R E V I E W S 33 Mah~,B. et al. (1995) Mol. Microbiol. 18, 33-43 34 Franza, T. et al. (1991) Mol. Microbiol. 5, 1319-1329 35 Franza, T. and Expert,D. (1991)J. Bacteriol. 173, 6874-6881 36 Expert, D., Sauvage, C. and Neilands,J.B. (1992) Mol. Microbiol. 6, 2009-2017 37 Sauvage, C., Franza, T. and Expert, D. (1996)]. Bacteriol. 178, 1227-1231 38 Neema,C., Laulh~re, J-P. and Expert, D. (1993) Plant Physiol. 102, 967-973 39 Masclaux,C., Hugouvieux-Cotte-Pattat, N. and Expert, D. (1996) Mol. Plant-Microbe Interact. 9, 198-205 40 Masclaux,C. and Expert,D. (1995) PlantJ. 7, 121-128 41 Sauvage, C. and Expert, D. (1994) Mol. Plant-Microbe Interact. 7, 71-77 42 Groisman,E.A. and Ochman,H. (1994) Trends Microbiol. 2, 289-294 43 Miller,J.F., Mekalanos, J.J. and Falkow,S. (1989) Science 243, 916-922 44 Nikaido,H. (1993) Trends MicrobioI. 1, 5-7 Do bacteria need to communicate with each other for growth? Arseny S. Kaprelyants and Douglas B. Kell T issue cultures of cells taken from higher, dif- ferentiated multicellular organisms normally need com- plex growth factors (cytokines) for successful cell division. The role of these factors is thought to involve binding at the cell membrane and the production of second messengers, such as cyclic GMP, which activate various segments of primary metabolism, possibly including those responsible for their own synthesis 1. By contrast, it is usually assumed that each bac- terial cell in an axenic (pure) culture can multiply indepen- dently of other bacteria, pro- vided that appropriate concen- trations of substrates, vitamins and trace elements are present in the culture medium, and It is usually assumed that most prokaryotes, when given appropriate nutrients, can grow and divide in the absence of other cells of the same species. However, recent studies have suggested that, for growth, prokaryotes need to communicate with each other using signalling molecules, and a variety of 'eukaryotic' hormones have been shown to stimulate bacterial growth. These observations have important implications for our understanding of bacterial pathogenicity. A.S. Kaprelyants is in the Bakh Institute of Biochemistry, Russian Academy of Sciences, Leninskii prospekt 33, Moscow 117071, Russia; D.B. Kell* is in the Institute of Biological Sciences, Edward LIwyd Building, University of Wales, Aberystwyth, Dyfed, UK SY23 3DA. *tel: +44 1970 622334, fax: +44 1970 622354, e-maih dbk@aber.ac.uk that physicochemical environmental parameters, such as pH, temperature, water activity and oxygen status, are at levels com- patible with the growth of the organism. Current laboratory experience seems to be consist- ent with this idea: the development of bacterial col- onies from single cells on agar plates is commonplace, and the most probable number method is based on the apparently correct assumption that a test tube contain- ing one viable cell will eventually show visible growth or turbidity. Although axenic bacterial cultures are not a homogeneous population 2a, these observations are most easily interpreted as being consistent with 'autonomous' growth. However, especially in the most probable number method, growth is almost always ana- lysed in the presence of culture supernatants or cell- adherent substances introduced with the inoculum, which may contain potent autocrine growth factors pro- duced by the cells during their previous growth phase that are necessary for the initiation of regrowth or division. There is increasing evidence for the widespread importance of chemically mediated inter- cellular communication in bac- terial cultures for some events, including sporulation, conju- gation, virulence and biolumi- nescence. A variety of different autocrine chemicals (phero- mones 4) produced as secondary metabolites are responsible for the social behaviour of prokary- otes, under these perhaps spe- cialized conditions involving obvious cellular differentiation (for reviews, see Refs 5-10). Pheromones differ from nutrients in that they are pro- duced by the organisms themselves, they are active at very low concentrations and, in the absence of pro- hormones, their metabolism is not necessary for growth (although they may of course ultimately be degraded). Are similar types of signalling of wider significance for cell multiplication in growing bacterial cultures? Here, we outline many examples in which an apparently op- timized nutrient medium is insufficient for cell growth, and where there is evidence for the involvement of hormones and pheromones in prokaryotic growth and division. Starvation, lag phase and outgrowth Starvation Much molecular information has recently become avail- able about the physiological changes accompanying Copyright © 1996 Elscvicr Science Ltd. All rights reserved. 0966 842X/96/$15.00 PII: S0966-842X(96)10035-4 TRENDS,N M,CROB OLO(;' 237 VOL 4 NO. 6 JUNE 1996