Competence

Competence is the condition of bacterial cells that enables them to take up naked DNA. When this DNA can integrate into the chromosome by recombination or, in the case of plasmids, establish itself in the cytoplasm, the cells are said to be transformed. Competence is manifested by several species of bacteria—representatives of 29 different genera according to a recent count (1)—but far from all. Even in transformable strains, competence generally depends on the establishment of a particular physiological state. Four species have served as the major experimental models for studies of competence, and we shall therefore confine discussion largely to these: Bacillus subtilis, Streptococcus pneumoniae (“pneumococcus”), Haemophilus influenzae, and Neisseria gonorrhoeae. In view of the universal utility of Escherichia coli, which does not acquire competence naturally, we shall also review briefly the artificial induction of competence in this species.

The basic mechanism of DNA uptake appears to be the same in all naturally transformable organisms. Duplex DNA binds to the cell surface, and is then processed so that just one of the two strands passes through a specialized transmembrane channel into the cytoplasm, where it undergoes recombination with the resident chromosome. In gram-negative bacteria, a means of first passing through the outer membrane is also needed. The systems that regulate the uptake process, however, as well as the environmental stimuli to which these systems respond, appear to be more idiosyncratic, presumably reflecting the very different natural habitats in which the capacity to take up DNA is induced. Furthermore, the regulatory pathways of competence intersect with those of other metabolic activities. It is with the regulation of competence that this entry is chiefly concerned, the uptake process itself being deferred largely to the entry Transformation. We begin with the bacterium for which most is known at both the mechanistic and regulatory levels.

1. B. subtilis

 B. subtilis survives and grows in many environments, but its natural reservoir is the soil. It was early appreciated that competence in this bacterium is a phenomenon induced by decelerating growth (2), such as occurs at the transition from exponential to stationary phase or upon depleting the medium of amino acids. Dependence of competence on transition to the stationary phase appears to reflect inhibition of competence by added amino acids (3). Prototrophs in minimal medium can attain competence in exponential phase (4). We now know that it is just one of a number of late-growth responses, which also include excretion of degradative enzymes, antibiotic production, development of motility, and sporulation (5). Regardless of the many overlapping inputs from these other pathways, induction of competence is mediated by a single transcriptional activator, the ComK protein (6). This protein stimulates expression of the genes that encode the 12 components of the DNA-uptake machine identified to date (7-9). These genes are distributed among four polycistronic transcription units, com G, C, E and F, which specify the DNA-binding apparatus and its assembly, as well as the functions that translocate a single strand of DNA into the cytoplasm. The nuclease that degrades the other strand has not been identified. ComK also stimulates (a) the synthesis of RecA, which catalyzes the final step of transformation, (b) the recombination of the imported strand with the chromosome, and (c) the synthesis of AadAB, a nuclease/DNA helicase that assists the process.

Several intersecting metabolic currents determine the net level of active ComK at a given moment, and hence the degree of competence.. What is remarkable about this system is that positive input from all these factors is necessary for high levels of competence; the absence of any one of them diminishes activation of ComK or reduces transcription of its gene. Stochastic variability, leading to limits on one or more of these factors, may explain why in B. subtilis cultures, unlike those of other species, only about 10% of the cells become competent, even with optimum inducing treatments. The special condition of these cells is underscored by assorted observations of their relatively quiescent metabolic state and reduced buoyant density (4). 

The finding that competence was attained earlier in cultures growing in medium already used for growth of cells to the competent state implied that extracellular signaling molecules promote development of competence (10), and this proved to be the case. Two such molecules are known, both generated by the cell and exported to the medium. The ComX pheromone interacts with the cell-surface protein ComP to initiate a cascade of interactions that culminate in competence (11, 12). Competence stimulating factor (CSF) is probably imported to stimulate this pathway. It is possible that the apparent redundancy of competence-inducing function reflects only the co-opting of the CSF sporulation factor to protect the primary product of the competence-specific pathway, the phosphorylated form of ComA, ComA ~ P (13). Unlike ComX, which seems to be specific for competence, CSF production is intimately linked with the system that regulates sporulation; transcription of its gene requires the Spo0H sigma factor and is stimulated, indirectly, by the Spo0A ~ P transcription

regulator. Grossman (14) has suggested that CSF is also a signaling molecule for sporulation.

 The external actions of ComX pheromone and of CSF have led to the suggestion that these factors communicate cell-density information to the system that regulates competence (14). ComA thus serves as a depot for these signals, and in its phosphorylated form it sends them on by activating transcription of comS (12, 15-18).

This general induction pathway is also subject to nutritional signals: amino acids via CodY (3), and glucose via SinI and SinR (19). How these, as well as signals from the sporulation and enzyme secretion pathways (AbrB, DegU ¼), are integrated to allow the cell to decide whether to commit itself to competence remains an intriguing issue.

2. S. pneumoniae

This organism is an obligate parasite of the respiratory tract. In contrast to B. subtilis, S. pneumoniae readily attains competence during exponential growth in rich medium. Moreover, all the cells become competent, albeit briefly. Competence in S. pneumoniae does share with that of B. subtilis, however, an association with a distinct physiological state. This takes the form of altered surface properties, cessation of cell-wall synthesis, increased cell-wall fragility, and release of DNA into the medium. The onset of competence is very sensitive to culture conditions: temperature, the nature of the medium buffer, the concentrations of magnesium and calcium, and pH. In low-pH medium, there is only one peak of competence, at high cell density, whereas in high-pH medium there is a

succession of peaks (20).

The competence induction pathway of S. pneumoniae shows notable similarities to that of B. subtilis, Induction of competence is mediated by an extracellular signaling molecule, the competence stimulating peptide (CSP) (21-23), and this initiates the induction cascade via a sensor-regulator protein pair (24-26). Induction is also modulated by imported peptides (27, 28,( though in this case the effect is negative (29).

Several proteins make their appearance at competence, some of them associated with the membrane (30); but the only one demonstrated to have a clear role in transformation, indeed in competence itself, is the RecA protein, which mediates homologous recombination (31).

 S. pneumoniae cultures exhibit a particular aspect of competence that, though detected long ago (32,( has received little attention. Filtrates of cultures that have passed their peak of competence prevent development of competence when added to another culture, suggesting that a specific inhibitor of competence is secreted into the medium. Such a factor might be induced at competence just in order to turn it off again, accounting for the brevity of the competence peak.

 3. H. influenzae

H. influenzae lives in the mucus of the respiratory tract, where it may be exposed to large amounts of naked DNA (33). Only its own DNA, however, is taken up with high efficiency. This selectivity results from the presence throughout the genome of numerous copies of a short, specific base sequence, the uptake-signal sequence (USS), which is recognized by the uptake apparatus (34, 35). The outer membranes of competent cells extrude vesicles, termed transformasomes (or “blebs”), which encapsulate the DNA and then release it into the periplasm to allow transport into the cell (36-38) . As in B. subtilis, growth-inhibiting conditions, such as transient anaerobiosis, nutritional downshift, and approach to stationary phase, induce competence. The level of competence attained varies and can be close to 100% of the population. Nevertheless, there appears to be no cell density sensing involved, each cell developing competence independently. H. influenzae cells implanted in the rat peritoneum rapidly develop the capacity to import DNA added with the inoculum and manifest other changes characteristic of the competent state (39).

Several factors important to the development of competence in H. influenzae have been identified, but the picture remains fragmented. Of crucial importance is the intracellular level of cyclic AMP (cAMP). Competence was abolished by mutations in the cyclic AMP receptor protein (CRP) and adenylate cyclase (Cya) genes, but could be restored to a cya mutant by addition of cAMP (40, 41). The cell senses sugar availability through a carbohydrate phosphotransferase system that increases cAMP synthesis in response to declining sugar availability: Mutations in two genes specifying components of this system, ptsI and crr, caused partial loss of competence, a deficiency that can be compensated for by added cAMP (42, 43). Competence in H. influenzae thus responds directly to nutritional state.

 Some clues as to how cAMP might act in competence induction have recently emerged. Notably, cAMP stimulates transcription from a divergent promoter region of two genes essential for transformation, tfoX and rec-1. TfoX was originally identified by a mutation, sxy-1, that increased transformation 100-fold (44). Null mutations in the gene eliminated DNA binding and transformation, whereas overexpression of tfoX caused constitutive competence (45, 46), implying that TfoX is a positive regulator of competence. The finding that TfoX is needed for expression of other competence genes (com101A, dprA, and rec-2) (46, 47) supports this suggestion and hints at a central role in competence similar to that played by the ComK regulator of B. subtilis. Whereas tfoX transcription normally increases only as cells approach stationary phase, addition of cAMP to a log-phase culture causes an immediate stimulation (46).

On the other hand, transcription of rec-1, which encodes the general recombination enzyme that is analogous to RecA of E. coli, is only weakly stimulated by cAMP-CRP (48). It is, however, induced by single-stranded DNA, as befits its function of integrating transforming DNA. This common property of Rec proteins has been proposed to endow Rec-1 with another quite distinct function, that of triggering competence. Induction of competence in H. influenzae may be associated with the conversion of up to 5% of the chromosomal DNA to single-stranded form. Observing that overproduction of Rec-1 led to early development of competence, Stuy (49) suggested that formation of Rec-1/single-stranded DNA complexes might induce competence gene expression, as the equivalent complexes induce the SOS regulon in E. coli. The idea remains untested.

Several new proteins appear in the outer membrane at competence. Some of them, rather than being synthesized de novo, are redistributed from the inner membrane. One mutant that fails to do this also fails to bind DNA. It carries a lesion in the gene por, which encodes a periplasmic disulfide oxidoreductase, suggesting the need to assemble or fold disulfide bond-containing proteins (50). Two proteins probably involved in conducting DNA into the cytoplasm are induced at competence: Rec-2, which shares homology with the presumed transmembrane channel protein ComEC of B. subtilis, and DprA (47). There is as yet no indication that H. influenzae competence responds to a specific extracellular signal. If it does, it now seems unlikely to do so via a sensor-regulator protein pair, such as ComP-ComA in B. subtilis: Mutation of the genes encoding four such systems, as deduced from the genome sequence, failed to affect competence (43). What is clear is that the identity and function of most of the components of the competence regulatory system are still unknown, and that the true place of the known ones in the system likewise remains to be shown.

4. N. gonorrhoeae

Unlike the bacteria already discussed, for which competence is a transitory response to growth conditions, transformable Neisseria strains are competent at all phases of growth (51). An early indication (52) that competence could be induced by a proteinase-sensitive factor present in conditioned medium has not been confirmed. The transformation mutants so far reported are defective in one or more of the individual functions known to be involved in transformation-pili and proteins for DNA binding, transport, and recombination (53-55), but none have suggested the existence of competence factors or regulators. The competent state may be so integrated into the general physiology of Neisseria that a specific regulatory system is not needed, or the regulatory elements are essential and therefore missed in mutant hunts. Transformation of cells in culture is influenced by pH and by the concentrations of mono- and divalent cations (53), but these factors probably affect DNA-cell envelope interactions directly, rather than by regulating competence gene output.

5. E. coli

In 1970, Mandel and Higa (56) discovered that E. coli could be induced to take up DNA by treatment of the cells with Ca²⁺ ions. Although this artificial competence has been a key technique in molecular genetics, its basis is not understood. Suspension of E. coli cells in cold CaCl₂ solution is known to induce the formation of transmembrane pores consisting of an outer shell of polyhydroxybutyrate, connected by Ca²⁺ to an inner sheath of polyphosphate (57). The role of these pores is unclear, because their internal diameter appears to be too small to allow the passage of double-stranded DNA. They do, however, have marked effects on temperature-induced membrane fluidity, and it is possible that, by anchoring an otherwise mobile membrane, they create tension that opens holes elsewhere in the membrane to admit the DNA (57, 58).

 6. Summary

One of the most striking aspects of research on competence to emerge in recent years is the interrelationships between the proteins and regulatory factors involved and those of other phenomena previously considered distinct, perhaps the clearest example being B. subtilis sporulation (14). Indeed, acquisition of competence is usually accompanied by other major physiological changes. Because of this, and because our knowledge of competence is based largely on laboratory observation, it is fair to question whether DNA uptake itself is relevant in natural bacterial habitats. In fact there are good reasons for thinking that it is, and this issue is taken up in the entry Transformation.

References

1. M. G. Lorenz and W. Wackernagel (1994) Microbiol. Rev. 58, 563–602. 

2. C. Anagnostopoulos and J. Spizizen (1961) J. Bacteriol. 81, 741–746. 

3. P. Serror and A. L. Sonenshein (1996) J. Bacteriol. 178, 5910–5915. 

4. D. Dubnau (1993) In Bacillus subtilis and Other Gram-Positive Bacteria (A. L. Sonenshein, J. A. Hoch, and R. Losick, eds.), ASM, Washington, D.C., pp. 555–584. 

5. D. C. Dooley, C. T. Hadden, and E. W. Nester (1971) J. Bacteriol. 108, 668–679. 

6. D. van Sinderen and G. Venema (1994) J. Bacteriol. 176, 5762–5770. 

7. D. van Sinderen, A. Luttinger, L. Kong, D. Dubnau, G. Venema, and L. Hamoen (1995) Mol. Microbiol. 15, 455–462. 

8. D. Dubnau (1997) Gene 192, 191–198. 

9. Y. S. Chung and D. Dubnau (1994) Mol. Microbiol. 15, 543–551. 

10. H. Joenje, M. Gruber, and G. Venema (1972) Biochim. Biophys. Acta 262, 189–199. 

11. R. Magnuson, J. Solomon, and A. D. Grossman (1994) Cell 77, 207–216. 

12. J. Solomon, R. Magnuson, A. Srivastava, and A. D. Grossman (1995) Genes Dev. 9, 547–558. 

13. J. Solomon, B. Lazazzera, and A. D. Grossman (1996) Genes Dev. 10, 2014–2024. 

14. A. D. Grossman (1995) Annu. Rev. Genet. 29, 477–508. 

15. N. M. Nakano and P. Zuber (1991) J. Bacteriol. 173, 7269–7274. 

16. J. Hahn and D. Dubnau (1991) J. Bacteriol. 173, 7275–7282. 

17. L. Kong and D. Dubnau (1994) Proc. Natl. Acad. Sci. USA 91, 5793–5797. 

18. T. Msadek, F. Kunst, and G. Rapoport (1994) Proc. Natl. Acad. Sci. USA 91, 5788–5792. 

19. U. Bai, I. Mandic-Mulic, and I. Smith (1993) Genes Dev. 7, 139–148.

20. J. D. Chen and D. A. Morrison (1987) J. Gen. Microbiol. 133, 1959–1967. 

21. A. Tomasz and J. L. Mosser (1966) Proc. Natl. Acad. Sci. USA 55, 58–66. 

22. L. S. Havarstein, G. Coomaraswami, and D. A. Morrison (1995) Proc. Natl. Acad. Sci. USA 92, 11140-11144.

23. L. S. Havarstein, D. B. Diep, and I. F. Nes (1995) Mol. Microbiol. 16, 229–240. 

24. E. V. Pestova, L. S. Havarstein, and D. A. Morrison (1996) Mol. Microbiol. 21, 853–862. 

25. L. S. Havarstein, P. Gaustad, I. F. Nes, and D. A. Morrison (1996) Mol. Microbiol. 21, 965–971.

971. 26. Q. Cheng, E. A. Campbell, A. M. Naughton, S. Johnson, and H. R. Masure (1997) Mol. Microbiol. 23, 683–692. 

27. G. Alloing, P. de Philip, and J.-P. Claverys (1996) J. Mol. Biol. 241, 44–58. 

28. B. J. Pearce, A. M. Naughton, and H. R. Masure (1994) Mol. Microbiol. 12, 881–892. 

29.G. Alloing, B. Martin, C. Granadel, and J.-P. Claverys (1998) Mol. Microbiol. 29, 75–83. 

30. M. N. Vijayakumar and D. A. Morrison (1986) J. Bacteriol. 165, 689–695. 

31. B. Martin, P. Garcia, M.-P. Castaniè, and J.-P. Claverys (1995) Mol. Microbiol. 15, 367–379. 

32. A. Tomasz and R. D. Hotchkiss (1964) Proc. Natl. Acad. Sci. USA 51, 480–487. 

33. L. Matthews, S. Spector, L. Lemm, and J. Potter (1963) Am. Rev. Respir. Dis. 88, 199–204. 

34. K. L. Sisco and H. O. Smith (1979) Proc. Natl. Acad. Sci. USA 76, 972–976. 

35. D. B. Danner, R. A. Deich, K. L. Sisco, and H. O. Smith (1980) Gene 11, 311–318. 

36. M. Kahn, M. Concino, R. Gromkova, and S. H. Goodgal (1979) Biochem. Biophys. Res. Commun. 87, 764–772. 

37. R. A Deich and L. C. Hoyer (1982) J. Bacteriol. 152, 855–864. 

38. M. E. Kahn, G. Maul, and S. H. Goodgal (1982) Proc. Natl. Acad. Sci. USA 79, 6370–6374. 

39. M. Dargis, P. Gourde, D. Beauchamp, B. Foiry, M. Jaques, and F. Malouin (1992) Infect. Immun. 60, 4024–4031. 

40.  M. S. Chandler (1992) Proc. Natl. Acad. Sci. USA 89, 1626–1630. 

41. I. R. Dorocicz, P. M. Williams, and R. J. Redfield (1993) J. Bacteriol. 175, 7142–7149. 

42. L. P. Macfadyen, I. R. Dorocicz, J. Reizer, M. H. Saier Jr., and R. J. Redfield (1996) Mol. Microbiol. 21, 941–952. 

43. M. L. Gwinn, D. Yi, H. O. Smith, and J. F. Tomb (1996) J. Bacteriol. 178, 6366–6368. 

44. R. J. Redfield (1991) J. Bacteriol. 173, 5612–5618. 

45. P. M. Williams, L. A. Bannister, and R. J. Redfield (1994) J. Bacteriol. 176, 6789–6794. 

46. J. J. Zulty and G. J. Barcak (1995) Proc. Natl. Acad. Sci. USA 92, 3616–3620. 

47. S. Karudapuram and G. J. Barcak (1997) J. Bacteriol. 179, 4815–4820. 

48. J. J. Zulty and G. J. Barcak (1993) J. Bacteriol. 175, 7269–7281. 

49. J. H. Stuy (1989) In Genetic transformation and expression (L. O. Butler et al., eds.), Intercept Inc., Andover, pp. 85–112. 

50. J.-F. Tomb (1992) Proc. Natl. Acad. Sci. USA 89, 10252–10256. 

51. P. F. Sparling (1966) J. Bacteriol. 92, 1364–1369. 

52. A. Siddiqui and I. D. Goldberg (1975) Biochem. Biophys. Res. Commun. 64, 34–42. 

53. G. D. Biswas, T. Sox, E. Blackman, and P. F. Sparling (1977) J. Bacteriol. 129, 983–992. 

54.  G. D. Biswas, S. A Lacks, and P. F. Sparling (1989) J. Bacteriol. 171, 657–664. 

55. I. J. Mehr and H. S. Seifert (1997) Mol. Microbiol. 23, 1121–1131. 

56. M. Mandel and A. Higa (1970) J. Mol. Biol. 53, 159–162. 

57. R. N. Reusch and H. L. Sadoff (1988) Proc. Natl. Acad. Sci. USA 85, 4176–4180. 

58. C. E. Castuma, R. Huang, A. Kornberg, and R. N. Reusch (1995) J. Biol. Chem. 270, 12980–12983.