Size Limits of Very Small Microorganisms

Abstract

The marine environment harbors enormous numbers of viruses and prokaryotes, existing in complex communities that span a wide spectrum of biotopes, lifestyles, and size ranges. Many naturally occurring marine bacterioplankton are extremely small, some measuring < 0.3 ^mm in their largest dimension, having estimated biovolumes as low as 0.027 ^mm³. Available data suggest that the majority of naturally occurring bacterioplankton resist cultivation, and have not been phylogenetically identified at the single cell level. Phylogenetic evidence for the evolution of major lineages that are characteristically small have not been reported (but they may exist). Because a large fraction of naturally occurring microorganisms have not been cultivated, their specific physiological traits are largely unknown. Consequently, the fraction of very small marine microbes that transiently and reversibly exist as "dwarf cells" is also unknown. Finally, although extremely small (< 0.1 ^mm) DNA-containing particles are very abundant in seawater and are thought to be viruses, the fraction of these particles that may actually represent cellular organisms is uncertain.

Introduction

Small microorganisms are ubiquitous in ocean waters, averaging about 5 x 10⁵ cells/ml in the upper 200 m, and 5 x 10⁴ cells/ml below 200 m depth. The total number of prokaryotic cells in ocean waters is about 1 x 10²⁹(1). Assuming a biomass of approximately 20 fg carbon per cell, this represents 2.2 x 10¹⁵ g of prokaryotic carbon in the world's oceans. This biomass represents an enormous pool of genetic variability, a large fraction of which is represented by very small cells (2,3). Extremely small cells (< 0.5 ^mm) may result from a genetically fixed phenotype maintained throughout the cell cycle. Alternatively, very small cells may reflect physiological changes associated with starvation, or other aspects of the cell cycle. Both explanations likely hold for different members of complex mixed populations of small cells found in the ocean. Extremely small (<0.1 ^mm) DNA-containing particles are also very abundant in seawater, reaching concentrations of about 1 x 10⁷ particles/ml in surface waters (4-6). These small particles are thought to consist largely, although not necessarily entirely, of viruses.

Cell dimensions of cultured or naturally occurring bacteria can be derived from several sorts of data, each with inherent limitations. A number of uncertainties can be associated with cell size estimates. Historically, the existence of very small bacteria and viruses was first documented by observations of infectious filterable agents. Indirect cell size estimates have more recently been derived from filter fractionation experiments using membrane filters of uniform pore size. These sorts of size estimates can be compromised by filter trapping effects, as well as differential retention of cells with varying shapes or cell wall compositions. Cell dimensions and biovolumes are now more frequently estimated via fluorescent nucleic acid staining and epifluorescence microscopy, or flow cytometry. Fluorescent DNA stains can also sometimes be misleading, because the visualized nuclear material may not accurately reflect the actual cytoplasmic volume (7). Most estimates by light microscopy, electron microscopy, and flow cytometry also involve the use of fixatives that may cause cell shrinkage or other artifacts (3). Nevertheless, it is apparent that the majority of naturally occurring prokaryotes in marine plankton are about 1.0 ^mm or less in their largest dimension, and a good number of these are 0.5 ^mm or less in diameter (2,3).

Critical Assessment of the Issue

1. What is the phylogenetic distribution of small bacteria?

This question can be broken down into several components:

A. What is the phylogenetic distribution of cultivated prokaryotes with a stable, very small cell size?

The ongoing efforts of microbiologists to cultivate new microbial groups are currently providing new perspectives and answers to this question. It is still an open-ended question, because new microbial groups continue to yield to cultivation efforts. Recently isolated bacteria having stable, maximal dimensions of around 0.5 ^mm, fall into the alpha Proteobacterial lineage, as well as the Bacterial order Verrucomicrobiales.

Very small bacteria in the order Verrucomicrobiales have been recently isolated. New strains isolated from an anoxic rice paddy displayed a stable cell size of about 0.5 ^mm in length and 0.35 ^mm in diameter yielding a cell volume of about 0.03 ^mm³ (8). These bacteria were oxygen-tolerant heterotrophs, exhibiting strictly fermentative growth on sugars. Other cultivated relatives, including Verrucomicrobium spinosum, are generally larger than 1 ^mm and possess prosthecae (9). Small cell size is therefore not an inherent property of members of this order.

A very small marine isolate with cell volume ranging from 0.03 to 0.07 ^mm³ was isolated using the dilution culture technique of Button and Schut (10). This isolate was found to be associated with the alpha Proteobacterial genus Sphingomonas (11). Sphingomonas sp. strain RB2256 is heterotrophic, contains about 1.5 fg DNA/cell, and grows at a maximal rate of about 0.16 hr-1. This marine Sphingomonas isolate showed very little variation in growth rate or cell size in response to 1,000-fold variation in nutrient supply, indicating the stability of the small cell phenotype (12). Other Sphingomonas species have larger, more typical cell sizes, so diminutive size is not a specific characteristic of the genus.

Nanobacteria species have been reportedly found in association with human and cow blood (13). They have been cultured in serum-free media, and have cell diameters, estimated from electron microscopy, of 0.2 to 0.5 ^mm (13). They have been reported to pass through 0.1 ^mm filters, apparently due to pleomorphic forms even smaller, about 0.05-0.2 ^mm (13). Ribosomal RNA sequences originating from these microorganisms are associated with the alpha subdivision of the Proteobacteria, and are most closely related to Phyllobacterium rubiacearum.

B. What is the phylogenetic distribution of cultured prokaryotes that undergo an induced cell cycle transition from a "typical" to very small cell size?

A significant number of bacteria have been observed to undergo a transition from a large, actively growing state, to a dormant state of much smaller cell size (14-16). Some of these physiologically induced small cells reduce to cell volumes as low as 0.03 ^mm³. Different bacterial genera have been reported to undergo a starvation-induced response resulting in cell miniaturization, including the gamma Proteobacteria genera Vibrio, Pseudomonas, Alcaligenes, Aeromonas, and Listonella (14). This reduction in cell size may be a common phenomenon for heterotrophic microorganisms adapted for growth at relatively high nutrient concentrations. In many of these microorganisms, the transition from large to dwarf cells is fully reversible upon nutrient upshift. This physiological strategy appears to be common, but its actual distribution among diverse bacterial phyla is poorly characterized. It is unknown what fraction of naturally occurring "ultramicrobacteria" represent typically larger cells that have experienced nutrient downshift and undergone cellular miniaturization. It is also not clear what fraction of these readily reverse to a large actively growing state (15), or alternatively have entered a hypothetical "viable but nonculturable" state (16).

C. What are the phylogenetic identities of (uncultivated) very small cells frequently observed in natural environmental samples?

This remains an open question. It has been estimated that only about 0.1-1% of naturally occurring prokaryotes have been cultivated from many specific habitats (17,18). Culture independent surveys have indicated the presence of many new, yet uncultivated, and previously unrecognized prokaryotic groups (19). Most of these have not yet been specifically identified at the single cell level. It will be interesting to determine whether a significant fraction of recently discovered, uncultivated prokaryotic groups represent some of the more diminutive cell forms. Are there inherent properties of very small cell lineage that render them recalcitrant to cultivation?

2. Is there a relationship between minimum size and environment?

In low-nutrient habitats in marine plankton, cells typically appear smaller in size than those of comparable higher nutrient habitats. To the extent that some cells undergo a starvation response that involves reduction in cell size, there may be a loose relationship between cell size and ambient nutrient concentration. However, it is still unknown what fraction of naturally occurring small cells represent physiologically induced forms, versus stable, diminutive phenotypes. Smaller cells have a greater surface area to volume ratio, postulated to be adaptive for low-nutrient environments (11). However, small cell size does not necessarily imply adaptation to an oligotrophic (low-nutrient) lifestyle. For instance, new Verrucomicorbiales isolates (8) grow well and maintain small cell size under relatively high nutrient growth conditions (e.g., 4 mM glucose, or 0.1% starch). Nanobacteria dwell (and are cultivated) in a relatively nutrient-rich environment, yet maintain their small cell dimensions (13). Symbiotic and parasitic bacteria are known that have reduced physiological capacities and genome sizes (20). It is possible that symbionts in environments rich with host-supplied growth factors may actually have reduced genetic and physiological demands, thereby facilitating cell size reduction. It is possible that small cell size is adaptive for free-living cells in low nutrient environments, but symbiotic species may tend toward small cell size in a nutrient-replete environment provided by the host.

3. Is there a continuum (or quanta?) of size and complexity that links conventional bacteria and viruses?

Direct examination of concentrated seawater samples by electron microscopy have revealed the presence of large numbers of viral-like particles (VLPs) in the world's oceans (4,5). Ranging in numbers from about 2 x 10⁵ to 5 x 10⁸ particles/ml, VLP numbers typically exceed bacterial cell numbers in aquatic samples by 10-fold. Most quantitative studies to date have employed ultracentrifugation or ultrafiltration coupled with electron microscopy, or filtration, fluorescent DNA staining, and epifluorescence microscopy. A few studies have succeeded in enumerating naturally occurring viable infectious particles (especially in marine Synechococcus sp.) to determine the host range, in situ titers, and ecological variability of naturally occurring cyanophages (21).

In the marine environment there is certainly a continuum of size in both bacterioplankton and virioplankton. Bacterioplankton can range from large filaments > 10 ^mm, to small coccoid cells with diameters approaching 0.3 ^mm (2). Marine virus isolates range in length from about 40 nm, to as large as 120 nm (5). Electron micrographs of naturally occurring infected cells suggest that some bacterial hosts are considerably less than 10-fold larger than their viral parasites, having a burst size of about 7! (6) The very smallest bacterial cells and the very largest viral particles fall into about the same size category, raising some questions about the accuracy of currently used methods for quantifying naturally occurring virus and prokaryotes. Commonly used epifluorescence techniques are convenient and reproducible, but the identity of the fluorescently stained particles is certainly subject to some uncertainty. What fraction of VLPs are actually viruses? What fraction of VLPs are viable viruses? What fraction of DNA-containing particles < 0.1 ^mm are actually cells, and not viruses? If some of the < 0.1 ^mm DNA-containing particles are cells, are they viable? These remain open-ended questions.

With regard to the complexity of these populations, the issue of cultivability is a serious one. It still appears from available data that a large fraction of naturally occurring microbes have resisted cultivation attempts. The specific physiological traits and life histories of these microorganisms remain unknown, as does that of their viral parasites. A major challenge to contemporary microbiology is to devise and implement approaches to better characterize this large and uncharacterized biota.

References

1. Whitman, W.B., Coleman, D.C., Wiebe, W.J. (1998), Proc. Natl. Acad. Sci. USA 95:6578-6583.

2. Watson, S.W., Novitsky, T.J., Quinby, H.L., Valois, F.W. (1977), Appl. Environ. Microbiol. 33:940-946.

3. Fuhrman, J.A. (1981), Mar. Ecol. Prog. Ser. 5:103-106.

4. Bergh, O., Borsheim, K.Y., Bratbak, G., Heidal, M. (1989), Nature (London) 340:467-468.

5. Borsheim, K.Y. (1993), FEMS Microb. Ecol. 102:141-159.

6. Steward, G.F., Smith, D.C., Azam, F. (1996), Mar. Ecol. Prog. Ser. 131:287-300.

7. Suzuki, M.T., Sherr, E.B., Sherr, B.F. (1993), Limnol. Oceanog. 38:1566-1570.

8. Janssen, P.H., Schuhmann, A., Morschel, E., Rainey, F.A. (1997), Appl. Environ. Microbiol. 63:1382-1388.

9. Hedlund, B.P., Gosnik, J.J., Staley, J.T. (1996), Appl. Environ. Microbiol. 46:960-966.

10. Schut, F., DeVries, E.J., Gottschal, J.C., Robertson, B.R., Harder, W., Prins, R.A., Button, D.K. (1993), Appl. Environ. Microbiol. 59:2150-2160.

11. Schut, F., Prins, R.A., Gottschal, J.C. (1997), Aquat. Microb. Ecol. 12:177-202.

12. Eguchi, M., Nishikawa, T., MacDonald, K., Cavicchioli, R., Gottschal, J., Kjelleberg, S. (1996), Appl. Environ. Microbiol. 62:1287-1294.

13. Kajander, E.O., Çiftçioglu, N. (1998), Proc. Natl. Acad. Sci. USA 95:8274-8279.

14. MacDonnell, M.T., Hood, M.A. (1982), Appl. Environ. Microbiol. 43:566-571.

15. Kjelleberg, S., Hermansson, M., Marden, P. (1987), Ann. Rev. Microbiol. 41:25-49.

16. Rozak, D.B., Colwell, R.R. (1987), Microbiol. Rev. 51:365-379.

17. Staley, J.T., Konopka, A. (1985), Ann. Rev. Microbiol. 39:321-346.

18. Amann, R.I., Ludwig, W., Schleifer, K.H. (1995), Microbiol. Rev. 59:143-169.

19. Pace, N.R. (1997), Science 276:734-740.

20. Fraser et al. (1995), Science 270:397-403.

21. Waterbury, J.B., Valois, F.W. (1993), Appl. Environ. Microbiol. 59:3393-3399.

Last update 12/28/00 at 3:53 pm