World Journal of Microbiology & Biotechnology 14, 1±5

Review: Bacterial population genetics

J.T. Trevors Bacterial population genetics is the study of natural bacterial genetic diversity arising from evolutionary processes. The roles of molecular mistakes, restriction±modi®cation, plasmids and gene transfer in bacteria are also important components of population genetics. These aspects are of considerable scienti®c importance from a fundamental perspective, because of the short generation times of bacteria, their microscopic cell size, the large population sizes bacteria can achieve and their different mechanisms of gene transfer. Key words: Bacteria, DNA, evolution, genetics, genome, molecular, population.

Introduction Bacterial population genetics (Hopwood & Chater 1989) can be considered to be the study of natural bacterial genetic diversity arising from evolutionary processes (see Figure 1). During the last decade, genetically-engineered microorganisms (GEMs) have been researched for their potential use in industrial and environmental applications. In addition, the release of GEMs into the environment, especially where repeated applications have taken place, means that some genes may move from GEMs released (intentionally or accidently) into indigenous bacteria/ microorganisms at varying frequencies. The impact of genetic engineering and the use of GEMs may have a considerable impact on future studies of bacterial population genetics, because of the actual numbers of cells and copies of genes that will be produced for use in industrial/environmental applications and the potential for the repeated release of these cells into the environment. Bacterial population genetics poses numerous problems to microbiologists. For example, natural mechanisms of bacterial gene transfer (transformation, conjugation and transduction) as well as recombination, mutation and genetic drift have been studied for selected bacterial species, yet there is a paucity of information on how gene transfer can be modelled or described at the population level (Baumberg et al. 1995) in the open environment. Reproduction by cell division cannot be easily linked to genetic recombination events. Moreover, bacterial populations may not be highly clonal but may The author is the supervisor of the Laboratory of Microbial Technology, Department of Environmental Biology, University of Guelph, Guelph, Ontario, Canada, N1G 2W1; fax: 519 837 0442.

be undergoing genetic recombination over a range of frequencies. More information is gradually becoming available. For example, a model for the transfer of the mer-operon (mercury volatilization) between microorganisms has been described by Childress & Sharpe (1992). This model describes the genetic states (capability of volatilizing mercury due to presence of plasmid carrying the mer-operon) as a result of gene transfer of this operon and the cell numbers present. The role of self-transmissible and mobilizable plasmids that have broad host ranges also contributes to the evolution of bacterial populations. A better understanding of bacterial population genetics will assist with advances in evolutionary microbiology. The short generation times of bacteria under optimal growth conditions and even under non-optimal growth conditions, combined with the immense numbers of cells in bacterial populations (Maynard Smith 1995) and their microscopic size (in the order of microns) means that the population genetics of most microorganisms is different from that of eukaryotic organisms (see Figure 1). These characteristics mean that evolutionary change on a time scale can be very rapid, occurring in a generation time of under an hour for some bacterial species. Conversely, since bacteria are the ®rst and oldest living organisms, they collectively have had more generations than any other group of organisms in which to evolve. Although today's bacteria may not be the same as bacteria in the distant past, many are probably not very different. Table 1 summarizes some of the factors to be considered when studying bacterial/microbial population genetics. These include the microscopic size (microns) and rapid generation times. In addition, numerous bacterial cells are viable but non-culturable (VBNC), which hampers investigations on population genetics. In

ã 1998 Rapid Science Publishers World Journal of Microbiology & Biotechnology, Vol 14, 1998

1

J.T. Trevors

Figure 1. Factors involved in bacterial population genetics.

Table 1. Some of the factors that cause dif®culties in the study of bacterial/microbial population genetics. 1. 2. 3. 4. 5. 6.

Microscopic size (lm) Rapid growth Viable but non-culturable state (VBNC) Dif®cult to recover viable cells from environmental samples Many species have not been isolated or identi®ed Plasmids are self-transmissible and mobilizable and can compose part of the genome in some species 7. Diverse methods of gene transfer: transformation, conjugation and transduction 8. Physicochemical heterogeneity of environmental samples such as soil, water and sediment ± dif®cult to obtain microbial population data

microbiological research, studies are often easier to conduct, and more historical literature is available if the species under investigation are culturable on laboratory media. Moreover, many bacterial cells that are viable and culturable, are not always easily isolated from environmental samples such as soil, sediment and water. The changing physicochemical parameters of the environment and its heterogeneity also hamper an understanding of bacterial population genetics, their numbers and diverse metabolic activities.

Molecular Mistakes and Evolution of Bacterial Populations Mistakes in the replication of bacterial genetic material (DNA, RNA, PNA-peptide nucleic acid, some other unknown primitive genetic material such as clay crystal genes or precursors of RNA/DNA) (Trevors 1995) would be a certainty during evolution, especially when the genetic material was transmitted over billions of years (Reanney 1984). Interacting physical, chemical and biological factors could have produced mutations, deletions, insertions and recombination events in the genetic material, especially in DNA once it became the master

2

World Journal of Microbiology & Biotechnology, Vol 14, 1998

template or genetic code. During replication of genetic material, changes (or mistakes) had a signi®cant role in prokaryotic evolution and therefore in bacterial population genetics. It would be virtually impossible for genetic information to be transmitted over time without changes (some being mistakes) in the copying process (replication). The ability of evolving bacterial genomes to remain somewhat constant while altering their genomes (Riley 1989) and gene orders provides a means for evolution of bacteria (Trevors 1995; see also Trevors 1996a, 1996b). Errors in DNA replication in eukaryotic organisms occur at a rate of about one incorrect nucleotide for every 109±1011 bases during replication (Reanney 1984). The present day situation is probably very different from the replication of primitive genetic material which may have exhibited replication errors of 10)1±10)2 per base per generation (Reanney 1984). This means that the occurrence of mistakes probably decreased in frequency, as DNA replication evolved to the present status found in bacteria and other organisms. This re®nement was necessary if microorganisms were to evolve and maintain relatively stable genome sizes and gene orders while at the same time possessing a mechanism for molecular evolution. This opposing balance between a degree of genome constancy (stability) and a degree of change (Riley 1989) is required for evolution and the diversity of species. This is possible as small changes in DNA sequences can cause signi®cant changes in gene products and, therefore, the organism. The realization that small changes in DNA or some other genetic material produce signi®cant differences in organisms is central to molecular evolution, especially since it has occurred over billions of years. If mistakes occurred during copying of genetic material it is possible that multiple copies of a single bacterial chromosome or extrachromosomal plasmid DNA would be of bene®t in evolution. For example, plasmids are an excellent structure for supplying multiple identical copies of DNA in host bacterial cells. The presence of multiple copies means that some copies may not contain errors. Some plasmids are also self-transmissible by conjugation between compatible donor and recipient bacterial cells. Moreover, possibly two or more, small bacterial genomes could have fused and resulted in a single larger prokaryotic genome (Trevors 1996b). Both plasmid transfer and genome fusion may have provided a mechanism for accelerated molecular evolution provided that the bacterial cell(s) could maintain the fused genome and that metabolic pathways were integrated to accommodate the fused genome. Also, some genetic information may have been present in more than one copy. This means that some copies could be lost without loss of the message, as long as one copy was retained. It is possible that evolving bacterial genomes were smaller

Population genetics than present day bacterial genomes and/or were composed of large segments of non-coding DNA. Biochemical pathways in bacteria were probably minimal because integrated metabolism and regulation of this metabolism are products of evolution. Having access to a diverse pool of genetic material may also be linked to restriction±modi®cation (R±M), de®ned as endonucleases and methylases catalytically active at the same DNA sequences (Price & Bickle 1986). DNA that is methylated cannot then be a substrate for restriction (Price & Bickle 1986). The normal substrate for restriction enzymes is foreign DNA entering cells as phages or plasmids, or naked DNA entering competent cells by genetic transformation (natural or induced competence). R±M systems are common in many microorganisms, which suggests an important role for them in the evolution of microbial populations. R±M systems are not necessary for bacterial growth, DNA synthesis, genetic recombination or repair of DNA (Price & Bickle 1986; Trevors 1995) but assist in maintaining a genome while cleaving foreign DNA. This may have made small fragments of DNA available for genetic recombination in the bacterial cells; as R±M systems became more ef®cient at digesting foreign DNA, recombination may have decreased in frequency and the bacterial genome became more constant or stable. A more stable genome may be less susceptible to replication errors in the genetic message over time. Errors would then be caused by external events such as radiation, ¯uctuating temperatures, changing nutrient status of the system and the presence of mutagenic chemicals. Prokaryotic cells have little control over these external environmental in¯uences. Cells may have been subjected to lethal chemicals and radiation with some resistant cells surviving and the evolutionary process continuing. Relative genome stability and some change are the opposite complementary forces required for evolution (Riley 1989).

Plasmids and Evolution Possibly, multiple copies of plasmid DNA could have eventually been produced from primitive RNA genome(s). Plasmids may not have a signi®cantly different guanine and cytosine (G + C) content from the bacterial host chromosome (Hardy 1987). However, this is not true in all microorganisms. Differences in the G + C content between plasmid(s) and the chromosome of the host cell can occur when a plasmid is transferred into other microbial host(s). Broad host-range plasmids like RP4 and R330B can be maintained in Gram-negative bacteria that have different G + C chromosomal ratios (Datta 1985). Datta suggested that the base composition of plasmids may indicate their origin within certain bacterial host cells, or a tendency of the original plasmid DNA to

change to conform to the G + C content of those cells. Conversely, the similarity between the G + C content of some plasmids and their host's bacterial chromosome may be due to the chromosome being an ancestor of earlier multiple plasmids. It is interesting to consider whether several small plasmids could have been assembled to produce small evolving bacterial genomes with limited function(s). Since plasmids can exist at more than one copy per host cell, some copies of the plasmid(s) may have assembled into evolving genomes while other copies of the same and different plasmids were dispersed throughout the environment and participated in evolving genomes at other suitable locations. The plasmid contained in primitive lipid or lipid-protein micelles would need to be stable while being dispersed. The relationship between G + C content of a plasmid and the host bacterial chromosome may be evolutionary if the earlier bacterial genome was a small plasmid genome that increased in size by cycles of restriction and new sequences being ligated into the growing genome together with rearrangements in the order of sequences. Enzymatic restriction and ligation were possibly not present. Therefore, other mechanism(s) that have yet to be determined, would be necessary to allow the small genome to grow in size and for sequences to change in order.

Gene Transfer and Genetically-Engineered Microorganisms (GEMs) in the Environment Natural gene transfer in the environment allows microbial populations to adopt and evolve (see Figures 2 and 3). However, the frequency of gene transfer events are dif®cult to estimate in the natural environment. Therefore, most data are from laboratory microcosm experiments. The frequency of gene transfer events in soil, sediment and water also depend on ¯uctuating and

Figure 2. Gene ¯ow in soil by transformation.

World Journal of Microbiology & Biotechnology, Vol 14, 1998

3

J.T. Trevors interacting biological, chemical and physical conditions. Table 2 contains a summary of selected factors that can exert a considerable in¯uence on gene transfer events in soil microbial communities. The physicochemical characteristics of the environment in which gene transfer is occurring is of paramount importance. The biological factors summarized in Table 2 also in¯uence bacterial/ microbial gene transfer events in soil. Transformation is the uptake and integration of DNA by competent microbial cells (Porter 1988; Trevors & van Elsas 1997). The state of competence means that the bacterial cells can bind and take up the DNA in a form resistant to intracellular digestion by DNases, followed by integration of the DNA into the host chromosome. The reader is referred to an excellent review of natural genetic transformation in the environment by Lorenz & Wackernagel (1994). Bacterial conjugation (Figure 3) is also an important mechanism of gene transfer in some bacterial species. A conjugation bridge is established between the donor and

Figure 3. Bacterial gene ¯ow in soil by conjugation.

recipient cells through which a copy of the plasmid is transferred to the recipient cell, which in turn becomes a transconjugant and is able to act as a donor cell to a suitable recipient cell. In transduction, a bacteriophage infects the host bacterial cell and can transfer a portion of the host's genome to another bacterial cell, where genetic recombination can

Table 2. Some of the factors that have an in¯uence on microbial gene transfer in soil. Factor Soil physicochemical factors Soil type Bulk density Soil water content, wetting/drying cycles Soil pH Aerobic or anaerobic Nutrient status Presence of other chemicals (e.g. salinity) Presence of toxic pollutant(s) Soil temperature, freezing±thawing cycles Duration of gene transfer experiment Soil biological factors Concentration of transforming DNA Availability of transforming DNA Length or size of transforming DNA Numbers and ratio of donor to recipient cells Survival of donor and recipient cells Repeated applications of same or different cells to the same environment Ingestion of donor and recipient cells by protozoa Presence of plant rhizosphere Density of transducing phages and bacterial hosts

Comments Textural class Especially important when plants are present or intact cores are being used Dry soils decrease gene transfer May not be optimum for the species being investigated and gene transfer is decreased Most studies on gene transfer have been conducted under aerobic conditions Amendments such as NPK, organic matter may in¯uence gene transfer Effect vary depending on chemical(s) and microbial species Toxic chemicals may decrease or increase gene transfer Minimum, optimal and maximum values for gene transfer Length of time must be suf®cient for optimal or at least detectable gene transfer Suf®cient naked or soil bound DNA to transform competent cells DNA may be digested by soil nuclease activity or complexed to soil particles and unavailable DNA may originate from plasmid or chromosomal DNA and should be able to enter competent recipient cells Conjugation is usually favoured when donor to recipient ratios are in the range 1:1±1:10* If cells do not remain viable after gene transfer, the frequency cannot be estimated May or may nor alter gene transfer frequency This may cause an underestimation of gene transfer and survival of cells will be decreased Plant exudates and root surfaces may increase frequency of gene transfer Must be suf®cient for transduction to be detected

* The probability of a donor cell physically contacting a recipient cell is dependent on the donor recipient ratio. For example, if the number of donor cells is high and the number of recipient cells is low, a donor cell is more likely to come in contact with another donor cell instead of a recipient cell. If there are more recipient cells than donor cells, there is a better chance of a donor cell contacting a recipient cell.

4

World Journal of Microbiology & Biotechnology, Vol 14, 1998

Population genetics occur with the host DNA. There is still a paucity of knowledge on transduction in soil, as it is often dif®cult to detect and quantify. More research has been conducted on transformation and conjugation in bacteria in soil. All three methods of gene transfer allow bacterial populations to evolve over very short periods of time (minutes to hours). This rapid pace of evolution and the immense numbers of bacterial cells in a gram of soil (up to billions) makes bacterial population genetics an important component of the biosphere and of our understanding of microbial processes. The amount of genetic variation may or may not increase if different recombination events occurred between GEMs and indigenous environmental microbial species. If a single GEM or naturally-occurring bacterial species is repeatedly introduced into an environmental site over an extended period of time (months to years) it has the potential to become the dominant species, especially if it is robust in the environment, competitive and has special advantages in the environment it is in. There is a paucity of information on repeated applications of one or more GEMs into environmental sites or in laboratory studies. This information is central to the safe and bene®cial uses of GEMs in the environment over extended periods of time. The population genetics of released microorganisms is central to environmental biotechnology, as genes will not always stay in the microorganisms they were originally placed in, and microbial populations will not remain unchanged even over short durations of time. The least predictable potential interaction is probably gene transfer (Young 1989). Also, as discussed by Reanny (1984), genetic information cannot be transmitted over time without mistakes.

Acknowledgements This research was supported by the Natural Sciences and Engineering Research Council of Canada (NSERC) operating grants program.

References Baumberg, S., Young, J.P.W., Wellington, E.M.H. & Saunders, J.R. (eds) 1995 Population Genetics of Bacteria, Symposium of the

Society for General Microbiology. Cambridge: Cambridge University Press. Childress, W.M. & Sharpe, P.J.H. 1992 A compartment model approach to bacterial population genetics and biodegradation. In Modelling the Metabolic And Physiologic Activities of Microorganisms, ed Hurst, C.J. (pp. 61±87). New York: John Wiley & Sons, Inc. Datta, N. 1985 Plasmids as Organisms. In Plasmids in Bacteria eds Helinski, D.R., Cohen, S.N., Clewell, D.B., Jackson, D.A. & Hollaender, A. pp. 3±16. New York: Plenum Press. Hopwood, D.A. & Chater, K.F. (eds) 1989 Genetics of Bacterial Diversity, New York: Academic Press. Lorenz, M.G. & Wackernagel, W. 1994 Bacterial gene transfer by natural genetic transformation in the environment. Microbiological Reviews 58, 563±602. Maynard-Smith, J. 1995 Do bacteria have population genetics? In Population Genetics of Bacteria eds Baumberg, S., Young, J.P.W., Wellington, E.M.H. & Saunders, J.R. pp. 1±12. Cambridge: Cambridge University Press. Porter, R.D. 1988 Modes of transfer in bacteria. In Genetic Recombination eds Kucheklapat, R. & Smith, G.F. pp. 1±41. Washington: American Society for Microbiology. Price, C. & Bickle, T.A. 1986 A possible role for DNA restriction in bacterial evolution. Microbiological Science 3, 296± 299. Reanney, D. 1984 Genetic noise in evolution. Nature 307, 318± 319. Riley, M. 1989 Constancy and change in bacterial genomes. In Bacteria in Nature eds Poindexter, J.S. & Leadbetter, R. pp. 359±388. New York: Plenum Press. Trevors, J.T. 1995 Molecular evolution in bacteria. Antonie van Leeuwenhoek 67, 315±324. Trevors, J.T. 1996a Genome size in bacteria. Antonie van Leeuwenhoek 69, 293±303. Trevors, J.T. 1996b DNA in soil: adsorption, genetic transformation and molecular evolution. Antonie van Leeuwenhoek 70, 1±10. Trevors, J.T. & van Elsas, J.D. (eds) 1995 Nucleic Acids in the Environment: Methods and Applications, Heidelberg; Germany: Springer-Verlag. Trevors, J.T. & van Elsas, J.D. 1997 Quanti®cation of gene transfer in soil and the rhizosphere. In Manual of Environmental Microbiology, eds Hurst, G.J. Knudsen, C.R. Mclnerney, M.L. Stetzenbach, L.D. & Walter, M.V. pp. 500±508. Washington, DC: American Society for Microbiology. Young, J.P.W. 1989 The population genetics of bacteria. In Genetics of Bacterial Diversity eds Hopwood, D.A. & Chater, K.F. pp. 417±438. New York: Academic Press.

(Received in revised form 30 January 1997; accepted 3 February 1997)

World Journal of Microbiology & Biotechnology, Vol 14, 1998

5