VIABILITY, TOXICITY, AND SURVIVAL

• Once a cell is explanted from its normal in vivo environment, the question of
viability, particularly in the course of experimental manipulations, becomes
fundamental.

• Furthermore, many experiments carried out in vitro are for the sole purpose of
determining the potential cytotoxicity of the compounds being studied, either
because the compounds are being used as pharmaceuticals or cosmetics and must
be shown to be nontoxic or because they are designed as anticancer agents and
cytotoxicity may be crucial to their action.

• New drugs, cosmetics, food additives, and so on go through extensive cytotoxicity

testing before they are released for use by the public.

• Toxicity is a complex event in vivo, where there may be direct cellular damage, as
with a cytotoxic anticancer drug, physiological effects, such as membrane transport
in the kidney or neurotoxicity in the brain, inflammatory effects, both at the site of
application and at other sites, and other systemic effects.

• Currently, it is difficult to monitor systemic and physiological effects in vitro, so

most assays determine effects at the cellular level, or cytotoxicity.
• Definitions of cytotoxicity vary, depending on
the nature of the study and whether cells are
killed or simply have their metabolism altered.

• Whereas an anticancer agent may be required

to kill cells, demonstrating the lack of toxicity
of other pharmaceuticals may require a more
subtle analysis of specific targets such as an
alteration in cell signaling or cell interaction
such as might give rise to an inflammatory or
allergic response.
IN VITRO LIMITATIONS
• It is important that any in vitro measurement can be interpreted in terms of the in
vivo response of the same or similar cells, or at least that the differences that exist
between in vitro and in vivo measurements are clearly understood.

• Pharmacokinetics. The measurement of toxicity in vitro is generally a cellular event.

For example, it would be very difficult to recreate the complex pharmacokinetics of
drug exposure in vitro, and between in vitro and in vivo experiments there usually
are significant differences in exposure time to and concentration of the drug, rate of
change of the concentration, metabolism, tissue penetration, clearance, and
excretion. Although it may be possible to simulate these parameters—e.g., using
multicellular tumor spheroids for drug penetration or timed perfusion to simulate
concentration and time (C × T) effects—most studies concentrate on a direct cellular
response, thereby gaining simplicity and reproducibility.

• Metabolism. Many nontoxic substances become toxic after being metabolized by

the liver; in addition, many substances that are toxic in vitro may be detoxified by
liver enzymes. For in vitro testing to be accepted as an alternative to animal testing,
it must be demonstrated that potential toxins reach the cells in vitro in the same
form as they would in vivo. This proof may require additional processing by purified
liver microsomal enzyme preparations, coculture with activated hepatocytes , or
genetic modification of the target cells with the introduction of genes for
metabolizing enzymes under the control of a regulatable promoter.
Tissue and systemic responses
• The nature of the response must also be considered carefully. A
toxic response in vitro may be measured by changes in cell
survival or metabolism whereas the major problem in vivo may be
a tissue response (e.g., an inflammatory reaction, fibrosis, kidney
failure) or a systemic response (e.g., pyrexia, vascular dilatation).
For in vitro testing to be more effective, models of these
responses must be constructed, perhaps utilizing organotypic
cultures reassembled from several different cell types and
maintained in the appropriate hormonal milieu.

• It should not be assumed that complex tissue and even systemic

reactions cannot be simulated in vitro. Assays for inflammatory
responses, teratogenic disorders, and neurological dysfunctions
may be feasible in vitro, given a proper understanding of cell–cell
interaction and the interplay of endocrine hormones with local
paracrine and autocrine factors.
NATURE OF THE ASSAY
• The choice of assay will depend on the agent under study, the nature of the
response, and the particular target cell. Assays can be divided into five major classes:

(1) Viability: An immediate or short-term response, such as an alteration in membrane

permeability or a perturbation of a particular metabolic pathway correlated with cell
proliferation or survival.

(2) Survival: The long-term retention of self-renewal capacity (5–10 generations or

more).

(3) Metabolic: Assays, usually microtitration based, of intermediate duration that can
either measure a metabolic response (e.g., dehydrogenase activity; DNA, RNA, or
protein synthesis) at the time of, or shortly after, exposure. Making the
measurement two or three population doublings after exposure is more likely to
reflect cell growth potential and may correlate with survival.

(4) Transformation: Survival in an altered state (e.g., a state expressing genetic

mutation, alterations in growth control, or malignant transformation).

(5) Irritancy: A response analogous to inflammation, allergy, or irritation in vivo; as yet

difficult to model in vitro, but may be possible to assay by monitoring cytokine
release in organotypic cultures.
Viability
• Viability assays are used to measure the proportion of viable cells after a potentially
traumatic procedure, such as primary disaggregation, cell separation, or cryostorage.

• Most viability tests rely on a breakdown in membrane integrity measured by the

uptake of a dye to which the cell is normally impermeable (e.g., Trypan Blue,
Erythrosin, or Naphthalene Black) or the release of a dye normally taken up and
retained by viable cells (e.g., diacetyl fluorescein or Neutral Red).

• However, this effect is immediate and does not always predict ultimate survival.
Furthermore, dye exclusion tends to overestimate viability—e.g., 90% of cells
thawed from liquid nitrogen may exclude Trypan Blue, but only 60% prove to be
capable of attachment 24 h later.

• Note that routine assessment of viability at subculture can be uninformative

regarding trypsinized cells as most of the nonviable cells will be lost in the discarded
medium and prewash before trypsinization.

• An accurate assessment of the viability status at subculture requires that all the cells
are recovered from the medium and prewash and combined with the trypsinate.
However, viability of reseeded cells will be accurately determined without this
recovery.
• Neutral red uptake. Living cells take up neutral red, 40
µg/mL in culture medium, and sequester it in the
lysosomes. However, neutral red is not retained by
nonviable cells.

• Uptake of neutral red is quantified by fixing the cells in

formaldehyde and solubilizing the stain in acetic ethanol,
allowing the plate to be read on an ELISA plate reader at
570 nm.

• Neutral red tends to precipitate, so the medium with stain

is usually incubated overnight and centrifuged before use.

• This assay does not measure the total number of cells, but it
does show a reduction in the absorbance related to loss of
viable cells and is readily automated.
Survival
• Although short-term tests are convenient and usually are quick and easy
to perform, they reveal only cells that are dead (i.e., permeable) at the
time of the assay. Frequently, however, cells that have been subjected to
toxic influences (e.g., irradiation, antineoplastic drugs) show an effect
several hours, or even days, later.

• The nature of the tests required to measure viability in these cases is

necessarily different, because by the time the measurement is made the
dead cells may have disappeared.

• Therefore, long-term tests are used to demonstrate survival rather than

short-term toxicity, which may be reversible. Survival implies the retention
of regenerative capacity and is usually measured by plating efficiency.

• Although not ideal, a plating efficiency of over 10% is usually acceptable.

Variable parameters in survival assay
Concentration of agent
• A wide range of concentrations in log
increments (e.g., 1 × 10−6 M, 1 × 10−5 M, 1 ×
10−4 M, 1 × 10−3 M, and control) should be
used for the first attempt and a narrower
range (log or linear), based on the results from
the first range, for subsequent attempts.
Invariate agent concentrations.
Some conditions that are tested cannot easily be
varied—e.g., the quality of medium, water, or an insoluble
plastic. In these cases, the serum concentrations can be
varied. As serum may have a masking effect on low-level
toxicity, an effect may only be seen in limiting serum.

Duration of exposure to agent.

Some agents act rapidly, whereas others act more slowly.
Exposure to ionizing radiation, for example, need last only a
matter of minutes to achieve the required dose, whereas
testing some cycledependent antimetabolic drugs may take
several days to achieve a measurable effect. Duration of
exposure (T) and drug concentration (C) are related,
although C × T is not always a constant. Prolonging
exposure can increase sensitivity beyond that predicted by
C × T, because of cell cycle effects and cumulative damage.
Time of exposure to agent

• When the agent is soluble and expected to be toxic,

when the quality of the agent is unknown, stimulation
is expected, or only a minor effect is expected (e.g.,
20% inhibition rather than severalfold), the agent may
be incorporated during clonal growth rather than at
preincubation.

• Confirmation of anticipated toxicity—e.g., for a

cytotoxic drug— requires a conservative assay, with
minimal drug exposure as compared to that in vivo, be
applied before cloning. Confirmation of the lack of
toxicity—e.g., for tap water or a nontoxic
pharmaceutical— requires a more stringent assay,
with prolonged exposure during cloning.
Cell density during exposure

• The density of the cells during exposure to an agent can

alter the response of the cells and the agent; for example,
HeLa cells are less sensitive to the alkylating agent mustine
at high cell densities.

• To determine the effects of cell density it should be varied

in the preincubation phase, during exposure to the drug.

Cell density during cloning

• The number of colonies may fall at high concentrations of a

toxic agent, but it is possible to compensate for this effect
by seeding more cells so that approximately the same
number of colonies form at each concentration.
Colony size

• Some agents are cytostatic (i.e., they inhibit cell

proliferation) but not cytotoxic, and during
continuous exposure they may reduce the size of
colonies without reducing the number of colonies.

• In this case, the size of the colonies should be

determined by densitometry automatic colony
counting, or visually counting the number of cells
per colony.
Solvents
• Some agents to be tested have low solubilities in aqueous
media, and it may be necessary to use an organic solvent to
dissolve them.

• Ethanol, propylene glycol, and dimethyl sulfoxide have been

used for this purpose, but may themselves be toxic to cells.

• Hence, the minimum concentration of solvent should be

used to obtain a solution.

• The agent may be made up at a high concentration in, for

example, 100% ethanol, then diluted gradually with BSS and
finally diluted into medium. The final concentration of
solvent should be
Assays Based on Cell Proliferation

• Cell counts after a few days in culture can also be used to

determine the effect of various compounds on cell proliferation,
but, at least in the early stages of testing, a complete growth curve
is required, because cell counts at a single point in time can be
ambiguous.

• Growth curve analyses, using cell counting, are feasible only with
relatively small numbers of samples, as they become cumbersome
in a large screen. In cases for which there are many samples, a
single point in time—e.g., the number of cells three to five days
after exposure—can be used.

• The time should be selected as within the log phase, and

preferably mid-log phase, of control cells. Any significant effect
should be backed up with a complete growth curve over the
whole growth cycle or by an alternative assay, such as a survival
curve by clonogenic assay or MTT assay.
Metabolic Cytotoxicity Assays

• Plating efficiency tests are labor intensive and time consuming to set up and analyze,
particularly when a large number of samples is involved, and the duration of each
experiment may be anywhere from 2 to 4 weeks.

• Furthermore, some cell lines have poor plating efficiencies, particularly freshly
isolated normal cells, so a number of alternatives have been devised for assaying
cells at higher densities, e.g., in microtitration plates.

• None of these tests measures survival directly, however. Instead, the net increase in
the number of cells the increase in the total amount of protein or DNA, or continued
metabolic activity, such as the reduction of a tetrazolium salt to formazan or the
synthesis protein or DNA, is determined.

• Survival in these cases is defined as the retention of metabolic or proliferative ability

by the cell population as a whole some time after removal of the toxic influence.

• However, such assays cannot discriminate between a reduction in metabolic or

proliferative activity per cell and a reduced number of cells, and therefore any novel
or exceptional observation should be confirmed by clonogenic survival assay.
Microtitration Assays
• The introduction of multiwell plates revolutionized the approach to replicate
sampling in tissue culture. These plates are economical to use, lend themselves to
automated handling, and can be of good optical quality.

• The most popular is the 96-well microtitration plate, each well having 28–32 mm2 of
growth area, 0.1 or 0.2 mL medium, and up to 1 × 105 cells. Microtitration offers a
method whereby large numbers of samples may be handled simultaneously, but
with relatively few cells per sample.

• With this method, the whole population is exposed to the agent, and viability is
determined subsequently, usually by measuring a metabolic parameter such as the
ATP or NADH/NADPH concentration. A range of kits are available from Promega .

• The end point of a microtitration assay is usually an estimate of the number of cells.
Although this result can be achieved directly by cell counts or by indirect methods,
such as isotope incorporation, cell viability as measured by MTT reduction is now
widely chosen as the optimal end point.

• MTT is a yellow water-soluble tetrazolium dye that is reduced by live, but not dead,
cells to a purple formazan product that is insoluble in aqueous solutions. However, a
number of factors can influence the reduction of MTT.
• Labeling with [3H]thymidine (DNA synthesis), [3H]leucine
[Freshney et al., 1975] or [35S]methionine (protein
synthesis), or other isotopes can be substituted forMTT
reduction.

• Quantitation is achieved by microtitration plate scintillation

counting.

• Two types of scintillation counting are available:

(1) The cellular contents may be aspirated onto filters by

trypsinization and onto glass fiber filters by suction transfer,
and the filters may be dried and counted in scintillant, or

(2) the whole plate may be counted on a specially adapted

scintillation counter (PerkinElmer).
APPLICATIONS OF CYTOTOXICITY ASSAYS

Anticancer Drug Screening

• Drug screening for the identification of new anticancer drugs can be a

tedious and often inefficient method of discovering new active
compounds.

• The trend is now more toward monitoring effects on specific molecular

targets. However, there have been attempts to improve screening by
adopting rapid, easily automated assays, like those based on the
determination of the number of viable cells by staining the cells with MTT.

• To further cut down on manipulations, the MTT incubation step may be

omitted and the end point determined by measuring the amount of total
protein with sulforhodamine B.

• Although this method is quicker and easier than the MTT assay, it should
be remembered that nonviable, and certainly nonreplicating, cells will still
stain, so the assay should be confirmed when activity is detected, using a
more reliable indicator such as clonogenicity or MTT reduction.
Predictive Drug Testing for Tumors

• The possibility has often been considered that measurement of

the chemosensitivity of cells derived from a patient’s tumor might
be used in designing a chemotherapeutic regime for the patient.

• As many of the cells within a tumor have a limited life span, the
main targets for chemotherapy are the clonogenic populations
with infinite repopulation capacity. Advances in stem cell
recognition may make isolation of tumor stem cells feasible.

• Routine isolation of tumor cell populations with long-term

repopulation efficiency has yet to be achieved, but when it is it
may be possible to improve targeting and specificity of anticancer
drugs, making predictive testing more meaningful.
Testing Pharmaceuticals

• A number of pharmaceutical companies

maintain a program of in vitro testing on the
assumption that it might prove more
economical and ethically acceptable than
animal testing.

• Legislation enforcing the use of animal tests is

difficult to introduce as the complexity of the
wide range of effects seen in vivo is still very
difficult to model in vitro.
TRANSFORMATION AND MUTAGENESIS

• Commonly used in vitro assays for

transformation include anchorage
independence reduced density limitation of
cell proliferation, and evidence of
mutagenesis.

• Mutagenesis can be assayed by sister

chromatid exchange.
Mutagenesis Assay by Sister Chromatid
Exchange
• Sister chromatid exchanges (SCEs) are reciprocal exchanges of
DNA segments between sister chromatids at identical loci during
the S-phase of the cell cycle. As SCEs are more sensitive indicators
of mutagenic activity than chromosome breaks, they have become
a major tool in mutagenesis research [Latt, 1981].

• With the development of the thymidine analog

bromodeoxyuridine (BUdR) and its subsequent use in DNA
labeling experiments, the resolution of SCEs greatly improved in
comparison to previous methods, which involved the
incorporation of radioactive nucleotides into replicating DNA
[Taylor, 1958].

• Later, the fluorescence plus Giemsa (FPG) technique of Perry and

Wolf [1974] for the scoring of SCEs was enhanced, and for the first
time, permanent staining of SCEs was demonstrated.
 • The FPG method involves two distinct steps:

(1) Cells are labeled with BUdR for two complete cycles and then treated with
colcemid to block the cells in metaphase.

(1) After BUdR exposure, the DNA of one chromatid of each chromosome contains
bromouracil in one strand, while the DNA of its sister chromatid contains
bromouracil in both strands.

(3) Chromosomes are then prepared from these cells and stained with the fluorescent
dye Hoechst 33258, and then the BUdR is photodegraded with ultraviolet light;
this is followed by Giemsa staining.

(4) These final steps highlight the differential incorporation of bromouracil into the
sister chromatids.

(5) DNA that contains bromouracil quenches the fluorescence of Hoechst-DNA

complexes.

(6) Therefore, the chromatid containing bromouracil substituted in both strands

fluoresces weakly and stains weakly with Giemsa, while the chromatid containing
bromouracil in only one strand fluoresces more intensely, degrades the BUdR, and
subsequently stains darkly with Giemsa .

(7) If any SCEs occur, this staining pattern produces what are called harlequin
chromosomes.
INFLAMMATION

• There is an increasing need for tissue culture testing to reveal the

inflammatory responses that are likely to be induced by
pharmaceuticals and cosmetics with topical application or by
xenobiotics that may be inhaled or ingested and may be
responsible for many forms of allergy.

• This is an area that is only at the early stages of development but

bears great promise for the future. It is a sensitive topic in more
ways than one.

• Animal rights groups are naturally incensed at the needless use of

large numbers of animals to test new cosmetics that have little
benefit except commercial advantage to the manufacturer,
particularly when the testing of substances (such as shampoos)
involves the Draize test, in which the compound is added to a
rabbit’s eye.

•
• More important, clinically, is the apparent increase in allergenic responses
produced by pharmaceuticals and xenobiotics. These responses are little
understood and poorly controlled, largely because of the absence of a
simple reproducible in vitro test.

• Since the advent of filter well technology, several models for skin and
cornea have appeared, utilizing the facility for coculture of different cell
types that the filter well system provides. In these systems, the interaction
of an allergen or irritant with a primary target (e.g., epidermis) is
presumed to initiate a paracrine response, which triggers the release of a
cytokine from a second, stromal component.

• This cytokine can then be measured by ELISA technology to monitor the

degree of the response. Although still in the early stages of development,
kits for the measurement of irritant responses are available.

• A protocol for corneal culture suitable for modeling irritant responses in

the eye was provided by Carolyn Cahn. It would seem that this type of
system may be a major area of development, with the real prospect that
allergen screening from patients’ own skin in organotypic culture may
become possible and that, ultimately, analysis of the GI tract will reveal
allergens responsible for irritable bowel syndrome. In each case, and in
many others, there is the possibility of specific mechanistic studies into
the processes of abnormal cell interaction that typify many allergic and
degenerative diseases.