83
Development 120, 83-89 (1994)
Printed in Great Britain  The Company of Biologists Limited 1994
Primordial germ cell migration in Drosophila melanogaster is controlled by
somatic tissue
Mariusz K. Jaglarz and Kenneth R. Howard
Roche Institute of Molecular Biology, Roche Research Center, 340 Kingsland Street, Nutley, NJ 07110, USA
SUMMARY
In Drosophila, as in many other organisms, primordial
germ cells show invasive and migratory behavior moving
from their site of origin to the somatic component of the
gonad. At a characteristic time in development, the primordial germ cells pass across the primordium of the gut
and migrate on its outer surface toward the mesoderm,
where they eventually associate with the somatic tissues of
the gonad. Here we demonstrate that the exit and
migration are specific behaviors of the primordial germ
cells and that they are controlled by the somatic tissue of
the embryo rather than by a germ cell autonomous clock.
Using mutations, we show that these controlling somatic
events probably occur in the tissue of the gut primordium
itself.
INTRODUCTION
mesoderm of parasegments 10, 11 and 12 and is marked by
virtue of its expression of the transposable element 412
(Brookman et al., 1992). After leaving the pocket, PGCs enter
the mesoderm and associate with the primordia of the somatic
gonad. Subsequently, the gonadal mesoderm ensheathes the
germ cells forming a compact structure lying in abdominal
segment 5 (Campos-Ortega and Hartenstein, 1985). Recently
genetic and morphological studies of the development of the
gonad (Brookman et al., 1992; Cumberledge et al., 1992;
Szabad and Nothiger, 1992), reviewed in Wilkins (1992),
suggest that the assembly of the gonadal mesoderm requires
specific cells in A5 and is independent of germ cells per se.
Here our interest is in the early events of PGC migration as
they leave the gut primordium. In order to study this process, we
used low-light optical imaging of dye-labeled transplanted germ
and labeled host germ cells and pocket tissue in living embryos.
These procedures allow us to determine the effect of the nature
and stage of development of the transplanted cells on their
behavior. We show that the exit and migration of the PGCs are
regulated by the somatic tissue, they are not determined by a
germ cell autonomous timing mechanism and are PGC-specific
events. Using mutants, we demonstrate that the tissue that the
pole cells first encounter, the posterior midgut primordium,
regulates the exit from the pocket and that the orientation of the
migration does not require the presence of the target of this
motion, the mesoderm. We suggest that this orientation depends
on cues residing in the primordium of the posterior midgut.
The events bringing primordial germ cells (PGCs) or pole cells
and gonadal mesoderm together in Drosophila melanogaster
were first characterized in the early 1940s (Rabinowitz, 1941;
Sonnenblick, 1941). The pole cells form at the very posterior
of the embryo early in development in response to a localized
determinant. These cells will contribute to the germ line
(Underwood et al., 1980; Technau and Campos-Ortega, 1986).
However, not all pole cells go on to do this, many of them
degenerate or become lost during their migration (Poulson,
1950; Sonnenblick, 1950; Underwood et al., 1980; Technau
and Campos-Ortega, 1986; Hay et al., 1988). This difference
in fate seems to correlate with specific splicing activity in these
cells (Kobayashi et al., 1993). There is at present no way to
distinguish these cells in living embryos. Here the terms ‘pole
cells’ and ‘PGCs’ are used interchangeably.
During gastrulation (stage 7-8), the pole cells are pushed
into the hindgut and posterior midgut primordium (here called
the pocket) where they remain, apparently immobile, until late
stage 10. At this time, they pass across the posterior midgut
primordium (PMG) in the distal region of the pocket and move
dorsally toward the gonadal mesoderm (Campos-Ortega and
Hartenstein, 1985). Recently, reagents that specifically stain
the PGCs have allowed better determination of the number of
pole cells and further characterization of their migratory route
in fixed material (Hay et al., 1988; Lasko and Ashburner,
1988). From the total number of ~40 pole cells only less than
half will eventually populate the germ line (Technau and
Campos-Ortega, 1986; Hay et al., 1988). The final target of the
PGC migration, the somatic tissue of the gonad forms in the
Key words: primordial germ cell, gonadogenesis, mutants, cell
motility, Drosophila
MATERIALS AND METHODS
In order to label PGCs and the PMG, embryos at stage 3 (stages
84
M. K. Jaglarz and K. R. Howard
Table 1. Summary of the transplantation experiments
Type
donor stage
host stage
time difference
1
2
3
early 4
late 6
+3 hours
7
early 4
−3 hours
6
6
0 hours
Three types of transplantation were performed: type 1 and 2 heterochronic
transplantation of pole cells; type 3 isochronic transplantation of blastoderm
cells.
according to Campos-Ortega and Hartenstein, 1985) were injected at
the posterior pole with nuclear targeted lissamine rhodamine dextran
(Mr 70,000, prepared essentially as described (Vincent and O’Farrell,
1992). The limited time for diffusion of this label resulted in its
inclusion only in posterior nuclei and labeling of the host pole cells
and pocket. These embryos were then observed either without further
manipulation or after transplantation of cells into the forming pocket.
These transplanted cells, either PGCs or somatic blastoderm cells,
were labeled with fluorescein so that they could be distinguished from
the tissues of the host embryo. In these experiments, the donor
embryos were injected at stage 2 with fluorescein-dextran (Mr 70,000,
Molecular Probes). Up to 10 cells were then transplanted to recipients. Generally embryos from a ru klarsicht stock were used as recipients and Ore R as donors. klarsicht is a maternal effect mutation,
which produces more transparent embryos (Wieschaus and NüssleinVolhard, 1986). In one set of experiments oskar301 embryos, which
lack pole cells were used as recipients (Lehmann and NüssleinVolhard, 1986). In addition to isochronic pole cell transplantations,
isochronic transplantations of blastoderm cells were performed as
well as heterochronic transplantations of pole cells where the stage of
development of the transplanted cells preceded or lagged that of the
host by approximately 3 hours. These procedures are summarized in
Table 1. All manipulations and observations were done at 18°C. Each
heterochronic transplantation experiment was repeated 10 times with
similar results.
Images were acquired using a Photometrics camera with a Peltier
cooled Kodak KAF1400 chip. The rhodamine and fluorescein signals
were obtained separately using a three-band dichroic and barrier filter
with separate excitation filters (Chroma Technologies) ensuring
perfect registration of these two images. The focus of the microscope
was swept through a predetermined depth (usually ~G of the embryo)
and the image allowed to integrate on the chip. Exposure times were
generally about 5 seconds using a 100 W tungsten illumination source.
The intervals between sets of collections was 1 minute in transplantation experiments and 20 seconds when observing only the signal
from the rhodamine tag. A Zeiss Axiovert 35 microscope equipped
with Narashigi hydraulic micromanipulators, Ludl focus control, a
filter wheel and Uniblitz shutters was controlled by BDS Image
software running on a Macintosh computer. The data were processed
to remove background noise and enhance contrast before display
either on the Macintosh screen or as an NTSC video sequence
generated from the Macintosh display signal using a Mediator encoder
and a Panasonic OMDR.
Mutant embryos were of the genotype huckebein (hkbA/TM3, ftz-βgal, constructed using a chromosome bearing the hkbA allele obtained
from Detlef Weigel via Jordi Cassanova); snail twist (sna11a,twis60
obtained from Kavita Arora); forkhead (fkhXT6) and tailless (tllg), both
obtained from the Drosophila stock center at Indiana. The germ cells
were examined in all cases using a rat polyclonal antiserum directed
against vasa protein (gift of Robin Wharton). In the case of hkbA,
homozygous embryos were identified by virtue of their lack of
expression of β-gal. Confocal images were acquired using a Bio-Rad
MRC 600 scanning confocal system with the Krypton Argon laser and
dual pass dichroic optics mounted on a Nikon Diaphot microscope.
The immunostained embryos were counterstained with Evans Blue (2
µg/ml in 20% ethanol) before being dehydrated in ethanol and
Fig. 1. Labeled cells can migrate to the gonadal primordium normally.
(A) A side view of a living early gastrula showing pole cells
(arrowhead) at the posterior pole lying over the host tissue that will
form the gut primordium (labeled red). Transplanted PGCs are labeled
green and the endogenous PGCs are labeled red. (B) Gonadal
primordia (arrowheads) of the host embryo (dorsal view, posterior pole
to the right), which have been populated by host PGCs (red) as well as
younger transplanted PGCs (green). Scale bars are both 50 µM.
mounted in methyl salicyalte. In the case of sna twi embryos from a
7-9 hour collection (at 25°C) were dehydrated in ethanol and then
taken to a mixture of methyl salicylate and Canada balsam. All mutant
embryos in the sample with extended germ bands were then identified by morphological criteria and mounted in fine glass capillary
tubes (Hilberg, Malsfeld. art 14 216 02) before observation under a
Zeiss axio-series ×25 oil immersion lens using Nomarski optics. This
arrangement (Prokop and Technau, in press) allows the embryo to be
viewed from any angle by rotating the glass tube. Each one of 27 germ
band extended >7 hour sna twi embryos was observed from lateral
and dorsal perspectives and recorded on film.
RESULTS
In order to test that our labeling techniques work and determine
their effects on development, embryos labeled only by
injection with nuclear tagged rhodamine dextran were
observed as well as pole cells labeled with fluorescein dextran
isochronically transplanted to such labeled hosts. The results
of these observations demonstrate that we can visualize the
events of migration and that our techniques are relatively
innocuous. These data are consistent with previous studies on
fixed material (Hay et al., 1988; Lasko and Ashburner, 1988).
The dynamics of the process, especially the PGCs exit from
the pocket, can be best assessed by watching the video display.
Interestingly, these data show that the exit from the PMG is
relatively rapid and that at 11 hours of development (at 18°C)
Primordial germ cell migration in Drosophila
the PGCs begin to leave the pocket in a process that takes only
20 to 30 minutes. In the following 150 minutes, the PGCs
migrate on the surface of the PMG moving dorsally. This
85
migration and the movement of the pocket allow the PGCs to
meet that part of the mesoderm where the somatic component
of the gonadal primordium eventually forms.
Fig. 2. Frames from time-lapse studies of heterochronically transplanted pole cells. The panels show dorsal views of a living embryo (posterior
pole to the right). A through H are collected at successively later stages of development. The host PGCs, hindgut and midgut (PMG) primordia
(the pocket) are labeled red. The transplanted pole cells are 3 hours younger than the recipient and are labeled green. (A) Embryo at stage 9;
most of the PGCs lie in the PMG, although two transplanted cells remain in the hindgut primordium (arrow). (B) 1 hour 15 minutes later, stage
10; the transplanted cells that lie in the distal part of the pocket leave along with the recipient PGCs. Over the next 2.5 hours the host and
transplanted PGCs migrate on the surface of the PMG (C, 1 hour 35 minutes after A; D, 3 hours after A) before migrating into the mesoderm
(E, 4 hours after A; F, 4 hours 25 minutes after A). Note that the transplanted PGCs that lie in the hindgut primordium (arrow in A) cannot be
seen to be motile until stage 12 (G, arrow: 6 hours 25 minutes after A) at which time the other PGCs have already begun to condense into the
gonadal primordia (arrowheads). This late migration continues during stage 13 (H, arrows, ~7 hours later than A). Scale bar is 50 µm. Similar
results have been obtained with transplantation of PGCs 3 hours older than the host.
86
M. K. Jaglarz and K. R. Howard
Fig. 4. PGCs fail to exit the gut primordium in huckebein mutants.
Confocal projection of lateral (A) and dorsal (B) views of a
huckebein embryo after germ band shortening showing germ cells
(green) trapped inside the reduced and mutant gut primordium
(posterior is to the right and dorsal is uppermost). Compare with the
dorsal view of a wild-type embryo at approximately the same stage
shown in C where the germ cells have already associated with the
gonadal mesoderm. Scale bar is 50 µm.
Fig. 3. Transplanted blastoderm cells do not migrate to the gonadal
primordia. The panels show dorsal views of a living embryo
(posterior pole to the right). (A) Isochronically transplanted
blastoderm cells (green) inside the posterior midgut primordium
(PMG) of an embryo at stage 9. (B) During stage 10, the host PGCs
(red dots on the surface of the PMG) begin migration but blastoderm
cells do not (compare with Fig. 3B-F where transplanted PGCs
behave similarly to the host PGCs). (C,D) Host PGCs migrate to
form gonadal primordia (arrowheads). Note the location of the
blastoderm cells (compare with Fig. 3G where some transplanted
PGCs enter the forming gonadal primordia together with the host
PGCs). Scale bar is 50 µm.
The labeling and visualization techniques do not interfere
with the ability of transplanted pole cells to enter the germ line.
An example of the preparation soon after transplantation of
pole cells is shown in Fig. 1A. After 20 hours of observation
of either isochronic or heterochronic transplantation experiments, a variable number of the transplanted cells (ranging
from 2/7 to 6/8) become incorporated into the gonad. An
example is shown in Fig. 1B. In some cases, the hatched larva
can develop into an adult which transmits the genetic markers
of the transplanted cells just as transplanted cells which are not
labeled with dye (van Deusen, 1976; Technau and CamposOrtega, 1986).
In experiments designed to determine if the factors regulating the time of their exit from the pocket reside in the soma of
the embryo or if this event was autonomously programmed in
the PGCs included in the pocket, pole cells were heterochronically transplanted so that the transplanted cells were either
three hours advanced or retarded in developmental time with
regard to their host. Despite the fact that, in their normal environment, the transplanted cells would have migrated at
different times, in all twenty cases (ten with positive and ten
Primordial germ cell migration in Drosophila
87
Fig. 5. The migration is oriented in the absence of mesoderm. Two examples of sna twi double mutant embryos are shown in both dorsal (A,C)
and lateral (B,D) views (posterior is to the right and dorsal is uppermost). These are representatives of 25/27 cases examined in which at least
half of the germ cells left the pocket and moved toward the dorsal surface. The remaining cells were trapped in the pocket. In only 2/27
examples in this 7-9 hour (25°C) collection did the cells that exited show an apparently random distribution in the dorsoventral axis. These are
likely to represent an earlier stage in the migratory process in these embryos. Scale bar is 50 µm.
with negative time shifts), the donor PGCs, which were associated with the PMG, migrate along with and were indistinguishable in their behavior from the host PGCs. This is illustrated by consideration of the panels in Fig. 2, which show a
set of frames from one such experiment. Note that in F the
stage of development of the transplanted germ cells, which
have already begun to move bilaterally in the mesoderm, is
equivalent to that of the host germ cells in panel B where these
cells have only just begun their migration on the PMG. This
behavior was consistent in all the experiments and establishes
that the onset of the migration is controlled by the developmental stage of the soma. Since our data collection technique
does not record z data, we cannot accurately quantitate the
coordinates of each cell throughout the experiment. Consequently our analysis has been restricted to largely qualitative
description of the data set observed as a time-lapse movie
either on the computer display or as a video signal.
The possibility that the PGCs of the recipient were affecting
the time of migration of the transplanted cells was ruled out by
using oskar301 hosts, which lack pole cells. Regardless of their
age the transplanted PGCs migrate according to the host developmental stage (not shown). It is worth noting here that in none
of our experiments were any differences between any of the
data sets observed that could be attributed to heterosexual
transplantation. This is consistent with other data that show
that male PGCs can colonize the female gonad and vice versa
(Steinmann-Zwicky, 1992).
It was common for some transplanted pole cells to lodge
either in the ectodermal hindgut primordium (Fig. 2A) or to
remain outside the embryo just under the vitelline membrane.
These ectopic cells behave in a different and informative
fashion. The cells outside the embryo become motile soon after
transplantation (data not shown) indicating that PGCs are
capable of active movement at an early stage in their development. In contrast, the PGCs trapped in the upper part of the
pocket (the hindgut primordium) do not move along with the
host cells but disperse and move actively after they have been
carried along for some time by the retracting germ band. This
later migratory activity occurs long after the PGCs of the recipients have associated with the gonadal mesoderm (compare
Fig. 2A-F with Fig. 2G,H). Interestingly, in no case do these
ectopic pole cells enter the forming gonads.
To test how specific to PGCs the migration is, blastoderm
cells taken from the middle-dorsal part of the donor were transplanted into the host embryo. The fate of these cells is to form
amnioserosa (Campos-Ortega and Hartenstein, 1985).
Although the blastoderm cells enter the pocket together with
PGCs (Fig. 3A), they do not leave along with the host PGCs
(Fig. 3B) and do not pass to the gonad (Fig. 3C,D). These data
indicate that the movements of the PGCs from the pocket to
the mesoderm are specific events requiring the properties of
PGCs and are not simply the consequences of mechanical
forces that displace the contents of the pocket dorsally toward
the mesoderm. The possibility that this is caused by the
inability of isolated blastoderm cells to move is unlikely since
these cells can move once released from the pocket during later
morphogenetic movements (not shown).
We have used mutants to investigate the source of the cues
governing both PGC exit and migration on the pocket. Here
we present data on two mutant genotypes: huckebein (hkb) and
the double mutant snail twist (sna twi), which are particularly
informative about the events described above. The mutation
hkb belongs to the zygotic terminal class of genes (Weigel et
al., 1990). The mutant phenotype involves a failure of the
development of the posterior midgut. In these embryos, the exit
of the germ cells from the pocket is dramatically affected.
Although the hkb phenotype is variable, it is common to find
mutant embryos in which no germ cells leave the pocket. An
example of this phenotype is shown in Fig. 4. Even in less
extreme hkb embryos only a few PGCs manage to migrate to
88
M. K. Jaglarz and K. R. Howard
the mesoderm. This indicates that the properties of the hkbdependent part of the PMG are necessary for exit. Two other
mutants affecting the development of the gut, tailless (Weigel
et al., 1990) and forkhead (Jürgens and Weigel, 1988), do not
perturb the exit (data not shown).
In order to determine if the orientation of the migration on
the pocket depends on the presence of the mesodermal target
of the migration, we investigated the outcome of lack of
mesoderm (sna twi) (Leptin, 1991; Ray et al., 1991). In this
case, many of the germ cells exit relatively normally (Warrior,
personal communication) and migrate dorsally toward the
surface of the pocket which would normally appose the
mesoderm (Fig. 5).
is correctly oriented in the absence of mesoderm (sna twi
mutant) rules out this sort of explanation. We favor the idea
that the signals orienting this migration reside in the posterior
midgut primordium itself.
We would like to thank Kavita Arora, Jordi Cassanova, Ruth
Lehmann, Detlef Wiegel and Eric Wieschaus for Drosophila stocks.
Robin Wharton for the anti vasa antibody. Helen Skaer for the gift of
microcapillary tubes and sharing with us her technical knowledge.
Mike Bate, Amy Bejsovec, Jose Campos-Ortega, Ruth Lehmann,
Helen Skaer, Colin Stewart, Rahul Warrior and Eric Wieschaus for
useful input. Waleed Danho (Roche, Nutley) for synthesizing the
nuclear targeting signal peptide. John Conner (RIMB), Lou Marrick
(BDS), Tim Mitchison and Paul Millman (Chroma Technologies) for
help with imaging and Duggal (NYC) and Jim Hempstead (RIMB)
for output.
DISCUSSION
Earlier work has demonstrated that the sole fate of the pole
cells is to contribute to the germ line (Underwood et al., 1980;
Technau and Campos-Ortega, 1986) and that this ability is
maintained during short periods in primary culture (Allis et al.,
1979). In our studies, we have focused on the early migratory
behavior of the PGCs, in particular their movement across and
navigation on the PMG. Previous experiments transplanting
pole cells directly into mesoderm demonstrated that passage
across the pocket is not an obligate step in the migration of
PGCs since under these circumstances they can populate the
gonadal primordia (Illmensee and Mahowald, 1976).
The experiments reported have shown that the exit from the
pocket and the migration towards the gonadal mesoderm are
specific properties of PGCs and that blastoderm cells cannot
migrate in the same way. In addition, we have demonstrated
that the initiation of movement through the pocket is not programmed autonomously in each pole cell. Instead the PGCs are
capable of responding to a signal from the somatic tissue of
the embryo at least three hours before or three hours after they
normally do so.
The onset of this migration requires some change in the hkbdependent part of the posterior midgut primordium. It maybe
that the germ cells are already competent to exit from the
pocket and that the exit is regulated by changes in the somatic
substratum. In this case, the signal would not affect the nature
of the germ cells but simply allow them to express an intrinsic
motility program. This sort of mechanism would be analogous
to that observed in the regulation of mouse PGC migration
where the embryo controls the behavior of these cells first by
preventing and then permitting their migration (FfrenchConstant et al., 1991). Alternatively, the hkb-dependent part of
the pocket may produce a signal that results in a change in the
state of the germ cells, which initiates migration. However,
such an explanation is inconsistent with our data showing that
pole cells transplanted to ectopic positions are prematurely
motile and that PGCs lodged in ectopic positions in the hindgut
begin to migrate long after the normally placed PGCs and
never enter the gonad.
Once they have left the pocket, the germ cells migrate on its
surface toward the mesoderm. This movement could be
directed by a diffusable factor produced by the developing
gonadal mesoderm in a mechanism formally similar to that
proposed for the guidance of murine PGCs to the genital ridge
(Godin et al., 1990). However, the fact that germ cell migration
REFERENCES
Allis, C. D., Underwood, E. M., Caulton, J. H. and Mahowald, A. P. (1979).
Pole cells of Drosophila melanogaster in culture. Normal metabolism,
ultrastructure and functional capabilities. Dev. Biol. 69, 451-465.
Brookman, J. J.Toosy, A. T., Shashidhara, L. S. and White, R. A. H. (1992).
The 412 retrotransposon and the development of gonadal mesoderm in
Drosophila. Development 116, 1185-1192.
Campos-Ortega, J. A. and Hartenstein, V. (1985). The Embryonic
Development of Drosophila melanogaster. Berlin: Springer-Verlag.
Cumberledge, S., Szabad, J. and Sakonju, S. (1992). Gonad formation and
development requires the abd-A domain of the bithorax complex in
Drosophila melanogaster. Development 115, 395-402.
Ffrench-Constant, C., Hollingsworth, A., Heasman, J. and Wylie, C. C.
(1991). Response to fibronectin of mouse primordial germ cells before,
during and after migration. Development 113, 1365-1373.
Godin, I., Wylie, C. and Heasman, J. (1990). Genital ridges exert long-range
effects on mouse primordial germ cell numbers and direction of migration in
culture. Development 108, 357-63.
Hay, B., Ackerman, L., Barbel, S., Jan, L. Y. and Jan, Y. N. (1988).
Identification of a component of Drosophila polar granules. Development
103, 625-40.
Illmensee, K. and Mahowald, A. P. (1976). The autonomous function of germ
plasm in a somatic region of the Drosophila egg. Exp. Cell Res. 97, 127-140.
Jürgens, G. and Weigel, D. (1988). Terminal versus segmental development
in the Drosophila embryo: the role of the homeotic gene fork head. Wilhelm
Roux’s Arch. Dev. Biol. 197, 345-354.
Kobayashi, S., Kitimura, T., Sasaki, H. and Okada, M. (1993). Two types of
pole cells are present in the Drosophila embryo, one with and one without
splicing activity for the third P-element intron. Development 117, 885-893.
Lasko, P. F. and Ashburner, M. (1988). The product of the Drosophila gene
vasa is very similar to eukaryotic initiation factor-4A. Nature 335, 611-7.
Lehmann, R. and Nüsslein-Volhard, C. (1986). Abdominal segmentation,
pole cell formation, and embryonic polarity require the localized activity of
oskar, a maternal gene in Drosophila. Cell 47, 141-52.
Leptin, M. (1991). twist and snail as positive and negative regulators during
Drosophila mesoderm development. Genes Dev. 5, 1568-76.
Poulson, D. F. (1950). Histogenesis, organogenesis and differentiation in the
embryo of Drosophila melanogaster Meigen. In Biology of Drosophila (ed.
M. Demerec), pp. 168-274. New York: Wiley.
Rabinowitz, M. (1941). Studies on the cytology and early embryology of the
egg of Drosophila melanogaster. J. Morphol. 69, 1-49.
Ray, R. P., Arora, K., Nüsslein-Volhard, C. and Gelbart, W. M. (1991). The
control of cell fate along the dorsal-ventral axis of the Drosophila embryo.
Development 113, 35-54.
Sonnenblick, B. P. (1941). Germ cell movements and sex differentiation of the
gonads in the Drosophila embryo. Proc. Natn. Acad. Sci. USA 27, 484-489.
Sonnenblick, B. P. (1950). The early embryology of Drosophila melanogaster.
In Biology of Drosophila (ed. M. Demerec), pp. 62-167. New York: Wiley.
Steinmann-Zwicky, M. (1992). How do germ cells choose their sex?
Drosophila as a paradigm. BioEssays 14, 513-518.
Szabad, J. and Nothiger, R. (1992). Gynandromorphs of Drosophila suggest
one common primordium for the somatic cells of the female and male gonads
in the region of abdominal segments 4 and 5. Development 115, 527-33.
Primordial germ cell migration in Drosophila
Technau, G. M. and Campos-Ortega, J. A. (1986). Lineage analysis of
transplanted individual cells in embryos of Drosophila melanogaster: III.
Commitment and proliferative capabilities of pole cells and midgut
progenitors. Wilhelm Roux’s Arch. Dev. Biol. 195, 489-498.
Underwood, E. M., Caulton, J. H., Allis, C. D. and Mahowald, A. P. (1980).
Developmental fate of pole cells in Drosophila melanogaster. Dev. Biol. 77,
303-314.
van Deusen, E.B. (1976). Sex determination in germ line chimeras of
Drosophila melanogaster. J. Embryol. Exp. Morphol. 37, 173-185.
Vincent, J. P. and O’Farrell, P. H. (1992). The state of engrailed expression is
not clonally transmitted during early Drosophila development. Cell 68, 923931.
89
Weigel, D., Jurgens, G., Klingler, M. and Jackle, H. (1990). Two gap genes
mediate maternal terminal pattern information in Drosophila. Science 248,
495-8.
Wieschaus, E. and Nüsslein-Volhard, C. (1986). Looking at embryos. In
Drosophila: a Practical Approach, pp. 199-227. Oxford: IRL.
Wilkins, A. S. (1992). Origins of the gonadal mesoderm in Drosophila.
BioEssays 14, 627-628.
(Accepted 30 September 1993)