Influence of initial L-asparagine and glycerol concentrations on the batch growth kinetics of Mycobacterium bovis BCG

Abstract

The influences of the L-asparagine and glycerol initial concentrations in Sauton medium on the productivities of biomass and colony forming units were studied. The submerged batch cultivations of Mycobacterium bovis were carried out in a 20 L bioreactor. The L-asparagine and glycerol initial concentrations of 4.54 g/L and 25 mL/L, respectively, corresponded to the best biomass productivity, namely 2.5 g/L.day. On the other hand, the concentrations of 2.27 g/L and 25 mL/L, respectively, led to the highest productivity in terms of colony forming units, namely 2.7. 10(6) colonies/mg.day. In addition, by means of the relative consumption analysis of L-asparagine and glycerol (50 and 26% respectively), it was concluded that the concentrations of such components could be reduced, with respect to the original Sauton medium composition, aiming the obtainment of an optimal BCG vaccine production in the bioreactor.

Full text

Brazilian Journal of Microbiology (2004) 35:337-344
ISSN 1517-8382
337
INFLUENCE OF INITIAL L-ASPARAGINE AND GLYCEROL CONCENTRATIONS ON THE
BATCH GROWTH KINETICS OF MYCOBACTERIUM BOVIS BCG
Maria Betania Batista Leal; Júlia Baruque-Ramos*; Haroldo Hiss; Marcelo Fossa da Paz; Maria Cristina Sakai;
Umbelina Macedo Vassoler; Luciana Juncioni de Arauz; Isaías Raw
Instituto Butantan, Centro de Biotecnologia, São Paulo, SP, Brasil
Submitted: December 08, 2003; Returned to Authors: August 23, 2004; Approved: September 27, 2004
ABSTRACT
The influences of the L-asparagine and glycerol initial concentrations in Sauton medium on the productivities
of biomass and colony forming units were studied. The submerged batch cultivations of Mycobacterium
bovis were carried out in a 20 L bioreactor. The L-asparagine and glycerol initial concentrations of 4.54 g/L and
25 mL/L, respectively, corresponded to the best biomass productivity, namely 2.5 g/L.day. On the other hand,
the concentrations of 2.27 g/L and 25 mL/L, respectively, led to the highest productivity in terms of colony
forming units, namely 2.7·10
6
colonies/mg.day. In addition, by means of the relative consumption analysis of
L-asparagine and glycerol (50 and 26% respectively), it was concluded that the concentrations of such
components could be reduced, with respect to the original Sauton medium composition, aiming the obtainment
of an optimal BCG vaccine production in the bioreactor.
Key words: BCG, Mycobacterium bovis, culture medium, vaccine, submerged batch cultivation
INTRODUCTION
The tubercle bacillus (Mycobacterium tuberculosis) can
grow in almost all human tissues, such as bones, joints and
genito-urinary tract. However, the majority of infections of
the human variety involve problems, often exclusively in the
lungs (1).
After a period of decreased disease cases, from the end of
the Second World War to the end of the 70s, one can observe
an alarming increase of new tuberculosis cases. In the last
decade, 1990-1999, 88 million new cases of tuberculosis were
estimated in the world (among them 8 million attributed to
infection by the HIV) (2).
Calmette and Guérin, assuming that the acquired immunity
depends upon the persistence of living bacilli in the body,
developed a vaccine against tuberculosis from a virulent strain
of Mycobacterium bovis (isolated from milk) after 230 passages
on a glycerol-bile-potato medium for a period of 13 years (from
1908 to 1920). After that period, the strain became avirulent and
hence was called BCG (Bacille Calmette-Guérin) (1). Besides its
action as a vaccine against tuberculosis, the BCG has proved to
be a non-specific stimulant of immune systems and it is one of
the clinic immunotherapic forms approved for the treatment
against superficial cancer of the bladder in the United States (3).
In Brazil, the sub-strain BCG-Moreau, received by Arlindo
de Assis in 1927 from the Pasteur Institute in Paris (4), was
initially used. Later, the sub-strain Copenhagen, currently used
in the production at Instituto Butantan, was used. In this
production process, which has continued until the present, the
BCG is cultivated in glass flasks on a shallow layer of non-
stirred Sauton medium (static cultivation) (5). The microorganism
grows as a coherent pellicle on the liquid medium surface. After
a pressing of the bacterial pellicle in a Birkhaug apparatus to a
damp cake (containing approximately 20% dry weight of cells),
a suspension is prepared in a 2% sodium glutamate solution by
shaking the cake in the presence of steel balls. This procedure
causes a destruction of 40 to 60% of the cells with a consequent
decrease in the vaccine efficiency (6).
*Corresponding author. Mailing address: Instituto Butantan, Centro de Biotecnologia. Av. Dr. Vital Brazil, 1500, Butantã. 05503-900, São Paulo, SP,
Brasil. Tel.: (+5511) 3726-7222, Ext. 2182. Fax: (+5511) 3726-1505. E-mail: jbaruque@butantan.gov.br

338
M.B.B. Leal et al.
The purpose of this work is to study the production of the
BCG vaccine in dispersed cultivation in the bioreactor
(avoiding thus the further bacillary dispersion with steel balls).
From this, a desirable increase of the viable cells follows, in
comparison to the static cultivation. As described by Gheorghiu
(7), the submerged cultivation also produces bacteria which
are more resistant to the freeze-drying process and storage at
room temperature than the bacteria produced by static
cultivation. More specifically, the purpose of this work is to
study the influence of the initial L-asparagine and glycerol
concentrations in Sauton medium on growth, in order to adapt
the composition of this medium for the submerged batch
cultivation.
MATERIALS AND METHODS
Microorganism
For the preliminary runs, carried out in shaken flasks,
ampoules of BCG were employed as inoculum (Moreau strain
from Copenhagen - secondary seed lot), each containing 5 mg
of the lyophilized microorganism dated July 1978.
For the runs in bioreactor, ampoules from the Production
Laboratory of the Instituto Butantan were used (lyophilized -
lot 3/95 BCG ID).
Culture medium
The Sauton medium was used (5) for the inoculum and
bioreactor cultivations with the following composition: L-
asparagine, 4.54 g/L; citric acid, 2.0 g/L; K
2
HPO
4
, 0.5 g/L;
MgSO
4
.7H
2
O, 1.0 g/L; ferriammonium citrate, 0.05 g/L; glycerol,
60.0 mL/L (v/v). The pH was adjusted to 7.2 with a 4 N NaOH
solution.
The cultivation media were sterilized by filtration through a
0.2 µm membrane for preliminary experiments 1 to 4. For the
assays carried out in the bioreactor (5 to 13) the cultivation
media were sterilized inside the vessel at 121ºC for 30 min. Besides
the fact that the sterilization by wet heat is easier than filtration,
it was verified that heating the medium did not cause alterations
in the growth kinetics (8).
The modifications of the initial L-asparagine and glycerol
concentrations with respect to the original Sauton medium
composition are expressed in Table 1.
Table 1. Factorial planning design and result analysis for preliminary experiments 1 to 4 carried out in shaker.
(X1) Tween-80 (v/v): 0 (-)/0.025 % (+)
(X2) Glycerol (mL/L): 25 (-)/60 (+)
1
(X3) L-asparagine.H2O (g/L): 2.27 (-)/4.54 (+)
1
Erlenmeyer Average Factorial Planning of Assays Interaction Effects Dry
Number Calculation
(X1) (X2) (X3) X1X2 X1X3 X2X3 X1X2X3
Biomass
2
1++++++++0
2+++-+--- 0
3++-+-+-- 0
4++----++ 0
5+-++--+-4.59
6+-+--+-+3.38
7+--++--+3.16
8 + --- +++ - 2.76
Divisor 8 4 4 4 4 4 4 4 ——
Average= 1.74± 0.25
Effect of (X1)= -3.48±0.51
Effect of (X2)= 0.51±0,51
Effect of (X3)= 0.40±0.51
Interaction Effect (X1X2)= -0.51±0.51
Interaction Effect (X1X3)= 0.40±0.51
Interaction Effect (X2X3)= -0.20±0.51
Interaction Effect (X1X2X3)= -0.20±0.51
1
Original concentrations in Sauton medium: 60 mL/L and 4.54 g/L of glycerol and L-asparagine respectively;
2
Average of experiments 1 to 4. Expressed in total dry biomass per erlenmeyer contents (g).

Growth kinetics of M. bovis
339
Preparation of inoculum
In a 300 mL erlenmeyer flask, 100 mL of Sauton medium were
inoculated with the contents of one ampoule. The flask was
agitated on a rotatory shaker at 198 min
-1
, 37ºC, for 13 days.
This culture was used as inoculum for the preliminary runs in
shaken flasks and the same procedure was adopted for the
inoculation of the bioreactor.
Cultivation conditions
Preliminary cultivation runs
Eight cultivation runs were carried out according to the
factorial planning design 2
3
(9) (see Table 1), repeated four times
(assays 1 to 4), each one in shaken flasks of 300 mL of capacity
with 100 mL of Sauton medium, agitated at 198 min
-1
and
temperature controlled at 37 ± 0.5ºC. The values adopted for the
initial concentrations of L-asparagine, glycerol and Tween 80 in
the Sauton medium are also listed in Table 1. The Tween 80 was
used as a dispersing agent.
The analysis of the influence indicated in Table 1 was based
on the dry biomasses formed in each flask contents after 13
days of cultivation.
Cultivation runs in bioreactor
The experiments were carried out in a 20 L bioreactor (LSL
Biolafitte S.A., Saint Germain en Laye, France) constituted of a
stirred cylindrical vessel of borosilicate glass with a half
spherical bottom (23 cm diameter) and a top-driven agitator. A
marine propeller promoted the agitation with six blades set at a
45º angle. This propeller was located 15 cm from the vessel
bottom and the aeration was done at the surface medium into
the vortex formed by the absence of baffles. The initial medium
volume was near 12 L, the aeration rate corresponded to 15 L/
min, the cultivation temperature was 37ºC and the manometric
head vessel pressure was 4 cm Hg. The agitation frequency
was 1200 min
-1
for experiment 5 only and 840 min
-1
for all other
experiments. The initial pH of the Sauton medium was previously
adjusted to 7.2 with a 4N NaOH solution.
The initial adopted concentrations of L-asparagine and
glycerol were equal to those of the preliminary experiments and
a factorial planning design 2
2
was done as shown in Table 2.
Analytical methods
Cell Concentration. Expressed as dry biomass, determined
by filtration of a sample in quantitative paper, followed by the
pellet drying at 60ºC for 48 h. The obtained cell concentration
value, multiplied by the medium volume, was transformed into
total dry biomass (M
x
) in order to exclude the medium
evaporation effect on the mentioned concentration (10).
Glycerol Concentration. The method used (11) was based
on the glycerol oxidation by sodium periodate, in which glycerol
reacts with sodium periodate in acid solution generating
aldehyde and formic acid. The formic acid was titrated with a
NaOH solution (0.125 N) and the volume spent corresponded
Table 2. Factorial planning design and result analysis for experiments 5 to 6 carried out in bioreactor.
(X2) Glycerol (mL/L): 25 (-)/60 (+)
(X3) L-asparagine.H
2
O (g/L): 2.27 (-)/4.54 (+)
Assay
Agitation
µ
max
P
CFU
.10
-6
P
Mx
L-asparagine Glycerol
Number
X2 X3 X2X3 Frequency
(day
-1
) (CFU/mg.day) (g/day)
consumption consumption
(min
-1
) (%) (%)
5 + + + 1200 0.08 —— 0.16 20.3 19.6
6 + + + 840 0.44 —— 2.25 30.5 23.0
7 + + + 840 0.24 1.68 1.83 41.1 27.7
8 - + - 840 0.38 1.21 2.47
2
38.6 33.3
9 - + - 840 0.41 0.25 2.53
2
34.7 30.8
10 + - - 840 0.33 1.42 2.41 42.1 25.0
11 + - - 840 0.40 —— 2.16 32.5 16.2
12 - - + 840 0.33 2.01
1
2.24 49.6
3
21.3
4
13 - - + 840 0.30 3.35
1
2.17 51.1
3
30.7
4
1
Maximum average for P
CFU
= 2.68.10
6
UFC/mg.day;
2
Maximum average for P
Mx
= 2.50 g/day;
3
Maximum average for L-asparagine consumption = 50%;
4
Maximum average for glycerol consumption = 26%.

340
M.B.B. Leal et al.
to the glycerol concentration. This result multiplied by the
medium volume was transformed into total mass of glycerol
(M
G
).
Supernatant Total Nitrogen Concentration. Estimated using
the Semi-Micro Kjeldahl method (12) through digestion with
sulfuric acid and copper sulfate as catalyst for 4 hours. The
ammonium sulfate formed was distilled in the presence of NaOH
10N to discharge the ammonia vapor (13), which was absorbed
in a boric acid solution in the presence of a mixed indicator
followed by titration with a standard solution of HCl. The HCl
expended corresponded to the nitrogen in the sample, according
to the equation:
Nit.(g/L) = 1.401.10
7
. V . N (1)
Al
where,
V = HCl expended volume (mL);
N = HCl normality;
A1 = sample volume (mL).
This value, multiplied by the medium volume, was
transformed in total supernatant nitrogen mass (M
N
).
Colony Forming Units (CFU). Determined by a counting of
the colonies grown in Loewenstein-Jensen agar-medium (4) at
37ºC for 28 days, of a sample previously diluted (1:4) with
sterilized Sauton medium. The calculation was based on the
statistic method recommended by the OMS (14).
Percentage of Dissolved Oxygen. Determined by an on-line
polarographic probe (Ingold model Oxygraf 400) installed and
sterilized together with the vessel. The saturation point (100%)
was calibrated before the inoculation, under the same cultivation
conditions (37ºC, 4 cm Hg head vessel pressure, agitation as
the cultivation condition).
Sterility Test. This control was done through the inoculation
of 0.5 mL of the sample in a test tube containing Brewer medium
(4), maintained at 37ºC (for bacteria presence detection), and in
another tube with casein soya broth (15) maintained at room
temperature (for fungi detection). The tubes were inoculated
for a period of one week or more.
Two staining methods: one, the Ziehl-Neelsen (4) for
microscopic mycobacteria detection and the other, the Gram
test (4) for eventual observation of contaminants, were also
employed.
Maximum specific growth rate (
µµ
µµ
µ
max
). It was calculated
according to the following equations for the exponential
phase:
log (M
x
) = a ⋅ (t-t
i
) + b (2)
ln (M
x
) = 2.303 ⋅ a ⋅ (t-t
i
) + 2.303 ⋅ b (3)
µ
max
= 2.303 ⋅ a = (1/M
x
) ⋅ (dM
x
/dt) (4)
where,
M
x
(g) has the same meaning as defined in analytical
methods at the cultivation time “t” (day);
“a” is defined by equation 4;
“b” represents the log of biomass M
xi
at beginning of
exponential phase t
i
.
Dry biomass productivity (P
Mx
). The time of the exponential
growth phase end was taken to calculate the dry biomass
productivity:
P
Mx
= (M
Xc
– P
Xi
) (5)
(T
c
– t
i
)
where,
M
Xc
- total dry biomass at the exponential growth phase
end (g);
T
c
- time of the exponential growth phase end (day).
CFU productivity (P
CFU
). The minimum values observed of
dissolved oxygen, during the process of cultivation coincided
with the maximum values of CFU. In this way, this productivity
was calculated in a similar way of equation (5):
P
CFU
= (CFU
m
– CFU
i
) (6)
(t
O
2
min
– t
i
)
where,
CFU
m
- maximum value of colony forming units,
corresponding to the minimum dissolved oxygen in
the medium, obtained after a semi-logarithmic fit
between CFU and fermentation time (colonies/mg).
CFU
i
- value of colony forming units at the beginning of
the exponential growth phase (colonies/mg);
t
O2min
- cultivation time corresponding to the minimum value
of dissolved oxygen in cultivation medium (day).
RESULTS
Analysis of the effects in the factorial planning of the
preliminary assays
The calculated effects and their interactions are presented in
Table 1. Considering that the highest negative effect is caused by
X1, the conclusion is that the Tween-80 exerts an inhibitory effect
on the cellular growth at an initial concentration of 0.025% (v/v).

Growth kinetics of M. bovis
341
Studies carried out by Sakai (8) show that such effect depends on
the relation: Tween-80 concentration / cellular concentration.
As for the L-asparagine and glycerol, the effects (X2 and
X3) are positive for the highest concentrations of these
components (Table 1: 4.54 and 60 mL/L respectively). Their
differences, however, are not significantly high.
These preliminary conclusions have, therefore, allowed for
the planning of assays 5 to 13, carried out in submerged
cultivations in the bioreactor. In these experiments no Tween-
80 was added, but the levels of concentration for L-asparagine
and glycerol were maintained.
Analysis of results obtained in experiments carried out in
bioreactor
The results are shown in Figs. 1 to 3 and Table 2.
A similar cellular growth behavior was observed in all
experiments, as it can be seen by comparing the growth curves.
The exception was cultivation run 5 (Fig. 1) with a near absence
of growth. In the same way, comparing the values of productivity
(P
Mx
) and maximum specific growth rate (µ
max
), shown in Table
2, they are lower for run 5 than for the other assays. The agitator
frequency of 1200 min
-1
in run 5, higher than in the other
experiments (840 min
-1
), explains such differences since
vigorous mixing provokes mechanical stress, which is
detrimental to growth (1).
The CFU productivity was based on the minimum dissolved
oxygen value (P
CFU
) because at this point it was the maximum
viable count (8,16,17). Such behavior was also confirmed in this
work, as it can be seen in figure 4: the maximum value of CFU
corresponds to the minimum value of oxygen concentration in
the medium since the population of viables is responsible for
the oxygen uptake. In addition, once a relationship between the
number of living bacilli and the immunogenic potency was
verified (18), the maximum value of colony forming units (CFU
m
)
is important for the calculation of the related productivity, which,
Figure 1. Total dry biomass (M
x
) obtained by the multiplication
of the cell concentration value and the medium volume for
experiments in bioreactors 5 to 13.
Figure 2. Total glycerol mass (M
G
) obtained by the multiplication
of the glycerol concentration value and the medium volume for
experiments in bioreactors 5 to 13.
Figure 3. Total supernatant nitrogen mass (M
N
) obtained by
the multiplication of the nitrogen concentration value and the
medium volume for experiments in bioreactors 5 to 13.

342
M.B.B. Leal et al.
therefore, corresponds to the cultivation harvest time and
consequently to a highly immunogenic BCG.
The results in Table 2 show that the lowest initial glycerol
and the highest L-asparagine concentrations implied the
maximum dry biomass productivity (runs 8 and 9, P
Mx
=2.50 g/
day), but not the maximum CFU productivity (runs 12 and 13,
P
CFU
=2.68⋅10
6
CFU/mg.day).
Moreover, concerning relative consumption of substrates,
shown in figures 1 to 3, it is possible to estimate the averages of
global yield factors (Y
X/G
=0.14 and Y
X/N
=5.25 respectively),
which led to an average relation (carbon/nitrogen) namely C/N
≅16 (w/w). The maximum variation coefficient for these
calculated averages was 30.0%.
DISCUSSION
This microorganism generally grows as a surface veil in non-
agitated culture flasks. In Instituto Butantan, the BCG lyophilized
vaccine is supplied in vials with 10 doses, containing 1 mg of
BCG lyophilized and between 2 to 10 million of live bacilli each.
The production is 70,000 vials or 700,000 doses. For this purpose,
96 L of Sauton medium are prepared weekly and distributed in
aliquots of 80 mL (in flasks of 300 mL), totaling 1,200 flasks for the
static cultivation of Mycobacterium bovis BCG (19). This
represents an approximated concentration of 0.7 g of bacteria per
liter of medium. In the present work higher values of dry biomass
productivity and CFU productivity are obtained (Table 2). These
data are truly promising regarding process transposition from
static to submerged cultivation in bioreactors.
The increase of the BCG production is a public necessity
considering it is the most widely used vaccine in the world against
tuberculosis. Besides this, it is a useful vaccine for delivering
protective antigens of multiple pathogens (20). Since the early
90’s the BCG has been studied for this purpose because of its
unique characteristics, including low toxicity adjuvant potential
and long-lasting immunity (21). These recombinant BCG strains
can elicit long-lasting humoral and cellular immune responses to
foreign antigens (22). Several studies were done with this
approach for developing vaccines against diverse diseases: AIDS
(23), hepatitis C (24), leishmaniosis (25) and several others. In
accordance with this approach, Instituto Butantan recently
developed a recombinant Mycobacterium bovis BCG expressing
the Sm14 antigen of Schistosoma mansoni which protected mice
submitted to cercarial challenge tests (26). Also, it is remarkable
that in Instituto Butantan there is another research line studying
the use of recombinant proteins for developing a vaccine against
tuberculosis which is more potent than the classical one (27).
However, very little data are found in the literature related to
suitable BCG process production in bioreactors. A suitable small
scale cultivation in the bioreactor has been recently described
(28). It was performed in a 3 L total capacity vessel, employing 2
L of Sauton medium, at a cultivation temperature of 37ºC, dissolved
oxygen controlled above 20% saturation and agitation frequency
varying from 360 to 500 rpm. The microorganism was M. bovis -
BCG (Copenhagen strain) cultivated for 164 hs (6 days and 20 h).
The maximum cell concentration achieved was equivalent to
an optical density (measured at 600 nm) of 2.2 units and a CFU
value of 1.02⋅10
9
/mL. According to that report, the bioreactor-
grown BCG bacteria exhibited a similar vaccine efficacy against a
challenge (in tests with mice) with M. tuberculosis with regard to
the BCG obtained from the classical surface-grown culture. Thus,
through those biological tests, the real possibility of process
transposition from static to submerged cultivation in bioreactors
was proven. However, still considering the great importance of
the information contained in that paper, their focus was not on
showing a clear growth curve, neither a clear dissolved oxygen
variation in the medium. Furthermore, a kinetic study of carbon
and nitrogen source consumptions was not presented.
For scale-up bioreactor cultivations, besides the c-GMP
norms, which recommend the use of chemically, defined medium
(29), the use of an optimized cultivation medium is advisable for
economic reasons. Thus, the present study describes, in a larger
scale, the preliminary kinetic analysis in order to contribute to
the search of such optimization.
In the present work, slow bacterial growth (average of µ
max
= 0.35 day
-1
- runs 6 to 13) was observed, even in the bioreactor
(Fig. 1) where the agitation is very efficient. The slow growth of
many mycobacteria is probably due to the hydrophobic
character of the cellular wall surface, which makes the transport
of the nutrients through the bacterial cell membrane difficult
(30). On the other hand, being the high lipid content a
characteristic of the mycobacteria (constituting 60% of the
cellular wall), the addition of glycerol and lipids to the cultivation
Figure 4. Dissolved oxygen in the medium (O
2
) and colony
forming units (CFU) as a function of cultivation time.

Growth kinetics of M. bovis
343
medium favors cellular growth (4), thus justifying the use of
glycerol in the cultivation medium. This specific necessity of
carbon source related to the high bacterial lipid content was
also evident in the high relation carbon/nitrogen as assessed in
the batch cultivations (see previous section).
In addition, there is a need for high initial concentrations of
the medium components in static cultures because the transport
of nutrients to the microorganism/medium interface occurs by
molecular diffusion, which in turn depends on the concentration
gradients. On the other hand, in the submerged culture with
constant stirring, such high concentration values are not
necessary because continuous mixing facilitates the transport.
An excessive presence of certain medium components may cause
the inhibition of bacterial growth. The results in Table 1 show
this effect on the final dry biomass production.
The detergent Tween-80 exerts an evident inhibitory effect
on the cellular growth at concentration of 0.025% (v/v) as
demonstrated in the result analysis (Table 1). For this reason,
the Tween-80 was not used in bioreactor cultivations. However,
it must be considered that this detergent exerts two distinct and
antagonistic effects: it stimulates bacterial growth by allowing
higher access of the microorganisms to the nutrients through
reduction of the bacterial clump size and consequent increase
of the specific area; on the other hand, it inhibits bacterial growth,
probably through the oleic acid action, which is toxic for the
bacilli. The inhibition depends on the relation between the
concentrations of Tween-80 and the cells (8). Curiously, in the
paper cited previously (28), the authors affirm that in their
process conditions the optimal dissociation of the bacilli was
achieved at 0.3% Tween-80. This value is greater than that
employed in the preliminary assays carried out in flasks. Thus,
although its use was discarded in bioreactor cultivations in the
present work, these apparently contradictory data lead to the
supposition that the Tween-80 concentration must be optimized
in future studies according to the process conditions adopted.
The results in Table 2 show that the lowest initial glycerol and
the highest L-asparagine concentrations implied the maximum
dry biomass productivity (P
Mx
=2.50 g/day), but not the maximum
CFU productivity. The condition above could be recommended
for purified protein derivative (PPD) production, which is
associated to biomass production. If, however, the vaccine
production is the aim, where its efficiency depends on the number
of viable units, the condition of lowest initial concentrations of
L-asparagine and glycerol possibly represents the recommended
composition of the cultivation medium (P
CFU
=2.68⋅10
6
CFU/
mg.day). Of course that appropriate product tests must be still
carried out to confirm or not the considerations above.
Figs. 2 and 3 show incomplete consumptions of glycerol
and nitrogen, considering that residual values of M
G
and M
N
were present after 10 days of cultivation. Based on the lowest
level of initial glycerol concentration (Table 2), the relative
consumptions of M
N
for pairs of experiments 12, 13 and 8, 9,
corresponded to 50% and 36.7% respectively. For the highest
level of this substance (pairs of assays 6, 7 and 10, 11) the
inhibitory effect of L-asparagine was not disclosed, taking into
account that the relative consumption of this last substrate
practically did not vary (36.0 and 37.3% respectively).
Thus, as already described above, considering the best
value of productivity in CFU (P
CFU
= 2.68⋅10
6
CFU/mg.day) and
the presence of L-asparagine and glycerol in their lowest initial
concentrations (assays 12 and 13), where the average consumed
is nearly 50 and 26% respectively, one concludes that the
reduction of concentrations of these substrates in relation to
the original Sauton medium composition leads to the
improvement of productivity in CFU. This way, if one takes the
possibility to reduce the initial levels of these two components
even more as a starting point, future studies will be carried out
in order to optimize the Sauton medium, aiming at an adequate
BCG vaccine production in the bioreactor.
ACKNOWLEDGMENTS
Financial support from Fundação Butantan is gratefully
acknowledged. We would like to thank CAPES and FUNDAP
for the research studentships. The authors would also like to
thank Mr. Lourivaldo Inácio de Souza, Mr. Máximo de Moraes,
Mr. Hélio Fernandes Chagas, Ms. Inês do Amaral Maurelli, Ms.
Salete Vargas and Ms. Fátima Aparecida Mendonça de Oliveira
for their technical support.
RESUMO
Influência das concentrações iniciais de asparagina e
glicerol sobre a cinética de crescimento submerso de
Mycobacterium bovis
Estudou-se a influência das concentrações iniciais, no meio
de Sauton, de asparagina e glicerol sobre as produtividades,
expressas em unidades formadoras de colônias e biomassa
microbiana, referentes aos cultivos submersos do Mycobacterium
bovis, em biorreator de 20 mL. As concentrações iniciais de 2,27
e 25 mL/L de asparagina e glicerol, respectivamente, conduziram
à maior produtividade, em unidades formadoras de colônias, a
saber 2,7.10
6
colônias/mg.dia. Por outro lado, as concentrações
de 4,54 e 25 mL/L dos mesmos componentes, corresponderam à
melhor produtividade em biomassa, a saber: 2,5 g/dia. Através
das análises dos consumos relativos de asparagina e glicerol (50
e 26% respectivamente), verificou-se também que as
concentrações destes componentes podem ser reduzidas na
composição original do meio de Sauton, com o objetivo de obter
uma produção otimizada de vacina BCG em bioreator.
Palavras-chave: BCG, Mycobacterium bovis, meio de cultura,
vacina, cultivo submerso descontínuo

344
M.B.B. Leal et al.
REFERENCES
1. Hemert, V. Vaccine production as a unit process. In: Progress in
Industrial Microbiology. Churchil Livingstone, Edinburgh and
London, 1974, v.13, p.227-233.
2. Dolin, P.S.; Raviglione, M.C.; Kochi, A. Global tuberculosis incidence
and mortality during 1990-2000. Bull. World Health Org., 72(2):213-
220, 1994.
3. Groves, M.J. Pharmaceutical characterization of Mycobacterium
bovis bacillus Calmette-Guérin (BCG) vaccine used for the treatment
of superficial bladder cancer. J. Pharm. Sciences, 82(6):555-562,
1993.
4. Bier, O. Microbiologia e Imunologia. 24
th
ed. Ed. Melhoramentos,
São Paulo, 1985, 1234p.
5. Cassagne, H. Millieux de Culture. In: Collection “Techniques de
Base”. Tome 2. Éditions de la Tourelle Saint-Mandé, Paris, 1961,
p.242.
6. Kim, T.H. High-viability lyophilized bacilli Calmette-Guérin vaccine
produced by deep-culture technique. Appl. Envir Microb., 34(5):495-
499, 1977.
7. Gheorghiu, M.; Lagrange, P.H.; Fillastre, C. The stability and
immunogenicity of a dispersed-grown freeze-dried Pasteur BCG
vaccine. J. Biol. Stand., 16:15-26, 1988.
8. Sakai, M.C.; Hiss, H. Influence of surface-active agent concentration
on the spectrophotometric biomass estimation during submerged
growth of Mycobacterium bovis. Braz. J. Chem. Eng., 14(3):299-
302, 1997.
9. Box, G.E.P.; Hunter, W.G.; Hunter J.S. Statistics for Experimenters.
John Wiley & Sons, New York, 1978, 653p.
10. Borzani, W.; Baralle, S.B.; Correction of results obtained in
laboratory-scale studies of batch cultivation kinetics. Biotech.
Bioeng., 25:3201-3206, 1983.
11. A.O.C.S. Method Ea 6-51. In: Official Methods and Recommended
Practices of the American Oil Chemistry Society, AOCS, Illinois,
1987, p.6-51.
12. Lima, L.S. et al. Doseamento de Nitrogênio. In: Farmacopéia dos
Estados Unidos do Brasil, Brazilian Health Ministry, Brasilia, 1959,
p.960-967.
13. Kemmerer, G.; Hallett, L.T. Improved micro-Kjeldahl ammonia
distillation apparatus. Ind. Eng. Chem., 19:1295-1296, 1927.
14. O.M.S. Revised Requirements for Dried BCG Vaccine. In: Vitro Assays
of BCG Products. Org. Mond. Santé (Ser. Rapp. Techn.), Geneve,
638:3-18, 1979.
15. Difco manual. Dehydrated Culture Media and Reagents for
Microbiology, 10. ed. Difco Laboratories, Detroit, 1985, p.1027.
16. Malucelli, M.I.C.; Niero, R.; Lucchiari, P.H.; Bacila, M. Evaluation
of the polarographic technique for assay of the viability of freeze-
dried BCG vaccine: I. The polarographic technique. Vaccine,
13(3):268-272, 1995.
17. Malucelli, M.I.C.; Niero, R.; Lucchiari, P.H.; Souza, M.D.C.; Bruzzo,
D.; Alves, R.C.B.; Miguel, O.; Bacila, M. Evaluation of the
polarographic technique for assay of the viability of freeze-dried
BCG vaccine: II. Viability of the vaccine assessed by polarography,
Warburg respirometry and colony counting. Vaccine, 13(3):273-
275, 1995.
18. Milinkovic, M.; Vlajinic, S. Relationship between oxygen
consumption and colony count test. International Symposium on
BCG Vaccine, Frankfurt, 1970, v.17, p.227-232.
19. Vancetto, M.D.C. Cultivo submerso do Mycobacterium bovis para a
produção do BCG imunoterápico. 2003, 97p. Doctoral thesis
presented to the “Programa de Pós-Graduação Interunidades em
Biotecnologia USP/IPT/Instituto Butantan, São Paulo (Brazil).
20. Stover, C.K.; Cruz, V.F.; Fuerst, T.R.; Burlein, J.E.; Benson, L.A.;
Bennett, L.T.; Bansal, G.P.; Young, J.F.; Lee, M.H.; Hatfull, G.F.;
Snapper, S.B.; Barletta, R.G.; Jacobs, W.R.; Bloom, B.R. New use of
BCG for recombinant vaccines. Nature, 351:456-460, 1991.
21. Lugosi, L. Analysis of variables of plasmid transformation of a
bacterial vaccine: studies on recombinant BCG. Vaccine, 8(2):145-
149, 1990.
22. Hanson, M.S.; Lapcevich, C.V.; Haun, S.L. Process on development
of the live BCG recombinant vaccine vehicle for combined vaccine
delivery. Ann. New York Acad. Sciences, 754:214-221, 1995.
23. Fuerst, T.R.; Stover, C.K.; Cruz, V.F. Development of BCG as a live
recombinant vector system: potential use as an HIV vaccine.
Biotechnol. Ther., 2(1-2):159-178, 1991.
24. Uno-Furuta, S.; Matsuo, K.; Tamaki, S. et al
. Immunization with
recombinant Calmette-Guerin bacillus (BCG)-hepatitis C virus (HCV)
elicits HCV-specific cytotoxic T lymphocytes in mice. Vaccine 21
(23):3149-3156, 2003.
25. Abdelhak, S.; Louzir, H.; Timm, J. et al. Recombinant BCG expressing
the leishmania surface-antigen GP63 induces protective immunity
against Leishmania major infection in Balb/C mice. Microbiology
UK (141):1585-1592, 1995.
26. Varaldo, P.B.; Leite, L.C.; Dias, W.O.; Miyagi, E.N.; Torres, F.I.;
Gebara, V.C.; Armoa, G.R.; Campos, A .S.; Matos, D.C.; Winter, N.;
Gicquel, B.; Vilar, M.M.; McFadden, J.; Almeida, M.S.; Tendler, M.;
McIntosh, D. Recombinant Mycobacterium bovis BCG expressing
the Sm14 antigen of Schistosoma mansoni protects mice from
cercarial challenge. Infect. Immun., 72(6):3336-3343, 2004.
27. Portaro, F.C.; Hayashi, M.A.; Arauz, L.J.; Palma, M.S.; Assakura,
M.T.; Silva, C.L.; Camargo, A.C. The Mycobacterium leprae hsp65
displays proteolytic activity. Mutagenesis studies that the M. leprae
hsp65 proteolytic activity is catalytically related to the HsIVU
protease. Biochemistry, 41(23):7400-7406, 2002.
28. Dietrich, G.; Mollenkopf, H.J.; Weber, H.; Knapp, B.; Diehl, K.D.;
Hess, J.; Blackkolb, F.; Bröker, M.; Kaufmann, S.H.E.; Hundt, E.
Cultivation of Mycobacterium bovis BCG in bioreactors. J. Biotech.,
96:259-270, 2002.
29. Zhang, J.; Greasham, R. Chemically defined media for commercial
cultivations. App. Microb. Biotechnol., 51:407-421, 1999.
30. Brock, T.D.; Modigan, M.T.; Martinko, J.M.; Parker, J. Biology of
Microorganisms., 7th ed., Prentice Hall, New Jersey, 1994, 909p.