Starvation-induced Mutations in Escherichia coli
Colby Stoddard
The University of Montana IBS-CORE Program
Mentor- Dr. Barbara Wright
INTRODUCATION- Our laboratory is investigating the effect of starvation on derepression and mutation in catabolic pathways. When a bacteria’s environment cannot supply the necessary nutrients required for cell growth, the cell responds by switching from a state of cell growth to that of survival. This state of survival is identified by a termination in the synthesis of cellular constituents associated with cell growth and replication such as cell wall components, rRNA, and tRNA(1). An increase in the production of proteins that can protect the starving cell from environmental damage is also seen at this time.
Both cya and crp have been shown to be required for
derepression during starvation (2). Cya codes for an enzyme, adenylate
cyclase, which catalyzes the formation of cyclic AMP(cAMP). The cAMP receptor protein, crp, is required to bind cAMP and
activate transcription. Transcription
of genes, coding for enzymes responsible in the use of other carbon sources, is
inhibited during growth in the presence of glucose. Glucose also inhibits the accumulation of cAMP during
growth. When glucose becomes depleted,
cAMP accumulates and can then be bound by crp
to allow derepression. This system
is concerned with derepression in malA and
lacZ, the genes associated with
enzyme production for maltose and lactose catabolism.
In order for genes to be transcribed, they must first be unwound from a double strand to two single strands by way of a polymerase. One of the two single strands is transcribed by RNA polymerase and becomes hybridized with mRNA while the other non-transcribed strand remains static. This hybridization of the transcribed strand results in a stable yet slightly more vulnerable complex than double stranded DNA. In comparison, the non-transcribed strand suffers the effects of mutation due to its un-hybridized nature. This strand has been shown to have a 100-fold increase in depurination (3) and deamination (4) as compared to DNA in the double stranded state. The dramatic increase in mutation is observed only in the derepressed genes, targeted by starvation, and not on a genome-wide scale. Many of the mutations will result in the production of defective cells that are unlikely to survive. However, a small percentage of these mutations will result in a cell that is superior to its parent. For example, if a mutation resulted in a protein with a higher affinity for its substrate, the cell would have a higher rate of survival due to the decreased need for substrate or its more efficient use. The increase in beneficial mutations is essential to the process of evolution because nature will then select the fittest of the new and improved cells at the phenotypic level.
The mutations
observed in this study must be distinguished from background mutations and
other mutations termed ‘adaptive’.
Background mutations occur during growth due to the misincorporation of
nucleotides by DNA polymerase, UV damage and other means (5). These mutations occur at a specific rate and
on a genome-wide scale. Adaptive
mutations occur after prolonged starvation, are random, and are also observed
on a genome-wide scale. The stress induced mutations studied here are different
from adaptive and background mutations due to their specific, non-genome wide
increase in mutation rates. They occur
in specific response to particular conditions of environmental stress.
Experimental Design
Mutation rates of the Cold Springs Harbor (lacZ-) and CP78 (mal-) cells will be compared to the double mutant (lacZ-,cya-), (lacZ-, crp-), (mal A-, cya-) and (mal A-, crp-) strains. In the presence of starvation conditions, strains encoding a defective cya or crp will not be able to initiate transcription. The parent mal A- and lacZ- cells are predicted to have a higher mutation rate than the double mutants, due to their ability to become derepressed. B-galactosidase is the enzyme product of the lacZ gene required for growth when lactose is the sole carbon source(6). Activity of the lacZ gene in the Cold Springs Harbor strain can be monitored through a B-galactosidase assay. Monitoring of lacZ before starvation should demonstrate the lacZ is not active and thus is not in an increased state of vulnerability.
Methods and
Materials
Mutant and wildtype CP78 strains are grown for 20 to 30 hours in minimal media consisting of 50 mM NaPi (pH=6.4), 13 mM KCL, 7.5 mM (NH4)2SO4, 1.7 mM MgSO4, 0.0005% B1, 1 mM threonine, and 0.3 mM histidine, arginine, and leucine. Growth media for the Cold Springs Harbor strain is the same as that for CP78 with respect to NaPi and salt concentrations, and requires the addition of 0.3 mM tryptophan. The concentration of glucose used to limit growth in all strains is 0.6 mM. Solid media used for plating after starvation is the same as the growth media but is supplemented with 0.5% lactose or maltose instead of glucose.
To
determine a mutation rate, 75 ml of 0.6 mM limiting glucose minimal media is innoculated with
cells from an overnight culture grown on a nutrient agar plate. 1.5 ml aliquots are then distributed to
46-50 ml culture tubes which are then shaken 20 to 30 hours at 37oC. Cell growth is followed by reading the OD550 of three tubes until the
cells are in stationary phase. At the
time of plating, the OD550 is read on three tubes to find two ‘average’
cultures. Serial, 10-fold dilutions are
then preformed on the two cultures and the 10-6 and 10-7
are plated to nutrient agar to determine the viable number of cells per
ml. Mutant colonies begin appearing at
40 hours after plating and final counts are made at 48 hours after plating
(7). Mutation rates are determined by
the ‘zero’ method according to the expression
MR = (-ln 2)/(ln Po/N) where Po is the proportion
of negative plates and N is the viable cell count(8).
The B-galactosidase assay was preformed during growth by harvesting cells and treating them with Complete Z buffer and a reducing agent, B-mercaptoethanol. Cell lysis and protein denaturation is accomplished by the addition of SDS and chloroform. After a 5 minute incubation, the cells are then centrifuged for 10 minutes at 1,500 RPM’s. The supernatent is added to an ortho-nitrophenyl galactopyranoside solution and incubated at 37oC for 30 minutes. The indicator reaction is stopped with the addition of sodium carbonate. The OD420 is then read to find the B-galactosidase activity in Miller’s units.
Transductions
using a P1 bacteriophage intermediate
were used to construct the desired strains.
Escherichia coli parent
strains X6161 and X6162 were used to supply the mutant cya and crp genes. Since these genes are ‘linked’ to
tetracycline resistance, strains with the desired genotype can be selected for
by their ability to grow on tetracycline media.
Results/Discussion
The results for the mutation
rates of CP78, CSH and their mutant strains is shown in table 1. The mutation rate, from mal- to mal+ in CP78, was
determined to be 24-fold higher than its crp
mutant and 9-fold higher than the cya
mutant. The effect was less dramatic in
the lac- to lac+ mutation rate of the CSH strains. The crp mutant showed a
6-fold decrease in mutation rate where as the cya mutant showed a 3-fold decrease. The mutation rate of the crp
mutant was shown to have a 4-fold decrease in mutation rate from CP78 to CSH
while the cya mutation rate was
decreased by 3-fold. A total of 80 cultures
for CP78, 120 cultures for CSH, and 160 cultures for each of the double mutants
was used to summarize the data in table 1.
|
|
Total cell
number
(X 108) |
P0* (#neg/tot plate) |
Mutation
Rate**
(X 10-10) |
|
CP78 (mal-) |
8.75 ± .32 |
.238 ± .053 |
11.36 |
|
CP78 (mal-, cya-) |
2.87 ± .20 |
.950 ± 0 |
1.24 |
|
CP78 (mal-, crp-) |
2.77 ± .11 |
.981 ± .013 |
0.48 |
|
CSH (lac-) |
3.70 ± .39 |
.742 ± .052 |
5.59 |
|
CSH (lac-, cya-) |
2.77 ± .89 |
.925 ± .065 |
1.95 |
|
CSH (lac-, crp-) |
2.97 ± .94 |
.963 ± .075 |
0.88 |
* If all plates were
negative the Po number was assumed to be 1.
** One MR is determined for all
cultures using an average Po and N= according to the Luria & DelBruck
expression.
The
B-galactosidase assay and the low
number of lac- to lac+ and mal- to mal+ colonies
observed on each positive plate indicated the mutations occur at or after the
cell becomes starved and derepressed. Growth of mutant cya/crp cells and the parent strains in lactose containing media
was monitored by reading the OD550.
B-galactosidase activity was
also monitored at this time. Fig. 1 compares these
activities to the activity of cells prior to plating in a mutation rate. The mutant cya/crp strains grown in lactose media and those grown in a
mutation rate experiment did not produce B-galactosidase during growth. The wildtype lac+ strains did show a dramatic increase in B-galactosidase
production which occured directly after derepression. The higher mutation rate of CSH, when compared to the crp/cya mutants, is believed to be due
to the elevated vulnerability to mutation upon activation of the lacZ gene.
This study has shown cya and crp to be required for the mutation rates seen in CP78 and the Cold Springs Harbor strains. The vulnerability of DNA to depurination and deamination during transcription can reasonably explain the stimulated the rate of mutation in parent strains. Transcription of the lacZ gene was shown not to occur prior to derepression indicating the mutations occur at or after starvation. In nature, it is likely for the metabolism of a cell to be that of starvation for a carbon source. Thus, activation of genes able to use an alternate carbon source such as maltose of lactose will induce those genes to become highly mutable. The resulting mutants should include some able to survive even better than their parent, e.g. by having a higher affinity for their carbon source. Thus, such variants would be selected to evolve.
I would like to thank Dr. Barbara Wright for providing excellent
technical and financial assistance during this fellowship. This study was made possible by a Project
IBS-CORE Undergraduate Research Fellowship, provided by a grant from the Howard
Hughes Medical Institute to the University of Montana.
References
1) Cashel, M., and Rudd, K.E. (1987) The Stringent Response. In Escherichia coli and Salmonella typhimurium. Vol.2. Neidhardt, F.C., Ingraham, J., Low, K., Magasinik, B., Schaechter, M., and Umbarger, H.E. (eds). Washington D.C: American Society for Microbiology, pp. 1410-1438
2) Saier, M.H., Ramseier, T.M., and Reizer, J. 1987. Regulation of Carbon Utilization. In Escherichia coli and Salmonella typhimurium. Vol.2. Neidhardt, F.C., Ingraham, J., Low, K., Magasinik, B., Schaechter, M., and Umbarger, H.E. (eds). Washington D.C: American Society for Microbiology, pp. 1325-1343
3) Lindahl, T. and Neiberg, B. (1972) Biochemistry 11: 3610-3618
4) Lindahl, T.
and Neiberg, B. (1974) Biochemistry 13: 3405-3410
5) Fehr, S., and Richter, D. (1981) Stringent respnse of Bacillus stearothermophilus: evidence for the existance of two distinct 3,5 polyphosphate synthetases. J. Bacteriol 145:68-73
6) Brock, T.D. (1997) Biology of Microorganisms. Eigth Edition. Madigan, M., Martinko, J., Parker, J. Upper Saddle River, New Jersey. pp.233
7) Wright, B.E. (1996) The effect of the stringent response on mutation rates in Escherichia coli K-12. Molecular Microbiology 19(2): 213-219
8) Luria, S.E., and DelBruck, M. (1943) Mutations of bacteria from virus sensitivity to virus resistance. Genetics 28: 491-511