Fermentation Methods for Protein Enrichment of Cassava

Page created by Willie Lloyd
 
CONTINUE READING
APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Jan. 1980, p. 41-47                                               Vol. 39, No. 1
0099-2240/80/01-0041/07$02.00/0

Fermentation Methods for Protein Enrichment of Cassava and
               Corn with Candida tropicalis
    EDGARD AZOULAY,'* FRAN,COISE JOUANNEAU,' JEAN-CLAUDE BERTRAND,' ALAIN
               RAPHAEL,' JACQUES JANSSENS,2 AND JEAN MICHEL LEBEAULT2
Laboratoire de Structure et Fonction des Biomembranes, U.E.R. de Luminy, 13288 Marseille Cedex 2,' and
  Laboratoire de Genie Biologique, Universite de Technologie de Compiegne, 60200 Compiegne,2 France

            Candida tropicalis grows on soluble starch, corn, and cassava powders without

                                                                                                                           Downloaded from http://aem.asm.org/ on January 26, 2021 by guest
          requiring that these substrates be previously hydrolyzed. C. tropicalis possesses
          the enzyme needed to hydrolyze starch, namely, an a-amylase. That property has
          been used to develop a fermentation process whereby C. tropicalis can be grown
          directly on corn or cassava powders so that the resultant mixture of biomass and
          residual corn or cassava contains about 20% protein, which represents a balanced
          diet for either animal fodder or human food. The fact that no extra enzymes are
          required to hydrolyze starch results in a particularly efficient way of improving
          the nutritional value of amylaceous products, through a single-step fermentation
          process.

   Corn flour contains about 10% protein and                    starch which has been soaked in an acidic solution at
80% polysaccharide, mostly starch. When used                    a temperature below that of gelatinization. In our
for animal feeding it has to be supplemented                    experiments, we used Merck (1257) soluble starch,
with soya flour, which contains about 50% pro-                  according to Zulkowski, containing no reducing sugars;
tein, so as to obtain a protein content in the final            Prolabo (France), soluble starch containing 10% re-
                                                                ducing sugars; and Merck Erg B6 starch.
mixture of about 20% protein. Cassava contains                     (ii) Dextrines (Amisol, France). Soluble starch
mostly starch (70%) and even less protein than                  must not be confused with dextrine, which is the
corn (3%); the rest is a poorly defined mixture                 product obtained by heating dry starch, usually corn
originating from the ligneous part of the root                  starch, sprinkled with 1% chlorhydric acid, to above
(6). An increased food value of these powders                   100°C in a slightly acidic medium. At about 100 to
can be obtained by increasing their protein con-                120°C, white dextrines are obtained; at 150°C, yellow,
tent through fermentation processes which use                   blond, or canary dextrines are produced; and above
fungi (2, 12).                                                  180°C, the product is British gums. In our experiments,
   The cultivation of a yeast such as Candida                   we used 99% soluble yellow Amisol 08077 corn dextrine
tropicalis on corn powder leads directly to a                   provided by the Societe des Produits du Mais, France.
                                                                   (iii) Cassava. We used pellets of cassava provided
balanced feed. The growth of these microorga-                   by Peter Cremer (Hamburg, Germany), containing
nisms requires the assimilation of the polysac-                 about 70% starch and 10% water. The remainder (un-
charides (7) contained in these powders, and the                identified) contains mainly ligneous parts from the
resultant protein content increase ofthis product               exterior part of the root. In all the experiments the
is due to the presence of the synthesized yeast                 pellets were finely crushed to a mean size of 50 ,um
biomass.                                                        with a Law hammer grinder type D.G. It was observed
   In this paper, we describe a process whereby                 that a nonnegligible fraction (5%) dissolved sponta-
a strain of C. tropicalis has been isolated which               neously and that solubility could be increased by add-
grows directly on starch (E. Azoulay, U.S. Pat-                 ing 0.1 N H2SO4. This was consistent with the known
                                                                results on the solubility of the various starches.
ent application no. 4,081,557, 18 March 1978),                     (iv) Corn. Grains of corn, provided by Funk, were
resulting in the improvement of com or cassava                  crushed with a Law hammer grinder type D.G. to
for its direct utilization as animal feed.                      obtain a powder or flour of which the grain dimension
          MATERIALS AND METHODS                                 was about 100 ,m (ultrafine corn).
                                                                   Growth medium. The growth medium contained
   Organism. Yeasts used were C. tropicalis CBS                 the following: MgSO4.7H20 (0.2 g), NaCl (0.1 g),
6947, first described as C. tropicalis 101 (3), and C.          NH4Cl (2.5 g), KH2PO4 (7.0 g), Na2HPO4 (1.2 g), FeCl3
tropicalis CBS 6948, isolated by Azoulay (Azoulay,              (0.05 mg), ZnSO4 (0.5 mg), MnCl2 (0.5 mg), CaCl2 (0.5
U.S. Patent application no. 4,081,557, 1978) from strain        mg), Na2MoO4 (0.5 mg), CUSO4 (1 mg), Difco yeast
CBS 6947 on a nutritional medium containing Merck               extract (0.1 g), and distilled water (1 liter); pH was
soluble starch as the only carbon source.                       adjusted to 5.1 by addition of 2 N NaOH. The carbon
   Products used. (i) Starches. Soluble starch, more            source was either soluble starch or ground corn powder
commonly known as acid-modified starch (13), is                 or ground cassava powder (2 g/liter of mineral me-
                                                           41
42       AZOULAY ET AL.                                                             APPL. ENVIRON. MICROBIOL.
dium). The media were autoclaved at 110°C for 1 h at        acid analyzer (model 116). The samples were prepared
a pressure of 21.7557 lb/in2 (1 lb/in2 = 6.894757 kPa).     by hydrolyzing whole cells in 6 N HCI for 22 h at
   Cultures. (i) Batch cultures were made in 2-liter        110°C under vacuum.
Fernbach flasks at 32°C agitated on a reciprocal               Determination of the growth parameters. The
shaker at 200 rpm for 48 h or in a 4-liter Ultroferm        yield with respect to the original amount of starch was
LKB fermentor agitated at 1,000 rpm and with an             calculated by dividing the resultant biomass by the
oxygen flow in the reactor of 30 vol/vol per h.             amount of starch used up by the yeast. The amount of
   (ii) Continuous cultures were made in the Ultroferm      residual starch was estimated by the zero method
LKB fermentor with the following parameters: Tem-           described below. The amount of residual starch was
perature, 32°C; agitation, 1,000 rpm; oxygen flow, 30       estimated after acid hydrolysis by measurement of
vol/vol per h; pH adjusted to 5.1 by continuous and         reducing sugars. The symbols and notations used are
controlled addition of 0.1 N NaOH; input and output         given in Table 1.
flow simultaneously controlled and equal to 0.2 or 0.4         Preparation of cell-free extracts. The yeast

                                                                                                                       Downloaded from http://aem.asm.org/ on January 26, 2021 by guest
liter/h. The dilution rate (see Table 1) varied from 0.2    cells, grown on Merck soluble starch, were harvested
to 0.5/h depending on the type of experiment.               at the end of the exponential phase and processed to
   The inoculum consisted of 1 liter of mineral medium      obtain an enzymatic extract essentially by the method
containing soluble starch as the carbon source, previ-      described by Gilewicz et al. (4), which was modified as
ously inoculated with C. tropicalis CBS 6948 and            follows: after being washed, the pellets were put in
incubated for 24 h to obtain a cell concentration of 0.3    suspension again in a 0.05 M phosphate buffer (pH
g (dry weight) of cells per liter. The corn or cassava      7.0), centrifuged, and resuspended in the same buffer.
suspensions were continuously agitated and were in-         After disruption in a French pressure cell at 10,000 lb/
troduced in the fermentor as a homogeneous suspen-          in2, this suspension was centrifuged at 1,600 x g. The
sion.                                                       supernatant constituted the crude extract, and it was
   When high concentrations (90 g/liter) of cassava         centrifuged at 27,000 x g for 10 min. The supernatant
powder were used, the continuous cultures were car-         obtained was centrifuged at 220,000 x g for 90 min.
ried out in a 14-liter fermentor (PEC, Chemap AG).          The pellet obtained constituted the particular extract,
The working broth volume was 9 liters and was kept          and the supernatant was used as the supernatant
constant by means of a dissipated power system. Me-         extract.
dium feed was controlled by a low flow rate peristaltic        Amylase assay. Amylase activity was measured in
pump (Desaga).                                              magnetically stirred reaction mixtures containing 2 ml
   Analytical methods. (i) Determination of the             of 1% soluble starch (Merck), 1 ml of 0.1 M sodium
biomass (cell concentration). Cell growth was fol-          acetate buffer (pH 5.6), and 0.6 ml of water. After
lowed by measuring the optical density at 450 nm and        equilibration was reached in a water vessel maintained
was expressed in milligrams (dry weight) of cells per       at 31°C, the reaction was initiated by adding 0.4 ml of
milliliter with reference to a calibration curve (optical   appropriately diluted enzyme. One-milliliter samples
density readings at 450 nm versus dry cell weight,          were drawn at zero time and after 10 min of incubation.
grams per liter). The calibration curve was the result      Accumulation of reducing sugars was estimated by the
of a good linear correlation between optical density        dinitrosalicylic acid method of Bernfeld (1). An amy-
and cell weight so that one optical density unit corre-     lase activity unit was defined as the release of 1 ,umol
sponded to 0.25 g of cells per liter. The microbial         of reducing sugar (as glucose) per min per mg of
growth on corn or cassava flour was evaluated accord-       protein, determined by the method of Lowry et al.
ing to the following method. Samples of the growth          (10).
medium on corn (10 ml) were removed at different
times and centrifuged at 4,000 x g. The pellet obtained                        RESULTS
after two washings in distilled water contained the           Growth of C. tropicalis CBS 6948 on sol-
transformed corn powder and the yeast cells. After          uble starch. The strain CBS 6948, having an
overnight desiccation at 60°C, this pellet was weighed,     amylolytic activity, was isolated by successive
and its protein nitrogen content was measured by the        subcultures of a population of C. tropicalis CBS
Nessler method after mineralization in the presence of      6947 grown on a nutritional minimal medium
concentrated H2SO4. In that way, the quantity of dry
matter corresponding to the available biomass and the       containing Merck soluble starch as the only
protein content of this biomass could be determined         source of carbon and energy (Azoulay, U.S. Pat-
at different times of growth. The protein content was       ent application no. 4,081,557, 1978). This yeast
determined from the amount of ammoniacal nitrogen           grows by forming star-shaped colonies, different
multiplied by 6.25.                                         from the smooth colonies of the initial strain
    (ii) Assay of the starch. Starch was first hydro-       (CBS 6947), which grows slowly on starch. CBS
lyzed with 0.5 M hydrochloric acid into reducing sugars     6498 grows in batch culture on soluble starch
 (mainly D-glucose), the concentration of which was         with growth yields between 0.45 and 0.55 g of
determined by the method of Bernfeld (1) with 3,5-          cells per g of transformed starch and a growth
dinitrosalicylic acid.                                      rate of 0.4/h. The permissive temperature of
    (iii) Amino acid analysis. Amino acids were an-
alyzed by the UCAAB nutritional laboratories                growth ranged from 20 to 37°C, with an opti-
 (France) on the desiccated and homogenized biomass         mum at 32°C. In these experiments, ammonia or
harvested after growth with a Beckman auto-amino            NH4' ion was the source of nitrogen. The pH
VOL. 39, 1980                                  PROTEIN ENRICHMENT WITH C. TROPICALIS                              43
                                       TABLE 1. Symbols and notations
Symbol                    Definition                     Unit                        Method of determination
  S.      Initial concn of the dry substrate (starch,   g/liter        Drying and weighing
             corn, cassava)
  Si      Concn of dry gelified cassava                 g/liter        Centrifugation, separation of the pellet,
                                                                         drying, and weighing
  S       Concn of the residual substrate of cas-       g/liter        Centrifugation, separation of the super-
            sava soluble in the fermentor                                natant, drying, and weighing
  S'      Concn of the residual substrate of starch     g/liter        Centrifugation, separation of the super-
            and soluble sugars in the fermentor                          natant, hydrolysis, and assay of the re-
                                                                         ducing sugars
  Se'     Starch and solubilized sugars concn           g/liter        Centrifugation, separation of the super-

                                                                                                                       Downloaded from http://aem.asm.org/ on January 26, 2021 by guest
                                                                         natant, hydrolysis, and assay of the re-
                                                                         ducing sugars
  P       Concn of the final products in the output     g/liter        Evaporation and weighing
            of the fermentor
  F       Input of water in the fermentor               liter/h        Volumetric pump
  D       Dilution rate (water), relation of the out-                  Calculated: F/V
            put of water to the useful vol of the
            reactor
  V       Useful vol of the reactor                      liter         Quantity of water introduced
  X       Yeast concn                                   g/liter        Centrifugation and weighing
  Y       Total yield of the fermentation (enriched                    Calculated: P/S,
            cassava wt/wt of the original cassava)
  (p)     Level of proteins of the product P                           Assay of nitrogen
  (s)     Level of proteins of the substrate S.                        Assay of nitrogen
  t       Duration                                        h            Timer
  (x)     Level of proteins of yeast x                                 Assay of nitrogen
  Y.      Elementary yield of the transformation                       Calculated: X/(Se'-S')
            of the substrate in yeast (wt of yeast/
            wt of used substrate)

values giving the best growth were between 4.0            TABLE 2. Growth of C. tropicalis CBS 6948 on
and 5.8. The two growth parameters (yield and                         different starches
rate) did not vary much with the substrate con-                            Initial Residual Assimi-
centration in the growth medium, if the aeration                            sub-     sub-    lated Yeast Yield
conditions were good, up to a starch concentra-               Starch                   strate   sub-S(
tion of 10 g/liter. In the experimental conditions
                                                                           s
                                                                            (Strat      (Strt      ()
                                                                                                strate(.

that permitted growth on soluble starch, we
found that the growth medium prepared as al-            Soluble starch      13.9         1.2    11.7     5.4   0.43
ready described (sterilization at 1100C for 1 h)          (Prolabo)
contained only negligible quantities of glucose         Dextrine            27.8        19.3      8.5    3.7   0.44
(less than 0.1%) which might be liberated by
starch hydrolysis. It could therefore be con-           attacked by the amylase present in the yeast.
cluded that this strain grows directly on starch        Besides, we have shown that the Merck starch,
and not on previously liberated hydrolysis prod-        Erg B6, which gives only colloidal suspensions
ucts. It should however be noted that at the end        in water, is not at all degraded by C. tropicalis.
of the growth, 25% of the initial starch had not        These results indicate that only the soluble frac-
been assimilated; this fraction could be deter-         tion of the starch can be used as source of
mined after acid hydrolysis by measuring the            carbon.
reducing sugars.                                           Development of C. tropicalis CBS 6948 on
   For comparison purposes, we studied the              cassava powder. The cultures on cassava pow-
growth of C. tropicalis CBS 6948 on starches            der were made using powder suspensions of var-
other than Merck soluble starch, Zulkowsky              ious concentrations which were previously ster-
type. The results (Table 2) show that the pres-         ilized at 110°C for 15 min.
ence of reducing sugars in Prolabo soluble starch          (i) Batch cultures. The growth of C. tropi-
(France) or dextrine does not inhibit growth.           calis is possible on cassava concentrations rang-
However, in spite of their solubility, only 30% of      ing from 2 to 100 g/liter. The maximum specific
the dextrines are assimilated by the yeast. This        growth rate obtained was 0.4/h, and the maxi-
indicated that a fraction of the starch is not          mum productivity of batch culture attainable
44      AZOULAY ET AL.                                                               APPL. ENVIRON. MICROBIOL.
was estimated to be about 0.038 g of cells per           with cassava powder are equivalent to those
liter per h. The changes in steady-state cell            obtained with soluble starch. The biomass had
concentration were measured with various con-            a weight ratio of 16 to 18%, which is six to seven
centrations of cassava, at pH 5.0 and 32°C (Fig.         times greater than that of natural cassava. Con-
1). The yield coefficient was determined as 0.55         tinuous types of culture were carried out to
g of cells per g of metabolized cassava starch.          optimize pH and temperature conditions. C.
This result compares favorably with the yield            tropicalis CBS 6948 grew best in the pH range
coefficient of 0.50 g of cells per g of starch found     of 4.0 to 5.0, but in nonaseptic conditions, a pH
with several microbial systems (14). Only part of        of 4.0 was optimal. With pH 4.0, it was found
the starch is accessible to the amylases this yeast      that a maximum cell concentration was attained
synthesizes. About 25% of the initial substrate          at 35°C. Figure 2 illustrates for a continuous
remains nonassimilated.                                  culture the results of cell concentration, biomass

                                                                                                                     Downloaded from http://aem.asm.org/ on January 26, 2021 by guest
   In some experiments we replaced the suspen-           productivity, and residual starch concentration.
sions of cassava powder in water with suspen-            The maximum biomass productivity obtained
sions of cassava powder in 0.1 N sulfuric acid,          was 3.6 g/liter per h at a dilution rate of 0.26
sterilized at 110°C for 15 min. We confirmed             h-'. At lower dilution rates, the cell concentra-
that the suspensions containing soluble starch           tion approached that of batch culture, and at
never had more than 0.1 to 0.2% of reducing              dilution rates higher than 0.26 h-', as it ap-
sugars.                                                  proached the maximum dilution rate, the cell
   (ii) Continuous cultures. In this case, the           concentration decreased sharply, as would be
cassava powder was used only as suspensions in           expected in a continuous culture. To make sure
water after sterilization at 110°C for 15 min, and,      that a new steady state was established, a period
contrary to the batch culture experiments, no            of 48 h was allowed after each change of dilution
sulfuric acid was used in the suspension. Table
3 shows the results of two fermentation runs              TABLE 3. Continuous culture of C. tropicalis CBS
made with C. tropicalis on cassava powder at                             6948 on cassava powdera
two different dilution rates. The yields obtained         Expt F           D      S. (s)(%) P (x) (%) Y
                                                            1       1.5 0.3       10    2.8 5.6 16.8 0.56
                                                            2       1     0.25 10       3.1 4.2 18.6 0.42
                                                            a Initial concentration, SO = 10 g/liter. For symbols,
                                                         see Table 1.

         0       20     40     60    80     100
             Metabolized starch concentration   (l)
  FIG. 1. Cell concentration and residual starch
concentration with different metabolized starch con-
centration media containing different cassava con-
centrations. The starch contained in the cassava was
determined at the beginning and end of the growth,                      005   01      015 02      Dm 03       035
and the metabolized starch concentration was cal-                                  Dilution rate, D ( hrK1)
culated from the difference between the two obtained
values. The slope of the cell concentration curve (0),      FIG. 2. Biomass productivity and cell concentra-
equal to 0.55, corresponds to the elementary yield of    tion versus dilution rate in continuous culture of C.
growth, Y0, as defined in Table 1.                        tropicalis on cassava powder.
VOL. 39, 1980                                     PROTEIN ENRICHMENT WITH C. TROPICALIS                               45
rate. The dilution rate for the production of                   Growth of C. tropicalis CBS 6948 on corn
single-cell protein, corresponding to the maxi-              powder. Corn powder in suspension in the min-
mum     productivity, was 0.25/h.                            eral medium was first sterilized at 1100C for 15
   (iii) Analysis of the cassava enriched by                 min. The yield of batch cultures of C. tropicalis
fermentation with C. tropicalis CBS 6948.                    in this medium relative to the assimilated solu-
In the final product the desired amount of pro-              bilized corn was about 0.47 at a substrate con-
tein should be relatively large, and at the same             centration of 20 g/liter. This yield, Y, and the
time the quality of protein in terms of its amino            elementary yield of growth, Y8 (0.6), are close to
acid profile should be well balanced. The protein            those obtained with cassava powder.
content of C. tropicalis grown on starch was                    In continuous cultures, the growth of C. trop-
about 45% (Table 4). In the enriched cassava, 5              icalis on corn powder made at relatively low
to 6% of the proteins corresponded to the protein            dilution rates (D = 0.05 - 0.1/h) gave enriched

                                                                                                                           Downloaded from http://aem.asm.org/ on January 26, 2021 by guest
contained in the natural cassava, and the rest               corn with a yield, Y, between 0.6 and 0.75. The
was made up by the yeasts. These results were                protein concentration at the outlet of the fer-
thus consistent with the usual yields of fermen-             mentor was about 20% (Table 6). Thus, it seems
tation on carbohydrates, where the yeast-starch              possible to enrich corn with high yields by vary-
yield is between 50 and 55% and the ratio of                 ing only the dilution rate, D.
weight yield of protein/starch is thus near 1:3.                The analysis of the corn enriched in this way
In the fermentation of cassava, the loss of 40%              shows that the increase in the protein content
of the starch therefore results in an increase of            parallels an appreciable enrichment in lysine
about 12% protein.                                           and cystine (Table 7).
   To confirm these results we studied the                      Assay of the a-amylase synthesized by C.
growth of C. tropicalis in continuous culture                tropicalis CBS 6948 grown on starch. Pre-
with high cassava concentrations (90 g/liter) and            liminary studies showed that, in the conditions
determined systematically the quantities of ge-              described in Materials and Methods, the crude
lified cassava, solubilized substrate (starch or             extract catalyzed the transformation of starch to
reducing sugar), and fermentation products at                reducing sugars with a specific activity of about
the end of the culture (42 h of growth). This                25 units. This transformation was linear only
made it possible to calculate the quantity of                during the first 10 min (Fig. 3), and, at the end
yeast resulting from the solubilized cassava and             of the reaction, 50 to 60% of the starch was not
to determine the growth yields (Table 5). From               hydrolyzed.
these measurements the quantity of product ob-                  The enzymatic activity exhibited by the crude
tained after fermentation with C. tropicalis,                extract remained in the supernatant after cen-
which corresponds to the enriched cassava and                trifugation at 200,000 x g. This indicates that
the elementary yield (Y1"; Table 1) of the fer-              the amylase present in the cells of C. tropicalis
mentation, can be determined; it was found to                is a soluble enzyme. Unlike some other micro-
be equal to 0.6.                                             organisms (11), C. tropicalis does not produce

  TABLE 4. Essential amino acid composition of                TABLE 6. Continuous culture of C. tropicalis CBS
    natural cassava and cassava enriched by                               6948 on corn powdera
    fermentation with C. tropicalis CBS 6948
                                                              Expt   F        D       S. (s)(%)    P     (x)    Y
                Nitro-  Protein Lysine Methi-     Cys-
 Cassava         gen      (%)   Lyn onine         tine         1    0.4 0.1       2    8.8        1.48   21    0.74
                       M M       (%O)   ~~~(%O)   M%o)         2    0.2 0.05 2 8.8                1.20   24    0.60
Natural          0.5        3.1   1.1     0.7     0.5          3    0.2 0.05       1   8.8        0.63   20    0.63
Enriched         3         18.8   7.7     2.7     2            a For symbols, see Table 1.

           TABLE 5. Fermentation characteristics of the cassava medium by C. tropicalis CBS 6948'
   Expt           S.                       Si         (Se'-S') P = (X + Si)       X           YSP/S
                                                                                                  = S')
    1             90        23          19             36         41         22           0.46          0.60
    2             90        10           9.5           50         34         24.5         0.36          0.50
  a Batch cultures were made in a 4-liter fermentor over 42 h. The results are expressed in grams of assimilated,
transformed, or residual substrate per liter of medium. (Se' - S'), Starch assimilated during the growth (grams
per liter); the other symbols are defined in Table. 1.
46         AZOULAY ET AL.                                                               APPL. ENVIRON. MICROBIOL.
     TABLE 7. Comparative analysis of initial and enriched by fermentation with C. tropicalis CBS 6948
       Corn          Humidity Nitrogen            Protein ( Carbon M     Starch M       Lysine     Methio-      Cystine
         Cor                                      Potin %)Cabon(%        Sarc ()            )      nin M%O       (%O)
     Initial            10           1.4             8.75     42.1          66           3.3          2.2         1.9
     Enriched           10           3.9            20.6      40.8          44.7         6.8          2.4         2.6

                                                              amylolytic activity (about 10%) seems to indi-
                                                              cate an inductional rather than a mutational
                                                              phenomenon. Supporting this hypothesis is the
                                                              fact that the amylolytic activity is found only in
                                                              cell-free extracts prepared from cells grown on

                                                                                                                             Downloaded from http://aem.asm.org/ on January 26, 2021 by guest
       z                                                      starch as substrate. This enzyme is thus a totally
                                                              inducible one.
                                                                 No amylolytic activity was found in the
                                                              growth medium after removal of the cells. This
                                                              and the presence of the activity in the cell-free
                                                              extracts strongly indicate an intracellular local-
           0    10     20       30    40     50      60       ization for this enzyme rather than an extracel-
                            TIME (MINUTES)
                                                              lular one as was suggested by Spencer-Martins
   FIG. 3. Time course of reducing sugar liberation           and van Uden (14).
from starch (1%) by amylase. The experimental system             The isolated strain, CBS 6948, is able to grow
contained 20 pg of crude extract protein of C. tropi-         on starch only when this substrate is soluble in
calis grown on starch. The assay was made at 32°C
atpH 5.6.                                                     water. Under these conditions, the growth yield
                                                              was equivalent to those obtained with different
                                                              sugars, particularly glucose. It probably grows at
extracellular amylase. Actually, no amylolytic                the expense of 1-4 polysaccharidic chains of the
activity was found in the growth medium at the                amylose type, with amylopectin representing the
end of the growth of this yeast on starch even                fraction of the starch that is not assimilated.
after concentration of the medium by ultrafiltra-                The interest in this strain lies in the fact that
tion on Amicon (PM10). Furthermore, the crude                 it can be used in enrichment processes of com
extracts of cells of C. tropicalis CBS 6948 grown             or cassava powders. Consequently, it can replace
on glucose or Sabouraud maltose broth (Difco)
                                                              the fungi used for the preparation of single-cell
had no amylase activity whatsoever. This indi-                protein from these two substrates, mainly cas-
cated that the amylase of the cells grown on                  sava (8, 14). The ability of the yeast to grow is
starch was an inducible enzyme.                               essentially linked to the fact that the com or
   A preliminary study of the properties of the               cassava powders were previously sterilized at
enzyme present in the supernatant extract of
cells grown on starch made it possible to estab-
                                                              110°C for 15 min. According to Leach et al. (9),
lish that the maximum of activity was obtained
                                                              who studied the swelling and gelatinization of
                                                              various starches, the sterilization leads to a more
at pH 4.8 in 0.1 M phosphate buffer and at 6.2                or less complete solubilization of the com or
in 0.1 M acetate buffer. At a higher molarity (1              cassava starches. This soluble form of the starch
M), a 50% diminution of the activity was ob-                  is available for the growth of C. tropicalis in the
served. Under the routine conditions assay sys-               same manner as Merck soluble starch. Besides,
tem, the optimal temperature for measuring the                we have shown that this type of sterilization
activity was 60°C. At pH 7.0, the enzyme was                  does not entail the formation of reducing sugars
rapidly inactivated by an incubation at 65°C.                 which could be used by this yeast for its growth.
                                                              The enrichment of corn or cassava powders re-
                 DISCUSSION                                   sults, after fermentation with C. tropicalis CBS
  It has been well established by Spencer-Mar-                6948, in an increase of 18 to 20% of the protein
tins and van Uden (14) that yeasts can assimilate             content. If we refer to the levels of lysine and
starches. In this study, a strain with a strong               methionine, the essential amino acids composi-
amylolytic activity (CBS 6948) was isolated from              tion is relatively balanced.
a low-activity population of C. tropicalis. The
                                                                                 ACKNOWLEDGMENTS
absence of genetic studies on C. tropicalis does
not allow us to distinguish whether these colo-                   We thank Marcelle Zacek and Nicolas Inchauspe for inter-
                                                               esting discussions.
nies are mutant cells or not. However, the im-                    This work was supported by Societe SPEICHIM, Bondy,
portant proportion of colonies having a high                   France, and Adour Entreprise, Pau, France.
VOL. 39, 1980                                         PROTEIN ENRICHMENT WITH C. TROPICALIS                                 47
                    LITERATURE CMD                                 8. Humphrey, A. E. 1974. Current developments in fermen-
                                                                        tation. Chem. Eng. 20:98-112.
 1. Bernfeld, P. 1955. Amylases, a and ,. Methods Enzymol.
      1:149-158.                                                   9. Leach, H. W. 1967. Gelatinization of starch, p. 289-307.
 2. Brook, E. J., W. R. Stanton, and A. Wallbridge. 1969.               In R. E. Whistler and E. S. Paschall (ed.), Starch
        Fermentation methods for protein enrichment of cas-             chemistry and technology, 1967. Academic Press, New
        sava. Biotechnol. Bioeng. 11:1271-1284.                         York.
 3.   Duvnjak, Z., B. Roche, and E. Azoulay. 1970. Isolation      10. Lowry, 0. H., N. J. Rosebrongh, A. L. Farr, and R. J.
        and study of strain of Candida tropicalis growing on n-         Randall. 1951. Protein measurement with the Folin
        alkanes. Arch. Mikrobiol. 72:135-139.                           phenol reagent. J. Biol. Chem. 193:265-275.
 4.   Gilewicz, M., M. Zacek, J. C. Bertrand, and E. Azou-        11. Pfuller, S. L, and W. H. Elliott. 1969. The extracellular
        lay. 1979. Hydroxylase regulation in Candida tropicalis         a-amylase of Bacillus stearothermophilus. J. Biol.
        grown on alkanes. Can. J. Microbiol. 25:201-206.                Chem. 244:48-54.
 5.   Glymph, J. L., and F. J. Stutzenberger. 1977. Produc-       12. Reade, A. E., and K. F. Gregory. 1975. High tempera-
        tion, purification, and characterization of a-amylase           ture production of protein enriched feed from cassava

                                                                                                                                   Downloaded from http://aem.asm.org/ on January 26, 2021 by guest
        from Thermomonospora curvata. Appl. Environ. Mi-                by fungi. Appl. Microbiol. 30:897-904.
        crobiol. 34:391-397.                                      13. Shildneck, P., and C. E. Smith. 1967. Production and
 6.   Grace, M. 1971. Processing of cassava. Agric. Serv. Bul-          uses of acid-modified starch, p. 217-235. In R. E. Whis-
        letin no. 8. Food and Agriculture Organization, Rome.           tler and E. S. Paschall (ed.), Starch chemistry and
 7.   Gregory, K. F., A. E. Reade, G. L Khor, J. C. Alex-               technology, 1967. Academic Press, New York.
        ander, J. H. Lumsden, and G. Losos. 1976. Conver-         14. Spencer-Martins, I., and N. van Uden. 1977. Yields of
        sion of carbohydrates to protein by high temperature            yeast growth on starch. Eur. J. Appl. Microbiol. 4:29-
        fungi. Food Technol. 3:30-35.                                    35.
You can also read