purification steps. Organic solvents, usually acetone, butanol, chloroform, or methanol, have been
used to liberate enzymes and other substances from microorganisms by creating channels
through the cell membrane. Simple treatment with alkali or detergents, such as sodium lauryl
sulphate or Triton X-100, can also be effective.

Several cell wall degrading enzymes have been successfully employed in cell disruption. For
example, lysozyme, which hydrolyses ß-1,4 glycosidic linkages within the peptidoglycan of cell
walls, is useful for lysing Gram-positive organisms. Addition of ethylene diamine tetraacetic acid
(EDTA) to chelate metal ions also improves the effectiveness of lysozyme and other treatments
on Gram-negative bacteria. This is because EDTA has the ability to sequester the divalent cations
that stabilize the structure of their outer membranes. Enzymic destruction of yeast cell walls can
be achieved with snail gut enzymes that contain a mixture of β-glucanases. These enzyme
preparations are also useful for producing living yeast spheroplasts or protoplasts.

The antibiotics penicillin and cycloserine may be used to lyse actively growing bacterial cells,
often in combination with an osmotic shock. Other permeabilisation techniques include the use of
basic proteins such as protamine; the cationic polysaccharide chitosan is effective for yeast cells;
and streptolysin permeabilizes mammalian cells.

Conclusion
A cell disruption process cannot be considered in isolation from further downstream processing.
This is because the cell disruption operation affects the physical properties of the cell slurry such
as viscosity, density, particle size, and settleability of suspension, which in turn affect subsequent
processing. The choice of appropriate cell disruption equipment and the extent of treatment in an
equipment is an important decision in order to facilitate further downstream processing of the
disrupted material.

Product Recovery
Recovery of extracellular proteins is from the clarified medium, whereas disrupted cell
preparations are used for both intracellular proteins and those held within the periplasmic space.
In the case of some recombinant proteins expressed at high levels, they sometimes form inclusion
bodies that are released by cell breakage.

Following cell disruption, soluble proteins are usually separated from cell debris by centrifugation.
The resultant supernatant, containing the proteins, is then processed in a similar way to growth
medium containing excreted proteins. This can involve several different types of methods, some
of which have been previously described, such as ultrafiltration. An older, but nevertheless
effective, method used at this stage is salting out of proteins, followed by the recovery of
precipitated proteins by centrifugation. Precipitation is achieved by the addition of inorganic salts
at high ionic strength, usually in the form of solid or saturated solutions of ammonium sulphate.
Ammonium sulphate is popular because of its high solubility, low toxicity and low cost, but it can
liberate ammonia at high pH values and is corrosive to metal surfaces, e.g., centrifuges. The
solubility of the salt varies with temperature, so strict temperature control is required.

69
Reduction of protein solubility can also be achieved by adding organic solvents, such as acetone,
ethanol and isopropanol. This is performed at low temperature and reduces the dielectric constant
of the medium, causing precipitation of the proteins.

Aqueous two-phase separation involves partitioning the protein between the two phases,
depending upon its molecular weight and charge. Commonly used systems include dextran and
polyethylene glycol (PEG), or PEG and potassium phosphate. The two phases can then be
separated by centrifugation. This method is cheap, gentle, and versatile, and can be scaled up
for industrial applications, including the purification of enzymes, e.g., RNA polymerase from E.
coli.

Many alkaloids, antibiotics, steroids, and vitamins are recovered by liquid–liquid extraction
methods using organic solvents. Antibiotics, for example, are extracted from culture media into
solvents such as amyl acetate.

Those solvents used should be non-toxic, selective, inexpensive, immiscible with broth and must
have a high distribution coefficient for the product, i.e., the ratio of the product in the two phases.
Efficient recovery of the solvent for reuse is an essential aspect of overall process economics.

2. Product Concentration
The result of separation is usually a rather dilute solution of the product, which must be
concentrated. Some of the same methods can also be used for concentration. Simply drying large
volumes of liquid is usually too expensive, so again ultrafiltration or reverse osmosis (both
membrane methods that keep the product on one side of the membrane whiles most of the water
goes through it to the other side) are popular. Other popular methods include; adsorption
methods, or extraction with another liquid.

Most biotechnology products are produced as mixtures by cells but are required pure. Purification
methods include chromatography, affinity methods, and various specific precipitation
methods. If the product is produced by genetic engineering, then it may be engineered to have
a molecular “hook” which makes it easier to isolate.

The role of downstream processing will continue to be one of the most challenging and demanding
parts of many biotechnological processes. Purity and stability are the hallmarks of most high value
biotechnological products. On the industrial side, the scale of operation will, for economic reasons,
mainly be very large, and in almost all cases the final success will require the closest cooperation
between the bioscientist and the process engineer; in this way demonstrating the truly
interdisciplinary nature of biotechnological processes.

Traditional separation techniques are well-developed, widely practiced, and comprehensively

described in standard references. The techniques are categorised according to their fundamental
mechanisms as:
• Physical (adsorption, crystallisation, extraction, etc.),
• Mechanical (filtration, centrifugation, etc.),
• Thermal (distillation),
• Chemical (chemisorption, chromatography).

70
Because separations often dominate the economics of biochemical processing, development of
energy-efficient separation methods is of especially great importance to the commercial success
of the low-value, high-volume bioproducts most relevant to pollution prevention.

Downstream processing of bioreactor contents involves special challenges due to the presence
of biomass and non-product biomolecules such as proteins and sugars, which promote fouling
(coverage with biofilms or clogging with bioparticles), and also due to the necessarily dilute nature
of fermentation broths, which causes the use of energy-intensive distillation to be prohibitively
expensive.

State of the Science

Research is currently underway in improving many bioseparation techniques, offering hope that
major improvements in the efficiency of bioengineered processes can be achieved by
incorporating new bioseparation technologies into process design, and several recent, excellent
compilations of modern techniques are now available.

71
Figure: Overview of Bioseparation

72
1. Physical Separations
Emerging physical bioseparation techniques include aqueous two-phase extraction, reverse
micellar extraction, cloud point extraction, and magnetic and electrophoretic separation.

(i) Liquid-Liquid Extraction

Liquid extraction is commonly used to separate inhibitory fermentation products such as ethanol
and acetone-butanol from a fermentation broth. Antibiotics are also recovered by liquid extraction
(using amylactate or isoamylactate). Ideally, the liquid extractant should be nontoxic, selective,
inexpensive, and immiscible with fermentation broth and should have a high distribution
coefficient for the product.
The extraction of a compound from one phase to the other is based on solubility differences of
the compound in one phase relative to the other. When a compound is distributed between two
immiscible liquids, the ratio of the concentrations in the two phases is known as the distribution
coefficient:

KD = Y/X
where Y and X are concentrations of solute in light and heavy phases, respectively. In most, but
certainly not all cases, the light phase will be the organic solvent and the heavy phase will be the
aqueous fermentation broth.

Most antibiotics are extracted from the fermentation broth by using solvents such as amylacetate
or isoamylactate. Usually, continuous centrifugal Podbielniak extractors are used for the
extraction of antibiotics. Penicillin is more soluble in organic phase at low pHs (pH = 2 to 3) and
is highly soluble in aqueous phase at high pHs (pH = 8 to 9). Therefore, penicillin is extracted
between organic and aqueous phases several times by shifting the pH to improve the purity of
the product.

Some products to be recovered from the fermentation broth are weak acids or weak bases. Since
compounds that are not ionised are soluble in the organic phase, the pH conditions are selected
such that the extracted compound is neutral and is soluble in an organic solvent. Therefore, weak
bases are extracted at high pHs and weak acids are extracted at low pH values in the neutral
form.

A particularly important device for liquid-liquid extraction for fermentation products is the
Podbielniak centrifugal extractor. Many fermentation products are unstable. The rapid rotation
of the Podbielniak extractor produces a centrifugal field that rapidly drives the two fluids
countercurrent to each other; a product can be extracted and returned to another aqueous phase
(e.g., a phosphate buffer) within minutes.

A generalised block diagram of downstream processing of bioreactor media is shown in Figure 5

(2). During primary recovery, preliminary separation of solid (biomass) and aqueous (product)
phases, as well as product extraction, occur. The fermentation broth is also reduced significantly
in volume during this stage. Products are further concentrated during intermediate recovery,
which also involves, in some cases, the redissolution of materials before further concentration is
possible. In the final purification stages, a series of polishing steps are used to raise product purity
to the final specifications.

73
 (i) Two-Phase Partitioning Bioreactors
Among emerging techniques, the two-phase partitioning bioreactor appears to have great
potential in enhancing the productivity of many bioprocesses. The approach integrates
fermentation with a primary product recovery step by incorporating both organic and aqueous
phases simultaneously, such that microbial growth occurs in the aqueous media and substrates
and products partition into the organic phase based on their affinities for it.

This approach allows controlled substrate delivery to the fermentation broth and effectively lowers
product concentrations in the fermentation broth as well, promoting microbial health and activity.
Although it is already practiced commercially, its effectiveness could still be improved by the
discovery of low-cost solvents that are non-toxic for microbial growth.

(ii) Aqueous Two-Phase Extraction

Aqueous two-phase extraction is an approach under active development for the extraction of
soluble proteins such as enzymes between two aqueous phases containing incompatible
polymers such as polyethylene glycol (PEG) and dextran. Dextrans are more than 75% water and
are immiscible. Typical aqueous phases used for this purpose are PEG-water/dextran-water and
PEG-water/K phosphate-water, PEG/dextran and PEG/K phosphate are reasonably immiscible.
The partition coefficient, Kp, varies with the molecular weight (MW) of the soluble protein in the
form of an exponential function.
Kp = eAM/T
Where M is MW of protein, T is the absolute temperature, and A is a constant. The partition
coefficient, Kp, of many enzymes between the two phases (CPEG – CDEX) varies between 1 and
3.7, resulting in poor separation in a single stage. The partition coefficient can be improved by
including ion-exchange resins or certain salts, such as (NH4)2SO4 and KH2PO4 in one of the
phases. More than tenfold increase in Kp values may be achieved by increasing the concentration
of KH2PO4 from 0.1 to 0.3 M. In PEG-salt systems, salting-out may result in protein precipitation
at the interface. Ion-exchange resins can be used to increase the value of a partition coefficient.
After the extraction step, the phases can be separated by centrifugation or decantation, and PEG
can be recovered by ultrafiltration.

(iii) Non-Solvent-Based Processing

A primary purpose of nonpolar organic solvents in bioprocessing is the dissolution of cell
membranes to release intracellular products.
However, organic solvents are undesirable for several reasons: they are frequently derived from
non-renewable resources such as petroleum; they are frequently toxic and/or carcinogenic, and
consequently expensive to treat in disposal; and they frequently form non-aqueous phases that
hinder biodegradation. As a result, efforts to develop efficient, non-solvent-based approaches to
cell disruption and product purification deserve high priority.

The purification of polyhydroxyalkanoate (PHA) polymers provides an excellent example of the

benefits possible with such approaches. Following fermentation, PHA-containing cells are
separated from culture media by centrifugation, filtration, and/or flocculation, and cells are then
disrupted to recover the polymer. Subsequent recovery, unfortunately, typically involves
extraction of the polymer from biomass with large amounts of toxic and inflammable organic
solvents (e.g., chloroform methylene chloride, propylene carbonate, or dichloroethane). Another
established method, again involving environmentally undesirable reagents, uses sodium
hypochlorite for the digestion of non-PHA cellular materials, carrying the additional disadvantage
that it partially degrades the PHA.

74
To improve the environmental friendliness of PHA processing, a non-solvent-based process has
been developed by Zeneca Agrochemicals (now part of Syngenta; www.syngenta.com) to assist
in the commercial production of poly[(R)-3-hydroxybutyrate] [poly(3HB)] and poly(3HB-co-3V)
by Alcaligenes eutrophus. In this process, cells are brought to 80°C and treated with a mixture
of hydrolytic enzymes, including lysozyme, phospholipase, and lecithinase that hydrolyse
cellular components without degrading the polymer. The polymer can then be recovered as a
white powder after washing and drying, demonstrating the technical feasibility of performing novel
separations utilising advanced bioengineering techniques. Cost of necessary reagents, such as
the enzymes involved in this case, remains an important consideration, however, and must be
addressed to enable such techniques to compete with existing routes to synthesis of plastics
based on petrochemical sources.

2. Mechanical Separations
A great variety of membrane-based techniques are not only under development, but are already
central to commercial integrated bioprocesses, enabling further avoidance of environmentally
deleterious reagents in bioprocessing. These approaches separate products from culture media
based on hydrophobicity, volatility, and/or affinity for membrane components, and offer the
potential for great selectivity as well as low-energy and waste-disposal requirements in
biorefining.

(i) Size-Based Separations

Membranes have traditionally been used for size-based separations that require high throughput
but relatively low resolution. Examples include microfiltration (MF) for clarification and
sterilisation as well as ultrafiltration (UF) for product concentration and buffer exchange.

Membranes separations are usually used to separate proteins after precipitation. Membranes
serve as a molecular sieve to separate solute molecules of different molecular size. Depending
on molecular size cutoff, different membranes can be used for the separation of different MW
proteins. Microfilters are used to separate species that range from 10 to 10-2 μm. Ultrafilters are
used for a molecule weight range of 2,000 to 500,000.

All these membranes sieving methods (microfiltration, ultrafiltration, reverse osmosis) are based
on the same theory with minor differences. However, ultrafiltration (UF) membranes are different
from most microporous filters because of their anisotropic structure. In an anisotropic
membrane, a thin skin with small pores is formed on top of a thick, highly porous structure.
The thin layer provides selectivity, while the thicker layer provides mechanical support. Many
microporous filters have an open, tortuous path structure, which causes particles entrapment
within the filter, while others have well-defined pores of uniform size.

Microfiltration and ultrafiltration promise to become major unit operations in the emerging
biorefinery arena. The development of new materials for UF and MF, including porous metals and
ceramics as well as polymers, is therefore an important priority. Similarly, nanofiltration and
reverse osmosis are becoming increasingly important, with recent developments in
nanotechnology promising to yield new materials with significantly improved fluxes and
selectivities.

Microfiltration is competitive with depth filtration, centrifugation, and expanded-bed

chromatography for the initial harvest of products from bacterial, yeast, and mammalian cell
cultures. Use of 0.2 µm-rated MF membranes can yield a particle-free harvest solution requiring

75
no additional clarification prior to purification, while larger pore-size membranes can be used to
improve product yield and throughput when the filtrate is subsequently treated with a normal flow
filter for final clarification. MF systems are generally operated at constant flux instead of constant
transmembrane pressure to improve product yield and throughput, although this practice tends to
exacerbate fouling problems. An emerging solution to this problem is high-frequency, back-
pulsing air scrubbing to clean the surface of the membrane continuously.

Figure: Types of Ideal Continuous Flows used in Membrane-Based Separations

Ultrafiltration, in turn, has become the method of choice for protein concentration, replacing size-
exclusion chromatography in this application. Plasmid DNA and virus-like particles can also be
purified using UF.

UF membranes are typically composed of polysulfone, polyethersulfone, or regenerated

cellulose, and experience fouling problems analogous to those that plague MF membranes.
Newly-developed composite cellulose membranes are less susceptible to fouling than are many
synthetic polymers, and are more easily cleaned, but still possess excellent mechanical strength,
thus outperforming other membrane materials. Nevertheless, cellulose has had a much longer
development history than have synthetic polymers in membrane applications, and promising
developments in the latter are still expected.

UF membranes are widely used in the pharmaceutical, chemical, and food industries for the
separation of vaccines, fermentation products, enzymes, and other proteins. Ultrafiltration is an
energy-efficient, economical separation method used to concentrate chemicals and biologicals at
a high degree of purity. A typical UF operation is a pressure-driven process in which low-MW
solutes and water pass through the filter and high-MW solutes are retained on the surface of the
membrane surface. Therefore, a concentration gradient builds up between the surface of the
membrane and the bulk fluid. This gradient results in concentration polarisation. As a result of
concentration polarisation, solute diffuses back from the membrane surface to the solution.

76
 Figure: Ultrafiltration Separates Molecules Based on Size and Shape.
(a) Diagrammatic representation of a typical laboratory-scale ultrafiltration system. The sample
(e.g., crude protein solution) is placed in the ultrafiltration chamber, where it sits directly
above the ultrafilter membrane. The membrane, in turn, sits on a macroporous support to
provide it with mechanical strength. Pressure is then applied (usually in the form of an inert
gas), as shown. Molecules larger than the pore diameter (e.g., large proteins) are retained
on the up-stream side of the ultrafilter membrane. However, smaller molecules (particularly
water molecules) are easily forced through the pores, thus effectively concentrating the
protein solution (see also (b)). Membranes that display different pore sizes, i.e., have
different molecular mass cut-off points, can be manufactured. (c) Photographic
representation of an industrial-scale ultrafiltration system.

(ii) Membrane Chromatography

An improved understanding of the interactions between culture media components and synthetic
polymers suitable for membranes would greatly facilitate the design of synthetic substrates for
use in membrane chromatography.

77
Among those, ligand-binding and sterically-interacting species should be investigated closely to
improve the selectivity of membrane chromatography while maintaining acceptably high
throughput.

Recent studies are generating renewed interest in membrane chromatography. Diffusion

limitations are less pronounced in membranes than in conventional bead packings, thus providing,
in principle, a binding capacity independent of flow rate. Although the high, internal surface area
provided by small pores in membranes is often offset by the reduction in convective flow, and it
is challenging to achieve binding capacities competitive with bead packings, research into
optimisation of membrane chromatography continues. Especially encouraging are manifold
designs with high membrane density and low retention volumes that can accomplish up to 100-
fold concentrations in a single stage.

Flow distribution within membrane modules is an important consideration for optimal

performance, especially in manifold designs, and careful design of entrance and exit regions is
essential to provide even, well-distributed flow. In addition, sensing and maintenance of constant
concentration of the retained species at the membrane surface is becoming a new robust control
strategy, enhancing product yield, and providing greater operational robustness with respect to
variations in feed quality. Finally, additional efforts are directed toward improving binding capacity,
membrane selectivity, flow distribution, and flow rate, through adjustments in pore size,
membrane chemistry, and membrane morphology.

3. Other Membrane Techniques

The development of new and/or improved membrane materials that provide increased selectivity
and specificity for the desired substances, as well as increased flux with stability and robustness,
is of central importance to the membrane-based techniques.

(i) Pervapouration
Pervapouration is the separation of various liquid mixtures by partial vapourisation through a
non-porous membrane. The use of pervapouration to remove either water or bioproducts from
bioreactor media appears promising. The membrane acts as a selective barrier between the two
phases: the liquid feed and the vapour permeate phases. It allows the desired component(s) of
the liquid feed to dissolve within it and then to transfer through it by vapourisation, resulting in a
separation based primarily on differences in polarity rather than volatility.
Continued support for new membrane materials, new module and process designs, and improved
theoretical understanding and modeling of the pervapouration process should therefore be
pursued.

Pervapouration has shown great promise in separating alcohols such as ethanol and butanol from
bioreactor media, as in the production of ethanol from rice straw by Pichia stipitis, as well as other
azeotropic and close-boiling mixtures, including isomers. A challenge associated with the use of
pervapouration is the accumulation of less volatile components (higher alcohols) in fermentation
media that can cause microbial growth inhibition. Nevertheless, industrial uses for this method
are widespread, and membrane technologies as well as integration of pervapouration with other
unit operations are still showing promising improvement.

78
 Figure: Pervaporation process

(iii) Electrodialysis
Electrodialysis is separation of ions occurs due to the imposed potential difference across the ion
selective cationic and anionic exchange membranes. The unit consists of a stack of compartment
formed by alternate cationic and anionic exchange membranes. Electrokinetic transport of
positively charged species through the cationic membranes and negatively charged through
anionic membrane occurs and selective separation of mixtures of ionic species depending on the
different ionic mobilities of species is facilitated.

Figure: Electrodialysis (1) cation selective membrane (2) anion selective membrane (3)
cathode (4) anode (5,6) spacers

Antifouling Techniques
Fouling is a persistent problem among membrane technologies, with the result that methods to
diminish fouling of membranes and ion exchange materials, as well as to remove impurities such
as salts or acids that cause complications in downstream processes, are high priorities in the
advancement of bioseparations.

79
4. Microbe Engineering
Sometimes bioprocessing procedures can be simplified greatly by engineering an organism
itself to produce a more convenient product, as in the development of DuPont’s bio-based
1,3-propanediol (3G) process. In nature, two separate microorganisms are required to convert
glucose to glycerol and then to convert glycerol to 3G. To avoid the difficulty of purifying the
glycerol, a genetically-engineered microorganism was developed that converted glucose directly
to 3G. With increasing advances in metabolic pathway engineering such avenues are expected
to become increasingly available.

Environmentally Benign Solvents

New renewable, biodegradable solvents are needed to support environmentally-friendly
extraction processes. Supercritical CO2, a highly compressed phase of CO2 possessing
properties of both liquid and gas phases, is one benign solvent that has already achieved great
popularity and that has the potential to contribute performance, cost-effectiveness, and
sustainability to separations of both biofuels and biomaterials.

Integrated Modules
Combined-unit or hybrid-unit operations, in which a bioreactor is integrated with a
bioseparation module, as in two-phase reactor systems, are particularly attractive as means to
overcome limitations inherent to bioprocessing. These are particularly desirable for their potential
to remove products as they are synthesized, alleviating the nearly universal problem of product
inhibition in culture media.

PURIFICATION METHODS: SMALL SCALE

Many biotechnological products must be extremely pure, for use as drugs or in producing fine
chemicals, so the relatively crude purification methods that isolate them from large-scale culture
are not good enough. A further purification step is needed. There are many such methods, but
since they are complicated and expensive, they are usually only used on a small scale
(micrograms to grams, rather than grams to tonnes). Most are chromatography methods. Here
the mixture is passes down a tube that is packed with some material to which some components
in the mixture stick and others do not. It does not matter whether the product sticks or not,
provided the contaminants do the opposite.

CHROMATOGRAPHY
Chromatographic techniques are usually employed for higher-value products. These methods,
normally involving columns of chromatographic media (stationary phase), are used for desalting,
concentration, and purification of protein preparations. In choosing a chromatographic technique
a number of considerations must be taken into account. For protein products these factors include
molecular weight, isoelectric point, hydrophobicity, and biological affinity. Each of these properties
can be exploited by specific chromatographic methods that may be scaled up to form an industrial
unit process.

The order and choice of technique will depend on the particular product, but the following
chromatographic parameters should be considered: capacity, recovery and resolving power
(selectivity). The capacity refers to the sample size that can be applied to the system in terms of
protein concentration and volume, and the recovery is the yield of product at each stage.

80
Yield values should be as high as possible otherwise the overall process will be uneconomic.
Resolving power and selectivity relate to the ability to separate two components, one being the
product and the other the impurity. This is particularly important at the final purification stage.
Care must be taken not to allow a protein to denature during purification, so columns are usually
run at 4°C. Changes of pH, excessive dilution and addition of certain chemicals can also affect
the stability of the protein.

Adsorption Chromatography
Adsorption chromatography separates according to the affinity of the protein, or other material,
for the surfaces of the solid matrix. Alumina (Al2O3), hydroxyapatite (Ca10(PO4)6(OH)2) or silica
(SiO2) are used for purifying non-polar molecules, whereas polystyrene-based resins are effective
matrices for polar molecules.

This technique involves hydrogen bonding and/or van der Waals forces. The adsorbed proteins
are usually eluted by increasing the ionic strength, often by using a gradient of increasing
phosphate ion concentration.

Figure: (a) Column chromatography (b) Planar chromatography

Affinity Chromatography (AFC).

Affinity chromatography is a particularly powerful and highly selective purification technique. It
can often give up to several thousand-fold purification in a single step. However, this
chromatographic method is expensive on an industrial scale. The technique involves specific
chemical interactions between solute molecules, such as proteins, and an immobilized ligand
(functional molecule). Ligands are covalently linked to the matrix material, e.g., agarose, via a
spacer arm to avoid steric hindrance. Some ligands interact with a group of proteins, notably

81
nicotinamide adenine dinucleotide, adenosine monophosphate, and Procion and Cibacron dyes;
other ligands are highly specific, especially substrates, substrate analogues and antibodies.

Since monoclonal antibodies have become more readily available, immunoaffinity

chromatography methods have been developed for the purification of various antigens. The
loading capacity of affinity columns is large, as the volume of the sample is unimportant and high
resolution can be attained. This is also a high-speed technique and elution is achieved using
specific cofactors or substrates; alternatively, non-specific elution may be performed with salt or
pH change. For industrial-scale operations, it is often necessary to sterilize the chromatographic
media in order to comply with various regulatory requirements. This can be problematical as some
ligands are sensitive to sterilizing agents.

Affinity chromatography is based on the specific chemical interactions between solute molecules
and ligands (a functional molecule covalently linked to a support particle) bound on support
particles. Ligand-solute interaction is extremely specific, such as enzyme-substrate interaction,
which may depend on covalent, ionic forces or hydrogen-bond formation. Affinity binding may be
molecular size and shape specific.

Some specific ligands are lectins such as con A for glycoprotein isolation. The choice of the ligand
depends mainly on two factors namely, (i) availability of chemically modified groups on the ligand
to facilitate its attachment to the matrix while retaining binding capacity with the biomolecule (ii)
affinity towards the counter-ligand with suitable K values to ensure maximum binding and
desorption. Spacer arm usage helps in proper access of the binding site to the molecule.
Commonly used spacer arm are linear aliphatic hydrocarbons with terminal functional groups to
facilitate the binding of the spacer molecule to the matrix as well as the ligand. For coupling the
ligand to the matrix, the following steps have to be adopted.

Chemical activation of the matrix,

Immobilization of the ligand via the chosen functional group and, Blocking or deactivating
the residual active groups. After coupling and blocking the matrix can be packed and used for
separation. In the economic point of view, the column is regenerated and reused.

Figure: Affinity Chromatography Matrix Preparation

82
Gel Filtration (Molecular Sieving) Chromatography
This is a chromatographic method in which the molecules are separated by size and is based on
the penetration of solute molecules into small pores of packing particles on the basis of molecular
size and the shape of the solute molecules. The gel can be considered as containing two types
of solvent; that within the gel particle and that outside it. Large particles which cannot penetrate
the gel appear in the column effluent after a volume equivalent to the solvent outside the gel has
emerged from the column. Small molecules which permeate the matrix appear in the effluent after
a volume equivalent to the total liquid volume within the matrix has emerged.

Gel filtration chromatography essentially involves separation on the basis of molecular size
(molecular sieving), although molecular shape can also influence separation performance.
Consequently, it is particularly useful for desalting protein preparations. The stationary phase
consists of porous beads composed of acrylic polymers, agarose, cellulose, cross-linked dextran,
etc., which have a defined pore size. These support materials should be sterilizable, chemically
inert, stable, highly porous, and hydrophilic, containing some ionic groups.

Mechanical rigidity is particularly important in order to maintain good flow rates. The initial choice
of stationary phase material is also a key factor, as some may interact with the target product,
e.g., carbohydrate-based matrices interact with glycoproteins. Solute molecules below the
exclusion size of the support material pass in and out of the beads. Molecules above the size
pass only around the outside of the beads through the interstitial spaces and the apparent volume
of the column is smaller for these larger excluded molecules. As a result, they flow faster down
the column, separating from smaller molecules and eluting first. Smaller molecules able to enter
the pores are then eluted in decreasing order of size.

Ion Exchange Chromatography (IEC)

Ion-Exchange Chromatography involves the selective adsorption of ions or electrically charged
compounds onto ion-exchange resin particles by electrostatic forces. The matrix material is often
based on cellulose substituted with various charged groups, either cations or anions. A commonly
used example is the anion-exchange resin diethylaminoethyl (DEAE) cellulose.

Proteins possess positive, negative or no charge depending on their isoelectric point (pI) and the
pH of the surrounding buffer. If the pH of the buffer is below the pI, a protein has an overall positive
charge, whereas a buffer at the pI results in the protein having no charge and pHs above its pI
produce an overall negative charge. For example, if a protein has a pI of 4.2, at pH4.2 this protein
is uncharged and will not bind to either positively or negatively charged resins. When the pH is
raised to pH7.0 the protein is negatively charged and binds to positively charged resins.

In conditions below pH4.0 the protein is positively charged and binds to negatively charged resins.
If the protein is in an anionic state able to adsorb to DEAE cellulose, for example, any
contaminants pass through the column. The product can then be desorbed as a purified fraction
by altering the ionic strength of the buffer, often by using a gradient of increasing phosphate buffer
concentration.

This is based on the adsorption of solute molecules onto solid particles, such as alumina and
silica gel, by weak van der Waals forces and steric interactions, and separates molecules
according to their charge. Since the charge of a molecule depends on the pH, a combination of
varying pH and ion exchange chromatography can prove very effective at purifying proteins.

83
In a column cationic matrix (negatively charged) when positively charged components are loaded,
more positively charged components bind strongly to the matrix whereas negative charged and
neutral components eluted without retention. The factors influencing the binding are net charge,
anisotropy (charge distribution on protein surface), ionic strength, pH of solvent, nature of ions
and other additives.
pH of the medium and solvent are important since they determine the effective charge on both
proteins and matrix. There are two types of ion exchangers.

Cation exchangers have acidic groups with net negative charge and positively charged
exchangeable ions.

Anion exchangers have basic groups with net positive charge and negatively charged
exchangeable ions. Charges on ion exchanger-based pH of solvent.

Figure: Ion exchange column

Hydrophobic Chromatography (HC)

This is based on hydrophobic interactions between solute molecules (e.g., proteins) and
functional groups (e.g., alkyl residues) on support particles. HC uses the different affinity that
different molecules have for hydrophobic materials, i.e., for materials that are hydrophobic (‘water
haters’), like plastics [as opposed to hydrophilic (‘water lover’) materials, like paper].

Hydrophobic chromatography relies on hydrophobic interaction between hydrophobic regions or

domains of a solute protein and hydrophobic functional groups of the support particles. These
supports are often agarose substituted with octyl or phenyl groups. Elution from the column is
usually achieved by altering the ionic strength, changing the pH or increasing the concentration
of chaotropic ions, e.g., thiocyanate. This technique provides good resolution and, like ion
exchange chromatography, has a very high capacity as it is not limited by sample volume.

Popular versions of chromatographic separation methods are FPLC and HPLC, which have been
scaled up from laboratory tools to production methods in some cases.

High Performance Liquid Chromatography (HPLC)

High-performance liquid chromatography (HPLC) was originally developed for the separation
of organic molecules in non-aqueous solvents but is now used for proteins in aqueous solution.
This method uses densely packed columns containing very small rigid particles, 5 to 50µm
diameter, of silica or a cross-linked polymer. Consequently, high pressures are required. The
method is fast and gives high resolution of solute molecules. Equipment for use in large-scale
operations is now available.

84
HPLC is based on the general principle of chromatography, with the only difference being high
liquid pressure applied to the packed column. The mixture is pumped through the chromatography
‘column’ at extremely high pressures, ensuring very precise separation in a short time due to the
high-pressure liquid (high liquid flow rate) and dense column packing. HPLC provides fast and
high resolution of solute molecules.

Fast Protein Liquid Chromatography (FPLC)

FPLC is a more specialised technique for separating proteins, which, because many
biotechnological products are proteins, has found widespread use. The pressures used in FPLC
are much lower than in HPLC, so the apparatus can be substantially cheaper.

Metal Chelate Chromatography

Metal chelate chromatography utilises a matrix with attached metal ions, e.g., agarose
containing calcium, copper, or magnesium ions. The protein to be purified must have an affinity
for this ion and binds to it by forming coordination complexes with groups such as the imidazole
of histidine residues. Bound proteins are then eluted using solutions of free metal-binding
ligands, e.g., amino acids.

Dialysis and Electrodialysis

These membrane separation techniques are primarily used for the removal of low molecular
weight solutes and inorganic ions from a solution. The membranes involved are size selective
with specific molecular weight cut-offs. Low molecular weight solutes move across the membrane
by osmosis from a region of high concentration to one of low concentration.

Electrodialysis methods separate charged molecules from a solution by the application of a direct
electrical current carried by mobile counter-ions. Membranes used contain ion-exchange groups
and have a fixed charge; e.g. positively charged membranes allow the passage of anions and
repel cations. They are essentially ion-exchange resins in sheet form and have also been used
for the desalination of water.

APPROPRIATE METHODS FOR LARGE SCALE PURIFICATION

One of the central parts of the downstream processing of a fermentation product is purification.
Large-scale purification methods are used to take a crude fermentation supernatant or cell
homogenate and isolate the product from it in a fairly pure form. Industrial enzymes are often sold
in this semi-pure form as bulk product. If they need to be really pure, then they have to go through
a second purification step. Purification of the cells from a culture is usually called harvesting and
relies on rather different methods.

The economics of producing almost anything are usually governed by the cost of purification and
distribution. For expensive biotechnology products, purification is the largest of these costs.
Growing E. coli in a tube or broth is very cheap. Purifying a recombinant drug from it needs time,
skill, and materials, all of which cost money. This is especially relevant in biotechnology, where
the products are also governed by strict regulations that demand not merely purity but proof of
purity. In general, then, the less purifying that is required the better. This usually means that the
more concentrated the material is to start with the cheaper it will be. For very dilute materials, only

85
a very high selling price can justify the extreme cost of extracting the pure material from the
starting bulk. Pharmaceutical proteins are an example of such materials.

There are a range of purification methods that are cheap enough to use on large volumes of
material. These include:

1. Precipitation.
The first step in the purification of intracellular proteins after cell disruption is usually the
precipitation of proteins. Proteins in a fermentation broth (before or after cell lysis) can be
separated from other components by precipitation using certain salts. Examples include
streptomycin sulfate and ammonium sulfate. The two major methods used for protein precipitation
are:
(i) Salting-out by adding inorganic salts such as (NH4)2SO4 at high ionic strength.
(ii) Solubility reduction at low temperatures by adding organic solvents (T< -5°C).

(i) “Salting Out”

Adding salt so that a particular group of proteins precipitates from solution is termed salt
precipitation, also sometimes called “salting out”. Salting-out of proteins is achieved by
increasing the ionic strength of a protein-containing solution by adding salts such as (NH4)2SO4
or Na2SO4. The added ions interact with water more strongly, causing protein molecules to
precipitate.

Simply adding water to the precipitate usually makes them dissolve again. Ammonium sulfate
precipitation (adding concentrated ammonium sulfate to a protein solution) is often used in
preparing proteins at small and medium scale. A low concentration of ammonium sulfate is added
to precipitate some unwanted proteins, the precipitate removed, then a higher concentration
added to precipitate a partially purified mixture containing the protein that is wanted. Such a
preparation is sometimes called an “ammonium sulfate cut”.

(ii) Solvent Precipitation

Organic solvent addition at low temperatures (T< -5°C) causes the precipitation of proteins by
reducing the dielectric constant of the solution. A reduction in the dielectric constant of a
solution results in stronger electrostatic forces between the protein molecules and
facilitates protein precipitation. The addition of solvents also reduces protein-water molecule
interactions and therefore decreases protein solubility. Solvents may cause protein denaturation.
However, denaturation of proteins in the salting-out is less likely.

Solvent precipitation can be used with salt addition, pH adjustment, and low temperature to
improve precipitation. Among other protein precipitation methods, the following methods are the
most widely used.

(iii) Isoelectric Precipitation

This is the precipitation of proteins at their isoelectric point, which is the pH at which proteins have
no net charge. Most proteins are quite insoluble at a particular pH (their isoelectric point or pI). If
one adds acid or alkali until the pH of the solution is at this isoelectric point, then those proteins
will precipitate. Again, adding water usually redissolves the precipitate. The isoelectric point of a
protein is defined as:
pI = ½(pK1 - pK2).

86
When pH = pI, protein becomes free of charge and precipitates. This method is effective for
proteins with high surface hydrophobicity (i.e., a nonpolar surface). The method is inexpensive,
but low pH values may result in the denaturation of proteins.

The use of ionic polyelectrolytes, such as ionic polysaccharides, polyphosphates and

polyacrylic acids, changes the ionic strength of the media and causes protein precipitation.
However, polyelectrolytes may cause protein denaturation and structural damage.
The use of nonionic polymers, such as dextrans and polyethylene glycol, reduces the amount of
water available to interact with protein molecules and results in protein precipitation. High
concentrations of polymers are most effective for low-MW proteins, and nonionic polymers do not
interfere with further protein recovery.

2. Heat Denaturation.
This is simple and effective if the protein is heat stable (thermostable): one just heats up the
mixture and most of the proteins denature, and so ‘coagulate’ and settle out of solution. The
thermostable desired product remains in solution. This only works with some proteins. It can also
be used under some conditions to separate proteins from non-proteins (e.g., metabolites).

3. Dialysis
Dialysis is a membrane separation operation used for the removal of low-MW solutes such as
organic acids (100<MW<500) and inorganic ions (10<MW<100) from a solution. A well-known
example is the use of dialysis membranes to remove urea (MW = 60) from urine in artificial kidney
(dialysis) devices. The membrane is selective and separates two phases containing low-MW and
high-MW solutions. Since the MW cutoff of a dialysis membrane is very small, low-MW solutes
move from a high- to low-concentration region.

To purify a protein is put on one side of a semi-permeable membrane, and water or salt solution
on the other. A semi-permeable membrane is one that lets through small molecules, but not large
ones. So, salts, impurities, etc., in a protein solution are let out, and the protein remains in. The
process can be accelerated by an electric current in electrodialysis, which drives the ions in
one direction. Electrodialysis is used to deionise (i.e., remove ions from) several industrial-scale
protein materials, including whey.

87
 Figure: Dialysis Tubing Set-Up (In Fridge)

Reverse Osmosis (RO)

For fermentation broths, osmosis is the transport of water molecules from a high- to a low-
concentration region (i.e., from pure water phase to a salt-containing aqueous phase) when these
two phases are separated by a selective membrane. The water passes the membrane easily,
while the salt does not. At equilibrium, the chemical potential of water must be the same on both
sides of the membrane. As the water passes into the salt solution, its pressure increases. This
pressure is called the osmotic pressure.

In reverse osmosis (RO), a pressure is applied onto a salt-containing phase, which drives
water (solvent) molecules from a low to a high-concentration region and results in the
concentration of solute (salt) molecules on one side of the membrane. The pressure
required to move solvent from a low- to high-concentration phase is equal to or slightly larger than
the osmotic pressure.

In some reverse osmosis applications, membranes may allow the passage of solute molecules
along with solvent. Solute transport can be by diffusion or convection. A reflection coefficient
(σ) for each solute can be defined as the fraction of solute molecules retained on one side of
the membrane in the presence of a solvent flux. Therefore, for σ = 0, complete solute
passage is obtained, and for σ = 1, no solute passage is achieved (i.e., perfect reflection).

88
 Figure: Reverse osmosis process

The magnitude of the required pressure varies depending on the concentration of solutions.
Pressures on the order of 30 to 40 atmospheres are required for a 0.6 M salt solution. The
salt level in fermentation fluids may be much higher than 0.6 M, requiring high pressures of RO
separations. The applications of RO in bioseparation are limited since the method requires high
pressures and is based on solvent removal. RO membranes are usually used for dewatering and
concentration purposes, but not for protein separation.

Another problem with RO membranes is the deposition of solute molecules on membrane

surfaces, resulting in large resistance for solvent flow. This phenomenon is known as
concentration polarisation and can be overcome by increasing the degree of turbulence on the
membrane surface. The osmotic pressure for multicomponent systems is equal to the sum of the
individual osmotic pressure.

4. Supercritical Fluid Extraction.

Supercritical fluids can also be used to extract biomolecules. Supercritical fluids are gasses that
have been compressed to the density of liquids above the temperature at which they can form
liquids. Their ability to dissolve chemicals depends strongly on their temperature and pressure,
and so they can be adjusted to dissolve the product desired.

Supercritical fluid extraction involves the dissolution power of super critical fluids i.e., fluids above
their critical temperature and pressure. Critical temperature is defined as the temperature above
which a distinct liquid phase cannot exist regardless of pressure. The vapour pressure of the
substance at its critical temperature is called the critical pressure. Alternately, pressure and
temperature required to liquefy a gas are critical temperature and pressure. At temperature and
pressure above but close to the critical point a substance exists as a supercritical fluid. For
example, Carbon dioxide, NO, SO2 are used in extraction of β-carotene, vanilla, vegetable oil,
etc.

Supercritical carbon dioxide has been used to extract the caffeine from coffee for many years: its
properties can be ‘tuned’ to dissolve the caffeine and leave virtually all other components of the
coffee in place. It is also being explored as a method of extracting and partially purifying many
other biomolecules, including proteins; some proteins can even function as enzymes in
supercritical fluids.

89
Figure: Supercritical fluid extraction process. CO2 is fluidised and used for extraction.

Distillation
Distillation is used to recover fuel alcohol, acetone, and other solvents from fermentation media,
and for the preparation of potable spirits. Batch distillation in pot stills continues to be used for the
production of some whiskies, but for most other purposes continuous distillation is the method of
choice. With ethanol, for example, the continuous system produces a product with a maximum
ethanol concentration of 96.5% (v/v). This azeotropic mixture is the highest concentration that can
be achieved from aqueous ethanol unless a dehydration step is introduced using a water entrainer
such as benzene or cyclohexane.

Some continuous stills may be in the form of four or five separate columns, but the Coffey-type
still comprises just two columns, the ‘rectifier’ and ‘analyser’, each containing a stack of 30 to
32 perforated plates (Figure).

Figure: A Coffey-type still (only six of the 30 column sections shown)

90
Incoming fermentation broth is heated, as it passes down a coiled pipe within the rectifier column,
by the ascending hot vapour produced by the analyser. The now hot broth is released into a
trough at the top of the analyser column and as it falls down the column it is heated by steam. Hot
vapours generated are then conveyed from the top of the analyser column to the bottom of the
rectifier column. As it passes upwards it is condensed on the coils carrying incoming broth. There
is a temperature gradient in the rectifier column and each volatile compound condenses at its
appropriate level, from where the fraction is collected.

• Liquid-Liquid Separation.

• Two-Phase Aqueous Extraction.

• Polymer Precipitation.

FINISHING STEPS FOR PURIFICATION

The major finishing steps in fermentation product purification are crystallisation and drying.

A. Crystallisation
It is a process where solid particles of specified size and shape are formed from a homogenous
phase. It is the most common method of final purification of a desired product because crystals
obtained are usually of exceptional purity and free from even closely related impurities.

91
Crystallisation can be brought about by:
(i) Cooling the solution with negligible evaporation,
(ii) Evaporation of the solvent with little or no cooling as in evaporator crystalliser, or
(iii) Combined cooling and evaporation as in adiabatic or vacuum crystallisers.

Crystallisation is usually the last step in producing highly purified products such as antibiotics.
Crystallisation operates at low temperatures, which minimise thermal degradation of heat-
sensitive materials. Operations are conducted at high concentrations and, therefore, unit costs
are low and separation factors are high. The determination of optimal crystallisation conditions is
a matter of empirical experimentation since phase diagrams and pertinent kinetics for specific
systems are usually not available.

High-purity crystals are recovered by using batch Nutsche-type filters or centrifugal filters. The
nature and size of crystals affect the centrifugation and washing rates; therefore, crystallisation
and recovery steps are interrelated and should be considered together in the optimisation of
crystallisation operations. After washing, the crystals are discharged for drying.

B. Drying
The removal of solvent from purified wet product (crystal or dissolved solute) is usually achieved
by drying. In selecting drying conditions, the physical properties of the product, its heat sensitivity,
and the desirable final moisture content must be considered. The parameters affecting drying can
be classified in four categories:
• Physical properties of the solid-liquid system;
• Intrinsic properties of the solute;
• Conditions of the drying environment;
• Heat-transfer parameters

Drying with evaporators removes the solvent, mainly water, from the desired product by heat
conduction. A typical evaporator has two principal functions to exchange heat and to separate
vapour that is formed from liquid. It has got three principal functional sections viz., the heat
exchanger, the evaporating sections where the liquid boils and evaporates and the separator in
which the vapour leaves the liquid and passes to the condenser.

Hot air drying uses a moving stream of hot air in contact with the product to be dried. Heat is
supplied to the product mainly by convection. Examples of such drying are Kiln driers, tray or
compartment driers, etc.

The major types of driers used for drying fermentation product are the following:

1. Vacuum Tray Drier consists of heated shelves in a single chamber and is usually used
in pharmaceutical products. This is a good method for small batches of expensive
materials where product loss and heat damage must be minimised.

2. Freeze Drying (Lyophilisation) is a method where water is removed by sublimation

(from solid ice to vapour) from the frozen solution. Freeze drying is the most complex and
expensive form of drying; its use is restricted to delicate, heat sensitive materials. In the
freeze-drying process, material is frozen by exposure to cold air followed by sublimation
of ice in vacuum from the frozen state to produce a dried product. The freezing can be
accomplished either outside or inside the vacuum chamber prior to drying. This method is
used for antibiotic, enzyme solutions and bacterial suspensions. In batch freeze driers, a

92
 vacuum cabinet, a vacuum system, and a heating system are present. Refrigerated
condensers backed by a mechanical pumping system are commonly used, commercially.
The pumping system is essential to pump the frozen materials down the cabinet pressure
initially in a short time to prevent melting of the frozen products. Heat may be supplied by
conduction or radiation or from a microwave radiator. The removal of the water by
sublimation results in a porous structured product which retains shape and size.

Figure: Photographic Representation of (a) Laboratory-Scale Freeze-Drier.

3. Rotary-Drum Driers are not good for crystal solutions. Water is removed by heat
conduction over a thin film of solution on the steam-heated surface of the rotating drum.
The dried product is scrapped from the drum with the aid of a knife at the discharge point.

Figure: Drum Drier

4. Spray Dryers employs atomisation and spraying of product solution into a heated
chamber through a nozzle. Hot gas inside the chamber provides the necessary heat for
evapouration of the liquid. Dried particles are separated from hot gases using cyclones.
Spray dryers are expensive to purchase but are the preferred method for heat-sensitive
materials.

93
 Figure: Conventional Spray Drying

5. Pneumatic Conveyor Driers use a hot air stream to suspend and transport particles.
The retention time of a particle in the gas stream is short, usually a few seconds. Such
systems work well when surface drying is critical, but do not provide sufficient exposure
times to dry large porous particles where water removal is diffusion controlled. Pneumatic
conveyor systems are well suited for heat-sensitive and easily oxidised materials.

Inclusion Bodies and the Role of Genetic Engineering in Downstream

Processing
Many recombinant proteins are formed as inclusion bodies, which are insoluble inactive
aggregates of over-expressed polypeptides. They arise due to the accumulation of partially folded
nascent polypeptides that have exposed hydrophobic surfaces. Resulting hydrophobic
interactions cause them to form insoluble aggregates. This is often fortuitous because it provides
a concentrated form of the protein, as the aggregates may contain over 50% of target protein.

Once the producer cells are broken open and the large cell debris is removed, the inclusion bodies
are easily recovered from the cell-free extracts by low-speed centrifugation at 5000 to 20000xg
for 15 to 60 min. The recovered protein must then be solubilized; however, at this point they are
not soluble in ordinary aqueous buffers. Their solubilisation requires concentrated solutions of
chaotrophic agents (protein denaturants), such as 5 to 10 mmol/L urea and 4 8 mmol/L guanidine-
HCl, or detergents. Where necessary, disulphide bonds are broken by the use of mercaptoethanol
or dithiothreitol.

Following dissolution, the denaturants and reducing agents are removed by dialysis, diafiltration
or gel filtration. This begins protein refolding (renaturation) to produce the functional conformation
and restore biological activity. Refolding may be achieved by the slow removal of denaturants,
along with additional treatments using specific buffers and various temperature regimes. For
some recombinant proteins disulphide bonds must also be reformed, which is often accomplished
by air oxidation or thiol-disulphide exchange reactions.

94
DSP can be aided by the introduction of specific aids at the upstream stages. Organisms can be
modified to suppress the production of byproducts and enzymes that may interfere with DSP
operations or degrade the target product. Recombinant protein products may be designed to be
excreted from the cell, and cell breakage and release of intracellular products can also be
assisted.

The cell envelope of an organism may be modified so that it is permeable or is at some point
induced to become permeable to the product. For example, recombinant E. coli are often lysed
by co-expressing lysozyme (under appropriate inducer control). Following lysozyme induction at
the end of the fermentation, cell breakage is promoted by a freeze-thaw step.

Recombinant protein products may be engineered with a high affinity for certain separation
matrices or joined to another molecule with desirable separation characteristics. A now well-
established technique is the construction of fusion (hybrid or chimeric) proteins (Table). For
example, the structural gene for the target recombinant peptide/protein can be fused/tagged
with additional DNA encoding a natural or synthetic polypeptide at the 5̍̍ or 3 end of the gene.
The tag can be a complete enzyme, such as ß-galactosidase, chloramphenicol
acetyltransferase or glutathione-S-transferase.

Table: Example of Fusion Proteins to Aid the Purification of Recombinant

Proteins

These tags may be employed to protect short peptides that are often rapidly degraded, or aid in
downstream processing and the specific assaying of the peptide/protein product. Tags may be
used to allow affinity, ion-exchange, hydrophobic covalent or metal-chelate separation of the
peptide/protein. Some methods are suitable for both secreted proteins and those forming
inclusion bodies. An enzyme or chemical cleavage site can also be included in the linker between
the target protein and its tag, in order to facilitate later removal of the tag. Enzyme cleavage sites
include those for endopeptidases, enterokinase and thrombin, whereas chemical sites are those
targeted by acid, cyanogen bromide or hydroxylamine.

Tags coding for a specific dipeptide or a sequence of amino residues, such as arginine,
phenylalanine, or cysteine, may be added at the C-terminal end forming a ‘tail’ on the target
protein. Polyarginine tails, for example, result in a particularly basic protein that unlike most other
cellular proteins binds to cation-exchange resins.

Once recovered from the resin, the polyarginine tails can be removed by passing the protein
through a column of an immobilised exopeptidase, e.g., carboxpeptidase A.

95
This mode of purification has been used for the preparation of urogastrone and interferon-g.
Alternatively, a gene for a protein that binds to IgG, such as staphylococcal protein A (SPA), can
be fused to the target protein gene. The product with the SPA tag can then be purified by
immunoaffinity chromatography using IgG as the ligand.

QC (Quality Control) and QA (Quality Assurance)

QC is quality control, testing a product to see that it meets the required quality. QA is quality
assurance, making sure that what is being used to make the product is appropriate for making a
product of the required quality. Both are general concerns in all industries. QC and QA are
especially important to a wide range of biotechnological processes, especially ones producing
pharmaceuticals or foods. Some of the QA/QC procedures that are specific to biologically based
products:
• Sterility (are there any living organisms in the product?) and bioburden (how many
organisms are there?), both performed using conventional microbiology.

• Endotoxin (or pyrogen) testing, done using the Limulus Amoebocyte Lysate (LAL)
test. Endotoxins are lipopolysaccharides from bacterial cell walls, which cause extreme
immune system responses, sometimes fatal.

GLP and GMP processes are meant to assure quality in product development and product
production, respectively. In fact, they assure that the product is always the same, not that it id
always of high quality.
Products produced by biotechnology are not usually attacked for their technical quality, but for
their deliberate performance: for example, no one disputes that BST is pure and highly effective,
but people argue strenuously about whether it should be used on cattle at all.

Lecturer: Prof. Hilary Domakyaara Zakpaa

References
1. Adkesson, D.M., Alsop, A.W., Ames, T.T. et al. (2004) Purification of biological‐produced
1,3‐propanediol. International Patent WO2004101479.
2. Agrawal,A.K. and P. Parihar. 2009. Industrial Microbiology: Fundamentals and
Applications. AbeBooks Seller,India.
3. Andersson, L., Yang, S., Neubauer, P. and Enfors, S.‐O. (1996) Impact of plasmid
presence and induction on cellular responses in fed‐batch cultures of Escherichia coli. J.
Biotechnol., 46, 255–263.
4. Angenent, L.T. and Wrenn, B.A. (2008) Optimising mixed‐culture bioprocessing to
convert wastes into bioenergy. in Bioenergy, A.Demain, C.S.Harwood, and J.D.Wall
(eds) Washington,DC: American Society of Microbiology, 179–194.
5. Ansorge, M.B. and Kula, M. (2000) Production of recombinant L‐leucine dehydrogenase
from Bacillus cereus in pilot scale using the runaway replication system E. coli[pIET98].
Biotechnol Bioeng, 68, 557–562.
6. Araujo, R., Casal, M. and Cavaco‐Paulo, A. (2008) Application of enzymes for textiles
fibers processing. Biocatal. Biotechnol., 26, 332–349.
7. Atmar, R.L., Metcalf, T.G., Neill, F.H. and Estes, M.K. (1993) Detection of enteric viruses
in Oysters by using the Polymerase Chain reaction. Appl. Environ. Microbiol., 59, 631–
635.

96
8. Atsonios, K., Kougioumtzis, M.‐A.D., Panopoulos, K. and Kakaras, E. (2015) Alternative
thermochemical routes for aviation biofuels via alcohols synthesis: Process
modeling,techno economic assessment and comparison. Appl. Energy, 138, 346–366.
9. Baer, Z.C., Bormann, S., Sreekumar, S. et al. (2016) Co‐production of acetone and
ethanol with molar ratio control enables production of improved gasoline or jet fuel
blends. Biotechnol. Bioeng.
10. Bannari, R., Bannari, A., Vermette, P. and Proulx, P. (2012) A model for cellulase
production from Trichoderma reesei in an airlift reactor. Biotechnol. Bioeng., 109, 2025–
2038.
11. Becker, F.U., Grund, G., Orschel, M. et al. (2012) Cells and Method for Producing
Acetone. US2A1. Washington, DC: U.S. Patent and Trademark Office.
12. Bernardez-Clark, E. D. (1998) Refolding of recombinant proteins. Current Opinion in
Biotechnology 9, 157–163.
13. Birol, G., Ündey, C. and Cinar, A. (2002a) A modular simulation package for fed‐batch
fermentation: penicillin production. Comput. Chem. Eng., 26, 1553–1565.
14. Birol, G., Ündey, C., Parulekar, S.J. and Çınar A. (2002b) A morphologically structured
model for penicillin production. Biotechnol. Bioeng., 77, 538–552.
15. Bormann, S., Baer, Z.C., Sreekumar, S. et al. (2014) Engineering Clostridium
acetobutylicum for production of kerosene and diesel blend stock precursors. Metab.
Eng., 25, 124–130.
16. Buckland, B.C., Gbewonyo, K., Dimasi, D. et al. (1988) Improved performance in viscous
mycelial fermentations by agitator retrofitting. Biotechnol. Bioeng., 31, 737–742.
17. Budzik, J.M. (2003) Phage isolation and investigation. Dartmouth Undergraduate J. Sc.,
3, 37 43.
18. Bull, A., Marrs, B. and Kurane, R. (1998) Biotechnology for clean industrial products and
processes. Towards industrial sustainability. OECD Publications, 7–139.
19. Burk, M.J. and Vandien, S. 2016. Biotechnology for chemical production: challenges and
opportunities. Trends Biotechnol., 34, 187–190.
20. Bylund, F., Castan, A., Mikkola, R. et al. (2000) Influence of scale‐up on the quality of
recombinant human growth hormone. Biotechnol. Bioengin., 69, 119–128.
21. Bylund, F., Collet, E., Enfors, E.O. and Larsson G. (1998) Substrate gradient formation
in the large scale lowers cell yield and increases byproduct formation. Bioproc. Eng., 18,
171–180.
22. Bylund, F., Guillard, F., Enfors S.O. et al. (1999) Scale down of recombinant production:
a comparative study of scaling performance. Bioproc. Engin., 20, 377–289.
23. Bylund, F.,E. Collet, S.O. Enfors and G. Larsson.1998. Substrate gradient formation in
the largescale bioreactor lowers cell yield and increases by-product formation. Biopro.
Eng.,18:171-180.
24. Calam, C.T. 1976. Starting investigational and production cultures. Pro. biochem.,11(3):
7-12. Laboratory Scale to Pilot Scale and to near Commercial Scale for Paste-Glue
Production. J. of Eng. and Tech. Sci.,45(1): 9-24.
25. Callanan, M. and Klaenhammer, T.R. (2002) Bacteriophages in industry. Encyc. Of Life
Sc., 1–8.
26. Cascaval, D., Oniscu, C. and Galaction, A. (2003) Rheology of fermentation broths 2.
Influence of the rheological behavior on biotechnological processes. Rev. Roum. Chim.,
48, 339–356.
27. Castan, A. and Enfors, S.O. (2001) Formate accumulation due to DNA release in
aerobic cultivations of Escherichia coli. Biotechnol. Bioengin., 77, 324–328.
28. Cervera, A.E., Petersen, N., Eliasson Lantz, A. (2009) Application of near‐infrared
spectroscopy for monitoring and control of cell culture and fermentation. Biotechnol.
Prog., 25, 1561–1581.

97
29. Cho, C., Choi, S.Y., Luo, Z. W., and Lee, S.Y. (2014) Recent advances in microbial
production of fuels and chemicals using tools and strategies of systems metabolic
engineering. Biotechnol. Adv., 33, 1455–1466. Council, N.R. (2015) Industrialization of
Biology: A Roadmap to Accelerate the Advanced Manufacturing of Chemicals.
Washington,DC: National Academies Press.
30. Christi, Y. & Moo-Young, M. (1986) Disruption of microbial cells for intracellular products.
Enzyme and Microbial Technology 8, 194–204.
31. Christi, Y. & Moo-Young, M. (1990) Large-scale protein separations: engineering
aspects of
32. chromatography. Biotechnological Advances 8, 699–708.
33. Das, R.K. and Brar, S.K. (2014) Enhanced Fumaric Acid Production from Brewery
Wastewater and Insight into the Morphology of Rhizopus oryzae 1526. Appl. biochem.
biotechnol., 172.
34. Datar, R. V., Cartwright, T. & Rosen, C. G. (1993) Review: Process economics of animal
cell culture and bacterial fermentations: a case study analysis of tissue plasminogen
activator. Bio/Technology 11, 349–357.
35. Daugulis, A. J. (1994) Integrated fermentation and recovery processes. Current Opinion
in Biotechnology 5, 192–195.
36. Demain, A. L. 1972. Cellular and environmental factors affecting the synthesis and
excretion of metabolites. J. Appl. Chem. Biotechnol. 22: 345-362.
37. Douma, R.D., Deshmukh, A.T., De Jonge, L.P. et al. (2012) Novel insights in transport
mechanisms and kinetics of phenylacetic acid and penicillin‐G in Penicillium
chrysogenum. Biotechnol. Prog., 28, 337–348.
38. Douma, R.D., Verheijen, P.J., De Laat, W.T. (2010b) Dynamic gene expression
regulation model for growth and penicillin production in Penicillium chrysogenum.
Biotechnol. Bioeng., 106, 608–618.
39. Duan, H., Yamada, Y. and Sato, S. (2015) Efficient production of 1,3‐butadiene in the
catalytic dehydration of 2,3‐butanediol. Appl. Catal. A Gen., 491, 163–169.
40. Dunlop, M.J., Dossani, Z.Y., Szmidt, H.L. et al. (2011) Engineering microbial biofuel
tolerance and export using efflux pumps. Mol. Syst. Biol., 7, 487.
41. Dunn, I. J., and J. R. Mor. 1975. Variable volume continuous cultivation. Biotechnol.
Bioeng. 17: 1805-1822.
42. Edwards, V. H., R. C. Ko, and S. A. Balogh. 1972. Dynamics and control of continuous
microbial propagation to subject inhibition Bioechnol. Bioeng. 15: 939-944.
43. Elander, R.P. (2003) Industrial production of β‐lactam antibiotics. Appl. Microbiol.
Biotechnol., 61, 385–392.
44. Elsworth, R., Miller, G., Whitaker, A. et al. (1968) Production of E. coli as a source of
nucleic acids. J. Appl. Chem., 17, 157–166.
45. Enfors, S.O., Jahic, M., Rozkov, A. et al. (2001) Physiological responses to mixing in
large scale bioreactors. J. Biotechnol., 85, 175–185.
46. Fan, Y.F. (2008) Breeding of yeast high producing ethanol and fermentation condition
with dynamics. Fujian Normal University: Fujian, 128-135.
47. Forberg, C. and Haggstrom, L. (1987) Effects of cultural conditions on the production of
phenylalanine from a plasmid‐harboring Escherichia coli strain. Appl. Microbiol.
Biotechnol., 26, 136–140.
48. Foster, D. (1995) Optimizing recombinant product recovery through improvements in cell
disruption technologies. Current Opinion in Biotechnology 6, 523–526.
49. Fraser, P.D., Roemer, S., Kiano, J.W. et al. (2001) Elevation of carotenoids in tomato by
genetic manipulation. J. Sci. Food. Agric., 81, 822–827.
50. Friehs, K. (2004) Plasmid copy number and plasmid stability. Adv. Biochem. Eng.
Biotechnol., 86, 47–82.

98
51. Gabelle, J.C., Augier, F., Carvalho, A. (2012) Effect of tank size on kLa and mixing time
in aerated stirred reactors with non‐Newtonian fluids. Can. J. Chem. Eng., 89, 1139–
1152.
52. George, K.W., Alonso‐Gutierrez, J., Keasling, J.D., and Lee, T.S. (2015) Isoprenoid
drugs, biofuels, and Chemicals‐Artemisinin,farnesene, and beyond. In Biotechnology of
Isoprenoids., eds J. Schrader and J. Bohlmann (Cham: Springer International
Publishing), 355–389.
53. George, S., Larsson, G., and Enfors, S.‐O. (1993) A scale‐down two‐compartment
reactor with controlled substrate oscillations: metabolic response of Saccharomyces
cerevisiae. Bioproc. Eng., 9, 249–257.
54. German, J. and Dillard, C. (2006) Composition, structure and absorption of milk lipids: A
source of energy, fat soluble nutrient and bioactive molecules. Food Sc. Nutr., 46, 57–
92.
55. Gernaey, K.V., Bolic, A., and Svanholm, B. (2012) PAT tools for fermentation processes.
Chem. Today, 30, 38–43.
56. Gibbs, P.A., Seviour, R.J., and Schmid, F. (2000) Growth of filamentous fungi in
submerged culture: Problems and possible solutions. Crit. Rev. Biotechnol., 20, 17–48.
57. Goeddel, D. V., E. Yelverton et al. 1980. Human leucocyte interferon produced by E. coli
is biologically active. Nature (London) 287: 411-416.
58. Goldberg, I., and Z. Er-el. 1981. The chemostat-an efficient technique for medium
optimisation. Process Biochem. (Oct./Nov.), p 2-8.
59. Grimm, L.H., Kelly, S., Hengstler, J. et al. (2004) Kinetic studies on the aggregation of
Aspergillus niger conidia. Biotechnol. Bioeng., 87, 213–218.
60. Harrison, S. T. L. (1991) Bacterial cell disruption: a key unit operation in the recovery of
intracellular proteins. Biotechnological Advances 9, 217–240.
61. Harrison, S. T. L., Dennis, J. S. & Chase, H. A. (1991) Combined chemical and
mechanical processes for the disruption of bacteria. Bioseparation 2, 95–105.
62. Harvey, B.G. and Meylemans, H.A. (2011) The role of butanol in the development of
sustainable fuel technologies. J. Chem. Technol. Biotechnol., 86, 2–9.
63. Henriksen, C.M., Christensen, L.H., Nielsen, J., and Villadsen J. 1996. Growth
energetics and metabolic fluxes in continuous cultures of Penicillium chrysogenum. J.
Biotechnol 45, 149–164.
64. Heveling, J., Nicolaides, C.P., and Scurrell, M.S. (1998) Catalysts and conditions for the
highly efficient,selective and stable heterogeneous oligomerisation of ethylene. Appl.
Catal. A Gen., 173, 1–9.
65. Holder, K.K. and Bull, J.J. (2001) Profiles of adaptation in two similar viruses. Genetics.,
159, 1393–1404.
66. Hood, E.E. (2002) From green plants to industrial enzymes. Enzyme. Microb. Technol.,
30, 279–283.
67. Hopkins, T. R. 1981. Feed-on-demand control of fermentation by cyclic changes in
dissolved oxygen tension. Biotechnol. Bioeng. 23: 2137-2143.
68. Hu, W. S., and A. L. Demain. 1979. Regulation of antibiotic biosynthesis by utilizable
carbon sources. Process Biochem. (Aug./Sept.) p 2-5.
69. Johnson, I.S. (1983) Human insulin from recombinant DNA technology. Science, 219,
632–637.
70. Jones, D.T., Shirley, M., Wu, X., and Keis, S. (2000) Bacteriophage infections in the
industrial acetone butanol (AB) fermentation process. J. Mol. Microbiol. And Biotechnol.,
2, 21–26.
71. Junker, B.H. (2004) Scale‐up methodologies for Escherichia coli and yeast fermentation
processes. J. Biosci. Bioeng., 97, 347–364.

99
72. Jüsten, P., Paul, G.C., Nienow, A.W., and Thomas, C.R. (1996) Dependence of mycelial
morphology on impeller type and agitation intensity. Biotechnol. Bioeng., 52, 672–684.
73. Kardos, N. and Demain, A.L. (2011) Penicillin: the medicine with the greatest impact on
therapeutic outcomes. Appl. Microbiol. Biotechnol., 92, 677–687.
74. Kirk, O., Borchert, T.V., and Fuglsang, C.C. (2002) Industrial enzyme applications. Curr.
Opin. Biotechnol., 13, 345–351.
75. Koizumi, J.‐I., Monden, Y., and Aiba, S. (1985) Effects of the temperature and the
dilution rate on the copy number of recombinant plasmid in continuous culture of Bacillus
staerothermophilus (pLP11). Biotechnol. Bioeng., 27, 721–728.
76. Kokjohn, T.A., Schrader, J.J., and Schrader, H.S. (2005) http://www.isb.vt.edu/brarg/
brasym94/kokjohn.html.
77. Köpke, M. and Havill, A. (2014) Lanza Tech’s route to bio‐butadiene. Catal. Rev., 27, 7–
12.
78. Köpke, M., Al‐Sinawi, B., Jensen, R.O. et al. (2015) Microorganisms and Methods for the
Production of Ketones. US20150152445. Washington,DC:U.S.Patent and Trade mark
Office.
79. Köpke, M., Simpson, S., Liew, F.M., and Hen, W. (2012) Fermentation Process for
Producing Isopropanol Using a Recombinant Microorganism. US20120252083A1.
Washington, DC: U.S. Patent and Trademark Office.
80. Koplove, H. M., and C. L. Cooney. 1979. Enzyme production during transient growth. Adv.
Biochem. Eng. 12: 1-40.
81. Krawczyk, T. (1996) Biodiesel‐alternative fuel makes inroads but hurdles remain. Inform,
7, 801–829.
82. Kresnowati, M.T.A.P., Van Winden, W.A., Van Gulik, W.M., and Heijnen, J. J. 2008.
Energetic and metabolic transient response of Saccharomyces cerevisiae to benzoic
acid. FEBS J., 275, 5527–5541.
83. Kwanmin, J.J. (1989) Scale‐down techniques for fermentation. Biopharm., 2, 30–39.
84. Kwint, K., Nachin, L., Diez, A., and Nyström, T. (2003). The bacterial universal stress
protein: Function and regulation. Curr. Opin. Microbiol., 6, 140–145.
85. Kywe, T. and Oo, M. (2009) Production of biodiesel from Jatropha oil (Jatropha curcas)
in pilot plant. WASET, 14, 78–86.
86. Lam, F.H., Ghaderi, A., Fink, G.R., and Stephanopoulos, G. (2014) Biofuels.
Engineering alcohol tolerance in yeast. Science, 346, 71–75.
87. Lane, J. (2015̍̍a) Alcohol to renewable jet fuel: the Digest’s 2015̍̍ 8‐Slide Guide to Gevo’s
ATJ,”in Biofuels Dig. Available online at:
http://www.biofuelsdigest.com/bdigest/2015/12/16/alcohol‐to‐renewable‐jet‐fuel‐the‐
digests‐2015‐8‐slide‐guide‐to‐gevos‐atj/.
88. Lanzatech (2015) Available on lineat:http://www.lanzatech.com/ innovation/
markets/fuels/[Accessed Jully, 2017]. Fuels.
89. Lapin, A., Klann, M., and Reuss, M. (2010) Multi‐scale spatio‐temporal modeling:
Lifelines of microorganisms in bioreactors and tracking molecules in cells. Adv. Biochem.
Eng., 121, 23–43.
90. Lara, A.R., E. Galindo, O.T. Ramirez and and L.A. Palomares. 2006. Living with
heterogeneities in bioreactors: understanding the effects of environmental gradients on
cells. Mol. Biotechnol.,34(3): 355-381.
91. Lara, A.R., Galindo, E., Ramírez, O.T., and Palomares, L.A. (2006) Living with
heterogeneities in bioreactors. Mol. Biotechnol., 34, 355–381.
92. Larsson, G., Törnkvist, M., Wernersson, E.S. et al. (1996a) Substrate gradients in
bioreactors: Origin and consequences. Bioprocess. Eng., 14, 281–289.
93. Larsson, G., Törnkvist, M., Wernersson, E.S. et al. (1996b) Substrate gradients in
bioreactors: origin and consequences. Bioproc. Engin, 14, 281–289.

100
94. Larsson, G.,M. Tornkvist, E. S. Wernersson, C. Tragardh, H. Noorman and S.O. Enfors.
1996. Substrate gradients in bioreactors: Origin and consequences, Biopro. Eng.,14(6):
281-289.
95. Latif, H., Zeidan, A.A., Nielsen, A.T., and Zengler, K. (2014) Trash to treasure:
production of biofuels and commodity chemicals via syngas fermenting microorganisms.
Curr. Opin. Biotechnol., 27, 79–87.
96. Lee, I. H., A. G. Fredrickson, and H. M. Tsuchiya. 1976. Damped oscillations in continuous
culture of Lactobacillus planetarium. J. Gen. Microbiol. 93: 204-209.
97. Lee, J., Lee, S.Y., Park, S., and Middelberg, A.P.J. (1999) Control of fed‐batch
fermentations. Biotechnol. Adv., 17, 29–48.
98. Lencastre Fernandes, R., Nierychlo, M., Lundin, L. et al. (2011) Experimental methods
and modeling techniques for description of cell population heterogeneity. Biotechnol.
Adv., 29, 575–599.
99. Lillly, M.D. 1983. Problems in process scale process, London, pp. 79-90.
100. Lin, H.Y. and Neubauer, P. 2000. Influence of controlled glucose oscillations on a
fed-batch process of recombinant Escherichia coli. J. Biotechnol.,79(1): 27-37.
101. Lin, H.Y. and Neubauer, P.I. (2000) Influence of controlled glucose oscillations
on a fed‐batch process of recombinant Escherichia coli. J. Biotechnol., 79, 27–37.
102. Ling, H., Teo, W., Chen, B. et al. (2014) Microbial tolerance engineering toward
biochemical production: from lignocellulose to products. Curr. Opin. Biotechnol., 29, 99–
106.
103. Los, M., Czyz, A., Sell, E. et al. (2004) Bacteriophage contamination: is there a
simple method to reduce its deleterious effect in the laboratory cultures and
biotechnological factories? J. Appl. Genet., 45, 111–120.
104. Lu, Z., Breidt, F., Plengvidhya, V., and Fleming, H.P. (2003) Bacteriophage
ecology in commercial sauerkraut fermentations. Appl. Environ. Microbiol., 9, 3192–
3202.
105. Luedeking, R., and E. L. Piret. 1959. Transient and steady states in continuous
fermentation. Theory and experiment. J. Biochem. Microbiol. Tech. Eng. 1: 453-459.
106. Luningprak, D.J., Jones, M.H., Trulove, P. et al. (2015) Physical and chemical
analysis of alcohol‐to‐jet(ATJ) fuel and development of surrogate fuel mixtures. Energy
and Fuels, 29, 3760–3769.
107. Lütke‐Eversloh, T. and Bahl, H. (2011) Metabolic engineering of Clostridium
acetobutylicum: Recent advances to improve butanol production. Curr. Opin.
Biotechnol., 22, 634–647.
108. Ma, N., Mollet, M., and Chalmers, J.J. (2006) Aeration, mixing and
hydrodynamics in bioreactors. in: Ozturk, S., Hu, W. (Eds.), Cell Culture Technology for
Pharmaceutical and Cell-Based Therapies, Taylor & Francis.
109. Magdouli, S., Brar, S.K. and Blais, J.F. (2016) Co‐culture for lipid production:
Advances and challenges. Biomass Bioenergy, 92, 20–30.
110. Magdouli, S., Brar, S.K., and Blais, J.F. (2017) Morphology and rheological
behaviour of Yarrowia lipolytica: Impact of dissolved oxygen level on cell growth and lipid
composition. Process. Biochem., https://doi.org/10.1016/j.procbio. 2017.10.021.
111. Magdouli, S., Yan, S., Tyagi, R.D., and Surampalli, R.Y. (2014) Heterotrophic
Microorganisms: A Promising Source for Biodiesel Production. Crit. Rev. Environ. Sci.
Technol., 44, 416–453.
112. Makshina, E.V., Dusselier, M., Janssens, W. et al. (2014) Review of old
chemistry and new catalytic advances in the on‐purpose synthesis of butadiene. Chem.
Soc. Rev., 43, 7917–7953.
113. Makshina, E.V., Janssens, W., Sels, B.F., and Jacobs, P.A. (2012) Catalytic
study of the conversion of ethanol into 1,3‐butadiene. Catal. Today 198, 338–344.

101
114. Marston, F. A. O., Angal, S., Lowe, P. A., Chan, M. & Hill, C. R. (1988) Scale-up
of the recovery and reactivation of recombinant proteins. Biochemical Society
Transactions 16, 112–115.
115. Martin, G. A., and W. P. Hempfling. 1976. A method for the regulation of microbial
population density during continuous culture at high growth rates. Arch. Microbiol. 107:
41-47.
116. Martinez‐Hernandez, E., Campbell, G., and Sadhukhan, J. (2013) Economic
value and environmental impact (EVEI) analysis of biorefinery systems. Chem. Eng.
Res. Des., 91, 1418–1426.
117. Mashego, M.R., Van Gulik, W.M., Vinke, J.L. et al. (2006) In vivo kinetics with
rapid perturbation experiments in Saccharomyces cerevisiae using a second‐generation
Bio Scope. Metab. Eng., 8, 370–383.
118. McGovern, A. C., D. Broadhurst, J. Taylor, N. Kaderbhai, M. K. Winson, D. A.
Small, J. J. Rowland, D. B. Kell, and R. Goodacre (2002). Monitoring of complex industrial
bioprocesses for metabolite concentrations using modern spectroscopies and machine
learning: application to gibberellic acid production, Biotechnol Bioeng 78:527-538.
119. Meerman, H.J., Kelley, A.S., and Ward, M. (2004) Advances in protein
expression in filamentous fungi. in: Baneyx, F. (Ed.), Protein Expression Technologies.
Current Status and Future Trends, 1st Edn., Horizon Bioscience, pp. 345–385.
120. Metzner, A.B. and Otto, R.E. (1957) Agitation of non‐Newtonian fluids. AIChE J.,
3, 3–10.
121. Michael J. W.,N. L. Morgan, J. S. Rockey, G. Higton. 2002. Industrial
Microbiology: An Introduction. Blackwell Science, Oxford, UK.
122. Minihane, B.J. and Brown, D.E. (1986) Fed‐batch culture technology. Biotechnol.
Adv., 4, 207–218.
123. Mogk, A., Tomoyasu, T., Goloubinoff, P. et al. (1999) Identification of
thermolabile Escherichia coli proteins: Prevention and reversion of aggregation by DnaK
and ClpB. EMBO J., 18, 6934–6949.
124. Moilanen, P., Laakkonen, M., and Aittamaa, J. (2006) Modeling aerated
fermenters
with computational fluid dynamics. Ind. Eng. Chem. Res., 45, 8656–8663.
125. Moulin, G., Broze, H., and Galzy, P. (1980) Inhibition of alcoholic fermentation by
substrate and ethanol. Biotechnology and bioengineering, 22, 2375–2381.
126. Moyer, A.J. (1948) Method for production of penicillin. US patent 2442141.
Murphy, T.D. (1977) Design and analysis of industrials experiments. Chem. Eng., 6,
168–182.
127. Naglak, T.J., Keith, M.G., and Omstead, D.R. (1994) Validation of fermentation
processes. Biopharm., 7, 28–36.
128. Neubauer, P., Lin, H.Y., and Mathiszik, B. (2003) Metabolic load of recombinant
protein production: inhibition of cellular capacities for glucose uptake and respiration
after induction of a heterologous gene in Escherichia coli. Biotechnol. Bioeng., 83, 53–
64.
129. Nienow, A.W. (1998) Hydrodynamics of stirred bioreactors. In: Fluid mechanics
problems in biotechnology. R. Pohorecki (ed.). App. Mech. Rev, 51, 3–32.
130. Nienow, A.W. (2006) Reactor engineering in large‐scale animal cell culture.
Cytotechnology, 50, 9–33.
131. Nienow, A.W. (2009) Scale‐up considerations based on studies at the bench
scale in stirred bioreactors. J. Chem. Eng. Jap., 42, 789–796.
132. Nienow, A.W., McLeod, G., and Hewitt, C.J. (2010) Studies supporting the use of
mechanical mixing in large scale beer fermentations. Biotechnol. Lett., 32, 623–633.
Novozymes (2002) AS: Annual Report 2001. Bagsvaerd, Denmark.

102
133. Ogata, S. 1980. Bacteriophage contamination in industrial process. Biotechnol.
Bioeng., 22, 177–193.
134. Ong, H.C., Mahlia, T.M.I., Masjuki, H.H., and Honnery, D. (2012) Life cycle cost
and sensitivity analysis of palm biodiesel production. Fuel, 98, 131–139.
135. Oosterhuis, N.M.G. and Kossen, N.W.F. (1984) Dissolved oxygen concentration
profiles in a production scale bioreactor. Biotechnol. Bioeng., 26, 546–550.
136. Oosterhuis, N.M.G., Kossen, N.W.F., Olivier, A.P.C., and Schenk, E.S. (1985)
Scale-down and optimization studies of the gluconic acid fermentation by Gluconobacter
oxydans. Biotechnol. Bioeng., 27, 711 ‐ 720.
137. Orgorzaly, H.J. 1958. The use of pilot plants in scale up. In: R.
Fleming,(Ed.),Scale up Practice.
138. Ostergaard, S., Olsson, L., and Nielsen, J. (2000) Metabolic engineering of
Saccharomyces cerevisiae. Microbiol. Mol. Biol. Rev., 64, 34–50.
139. Palluzi, R. P. 1992. Pilot plant design construction and operations. Mcgraw-Hill
publishers.
140. Park, J.H., Lee, K.H., Kim, T.Y., and Lee, S.Y. (2007) Metabolic engineering of
Escherichia coli for the production of L‐valine based on transcriptome analysis and in
silico gene knockout simulation. Proc. Natl. Acad. Sci. USA, 104, 7797–7802.
141. Pollard, D.J., Kirschner, T.F., Hunt, G.R. et al. (2007) Scale up of a viscous
fungal fermentation: Application of scale‐up criteria with regime analysis and operating
boundary conditions. Biotechnol. Bioeng., 96, 307–317.
142. Posch, A.E., Herwig, C., and Spadiut, O. (2013) Science‐based bioprocess
design for filamentous fungi. Trends. Biotechnol., 31, 37–44.
143. Posno, M., Leer, R.J., Van Luijk, N. et al. (1991) Incompatibility of Lactobacillus
vectors with replicons derived from small cryptic Lactobacillus plasmids and
segregational instability of the introduced vectors. Appl. Environ. Microbiol., 57, 1822–
1828.
144. Primrose, S.B. (1990) Controlling bacteriophage infection in industrial processes.
Adv. Biochem. Eng. Biotechnol., 43, 1–10.
145. Prokop, T. (2002) Imperial western products. Communication. P., 14970.
Chandler: St., Coachella.
146. Prueksakorn, K., and Gheewala, S.H., (2006) Energy and greenhouse gas
implications of biodiesel production from Jatropha curcas L. In: The 2nd Joint
International Conference on “Sustainable Energy and Environment (SEE 2006)”,
Bangkok, Thailand; 21–23 November 2006.
147. Ramasubramaniyan, P., C. S. Raj, O. Nagarajan, D., Sherly, L., Subramanian
and P. Solairaj. 2014. Pilot scale-up techniques for solid dose form-An overview for
tablets. World Journal of Pharmaceutical Research., 3(8): 925-931.
148. Reinikainen, P. and Virkajärvi, I. (1989) Escherichia coli growth and plasmid copy
numbers in continuous cultivations. Biotechnol.Lett., 11, 225–23.
149. Reuss, M. (1993) Oxygen transfer and mixing: Scale‐up implications. In: H.J.
Rehm
and G. Reed (eds), Biotechnology, Vol. 3, 2nd Edn., VCH pp. 185–213.
150. Richter, L., Wanka, F., Boecker, S. et al. (2014) Engineering of Aspergillus niger
for the production of secondary metabolites. Fungal. Biol. Biotechnol, 1, 1–13.
151. Riesenberg, D. and Schulz, V. (1991) High cell density cultivation of E. coli at
controlled specific growth rates. J. Biotechnol., 20, 17–28.
152. Rios, M. 2012. Anniversary Upstream BioProcss International., 10(6): 10-13.
153. Rokem, J.S., Lantz, A.E., and Nielsen, J. (2007) Systems biology of antibiotic
production by microorganisms. Nat. Prod. Rep., 24, 1262–1287.

103
154. Rønnest, N.P., Stocks, S.M., Lantz, A.E., and Gernaey, K.V. (2012) Comparison
of laser diffraction and image analysis for measurement of Streptomyces coelicolor cell
clumps and pellets. Biotechnol. Lett., 34, 1465–1473.
155. Rossi, F.G., Ribeiro, M.Z., Converti, A. et al. (2003) Enzyme Microb. Technol.,
32, 107-113.
156. Russo, E. (2001) Turning trash into treasure. Can organic waste become the
nation’s next big power source? The Scientist, 15, 1–4.
157. Sanchez Miron, A., Contreras Gomez, A., Garcia Camacho, F. et al. (1999).
Comparative evaluation of compact photobioreactors for large‐scale monoculture of
microalgae. J. Biotechnol., 70, 249–70.
158. Sassanfeld, H. M. (1990) Engineering proteins for purification. Trends in
Biotechnology 8, 88–92.
159. Scarff, M., Arnold, S.A., Harvey, L.M., and McNeil, B. (2006) Near Infrared
spectroscopy for bioprocess monitoring and control: Current status and future trends.
Crit. Rev. Biotechnol., 26, 17–39.
160. Schiel‐Bengelsdorf, B. and Dürre, P. (2012) Pathway engineering and synthetic
biology using acetogens. FEBS. Lett., 586, 2191–2198.
161. Schmalzriedt, S., Jenne, M., Mauch, K., and Reuss, M. (2003) Integration of
physiology and fluid dynamics. Process integration in biochemical engineering. Springer,
Berlin,, pp 19–68.
162. Schmidt, A.S., Garde, A., Klinke, H.B. et al. (2000) Lactic acid production from
wheat straw: Effect of pretreatment conditions. In: Proceedings of the First World
Conference and Exhibition on Biomass for Energy and Industry. Sevilla, Spain.
163. Schmidt, F. R. 2005. Optimization and scale-up of industrial fermentation
processes. App. Micro. Biotech., 68(4): 425-435.
164. Schmidt, F.R. (2005) Optimization and scale up of industrial fermentation
processes. Appl. Microbiol. Biotechnol., 68, 425–435.
165. Schweder, T., E. Kruger, B. Xu, B. Jurgen G. Blomsten, S. O. Enfor and H.
Hecker. 1999. Monitoring of genes that respond to process-related stress in larg-scale
bioprocesses. Biotech. Bioeng., 65(2): 151-159.
166. Schweder, T., Krüger, E., Xu, B. et al. (1999) Monitoring of genes that respond to
process related stress in large‐scale bioprocesses. Biotechnol. Bioeng., 65, 151–159.
167. Senior, P. J., G. A. Beech. G. A. F. Ritchie, and E. A. Dawes. 1972. The role of
oxygen limitation in the formation of poly-β-hydroxybutyrate during batch and continuous
culture of Azotobacter beijerinckii. Biochem. J. 128: 1193-1198.
168. Shu, P. and Johnson, M.J. (1948) Citric acid. J. Ind. Eng. Chem., 40, 1202–1205.
169. Sikander Ali et al. A review scale up fermentation procedure. Int J S Res Sci. Tech.
2018 Mar-Apr;4(5) : 1301-1307
170. Sipiczki, M. (2011) Diversity, variability and fast adaptive evolution of the wine
yeast (Saccharomyces cerevisiae) genome ‐ a review. Ann. Microbiol., 61, 85–93.
171. Smith, J.J., Lilly, M.D., and Fox, R.I. (1990). The effect of agitation on the
morphology and penicillin production of Penicillium chrysogenum. Biotechnol. Bioeng.,
35, 1011–1023.
172. Spohr, A., Dam‐Mikkelsen, C., Carlsen, M. et al. (1998) On‐line study of fungal
morphology during submerged growth in a small flow‐through cell. Biotechnol. Bioeng.,
58, 541–553.
173. Sreekumar, S., Baer, Z.C., Pazhamalai, A. et al. (2015) Production of an
acetone‐ butanol ethano lmixture from Clostridium acetobutylicum and its conversion to
high‐value biofuels. Nat. Protoc., 10, 528–537.
174. Sridhar, J., Eiteman, M.A., and Wiegel, J.W. (2000) Appl. Environ. Microbiol., 66,
246-251. Stocks, S.M. (2013) Industrial enzyme production: Process scale up/scale

104
 down. in: McNeil, B., Archer, D., Giavasis, I., Harvey, L. (Eds.), Microbial Production of
Food Ingredients, Enzymes and Nutraceuticals, Woodhead Publishing 2013, pp. 144–
172.
175. Srivasta, M. L. 2008. Fermentation technology. Alpha Science publishers, pp
100-105.
176. Stanbury, P. A., Whitaker, and S. Hall. 2016. Principles of fermentation
technology. (3rd Ed.). Butterworth-Heinemann publishers’ pp. 269-272.
177. Takors, R. (2012) Scale‐up of microbial processes: impacts, tools and open
questions. J. Biotechnol., 160, 3–9.
178. Thomas, C.R. (1990) Problems of shear in biotechnology. In Chemical
Engineering Problems in Biotechnology, M.A. Winkler, (ed.), pp 23–94. Elsevier Applied
Science, United Kingdom.
179. Thomas, K. and Ingledew, W.J. (1992) Ind. Microbiol. Biotechnol., 10, 61-68.
180. Todd, P., S. R. Rudge, D. P. Petrides, and R. G. Harrison, Eds. (2003).
Bioseparations Science and Engineering. Oxford University Press.
181. Tufvesson, P., Lima‐Ramos, J., Nordblad, M., and Woodley, J.M. (2011)
Guidelines and cost analysis for catalyst production in biocatalytic processes. Org.
Process Res. Dev., 15, 266–274.
182. Tzanov, T., Calafell, M., Guebitz, G.M., and Cavaco‐Paulo, A. (2001) Bio‐
preparation of cotton fabrics. Enzyme Microb. Technol., 29, 357–362.
183. Uihlein, A. and Schebek, L. (2009) Environmental impacts of a lignocellulose
feedstock biorefinery system: an assessment. Biomass. Bioenergy, 33, 793–802.
184. Umakoshi, H., Kuboi, R., Komasawa, I. et al. (1998) Heat Induced translocation
of cytoplasmic galactosidase across inner membrane of E. coli. Biotechnol. Prog., 14,
210–217.
185. Utrilla, J., Licona‐Cassani, C., Marcellin, E. et al. (2012) Engineering and
adaptive evolution of Escherichia coli for D‐lactate fermentation reveals GatC as a
xylose transporter. Metab. Eng., 14, 469–476.
186. Van Gulik, W.M., De Laat, W., Vinke, J.L., and Heijnen, J.J. (2000) Application of
metabolic flux analysis for the identification of metabolic bottlenecks in the biosynthesis
of penicillin‐G. Biotechnol. Bioeng., 68, 602–618.
187. Wang, D. I. C., and A. J. Sinskey. 1970. Collection of microbial cells. Adv. Appl.
Microbiol. 12: 121-132.
188. Wang, Y. and Van Ness, B. (1989) Site specific cleavage of supercoiled DNA by
ascorbate /Cu(II). Nucl. acid research, 17, 6915–6926.
189. Watson, T. G. 1969. Steady state operation of a continuous culture at maximum
growth rate by control of carbon dioxide production. J. Gen. Microbiol. 59: 83-87.
190. Watson, T. G. 1972. The present status and future prospects of the turbidostat. J.
Appl. Chem. Biotechnol. 22: 229-243.
191. Woolston, B.M., Edgar, S., and Stephanopoulos, G. (2013) Metabolic
engineering: Past, and future. Annu. Rev. Chem. Biomol. Eng., 4, 259–288. World
Enzymes (2011) Cleveland, Ohio, United States of America. 12–26.
192. Wu, B. (2012) Integration of mixing, heat transfer, and biochemical reaction
kinetics in anaerobic methane fermentation. Biotechnol. Bioeng., 109, 2864–2874.
193. Wunsche, L. (1989) Importance of bacteriophages in fermentation processes.
Acta Biotechnol., 9, 395–419.
194. Xu, B., Jahic, M., Bomsten, G., and Enfors, S. O. (1999) Glucose overflow
metabolism and mixed acid fermentation in aerobic large‐scale fed‐batch processes with
Echerichia coli. Appl. Microbiol. Biotechnol., 51, 564–571.

105
 195. Xu, B., M. Jahic, G. Blomsten and S. O. Enfors. 1999. Glucose overflow
metabolism and mixed acid fermentation in aerobic large scale fed-batch processes with
E. coli. App. Microbio. And Biotech., 51(5): 564-571.
196. Yamada, S., M. Wada, and I. Chibata. 1979. Oxygen transfer as a parameter of
automatic control of the continuous cultivation for the conversion of sorbitol to sorbose by
Acetobacter suboxydans. J. Ferment. Technol. 57: 210-225.
197. Ye, X.H., Chu, J., Zhuang, Y.P., and Zhang, S.L. (2005) Multi‐scale
methodology: a key to deciphering systems biology. Front. Biosci., 10, 961–965.
198. Zhang, M. and Yu, Y. (2013) Dehydration of ethanol to ethylene. Ind. Eng. Chem.
Res., 52, 9505–9514.
199. Zhang, S.L., Chu, J., and Zhuang, Y.P. (2004). A multi‐scale study of industrial
fermentation processes and their optimization. Biomanufacturing. Springer, Berlin, pp
97–150.
200. Zhang, S.L., Ye, B.C., Chu, J. et al. (2006). From multi‐scale methodology to
systems biology: to integrate strain improvement and fermentation optimization. J.
Chem. Technol. Biotechnol., 81, 734–745.
201. Zhang, T. (2015) More efficient together. Science, 350, 738–739.
202. Zhou, K., Qiao, K., Edgar, S., and Stephanopoulos, G. (2015) Distributing a
metabolic pathway among a microbial consortium enhances production of natural
products. Nat. Biotechnol., 33, 377–383.
203. Zhu, H., Swierstra, J., Wu, C. et al. (2014) Eliciting antibiotics active against the
ESKAPE pathogens in a collection of actinomycetes isolated from mountain soils.
Microbiology, 160, 1714–1725.

Lecturer: Prof. Hilary Domakyaara Zakpaa

106