Journal of Experimental Botany, Vol. 51, No. 346, pp. 831–838, May 2000
REVIEW ARTICLE
The generation and modification of cell polarity
David J. Cove1
Leeds Institute of Plant Biotechnology and Agriculture, University of Leeds, Leeds LS2 9JT, UK
Received 4 February 2000; Accepted 24 February 2000
Abstract
The generation of polarity in the cells of eukaryotes in
most cases requires an external asymmetrical input.
This signal may originate from a random event such
as the point of penetration of an egg by a sperm or
from asymmetries present in the maternal environment
as, for example, in megasporogenesis in flowering
plants, or from an asymmetry in the physical environment usually involving either light or gravity. The latter
affords the best opportunities for experimental manipulation and single cell systems in algae, mosses and
ferns allow both cell biological and genetical investigation of the mechanisms establishing polarity. These
simple systems have already shown that the generation of polarity involves two distinguishable processes,
axis alignment and axis orientation, but a detailed
understanding of the mechanisms by which the environmental inputs are transduced to set up a polar axis
is still lacking. Non-motile organisms must also be able
to modify cell polarity in response to changed environmental inputs, and here too, lower plant systems afford
the best prospect for understanding the mechanisms
responsible.
Key words: Cell polarity, polar axis generation, polar axis
modification, gravitropism, phototropism.
Defining polarity in biological terms
Polarity is central to the development of animals and
plants both in their evolution and for the production of
individual complexity. This paper will focus particularly
on the generation and modification of polarity at the
level of individual cells. Polarity at this level is at the
core of the development of complex multicellular organisms, since the division of a polar cell generates nonequivalent daughter cells, allowing subsequent differential
development.
1 Fax: +44 113 233 3094/ E-mail: [email protected]
© Oxford University Press 2000
The term ‘polarity’ is often used loosely and needs
more rigorous definition. Recent studies of polarity which
will be reviewed here have led to the need for further
clarification of terminology related to polar axes. I myself
have previously used the terms ‘axis polarity’, ‘axis alignment’ and ‘axis orientation’ inconsistently, and dictionary
definitions suggest that the terms, ‘polarity’, ‘alignment’
and ‘orientation’ may be interchangeable in common
usage. However, here these three terms will be defined
differentially to allow more critical analysis of systems in
which the generation of polarity may be studied. Polarity
has recently been defined in a biological context, as
‘persistent asymmetrical and ordered distribution of structures along an axis’ (Cove et al., 1999) and this definition
is not at variance with its general usage and applies
equally to polarity at the level of the organelle, cell, organ
or whole organism. Thus, most flowering plants have, at
the level of the whole organism, a single major polar axis
with roots below the soil and shoots and leaves above
while a zygomorphic flower has two. I now wish to define
more precisely, two further properties of a polar axis,
axis alignment and axis orientation. Axis alignment does
not specify the direction of polarity, only the relationship
between the axis and some external reference. The direction in which the ‘asymmetrical and ordered distribution
of structures’ occurs will be defined as the orientation of
the axis. Thus two polar objects may both be oriented to
an external gradient, for example, gravity, but one may
be more closely aligned than the other. On the other hand
the two may be equally aligned, but oriented in opposite
directions (see the legend to Fig. 1 for the use of these
terms).
Is a polar input needed to achieve cell polarity?
Although in many systems, the setting up of a polar axis
requires an asymmetrical input, some polar biological
structures develop autonomously. Asymmetrical virus
832 Cove
Fig. 1. Response of protonemal filaments of wild-type and mutant
strains of Ceratodon purpureus to gravity. Cultures have been grown on
Petri dishes, oriented with the agar surface upright. The downward
direction is towards the bottom of the page. Scale bar=5 mm. (a)
Cultures of wild type (wt) and a leaky gravitropic mutant (gtr). The
response of the mutant to gravity is slower than that of the wild type,
but shows the same orientation, i.e. growth is upwards. Mutant filaments
finally show the same degree of alignment to the gravity vector as the
wild type, but this level of alignment is achieved only after a longer
period of growth. (b) Cultures of wild type (wt) and a wrong way
response mutant (wwr). The response of the mutant is in the opposite
orientation to that of the wild-type, i.e. the mutant grows downwards.
Its filaments however show the same level of alignment to gravity as
the wild type.
capsids such as those of the T even bacteriophages, have
a highly polar structure that is achieved by assembly of
the capsid components without a requirement for an
external polar input (Mosig and Eiserling, 1988).
Organelles such as ribosomes may also achieve polarity
without an external input.
There is evidence that, in some systems, a polar axis
may be generated in a cell without an external polar
input. It has been proposed that the differentiation of
one daughter cell into a pre-spore, and the other into a
spore mother cell, at the onset of sporulation in the sporeforming bacterium, Bacillus subtilis, is achieved independently of external inputs ( Errington, 1992; Jacobs and
Shapiro, 1998). There are also examples where eukaryote
cells appear to be able to develop a polar axis in the
absence of any external input. The polarity of the microgametophyte (pollen grain) of flowering plants is related
to the proximal-distal axis of the microspore from which
it develops ( Twell et al., 1998). Although this relationship
may vary between species, for a particular species it is
constant, indicating that the polarity of the pollen grain
owes its origin to the topology of microspore tetrad
formation.
At the 32-cell stage of asexual development of the green
alga, Volvox carteri, 16 cells undergo the first of three
crucial asymmetrical cell divisions that result in each
producing one large, asexual propagule, the gonidium,
and numerous small cells that differentiate into somatic
cells ( Kirk, 1997). The 32-cell embryo is already a polar
structure and the 16 cells that undergo an asymmetrical
division are at the anterior of the embryo. The generation
of the embryo’s anterio-posterior axis does not appear to
require an external input. The alignment and orientation
of the polar division is stereotypical and related to the
topology of the embryo. The polar axis is such that the
gonidia-forming cells are produced to the outside of the
embryo. Subsequently in development, inversion occurs
so that the gonidia come to lie in the interior of the
mature plant. Numerous mutants affecting this pattern
of development, and the related patterns of sexual development, have been isolated and the establishment of a
transposon-tagging system now allows their molecular
basis to be investigated (Miller et al., 1993). Thus the
glsA gene, loss of function of which leads to the normallyasymmetrical division becoming symmetrical, has now
been isolated and sequenced (Miller and Kirk, 1999).
Although it can be argued that this gene is involved in
polar axis interpretation rather than polar axis generation
(see below), it is interesting to find that the GlsA protein
product is found to be associated with the mitotic spindle
and Miller and Kirk propose that it may play a role in
altering the plane of cell division.
Isolated protoplasts of the moss Ceratodon purpureus
regenerate into cell filaments by polar outgrowth.
Although the initial alignment of the regeneration axis
can be polarized by a directional light input (Cove et al.,
1996), protoplasts of this species are able to regenerate
in darkness. In these conditions, there is only a very weak
alignment to the gravity vector, suggesting that some
internal cue is sufficient to allow polar development.
Spores of the fern, Ceratopteris richardii, normally align
their germination axis to gravity, but recently have been
successfully germinated in microgravity in a space craft
in orbit around the earth (S Roux, unpublished data),
again indicating that an external cue may not be essential
for the acquisition of polarity.
External signals are usually required to establish a
polar axis
The development of a polar axis often requires an input
provided by a biological signal or by the physical environment, particularly a light gradient or gravity stimulus.
In many animal systems, including the nematode,
Caenorhabditis elegans, and amphibia such as Rana, the
point of sperm entry into the egg provides the stimulus
that breaks symmetry. The point of sperm entry is also
the default stimulus for setting up the polar axis in the
developing zygote of the fucoid alga, Pelvetia compressa
( Kropf et al., 1999), although, if the zygote develops in
a light gradient, the axis may be re-specified with a
different alignment. It appears that the sperm entry point
is in many systems random, so a chance biological signal
is responsible for the initiation of the programme leading
to the establishment of the polar axis.
The determination of the alignment of the dorso-ventral
Generation of cell polarity 833
axis in the embryo of the fruit fly, Drosophila melanogaster, is another example where it appears that a chance
biological event generates polarity. Early in maturation,
the oocyte nucleus migrates from its original median
position to an anterior position. The asymmetry leading
to the establishment of the dorso-ventral axis is achieved
by the chance position occupied by the oocyte nucleus
after migration. Transcription in the oocyte nucleus of
the gurken gene, leads to the localized production of a
secreted molecule, which signals to the maternal cells
nearest to it, setting in train the mechanisms that specify
the dorsal side of the embryo ( Klinger and Tautz, 1999).
The biological input may, however, often be more
difficult to pinpoint, the origins of asymmetry remaining
obscure. Even for the most studied animal system,
Drosophila melanogaster, detail of how the anterioposterior axis of the embryo is generated is still lacking.
The establishment of the axis involves the setting up of
an asymmetrical distribution of mRNAs in the oocyte
( Klinger and Tautz, 1999). These mRNAs are transcribed
in the nuclei of maternal tissue surrounding the oocyte
and are then transported into the oocyte where they come
to be localized. The product of the nannos gene, for
example, is localized at the end of the oocyte destined to
become the animal’s posterior. Several lines of evidence
indicate that the required polar transport is dependent
on pre-existing polarized microtubules. If this is the case,
the mechanism responsible for establishing the polarity
of the cytoskeleton needs in turn, to be explained. It is
probable that ultimately the programme for setting up
the anterio-posterior axis owes its origin to the maternal
anatomy. This example illustrates a problem facing developmental geneticists. The characterization of nannos mutants established that the nannos gene product was
necessary for setting up the anterio-posterior axis. Similar
work, also showed that products of other genes, such as
bicoid, were also required. However, none of these gene
products can be said to be involved in the origin of the
anterio-posterior polarity, but are instead required for
axis elaboration. In many systems, the origin may prove
impossible to describe with precision since it may be that
vital asymmetry is simply passed down the maternal line
from mother to daughter. Although less well understood
than Drosophila, the origins of polarity in the megaspore
of flowering plants may be similar, although the occurrence of somatic embryogenesis indicates that plants may
have more capacity for self organization. The review for
the June issue (Souter and Lindsey, 2000) deals specifically with polarity in flowering plant embryogenesis and
so this topic will not be reviewed here.
The mechanisms leading to the acquisition of a consistently-oriented third polar axis (in addition to anterioposterior and dorso-ventral axes) pose a greater conceptual problem. Such third axis polarity, which leads to
chirality, is found in plants. Although the direction of
phyllotaxis appears to be random in many plants, there
is some evidence for a genotypic component of consistent phyllotactic bias in some species of Tifolieae
(Rijven,1968). In species with twining stems, the chirality
of the twine is in most cases consistent (Hart, 1990). It
seems unlikely that consistent chirality can be achieved
by external inputs and so some intrinsic property of the
biological system may be responsible. The characterization of genes that mutate to reverse chirality would be
revealing. Similar genes have already been characterized
in mammals, including man, where mutant phenotypes
showing a reversal of the normal arrangement of organs
along the left-right axis are known. Such an arrangement,
in which for example, the heart is on the right rather
than the left, is called situs inversus. Two mouse mutations
that lead to a situs inversus phenotype have been characterized ( Yokoyama et al., 1993; Supp et al., 1998). All
inv/inv homozygous mice have a situs inversus phenotype,
whereas 50% of iv/iv mice are situs inversus and 50% are
normal. In both cases alignment along the left-right axis
is nearly normal, but in inv/inv individuals, the orientation
of the left-right polar axis has been reversed, while it
appears that homozygosity for the iv mutation leads to an
inability to orient the axis. The sequence of the iv gene
suggests that it encodes a dynein heavy chain, while that
of the inv gene shows homology with ankyrins
(Mochizuko et al., 1998). Thus in this case, evidence is
again emerging for a role of the cytoskeleton in determining the orientation of a polar axis, and this suggests that
similar studies of chirality in plants will be worthwhile.
The different developmental demands of polarity
in relation to the external environment
Although polarity is universal in multicellular organisms,
allowing differential cell function and organ formation,
the requirements for the generation of polarity will depend
on whether the developing organism is motile or nonmotile. A motile organism can adjust its orientation to
the physical world. In motile organisms, a polar axis is
of adaptive value, with organs of sensory perception and
analysis concentrated in the part of the body that encounters the environment first, conventionally its anterior. For
motile terrestrial animals, locomotory organs are ventral
where they can interact with the ground. However, there
is no need for a motile organism to develop oriented to
the gravity vector, since it can adjust its orientation
subsequently. Thus, although the alignment of the dorsoventral axis in Drosophila appears to rely on chance, it is
only necessary to set this axis up at right angles to the
anterio-posterior axis; alignment within the perpendicular
plane being unimportant.
Plants being non-motile, have a more rigorous need
for environmental inputs, but in many cases the need is
not for a polar axis to be aligned and oriented to the
834 Cove
physical environment at the time of its establishment, but
to be amenable to modification in response to environmental inputs. Thus the embryo within a seed will not
require its polar axis to be aligned to light or gravity at
the time of its establishment, but it is essential that it
should be responsive to these inputs during germination.
Only where no further movement will occur following
polar development, is there likely to be a requirement for
direct monitoring of the external environment during
polar axis generation, and even here the axis may be
initially established without environmental inputs and
alignment modified subsequently in response to light
or gravity.
Environmental inputs and the generation of cell
polarity
The relationship between experimental material and physical environment is usually easy to manipulate and so
experimental studies of the effects of light and gravity on
the generation of cell polarity have been particularly
rewarding. The relative direction of either a light source or
the gravity vector can be changed, and light quality can be
manipulated both in intensity and wavelength. With access
to orbiting spacecraft, gravitational forces less than that on
earth have become available. A number of different singlecelled plant systems are amenable to study.
Zygotes of brown algae, Fucus distichus and Pelvetia
compressa, have been particularly amenable to cell biological study (Fowler and Quatrano, 1995). Fucus eggs are
about 80 mm in diameter, Pelvetia eggs somewhat larger.
Fertilization activates zygote development leading to the
polar outgrowth of a rhizoid. The first cell division generates
a cell plate that is perpendicular to the long axis of the
developing zygote, separating a rhizoid cell which divides
to generate the holdfast, from the cell which gives rise to
the thallus. As outlined above, the point of sperm entry is
a sufficient cue to set up the polar axis ( Kropf et al., 1999).
The alignment of polar development can, however, be
changed if zygotes, once they have settled on the substratum, are exposed to a directional light gradient. The
rhizoid will now emerge on the shaded side of the zygote.
Re-orientation of zygotes with respect to a light gradient at
varying times, shows that the alignment of the polar axis
remains plastic until axis fixation occurs, in Fucus at about
12 h post-fertilization. Once the axis becomes fixed, further
changes in light direction do not affect axis alignment
(Quatrano and Shaw, 1997). There is evidence that the cell
wall plays an integral role in axis fixation ( Kropf et al.,
1988) and it is proposed that fixation occurs as a result of
cross linking between the actin cytoskeleton and the cell
wall (Quatrano and Shaw, 1997). The earliest observable
evidence of ordered asymmetry in the polar development
of the zygote is the localization of F-actin at the future site
of rhizoid outgrowth. In Pelvetia, localization resulting
from sperm entry is observed within 3 h of fertilization, but
is revised following exposure to a light gradient, and reversal
of this gradient before axis fixation results in the repositioning of the localized F-actin in less than an hour (Alessa
and Kropf, 1999). The mechanisms bringing about the
formation of the F-actin patch remain unknown. It has
been suggested that sperm proteins, introduced upon fertilization, may be the instigators of the recruitment of F-actin
( Kropf et al., 1999), but even if this is the mechanism by
which sperm entry cues polar axis alignment, the response
to a light stimulus must involve intrinsic mechanisms.
Details of the mode of transduction of a light stimulus
are emerging. Jaffe established that polarization of the
germination axis was induced by blue light, but not by
red, and that the photoreceptors concerned were located
in or near to the cell membrane (Jaffe, 1958). Recently,
all trans retinal has been isolated from zygotes of Pelvetia
fastigiata, and it has therefore been proposed that the
photoreceptor is a rhodopsin-like protein. (Robinson
et al., 1998). Rhodopsin is associated with the control of
cyclic GMP (cGMP) levels in other systems, and so a
role for a rhodopsin-like protein as the photoreceptor for
axis generation is supported by the finding (Robinson
and Miller, 1997) that the establishment of a cGMP
gradient may be one of the earliest events in the generation
of polarity. Blue light irradiation leads to increases in
cGMP levels, inhibitors of cGMP formation block the
development of polarity, bathing in an excess of a cGMP
analogue reduces polarization, and allowing zygotes to
germinate in darkness in the presence of a gradient of a
cGMP analogue leads to a tendency for rhizoid emergence
to the low side of the concentration gradient. It is likely
that the localization of Ca2+ channels that leads to a
Ca2+ influx in the region of future rhizoid outgrowth, is
an early event in axis determination (Shaw and Quatrano,
1996). In this work, evidence for the localization of Ca2+
channels was provided by the use of fluorescently-labelled
dihydropyridine (DHP), which is known to bind to
calcium channels in mammalian cells. The same work,
showed that the localization of DHP receptors was disrupted by cytochalasin B, a drug that interferes with actin
function, and the establishment of a calcium gradient is
therefore likely to be downstream of F-actin localization.
The fern, Ceratopteris richardii, provides another system
for the experimental investigation of the way in which the
physical environment determines the polarity of an individual isolated cell. Ceratopteris spores, like fucoid zygotes,
divide unequally to produce a small rhizoid cell and a larger
cell that develops into the thallus. The alignment of the
germination axis can be cued by gravity, but spores only
become sensitive to gravity after germination has been
triggered by light. Repositioning experiments, in which
spores are rotated 180° with respect to the gravity vector at
varying times following the initiation of germination, show
that spores remain sensitive to the changes in the direction
Generation of cell polarity 835
of the gravity vector up to about 10–12 h into the germination programme after which time, the axis is fixed for most
spores (Edwards and Roux, 1994). Within 1 h of the
initiation of germination by light, a polar calcium current
can be detected which is aligned to the gravity vector, such
that calcium moves into the cell on its lower side, and out
of the cell from its upper side. When treated with the
calcium blocking drug, nifedipine, the calcium current is
reduced and the spore is no longer sensitive to the gravity
vector. Nifedipine also leads to a partial inhibition of polar
axis formation (Chatterjee et al., 2000), and so there is
evidence that, as in Fucus and Pelvetia, an early event in
polar axis formation is the establishment of an internal
Ca2+ gradient.
The first morphological indication of polarity in
Ceratopteris spores is the downward migration of the
spore nucleus, and the position of the nucleus predicts
the subsequent direction of rhizoid outgrowth. Nuclear
movement occurs in material germinated on a klinostat,
but the final position of the nucleus still predicts the
position of rhizoid outgrowth, suggesting that the events
are causally related. Light also affects the determination
of axis polarity but only weakly, the rhizoid tending to
develop away for the direction of a light source (Edwards
and Roux, 1998). Protoplasts isolated from Ceratopteris
gametophytes regenerate by polar outgrowth producing
a rhizoid in a manner similar to spore germination. The
alignment of the regeneration axis is sensitive to light and
to gravity, but in contrast to their effects on spore
germination, light is now much the stronger morphogenetic influence. Regeneration in a light gradient results
in over 80% of regenerating protoplast producing a
rhizoid on their shaded side. Cell biological and genetic
studies of Ceratopteris spore germination and protoplast
regeneration are at an early stage, but the system should
allow progress to be made in the understanding of how
light and gravity generate cell polarity.
Regeneration of moss protoplasts shows many similarities to the germination of fucoid zygotes and the regeneration of Ceratopteris protoplasts. Isolated protoplasts
of the mosses, Physcomitrella patens and Ceratodon purpureus regenerate by polar outgrowth (Stumm et al., 1975;
Cove et al., 1996). In Physcomitrella, protoplast regeneration only occurs at high light intensities (Jenkins and
Cove, 1983a), but Ceratodon protoplasts are able to
regenerate in darkness. Ceratodon protoplasts regenerated
in darkness show a weak alignment of their regeneration
axes to the gravity vector. Because there is no absolute
requirement for light during protoplast regeneration in
Ceratodon, it is possible to investigate its role in axis
formation in more detail. Light quality has an effect on
the overall response of populations of regenerating protoplasts and their behaviour can best be interpreted by
considering axis alignment and axis orientation separately
(Cove et al., 1996). In red light, the axes of most
regenerating protoplasts are closely aligned to the direction of the light source, whereas in blue light, axis
alignment shows much more variation. Axis orientation
in red light is, however, much poorer than in blue light.
In red light, only about two-thirds of protoplasts orient
outgrowth towards the light source, the remaining onethird growing away, while in blue light almost all protoplasts orient towards the source of light. Reorientation
experiments, similar to those carried out for Fucus, show
that there is a period before which axis fixation occurs,
during which a change of light direction results in a
revision of the polar axis. These experiments allow the
independent investigation of the effect of changing the
direction of the light input on axis alignment and orientation. Results are consistent with axis alignment being
fixed before axis polarity. There is thus a period during
which the alignment of the axis with respect to the light
direction is unaffected by changing the direction of the
light source, but during which the orientation of the axis
is still labile and so is disturbed by the change. As a result
of these observations, it has been proposed that the
mechanisms for the determination of axis alignment and
orientation are at least partially independent, perhaps
involving different morphogen gradients (Cove et al.,
1996). There is evidence that the detection of light direction is by way of phytochrome (D Cove, unpublished
data), but no cell biological investigations, similar to
those carried out in Fucus and Pelvetia, have yet been
made of the role of the cytoskeleton in axis development
or of the cell wall in axis fixation.
The recognition that during the regeneration of
Ceratodon protoplasts, axis alignment and orientation
may involve different processes, was due both to a biological feature of the system and also to what may seem a
minor difference in experimental design. Ceratodon protoplasts do not regenerate synchronously; there is about
30 h difference between the most rapidly regenerating
protoplasts and those that are slowest. This allows the
kinetics of repositioning to be studied in detail, albeit by
the use of computer model fitting (Cove et al., 1996). In
addition, Ceratodon experiments involving the repositioning of protoplasts with respect to light, rotated material by 90° relative to the direction to the light source.
Fucus and Pelvetia zygotes show more synchronous development and, in repositioning experiments, 180° rotation
has been used. As a result, the type of analysis that
revealed the complexities of axis determination in
Ceratodon protoplasts has not yet been carried out in
Fucus or Pelvetia.
Environmental inputs and the modification of cell
polarity
As discussed above, it is to be expected that plants,
lacking motility at the level of both the whole organism
836 Cove
and the individual cell, will often show an ability to
modify polarity in response to environmental inputs. At
the multicellular level, tropic responses of the shoots and
roots of flowering plants have been extensively studied.
The modification of the polarity of individual cells in
response to light and gravity is however, as with the
generation of cell polarity, often more easily studied in
non-flowering plant species.
The protonemal stage of moss gametophyte development provides particularly good material for the investigation of the effects of light and gravity on the
modification of cell polarity. Protonemal filaments extend
by the growth and division of their apical cells, and the
response of individual living cells to environmental signals
can readily be observed using time-lapse video microscopy. In Physcomitrella patens, there are two types of
protonemal filaments, chloronema and caulonema.
Chloronemal filaments, which are produced following
spore germination or tissue regeneration, only grow in
the light. Chloronemal filaments developing from germinating spores are sensitive to light direction, growing
towards a low intensity light source, at right angles to
the direction of a high intensity red light source and away
from a high intensity blue light source (Jenkins and Cove,
1983b). Caulonemal filaments are produced from chloronema and grow much more rapidly. The direction of
extension of their apical cell is sensitive both to directional
light inputs and to the gravity vector, but sensitivity to
gravity is only shown by cells growing in darkness (Cove
and Knight, 1987). As with chloronemal apical cells, the
response of caulonemal apical cells to light depends on
wavelength and intensity. No effect of gravity on the
apical cells of chloronemal filaments has been observed,
but since chloronemal cell division and growth only occur
in the light, this is not unexpected.
Mutants of Physcomitrella abnormal in the response of
their caulonemal apical cells to light or gravity have been
isolated, but no mutant showing an altered response to
both light and gravity has been isolated, indicating that
the responses must be to a large extent independent.
Physcomitrella mutants that are unable to align their
caulonemal apical cells to light ( ptr mutants) comprise at
least three complementation groups. ptr mutants produce
gametophores that are also insensitive to light direction,
showing that the gene products concerned are involved
in response to light at both the level of the individual cell
and the multicellular shoot (Cove and Knight, 1987). The
response of the chloronemal apical cells of ptr mutants
to directional light has been investigated, and surprisingly,
they show a normal response to both low and high light
intensities, but the change from a low-light to a highlight response occurs at lower intensities (Jenkins and
Cove, 1983c). No explanation has been suggested for this
pleiotropic phenotype.
Physcomitrella mutants having caulonemal apical cells
that are insensitive to gravity ( gtr mutants) have also
been obtained and comprise at least two complementation
groups. All gtr mutants so far obtained, produce gametophores that retain the wild-type gravity response, showing that in contrast to light, the response to gravity of
the individual cell and of the multicellular shoot must be
at least partially independent. In addition to mutants
insensitive to gravity, there are mutants of Physcomitrella
(designated gtrC ) that show a reversal in the orientation
of their gravity response while retaining the wild type’s
ability to align to the gravity vector ( Knight and Cove,
1991a). These mutants therefore have caulonemal filaments that grow downwards rather than away from
gravity.
Protonemal filaments of Ceratodon purpureus do not
show clear differentiation into chloronemal and caulonemal cell types and studies of the response to light and
gravity have concentrated on the apical cells of filaments
on older cultures. The responses of these are similar to
those of caulonemal apical cells of Physcomitrella. Plastid
sedimentation in a sub-apical zone of the Ceratodon apical
cell is clearly observable and it is proposed that this is
part of the gravi-sensing apparatus (Sack, 1993). Mutants
of Ceratodon with altered responses to light and gravity
have also been isolated, although little genetic analysis
has yet been carried out on them. Ceratodon ptr mutants
fall into two classes (Lamparter et al., 1996). Some have
their ability to respond to directional light restored by
growth in the presence of biliverdin, a precursor of the
phytochrome chromophore. These mutants are likely to
be deficient in the biosynthesis of phytochrome, and
provide evidence that light direction must be detected by
way of phytochrome, as was proposed earlier for
Physcomitrium turbinatum by Nebel (Nebel, 1968). This
class of Ceratodon ptr mutant show other pleiotropic
effects including reduced chlorophyll synthesis
(Lamparter et al., 1997), and a continuing response to
gravity when grown in light. It is therefore proposed that
the response of apical cells to gravity in the wild type is
actively turned off in light, detected by way of phytochrome. The second class of Ceratodon ptr mutants, show
no other pleiotropic effects, and the products of the genes
concerned are therefore likely to be acting downstream
of light detection, to bring about the regulated polar
extension of the apical cell. The identification and molecular characterization of these genes should therefore be a
high priority.
Mutants of Ceratodon having protonemal apical cells
that respond abnormally to gravity ( gtr) include completely insensitive mutants and mutants which respond to
gravity more slowly than the wild type ( Fig. 1). Ceratodon
mutants resembling Physcomitrella gtrC mutants, in
which the orientation of the gravity response is reversed,
have also been isolated. This latter class of mutant have
consequently been designated wrong way response (wwr)
Generation of cell polarity 837
mutants ( Wagner et al., 1997). Wild-type strains of both
Physcomitrella and Ceratodon also show reversal in the
orientation of their response to gravity Knight and Cove,
1991b; Young and Sack, 1992). Reversal occurs immediately upon repositioning cultures with respect to gravity
and during cell division. wwr Mutants show similar
reversals of the orientation of their responses, mirroring
the wild type, i.e. growth is down when the wild type
shows upward growth, and up when the wild type grows
downwards ( Wagner et al., 1997). Plastid sedimentation
in wwr mutants is also similar to that in the wild type,
indicating that the wild-type wwr gene must play a role
in establishing axis orientation in response to gravity but
not axis alignment.
Prospects
To understand how a polar axis of an individual cell is
generated or its polarity is modified in response to the
physical environment, requires the study of systems that
are accessible to both cell biological techniques and
genetic analysis using molecular and genomic tools.
Mutant isolation in the fucoid algae does not seem to be
a near prospect, but the amenability of Fucus and Pelvetia
to cell biological studies including imaging and microinjection will ensure that their continued study will contribute to the knowledge of polarity generation. Fern
spore germination and moss protonemal growth may,
however, be able to provide the necessary combination
of accessibility to both cell biology and genetic analysis.
To exploit these resources further, cell biological study of
mosses and ferns need to be extended. Future genetic
studies in both mosses and ferns must include the molecular characterization of genes involved in axis development.
Recently, the aphototropic phenotype of a Ceratodon
mutant blocked in phytochrome chromophore synthesis,
has been repaired by the micro-injection of appropriate
cDNA (Brücker et al., 2000). This micro-injection procedure has considerable potential in the analysis of cell
polarity and promises to add to the repertoire of techniques available to study gene function in mosses.
Molecular genetic studies of mosses are progressing
rapidly. With the establishment that homologous recombination occurs with high efficiency between transforming
DNA and genomic sequences ( Kammerer and Cove,
1996; Schaefer and Zryd, 1997), the techniques of gene
knockout and allele replacement are now available in
Physcomitrella (see, for example, Strepp et al., 1998),
allowing the detailed study of gene function. However,
no method of gene tagging is yet available and so it has
not yet been possible to isolate and characterize the genes
involved in the responses to light and gravity that have
already been identified by forward mutation. If such a
technique becomes available, it should be possible to
make rapid progress in understanding the mechanisms
needed for polar axis development at the level of the
individual cell.
Acknowledgements
I am grateful to Drs Ian Hope and Ralph Quatrano, as well as
my students, for stimulating me to think through my ideas
more clearly and to Drs Steve Miller, Ralph Quatrano and
Stan Roux for their very helpful comments on parts of the
manuscript.
References
Alessa L, Kropf DL. 1999. F-actin marks the rhizoid pole in
living Pelvetia compressa zygotes. Development 126, 201–209.
Brücker G, Zeidler M, Kohchi T, Hartmann E, Lamparter T.
2000. Microinjection of heme oxygenase genes rescues
phytochrome-chromophore-deficient mutants of the moss
Ceratodon purpureus. Planta (in press)
Chatterjee A, Marshall Porterfield D, Smith PS, Roux SJ. 2000.
Gravity-directed calcium current in germinating spores of
Ceratopteris richardii. Planta (in press).
Cove DJ, Hope IA, Quatrano RS. 1999. Polarity in biological
systems. In: Russo VEA, Cove DJ, Edgar LG, Jaenisch R,
Salamini F, eds. Development. Genetics, epigenetics and
environmental regulation. Berlin: Springer-Verlag, 507–524.
Cove DJ, Knight CD. 1987. Gravitropism and phototropism in
the moss, Physcomitrella patens. In: Thomas H, Grierson D,
eds. Developmental mutants in higher plants. Cambridge:
Cambridge University Press, 181–196.
Cove DJ, Quatrano RS, Hartmann E. 1996. The alignment of
the axis of asymmetry in regenerating protoplasts of the
moss, Ceratodon purpureus, is determined independently of
axis polarity. Development 122, 371–379.
Edwards ES, Roux SJ. 1994. Limited period of gravireponsiveness in germinating spores of Ceratopteris richardii. Planta
195, 150–152.
Edwards ES, Roux SJ. 1998. Influence of gravity and light on
the development of polarity of Ceratopteris richardii spores.
Planta 205, 553–560.
Errington J. 1992. Bacillus subtilis sporulation: a paradigm for
the spatial and temporal control of gene expression. In:
Russo VEA, Brody S, Cove DJ, Ottolenghi S, eds.
Development. The molecular genetic approach. Berlin:
Springer-Verlag, 28–44.
Fowler J, Quatrano RS. 1995. Cell polarity and cell fate
determination on brown algal zygotes. Seminars on Cell and
Developmental Biology 6, 347–358.
Hart JW. 1990. Plant tropisms and other growth movements.
London: Unwin Hyman.
Jacobs C, Shapiro L. 1998. Microbial asymmetric cell division:
localization of cell fate determinants. Current Opinion in
Genetics and Development 8, 386–391.
Jaffe LF. 1958. Tropistic responses of zygotes of the Fucaceae
to polarized light. Experimental Cell Research 15, 282–299.
Jenkins GI, Cove DJ. 1983a. Light requirements for the
regeneration of protoplasts of the moss, Physcomitrella
patens. Planta 157, 39–45.
Jenkins GI, Cove DJ. 1983b. Phototropism and polarotropism
of primary chloronemata of the moss, Physcomitrella patens:
responses of the wild type. Planta 158, 357–364.
Jenkins GI, Cove DJ. 1983c. Phototropism and polarotropism
of primary chloronemata of the moss, Physcomitrella patens:
responses of mutant strains. Planta 158, 432–438.
838 Cove
Kammerer W, Cove DJ. 1996. Genetic analysis of the result of
re-transformation of transgenic lines of the moss,
Physcomitrella patens. Molecular and General Genetics 250,
380–382.
Kirk DL. 1997. The genetic program for germ-soma differentiation in Volvox. Annual Review of Genetics 31, 359–380.
Klinger M, Tautz D. 1999. Formation of embryonic axes and
balstoderm patterns in Drosophila. In: Russo VEA, Cove DJ,
Edgar LG, Jaenisch R, Salamini F, eds. Development.
Genetics, epigenetics and environmental regulation. Berlin:
Springer-Verlag, 311–330.
Knight CD, Cove DJ. 1991a. Genetic analysis of a mutant class
of Physcomitrella patens in which the polarity of gravitropism
is reversed. Molecular and General Genetics 230, 12–16.
Knight CD, Cove DJ. 1991b. The polarity of gravitropism in
the moss Physcomitrella patens is reversed during mitosis and
after growth on a clinostat. Plant, Cell and Environment 14,
995–1001.
Kropf DL, Bisgrove SR, Hable WE. 1999. Establishing a growth
axis in fucoid algae. Trends in Plant Science 4, 490–494.
Kropf DL, Kloareg B, Quatrano RS. 1988. Cell wall is required
for fixation of the embryonic axis in Fucus zygotes. Science
239, 187–194.
Lamparter T, Esch H, Cove D, Hartmann E. 1997. Phytochrome
control of phototropism and chlorophyll accumulation in the
apical cells of protonemal filaments of wild type and an
aphototropic mutant of the moss Ceratodon purpureus. Plant
Cell Physiology 38, 51–58.
Lamparter T, Esch H, Cove D, Hughes J, Hartmann E. 1996.
Aphototropic mutants of the moss Ceratodon purpureus with
spectrally normal and with spectrally dysfunctional phytochrome. Plant, Cell and Environment 19, 560–568.
Miller SM, Kirk DL. 1999. glsA, a Volvox gene required for
asymmetrical division and germ cell specification, encodes a
chaperon-like protein. Development 126, 649–658.
Miller SM, Schmitt R, Kirk DL. 1993. Jordan, an active Volvox
transposable element similar to higher plant transposons. The
Plant Cell 5, 1125–1138.
Mochizuki T, Saijoh Y, Tsuchiya K et al. 1998. Cloning of inv,
a gene that controls left/right asymmetry and kidney
development. Nature 395, 177–181
Mosig G, Eiserling F. 1988. Phage T4 structure and metabolism.
In Calender R, ed. The bacteriophages, Vol. 2. New York:
Plenum Press, 521–606.
Nebel BJ. 1968. Action spectra for photogrowth and phototropism in protonemata of the moss Physcomitrium turbinatum.
Planta 81, 287–302.
Quatrano RS, Shaw SL. 1997. Role of the cell wall in the
determination of cell polarity and the plane of cell division
in Fucus embryos. Trends in Plant Science 2, 15–21.
Robinson KR, Lorenzi R, Ceccarelli N, Gualtieri P. 1998. Retinal
identification in Pelvetia fastigiata. Biochemical and
Biophysical Research Communications 243, 776–778.
Robinson KR, Miller BJ. 1997. The coupling of cyclic GMP
and photopolarization of Pelvetia zygotes. Developmental
Biology 187, 125–130.
Rijven AHGC. 1968. Randomness in the genesis of phyllotaxis.
I. The initiation of the first leaf in some Trifolieae. New
Phytologist 67, 247–256.
Sack FD. 1993. Gravitropism in protonemata of the moss
Ceratodon. Memoirs of the Torrey Botanical Club 25, 36–44.
Schaefer DG, Zrÿd J-P. 1997. Efficient gene targeting in the
moss Physcomitrella patens. The Plant Journal 11, 1195–1206.
Shaw SL, Quatrano RS. 1996. Polar localization of a dihydropyridine receptor on living Fucus zygotes. Journal of Cell
Science 109, 335–342.
Souter M, Lindsey K. 2000. Polarity and signalling in plant
embryogenesis. Journal of Experimental Botany 51, (in press).
Strepp R, Scholz S, Kruse S, Speth V, Reski R. 1998. Plant
nuclear gene knockout reveals a role in plastid division for
the homolog of the bacterial cell division protein FtsZ, an
ancestral tubulin. Proceedings of the National Academy of
Sciences, USA 95, 4368–4373.
Stumm I, Meyer Y, Abel WO. 1975. Regeneration of the moss
Physcomitrella patens (Hedw.) from isolated protoplasts.
Plant Science Letters 5, 113–118.
Supp DM, Brueckner M, Potter SS. 1998. Handed asymmetry
in the mouse: Understanding how things go right (or left) by
studying how they go wrong. Seminars in Cell and
Developmental Biology 9, 77–87.
Twell D, Park SK, Lalanne E. 1998. Asymmetric division and
cell-fate in developing pollen. Trends in Plant Science
3, 305–310.
Wagner TA, Cove DJ, Sack FD. 1997. A positively gravitropic
mutant mirrors the wild-type protonemal response in the
moss, Ceratodon purpureus. Planta 202, 149–154.
Yokoyama T, Copeland NG, Jenkins NA, Montgomery CA,
Elder FFB, Overbeek PA. 1993. Reversal of left-right
asymmetry: A situs inversus mutation. Science 260, 679–682.
Young JC, Sack FD. 1992. Time-lapse analysis of gravitropism
in Ceratodon protonemata. American Journal of Botany 79,
1348–1358.