/. Embryol. exp. Morph. Vol. 33, 1, pp. 95-115, 1975
Printed in Great Britain
95
Morphogenesis in Micrasterias
I. Tip growth
By T. C. LACALLI 1
From the Department of Biology, McGill University
SUMMARY
Observations on lobe growth in the lateral wings of the developing primary cell wall in the
desmid Micrasterias rotata are reported and discussed. Patterns of incorporation of methyl[3H>methionine and C-1-[3H] glucose into the primary wall as revealed in autoradiograms
indicate that formation of the new wall is concentrated at the tips of lobes. Patterns follow
the predictions of Robertson's model for tip growth in fungal hyphae; thus they link growth
in M. rotata lobes with mechanisms of cell elongation found in other cells. Damage done to
selected regions of the cell surface with a laser microbeam demonstrates that only certain
regions are required for continued growth and morphogenesis while much of the surface
plays only a passive role. In growth stages at which lobes are already well defined (stage 4
and later) continued growth of each lobe requires the participation of an area no more than
4-5 /tm in diameter, here termed a singularity, at its tip. At early stages (prior to stage 3)
singularities .per se cannot be demonstrated. At these stages the capacity to initiate lobes and
hence to form singularities is not fixed at specific points, but is distributed over an area of the
surface no less than 10 /tm in diameter. Singularities, by their persistence and repeated duplication, are directly responsible for the spatial form of the two lateral wings.
INTRODUCTION
Each of the two semicells comprising a mature Micrasterias cell is characterized by a central polar lobe flanked on either side by a flat, fan-shaped wing
notched deeply about its perimeter (Fig. ID). In the several hours immediately
following mitosis the elaborate semicell pattern is reproduced in each daughter
cell by extensive growth of the septum of primary cell wall material that forms
at the cell isthmus to separate the two daughters. The septum, at first a simple
circular plate, bulges to a hemispherical form. A central bump appears upon the
hemisphere and develops, by elongation, into a polar lobe. Simultaneously
lateral bumps give rise to the two wings by elongation and repeated branching
(Figs. 1A-C, 2). Under normal circumstances, the two lateral bumps are positioned so that the wings and polar lobe that develop from them are restricted
to the single plane in which the wings and polar lobe of the parent semicell also
lie. In this way the biradial symmetry of the parent cell is transmitted to its
daughters.
1
Author's address: Bellairs Research Institute, St James, Barbados.
96
T. C. LACALLI
C
Fig. 1. Cell development in M. rotata. (A-C) A dividing cell 60, 140 and 200 min
respectively following first appearance of the septum (stages 3, 7 and 10 in the
notation used in Fig. 2). (D) Scanning electron micrograph of an interphase cell.
Fig. 2. Profiles of a semicell during development. The numbering of stages shown
here is used throughout the text. Stages are separated by 20 min from one another.
Note that each lateral wing is composed of two major lobes, an upper and a lower one.
Morphogenesis in Micrasterias. /
97
A considerable literature exists dealing with the role of cell nucleus and
cytoplasm in determining final semicell form in Micrasterias. If the nuclear
contribution is interfered with wing lobes can continue to elongate, but lobe
branching does not occur (Kallio, 1951; Waris & Kallio, 1964; Selman, 1966).
Surprisingly, cells enucleated at very early developmental stages can always
produce at least a complement of three unbranched lobes corresponding in
position to the normal polar lobe and lateral wings. This phenomenon led Waris
and Kallio to postulate the existence of a cytoskeleton of plasmatic axes responsible for the fundamental threefold pattern of development in enucleate cells.
From the above evidence and from the observation that cells with increased
nuclear ploidy have more fully developed semicells, they concluded that the
further elaboration of the fundamental threefold pattern is under nuclear
control. The existence of species-specific differences in semicell form reinforces
this conclusion. More recently, Kallio & Lehtonen (1973) determined that the
potential for axis formation is transmitted from the polar lobe and wings of the
parent semicell to the forming semicell via non-nuclear 'growth centers' located
at the cell isthmus. The transmission process showed dependence on the cell
nucleus, but at a time considerably before the nuclear determinants of lobe
form appeared.
It is still unclear how the various cytoplasmic and nuclear determinants act
to cause patterned growth in new semicells. Teiling (1950) proposed that areas
of 'meristematic cytoplasm' might contribute to patterned growth. Kiermayer
(Kiermayer & Jarosch, 1962; Kiermayer, 1964) demonstrated patterns of differential wall deposition and of the strength of membrane attachment to cell wall
in new semicells. He considered the septum especially relevant to the problem
of semicell pattern and suggested that it could act as a morphogenetic template
(Kiermayer, 1967, 19706). Kallio & Lehtonen (1973) and Kies (1970) have
mentioned the possible importance of organized arrays of organelles or other
structures. Unfortunately electron microscopical studies reveal no obvious
ultrastructural basis for morphogenesis (Kiermayer, 1968 a, b, 1970a). No particular role for the cell wall in morphogenesis has as yet been demonstrated
though Waddington (1966) has alluded to the possible importance of local
specializations of the cell surface.
In the studies reported here, laser irradiations were performed as a means of
damaging selected parts of the growing primary wall to determine the contribution of each part to the final form. These studies demonstrated both the normal
fate and the developmental potential of specific regions of the wall. In addition,
patterns of radioactive labeling were studied in autoradiograms in order that
patterns of cell wall synthesis might be understood. The results of these two
approaches are complementary and together implicate a particular type of cell
wall growth, tip growth, as the mechanism by which lobes elongate. Some
implications of this work with regard to the organization of tip growth are
mentioned.
7
EMB
33
98
T. C. LACALLI
MATERIALS AND METHODS
Micrasterias rotata was used exclusively in this study. Cells were obtained
from Beaver Lake, Vancouver, B.C., in February 1970. Identification was made
by reference to West & West (1905) and cultures were established from single
cells. Cells were grown in the MS medium of Waris (1953) on a 12 h light/12 h
dark cycle (500-1200 ft lux). On this light regime the cells became roughly
synchronized so that about 15% of the cells divided each day. Those dividing
on any given day began division from 1 to 2 h before the beginning of the light
period.
Laser irradiation. Dividing cells were irradiated with a microbeam for which
a helium-neon gas laser (model M 610, Metrologic Inst., 143 Harding Ave.,
Bellmawr, New Jersey) served as source. The microbeam apparatus has been
previously described (Lacalli & Acton, 1972) and was constructed so as to make
the laser beam coaxial with the source of illumination. The beam (0-5 mW
continuous beam, 633 nm) was focused at the object plane by the microscope
condenser and viewed through standard microscope optics. Laser damage to
the eyes is always a possibility, even with a helium-neon laser of very low power
such as the one used here (Bloom, 1968). For this reason, a green interference
filter was introduced into the eyepiece to reduce the beam intensity substantially.
In addition, the beam was viewed obliquely, or its image was projected through
a suitable eyepiece and observed. These precautions would not be sufficient to
protect an experimenter were a substantially more powerful laser being used.
Cells to be lased were placed on the microscope stage supported on a large
coverglass. Glass slides were not used for this purpose because the laser beam
was routinely focused at a point closer to the surface of the condenser front
lens than the thickness of a slide.
The continued normal growth of semicells was unaffected by laser irradiation
of the semicells themselves. Therefore a number of blue and green dyes were
tested to determine whether they would be taken up by the cells, and whether
this uptake would promote absorption of laser energy by the semicell. Of 25
dyes tested, five were clearly taken up either into the cell or by the cell wall
(malachite green, alcian blue, janus green, nile blue and methylene blue) and
of these five, alcian blue was the most interesting, promoting damage that
depended very specifically on the exact point of irradiation. All the laser experiments reported here were done on cells treated in solutions of alcian blue (Alcian
Blue 8XG, Allied Chemical Co.) in culture medium made by dilution of a
stock solution of alcian blue in distilled water. Dye concentrations up to 0-010 %
were used in the experiments. M. rotata cells can in fact develop normally and
survive for several days at this concentration. At 0-020 %, however, cells become
retarded in their development and show other signs of toxicity (e.g. contraction
of the chloroplasts). In a typical laser experiment, five to ten dividing cells were
placed in the dye solution and left for 20 min. Cells were then lased one at a
Morphogenesis in Micrasterias. /
99
time and photographic recordings made of each lasing. Lased cells were removed
to spot-plate depressions so that subsequent development and cell divisions
could be observed. Because alcian blue precipitates slowly when mixed with
culture medium, only four or five cells were lased in any one experiment before
beginning again with new cells and fresh dye solution. In most lasings, one
daughter cell served as the experimental cell and the other as control. In all,
about 500 lasings were done; the observations reported here depend on detailed
interpretation of about half of these. All experimental cells were kept and
followed through at least one further cell division, and the most damaged cells
were followed through several. In no case was a laser-induced shape abnormality
inherited. Though the experiments described refer to work on M. rotata, similar
results were obtained with cells of M. radiata.
Autoradiography. Autoradiograms were prepared from cell wall ghosts of cells
exposed to tritiated compounds for varying lengths of time during development.
Usually 30-50 developing cells were placed in 0-3-0-4 ml of culture medium to
which a labeled compound (International Chemical & Nuclear) had been added
to a concentration of 200 /iCi/ml (D-glucose-l-[3H], methyl-[3H]-methionine or
generally labeled [3H]-proline) or 50 mCi/ml ([3H]-water). Cells were removed,
washed in changes of distilled water and ruptured using an ultrasonic cleaning
device (20 kHz). The time from first wash to rupture was standardized at 5 min;
this 5 min is uniformly included wherever reference is made in the results to
the length of time a cell was labeled. Drops of 0-1 % crystal violet were added
to the preparation of ruptured cells so that the primary walls could be seen and
picked out individually. These were dried onto slides for standard autoradiographic processing using Ilford L-4 nuclear emulsion at a 1/5 dilution. Completed autoradiograms were stained with crystal violet. Drawings made of
autoradiograms to compare areas of label incorporation were done with the
aid of a camera lucida.
RESULTS
It was immediately apparent that irradiations of growing semicells treated
with alcian blue had interesting and specific consequences only when regions of
the cell surface were included in the lasing. Lasing the septum at a point exactly
between the two developing semicells, for example, prevented formation of the
polar lobe in both daughter cells without altering greatly the form of the wings
(Fig. 3). Laser damage was most apparent in lasings of semicells sufficiently
well developed as to have their major lobes and notches already defined (i.e.
stage 6 and older). Lasings of the tips of lobes in such cells caused portions of
the final complement of lobes to be missing (Fig. 4). The missing lobes were
always those which would normally have arisen by further growth and branching
of the irradiated lobe. In these lasings of late stage semicells, interpretation of
each experiment was facilitated because the lobes and notches already present
served as landmarks. From Fig. 4, for example, it is clear that no part of the
7-2
100
T. C. LACALLI
Fig. 3. (A) An 8 sec lasing (arrow) of a stage-4 semicell treated in 0-001 % alcian
blue. (B) One of the daughter cells on the following day. Both daughters were
similar in form; note absence of the polar lobe (arrow). Scale as in Fig. 1.
irradiated lobe participated in further morphogenesis because the entire stage-6
lobe can be accounted for in the completed semicell (Fig. 4C). The effects of
lasing notches or the sides of lobes were far less striking. In these cases the lased
region failed to participate in morphogenesis and hence was often not normally
disposed in the completed semicell (Fig. 5). The normal complement of lobes
was always present, however. Lobes near to the lasings were in some cases
retarded in their growth, but this retarding was usually only temporary. From
a series of about 40 similar lasings (using 0-001 % alcian blue, 2-3 sec lasings
with a 3 fim beam diameter) of cells of stage 6 or later, the following conclusions
were drawn. (1) Semicell lobes grow and develop entirely independently of one
another. A lobe is not adversely affected by damage inflicted on neighboring
lobes, nor does it compensate by its own growth for the absence of the neighbor.
(2) Certain regions of the cell surface, the lobe tips, play a more crucial role in
lobe growth than other regions of the surface. Morphogenesis is especially
sensitive to laser damage at these specific regions.
In theory it should have been possible to map the cell surface at each develop-
Morphogenesis in Micrasterias. /
101
Fig. 4. (A) A 3 sec lasing (arrow) at the tip of a lobe in a stage-6 semicell treated in
0001 % alcian blue. Daughter cells are shown 30 min after the lasing (B) and on
the following day (C, D). Arrows indicate the point of lasing. Scale as in Fig. 1.
mental stage to determine the position and size of these sensitive regions and,
through a series of stages, to trace their developmental history. This was attempted
for stages younger than stage 6, in particular for very early stages, but proved
to be difficult for a variety of reasons. On the one hand, lobes became less
independent of one another in their growth at progressively earlier stages. Thus
an undamaged lobe might respond by hypertrophy when the growth of a neighboring lobe was severely retarded or if it was eliminated altogether. The righthand wing shown in Fig. 3B and the left-hand wing in Fig. 8E are examples of
such compensatory growth. Each of these wings appears to have three major
lobes rather than the usual two; the third and extra lobe is the one lying closest
to the isthmus. Because such compensation could occur, it was difficult to
determine the effect of a lasing from the completed shape of the wings only.
Complete elimination of one of the two major wing lobes at an early stage
could evoke hypertrophy of the remaining lobe, whereas damage that eliminated
neither lobe could sufficiently retard one or the other of the two so as to make
Fig. 5. (A) A 3 sec lasing (arrow) at a notch of a stage-6 semicell treated in 0001 %
alcian blue. Daughter cells are shown 40 min after lasing (B) and on the following
day (C, D). Arrows indicate the point of lasing. Scale as in Fig. 1.
it abnormally small. In either case the resulting wings would be similar, their
size would be intermediate between that of a normal wing and that of a wing
with only a single major lobe of normal size. Such intermediate wings were
frequently produced by lasings of semicells younger than stage 6. Difficulty of
interpretation was aggravated by an apparent plasticity of the semicell wall at
early stages. Lasings, whether they had eliminated lobes or not, did not remain
as permanent landmarks, but became incorporated into notches as development
proceeded in the surrounding undamaged wall. In practice, then, continuous
observation of semicell development was required following lasing. The number
of major lobes first formed and their position relative to the point of lasing had
to be known if interpretations were to be made with any confidence. Finally,
there was a problem in selecting the appropriate dose of laser radiation for each
experiment. Too great a dose eliminated the differential sensitivity of regions
of the semicell surface. If a stage-6 cell were colored in 0-005 % alcian blue
rather than a 0-001 % solution, for example, a lasing at the wing notch as shown
in Fig. 5 would have prevented all further development of the entire wing;
Morphogenesis in Micrasterias. /
103
Fig. 6. Two examples of developing cells in which lobes have recovered from lasings
of insufficient strength. The cell in (A) has been plasmolysed to show the wall more
clearly. Scale as in Fig. 1.
Fig. 7. Composites showing the positions of numerous lasings of the cell perimeter
of developing cells at stages from 3 to 6. Cells were treated with 0001-0005 %
alcian blue, lased for from 2 to 8 sec with a 3 /tm spot. (A) Lasings which failed to
eliminate either of the two major wing lobes or the polar lobe. Those marked with
a vertical bar resulted in partial damage to the polar lobe, but did not eliminate it
altogether. (B) Lasings which resulted in the elimination of one of the two major
wing lobes, or a branch of one of these (after stage 5 only) or the polar lobe.
with cells colored in solutions much less concentrated than 0-001 %, lasing of
lobe tips appeared to arrest lobe growth only temporarily, and partially recovered lobes could be observed (Fig. 6). The dose at which specific laser-sensitive
regions could be demonstrated was found also to be different at different developmental stages; increased doses were required at increasingly early stages. In
practice the dose was adjusted by changing both the concentration of the alcian
104
T. C. LACALLI
C
Fig. 8. (A) An 8 sec lasing (arrow) of a stage-l cell treated in 0-005 % alcian blue.
(B-D) The daughter cells 45, 75 and 105 min respectively after the lasing. Note that
a mark (arrow) remains at the point of lasing. (E) One daughter cell on the following
day. Both daughter cells were similar. Note the complete but abnormally small
wing on the lased side which must have arisen from 'above' the point of lasing.
Scale as in Fig. 1.
blue solution used for coloring cells and the duration of lasing. For stages 1-3,
a 0-005 % alcian blue solution and 5-8 sec lasings were used. For stages 4-10,
a 0-001-0-002 % solution and 2-3 sec lasings were used. The spot size of the
microbeam was kept constant at 3 /tm.
Despite the difficulties described, lasings that eliminated one or the other of
the major wing lobes or the polar lobe could be distinguished from those that
did not for stages as early as stage 4 (Fig. 7). The lasings unambiguously
revealed two sensitive regions between 3 and 5 jum in diameter on each side
of the polar lobe. At progressively earlier stages these two were progressively
closer together until, at stages earlier than stage 4, only one such region could be
distinguished on either side. From Fig. 7 one might predict that a single lasing
at an appropriate point at stage 3 would be sufficient to eliminate both sensitive
regions on one side and thus would eliminate an entire wing. This could in fact
be accomplished by lasing at stage 3 with a spot 10 /im or more in diameter,
Morphogenesis in Micrasterias. /
105
20 /mi
Fig. 9. Results of lasings of stage-2 semicells treated in 0005 % alcian blue for 8 sec
with a 3 /tm spot. Numbers in parentheses indicate the number of cells in which each
effect was observed in the 25 cells studied. The point of contact of the two semicells
serves as a convenient landmark. Lasing at points below the point of contact (A)
results in all cases in a semicell wing which has developed from above the point of
lasing by a process similar to that shown in Fig. 8. Lasing of a region including the
point of contact (B), even if the spot is large, does not prevent wing formation. The
wing may again arise from above the point of lasing as in (A) or may arise laterally
so that it is shifted out of the plane of biradial symmetry of the parent semicell.
Lasing of the area above the point of contact (C) causes lobes to arise from the area
below the point of contact, the same area from which lobes were shown not to arise
in (A).
but could not with a 3 /*m spot. It appeared to be generally true of stages younger
than stage 4 that an entire wing or the polar lobe (or wing and polar lobe together)
could be eliminated by lasing with a suitably large spot. Certainly then, the laser
was not ineffective at these early stages. The ability to unambiguously demonstrate laser-sensitive regions using a 3 /an. spot, however, disappeared some time
between stage 3 and stage 4. That this resulted not from ineffectiveness of the
small laser spot, but instead from a fundamental change in the properties of
the laser-sensitive regions, was shown by detailed study of stages 1 and 2.
In a series of 65 lasings of stage-1 and stage-2 semicells, damage was in 25
cases sufficient to produce a mark at the point of lasing which persisted and
served as a landmark (Fig. 8 is an example) delimiting the area from which any
wing lobes formed could have originated. Wings could not be prevented from
forming by increasingly encroaching on such an area in successive cells, however
(Fig. 9). Wings instead arose from areas which had been lased previously in
other cells without affecting damage or preventing formation of the wings in
these other cells, or from positions lateral to the lasing so that the resulting
106
T. C. LACALLI
6
40 //
////// /mm
Fig. 10. An experimental lasing (arrow) shown from above (A), with a reconstruction
of the dimensions as seen from the side (B).
wing projected out of the normal plane of semicell symmetry. Hence, at stages
1 and 2, an area considerably larger than the 3 /tm laser spot has the capacity
to initiate lobes, but only a fraction of this area need participate directly in the
event.
A note is required here concerning the mechanism by which laser damage is
affected. The laser beam is itself a double cone of light, the dimensions of which
were measured (Fig. 10). Power output of the helium-neon laser used is in the
order of 108 times less than that of a pulsed ruby laser. Sufficient energy can
be delivered with the helium-neon system to damage cells, however, if an
absorbing substance or natural pigment is present in the target. In M. rotata,
alcian blue probably served to color the cell wall, as the cell perimeter took on
a distinct blue tinge in the solutions used. It was not determined whether or
not the dye penetrated the plasma membrane. If a cell was plasmolysed immediately after lasing, the cell wall and plasma membrane were found to be stuck
together at the point of the lasing (Fig. 11). The area of fusion was usually
sufficiently small to show that laser damage was strictly localized. In cells which
were more intensely colored with alcian blue, the area of fusion could be demonstrated to be larger by several times than the diameter of the laser spot. Marks
left by lasing appeared over a period of from several to 10 min and appeared
to be due to optical refraction: the lased areas possibly developed different
Morphogenesis in Micrasterias. /
107
B
Fig. 11. (A) A 2 sec lasing (arrow) of a stage-8 semicell treated in 0-001 % alcian
blue and lased with a 3 /*m spot. (B-D) Plasmolysis of the same cell in 0-2 M sucrose
to show fusion between the cell membrane and the cell wall at the point of lasing
(arrow). Scale as in Fig. 1.
physical properties than the surrounding more normal wall or were subject to
different stresses.
Of the four tritiated compounds tested, only methionine and glucose were
incorporated into the primary cell wall and could be demonstrated in autoradiograms of primary cell wall ghosts. Patterns of incorporation were similar
in the two cases though methionine labeling was generally denser and the patterns
more distinct. Examples are shown for methionine (Fig. 12) and glucose (Fig. 13)
for various stages and durations of exposure to the two compounds. Label
was generally confined to the tips of lobes, but to ever-enlarging areas of the
tips as the duration of exposure to label was increased. This can be seen more
readily in a series of composite diagrams (Fig. 14) prepared by tracing the
extent of tip label of uniform density in about 100 cells given pulses of different
duration. These composites could be drawn with considerable confidence, for
though individual walls did vary greatly in density of their labeling, the area
labeled was always very constant among cells of a given stage exposed to label
for similar lengths of time. From the diagrams it can be clearly seen that the
108
T. C. LACALLI
H
Fig. 12. Methionine autoradiography of primary cell wall ghosts. Walls are from
cells labeled for lOmin (A), 15 min (B, C), 20min (D-F) and 35 min (G-l).
Darkened notches in this figure and in Fig. 13 result from folds in the primary
wall and not from autoradiographic labeling. Scale as in Fig. 1.
methionine label is more localized in its initial incorporation than glucose label.
If extrapolations are made from the 10 min contours, hypothetical zero time
contours may be drawn to indicate the maximum possible extent of the region
of initial incorporation. For the case of glucose label, this hypothetical zero
time contour always encloses a considerable area of the tip. For the case of
Morphogenesis in Micrasterias. /
109
Fig. 13. Glucose autoradiography of primary cell wall ghosts. Walls are from cells
labeled for 10 min (A), 30 min (B, C) and 65 min (D-F). Scale as in Fig. 1.
methionine, there are only a series of points being labeled at such a hypothetical
zero time. It appears in general, from the diagrams, that glucose initially extends
to about that region of wall which will contain methionine label only after
20-25 min exposure.
In addition to the label at lobe tips, semicells also exhibited a faint veined
pattern of label over their surface. This pattern was prominent in semicells of
very late stages (stages 11-13; Fig. 13 F is an example) which showed particularly dense labeling of lobe tips. The density of label appeared in general to be
about 50-100 times less per unit area in veined areas than at lobe tips. For
younger stage semicells, most of which had less densely labeled tips, the veined
pattern was present but was relatively inconspicuous.
Extractions performed on labeled walls indicated that the patterns seen were
not artifacts resulting from cytoplasmic residues or insufficient washing of the
preparations, but represented real incorporation of polysaccharide substance
into the primary wall (Lacalli, 1974). In the case of glucose, labeling indicated
110
T. C. LACALLI
12
Fig. 14. Composite diagram showing the extent of uniform tip label from tracings
made of the upper half of the upper of the two major wing lobes at stages 9,11 and 13
(see inset). Patterns are shown for walls labeled with methionine (left-hand column)
and with glucose (right-hand column). Numbers indicate in each case the duration
of labeling in min.
synthesis and incorporation of cellulosic microfibrils. Methionine incorporation,
on the other hand, was an indication of methylation and showed regions of
incorporation of acid matrix substances bearing labeled methyl esters.
DISCUSSION
Tip growth is well known as the major means by which elongation occurs in
fungal hyphae, root hairs and some algal cells. Growth of the lobes in the two
lateral wings of Micrasterias rotata is clearly also of this type. If it is assumed
that the change in area of M. rotata semicells arises only by addition of material
at the tips of lobes, the plots of perimeter growth for semicells shown in Fig. 2
Morphogenesis in Micrasterias. /
111
Fig. 15. Idealized diagrams of cell wall fate for the upper half of the upper of the two
major wing lobes at stages 7, 9 and 11 (see inset). It is assumed that all new wall is
introduced at the lobe tip and is pushed backward by subsequent addition of wall
material at a rate corresponding to that at which the wall perimeter is increased as
measured from Fig. 2. Contours divide the wall into age categories; wall within the
first contour is less than 10 min old, wall within the second contour is less than
20 min old and so on.
can be converted to fate maps showing by contour the relative age of different
areas of the wall surface. This is done by computing the rate of increase of the
perimeter of a lobe and then marking off equal time intervals backwards from
the lobe tip. Examples for which connecting contours have been estimated are
shown in Fig. 15 for three stages for the upper half of the upper of the two
major wing lobes of the semicell. The close correspondence between this figure
and the observed patterns of methionine incorporation into the primary wall
(Fig. 14) demonstrates that the assumption of tip growth is essentially a valid
one if methionine incorporation may be thought to indicate the areas of new
wall deposition. The latter supposition is reasonable and requires only that all
new wall include a newly methylated fraction. Patterns of incorporation differing
from the methionine tip pattern are observed in semicells, but their existence
does not challenge the assumption of tip growth. Incorporation of glucose, for
example, occurs initially over a larger area of the lobe tip. This fact has been
considered previously (Lacalli & Acton, 1974) and serves to further link lobe
growth in Micrasterias to tip growth in hyphae because direct comparison can
be made with models of hyphal tip growth proposed by Robertson (1959,1968).
Robertson divided the tip into two regions: a central one at the very apex where
wall extension dominates all other growth-related activities, and a larger region
covering the remainder of the growth hemisphere characterized by wall hardening. Comparison suggests that the hardening reaction in Micrasterias would be
associated with glucose incorporation (microfibril synthesis) and the process of
wall extension with methionine incorporation (matrix synthesis). In addition
to the tip label, there is a vein-like pattern of labeling over the surface of the
semicells both in the case of methionine and of glucose labeling. Because this
112
T. C. LACALLI
pattern is quantitatively less important than the tip label, it car^be disregarded
for the moment as a secondary process. Veining will be dealt with in the second
article of this series.
Considered simply as marking experiments, the laser irradiation work shows
that semicell notches are relatively fixed in their position compared to lobe tips.
This reinforces the conclusion that lobes grow by tip growth. The laser experiments go further than this, however. They demonstrate that morphogenetic
potential is not uniformly distributed over the cell surface, but rather is associated with discrete regions: the participation of regions at the lobe tips is
absolutely required for continued growth while other regions of the cell surface,
notably the semicell notches, play a passive role. The laser irradiation experiments
summarized in Fig. 7 supply the regions of morphogenetic potential with both
size and positional continuity. This is sufficient to warrant that they be accepted
as well-defined functional components of the Micrasterias semicell and that
they be given a proper name for purposes of reference. In their general properties,
the regions of potential are not unlike the singularities discussed by Tokunaga
& Stern (1965). By 'singularities' Tokunaga and Stern meant the specific points
on the otherwise undifferentiated surface of an insect leg at which potential to
produce bristles was localized.To use 'singularity' in reference to laser-sensitive
regions in Micrasterias does not do violence to this original usage (Stern, personal communication), for, in using the word only the fact of the localization
of morphogenetic potential at discrete points is meant to be referred to, with
nothing implied as to the structure of singularities or the means by which
they might act.
It must be emphasized that singularities are far smaller than the growing
region of cell surface, the tip, of which they are part. Singularities occupy only
a fraction of the curvature of the tip; this is particularly evident at early stages
when tips are 15 jam. or more in breadth. The exact size of a singularity is difficult
to determine. From Fig. 7 it is clear that singularities can be no larger than
4-5 /*m in diameter, but they could in fact be considerably smaller. The singularity associated with the pointed lobe shown in Fig. 6 A must have been no
more than a fraction of a micron in diameter, for example. It has not been
possible to determine for M. rotata whether singularities are situated exactly
at the cell surface (i.e. in the primary wall) or are part of the underlying plasma
membrane and cytoplasm. Experiments were designed to solve this problem;
an attempt was made to lase the wall and membrane separately in plasmolysed
cells and to observe the morphogenetic consequences. Because morphogenesis
of plasmolysed cells is characteristically abnormal, these experiments were
inconclusive. At this time it can be said only that singularities are associated
with peripheral structures of the cell; the cell wall, the plasma membrane and
the immediately adjacent cytoplasm with whatever structures or organization
it may contain.
An idealized plot of the position of singularities for stages 3-6 can be
Morphogenesis in Micrasterias. /
113
15 //in
Fig. 16. An idealized diagram showing the progress of morphogenetically necessary
regions of the cell surface (singularities) demonstrated by laser experiments (Fig. 7)
for stages 3-6.
constructed from the laser experiments and is shown in Fig. 16. By implication,
if the singularities are responsible for morphogenesis in any way, this is because
each singularity is responsible for the growth of a single lobe at whose tip the
singularity is located. Further, singularities duplicate in association with lobe
branching. During semicell development, singularities can first be demonstrated
when two become established on opposite sides of the stage-3 semicell and a
third at its center. The two lateral singularities are responsible for the initial
lateral growths which, through repeated branching, develop into wings. Prior to
stage 3, the positions of the two initial singularities are not irreversibly determined; in fact singularities per se probably do not as yet exist. For during this
time, damage to the normal point of appearance of an initial singularity will
not prevent a lobe from arising from the surrounding area of wall. Once a
singularity has been established at its normal position (i.e. after stage 4), its
destruction no longer evokes such a response from the surrounding wall. The
semicell has therefore lost its ability to generate new singularities by stage 4.
These observations on singularities and the fashion in which they arise closely
parallel observations on the determination of structure within morphogenetic
fields in developing multicellular organisms, both plant and animal. The ability
to demonstrate processes of determination of structure in a single cell such as
Micrasterias has important consequences for the way in which developmental
phenomena are to be understood in general and suggests that some operational
terms used in the field of experimental embryology may be equally well applied
to developmental processes for which only a part of a cell rather than a whole
cell is considered to be the unit of life involved.
It should be noted that the existence of singularities alters the way in which
the role of the cytoplasm in tip growth can be viewed in Micrasterias and, by
implication, in other cells exhibiting tip growth. Cytoplasmic events, particularly those resulting in vesicle accumulation, may be necessary to tip growth in
Micrasterias, but are clearly not sufficient; otherwise damage to only very small
S
EMB
33
114
T. C. LACALLI
regions of the cell wall would not be so decisive as is the case in Micrasterias.
It can be seen from patterns of methionine incorporation that formation of
new cell wall at the tips of Micrasterias lobes is organized on a scale of only a
few microns; because the cytoplasmic activities and polarities so far observed
in this cell appear to operate on a scale of tens of microns or more, the cytoplasm
is a poor candidate for an organizer of tip growth. It is instead more satisfying
to postulate that something approaching in its own dimensions the demonstrated
scale of wall organization must be responsible, particularly if a suitable entity,
such as the singularity, can be demonstrated. In the extreme it would be possible
to imagine the cytoplasm playing a strictly supportive role in tip growth as a
supplier of raw materials, with final organization of the growth process dependent entirely on the singularity. The evidence presented here, however, proves
only that singularities are necessary for tip growth.
This study was supported by the National Research Council of Canada and was carried
out at the University of British Columbia. I thank A. B. Acton for his interest and support
and for bringing the published discussions of singularities to my attention.
REFERENCES
York: John Wiley & Sons.
nuclear quantity in the genus Micrasterias. Suomai.
elain-ja kasvit. Seur. van kasvit. Julk. {Ann. Bot. Soc. zool.-bot. fenn. Vanamo) 24 (2), 1-122.
KALLIO, P. & LEHTONEN, J. (1973). On theplasmatic template system in Micrasterias morphogenesis. Suomai. Tiedeakat. Toim. (IV) {Ann. Acad. Sci. fenn. A. iv.) Biologica 199, 1-6.
KIERMAYER, O. (1964). Untersuchungen iiber die Morphogenese und Zellwandbildung bei
Micrasterias denticulata Breb. Protoplasma 59, 76-132.
KIERMAYER, O. (1967). Das Septum-Initialmuster von Micrasterias denticulata und seine
Bildung. Protoplasma 64, 481-484.
KIERMAYER, O. (1968 a). The distribution of microtubules in differentiating cells of Micrasterias
denticulata Breb. Planta 83, 223-236.
KIERMAYER, O. (19686). Hemmung der Kern- und Chloroplastenmigration von Micrasterias
durch Colchizin. Naturwissenschaften 55, 299-300.
KIERMAYER, O. (1970a). Elektronenmikroskopische Untersuchungen zum Problem der
Cytomorphogenese von Micrasterias denticulata Breb. I. Allgemeiner tJberblick. Protoplasma 69, 97-132.
KIERMAYER, O. (1970/?). Causal aspects of cytomorphogenesis in Micrasterias. Ann. N.Y.
Acad. Sci. 175, 686-701.
KIERMAYER, O. & JAROSCH, R. (1962). Die Formbildung von Micrasterias rotata Ralfs und
ihre experimentelle Beeinflussung. Protoplasma 54, 382-420.
KIES, L. (1970). Elektronenmikroskopische Untersuchungen iiber Bildung und Struktur der
Zygotenwand bei Micrasterias papillifera {Desmidiaceae). I. Das Exospor. Protoplasma
70,21-47.
LACALLI, T. C. (1974). Composition of the primary wall in Micrasterias rotata. Protoplasma
80, 269-272.
LACALLI, T. C. & ACTON, A. B. (1972). An inexpensive laser microbeam. Trans. Am. microsc.
Soc. 91, 236-238.
LACALLI, T. C. & ACTON, A. B. (1974). Tip growth in Micrasterias. Science, N.Y. 183,
665-666.
ROBERTSON, N. F. (1959). Experimental control of hyphal branching and branch form in
hypomycetous fungi. /. Linn. Soc. {Bot) 56, 207-211.
BLOOM, A. L. (1968). Gas Lasers. New
KALLIO, P. (1951). The significance of
Morphogenesis in Micrasterias. /
115
ROBERTSON, N. F. (1968). The growth process in fungi. A. Rev. Phytopath. 6, 15-136.
G. G. (1966). Experimental evidence for nuclear control of differentiation in Micrasterias. J. Embryo/, exp. Morph. 16, 469-485.
TEILING, E. (1950). Radiation of desmids, its origin and its consequences as regards taxonomy
and nomenclature. Bot. Notiser. pp. 299-327.
TOKUNAGA, C. & STERN, C. (1965). The developmental autonomy of extra sex combs in
SELMAN,
Drosophila melanogaster. Devi Biol. 11, 50—81.
C. H. (1966). Fields and gradients. In Major Problems in Developmental
Biology, 25th Symp. Soc. devl Biol. (ed. M. Locke), pp. 105-124. New York: Academic
Press.
WARIS, H. (1953). The significance for algae of chelating substances in the nutrient solutions.
Physiologic PI. 6, 538-543.
WARIS, H. & KALLIO, P. (1964). Morphogenesis in Micrasterias. Adv. Morphogen. 4, 45-80.
WEST, W. & WEST, G. S. (1905). British Desmidiaceae, vol. n. London: Adlard & Sons for
the Ray Soc.
WADDINGTON,
{Received 26 April 1974, revised 28 August 1974)
8-2