Axolotl retina and lens development: mutual tissue stimulation and autonomous failure in the eyeless mutant retina
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
J. Embryol. exp. Morph. 96,151-170 (1986) 151
Printed in Great Britain © The Company of Biologists Limited 1986
Axolotl retina and lens development: mutual tissue
stimulation and autonomous failure in the eyeless
mutant retina
ROBERT CUNY* AND GEORGE M. MALACINSKI
Department of Biology, Indiana University, Bloomington, Indiana 47405, USA
SUMMARY
During eye development in the axolotl {Ambystoma mexicanum Shaw), morphogenetic
movements bring together tissues from head epidermis, neuroectoderm and neural crest. The
stages 0 to 14 of axolotl eye development were expanded from Rabl's (1898) stages 1 to 10 and
correlated with Harrison's (1969) stages. At the onset of neurulation (stage 13 of Harrison), the
head epidermis is already determined to form skin, and the neuroectoderm is committed to form
brain, because these tissues develop autonomously in 60% Leibovitz L-15 culture medium.
However, a sequence of mutual tissue interactions is necessary to stimulate eye development.
When head epidermis and neuroectoderm were cocultured, eyes developed, containing retinas
with photoreceptors (stage 8) and lenses with secondary lens fibres (stage 8). The first event
needed in this case appears to be the secretion of a growth factor from the head epidermis which
stimulates retina development from the neuroectoderm. When neuroectoderm cultures were
exposed to nondialysable extracts (30/igml"1) of an adult epidermis derivative, the bovine
cornea, pigmented retinas (stage 6) and at higher concentrations (SOOOjUgmP1) neural retinas
developed (stage 6). In turn, lens formation is stimulated in the head epidermis by a retina-
derived growth factor. A mutation that causes adult eyelessness (e eyeless, nonlethal, recessive)
affects the earliest event in eye development (stage la), while a mutation that causes arrest of
eye development (mi microphthalmic, lethal, recessive) acts in a later event (stage 8). Two
possibilities have been considered in the case of mutation e: either the head epidermis does not
secrete sufficient amounts of active growth factor, or the presumptive retina itself is defective.
The latter statement turned out to be correct, because mutant e neural plates rarely developed
early retina stages (stage 5) in organ culture when combined with wild-type head epidermis. On
the other hand, wild-type neural plates formed advanced retinas (stage 8) in all cases when
combined with mutant e head epidermis. As expected, no retina or lens developed when both
neural plate and head epidermis were from mutant e donors. The heterozygous presence of
genes e and r (renal insufficiency, lethal, recessive) produces duplications of the presumptive
retina at the optic stalk. This observation is consistent with the notion that the mutation e,
assisted by the r locus, causes a primary failure in the presumptive retinal region.
INTRODUCTION
The eye tissues of the axolotl (Ambystoma mexicanum Shaw) are derived from
distant embryonic regions, which are brought together by the morphogenetic
movements (Reyer, 1977). These embryonic regions involve first the head
* Present address: Unite de Recherches GSrontologiques, INSERM 118, 29 rue Wilhem,
75016 Paris, France.
Key words: e (eyeless), mi (microphthalmic), r (renal insufficiency) mutations, axolotl, retina,
lens, eye development stages, retina-stimulating factor.152 R. CUNY AND G. M. MALACINSKI epidermis and neuroectoderm and later neural crest pigment cells (Frost, Epp & Robinson, 1984) and neural crest mesoderm. At the late gastrula stage, the prechordal plate mesoderm secretes a neuralizing protein factor (Tiedemann & Born, 1978) which stimulates the cells of the overlying neural plate, thus separating the already determined optic fields (Mangold, 1929; Adelmann, 1937). In these fields the presumptive dorsal retinal cells are derived from the contralateral body side (Jacobson & Hirose, 1978). The optic fields are located in the anterolateral neural plate (Fischel, 1921; Manchot, 1929; Woerdeman, 1929; Adelmann, 1929) but they rapidly move to the neural folds (Spemann, 1938; Jacobson, 1962; Brun, 1981) where they contact the head epidermis. The latter tissue is necessary for retina development (Filatow, 1926; Detwiler & Van Dyke, 1954; Rollhauser- Terhorst, 1981) to the rhodopsin-containing photoreceptor stage (Papermaster & Schneider, 1982). The nature of the retina-stimulating effect of the head epi- dermis has been unknown. Using neural plate cultures we could stimulate retina development with a macromolecular extract of bovine cornea, a head-epidermis derivative. The retina in turn stimulates the development of a lens (Le Cron, 1907; Liedke, 1955; Karkinen-Jaaskelainen, 1978) with lens-specific proteins, the crys- tallins (Brahma & McDevitt, 1974; Brahma & Bours, 1976), by secreting protein growth factors (Arruti & Courtois, 1978; Beebe, Feagans & Jebens, 1980; Courty etal. 1985). Two mutations (Malacinski & Brothers, 1974; Malacinski, 1978) are known to disrupt eye development in the axolotl; mi (microphthalmic) (Humphrey & Chung, 1977) acts late in eye development, and causes degeneration of already differentiated cells derived from the head ectoderm at hatching time, while e (eyeless) (Humphrey, 1969, 1975; Briggs, 1973) acts early in eye development, suppressing growth of the presumptive eye tissues and some hypothalamic cells (Van Deusen, 1973). Both mutations are recessive. While the first is lethal, the latter produces secondary failures in the pituitary gland and the gonads, thus causing sterility. One purpose of our study has been to describe the normal stages 0 to 14 of axolotl eye development, expanded from Rabl's (1898) stages 1 to 10, and to correlate them with the whole body developmental stages (Harrison, 1969; Bordzilovskaya & Detlaf, 1975; Schreckenberg & Jacobson, 1975). Next we wanted to determine the morphological stages at which eye development was disturbed by the mutations mi and e. In the mutant e the question arose whether the primary failure resides in the presumptive retinal cells themselves or whether perhaps the head epidermis does not produce sufficient amounts of active retina-stimulating factor. We used mutant e-wild-type tissue combinations in organ culture to answer this question. MATERIALS AND METHODS Eye stages Axolotls were mated overnight. They were either wild type or heterozygous for e, r or mi. Spawnings were raised in 10% (v/v) Steinberg's saline. Stages were monitored according to
Axolotl eyeless mutation 153 Bordzilovskaya & Detlaf (1975) and Harrison (1969). At all stages, individuals or their eyes were fixed in Bouin's fixative (Conn etal. 1960) for 2-5 h. To avoid pain, old embryos, larval and adult axolotls were anaesthetized in 0-1 % (w/v) ethane-m-aminobenzoate methanesulphonate (Sigma) prior to fixation. Tissues were dehydrated in a graded ethanol series, amylacetate, toluene, embedded in paraplast and sectioned at 7jum. Rehydrated sections were stained with haematoxylin-eosin (Conn et al. 1960) and mounted in Permount (Fisher). Staging of eye cross sections was according to Rabl (1898) (Fig. 1) as far as they were treated there. Both Harrison's and Rabl's stages had to be extended. In mutants (Figs 2, 3) and organ-cultured eyes, develop- ment was not always uniform, and stages were assigned according to the most developed structures. Bovine cornea extracts Adult bovine corneas were homogenized in 0-005 M-sodium phosphate buffer, pH7-2, and centrifuged at 12000g, 5°C, for 30min. The supernatant was lyophilized. Samples of cornea powder were sterilized in 70% (v/v) ethanol for 15min at 23°C, and dialysed (Spectrum, relative molecular mass cut-off 12-14X103) overnight against deionized water, pH5-l, 5°C. The sterile contents of the dialysis bag were diluted with sterile water to reach 40 % of the desired volume and were mixed with 60 % sterile full-strength Leibovitz L-15 medium (Grand Island) with glutamine and lOOmgml"1 gentamicin sulphate (Sigma). Organ culture Jelly coats of axolotl embryos at late gastrula (Harrison stage 12) were oxidized with 10 mg I"1 KMnO7 in deionized water at 23°C, 1 h. Embryos with their jelly coats were briefly immersed in 70 % (v/v) ethanol using a wide-mouth pipette, and washed in two changes of autoclaved deionized water, pH7-8, and one change of sterile-filtered 50% (v/v) modified amphibian Ringer's solution (100%: 0-lM-NaCl, 0-002M-KC1, 0-002M-CaCl2, 0-001 M-MgSO4, 0-005M- Hepes (Sigma), SOmgl"1 gentamicin sulphate, pH7-8). Under sterile conditions, embryos were dejellied with forceps and transferred to a dish with fresh 50 % modified Ringer's solution. When they had reached the earliest stage of neurulation (Harrison stage 13), the vitelline membrane was removed with forceps. The right body side of the embryo was left intact for phenotypic identification at hatching time. On the left body side the anterolateral neural plate with adhering prechordal mesoderm was excised with tungsten needles to serve as a source of the retina, or presumptive head epidermis was peeled off as a source of the lens (Fig. 4). Any cell type contamination across the neural plate margin resulting from jagged cuts will lead to false results and must be carefully avoided. Excised tissues were transferred using a Pasteur pipette to the central well of plastic organ culture dishes (Falcon) with 60 % (v/v) Leibovitz L-15 culture medium supplemented with glutamine and SOmgl"1 of gentamicin sulphate. The culture dishes were humidified by the surrounding sterile water rings (Fig. 4). Tissues were cultured for 9 days, either as tissue sandwiches, alone as controls, or in the presence of bovine cornea extract. Cultured tissues and donor embryos, the latter raised in 50% modified Ringer's solution, werefixedat hatching age after 9 days, sectioned and stained as described above. After staging according to Rabl (1898), explants were grouped into eye-positive (stages 1 to 9) and eye-negative (stages 0 or —, not identifiable), independently for pigmented retina, neural retina, and lens. Results were also evaluated using a four-table x1 test for independence. RESULTS Stages of wild-type axolotl eye development Stages 1 to 10 are those of Rabl's drawings (1898), stages 0 and 11 to 14 are new, having never been published before. Stages 1 to 10 are described in more detail, especially concerning the retina and accessory tissues. Eye development (25°C) (Fig. 1) is correlated with whole body stages 23 to beyond 48 of Harrison (1969).
154 R. CUNY AND G. M. MALACINSKI
Axolotl eyeless mutation 155
Fig. 1. Stages 0 to 13 (opposite) and 14 (above) of eye development in wild-type
axolotls. Sectioned at 7jUm in paraplast and stained with haematoxylin-eosin (except
stage 13 sectioned at ljum in Araldite and stained with pararosaniline-methylene
blue). See Results for explanation of the annotations and a description of the stages. At
stage 0, the neural fold of the head region is curved downward and the most anterior
portion appears on the ventral side in cross section. Stage 14 is from an axolotl
simultaneously heterozygous for genes e and r. Vision was normal, although several
retinal vesicles (arrows) had formed at the optic stalk. Scale bars, 200jum; 0 to 3,
4 to 10, 11 to 12, 12 to 14 equal magnifications.
Stage 0 (late tail bud, 7 to 11 somites: Harrison stages 23 to 26, 60 to 75 h). Pair of
optic pouches (op) on lateroventral presumptive diencephalon wall. Presumptive
brain ventricle closing at neural folds (nf).
Stage la (branchial swelling with clefts, 11 to 16 somites: Harrison stages 26 to 29,
70 to 85 h). Pair of optic vesicles (ov) with central optocoel (oc) on lateroventral
diencephalon in contact with head epidermis (he). Optocoel continuous with
neurocoel (nc) at optic constriction or presumptive optic stalk.
Stage lb (dorsal fin rudiment, 16 to 20 somites: Harrison stages 29 to 32, 80 to
100 h). Optic vesicles expanded dorsad, optocoel flattened by pressure of brain
and head epidermis. Optic constriction lengthened to form optic stalk (si). Wall of
optic vesicle facing brain flattened to form tapetum nigrum (tn), wall facing head
epidermis thickened to form presumptive neural retina (nr). Head epidermis flat,
two cell layers thick.
Stage 2 (body axis straight, dorsal fin growing: Harrison stages 32 to 36, 90 to
130 h). Planar lens placode (Ip) formed by thickened internal layer of head epi-
dermis, external layer not thickened. Optocoel completely collapsed by thickened
presumptive neural retina, except in lengthened optic stalk.
Stage 3 (branchial branches: Harrison stages 36 to 39,120 to 155 h). Lens placode
contracted to form lens hemisphere (Ih). Outer layer of head epidermis not
thickened. Periphery of flat optic vesicle contracted and presumptive neural retina
concave, engulfing lens hemisphere. Choroid fissure (cf) ventral of optic stalk
wide open. Ventral neural retina rudimentary.156 R. CUNY AND G. M. MALACINSKI
Stage 4 (branchial filaments: Harrison stages 39 to 40, 140 to 190 h). Lens vesicle
(Iv) with small central lentocoel (Ic) still attached to outer layer of head epidermis
or presumptive epicornea. Vitreous chamber (vc) between lens vesicle and neural
retina opened. Papilla nervi optici still indistinct. Embryonic pigment granules and
yolk platelets in all ocular cell types.
Stage 5 (branchial covers united: Harrison stages 40 to 41,180 to 220h). Primary
lens fibres (pf) elongating on retinal side; their pigment granules and yolk
platelets disappearing. Lens vesicle still attached to epicornea (ec). Some new
totally pigmented cells in dorsal tapetum nigrum. Dorsal neural retina with
internal plexiform layer (ip) in central two thirds. Papilla nervi optici (pn) well
developed. Retina ventral of optic stalk half the length of dorsal retina. Nick
between ciliary epithelium (ce) and neural retina at presumptive ora serrata (os).
Space between optic cup and epicornea open.
Stage 6 (branchial branches elongated: Harrison stage 41, 200 to 230 h). Lens
vesicle with primary lens fibres, detached from one-cell-layer thick epicornea.
Lens capsule (c) thickened. Lentocoel maximal. Optic stalk forms sheath around
thin optic fibre tract {pi). Dorsal tapetum nigrum completely pigmented; ventral
tapetum not pigmented. Ganglion cell layer (gc) two-cell-layers thick. Internal
nuclear layer (in) three-cell-layers thick.
Stage 7 (mouth open: Harrison stage 42, 220 to 260 h). Primary lens fibre hillock
large, almost filling lentocoel. Nuclei of primary lens fibres degenerating, baso-
philic. Neural retina with rods (r) and cones. Internal plexiform layer complete in
dorsal retina, external plexiform layer (ep) irregular and thin. Nerve fibre layer
still indistinct. Ciliary epithelium about five cells long in cross section; external
layer pigmented, internal layer not pigmented. Internal limiting membrane (//),
Bruch's membrane (Br), and Bowman's membrane (Bo) thin. Ciliary stromal
fibroblasts, iridophores, and xanthophores present. Epicornea one- to two-cell-
layers thick. Sclera (s) cells on external side of tapetum nigrum. Sclerocorneal
junction (sj) very thin, anterior chamber (ac) between epicornea and lens still
partially open.
Stage8 (hatched: Harrison stages 42 to 43,250 to 330h = 10 to 14 days). Lentocoel
collapsed; lens epithelium (le) in direct contact with lens fibre core. Nuclei in
primary lens fibres lost. Nuclei in secondary lens fibres (sf) degenerating, baso-
philic, except in subequatorial zone. Ventral tapetum nigrum now also completely
pigmented. Ventral retina two thirds as long as dorsal retina when viewed in cross
section. Central portion of ganglion cell layer flattened, one-cell-layer thick.
External plexiform layer distinct. Sclerocorneal junction complete, although still
very thin. Anterior chamber closed.Axolotl eyeless mutation 157 Stage 9 (forelimb stump with up to three digits: Harrison stages 43 to 45, 12 to 23 days). Lens epithelium flattened. Lens fibre orientation spherical. Internal plexi- form layer (ip) complete in ventral retina. Ventral retina as long as dorsal retina in cross section. Nuclei of outer nuclear layer (on) of neural retina elongating and aligned. Iris (i) rudiment one to three cells long in cross section. Stage 10 (shoulder and elbow flexible, hind limb stump: Harrison stages 45 to 46, 20 to 40 days). Lens bigger. Nerve fibre layer (nf) of neural retina distinct. Ganglion cell layer one-cell-layer thick. Nuclei of outer nuclear layer 1-5 times as long as wide. Ora serrata constricted. Ciliary (cs) and iris (is) stromata irregular or missing. Stage 11 (hind limb with three digits, knees may be flexible: Harrison stages 47 to 48, 35 to 80 days). Nuclei in secondary lens fibres largely lost, except in sub- equatorial zone. Ciliary and lens epithelia flattened. Iris still only three cells long in cross section. Epicornea still only one- to two-cell-layers thick. Bowman's and Bruch's membranes thin. Dorsal sclera with orbital cartilage (ca), although scleral cells still sparse and scattered. Canal of Schlemm missing, and choroid (ch) still consisting only of a loose net of capillaries. Stage 12 (larva 8 to 12cm long: Harrison stages ?, 50 to 150 days). Nuclei of lens epithelium at most 1-5 times flatter than high. Secondary lens fibres flattened and oriented spherically. Zonule of Zinn (Zi) weak, enclosing new posterior chamber (pc) between iris and enlarged vitreous chamber (vc). Iris in cross section elongated, about six cells long. Ciliary and iris stromata completely coat ciliary and iris (ie) epithelia on their external sides. Stromal capillaries present. Trabecular mesh (tm) in corneociliary angle present, but canal of Schlemm possibly absent. Capillaries of choroid condensed, one-capillary-layer thick, unpigmented. Tensor choroidae inside of thin sclerocorneal junction indistinct. Bruch's membrane thicker. Retinal internal limiting membrane distinct. Retroorbital muscles (rm) proximal to orbital cartilage and sclera (s). Cartilage still incomplete on ventral sclera. Sclera with three flat cell strata. Epicornea with three cell layers. Bowman's membrane thick, coated on its internal side with three strata of corneal stroma (cos) cells. Eyelids absent, although small rudiment of Harderian gland (Ha) appearing at dorsal presumptive conjunctiva. Stage 13 (juvenile 20 to 25cm long: Harrison stages ?, 300 to 400 days). Lens epithelial nuclei three to four times flatter than wide. Secondary lens fibre cell boundaries indistinct. Lens slightly wider than vitreous chamber. Zonule of Zinn still weak, but more extensive. Nerve fibre layer of neural retina thick near papilla nervi optici. Nuclei of internal nuclear layer three-cell-layers spanning. Rod cells of outer nuclear layer and photoreceptor layer (pr) with rod and cone outer segments three times as long as their nuclei. Cone and possibly double cone cells more internal in these layers. Melanophores frequent in iris and ciliary stromata.
158 R. CUNY AND G. M. MALACINSKI Tensor choroidae muscle (tc) weak. Orbital cartilage complete on ventral sclera, about three-cell-layers thick. External cell layer of epicornea flattened. Corneal stroma five-cell-strata thick. Eyelid (/) rudiment on dorsal side. Harderian gland with few large acini. Stage 14 (adult 30 to 35 cm long: Harrison stages ?, 7 years at 18°C). Lens twice as wide as liquid-filled vitreous chamber. Zonule of Zinn strong, with several in part- torn fibre sheets. Nerve fibre layer thick in most parts of neural retina. Internal layer of ciliary epithelium unpigmented, the external pigmented. Ciliary caecae absent. Choroid still one-capillary-layer thick. Trabecular mesh in corneociliary angle extensive both dorsally and ventrally. Tensor choroidae ventrally strong. Sclera external of cartilage about five-cell-strata thick. Epicornea four-cell-layers thick. Corneal endothelium and Descemet's membrane absent. Dorsal eyelid only slightly enlarged, ventral eyelid missing. Dorsal Harderian gland branched with many acini. Conjunctiva distinct dorsally. Eye development in mutant e (eyeless) In axolotls homozygous for gene e (nonlethal, recessive) retina development is arrested at the early optic vesicle stage (Rabl stage la) (Fig. 2A), although sporadic cell divisions are still observed later. At hatching, an irregular cell mass (abnormal stages lb to 4) (Fig. 2B) represents the neural retina, and the pig- mented retina is represented by a few pigmented cells. In rare cases a pigmented retinal epithelium (stage 6) persists until hatching (Fig. 2C). Small normally shaped lenses (stage 6) (Fig. 2B) may form. They are not stimulated to grow by the retina and are pushed aside by the immigrating head mesenchyme. These lenses may thus become located close to the head epidermis or close to the retinal rudiment (Fig. 2B). Both retinal rudiments and lenses disintegrate and disappear (Fig. 2D) during larval life. In axolotls heterozygous for both e and r (renal in- sufficiency, lethal, recessive) (Humphrey, 1964) eyes develop to stage 14 (Fig. 1); however, often several optic vesicles develop into a stack of retinal cups (Fig. 1), whereby the lens may become dislocated (Epp, 1978). Eye development in mutant mi (microphthalmic) Early eye development in axolotls homozygous for the mutation mi (lethal, recessive) is indistinguishable from that in the wild type almost up to hatching (Rabl stage 8). Cells in the late embryonic to larval eye (Rabl stages 7 to 9), some brain areas, the gill branches and filaments begin to die, and cell growth and mitosis are halted. At that stage the eyes look dwarfed and crippled. The eye cup is often too small to harbour the lens behind the iris (Fig. 3). Cell differentiation is not impeded directly, and lens fibre cells, retinal photoreceptor cells with rod outer segments, retinal pigment epithelial cells, retinal ganglion cells with axons in the optic nerve are all present in reduced numbers. Scleral cartilage, corneal stroma, iris stroma, eyelid and Harderian gland never develop, because the larva
Axolotl eyeless mutation 159 dies before that stage, usually within 1 to 2 weeks after hatching. The larva swims but does not feed. Necessity for mutual head epidermis-neural plate stimulation When wild-type anterolateral neural plate neuroectoderm or head epidermis from stage-13 early neurula were cultured for 9 days in isolation in serum-free medium, they never developed retinas or lenses in several hundred cases examined. Instead, neural plates developed only into brain tissue with nerve tracts, small dense nuclei and still remaining yolk platelets. Head epidermis formed only skin (Figs 5, 6B,C) with intracellular slime vacuoles, large nuclei, and a thick outer cell surface. When those two tissues were cultured together, however, eyes developed in almost all cases (Figs 5, 6A) in addition to brain and Fig. 2. Eye development in vivo in axolotls homozygous for gene e. (A) Retarded optic vesicle (ov) (stage la, but smaller) in a 4-day embryo (Harrison stage 30). By that time, wild-type eyes would reach stage 2. (B) Small lens vesicle (Iv) (stage 5) next to disorganized neural retina (nr) mass (abnormal stage lb to 4) in a 10-day embryo at hatching (Harrison stage 42). Wild-type eyes would be at stage 7. (C) A pair of retinal rudiments with hypothalamus rudiment (hy) in between at time of hatching. Tapetum nigrum (tn) pigmented (stage 6), neural retina (nr) remaining a flat epithelium (resembling stage lb). (D) Eyeless orbital area of 56-day-old larva (Harrison stage 47). A small necrotic retinal cell cluster (arrow) may still be discernible. Wild-type eyes would be at stage 11. Scale bars, 100/im.
160 R. CUNY AND G. M. MALACINSKI
sf
Fig. 3. Eye development in axolotl homozygous for gene mi. Small eye at same stage
as wild-type eyes (stage 9, but smaller) in a 20-day-old larva (Harrison stage 43). Cell
number is reduced by cell death in cornea, lens, retina, brain areas, and gills. The lens
is outside the optic cup, because of the drastic space reduction in the vitreous chamber.
Annotations as in Fig. 1. Scale bars, 100 fim.
Fig. 4. Organ culture design. Embryonic tissues were explanted at the early neural
plate stage (Harrison stage 13) and cultured for 9 days to reach the hatching age
(Harrison stage 42). (A) Left anterolateral neural plate neuroectoderm (dotted)
developed into brain tissue, and (C) left anterolateral epidermis ectoderm (hatched)
developed into skin tissue. When the two tissues were cultured together as a sandwich
(B), and were both derived from wild-type donors, eyes with pigmented and neural
retina and lens developed. The right body side of the donor embryo was left intact to
determine the phenotype at hatching.
skin tissue. Retinas with dense nuclei but without remaining yolk platelets, and
often with rod outer segments, and lenses with a core of enucleated lensfibresboth
reached up to Rabl stage 8. All tissue explants had some mesodermal cells, which
could not be removed. Lens development was never observed in the absence of the
retina, while retina development frequently occurred in the absence of a lens.
It may be concluded that retina development depends on the presence of an
epidermal component, and lens development in turn depends on a retinal factor.
Bovine cornea extracts stimulate axolotl retina development
Embryonic axolotl head epidermis is too small for growth factor extraction.
Therefore, an adult head epidermis derivative, the bovine cornea was used.Axolotl eyeless mutation 161
Neural plates from wild-type axolotls, when cultured in the presence of non-
dialysable bovine cornea extract (3000 jug ml" 1 ), formed neural retinas (stage 6)
preferentially (Figs 7, 8B). At a lower concentration (30/igmr 1 ), the differen-
tiation of pigmented retinas (stage 6) was prevalent (Figs 7, 8A). Overall retina
development is relatively infrequent in the presence of corneal extract. However,
in several hundred cases, retina development was never observed in control
neural plates cultured alone. The results suggest that head epidermis derivatives,
such as the cornea, produce a macromolecular factor which stimulates retinal
differentiation from wild-type neural plates.
Primary failure in mutant e neural plate and retina
In heterozygous spawnings, embryos homozygous for gene e have a frequency
of 1/4, and the probability of an e-e tissue combination in sandwich cultures is
thus only 1/16. In the rare cases where both anterolateral neural plates and head
Neural plate (wild type) + epidermis (wild type)
Neural retina Pigmented retina Lens
St ages Stages Stages
1 2 3 4 5 6 7 8 9 - 1 2 3 4 5 6 7 8 9 - 1 2 3 4 5 6 7 8 9
Total
•f14 Total 13 Total
1
1
22 J 22 7
100%
^^J 1
P< 0-0005
1
95%
II | 1
P< 0-0005
Neural plate (wild type) E p i d e r m i s (wild type)
Neural retina Pigmented retina Lens
Stages Stages Stages
-112 3 4 5 6 7 8 9 -112 3 4 5 6 7 8 9 - 1 1 2 3 4 5 6 7 8 9
Fig. 5. Frequency of eye stages of neural retina, pigmented retina, and lens in explants
of wild-type anterolateral neural plate and, or, anterolateral head epidermis after
9 days of culture. See Fig. 4 for experimental design. Histograms on line 1 exhibit
results with wild-type embryos from a heterozygous spawning (same as in Figs 6, 9),
while those on line 2 are from a pure wild-type spawning (same as in Figs 7, 8). The
vertical lines separate the unidentifiable negative cases (-) on the left from the positive
cases (1 to 9) on the right in each histogram. Stages 1 to 3 are hard to recognize in
culture. The total of cases examined is given on the left, the percentage positive is on
the lower left, and the level of significance by the x1 test of independence appears on
the lower right of the histograms. The low level of significance (P162 R. CUNY AND G. M. MALACINSKI
Fig. 6. Mutant e and wild-type tissue explants after 9 days of culture. (A) Sandwich of
wild-type neural plate and wild-type epidermis showing differentiation of an eye with
pigmented retina (tn) (stage 8), neural retina (nr) (stage 7) with rod outer segments (r),
and a lens with lens epithelium (le) and secondary lensfibres(sf) (stage 8). Brain tissue
in the upper left, skin tissue in the lower right, and some mesodermal cells between eye
and skin. (B) Wild-type neural plate cultured alone developed into brain tissue only.
(C) Wild-type epidermis cultured alone formed skin. A few mesodermal cells are
apparent in the centre. (D) Sandwich of wild-type neural plate and gene e epidermis
with pigmented retina (tn) (stage 6), neural retina (nr) (stage 6) and lens vesicle (Iv)
(stage 4). Brain tissue just above pigmented retina and skin in the upper left area.
(E) Sandwich of gene e neural plate and wild-type epidermis without eye tissues, but
with brain tissue in upper right, skin in lower left, and some mesodermal cells in the
centre. (F) Sandwich of gene e neural plate and gene e epidermis. Eye tissues wanting,
brain tissue on the left, and skin tissue on the right. Scale bars, 100/mi.Axolotl eyeless mutation 163
epidermis were derived from donors homozygous for gene e, only brain and skin
tissues differentiated (Figs 6F, 9). This same result was obtained in several
repetitions of this experiment. All combinations of wild-type neural plates with
homozygous gene e epidermis developed a retina, reaching up to Rabl stage 8.
However, lenses formed less frequently, reaching stage 4 (Figs 6D, 9). When
Neural plate (wild type) + 3000/^g ml"1 aqueous bovine cornea extract
Neural retina Pigmented retina
Stages Stages
1 2 3 4 5 6 7 i* 9 - 1 2 3 4 5 6 7 8 9
—1
Total Total
171
1
19 15 19
1 21%
1
•
2
••
1
P164 R. CUNY AND G. M. MALACINSKI
Neural plate (wild type) + epidermis (eyeless)
Neural retina Pigmented retina Lens
Stages Stages Stages
- 1 2 3 4 5 6 7 8 9 i 2 3 4 b 6 7 t{ 9
-
Total 1
Ii 1
••
1
i ,
100% PAxolotl eyeless mutation 165 focus on the eye. As eyes of other anamniotic vertebrates, the axolotl eye lacks a retinal fovea and supraretinal blood vessels. In urodeles, the corneal endothelium and Descemet's membrane are absent. The lens is spherical and the retinal photoreceptors are more diverse and complex (Reyer, 1977) than those of mammals. In the axolotl, the choroid remains very thin and unpigmented, while in newts (genus Notophthalmus) it is pigmented and several capillary layers thick. Neither anurans nor urodeles have a vitreous body in their vitreous chamber. In the axolotl, a scleral cartilage protects the protruding eyeball, while adult newts have no cartilage in their sclera. On the other hand, newts have well-developed dorsal and ventral eyelids, while axolotls only have a dorsal eyelid rudiment. Nictitating membranes, while present in frogs (genus Rand), are absent in urodeles. Embryonic eye tissues are laden with yolk platelets in lungfish, urodeles and anurans, while they are devoid of yolk platelets in mammals and birds. The mutation ml (Humphrey & Chung, 1977) is expressed late during axolotl eye development (stage 8), and also affects other head tissues. In contrast, mutation e is already expressed at the beginning of eye development (stage la) in the optic vesicle and apparently in some cells of the hypothalamus (Van Deusen, 1971, 1973). Disorganized mitotic divisions still continue at a reduced rate (Ulshafer & Hibbard, 1976,1979) in later stages, and some pigmented retinal cells may differentiate (Van Deusen, 1973) in rudimentary optic vesicle cell masses (Brun, 1983,1985). A very similar eyeless mutation is also known in the rat, except that homozygous animals are fertile (Kinney et al. 1982). Van Deusen (1973) proposed that in the axolotl e mutation the retina- and hypothalamus-forming neuroectoderm regions have an intrinsic defect. If this hypothesis is correct, the insufficient production of retina-derived growth factors would arrest lens development as a direct consequence (Cuny & Malacinski, 1983). Other accessory tissues, such as cornea, scleral cartilage, Harderian gland, and sclera, never develop in the mutant e, probably also because they would depend on the retinal growth factors. While the preoptic nuclei in the brain are histologically normal (Eagleson & Malacinski, 1983) in the mutant e, the super- ficial optic layer of the tectal neuropil is hardly developed (Gruberg & Harris, 1981; Harris, 1982, 1983). The mutant e tectum develops normally, however, and full vision is restored (Epp, 1972; Hibbard, Ulshafer & Ornberg, 1975; Hibbard & Ornberg, 1976; Schwenk & Hibbard, 1977; Harris, 1981), if a wild-type eye is implanted in embryos homozygous for gene e. This demonstrates that the mutant e optic tectum is not genetically defective, and only suffers from a lack of retinal innervation. Implantation of a wild-type eye causes skin melanocytes to contract in white (d) strain (Keller, Lofberg & Spieth, 1982) homozygous carriers of mutation e, while parabiosis causes a spreading of melanocytes in d strain axolotls, if they are connected with an axolotl homozygous for gene e (Epp, 1972). These data suggest that the hypothalamic melanocyte-inhibiting factor is not secreted in sufficient amounts, which causes an overproduction of pituitary melanocyte-stimulating hormone (MSH) (Epp, 1972,1978). The pineal gland (Brick, 1962) and the retinal
166 R. CUNY AND G. M. MALACINSKI photoreceptors (Gern, Owens & Ralph, 1978) both secrete melatonin, which also stimulates skin melanocyte contraction. At any rate, the effect on skin pigmentation of mutant e seems to be a direct consequence of the eyeless condition. The sterility (Humphrey, 1969) of axolotls homozygous for gene e is apparently due to insufficient secretion (Van Deusen, 1973) of hypothalamic gonadotropin- releasing hormone (Kubo, Watanabe, Ibata & Sano, 1979). While the implan- tation of a wild-type eye has no effect on mutant gonads, the implantation of a wild-type hypothalamus region in a gene e embryo restores fertility (Van Deusen, 1973; Harris, 1983). The hypothalamic cells develop adjacent to the retinal cells in the neural plate, and may thus belong to the same stem cell lineage. From the presented evidence one may assume that all pleiotropic effects of the mutation e are consequences of the primary genetic defect in the retina area, with perhaps the exception of some hypothalamic cells. It remained still unclear, however, whether the retinal tissue does not develop, because the interacting tissues in the surroundings of the neural plate cells do not produce enough retina-stimulating growth factors. Tissue-grafting experiments (Van Deusen, 1971, 1973) between homozygous e and wild-type gastrulae have shown that the mutant e prechordal mesoderm can sustain normal eye develop- ment in wild-type embryos, while wild-type prechordal mesoderm cannot restore eye development in mutant e embryos. Hence, the putative mesoderm-derived neuralizing factor (Tiedemann & Born, 1978) is probably normal in the mutant. We found that in the mutant e paired optic vesicle rudiments always formed, which made contact with the head epidermis (Fig. 2A). This rules out a primary barrier function of the neural crest head mesenchyme in this mutant. Brun (1978, 1980) suggested that retina development might fail, because of a failure in the over- lying mutant e head epidermis. It is known that retina development from cultured wild-type neural plate tissue depends on the presence of head epidermis (Filatow, 1926; Rollhauser-Terhorst, 1981). The nature of this tissue interaction has been unknown however. Understanding the nature of this tissue interaction is crucial for the investigation of the failure in the mutant e retinal region. Unfortunately, embryonic axolotl head epidermis is too small for biochemical extractions, and we therefore resorted to extracts of adult bovine corneas, which are head epidermis derivatives. Dialysed cornea extracts stimulate development of pigmented or neural retinas, depending on the extract concentration, in cultured wild-type neural plates. In the embryo, the neural retina directly faces the head epidermis and thus receives a higher concentration than the pigmented retina behind it. The results suggest that the head epidermis stimulates retina development via a macro- molecular growth factor. In our cultures, mutant e head epidermis was fully capable of wild-type retina stimulation, while mutant e neural plates hardly developed retinas, even in the presence of wild-type epidermis. This demonstrates that the mutant e neural plate and retina are intrinsically defective, and cannot be rescued by the head-epidermis-derived growth factor.
Axolotl eyeless mutation 167
Knowing that the presumptive retina cells are the primary target of the mutation
e, we would like to know which molecular system is disrupted in these cells. At this
point we can only offer some speculations. Ulshafer & Hibbard (1979) argued that
the observed premature thickening and maturation of the basal lamina, that
surrounds the mutant e optic vesicles, could suppress eye development. The
composition of basal laminae modifies the cellular responsiveness to growth
factors (Gospodarowicz, Gonzales & Fuji, 1983). Basal laminae can act as filters
or adsorbants for macromolecules. Wild-type retina development can also be
prevented by intracellular injection of antibodies directed against gap junction
proteins (Warner, Guthrie & Gilula, 1982). Therefore, a reduction or mal-
functioning of gap junctions in the mutant e presumptive retinal region could be
the primary reason for the eyeless condition. A third possibility would be that a
shift in embryonic to retina-specific gene expression takes place during retina
development, involving two isoforms of proteins needed for cell proliferation. If
the retina-specific isoform is malfunctioning or unstable, this could lead to a
decrease in mitotic activity of the retina, and eventually to eyelessness. These
ideas may help to design experiments that will elucidate the primary cause of
eyelessness in the mutant e of the axolotl.
We thank Franchise Briggs and Leah Dvorak (Indiana University Axolotl Colony) for the
heterozygous spawnings of gene e axolotls. Herve Coet helped with the photo-reproduction.
Support for these studies was provided by the Swiss National Fund for Scientific Research and
the US National Science Foundation PC 83-15899.
REFERENCES
ADELMANN, H. B. (1929). Experimental studies on the development of the eye. I. The effect of
the removal of median and lateral areas of the anterior end of the urodelan neural plate on the
development of the eyes (Triton teniatus and Amblystoma punctatum). J. exp. Zool. 54,
249-290.
ADELMANN, H. B. (1937). Experimental studies on the development of the eye. IV. The effect of
the partial and complete excision of the prechordal substrate on the development of the eyes
of Amblystoma punctatum. J. exp. Zool. 75, 199-237.
ARRUTI, C. & COURTOIS, Y. (1978). Morphological changes and growth stimulation of bovine
epithelial lens cells by a retinal extract in vitro. Expl Cell Res. 117, 283-292.
BEEBE, D. C , FEAGANS, D. E. & JEBENS, H. A. H. (1980). Lentropin: a factor in vitreous humor
which promotes lens fiber cell differentiation (chicken embryo). Proc. natn. Acad. Sci. U.S.A.
77, 490-493.
BORDZILOVSKAYA, N. P. & DETLAF, T. A. (1975). XVI. Aksolotl Ambystomamexicanum Cope. In
Obzhekti biologyi razvitiya (ed. T. A. Detlaf, A. E. Gaysinovich, V. Ya. Brodskyi, A. P.
Diedan, G. V. Lopashov & B. P. Tokin), pp. 370-391. Akademiya Nauk SSSR Moskva,
Problemi biologyi razvitiya (ed. V. Astaurov), 579 p. (Translation: Axolotl News Letter 7,
2-22, 1979).
BRAHMA, S. K. & BOURS, J. (1976). Thin-layer isoelectric focusing of soluble and insoluble lens
extracts from cataractous and normal Mexican axolotl {Ambystoma mexicanum). Expl Eye
Res. 23, 57-63.
BRAHMA, S. K. & MCDEVITT, D. S. (1974). Ontogeny and localization of gamma-crystallins in
Rana temporaria, Ambystoma mexicanum, and Pleurodeles waltlii normal lens development.
Expl Eye Res. 19, 379-387.
BRICK, I. (1962). Relationship of the pineal to the pituitary-melanophore effective system in
Ambystoma opacum. Anat. Rec. 142, 229 (Abstract).168 R. CUNY AND G. M. MALACINSKI BRIGGS,R. (1973). Developmental genetics of the axolotl. In Genetic Mechanisms of Development (ed. F. H. Ruddle), pp. 169-199. Soc. devl. Biol. Symp. 31. New York: Academic Press. BRUN, R. B. (1978). Experimental analysis of the eyeless mutant in the Mexican axolotl (Ambystoma mexicanum). Am. Zool. 18, 273-279. BRUN, R. B. (1980). Eye formation in the eyeless mutant axolotl (Ambystoma mexicanum). Am. Zool. 20, 838 (Abstract 602). BRUN, R. B. (1981). The movement of the prospective eye vesicles from the neural plate into the neural fold in Ambystoma mexicanum and Xenopus laevis. Devi Biol. 88, 192-200. BRUN, R. B. (1983). The eyeless mutant axolotl (Ambystoma mexicanum) produces small and delayed optic vesicles. Am. Zool. 23, 998 (Abstract 716). BRUN, R. B. (1985). The eyeless gene e in the Mexican salamander (Ambystoma mexicanum) might interfere with a head gradient. Cell Differentiation 16 (Suppl), 42S (Abstract 100). CONN, H. J.,' DARROW, M. A., EMMEL, V. M. et al. (1960). Staining Procedures used by the Biological Stain Commission, 2nd edn, 289pp. Biological Stain Commission, University of Rochester, Medical Centre, New York. Baltimore: Williams & Wilkins. COURTY, J., LORET, C , MOENNER, M., CHEVALLIER, B., LAGENTE, O., COURTOIS, Y. & BARRTTAULT, D. (1985). Bovine retina contains three growth factor activities with different affinity to heparin: eye derived growth factor I, II, III. Biochimie 67, 265-269. CUNY, R. & MALACINSKI, G. M. (1983). Gene e (eyeless) in the axolotl decreases growth ability of cultured embryonic retina and lens. Can. Fed. Biol. Soc. Proc. 26,117 (Abstract PA-307). DETWILER, S. R. & VAN DYKE, R. H. (1954). Further experimental observations on retinal induction. /. exp. Zool. 126, 135-156. EAGLESON, G. W. & MALACINSKI, G. M. (1983). Development of the preoptic neurosecretory centers in the eyeless mutant axolotl (Ambystoma mexicanum). Am. Zool. 22, 929 (Abstract 443). EPP, L. G. (1972). Development of pigmentation in the eyeless mutant of the Mexican axolotl, Ambystoma mexicanum, Shaw. /. exp. Zool. 181,169-180. EPP, L. G. (1978). A review of the eyeless mutant in the Mexican axolotl. Am. Zool. 18, 267-272. FILATOW, D. (1926). Uber die Entwicklung des Augenkeimes einiger Amphibien in vitro. Wilhelm Roux' Arch. EntwMech. Org. 107, 575-582. FISCHEL, A. (1921). Uber normale und abnorme Entwicklung des Auges I. Uber Art und Ort der ersten Augenanlage sowie uber die formale und kausale Genese der Cyklopie. II. Zur Entwicklungsmechanik der Linse. Wilhelm Roux' Arch. EntwMech. Org. 49, 383-462. FROST, S. K., EPP, L. G. & ROBINSON, S. J. (1984). The pigmentary system of developing axolotls. II. An analysis of the melanoid phenotype. /. Embryol. exp. Morph. 81,127-142. GERN, W. A., OWENS, D. W. & RALPH, C. L. (1978). The synthesis of melatonin by the trout retina. J. exp. Zool. 206, 263-270. GOSPODAROWICZ, D., GONZALES, R. & FUJI, I. (1983). Are factors originating from serum, plasma or cultured cells involved in the growth-promoting effect of the extracellular matrix produced by cultured bovine corneal endothelial cells? /. Cell Physiol. 114, 191-202. GRUBERG, E. & HARRIS, W. A. (1981). The serotonergic somatosensory projection to the tectum of normal and eyeless salamanders. /. Morph. 170, 55-69. HARRIS, W. A. (1981). The transplantation of eyes to genetically eyeless salamanders. Soc. Neurosci. Symp. 7, 543 (Abstract). HARRIS, W. A. (1982). The transplantation of eyes to genetically eyeless salamanders: visual projections and somatosensory interactions. /. Neurosci. 2, 339-353. HARRIS, W. A. (1983). The eyeless axolotl: experimental embryogenetics and the development of the nervous system. Trends in Neurosciences 6, 505-510. HARRISON, R. G. (1969). Harrison stages and description of the normal development of the spotted salamander, Ambystoma punctatum (Linn.). In Organization and Development of the Embryo (ed. S. Wilens), pp. 44-66. New Haven, CT: Yale University Press. HIBBARD, E. & ORNBERG, R. L. (1976). Restoration of vision in genetically eyeless axolotls (Ambystoma mexicanum). ExplNeurol. 50,113-123. HIBBARD, E., ULSHAFER, R. J. & ORNBERG, R. L. (1975). Establishment of tectal connections by optic nerve fibers from grafted eyes in eyeless mutants of axolotls (Ambystoma mexicanum). Anat. Rec. 181, 375 (Abstract).
Axolotl eyeless mutation 169
HUMPHREY, R. R. (1964). Genetic and experimental studies on a lethal factor (r) in the axolotl
which induces abnormalities in the renal system and other organs. /. exp. Zool. 155,139-150.
HUMPHREY, R. R. (1969). A recently discovered mutant, "eyeless", in the Mexican axolotl
{Ambystoma mexicanum). Anat. Rec. 163, 306 (Abstract).
HUMPHREY, R. R. (1975). The axolotl, Ambystoma mexicanum. In Handbook of Genetics, vol. 4
(ed. R. C. King), pp. 3-17. New York: Plenum Press.
HUMPHREY, R. R. & CHUNG, H. M. (1977). Genetic and experimental studies on three
associated mutant genes in the Mexican axolotl: st (for stasis), mi (for microphthalmic) and h
(for hand lethal). /. exp. Zool. 202, 195-202.
JACOBSON, C O . (1962). Cell migration in the neural plate and the progress of neurulation in the
axolotl larva. Zool. Bidrag (Zoon) Uppsala 35, 433-449.
JACOBSON, M. & HIROSE, G. (1978). Origin of the retina from both sides of the embryonic brain: a
contribution to the problem of crossing at the optic chiasma. Science 202, 637-639.
KARKINEN-JAASKELAINEN, M. (1978). Transfilter lens induction in avian embryo. Differentiation 12,
31-37.
KELLER, R. E., LOFBERG, J. & SPIETH, J. (1982). Neural crest cell behavior in white and dark
embryos of Ambystoma mexicanum: epidermal inhibition of pigment cell migration in the
white axolotl. Devi Biol. 89, 179-195.
KINNEY, H. C , KLINWORTH, G. K., LESIEWICZ, J., GOLDSMITH, L. A. & WILKENING, B. (1982).
Congenital cystic microphthalmia and consequent anophthalmia in the rat: a study in
abnormal ocular morphogenesis. Teratology 26, 203-212.
KUBO, S., WATANABE, K., IBATA, Y. & SANO, Y. (1979). Luteinizing hormone releasing hormone
neuron system of the newt (Cynopspyrrhogaster) by immunohistochemical study. Arch, histol.
Jap. 42, 235-242.
LE CRON, W. L. (1907). Experiments on the origin and differentiation of the lens in Amblystoma.
Am. J. Anat. 6, 245-256.
LIEDKE, K. B. (1955). Studies on lens induction in Amblystoma punctatum. J. exp. Zool. 130,
353-380.
MALACINSKI, G. M. (1978). The Mexican axolotl, Ambystoma mexicanum: its biology and
developmental genetics, and its autonomous cell-lethal genes. Am. Zool. 18, 195-206.
MALACINSKI, G. M. & BROTHERS, A. J. (1974). Mutant genes in the Mexican axolotl. Science 184,
1142-1147.
MANCHOT, E. (1929). Abgrenzung des Augenmaterials und anderer Teilbezirke in der
Medullarplatte; die Teilbewegungen wahrend der Auffaltung (Farbmarkierungsversuche an
Keimen von Urodelen). Wilhelm Roux' Arch. EntwMech. Org. 116, 689-708.
MANGOLD, O. (1929). Experimente zur Analyse der Determination und Induktion der
Medullarplatte. Wilhelm Roux' Arch. EntwMech. Org. 117, 586-696.
PAPERMASTER, D. S. & SCHNEIDER, B. G. (1982). Biosynthesis and morphogenesis of outer
segment membranes in vertebrate photoreceptor cells. In Cell Biology of the Eye, ch. 10 (ed.
D. S. McDevitt), pp. 475-531. New York: Cell Biology Series, Academic Press.
RABL, C. (1898). Uber den Bau und die Entwicklung der Linse. I. Theil (Selachier und
Amphibien). Z. wiss. Zool. 63, 496-572.
REYER, R. W. (1977). The amphibian eye: development and regeneration. In Handbook of
Sensory Physiology, vol. VII (5). The Visual System in Vertebrates (ed. F. Crescitelli),
pp. 309-390. Berlin: Springer-Verlag.
ROLLHAUSER-TERHORST, J. (1981). Artificial neural induction in amphibia. 3. Retina formation
and autoorganization in single and double layer Triturus explants. Anat. Embryol. 162, 69-80.
SCHRECKENBERG, G. M. & JACOBSON, A. G. (1975). Normal stages of the axolotl, Ambystoma
mexicanum. Devi Biol. 42, 391-400.
SCHWENK, G. C. & HIBBARD, E. (1977). An autoradiographic study of optic fiber projections
from eye grafts in eyeless mutant axolotls. Expl Neurol. 55, 498-503.
SPEMANN, H. (1938, reprinted 1962). Embryonic Development and Induction. New Haven, CT
(Hafer NY): Yale University Press.
TIEDEMANN, H. & BORN, J. (1978). Biological activities of vegetalizing and neuralizing inducing
factors after binding to BAC-cellulose and CNBr-Sepharose. Wilhelm Roux' Arch, devl Biol.
184, 285-299.170 R. CUNY AND G. M. MALACINSKI
ULSHAFER, R. J. & HIBBARD, E. (1976). Morphology of the optic rudiment in eyed and eyeless
axolotls. Anat. Rec. 184, 552 (Abstract).
ULSHAFER, R. J. & HIBBARD, E. (1979). An SEM and TEM study of suppression of eye
development in eyeless mutant axolotls. Anat. Embryol. 156, 29-35.
VAN DEUSEN, E. B. (1971). Studies on embryonic determination: effects of gene "e" on
forebrain ectoderm in the axolotl. Am. Zool. 11, 678A (Abstract).
VAN DEUSEN, E. B. (1973). Experimental studies on a mutant gene (e) preventing the
differentiation of eye and normal hypothalamus primordia in the axolotl. Devi Biol. 34,
135-158.
WARNER, A. E., GUTHRIE, S. C. & GILULA, N. B. (1984). Antibodies to gap-junctional protein
selectively disrupt junctional communication in the early amphibian embryo. Nature, Lond.
311, 127-131.
WOERDEMAN, M. W. (1929). Experimented Untersuchungen iiber Lage und Bau der
augenbildenden Bezirke in der MeduUarplatte beim Axolotl. Wilhelm Roux' Arch. EntwMech.
Org. 116, 220-241.
(Accepted 6 February 1986)You can also read
