Section C: Developmental Changes and Cellular Transformations


Both amoebae and plasmodia can develop or reversibly transform
into other cell types (Dove et al. 1986; Burland et al., 1993a).
So far, the changes that have been most thoroughly analysed
biochemically are those exhibited by plasmodia, since these can
be induced to occur in large cell masses with good synchrony.
Spherulation is a reversible transformation of plasmodia, in
response to conditions of nutrient deprivation. It involves the
formation of resistant, cyst-like spherules or sclerotia, which
can survive drying and other extreme conditions. The spherules
are multinucleate and they yield plasmodia when placed in moist,
nutrient conditions.  Sporulation is a developmental phase,
involving complex changes in morphology, cytoplasmic cleavage,
melanisation and meiosis, leading to the formation of spores
containing haploid uninucleate amoebae.  Sporulation can be
induced to occur by exposing plasmodia to light in conditions of
nutrient deprivation. Precise conditions for inducing sporulation
with good synchrony have been described (Raub & Aldrich, 1982;
Starostzik & Marwan, 1995). Studies of spherulation and
sporulation have resulted in descriptions of many of the
processes involved and have shown that changes in gene expression
occur when a plasmodium ceases vegetative growth and enters
either of these transitional phases. Since plasmodia are
multinucleate, it has been difficult to screen for mutants in
these processes by classical methods. Some of the genes involved
are now also being identified by molecular methods.
 
In contrast, the transformations of amoebae have been less
studied biochemically but have been easier to approach
genetically. Amoebae can change into three alternative cell-
types:  cysts, flagellates, and plasmodia (Anderson et al. 1986).
The transformations of amoebae into cysts or flagellates are
reversible and can apparently be induced repeatedly in the same
cells. 

Encystment
		                             
Amoebae cultured on bacterial lawns form walled cysts when the
food supply is exhausted and in other conditions unfavourable for
growth, such as refrigeration. Contrary to previous reports, a G1
stage is not involved during this process (Anderson et al., 1997) 
Cysts are valuable for the storage of amoebal strains, although
it is not clear whether encystment is essential for amoebae to
survive freezing in glycerol.  Amoebae cultured in liquid axenic
medium may be induced to form rounded cells by the addition of
mannitol, but these cells are not walled cysts and they revert
rapidly to active cells when growth conditions are restored (Dee
et al., 1997). Encystment can be induced in axenic conditions but
it apparently requires a solid substratum as well as nutrient
deprivation. Strains differ in their ability to encyst and this
property can be lost during log-term culture of amoebae (see
Section B). There is little doubt that encystment requires
specific gene expression though the genes have not yet been
identified by either classical or molecular methods. Excystment
can be readily induced in nutrient conditions or suspension in
water or buffer (Gorman et al., 1977; Dee et al, 1997; see
Section B).

The amoeba-flagellate transformation

Amoebae cultured on bacterial lawns can be induced to transform
into flagellates with good synchrony when the plates are flooded
with water or various buffers (Pagh & Adelman, 1982). 
Flagellates can neither feed nor divide, but they can transform
into amoebae again in nutrient conditions, particularly when
allowed to attach to a surface.  The amoebae then resume feeding
and proliferating. Although the changes are quite easily
controllable in most amoebal strains, the precise signals
triggering either transformation are not understood. Suspension
in liquid is not enough to cause amoebae to transform into
flagellates, for example; there is evidence that ionic strength
is an important factor.

Studies of changes in gene expression during the amoeba-
flagellate transformation have been carried out on cells cultured
on bacteria or in axenic medium (Green & Dove, 1984; Paul et al,
1992). The reverse transformation has been less well studied but
there has been at least one detailed microscopic examination of
this (Glyn & Gull, 1990). It should be noted that both
transformations occur extremely rapidly, so that great care is
necessary to ensure that the cells have not begun to change
without the investigator's knowledge. For example, amoebae washed
off plates may start to form flagellates while a supposedly
'amoebal' population is being harvested prior to biochemical
analysis. Conversely, a suspension of 'flagellates' may be
rapidly transforming to amoebae if allowed to stand for some
minutes.

Mutant strains of amoebae unable to transform into flagellates
apparently occur at quite high frequencies and are not difficult
to isolate (Mir et al. 1979);  few have yet been analysed
genetically or biochemically, however.  Some strains of amoebae
able to grow in axenic liquid medium (e.g. CLd-AXE) were also
found to be unable to form flagellates, but further
investigations have shown that this defect is not necessarily
associated with the ability to grow in liquid. Following the
isolation of new Axe strains from crosses (see Section A), a
method was worked out for inducing some of these strains to form
flagellates with high efficiency in axenic conditions (Blindt,
1987;  Dee et al, 1989).  This method has facilitated biochemical
studies of the amoeba-flagellate transformation, but the
following precautions should be observed in using Axe strains for
such investigations.

First, to induce flagellate formation in axenic conditions,
amoebae must be transferred very gradually from the growth medium
into a medium of low ionic strength, as described in the
published method;  rapid transfer will result in death,
apparently due to osmotic shock. The same strain of amoebae may
be found to form flagellates quite successfully if suddenly
transferred to water from a bacterial lawn, but in these growth
conditions, the cells are apparently in a more robust state.
Second, the published method for inducing flagellate formation in
axenic conditions was worked out for a particular Axe strain
(LU352) and was found to be unsuccessful when applied to some
other Axe strains (e.g. LU353). The latter strains appear to be
even more sensitive to osmotic shock, and their failure to form
flagellates is perhaps due to this sensitivity. When suspended in
water from cultures on bacterial lawns, strains such as LU353
show almost normal capacity to transform into flagellates (Dee et
al. 1989).

Third, if a strain such as LU352 is maintained in liquid, axenic
culture for long periods, the amoebae apparently lose the ability
to transform into flagellates (Glyn, 1989).  This change may be
due to inadvertent selection for mutant cells unable to transform
into the non-feeding flagellate stage; such mutants may be at a
selective advantage during culture in a liquid medium, perhaps
because flagellates do not feed.  Alternatively, the change may
be due to a modification of the amoebae which is inherited during
vegetative propagation. In any case, it will be advisable to
return to stored stocks of the Axe amoebal strain at intervals if
ability to undergo the transformation efficiently is to be
maintained.
		                             
The amoebal-plasmodial transition

In natural isolates of P. polycephalum, plasmodium development
involves mating of amoebae to form zygotes which then develop
into multinucleate, diploid plasmodia. For studies of changes in
gene expression or cellular organization during development,
however, apogamic strains have usually been preferred because
amoebae form haploid plasmodia without changing in DNA content
(Bailey et al, 1992; Solnica-Krezel et al, 1990, 1991). Although
they are a sensible choice for this purpose, it should not be
forgotten that apogamic strains arose in laboratories as the
result of mutations. Some of their characteristics may therefore
be indirect consequences of the mutations and may have no
significance in relation to wild-type strains. For example,
apogamic development is temperature-sensitive but sexual
development is not. Where possible, it is desirable that wild-
type, sexual development should be studied also.  It must also be
remembered that apogamic (e.g. npfC+) amoebae are liable to lose
their ability to develop if cultured in conditions permissive for
development (see Section B). This loss can occur extremely
rapidly, within a few subcultures, and the characteristics of the
strain must therefore be checked frequently, for example, by
plating out a sample of the amoebae to give single colonies and
checking that plasmodium formation occurs in all colonies after
the expected number of days.

Amoebae blocked in apogamic development, such as npfC mutants of
npfC+ strains, may be useful for comparisons of gene expression
in amoebae and plasmodia, since they are more easily maintained
in vegetative growth, as explained in Section B. Although
plasmodia of genotype npfC cannot be obtained, revertant npfC+
plasmodia can be used for the comparison since they may be
assumed to differ from the amoebae at this single locus only. For
studies of changes occurring during the developmental transition,
however, npfC+ or similar strains, capable of efficient apogamic
development, must be used.

The changes associated with plasmodium development in individual
cells have been studied by analysis of time-lapse films of
cultures maintained in cavity slides on lawns of bacteria (see
section B). Both apogamic development and sexual development have
been studied in this way (Bailey et al, 1987, 1990). The lesions
in some mutants blocked in apogamic development have also been
analysed (Bailey et al, 1992, Solnica-Krezel et al, 1995). Other
important approaches have involved assays of cell populations
developing from amoebae to plasmodia with partial synchrony.
Several different methods may be used to estimate the frequencies
in a single population of (a) cells committed to plasmodium
development, (b) cells able to transform into flagellates, and
(c) cells that have become binucleate or multinucleate. Cells
committed to plasmodium development can be recognized by plating
out cell samples at low densities on bacterial lawns; the
committed cells grow to give individual, multinucleate plasmodia,
while the cells that have not become committed multiply to give
colonies of amoebae (Youngman et al. 1977).  By combining the
results of these different assays with film analyses, it has been
deduced that apogamic amoebae cultured at permissive temperature
proliferate until they reach a critical cell density. The amoebae
then cease to divide and enter an extended cell cycle, more than
twice the length of the amoebal cell cycle, during which they
continue to grow and finally undergo mitosis without cytokinesis
to give binucleate cells (Bailey et al. 1987). About halfway
through this extended cell cycle, an amoeba loses the ability to
form a flagellate, and at about the same time, it becomes
committed to plasmodium development (Blindt et al. 1986;  Bailey
et al. 1987). Thus, in an asynchronous population of cells, at
early stages of development, soon after the critical cell density
is reached, the majority of committed cells are uninucleate. At
later stages, however, many committed cells have two or more
nuclei.

So far, there have been only a few attempts to follow biochemical
changes during the amoebal-plasmodial transition, and these have
involved asynchronous populations of apogamic amoebae developing
on bacterial lawns (Solnica-Krezel et al., 1990, 1991; Bailey,
1997). If the timing of changes in relation to the extended cell
cycle is to be established, it is obviously important for each
cell sample to be assayed for the proportion of committed cells
and the proportions of cells with different numbers of nuclei.
Even where this has been done, however, there is still much
uncertainty about the timing of a change in gene expression,
because of the lack of synchrony and the resulting contamination
of each cell sample with cells at other stages of development.
Ideally, we should like to be able to control development and
initiate it at a specified time. There have been several
different approaches to this problem, none of them as yet
successful.

The simplest approach to controlling the onset of development has
been to inoculate a lawn of bacteria with an evenly-spread lawn
of amoebae, at a specific cell density;  the denser the amoebae,
the earlier the onset of the transition. High frequencies of
committed cells have been achieved by this method (Solnica-Krezel
et al, 1988) but a proportion of cells with two or more nuclei
will usually be present among them.  To avoid including cells
with more than one nucleus, and thus to restrict analysis to
cells still passing through the extended cell cycle, populations
of cells at early stages of development, containing low
frequencies of committed cells, have been enriched for the
presence of cells unable to transform into flagellates (Blindt et
al. 1986). The enrichment process involves passing cells through
a glass bead column. Very high densities of committed cells have
been obtained by this method, but there are problems in obtaining
large enough samples for some types of analysis, and the method
also has the disadvantage that several hours must elapse between
harvesting the developing culture and analysing the committed
cells.  n non-axenic strains, such as CL, developing cells
acquire the ability to grow in liquid, axenic medium (SDM) at
about the time of commitment, whereas cells at all earlier stages
of development fail to survive in SDM. Thus it is possible to
selectively culture and study later stages of development by
inoculating cells into liquid SDM from asynchronous populations.

The reason for the asynchrony of development in bacterial
cultures is that the amoebae distribute themselves unevenly as
they graze on the bacterial lawn. Since development depends upon
a critical cell density, it is initiated at different times in
different areas of the culture. Apogamic strains able to grow and
develop in axenic medium were constructed in the hope that more
homogeneous conditions, and consequently more synchronous
development, could be achieved. So far, however, these strains
have not been induced to develop with good efficiency in axenic
conditions, even though they do show efficient apogamic
development on bacterial lawns  (Dee et al. 1989). 

Another approach to controlling plasmodium development has been
an attempt to identify the diffusible inducer responsible for the
density-dependent onset of the amoebal-plasmodial transition.
Although partial purification of the inducer has been achieved
(Nader et al. 1984) however, preparation of quantities sufficient
to induce development reliably has not been described. It should
also be noted that the inducer has been extracted from cultures
of Didymium amoebae, not from Physarum.  Since cultures of
amoebae always produce inducer during growth, it may also be
necessary to isolate an 'inducerless mutant' before development
can be controlled cleanly by external application of the
substance.

Biochemical studies of sexual development are harder to interpret
since this involves changes in gene content as well as
expression. The formation of cells committed to sexual
development apparently coincides with cell fusion (Shipley &
Holt, 1982), and this event can be precisely controlled by mixing
two strains of amoebae at a specific time. If strains differing
in matA, matB and matC are used, mating occurs with high
frequency. The frequency of committed cells in a mating
population remains lower than in an apogamic strain, however,
since it depends on the meeting of compatible amoebae.  Filming
analysis has shown that following mating, sexual development,
like apogamic development, involves an extended cell cycle,
preceding the formation of a binucleate cell which develops into
a plasmodium (Bailey et al. 1990).


Back to PhysarumPlus HomePage


Last modified: Thursday, December 10, 1998