Download Were Gram-positive rods the first bacteria?

166
Opinion
TRENDS in Microbiology
Vol.11 No.4 April 2003
Were Gram-positive rods the first
bacteria?
Arthur L. Koch
Biology Department, Indiana University, Jordan Hall 142, 1001 East Third Street, Bloomington, IN 47405– 6801, USA
At some point in the evolution of life, the domain Bacteria arose from prokaryotic progenitors. The cell that
gave rise to the first bacterium has been given the
name (among several other names) ‘last universal
ancestor (LUA)’. This cell had an extensive, well-developed suite of biochemical strategies that increased its
ability to grow. The first bacterium is thought to have
acquired a covering, called a sacculus or exoskeleton,
that made it stress-resistant. This protected it from rupturing as a result of turgor pressure stress arising from
the success of its metabolic abilities. So what were the
properties of this cell’s wall? Was it Gram-positive or
Gram-negative? And was it a coccus or a rod?
Bacteria evolved as a separate domain some time before
Archaea or Eukarya [1]. They developed at a critical time
that defined the ‘last universal ancestor (LUA)’, and arose
after the basic groups of physiological processes had
evolved to a functional point. These major processes
must have included DNA replication, ribosome and
protein synthesis, central intermediate metabolism, cell
division, DNA repair and responses to fluctuation in the
environment. Such properties are possessed by almost all
modern cells [2] and each would have improved their
ability to grow compared with the first living cell on this
planet. The differences among the bacterial, archaeal and
eukaryotic cells are slight compared with their similarities
[2]. One important difference between the immediate
precursor of the three cell domains and the first bacterium
was the latter’s development of a functional and stressresistant wall [2,3], which allowed it to support a higher
internal osmotic pressure and remain intact in a low
osmotic pressure environment. The issue here is the
nature of this cell wall. Was the wall thick or thin? Was it a
coccus or a rod-shaped bacillus (Figs 1 and 2)? In the
Kandler and Schleifer designation system [4,5], what
category would it be? That is, what was the chemical
structure of the original bacterium’s cell wall? This
classification system uses three symbols. Thus, for Escherichia coli, it is A1g. The A indicates that the third amino-acid
residue is meso-diaminopimelyl; the 1 indicates that there is
no intervening peptide that bridges between the two
peptides at the tail-to-tail linkage; and the g indicates that
the third amino acid is meso and not LL.
Corresponding author: Arthur L. Koch ([email protected]).
The first cells: Gram-positive or Gram-negative?
Arguments for Gram-positive rod-shaped cells
Arguments from three groups suggest that the first cell to
separate from the monophyletic prokaryotic predecessor of
the bacteria was a Gram-positive, rod-shaped organism.
Seifert and Fox [6] noted that rod-shaped structures
cluster at the base of the bacterial phylogenetic tree. They
compared the morphology of the cells in the various
branches of the Woese’s 16S rRNA tree (see Ref. [1] and
Olsen et al. [7]). Seifert and Fox stated: ‘It [also] seems
likely that the last common ancestor of the domain
Bacteria was rod-shaped’.
Tamames et al. [8] drew conclusions from the analysis of
the dcw (division and cell wall) clusters present in many
bacteria. There are 15 genes in the dcw cluster of
Escherichia coli and around the same number in other
species. Tamames et al. compared the coding order of the
genes (i.e. the sequence of genes on the chromosome) in a
variety of species. They found that there was a pattern,
and that the dcw cluster was more compact and conserved
in bacilli, implying that rod-shaped bacteria came first. In
the bacilli, in contrast to the cocci that they studied, there
was a more constant order on the chromosome. These
bacilli encompassed both Gram-positive and Gramnegative forms.
Gupta et al. [9– 11] analyzed the completed, published
sequences of many genomes, both bacterial and archaeal,
and concluded that Gram-positive bacteria arose first, and
that Gram-negative bacteria arose from Gram-positive
bacteria through a sequence of several other groups.
Gupta’s phylogenetic tree [11] corroborates the standard
16S rRNA tree [12]. However, the Woese group has
presented convincing evidence from the 16S rRNA
Rod-shaped Bacillus
Coccus
Gram-positive
Gram-negative
TRENDS in Microbiology
Fig. 1. Four possibilities for the wall of the first bacterium. These four types represent a majority of organisms. There are other shapes (curved, spiral and tapered)
but these are probably less likely than the initial bacterial form.
http://timi.trends.com 0966-842X/03/$ - see front matter q 2003 Elsevier Science Ltd. All rights reserved. doi:10.1016/S0966-842X(03)00063-5
Opinion
TRENDS in Microbiology
167
arose only when there were plants, animals and fungi to be
parasitized and resisted. As a consequence, their evolution
would have been a later event.
Gram-positive
Older wall
Cytoplasmic membrane
Gram-negative
Penta-muropeptides
Nonamuropeptide
Up
peptide
To other
tessera
Down
peptide
TRENDS in Microbiology
Fig. 2. Wall-side wall growth mechanisms. In Gram-positive organisms, the wall is
formed by the continuous laying down of layers of murein just outside of the cytoplasmic membrane. After they are hydrolysed the layers become highly stretched
and peripheral. This mode of the addition of underlying layers that move to the
outside of the cell wall is called the inside-to-outside mechanism. In Gram-negative organisms, the wall grows by the extrusion of penta-muropeptide through the
cytoplasmic membrane and their incorporation into a functioning wall structure
under tension. This mode is called the nana-muropeptide stress mechanism.
sequences to show that Archaea and Eukarya separated
from a prokaryotic precursor and are not derivatives of the
Bacteria [1,12], as Gupta believes [9]. Although the origin
of Archaea or Eukarya is not directly relevant to the
present discussion, this conflict is pertinent.
Gupta et al. [9– 11] looked for corresponding regions in
the available sequenced genomes and for differences
characteristic to all members of a taxonomic group and
that, additionally, were unique to some, but not other,
taxonomic groups. Such differences of ‘significance’ they
called ‘indels’ (insertions/deletions); consequently, these
indels correlate with a taxonomic group. When these
persisted in all members of a taxonomic group, Gupta
concluded that the group originated from a founder cell
generated from another phylogenetic group that, by
chance, happened to have this particular indel. The
indel was therefore present in all members of the group,
and then was passed on when a founder from this group led
to a newer group. Thus, the indel was common to all
members of a group no matter what other lines of
diversification occurred within it later. The important
conclusion drawn was that major bacterial taxa arose
linearly from each other [9,11] and that cells with one
membrane, such as Gram-positive organisms, called
‘monoderms’, are the precursors of the cells with both a
cytoplasmic membrane and an outer membrane. These are
Gram-negative organisms, such as E. coli. All of these are
called ‘diderms’ by Gupta’s group.
A supporting argument for this order of evolution is that
the Gram-negative cells are structurally better able to
function as pathogens and to resist antibiotics. This would
be in accordance with the idea that the Gram-negative cell
http://timi.trends.com
Vol.11 No.4 April 2003
Arguments for Gram-negative cells
Woese [1] argued that Bacteria arose some time earlier than
the split that led to the separation of Archaea and Eukarya
from the ‘progenote’ or LUA. The derived phylogeny of
Bacteria, which is currently well accepted, does not group
the Gram-positive organisms together and does not group
the rod-shaped cells separately from the cocci.
A major proponent for the idea that Gram-negative cells
arose first is Cavalier-Smith, who has presented extensive
discussions of the origin and evolution of life [13,14]. He
argued first that life started on the outside of a bilayered
lipid vesicle that had been produced abiotically [15]. If life
started on the outside of the lipid membrane, the problem
of transport across a lipid bilayer of hydrophilic material is
avoided but the theory requires that biomolecules of
crucial importance remain attached to the outer surface.
Later, when life had developed adequately, this phospholipid vesicle engulfed the living portion. When fusion was
complete, this formed the first cell that was surrounded by
two bilayers, and was thus a Gram-negative organism.
The space between these layers corresponds to the
periplasmic space.
Cavalier-Smith’s suggestion [15] is a modification of one
proposed by Blobel [16], who originally suggests that the
cell started ‘inside out’. Blobel presumed that life arose on
a solid rock surface. Cavalier-Smith proposes that this
location solves the energy problem because of the presence
of polyphosphates in the rocks, and that only later did
invagination take place, he assumes to convert this ‘obcell’
(obverse cell) to the usual arrangement [15]. Some
descendents of this cell later degenerated to become the
single-membraned, thick-cell-walled Gram-positive cells.
Cell morphology: cocci or rod-shaped
Until the cell developed mechanisms to establish its shape,
the default appears to be that the phospholipid- or lipidbilayer-enclosed cell formed an ever-increasing sphere as
it grew. Thus, unless the cell had already developed a
cytoskeleton or a strong wall (that is, unless it had an exoor endo-skeleton), there was no mechanism to make it
divide to produce new cells or cells of a constant mean size
or particular shape. If the cell was not attached to any
object, it must have grown larger and larger without limit.
This certainly is not a working growth strategy. Attachment to surfaces could have helped such a cell to divide,
although this would have happened irregularly and
formed daughters of irregular shape. Only one alternative
to this theory has been suggested in the literature as a
possible mechanism for division during the time between
the first cell and the LUA [17]. This proposal is that the
lipid constituents of the bilayer were generated within the
cell, and that they became inserted into the inner layer of
the bilayer and this uneven growth caused the bilayer to
invaginate. Eventually, this would lead to cell division.
This strategy might have worked but would have been
irregular and would not have been very effective [17]. It
was possibly up to 800 million years later, when a
168
Opinion
TRENDS in Microbiology
prokaryote completed the evolution of a stretch-resistant
‘fabric’, that an effective alternative arose, which I argue
was the creation of the domain Bacteria. Similarly, the
development of pseudomurein could have led to
the creation of the first Archaea. Of course, later, the
cytoskeleton arose together with contractile proteins used
by Eukarya.
It is not enough in the development of bacteria to
develop a means of forming an enclosing strong murein
sacculus. Several other mechanisms must also have
arisen. A mechanism that prevents wall growth in the
established poles of cells is a key requirement. A
mechanism that causes a pole to be metabolically inert
provides a way to maintain the size and shape of cells in
succeeding generations. For rod-shaped cells, inert poles
provide support for the elongation of cylindrical growth
[18]. However, this can only function to the degree that the
poles are metabolically inert and rigid [19]. Thus,
the inertness of the poles can also be the basis for the
maintenance of the diameter of cylinder-shaped cells in
balanced growth. Cell mechanisms must also function to
foster wall growth in an amount consistent with the ongoing rate of protoplasmic synthesis. That is, cell
biochemical growth must drive the enlargement of the
cell wall. These aspects are, together, probably the answer
to the question: why do bacteria not grow larger and larger
and rounder and rounder?
Experimental evidence for the poles of bacteria started
with Cole’s and Hahn’s studies [20] of Streptococcus
pyogenes, and Doyle’s early unpublished studies of
Bacillus subtilis (published in [21]).These studies found
that the established poles of both this Gram-positive
coccus and rod were metabolically inert. There is now
definitive evidence for the inertness of the poles of
B. subtilis [22– 24] and of E. coli ([25]; A.L. Koch and
M.A. De Pedro, unpublished). If these findings apply to
other bacteria and it is found that the poles of bacteria,
other than mycoplasma, are rigid, metabolically inert and
cannot stretch further, then this rigidity is probably the
defining feature of the domain of bacteria.
The pole metabolisms of B. subtilis and E. coli only have
been studied in sufficient detail to consider the question of
metabolic turnover in the poles. The experimental finding
is that the turnover at the poles is negligible [22,24].
However, the sidewalls of rod-shaped cells turn over with a
half-life equal to their growth doubling time [24]. The inert
nature of the poles is surprising from the biochemical point
of view because attempts to find a significant chemical
difference in the murein wall of the poles and of the
sidewall regions of B. subtilis have failed [26].
Rigid poles allow the cells to maintain their size during
growth because the poles provide a template for wall
enlargement, and this determines the diameters of the
next generation of poles. When initially linked into the
wall of stress-bearing organisms, newly polymerized wall
is not extended to its maximum size. Although the murein
is elastic when inserted, it is unstretched. Of course, it will
come to be stretched during growth. If there were no
special controls on insertion and it occurred randomly over
the cell surface, the cell would bulge like a soap bubble
(as various bacteria do when treated with appropriate
http://timi.trends.com
Vol.11 No.4 April 2003
antibiotics [27]). The physical law, derived by LaPlace two
centuries ago, would lead to cylindrical extension [28]
if the poles were rigid. His law related the radius of a
‘bubble’ of arbitrary shape to the pressure difference
across the wall, the amount of work needed to increase
the surface by a unit amount and constraints upon its
surface (see Ref. [34]).
The implication of bacterial poles being metabolically
inert is that the first bacterium, like all bacteria (with the
exception of mycoplasma), formed new poles entirely by
new synthesis and, subsequently, did not enlarge or alter
them. This implies that the biochemistry and biophysics of
bacterial cells must be such that the mature poles are
blocked from further metabolism or turnover. This
prevents the size of completed poles from being further
modified by insertions or being turned over in future
generations. It therefore appears likely that cocci only
need to be able to form a septum centrally, and grow
without changing their maximum diameter and the
dimensions of a mature pole. They then only needed to
allow or aid the septum to split to form two new daughter
cells of the same diameter.
This is the way that the coccus Enterococcus hirae and
the rod Bacillus subtillis divide [28– 32]. Evidence from
the electron microscope is that, at some critical stage of the
cell cycle, a septum starts to grow inwards (from the site of
the previous septa or in the middle of the cylindrical
region). As it is formed, it starts to split from the outside in
and the intervening split septal wall bulges outwards,
forming two new poles. Consequently, the new pole has the
same diameter as the old one and, in the next generation,
these poles are the templates for the new nascent poles.
This means that the diameter of poles in a culture in
balanced growth is remarkably constant (^ 5%; Ref. [29]).
The growth of cocci is clearly simpler than that of rodshaped organisms, which must form the cylindrical walls
by a separate process in addition to septal formation and
splitting. This would suggest that that the first bacterium
was a coccus [33]. This is not in accordance with the logical
extension of the ideas of Woese [1], or those of Seifert and
Fox [6], Vicente’s group [8] and Gupta’s group [9].
Consequently, other roles for rod-shaped cells and their
advantages will be considered later in this article to
support their points of view.
Wall growth mechanisms for the first bacterium
The strategies for wall growth of Gram-positive and Gramnegative cells briefly presented here have been studied
both theoretically and experimentally. The mechanism for
Gram-positive cells is well-established [34]. Briefly, new
layers of wall are added from the cytoplasmic membrane
surface and autolysis removes the oldest wall. This is
called the Gram-positive ‘inside-to-outside’ mechanism.
By contrast, the mechanism for Gram-negative bacteria
requires that new wall units be synthesized, inserted
through the cytoplasmic membrane and covalently
inserted into the stress-bearing wall. The most recent
proposal depends on stress in the growing wall, altering
the conformation of the new wall. This model is called the
‘nona-muropeptide stretch’ mechanism. It can be
appreciated that both sidewall and pole formation for
Opinion
TRENDS in Microbiology
Gram-positive organisms are quite simple compared with
the process used by the thin-walled Gram-negative cell.
What are the possibilities for how saccular growth
occurred in the first place? Could either the mechanisms
for Gram-positive or Gram-negative cells be those that
functioned for the progenote LUA cell that had just
perfected methods to form a cross-linked ‘fabric’ outside
of its cytoplasmic membrane? Being able to form a polymer
outside the cytoplasmic membrane is very complex [34]
and there must have been many problems to overcome.
However, in addition for bacterial growth, there must be
mechanisms directing the insertion and leading to cell
division. It is here at the LUA stage that an additional set
of mechanisms had to be generated. Although the early
mechanisms may not have been as sophisticated as those
used in modern bacteria, at the start of the bacterial
domain they had to be sufficient, simple and functional.
On the basis of the Gram-positive and Gram-negative
possibilities, it would appear plausible that something like
a Gram-positive ‘inside-to-outside’ mechanism would be
much simpler than the ‘nona-muropeptide stretch’ mechanism for Gram-negative cells. Moreover, it seems selfevident that a coccus instead of a rod-shaped organism is
simpler and should have arisen earlier. So the first hunch
could be that the original bacterium was a Gram-positive
coccus. For division, this cell must have been capable of
forming a septum or crosswall, possibly with a mechanism
similar to that used for crosswall formation in B. subtilis or
Enterococcus hirae. The septal crosswalls of Gram-positive
cells are much thicker than a monolayer of murein and are
around the same thickness as the Gram-positive sidewall
of B. subtilis.
Conclusions
The first bacterial cell appears to have had a sacculus
[2,3,34]. But other thoughts about the first member of
the domain of Bacteria are more varied. It may have been a
Gram-positive organism, that is, if Gram-positivity only
denotes a thick peptidoglycan layer. I argued earlier in this
article that the Gram-positive strategy for growth is much
simpler than that of Gram-negative cells. Cavalier-Smith
argued that a Gram-negative cell appeared first, although
he considered mostly the membranes and not the murein
wall and its role in bacterial growth [14].
It is probably the case that, during the time between the
origin of the first cell and the first bacterium, cells had no
murein layer. A wall was needed after organisms gradually
became more successful because only then did their turgor
pressure increase. The simplest mode for Gram-negative
growth suggested so far is the nona-muropeptide stretch
model, which is much simpler than earlier models [34]. If it
functioned at a time when cells had only a partial
functional sacculus and did not involve enzymes passing
through the cytoplasmic membrane, proteins in holoenzyme clusters would not have been required, unlike in
Höltje’s ‘three-for-one’ model [35]. Höltje’s model postulates that penta-muropeptides are secreted through the
cytoplasmic membrane and are linked by glycan bonds to
form oligopeptides of the same length as a template portion
of the stress-bearing wall. Three of these chains are linked
together by tail-to-tail bonds and the resultant ‘raft’
http://timi.trends.com
Vol.11 No.4 April 2003
169
inserted around the template chain that is then removed
from the wall. This enlarges the wall by doubling the area
of the template strand. At present this model is the most
well known model in the literature. However, with either
model it is difficult to imagine from known biochemistry
that the first cell was a Gram-negative cell, as CavalierSmith would postulate [13– 15].
The coccus is the simplest of possible cell shapes. If
E. hirae were taken as the model for the first bacterium,
such a cell would only need to form a thick septum and
bisect it and then let the physical forces do the rest. The
semiconservative process described above for E. hirae
would not be too difficult to implement simply with a few or
no extracytoplasmic enzymes.
The reasons why a rod-shaped bacterial organism
might have come first have been presented in this Opinion.
This is the conclusion of three groups based on three quite
different arguments. There are biophysical arguments,
based on the kinetics of uptake issues [34], for the
advantages of rod-shaped organisms over coccal-type
cells. There are several different cases for diffusion up to
the cell. In an environment with significant concentrations
of resources, if every unit of surface area has the same
concentration of uptake sites, the only relevant factor is
the ratio of surface to volume. Smaller cells have higher
ratios. For a fixed volume, thin rods or flat leaf-like
structures have higher ratios. By contrast, spheres have
the lowest ratio. Therefore, a narrow rod would have been
an optimal shape.
If different regions of the cell under different environmental conditions have a different number of absorbing
sites, then the actual surface area of the cell is not the only
significant quantity, and the number and kind of uptake
systems per unit of surface area must also be taken into
consideration. However, when nutrients are present in
very low concentration, the factor of prime significance is
the diffusion process from the bulk medium to the cell. Now
the mathematics for different-shaped objects is quite
different. But the effect of cell shape is actually minor in
either low or high substrate concentration. Of course, even
if the effect on growth rate is very slight over many
generations, it might still be important. The general
conclusion from these considerations is that a rod shape is
better than a coccoid shape for a cell of fixed volume.
Are there advantages of rod-shaped growth over coccaltype growth? Two potential advantages can be suggested.
One is that the individual cells in a massed aggregate of
cells might have a more effective exposure to the
environment if the aggregate is an assortment of randomly
oriented rod-shaped cells than if it is a compact mass of
more spherical particles. This is a possibility that must be
mathematically explored.
The second, and more probable, suggestion is that a rod
shape is useful when cell division is controlled by cell
growth. At some point in the E. coli cell cycle, the initiation
of DNA replication takes place and this, in turn, controls
the timing of the subsequent cell division [36]. This
strategy may have been selected to preclude the possibility
that the cell division event would interfere and break a
chromosome. Consequently, the rod-shaped mode of cell
growth is safer for the genetic integrity of the cell.
170
Opinion
TRENDS in Microbiology
Contrast this with the behavior of cocci. In E. hirae, it
appears that splitting of the septum is initiated when wall
growth cannot occur fast enough for the cell’s needs for
space for its cytoplasm [29– 31]. Because of the disparity of
the ratio of volume growth rate to wall surface growth rate
throughout the cell cycle, the trigger might be turgor
pressure. Turgor pressure must increase inside the cell
towards the end of septal closure and it is the pressure that
triggers the wall splitting event. Such a system could err if
the chromosome and the cell division cycle were not in
proper temporal relationship, but would require an
additional regulatory system.
Relevant to this, Seifert and Fox [6] observed that, in
several branches of the 16S rRNA evolutionary tree, the
cells switched from rod-shape to cocci but then did not
revert in further evolution. The first bacterium might have
been a rod, to allow safe growth but much later in the
development of the bacterial clades, genetic deletions
occurred to eliminate the sidewall region. This could have
been dangerous, unless new developments of various kinds
occurred to control the chromosome cycle to be in
accordance with the cell division cycle.
This is an attractive possibility but more sequence data
for more species will be needed to check the mode of cell
growth. However, for now, this is the best guess for the
apparent initial role of rods and subsequent emergence of
cocci. This hypothesis would require early and sophisticated linkages between the morphological aspects of the
cell cycle with the chromosome cycle.
Vol.11 No.4 April 2003
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
References
1 Woese, C.R. (1987) Bacterial evolution. Clin. Microbiol. Rev. 51,
221 – 271
2 Koch, A.L. (1994) Development and diversification of the last universal
ancestor. J. Theor. Biol. 168, 269 – 280
3 Koch, A.L. (1995) Origin of intracellular and intercellular pathogens.
Q. Rev. Biol. 70, 423 – 437
4 Schleifer, K.H. and Kandler, O. (1972) Peptidoglycan types of bacterial
cell walls and their taxonomic implications. Bacteriol. Rev. 36,
407 – 477
5 Tipper, D.J. and Wright, A. (1979) Structure and biosynthesis of
bacterial cell walls. In The Bacteria (Sokatch, J.R. and Ornston, L.N.,
eds) pp. 291 – 426, Academic Press
6 Seifert, J.L. and Fox, G.E. (1998) Phylogenetic mapping of the bacteria
morphology. Microbiology 144, 2803– 2808
7 Olsen, G.J. et al. (1994) The winds of (evolutionary) change: breathing
new life into microbiology. J. Bacteriol. 176, 1 – 6
8 Tamames, J. et al. (2001) Bringing gene order into bacterial shape.
Trends Genet. 17, 124 – 126
9 Gupta, R.S. (2002) Phylogeny of bacteria, are we now close to
understanding it? ASM News 68, 284 – 291
10 Gupta, R.S. and Griffiths, E. (2002) Critical issues in bacterial
phylogeny. Theor. Popul. Biol. 61, 423 – 434
11 Gupta, R.S. (2001) The branching order and phylogenetic placement of
27
28
29
30
31
32
33
34
35
36
species from completed bacterial genomes, based on conserved indels
found in various proteins. Int. Microbiol. 4, 187– 202
Olsen, G.J. (2001) The history of life. Nat. Genet. 28, 197– 198
Cavalier-Smith, T. (2001) Obcells as proto-organisms: membrane
heredity, lithophosphorylation, and the origins of the genetic code, the
first cells, and photosynthesis. J. Mol. Evol. 53, 555 – 595
Cavalier-Smith, T. (2002) The neomuran origin of Archaebacteria, the
negibacterial root of the universal tree and the bacterial megaclassification. Inter. J. System. Evol. Microbiol. 52, 7 – 76
Cavalier-Smith, T. (1987) The origin of cells: a symbiosis between
genes, catalyst, and membranes. Cold Spring. Harbor Symp. Quant.
Biol. 52, 805 – 824
Blobel, G. (1980) Intracellular membrane topogenesis. Proc. Natl.
Acad. Sci. U. S. A. 77, 1496 – 1500
Koch, A.L. (1985) Primeval cells: possible energy-generating and celldivision mechanisms. J. Mol. Evol. 21, 270– 277
Koch, A.L. (1982) On the growth and form of Escherichia coli. J. Gen.
Microbiol. 128, 2527 – 2540
Koch, A.L. (2002) Why are rod-shaped bacteria rod-shaped? Trends
Microbiol. 10, 452– 455
Cole, R.M. and Hahn, J.J. (1962) Cell wall replication in Streptococcus
pyogenes. Science 135, 722
Koch, A.L. et al. (1981) The coupling of wall growth and chromosome
replication in Gram-positive rods. FEMS Microbiol. Lett. 12, 201 – 208
Merad, T. et al. (1989) Cell wall assembly in Bacillus subtilis:
visualisation of old and new material by electron microscopic
examination of samples selectively stained for teichoic acid and
teichuronic acid. J. Gen. Microbiol. 135, 645– 655
Clarke-Sturman, A.J. et al. (1989) Cell wall assembly in Bacillus
subtilis: partial conservation of polar wall material and the effect of
growth conditions on the pattern of incorporation of new material at
the polar caps. J. Gen. Microbiol. 135, 657 – 665
Koch, A.L. and Doyle, R.J. (1985) Inside-to-outside growth and the
turnover of the Gram-positive rod. J. Theor. Biol. 117, 137– 157
De Pedro, M.A. et al. (1997) Murein segregation in Escherichia coli.
J. Bacteriol. 179, 228– 234
Buchanan, C. (1979) Altered membrane proteins in a mini-cell mutant
of Bacillus subtilis. J. Bacteriol. 139, 305– 307
Schwarz, U. et al. (1969) Autolytic enzymes and cell division of
Escherichia coli. J. Mol. Biol. 41, 419 – 429
Koch, A.L. et al. (1981) Surface tension-like forces determine bacterial
shapes: Streptococcus faecium. J. Gen. Microbiol. 123, 151 – 161
Higgins, M.L. and Shockman, G.D. (1970) Model for cell wall growth of
Streptococcus faecalis. J. Bacteriol. 101, 643– 648
Koch, A.L. and Burdett, I.D.J. (1986) Biophysics of pole formation of
Gram-positive rods. J. Gen. Microbiol. 132, 3451– 3457
Koch, A.L. et al. (1982) The role of surface stress in the morphology of
microbes. J. Gen. Microbiol. 128, 927 – 945
Koch, A.L. and Higgins, M.L. (1984) Control of wall band splitting
in Streptococcus faecalis ATCC 9790. J. Gen. Microbiol. 130,
735 – 745
Koch, A.L. (1996) What size should a bacterium be? A question of scale.
Annu. Rev. Microbiol. 50, 317 – 334
Koch, A.L. (2001) Bacterial growth and form, 2nd edn, Kluwer
Höltje, J-V. (1993) ‘Three for One’: A simple growth mechanism that
guarantees a precise copy of the thin, rod-shaped sacculus of E. coli. In
Bacterial Growth and Lysis, Metabolism and Structure of the Bacterial
Sacculus (de Pedro, M.A., Höltje, J-V. and Löffëlhardt, W., eds)
pp. 419– 426, Plenum Press
Cooper, S. (1991) Bacterial Growth and Division, Academic Press
News & Features on BioMedNet
Start your day with BioMedNet’s own daily science news, features, research update articles and special reports. Every two weeks, enjoy
BioMedNet Magazine, which contains free articles from Trends, Current Opinion, Cell and Current Biology. Plus, subscribe to
Conference Reporter to get daily reports direct from major life science meetings.
http://news.bmn.com
http://timi.trends.com