Biochemical Engineering Journal 47 (2009) 87–92

Contents lists available at ScienceDirect

Biochemical Engineering Journal

journal homepage: www.elsevier.com/locate/bej

Culture medium improvement for Isaria fumosorosea submerged

conidia production
Ali Asaff ∗ , Francisco Escobar, Mayra de la Torre
Coordinación de Ciencia de los Alimentos, CIAD A.C., Hermosillo, Sonora 83000, Mexico

a r t i c l e i n f o a b s t r a c t

Article history: Submerged conidia and blastospores of the entomopathogenic fungus Isaria fumosorosea are produced in
Received 20 January 2009 several liquid culture media. However, yields and the ecological fitness of these propagules vary according
Received in revised form 11 June 2009 to culture media composition. In most culture media, hyphae, blastospores and submerged conidia are
Accepted 8 July 2009
white but we found that in some media they develop a brown pigmentation. A dark pigment was extracted
from brown-pigmented propagules and analyzed by IR spectroscopy. Adsorption bands coincided to those
characteristics of melanins.
Keywords:
Hadamard’s matrices were employed in order to increase submerged conidia yields and brown pigmen-
Optimization of culture medium
Submerged fermentation
tation of fungal propagules. Media containing 20–30 mg/l of FeSO4 ·7H2 O and 6–12 mg/l of CuSO4 ·5H2 O
Shake-flask culture allowed reaching the highest pigmentation (9 in a hedonic scale). A maximal concentration of submerged
Biomass production conidia of 1.0 (±1.2) × 1012 cell/l was achieved after 120 h of liquid culture in a improved culture medium,
Submerged conidia containing 25 ml/l of Polyethylene glycol (MW 200), substance which enhanced submerged conidia pro-
Isaria fumosorosea duction, reducing free mycelia or mycelial pellets formation. In the improved medium, it was estimated
that more than 60% of produced biomass corresponded to submerged conidia and blastospores, while in
other media, mycelia were the main product (80–97%).
© 2009 Elsevier B.V. All rights reserved.

1. Introduction of both growth conditions and culture media. Historically, one-at-a-

Isaria fumosorosea (Wize) (formerly Paecilomyces fumosoroseus)
is a fungal biocontrol agent with potential for controlling several
insect pests [1]. When grown on solid media it produces differen-
tial asexual conidia, which are genetically stable [2] and resistant to
adverse environmental conditions, like other Hyphomycetes. Aerial
conidia of I. fumosorosea have a typical brown color [3] attributed to
melanins, pigments which may contribute to fungal spores protec-
tion from solar radiation, particularly from UV light incidence, high
temperatures, desiccation, enzymatic lysis and fungicides [4]. In
submerged culture, I. fumosorosea produces usually non-pigmented
propagules which include freely dispersed mycelia, mycelial pellets,
short hyphae, blastospores and submerged conidia [5–7]. Blas-
tospores and submerged conidia are infective propagules, but they
are lesser resistant to environmental stresses than aerial conidia
[6]. Cell wall composition [8–10], osmolites [11,12] and pigments
like melanins may contribute to the resistance of the fungal spores
to abiotic stresses.
One approach to improve fungal biopesticides is to enhance the
ecological fitness of aerial conidia, submerged conidia and/or blas-
tospores by physiological manipulation [13], through modifications
time strategy has been one of the most popular choices for improv-
ing fungal medium composition [6,10,14–16]. The rationale behind
of this strategy is keeping the concentration of all culture medium
components constant except one. Nevertheless, the technique has
some major flaws; for example the optimum can be completely
missed, and a relatively large number of experiments is needed [17].
In this study we assessed the critical factors involved in
submerged conidia yield of I. fumosorosea. We also identified
which media components are important for pigmentation of these
propagules and we report the improvement of the culture medium
through the use of Hadamard’s matrices. These are most commonly
employed for identifying important factors of a system and optimiz-
ing medium compositions and growth conditions, both in closed
and open improvement strategies [17,18]. The improved medium
resulted in high concentration of brown-pigmented submerged
conidia, scarce production of blastospores and disperse-short-
mycelium. Melanin was identified in brown-pigmented fungal
propagules.
2. Materials and methods
2.1. Microorganism and growth conditions

∗ Corresponding author. Tel.: +52 662 2892400; fax: +52 662 2800381. A stock culture of the I. fumosorosea Pfrd strain obtained from
E-mail address: aliasaff@hotmail.com (A. Asaff). the culture collection (a public-access culture collection) of the

1369-703X/$ – see front matter © 2009 Elsevier B.V. All rights reserved.
doi:10.1016/j.bej.2009.07.007
88 A. Asaff et al. / Biochemical Engineering Journal 47 (2009) 87–92
Centro Nacional de Referencia de Control Biológico (DGSV-
SAGARPA, Mexico) was maintained in slant tubes, using SDY
medium [65 g/l Sabouraud dextrose agar (Difco) supplemented
with 10 g/l yeast extract (Difco)]. The fungus was grown on SDY-
Petri dishes for 14 days at 27 ◦ C with a 12:12 light:dark photoperiod;
dishes were sealed with parafilm from inoculum up to day 7, then
the seal was removed to allow dry-out in order to stimulate sporu-
lation. Aerial fresh conidia were harvested in sterile 0.05% Tween
80 and the suspension, containing 1 × 108 conidia/ml, was stored in
5-ml cryogenic tubes at −40 ◦ C. This spore suspension was utilized
as inoculum for flask cultures.
2.2. Submerged culture
Erlenmeyer flasks (125 ml) containing 30 ml of designed
medium were inoculated with 0.5 ml of the 1 × 108 conidia/ml sus-
pension stored in the cryogenic tubes. Flasks were incubated at
28 ◦ C and 150 rpm for 5 days.
2.3. Analytical methods
Total biomass was determined by gravimetric analysis after fil-
tration of the sample through pre-weighed nitrocellulose filters
(45 mm diameter, pore size 0.45 ␮m), rinsing twice with 10 ml dis-
tilled water aliquots and drying at 95 ◦ C to constant weight. Biomass
of blastospores and submerged conidia was estimated in a simi-
lar way after filtering a sample aliquot through gauze to remove
mycelia.
Blastospores and submerged conidia were counted in a hema-
cytometer using a light microscope (400× magnification). A
propagule was considered as blastospore when a monocellu-
lar hyphal body presenting an oblong to cylindrical shape was
observed and as submerged conidia when the morphological char-
acteristics were close to those of aerial conidiospores [19]. The
estimated average value of number of blastospores and submerged
conidia per gram of them was 1.2 × 1011 cell/gcell .
Color was measured by comparing the color of the biomass
retained in nitrocellulose filters (45 mm diameter, pore size
0.45 ␮m) and rinsed twice with 10 ml distilled water aliquots, with
the Munsell color chart (10R), choosing 9 possibilities (8/1, 8/2,
8/3, 8/4, 7/6, 7/4, 6/3, 5/3 and 5/1), where 8/1 correspond to white
and 5/1 to the darkest brown color. To each one of these colors
a numerical value (1–9) was assigned in order to develop further
mathematical and statistical analysis.
Extraction and purification of pigments from the fungal biomass
were performed according the procedure for plant melanins
described by Sava et al. [20] with minor modifications. First, the
fungal biomass was treated with 2 M NaOH, pH 10.5, for 36 h. There-
after, the mixture was centrifuged at 4000 × g for 15 min and the
supernatant was acidified with 2 M HCl to pH 2.5, incubated for 2 h
at room temperature, and centrifuged at 4000 × g for 15 min. The
precipitate was purified by acid hydrolysis using 6 M HCl at 100 ◦ C
for 2 h to remove carbohydrates and proteins and treated with ethyl
acetate to wash away lipids. The precipitate was then dried at room
temperature, re-dissolved in 2 M NaOH and centrifuged at 4000 × g
for 15 min. The supernatant was precipitated by the addition of 1 M
HCl, washed with distilled water and lyophilized. This procedure
yielded about 1.1 mg of pigment per g of pigmented fungal mass.
Pigment was mixed with KBr (1:100, w/w), and the mixture was
ground in a mortar and pestle until uniform color was achieved.
Aliquots were pressed into 5-mm pellets and characterized with IR
spectroscopy in a FTIR Nicolet spectrophotometer, Protege 460 ESP
model (Nicolet Co., Madison, WI).
Histological identification of melanins was performed using
Fontana-Masson-melanin-specific stain. Aliquots of 10 ml of brown
and white cultures were centrifuged at 4000 × g for 15 min to
remove supernatant. Precipitate fungal biomass was washed twice
with 10 ml of water. Then, biomass was resuspended in 10 ml of
water and 200 ml of stain solution were added, maintained at 62 ◦ C
for 15 min, and observed in a light microscope (400× magnifica-
tion).
2.4. Experimental design
Hadamard’s matrices with seven parameters (k = 7; N = 8) vary-
ing between a high and a low level were used to assess the influence
of mineral culture medium components on color development of
propagules and blastospores and submerged conidia production in
order to improve the culture medium. These matrices have only −1
and +1 values, and are constructed with an equal number of +1 and
−1 by column. Each column represents one variation factor (k) of
the study and each line, one experiment (N). For a given column,
+1 and −1 are the two chosen levels for the corresponding factor.
From a mathematical point of view, these matrices are orthogonal,
which guarantees the non-confusion of facts in the estimate of the
factors’.
The responses are polynomial models represented as

Yi = b0 + bi Xi (1)
with Yi = response; b0 = mean of the response; bi = coefficient;
Xi = factors studied.
The response mean value is
Y1 + Y2 + · · · + YN
b0 = (2)
N
The bi coefficients to be determined are given by
Xi Yi
bi = (3)
N
where Xi correspond to the transposed matrix.
A medium containing (per liter) 50 g of glucose, 1.84 g of
NH4 NO3 , 0.39 g of KH2 PO4 , 1.42 g of Na2 HPO4 ·12H2 O, 0.60 g
of MgSO4 ·7H2 O, 0.1 mg of KCl, 2 mg of MnSO4 ·H2 O, 10 mg of
CoCl2 ·6H2 O, 2 mg of Na2 MoO4 ·2H2 O, 20 mg of ZnSO4 ·7H2 O, 3 mg
of CuSO4 ·5H2 O, 2 mg of FeSO4 ·7H2 O and 60 mg of EDTA (disodium
salt) was used as basal medium.
The aim of the first experimental design (EP1) was to evalu-
ate the effect of salts (except NH4 NO3 , ZnSO4 ·7H2 O and EDTA) at
+1 or −1 level. In the second experimental plan (EP2); the influ-
ence of Cu2+ , Fe2+ , Mg2+ , K+ (as chloride), and Mn2+ and Mo6+ at
the new concentrations were assessed. Additionally, KH2 PO4 was
used as the sole phosphate source and evaluated as a new fac-
tor, excluding Na2 HPO4 . The third plan (EP3) was similar to the
two preceding ones, the effect of new concentrations of phosphate,
K+ (as chloride), Cu2+ , Fe2+ and the inclusion of Ca2+ , Polyethy-
lene Glycol (PEG) MW 200 (Purchased from Sigma) and EDTA
in the culture medium were assessed. Final adjustments to esti-
mate the optimal PEG concentration were made by one-at-a-time
procedure assaying different concentrations of the compound. In
these experiments, mean comparisons were conducted using the
Tukey–Kramer adjustment [21] for multiple comparisons; letters
indicating statistical significance or similarity were obtained using
NCSS 2000 software.
In all experiments pH was adjusted to 5.5 before sterilization.
Two replicate flasks were used for each trial. All experiments were
repeated at least twice.
3. Results and discussion
The maximal concentration of submerged conidia in the basal
medium was 2(±1) × 108 cell/ml and blastospores were just around
2(±1) × 107 cell/ml. The biomass had pale brown color [22]. In the
 A. Asaff et al. / Biochemical Engineering Journal 47 (2009) 87–92 89

Table 1
EP1 experimental results and plan coefficients.

Trial no. Biomass (g/l)a SC + B (cell/ml)a Colora Factors Units −1 +1 bi − b0 (g/l) (biomass) b i − b0 (cell/ml) (IP) bi − b0 (color)

1 8.13 (±0.71) 6.85 (±0.49) × 10 8

1.0 (±0.0) MgSO4 ·7H2 O g/l 0.20 1.0 −0.10 −5.83 × 10 7
−0.13
2 9.21 (±0.44) 4.30 (±0.28) × 108 8.5 (±0.7) KCl mg/l 0.50 1.50 0.01 −2.00 × 106 0.12
3 10.29 (±0.30) 5.10 (±0.14) × 108 7.0 (±0.0) MnSO4 ·H2 O mg/l 0.00 2.00 0.55 −1.02 × 108 −0.51
4 8.19 (±0.47) 3.90 (±1.13) × 108 2.0 (±1.4) CoCl2 ·6H2 O mg/l 0.00 10.00 −0.09 −5.55 × 107 −1.13
5 10.88 (±0.47) 3.50(±1.41) × 108 5.0 (±1.4) Na2 MoO4 ·2H2 O mg/l 0.00 2.00 0.14 3.55 × 107 −0.76
6 9.73 (±0.08) 1.50 (±0.97) × 108 1.5 (±0.7) CuSO4 ·5H2 O mg/l 0.00 3.00 0.27 5.18 × 107 0.25
7 9.15 (±0.16) 2.95 (±0.35) × 108 4.0 (±1.4) FeSO4 ·7H2 O mg/l 0.00 20.00 0.70 2.18 × 107 2.00
8 10.44 (±0.02) 4.05 (±0.21) × 108 4.0 (±0.0)
b0 = 9.50 b0 = 4.02 × 108 b0 = 4.13

Data correspond to means (±SD) values. Besides media contained per liter: 50.00 g of glucose, 1.84 g of NH4 NO3 , 0.39 g of KH2 PO4 , 1.42 g of Na2 HPO4 ·12H2 O, 20 mg of
a

ZnSO4 ·7H2 O and 60 mg of Na2 EDTA, initial pH 5.5. SC = Submerged conidia; B = blastospores.

present study, the culture medium was improved to increase the
concentration of submerged conidia and their pigmentation. As the
same time, the main factors of the culture medium that contribute
to pigmentation of propagules and increment of submerged conidia
concentration were identified.
Since Zn2+ was signed as one of the most important cations for
the growth of I. fumosorosea [23] and in an earlier publication was
established that a range of 10–100 ␮M Zn2+ was optimal for biomass
production [24], a concentration of 20 mg/l of ZnSO4 ·7H2 O (70 ␮M
Zn2+ ) was assumed as adequate.
3.1. First experimental plan (EP1)
As is shown in Table 1, Cu2+ and Fe2+ had positive influence
on total biomass production; thus both cations were increased in
the following experimental plan. Interestingly, the biomass was
pale-brown in those media that contained just Fe2+ or Cu2+ , but in
those were both cations were added, it was dark brown, suggesting
that both are needed to enhance pigmentation. Blastospores, sub-
merged conidia and mycelia were stained by the Fontana-Masson
procedure. Those from dark brown cultures showed strong inten-
sity, whereas the propagules from white cultures did not. Since the
Fontana-Masson stain is used to detect melanin of dematiaceous
fungi in tissues [25], we can conclude that melanins contribute
to the brown pigmentation of fungal propagules. Besides, just
brown-colored-biomass allowed to extract a dark pigment, whose
IR spectrum (Fig. 1) coincides with those reported by Moses et al.
[26] for Glycera jaws melanins.
Metabolic pathways for melanin biosynthesis include several
enzymes which require Fe2+ and Cu2+ [27–29]; thus, both cations
together should work better than each one alone.
As the coefficients of Co2+ were always negative, it was removed
from the medium. Also Mg2+ had a negative effect but since it
is essential for many fungal enzymes [30], its concentration was
diminished in the subsequent plan. Although K+ (as chloride) had
negligible effect in the response variables, given that it is included
in high quantities in other culture media [31], its concentration
was increased in the next plan. On the other hand, Mn2+ had
a positive effect on biomass (b3 = +0.55) but a highly negative
effect on submerged conidia and blastospores production (b3 =
−1.02 × 108 ) and pigmentation (b3 = −0.64); while Mo6+ showed a
slightly positive effect on biomass (b5 = +0.14) and submerged coni-
dia and blastospores production (b5 = +3.55 × 107 ) but it had a
high negative effect on pigmentation (b5 = −0.89). Therefore, the
concentration of both cations was diminished.
3.2. Second experimental plan (EP2)
Table 2 shows that KH2 PO4 as unique source of phosphate had
a slightly negative effect on biomass (b1 = −0.29) but a positive one
on submerged conidia and blastospores production (b1 = +1.50 ×
106 ) and pigmentation (b1 = +1.25); thus, it was increased in the

Fig. 1. IR absorption of pigment extracted from brown-pigmented fungal biomass of Isaria fumosorosea. Band assignments are: peak a (3400 cm−1 ) phenolic–OH stretches;
peak b (2950 cm−1 ) aliphatic stretches, CH3 and CH2 ; peak c (1600–1650 cm−1 ) aromatic C C stretches, COO stretches; peak d (1380–1400 cm−1 ) phenolic COH bends, indolic
and phenolic NH stretches; peak e (1260 cm−1 ) phenolic COH stretches; peak f (1170 cm−1 ) CO stretches; peak g (1100 cm−1 ) alcohol OH stretches, water; peak h (850 cm−1 )
aromatics, secondary amines, 500–750 cm−1 unknown.
90 A. Asaff et al. / Biochemical Engineering Journal 47 (2009) 87–92

Table 2
EP2 experimental results and plan coefficients.

Trial no. Biomass (g/l)a SC + B (cell/ml)a Colora Factors Units −1 +1 bi − b0 (g/l) (biomass) bi − b0 (cell/ml) (IP) bi − b0 (color)

1 11.62 (±1.10) 3.35 (±0.35) × 10 8

4.0 (±1.4) KH2 PO4 g/l 1.00 2.00 −0.29 1.50 × 10 6
1.25
2 11.20 (±1.13) 5.20 (±0.14) × 108 9.0 (±0.0) MgSO4 ·7H2 O g/l 0.20 0.50 0.02 2.28 × 107 −0.88
3 12.58 (±0.40) 4.50 (±0.98) × 108 4.0 (±1.4) KCl g/l 0.10 1.00 0.21 6.53 × 107 0.62
4 10.44 (±0.46) 5.90 (±1.70) × 108 7.0 (±0.0) MnSO4 ·H2 O mg/l 0.00 1.00 −0.01 −9.73 × 107 0.00
5 11.64 (±0.59) 4.25 (±0.21) × 108 8.5 (±0.7) Na2 MoO4 ·2H2 O mg/l 0.00 1.00 0.35 −4.23 × 107 −0.76
6 11.37 (±1.24) 6.35 (±1.23) × 108 8.5 (±0.7) CuSO4 ·5H2 O mg/l 3.00 6.00 0.17 1.40 × 107 −0.38
7 12.03 (±0.36) 7.90 (±0.71) × 108 5.0 (±1.4) FeSO4 ·7H2 O mg/l 20.00 30.00 0.31 −4.85 × 107 0.50
8 12.49 (±1.19) 4.15 (±0.35) × 108 7.0 (±0.0)
b0 = 11.67 b0 = 5.20 × 108 b0 = 6.63
a
Data correspond to means (±SD) values. Besides media contained per liter: 50.00 g of glucose, 1.84 g of NH4 NO3 , 20 mg of ZnSO4 ·7H2 O and 60 mg of Na2 EDTA, initial pH
5.5. SC = submerged conidia; B = blastospores.

Table 3
EP3 experimental results and plan coefficients.

Trial no. Biomass (g/l)a SC + B (cell/ml)a Colora Factors Units −1 +1 bi − b0 (g/l) (biomass) b i − b0 (cell/ml) (IP) bi − b0 (color)

1 11.34 (±0.28) 6.05 (±1.61) × 108 6.0 (±0.0) KH2 PO4 g/l 2.00 3.00 −0.18 −3.15 × 107 0.63
2 12.05 (±0.06) 3.10 (±1.84) × 108 8.0(±0.0) KCl g/l 1.00 2.00 0.13 −1.65 × 106 0.38
3 12.86 (±0.26) 4.90 (±0.28) × 108 8.5 (±0.7) CaCl2 g/l 0.00 0.10 0.28 −3.15 × 107 −0.63
4 11.12 (±0.15) 5.35 (±0.21) × 108 7.0 (±0.0) PEG mg/l 0.00 100.00 −0.87 6.35 × 107 −1.25
5 11.27 (±0.25) 5.25 (±1.06) × 108 3.0 (±0.0) CuSO4 ·5H2 O mg/l 6.00 12.00 0.08 3.10 × 107 0.50
6 13.31 (±0.53) 4.40 (±1.11) × 108 8.5 (±0.7) FeSO4 ·7H2 O mg/l 30.00 60.00 −0.02 −4.40 × 107 −0.01
7 13.72 (±0.81) 3.20 (±1.69) × 108 7.0 (±0.7) Na2 EDTA mg/l 0.00 60.00 −0.255 −2.15 × 107 −0.38
8 11.24 (±0.71) 4.05 (±0.92) × 108 6.0 (±0.7)
b0 = 12.11 b0 = 4.54 × 108 b0 = 6.75

Data correspond to means (±SD) values. Besides media contained per liter: 50.00 g of glucose, 1.84 g of NH4 NO3 , 0.20 g of MgSO4 ·7H2 O and 20 mg of ZnSO4 ·7H2 O, initial
a

pH 5.5. SC = submerged conidia; B = blastospores.

next plan. Although the diminished +1 value of Mg2+ had a pos-
itive effect on production of submerged conidia and blastospores
(b2 = +2.28 × 107 ), it still had a negative effect on pigmentation
(b2 = −0.75); therefore, the lower value (0.2 g/l) at −1 level was set
as optimal. K+ (as chloride) had a positive effect on all coefficients;
hence its concentration was increased in the next plan. Diminu-
tion of Mn2+ had no effect on biomass and pigmentation but had a
negative effect on submerged conidia and blastospores production
(b4 = −9.73 × 107 ). Also even at lesser concentration, Mo6+ had a
negative effect on pigmentation (b5 = −0.75); consequently, both
ions were excluded. On the other hand, Cu2+ and Fe2+ had a posi-
tive effect on biomass (b6 = +0.17; b7 = +0.31) but Cu2+ had a slightly
negative influence on pigmentation (b6 = −0.25) and Fe2+ on sub-
merged conidia and blastospores production (b7 = −4.85 × 107 ).
Since both cations could not interact with negative factors, like
Mn2+ and Mo6+ , higher concentrations of them were assessed in
the next plan.
3.3. Third experimental plan (EP3)
Table 3 shows that phosphate at +1 level, though positive for
color (b1 = +0.62) had a negative influence on biomass (b1 = −0.18)
and submerged conidia and blastospores production (b1 = −3.15 ×
107 ); thus its concentration was setting up at 2 g/l. The increment in
K+ (as chloride) was favorable for biomass (b2 = 0.13) and color (b2 =
+0.37) and negligible for submerged conidia and blastospores pro-
duction, therefore it was set at 2 g/l. Although Ca2+ had a positive
influence on biomass (b3 = +0.28), it had a negative effect on sub-
merged conidia and blastospores production (b3 = −3.15 × 107 )
and pigmentation (b3 = −0.63); hence, it was excluded. On the
other hand, PEG had a highly negative effect on biomass (b4 = −0.87)
and pigmentation (b4 = −1.38), but a positive one on submerged
conidia and blastospores production (b4 = +6.35 × 107 ). Besides,
fermented media containing PEG were highly homogeneous and
seemed lesser viscous than others, free mycelia was shorter and
no pellets were observed; since all these effects are desirable,
PEG concentration was further optimized by one-at-a-time strat-
egy. An increment in Cu2+ was negligible for biomass but had a
positive effect on submerged conidia and blastospores production
(b5 = −3.10 × 107 ) and pigmentation (b5 = +0.62). A subsequent
assay showed that concentrations above 12 mg/l have negligible
effects; thus this value was set as the optimal. Fe2+ and EDTA,
at +1 level, had negative effects on the three variables, therefore,
30 mg/l was considered as the optimal concentration and EDTA was
removed.
Along EP1, EP2 and EP3, other cations evaluated showed that
although they are included in several culture media, actually
its effects on I. fumosorosea biomass, submerged conidia and
blastospores production and propagule pigmentation are small,
negligible or in some cases, negative; therefore is advisable to
exclude them from the culture medium.
3.4. Final adjustments
According to the results the improved mineral medium con-
tains (per liter) 0.2 g MgSO4 ·7H2 O, 2 g KH2 PO4 , 2 g KCl, 20 mg
ZnSO4 ·7H2 O, 12 mg CuSO4 ·5H2 O and 30 mg FeSO4 ·7H2 O, with 50 g
of glucose and 1.8 g of NH4 NO3 as carbon and nitrogen sources.
Since PEG had desirable effects, assays to optimize its concentra-
tion were carried out and results are given in Table 4. As the highest
submerged conidia and blastospores production and pigmentation
were obtained in the medium containing 25 ml/l of PEG, this con-
centration was assumed as optimal. Even though mycelial pellets
were not quantified, fermented media containing small quanti-
Table 4
Effect of PEG concentration.
PEG (ml/l) Biomass (g/l) SC + B (×108 cell/ml) Color
0 12.86 (±0.35)a 4.50 (±0.42)a 9.0 (±0.0)a
25 12.88 (±0.28)a 10.72 (±1.21)b 9.0 (±0.0)a
50 12.94 (±0.30)a 7.80 (±0.71)bc 8.0 (±0.7)a
100 10.96 (±0.63)b 4.85 (±0.07)ac 4.0 (±0.7)b
Data correspond to means (±SD) values. Numbers within a column followed by
different letters are significantly different using Tukey–Kramer’s adjustment for
multiple comparisons. SC = submerged conidia; B = blastospores.
 A. Asaff et al. / Biochemical Engineering Journal 47 (2009) 87–92
Table 5
Proportion of submerged conidia and blastospores in Isaria fumosorosea biomass in different culture media.
Medium Total biomass (g/l) SC + B concentration (cell/l)
Jackson b
60.3 5.5 × 1011
Catrouxb 38.5 1.4 × 1011
Parisb 12.3 3.3 × 1011
Goralb ≈10 3.7 × 1010
Farguesc ≈10 5.0 × 1010
Improved 12.9 1.0 × 1012
a
Assuming that 1 g of SC + B has 1.2 × 1011 cells. SC = submerged conidia; B = blastospores.
b
Data were taken from Vidal et al. [19].
c
Data were taken from de la Torre and Cárdenas-Cota [32].
ties of glucose and NH4 NO3 (10 g/l and 0.5 g/l respectively) mostly
included this kind of aggregates and very few submerged conidia or
blastospores. This fact could be linked to water activity (aw ), since
media supplied with PEG avoided pellets formation and promoted
submerged conidia, blastospores and short free mycelia production,
resulting in highly homogeneous cultures (data not shown). Besides
to enhance submerged conidia and blastospores production, PEG
was signed to enhance submerged spores ecological fitness and
pathogenicity [16] as a consequence of an increment of media
osmolarity or aw diminution.
Table 5 shows that the culture medium improved in this
work allows a higher proportion of submerged conidia and blas-
tospores in the total biomass of I. fumosorosea Pfrd strain than the
medium used by de la Torre and Cárdenas-Cota [32]. Also this ratio
was higher than those estimated from data of Vidal et al., for I.
fumosorosea 92117 strain, cultured in several media, where mycelia
were the main products [19]. Although in all trials submerged coni-
dia and blastospores production were reported together, almost in
all of them, submerged conidia were the main product (80–90%).
4. Conclusions
Hadamard’s matrices allowed a fast distinction of the critical
factors that improve biomass, submerged conidia production and
propagules pigmentation by melanin formation. Co2+ , Mn2+ , Mo2+ ,
Ca2+ and EDTA had negligible or negative effects on submerged
conidia and blastospores production and their pigmentation, while
phosphate, K+ and Mg2+ had moderate positive influences. On
the other hand, the interaction between Fe2+ and Cu2+ promoted
propagules pigmentation, and PEG increased submerged conidia
yield, decreased mycelium length, thus improving medium homo-
geneity.
Also, the improved medium allows the highest submerged
conidia and blastospores production compared with other cul-
ture media, being brown-pigmented-submerged conidia the main
product. Furthermore, the fermented culture medium is highly
homogeneous, which would improve mixing and mass transfer,
mainly oxygen transfer, during fermentations, decreasing energy
consumption.
Acknowledgements
This work was supported by a grant from CONACyT (SAGARPA-
2004-C01-5) to M.T. We thank QB Socorro Vallejo Cohen for her
technical assistance.
References
[1] C. Dunlap, G. Biresaw, M. Jackson, Hydrophobic and electrostatic cell sur-
face properties of blastospores of the entomopathogenic fungus Paecilomyces
fumosoroseus, Colloids Surf. B 46 (2005) 261–266.
[2] F. Mangenot, O. Reisinger, Form and function of conidia as related to their devel-
opment, in: The Fungal Spore, Form and Function, John Wiley & Sons, New York,
1976, pp. 789–847.
91
SC + B biomass (gcell /l)a SC + B biomass proportion (%)
4.6 7.6
1.2 3.1
2.7 21.9
0.3 3.0
0.4 4.0
8.2 63.6
[3] R.I. Sánchez-Murillo, M. de la Torre-Martínez, J. Aguirre-Linares, A. Herrera-
Estrella, Light-regulated asexual reproduction in Paecilomyces fumosoroseus,
Microbiology 150 (2004) 311–319.
[4] M.J. Buttler, A.W. Day, Fungal melanins: a review, Can. J. Microbiol. 44 (1998)
1115–1136.
[5] M.S. Goettel, D.W. Roberts, Mass production, formulation and field application
of entomopathogenic fungi, in: Biological Control of Locusts and Grasshoppers,
CABI, Wallingford, UK, 1992, pp. 230–238.
[6] M.A. Jackson, M.R. McGuire, L.A. Lacey, S.P. Wraight, Liquid culture production
of desiccation tolerant blastospores of the bioinsecticidal fungus Paecilomyces
fumosoroseus, Mycol. Res. 101 (1997) 35–41.
[7] E.V. Lopez-y-Lopez, N. Chavarria-Hernandez, P. Fernández-Sumano, M. de la
Torre, Fermentation processes for bioinsecticide production. An overview,
Recent Res. Dev. Biotechnol. Bioeng. 3 (2000) 1–20.
[8] J. Leland, D. Mullins, L. Vaughan, H. Warren, Effects of media composition on
submerged culture spores of the entomopathogenic fungus, Metarhizium aniso-
pliae var. acridum. Part 1. Comparison of cell wall characteristics and drying
stability among three spore types, Biocontrol. Sci. Technol. 15 (2005) 379–392.
[9] D.J. Holder, N.O. Keyhani, Adhesion of the entomopathogenic fungus Beau-
veria (Cordyceps) bassiana to substrata, Appl. Environ. Microbiol. 71 (2005)
5260–5266.
[10] Y. Sheng-Hua, F. Ming-Guang, Means to mediating accumulation of
hydrophobin-like proteins in the wall of Beauveria bassiana conidia for
improved tolerance to thermal stress, J. Gen. Appl. Microbiol. 53 (2007)
309–314.
[11] V.M. Tereshina, Thermotolerance in fungi: the role of heat shock proteins and
trehalose, Microbiology 74 (2005) 247–257.
[12] M. Andersen, N. Magan, A. Mead, D. Chandler, Development of a population-
based threshold model of conidial germination for analyzing the effects of
physiological manipulation on the stress tolerance and infectivity of insect
pathogenic fungi, Environ. Microbiol. 8 (2006) 1625–1634.
[13] N. Magan, Physiological approaches to improving the ecological fitness of fungal
biocontrol agents, in: Fungi as Biocontrol Agents, CAB International, Walling-
ford, Oxford, UK, 2001, pp. 239–251.
[14] F. Vega, M. Jackson, G. Mercadier, T. Poprawski, The impact of nutrition on spore
yields for various fungal entomopathogens in liquid culture, World J. Microbiol.
Biotechnol. 19 (2003) 363–368.
[15] N. Issaly, H. Chauveau, F. Aglevor, J. Fargues, A. Durand, Influence of nutrient, pH
and dissolved oxygen on the production of Metarhizium flavoviride Mf189 blas-
tospores in submerged batch culture, Process Biochem. 40 (2005) 1425–1431.
[16] J.E. Leland, D.E. Mullins, L.J. Vaughan, H.L. Warren, Effects of media composition
on submerged culture spores of the entomopathogenic fungus, Metarhizium
anisopliae var. acridum. Part 2. Effects of media osmolality on cell wall char-
acteristics, carbohydrate concentrations, drying stability, and pathogenicity,
Biocontrol. Sci. Technol. 15 (2005) 393–409.
[17] M. Kennedy, D. Krouse, Strategies for improving fermentation medium perfor-
mance: a review, J. Ind. Microbiol. Biotechnol. 23 (1999) 456–475.
[18] M.D. White, Z.P. Shalita, D. Marcus, S. Reuveny, A systematic route for opti-
mization of media composition and growth conditions, in: Biologicals From
Recombinant Microorganisms and Animal Cells: Production and Recovery, 34th
Oholo Conference, Eilat, Israel, 1990, pp. 101–116.
[19] C. Vidal, J. Fargues, L.A. Lacey, M.A. Jackson, Effect of various liquid culture
media on morphology, growth, propagule production, and pathogenic activ-
ity to Bemisia argentifolii of the entomopathogenic Hyphomycete, Paecilomyces
fumosoroseus, Mycopathologia 143 (1998) 3346.
[20] V. Sava, B. Galkin, M. Hong, P. Yang, G. Huang, A novel melanin-like pigment
derived from black tea leaves with immuno-stimulating activity, Food Res. Int.
34 (2001) 337343.
[21] C. Kramer, Extension of multiple range tests to group means with unequal
numbers of replications, Biometrics 12 (1956) 309–310.
[22] A. Asaff, Insecticidal metabolites production by Paecilomyces fumosoroseus
in solid and submerged culture, Doctoral thesis, Universidad Autónoma
Metropolitana-Iztapalapa, México D.F., México, 2005.
[23] S. Cliquet, M.A. Jackson, Influence of culture conditions on production and
freeze-drying tolerance of Paecilomyces fumosoroseus blastospores, J. Ind.
Microbiol. Biotechnol. 23 (1999) 97102.
[24] A. Asaff, C.M. Cerda-García-Rojas, G. Viniegra-González, M. de la Torre, Carbon
distribution and redirection of metabolism in Paecilomyces fumosoroseus during
solid-state and liquid fermentations, Process Biochem. 41 (2006) 13031310.
92 A. Asaff et al. / Biochemical Engineering Journal 47 (2009) 87–92
[25] M. Kimura, M.R. McGinnis, Fontana-Masson-stained tissue from culture-proven
mycoses, Arch. Pathol. Lab. Med. 122 (1998) 1107–1111.
[26] D. Moses, J. Harreld, G. Stucky, J. Waite, Melanin and Glycera jaws: emerg-
ing dark side of a robust biocomposite structure, J. Biol. Chem. 281 (2006)
34826–34832.
[27] N.E. Zhukhlistova, Y.N. Zhukova, A.V. Lyashenko, V.N. Zaitsev, A.M. Mikhailov,
Three-dimensional organization of three-domain copper oxidases: a review,
Crystallogr. Rep. 53 (2008) 92109.
[28] H. Decker, T. Schweikardt, D. Nillius, U. Salzbrunn, E. Jaenicke, F. Tuczek, Similar
enzyme activation and catalysis in hemocyanins and tyrosinases, Gene 398
(2007) 12.
[29] H.P. Hersleth, U. Ryde, P. Rydberg, C.H. Goerbitz, K.K. Andersson, Structures of
the high-valent metal-ion haem-oxygen intermediates in peroxidases, oxyge-
nases and catalases, J. Inorg. Biochem. 100 (2006) 460476.
[30] E. Moore-Landecker, Fundamentals of the Fungi, Prentice-Hall, Inc., NJ, USA,
1996.
[31] S. Paris, Etude physiologique, biochimique et génétique des caractères de Beau-
veria brongniartii (Sacc.) Petch liés à la pathogénicité de ce champignon pour
le hanneton común Melolontha melolontha, Doctorat thesis, University Paris XI,
Paris, France, 1980.
[32] M. de la Torre, H. Cárdenas-Cota, Production of Paecilomyces fumosoroseus coni-
dia in submerged culture, Entomophaga 41 (1996) 443–453.