Journal of Experimental Microbiology and Immunology (JEMI) Vol.

2:1-5
Copyright  April 2002, M&I UBC

The Effect of Fructose, Galactose, and Glucose on the Induction of

β-Galactosidase in Escherichia coli
CHARLENE CHU, CHRISTINA HAN, HIROMI SHIMIZU, AND BONNIE WONG

Department of Microbiology and Immunology, UBC

Carbon catabolite repression (CCR) refers to the ability of bacteria to preferentially utilize
one carbon source over another. Phosphoenolpyruvate-dependent carbohydrate:
phosphotransferase system (PTS) is the mechanism by which CCR is exerted. In the absence
of preferred PTS-carbon sources, cAMP-CAP complexes are abundant and thus, the
enzymes encoded in the lac operon can be expressed maximally. In our experiment, IPTG
was used to induce expression of the lac operon in E. coli cells grown in 0.2% glycerol and
then transferred to 0.1% of fructose, galactose, and glucose. β-galactosidase levels of each
culture were measured by absorption spectroscopy of nitrophenol produced from the ONPG
substrate. Results showed that the addition of galactose exhibited a decrease in β-
galactosidase induction. However, both glucose and fructose displayed lower, but similar
enzyme induction patterns to galactose. By the end of the experiment, enzyme induction in
all three conditions (glucose, fructose, and galactose) was approximately 33% that of the
control (glycerol). Initially, according to our “composition theory,” we hypothesized that
induction levels of galactose would follow that of lactose and induction levels with fructose
would follow that of sucrose. However according to our results, galactose may exhibit
feedback inhibition when there are higher levels, while there seems to be no preferential
usage between glucose and fructose; thus, monosaccharides derived from dissacchrides are
not necessarily utilized by bacteria in the same way.

The ability of bacteria to preferentially utilize one carbon source over another is termed carbon catabolite
repression (CCR). It is also known as the “glucose effect”, but can be exhibited by other carbon sources as well.
Carbon sources are grouped into two classes: Class A and Class B. When grown in a mixture of carbon sources,
bacteria preferentially utilize Class A substrates over Class B substrates. The system by which Class A substrates
exert repression is called the phosphoenolpyruvate (PEP)-dependent carbohydrate:phosphotransferase system (PTS)
(4).
The PTS transports and phosphorylates a large number of carbohydrates. PTS carbohydrates, especially glucose,
result in a decrease in the phosphorylated form of Enzyme II, EIIAGlc~P. EIIAGlc~P activates adenylate cyclase
which makes cAMP from ATP. A decrease in EIIAGlc~P results in a decrease cAMP, which represses CAP-
dependent operons – genes whose transcription is regulated by the cAMP-CAP complex (7). One example of a
CAP-dependent operon is the lac operon in Escherichia coli.
When E. coli are grown in rich media, the cAMP concentration is low and there are few cAMP-CAP complexes to
activate the lac operon. Thus, even in the presence of an inducer, such as lactose or its analog, enzyme induction
remains low. However, when preferred carbon sources are absent, cAMP-CAP complexes are abundant and the
capacity to synthesize lac operon encoded enzymes is maximal upon addition of an inducer.
Experiment A4 (6) utilizes glycerol, glucose and the two disaccharides lactose and sucrose to study β-
galactosidase induction levels in E. coli. It showed strong reduction of activity due to lactose treatment but slower,
more limited reduction of activity due to sucrose treatment. Lactose is composed of glucose and galactose, whereas
sucrose is composed of glucose and fructose. Therefore we initially hypothesized that induction levels of galactose
would follow that of lactose and induction levels of fructose would follow that of sucrose. However, upon further
research we found that our “composition theory” does not make sense when the PTS system is taken into account.
This is because the disaccharides are not split into the components until after transport so composition would have
no bearing on the levels of induction. Even so, we went ahead with our initial experiment and investigated CCR by
growing E. coli in glucose, fructose, galactose and glycerol. β-galactosidase activity was measured as an indication
of cAMP levels and CCR relationship with respect to the different carbon sources.

1
Journal of Experimental Microbiology and Immunology (JEMI) Vol. 2:1-5
Copyright  April 2002, M&I UBC

MATERIALS AND METHODS

All materials and methods used in this experiment can be found listed in the “Materials and Equipment” section of Experiment A4 (6). E. coli
K-12 wild-type strain B23 in M9 media containing 0.2% glycerol was incubated overnight in a 37oC shaking waterbath. The overnight culture’s
OD460 was then adjusted in the morning so that the OD460 reading at the start of the experiment was 0.15.
The protocol for this experiment followed that of Experiment A4 (6) with a few modifications. This experiment was carried out utilizing the
following carbon sources: 10% glycerol, glucose, galactose and fructose. In general, the protocol involved growing E. coli in M9-0.2% glycerol
(the control) and adding the lactose analogue isopropyl thiogalactoside (IPTG) to induce β-galactosidase formation. The various sugars listed
above were then added to investigate their affect on bacterial growth and β-galactosidase activity. Bacterial growth was then monitored by taking
E. coli samples at various time points and measuring the turbidity with a Spectrophotometer (OD460). O-nitrophenyl galactoside (ONPG) was
then added to those samples, and the amount of nitrophenol was then analyzed at A420 to measure β-galactosidase activity.

RESULTS

Growth of E. coli diverged shortly after transferring the culture into the three different conditions (glucose,
fructose, and galactose), as indicated in Figure 1. Growth in each condition extensively overlaps with one another
such that it is difficult to tell clearly which condition exhibited the highest or the lowest rate. Trends were not
straight and there were several points that went off the line. Glucose exhibited a biphasic growth: growth slowed
from around 42 minutes (~0.045 OD) to 48 minutes (~0.043 OD) after induction, and resumed at a rate similar to the
first phase. However, no sample was taken in the 12 to 25 minutes after induction. Thus it is difficult to tell when
the first phase began, although growth seemed to increase abruptly from 25 minutes to 30 minutes after induction.
Glycerol exhibited a relatively lower average growth rate as compared to the other three conditions. Samples gave
more erratic results; the measurement at 43 minutes after induction seemed to be an experimental error. Growth
remained lower until it increased from around 0.042 OD at 48 minutes to 0.060 OD at 55 minutes, an almost 50%
increase. Growth then seemed to stop near the end of the experiment around 68 minutes. The trends of fructose and
galactose were similar; the trends were straighter and more reproducible. Both exhibited a steady increase in
growth.

Figure 1: Effect of various sugars on E. coli B23 growth in M9 Media

-1
turbidity (log OD460nm)

-1.5

-2
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70
time after induction (min)

2
Journal of Experimental Microbiology and Immunology (JEMI) Vol. 2:1-5
Copyright  April 2002, M&I UBC

Figure 2: Effect of various sugars on enzyme activity of B-

galactosidase from E. coli B23 growing in M9 Media
3.4
3.2
3
enzyme activity (log mUnits/ml)

2.8
2.6
2.4
2.2
2
1.8
1.6
1.4
1.2
1
0.8
0.6
0.4
0.2
0
-0.2
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70

time after induction (min)

Figure 2 shows the enzyme activity of β-galactosidase in E. coli grown in the different conditions.
Glycerol, the control, showed the highest enzyme activity level as expected. The trend was straight up till around 30
minutes when it shifted to another lower level. The addition of galactose caused a decrease in enzyme induction and
induction remained low throughout the course of the experiment; enzyme induction tripled at the end of the
experiment since the first time-point. However, decrease in induction occurred around 28-29 minutes (at least 10
minutes after the start of treatment). Samples had not been taken during the first 10 minutes, so the time and
strength of the decrease at the start cannot be discerned. Glucose and fructose exhibited similar enzyme induction
patterns, both causing a reduction in β-galactosidase activity that remained low throughout the course of the
experiment. Enzyme induction increased 10-fold by the end of the experiment. In all three conditions (glucose,
fructose, and galactose), enzyme induction by the end of the experiment was approximately 33% that of the control.

DISCUSSION

As shown in Figure 1, the culture grown in glycerol alone stopped growing after 55 minutes and resulted in
the lowest turbidity reading at the end of the experiment. This indicates that glycerol alone is not sufficient to
support the growth of E. coli after a certain point (0.06 OD at 55 minutes). It also implies that other nutrients may
be needed to support the continuation of cell growth after the cell number of a culture has reached to a particular
level. Under three other conditions (glucose, galactose and fructose), cells kept growing throughout the course of
the experiment indicating that the three sugars can be utilized as sole carbon sources by E. coli and are sufficient to
support its growth. Among them, glucose is the most preferential sugar to be utilized by E. coli as the average
reading of turbidity of the particular culture is the highest. Since glucose and fructose exhibit a similar trend that is
found in between that of galactose and glycerol, they may be assumed as intermediately preferential sugars utilized
by E. coli in the experiment.
Figure 2 shows that glycerol did not repress nor slightly repress β-galactosidase induction. Glycerol is
transported into E. coli through facilitated diffusion. Since it does not use the PTS, Enzyme II would be primarily
phosphorylated and cAMP levels would increase. Upon addition of IPTG, a lactose analog, transcription of the lac

3
Journal of Experimental Microbiology and Immunology (JEMI) Vol. 2:1-5
Copyright  April 2002, M&I UBC

operon occurred utilizing available cAMP-CAP complexes, resulting in high levels of the enzyme (8). On the other
hand, the addition of either of the three other sugars (glucose, galactose or fructose) all caused a decrease in the
enzyme induction. Among glucose, galactose, and fructose, galactose exhibited the least decrease in β-
galactosidase induction until around 57 minutes. A slightly higher repression in β-galactosidase induction by
galactose after 57 minutes indicates that galactose represses β-galactosidase activity in a non-steady pattern. An
explanation to this observation could be due to galactose’s ability to induce β-galactosidase and also its ability to
inhibit hydrolysis of lactose by β-galactosidase. When the concentration of galactose inside E. coli is low, β-
galactosidase will be induced. However, when there is too much galactose around and being transported (by the
CAP-dependent Gal P permease instead of PTS (2)) into the cell, the cell does not need to produce β-galactosidase
to hydrolyze so much lactose—feedback inhibition (5).
Fructose and glucose exhibit a very similar trend in repressing β-galactosidase activity. According to our
proposed “composition theory”, fructose should exhibit a lesser ability to repress β-galactosidase activity than
glucose. However, in Figure 2, their trends are too similar to be distinguished from each other. As a result, this
Figure indicates that β-galactosidase activity does not follow the “composition theory” but does not necessarily
contradict it either. The similarity of the fructose and glucose induction patterns might be due to the fact that
fructose and glucose both utilize that PTS, leading to low levels of cAMP and low β-galactosidase induction (3).
Further, it has also been shown that fructose and glucose are equally capable of exerting CCR over sucrose in
Saccharomyces cerevisiae, both suppress sucrose metabolic enzyme expression at 5g/L (1). Thus, there seems to be
no hierarchy of preferred usage between fructose and glucose.

ACKNOWLEDGEMENTS

We would like to thank Dr. Ramey for providing us with E. coli K-12 wild-type strain B23, Karen and Andre for
assisting us in the lab and the media room people for providing us with various equipment. We would also like to
thank our classmates for sharing their insights and equipment.

REFERENCES

1. Dyensen, J., H. P. Smits, L. Olsson, and J. Nielsen. 1998. Carbon catabolite repression of invertase during batch cultivations of
Saccharomyces cerevisiae: the role of glucose, fructose, and mannose. Appl. Microbiol. Biotech. 50:579–582.
2. Hernandez-Montalvo, V., F. Valle, F. Bolivar, and G. Gosset. 2001. Characterization of sugar mixtures utilization by an Escherichia
coli mutant devoid of the phosphotransferase system. Appl. Microbiol. Biotech. 57:186–191.
3. Kornberg, H. L. 2001. Routes for fructose utilization by Escherichia coli. J. Mol. Microbiol. Biotech. 3:355–359.
4. Lengeler, J. W. 1996. The Phosphoenolpyruvate-dependent carbohydrate: phosphotransferase system (PTS) and control of carbon
utilization, p. 231–254. In E.C.C. Lin and A.S. Lynch (eds.) Regulation of Gene Expression in Escherichia coli. R. G. Landes Company.
New York, NY.
5. Lianes, B. and E. McFall. 1969. Effect of galactose on beta-galactosidase synthesis in Escherichia coli K-12. J. Bacteriol. 97:217–222.
6. Ramey, W. D. 2002. Microbiology 421: Laboratory of Experimental Microbiology. University of British Columbia, Vancouver, BC.
7. Stulke, J. and W. Hillen. 1999. Carbon catabolite repression in bacteria. Curr. Op. Microbiol. 2:195–201.
8. Wang, J., E. D. Gilles, J. W. Lengeler, and K. Jahreis. 2001. Modeling of inducer exclusion and catabolite repression based on a PTS-
dependent sucrose and non-PTS-dependent glycerol transport systems in Escherichia coli K-12 and its experimental verification. J.
Biotech. 92:133–158.

4
Journal of Experimental Microbiology and Immunology (JEMI) Vol. 2:1-5
Copyright  April 2002, M&I UBC

APPENDIX

Sample calculation for Enzyme Activity (glycerol time point: 60)

Enzyme Activity = A x 1 x 106 x Nv x 1

ml of enzyme t 15,000 Ev

= 0.090 x 1 x 106 x 3.8 x 1

60 15,000 0.4

= 0.95mUnits