Effect of Arabinose on Glucose Fermentation by

Saccharomyces cerevisiae
W. BORZANI AND E. AQUARONE'
Department of Chemistry, Escola Politecnica, University of Sdo Paulo, Sdo Paulo, Brazil
Received for publication October 21, 1957

It has recently been demonstrated (Borzani et al., 0.4 per cent and 1.8 to 2.5 per cent, respectively, for
1956) that the mash obtained from Araucaria brasili- 60 to 70 hr incubation times. These e-vaporation losses
were subtracted from the observed weight losses during

Downloaded from http://aem.asm.org/ on November 2, 2020 by guest

ensis seeds by acid hydrolysis, when fermented by
Saccharomyces cerevisiae, give alcohol yields practically the fermentation. Final pH was measured potentio-
constant at pH values from 2.9 to 7.2. This fact repre- metrically. Arabinose concentrations were determined
sents a considerable increase of the yeast optimum by the Youngburg's method (Browne and Zerban,
pH interval. It has also been observed (Borzani et al., 1941) and the correction due to the presence of glucose
1956) that glucose and arabinose were produced by was considered. Alcohol content was measured by
the acid hydrolysis of the A. brasiliensis seeds. common distillation process (Borzani and Falcone,
The hypothesis that arabinose could be responsible 1952), and yeast cell concentration was measured by
for the observed facts was suggested to us by Professor direct counting with the hemocytometer chamber
Renato Fonseca Ribeiro, and was confirmed in some (White, 1954). The results presented are the averages
preliminary experiments. of two or three measurements. Only typical experiments
This study deals with the effect of arabinose on the are indicated in this report.
optimum pH interval of glucose fermentation by Sac-
charomyces cerevisiae. This report also shows that the RESULTS
presence of arabinose increases the fermentation rate. Table 1 and figure 1 show the influence of L-arabinose
on the glucose fermentation rate. Figure 2 shows the
EXPERIMENTAL METHODS effect of L-arabinose on the final pH of the fermented
Culture medium had the following composition: medium. Table 2 and figure 3 show the effect of L-
meat extract 3 g, peptone 5 g, glucose 30 g; L-arabinose arabinose on the yeast optimum pH interval.
added in quantities corresponding to 5, 15, and 25 per The influence of L-arabinose was observed in a cul-
cent of glucose; water to 1 L. Blanks were run with ture medium sterilized by heat (120 C, 15 min) after
culture media containing only one of the carbohydrates.
Meat extract and peptone solution was sterilized by TABLE 1
heat (120 C, 15 min); glucose and L-arabinose solutions Effect of L-arabinose on glucose fermentation rates
were sterilized by means of a Seitz filter. Initial pH was Decrease in Weight Due to Fermentation (Per Cent)*
adjusted with HCI and NaOH solutions and was Fermentation Medium containing glucose and arabinose
potentiometrically measured. Culture medium for inoc-
ulum had the following composition: meat extract 3 g, pH 2.5 pH 4.7 pH 2.5 pH 4.7
peptone 5 g, glucose 10 g; water to 1 L. The inoculum hr
was prepared with Saccharomyces cerevisiae NRRL Y- 5 0.000 0.000 0.000 0.000
684 at 29 to 30 C. Ten ml portions of the culture me- 10 0.000 0.000 0.022 0.000
dium were distributed in test tubes, inoculated with 15 0.000 0.000 0.017 0.037
20 0.000 0.015 0.017 0.121
0.5 ml of the inoculum, and incubated at 29 to 30 C. 25 0.000 0.063 0.035 0.270
The fermentation rates were compared by periodic 30.5 0.000 0.189 0.095 0.512
weighing of the test tubes (Borzani and Falcone, 1953). 35.5 0.000 0.324 0.198 0.709
In some experiments, 2 ml of sterilized liquid Vaseline 42.5 0.000 0.523 0.375 0.878
were added to each test tube in order to reduce evap- 49 0.000 0.677 0.564 1.019
59.5 0.000 0.900 0.810 1.151
oration losses. The percentage losses due to evapora- 67.5 0.000 0.929 0.919 1.190
tion with and without the Vaseline seal were 0.2 to
* Test-tubes without liquid Vaseline were used. The quan-
1 Studies carried out during the tenure of a specialization tity of arabinose was 5 per cent of the glucose. The decrease
scholarship of the Reitoria da Universidade de Sao Paulo. in weight due to fermentation was calculated in per cent of
Present address, Faculdade de FarmAcia e Odontologia, the initial culture medium weight.
University of Sao Paulo, Sao Paulo, Brazil. Incubation at 29 to 30° C.
225
226 W. BORZANI AND E. AQUARONE [VOL. 6

the pH values were adjusted and measured with pH 12

indicator paper. Figure 4 shows the results obtained.
One experiment was run with a culture medium con-
taining meat extract, 9 g per L; peptone, 15 g per L;
glucose, 100 g per L; and L-arabinose, 15 per cent of
the glucose. Blanks without glucose were run. The 8
culture media were sterilized by heat (120 C, 20 min),
distributed in 250 ml cylinders, inoculated with 10 g 0.
per L (8.4 X 1010 cells per L) of pressed yeast (S.
CLa
TABLE 2 z 4
Effect of L-arabinose on the optimum pH interval of
Saccharomyces cerevisiae NRRL Y-684

Downloaded from http://aem.asm.org/ on November 2, 2020 by guest

Decrease in Weight Due to Fermentation (Per Cent)*

Initial pH Culture medium containing: 0

Glucose and arabinose
Glucose and Glucose and 0 4 8 12
Glucose arabinose (t) ($) arabinose (§)
INITIAL pH
1.5 0.000 0.000 0.000 0.000 Figure 2. Effect of L-arabinose on the final pH of the fer-
2.5 0.239 0.849 1.470 1.452 mented media. Experiments were carried out in test tubes
4.7 0.573 1.234 1.208 1.657
6.4 0.739 1.069 1.296 1.112 containing the inoculated culture media, without liquid Vase-
9.2 0.459 0.759 1.050 line. Temperature 29 to 30 C. Results without L-arabinose
1.163 are shown by the solid line. The results obtained with culture
10.7 0.000 0.921 0.250 0.000
media containing L-arabinose (1.5, 4.5 and 7.5 g per L) were
11.3 0.000 0.000 0.000 0.000
practically the same and are represented by the dotted curve.
*
Test-tubes without liquid Vaseline were used. The de- Time of incubation 67.5 hr.
crease in weight due to fermentation was calculated in per
cent of the initial culture medium weight, after 67.5 hr of
incubation at 29 to 30 C.
t Arabinose at 5 per cent of the glucose concentration. 1.2
t Arabinose at 15 per cent of the glucose concentration.
§ Arabinose at 25 per cent of the glucose concentration.

0 1* II '~ ' ' 0.8 1

LaJ
to
LAJ
0.8 1--
LJ z 0.4 F
I LJ
0
IAJ
O.4
LOJ
LI
3: 0F
0
0 20 40 60 80 0 2 4 6 8 10 12
HOURS
INITIAL pH
Figure 1. Effect of L-arabinose on the CO2 production rate.
Experiments were carried out in test tubes containing the Figure S. pH-activity curves of Saccharomyces cerevisiae
inoculated culture medium and 2 ml of liquid Vaseline to re- NRRL Y-684. Experiments were carried out in test tubes
duce evaporation losses. The weight decrease is calculated in containing the culture media and 2 ml of liquid Vaseline to
per cent of the initial culture medium weight. Temperature reduce evaporation losses. Temperature 29 to 30 C. Culture
29 to 30 C. The dotted curves represent the results obtained medium containing 7.5 g per L of L-arabinose, *; culture
without L-arabinose (El, pH 4.2; 0, pH 6.4; A, pH 8.7); the medium without L-arabinose, El; A represents the average
solid-line curves represent the fermentation rates with culture of four different tests: (a) inoculated L-arabinose, (b) uninocu-
medium containing 7.5 g per L of L-arabinose (, pH 4.2; lated L-arabinose, (c) uninoculated glucose, (d) uninoculated
0, pH 6.4; A, pH 8.7). glucose + i,-arabinose. Time of incubation 62 hr.
1958] ARABINOSE IN GLUCOSE FERMENTATION BY S. CEREVISIAE 227

cerevisiae), and incubated at 29 to 30 C. Initial pH was
measured with pH paper. The following results were
obtained. In culture media without glucose, the yeast
cells are almost completely destroyed and no alcohol
is produced; final pH values are similar to those already
indicated in figure 2; and the presence of L-arabinose
does not influence the alcohol yields (table 3), but has
a considerable effect on yeast reproduction (figure 5).
Figure 6 represents the variation of L-arabinose con-
centrations after the fermentations were completed.
DIscussION
The results show that arabinose has a considerable
also effected. Arabinose destruction was observed in
culture media of low and high pH values. A great re-
duction of arabinose concentration was also observed
in media where glucose was fermented by the Sac-
charomyces.
These experimental facts can be explained by the
results of Conway and Downey (1950), and also yield
additional evidence to their hypothesis. Admitting that
the arabinose penetrates in the outer metabolic region
of the yeast cells (Conway and Downey, 1950), we
may suppose that this penetration modifies the physio-
18

Downloaded from http://aem.asm.org/ on November 2, 2020 by guest

effect on the fermentation of glucose by S. cerevisiae
NRRL Y-684. The rate of CO2 production, as well as :
the optimum pH interval of the yeast, are greatly in- 1-
creased. The final pH values of the fermented media are -J 14
0.6
LJ

0
0-%
> o 10
LaJ
t&J 0.41
(A 0
aJJ 0

4
0 2 4 6 8 10
0.2
INITIAL pH
Figure 5. Effect of L-arabinose on pH optimum of cell re-
to3 production. Experiments were carried out in 250 ml cylinders,
with culture media inoculated with pressed yeast. Medium
0I I I containing only glucose as carbon source, 0; medium con-
0 0
4 8 12 taining glucose and 15 g per L of L-arabinose (15 per cent of the
glucose concentration), *. Temperature 29 to 30 C. Time of
INITIAL pH incubation 50 hr.
Figure 4. pH activity curves of Saccharomyces cerevisiae
NRRL Y-684. Experiments were carried out in test tubes con- 8
taining the inoculated culture media, without liquid Vaseline.
Temperature 29 to 30 C. Glucose as carbon source, @; glucose
and 1.5 g per L of L-arabinose, A. Time of incubation 69 hr. I.-. 6t
-
TABLE 3 i
Effect of L-arabinose on alcohol yields
o
_ _J
4 .
Alcohol Yields, Per Cent by Volume at 15 C
m
I
Initial pH Culture medium containing:
Glucose Glucose and arabinose*
0 L 12
1.0 0.0 0.0 0 4 8
3.0 4.0 4.4 INITIAL pH
5.0 4.1
5.5 - 4.1 Figure 6. L-Arabinose concentrations after complete fer-
6.5 4.0 mentation. Experiments carried out in test tubes containing
7.0 - 3.9 the inoculated culture medium and 2 ml of liquid Vaseline to
9.0 3.1 3.2 reduce evaporation losses. Temperature 29 to 30 C. Time of
incubation 125 hr. Initial arabinose concentration, 6.7 g per L.
* The quantity of arabinose was 15 per cent of the glucose. Medium containing only L-arabinose as carbon source, 0;
Incubation 29 to 30 C. medium containing glucose and L-arabinose, *.
228 J. H. SILLIKER AND W. I. TAYLOR
logical properties of that outer region and it will be
responsible for the experimental facts described in this
report. However, this question will require further de-
tailed study, chiefly in order to observe if other pentoses
have the same effects.
ACKNOWLEDGMENTS
The authors wish to express their appreciation to
Professor Lucio P. Carvalho Lima and to Professor
Aristoteles Orsini for providing laboratory facilities.
SUMMARY
Experiments were carried out in an attempt to de-
termine the influence of arabinose on the fermentation
of glucose by Saccharomyces cerevisiae NRRL Y-684.
The presence of arabinose increases the rate of CO2
production and the optimum pH interval of the yeast.
The alcohol yields, with and without arabinose, are
the same. The reduction of the medium pH during
fermentation in the presence of arabinose is smaller
than that observed on culture medium containing only
Isolation of Salmonellae from Food Samples
II. The Effect of Added Food Samples upon
J. H. SILLIKER AND W. I. TAYLOR
Research Laboratories, Swift and Company, Chicago, Illinois
Received for publication October 29, 1957
Many of the differences that exist between the needs
of the clinical bacteriologist and the food analyst have
been defined and discussed previously (Taylor et al.,
1958). Nowhere are these differences any more pro-
nounced than in the need for the food bacteriologist to
quantify the salmonellae in foods. The small numbers
which characterize the occurrence of salmonellae in
foods, often less than one per gram, prohibit the use
of direct counting methods. The problem posed by the
analysis of large amounts of food sample for small
numbers of organisms has been attacked traditionally
by resorting to a modification of the most probable
number (MPN) technique (Hoskins, 1934; Ayres, 1949).
As routinely performed, this technique involves the
inoculation of triplicate 90 ml aliquots of selenite or
tetrathionate enrichment broths with 10, 1 and 0.1 g
amounts of food material. Successively greater dilutions
are used when the food is known or thought to be more
heavily contaminated. It was particularly in these
latter cases during the course of analysis of a great
many food samples that a recurring anomaly was ob-
[VOL. 65
glucose. Arabinose concentration decreases when the
yeast ferment gluicose. Arabinose is not fermented by
the yeast studied; the inoculation of media containing
only arabinose as a major carbon source results in an
almost complete destruction of the yeast cells.
REFERENCES
BORZANI, W. AND FALCONE, M. 1952 Estudo sbbre os pro-
cessos de dosagem de Alcool em vinhos resultantes da
fermentagao de mosto de melago. Bol. assoc. brasil.
quim., 10, 5-9.
BORZANI, W. AND FALCONE, M. 1953 Control of starter
preparation in fermentation of molasses. J. Agr. Food
Chem., 1, 1070-1071.
Downloaded from http://aem.asm.org/ on November 2, 2020 by guest
BORZANI, W., DOLES, J. M., ITINOSE, S., AND ALBUQUERQUE,
L. M. C. 1956 Estudo da hidr6lise 6cida do pinhao.
Engenharia, 14, 276-278.
BROWNE, C. A. AND ZERBAN, F. W. 1941 Physical and chemi-
cal methods of sugar analysis, Ed. 3, pp. 915. John Wiley
& Sons, Inc., New York, New York.
CONWAY, E. J. AND DOWNEY, M. 1950 An outer metabolic
region of the yeast cell. Biochem. J., 47, 347-355.
WHITE, J. 1954 Yeast technology, p. 135. Chapman & Hall,
London, England.
the Performance of Enrichment Broths
served. It was found quite frequently that Salmonella
type organisms were not isolated from enrichment
broth aliquots representing the greatest amounts of
inocula, but were detected with ease in those which
had received much smaller quantities of samples. The
levels at which these "skips" occurred were usually the
10, 1 and 0.1 g replicates. Since the MPN technique is
in effect a series of qualitative determinations from
which quantitative data may be obtained from proba-
bilities of frequency of occurrence, the underlying
assumption is that the introduction of a single viable
Salmonella type cell into a tube of enrichment broth
will ultimately manifest itself as a Salmonella-positive
test upon subsequent plating. The frequency and the
extent of the skips belied this assumption. If, as we
began to suspect, the salmonellae were undetectable at
the aforementioned levels, this was a severe limitation
of the method, because those were the most frequently
used concentrations. Also, if the salmonellae did not
occur in numbers great enough to show positives in
subsequent dilutions, one would be led to believe that