TERATOLOGY 42:171-182 (1990)
Delayed Neural Crest Cell Emigration From Sp and Spd Mouse Neural Tube Explants CONNIE E. MOASE AND DAPHNE G. TRASLER Department of Biology and Centre for Human Genetics, McGill University, Montreal, Quebec, Canada H3A 1B1
ABSTRACT Splotch (Sp) and splotch-delayed (Spd)are allelic mutations on chromosome 1 of the mouse. Embryos homozygous for either allele have neural tube defects (NTDs) and deficiencies in neural crest cell (NCC) derived structures. The fact that Spd mouse mutants sometimes have deficiencies in NCC derivatives in the absence of an NTD led to the hypothesis that neurulation and the release of NCCs may depend on a regulatory event that is common to both processes. Therefore, it may be possible to understand the cause of NTDs in these mutants by examining the basis of aberrant NCC derivatives. Caudal neural tubes were excised from day 9 Sp and Spd embryos and placed into gelatin-coated tissue culture dishes, or 3-dimensional basement membrane matrigel, and cultured for 72 hours. A cytogenetic marker was used to genotype the embryos, In planar cultures, no morphological differences were observed between NCCs from neural tube explants of Spd mutants compared to those from heterozygous or wild-type embryos. However, there appeared to be a delay in the release of NCCs from the neural tube in both Sp and Spd mutants, which was particularly evident in Sp. After 24 hours in culture, the extent of NCC outgrowth, as well as the number of NCCs emigrating from explanted neural tubes, was significantly lower in Sp and Spd mutant cultures than in controls. No differences were observed in the mitotic indices among cells which had emigrated. By 72 hours, mutant cultures and their non-mutant counterparts were similar in terms of outgrowth, cell number, and migratory capability. After 24 hours in 3-dimensional basement membrane matrigel, cell outgrowth from Sp explants was also significantly less than controls. The pattern of NCC outgrowth in both types of culture conditions indicates a 24 hour delay in mutant cultures compared to controls. This stems from a delay in the release of NCCs from the neural tube, suggesting that the defect lies within the neuroeptihelium with respect to the release of NCCs. Neurulation and the release of neural crest cells (NCCs) from the neural tube, are two fundamental processes which occur either concomitantly or in close succession during the earlier stages of vertebrate development. Examination of mutants with both neural tube closure and NCC emigration defects may reveal mechanisms on which both these processes depend for normal development. Splotch (Sp) and splotchdelayed (Spd) are allelic mutations located on chromosome 1 in the mouse. Embryos that are homozygous for either of these alleles have neural tube defects (NTDs), as well as deficiencies in specific NCC deriva0 1990 WILEY-LISS, INC
tives including spinal ganglia and melanocytes (Auerbach, '54; Moase and Trasler, '89). However, these defects are more severe in Sp than in Spd mutants. Heterozygotes rarely develop a n NTD and have deficiencies in NCC derivatives that are less severe. These include a shortage of melanocytes that results in a white belly patch and white extremities in the adult, as well as reduced spinal ganglia volume in the lumbosacral region of heterozygous embryos (Moase and "rasler, '89). Received October 20, 1989; accepted January 30, 1990.
172
C.E. MOASE AND D.G. TRASLER
The use of a biochemical marker to distinguish homozygous Sp or Spd mutants from their heterozygous and wild-type littermates has revealed the presence of a few embryos (11%) with the mutant genotype that fail to display an NTD (Moase and Trasler, '89). However, deficiencies in NCC derived structures were present in all mutants. This, in conjunction with other evidence (Kapron-Bras and Trasler, '88),led to the hypothesis that there is a regulatory event during neurulation that is necessary for both neural tube closure and the release of NCCs from the neural tube. Elucidation of the mechanism responsible for deficient NCC derivatives could, in turn, reveal the cause of NTDs in these mutants. Histological studies have shown that in Sp homozygotes, a t the time of NCC emigration from the neural tube, the surface ectoderm is in tight apposition with the underlying neural tube in the trunk region of the embryo where the neural tube has managed to close (Kapron-Bras and Trasler, '88). This was in contrast to the normal situation where a n extracellular matrix (ECMI-filled space separated the two tissues, thus providing room for the emigrating NCCs to move out from the neural tube and commence migration. In addition, Kapron-Bras and Trasler observed a significant reduction in the number of NCCs that migrated from the Sp mutant neural tube. However, it was unclear whether the NCCs themselves were at fault; whether the abnormality resided in the neuroepithelium from which they arose, or in the environment in which they were released and migrated. It is unknown what triggers the release and subsequent migration of NCCs, although a specified set of conditions must exist in order for this to occur (Newgreen and Gibbins, '82; Thiery et al., '85). Individual analyses of some of these factors have added to the confusion. Lofberg et al. ('85), using ECM-adsorbed microcarriers, have demonstrated that NCCs can be stimulated to migrate prematurely. However, Morriss-Kay et al. ('86) report th at NCC migration still occurs in areas where the ECM has been depleted. Others maintain that the cellular adhesion system plays a key role (Innes, '85; Martins-Green and Erickson, '87). This study was designed to monitor the release and subsequent migration of NCCs from Sp and Spd mutant and non-mutant (i.e., heterozygous and wild-type) neural
tubes in vitro. Our aim was to determine whether the effect of the mutant gene was intrinsic or extrinsic to the neural tube from which NCCs are derived. It was also of interest to determine whether both Sp and Spd mutants would provide similar results in vitro, or differ in accordance with the differences they exhibit in vivo. The results indicate that there is a n initial delay in the release of NCCs from mutant neural tubes explanted onto a flat gelatin-coated substrate or into a 3-dimensional matrix containing basement membrane components, with Sp being more severely affected than Spd.However, once mutant NCCs have been released, their subsequent migration appears to occur normally in vitro. This suggests that the faulty mechanism resides within the neuroepithelium with respect to its ability to release NCCs at the appropriate time in development, rather than in the environment external to the neural tube. MATERIALS AND METHODS
Mice The splotch and splotch-delayed stocks were obtained in 1979 (Jackson Laboratory, Bar Harbor, ME), and have since been maintained in our laboratory by brothersister matings. The inversion marker stock (In(1)lRk) was obtained in 1984 from Dr. T.H. Roderick and is currently maintained through random mating. The breeding scheme used to incorporate the In(1)lRk marker into the mutant lines has been published elsewhere (Moase and Trasler, '87). Briefly, mice heterozygous for the mutant allele were crossed to In(1)lRk mice, which are homozygous for a n inverted segment on chromosome 1. This inverted segment encompasses the Sp (or Spd) locus but carries the wild-type allele. Double heterozygotes (carrying the mutant allele and the inversion) were then intercrossed to produce a n F, with three possible genotypes, which are distinguishable cytogenetically (see below). Mice were housed in plastic cages with wood-chip bedding and fed Purina Lab Chow and tap water ad libitum. A 12 hour darkllight cycle was maintained with a con1°C. One stant room temperature of 22 male was placed in a cage overnight with several females and, in the presence of a vaginal plug, the following day was designated as day 0 of gestation. Females were sacrificed on day 9.5 of gestation, and the uterus transferred to sterile Hanks' bal-
*
DELAYED EMIGRATION OF MUTANT NCCs
173
culture 1 % trypsin
4 OC
20 min
day 9 embryo
remove: -ectoderm -mesoderm -somites
\
\
cytogenetics
\
Fig. 1. Schematic representation of caudal neural tube isolation from day 9.5 embryos, as well as a chromosomal spread prepared from a n Spd heterozygote. G-
anced salt solution (HBSS). Further manipulations were conducted in a sterile flow cabinet. Embryos were removed from the uterus with watchmaker's forceps and transferred to fresh HBSS for the preparations described below.
banded chromosomes show the pattern associated with Sp or Spd gene carriers (a), or carriers of the wild-type allele on the inverted chromosomal segment (b).
dried slides (aged 4-7 days) were treated with 0.02% trypsin in 0.15M NaCl for 12-16 sec at 37°C and processed according to standard G-banding techniques (Davisson and Roderick, '73). Slides were scanned on a Leitz Laborlux K microscope, and a minimum of three metaphase spreads showing a conCytogenetics sistent genotype were scored per embryo. A Cytogenetic preparations were based on particular G-banding pattern on chromothe methods described by Evans et al. ('72). some 1 indicates the presence or absence of The anterior portion of each day 9 embryo, the Sp (or Spd) allele. Chromosome 1 is the including the amnion and embryonic yolk largest chromosome in the mouse karyotype sac, was placed in a separate 35 mm petri and, with the aid of G-banding, is easily disdish containing RPMI 1640 (Flow Laborato- tinguishable in a chromosomal spread. The ries, McLean, VA) with 5%fetal calf serum two heavy bands close together on the distal (FCS) plus 1%colcimid (GIBCO Laborato- part of chromosome 1 indicate that the Sp ries, Grand Island, NY), incubated for 1.5 allele is present (Fig. la). However, a 180" hours a t 37"C, transferred to 0.5% KC1 (15 rotation of the proximal band indicates that min at 37"C), and then fixed (3:l methanol: the wild-type allele from the inverted segacetic acid) for a minimum of 45 min a t 4°C. ment is present (Fig. lb). Thus the presence The tissue was dissociated in 60% acetic of both banding patterns in a single chroacid for 4 min and cell suspensions were mosomal spread typifies a heterozygous emdropped onto glass slides heated to 37°C. Air bryo.
174
C.E. MOASE AND D.G. TRASLER
Neural tube explants Primary neural tube explants were prepared and cultured according to the methods outlined by Ito and Takeuchi ('84) and Jaenisch ('851, slightly modified. The caudal portion of day 9 embryos (14-28 somite pairs) containing the pre-migratory NCC region was excised and placed in 1% trypsin (GIBCO, 1:250) in HEPES buffer (Flow) at 4°C for 20 min, a s illustrated in Figure 1. (Each genotype was represented by a n even distribution of embryos at various developmental stages.) The tissue was rinsed twice in HEPES containing 10%FCS, transferred to Dulbecco's modified Eagle's medium (DMEM),and dissected free of the surrounding tissues by using sharpened tungsten needles. A Wild M7A stereomicroscope was used to ascertain the complete removal of somites, notochord, mesoderm, and ectoderma1 tissues. The neural tubes were transferred to 35 mm plastic tissue culture dishes that had been coated with 0.5% gelatin, and cultured in 1.5 ml of DMEM supplemented with 5% FCS and 5% horse serum for 72 hours a t 37°C in a humidified atmosphere of 5% co,. Neural tubes for explantation into a 3dimensional basement membrane matrix were prepared a s above, with the addition of a 1 hour pre-incubation of the explanted neural tubes in DMEM containing 10% FCS prior to transfer into gelling matrix (Bilozur and Hay, '88). For this experiment, only Sp mutant cultures, and not Spd, were compared to their heterozygous and wild-type counter arts. Basement membrane matrigelT' (Collaborative Research, Bedford, MA) was thawed on ice, diluted 2:l with DMEM, and dispensed into microtiter plates just prior to the addition of individual neural tube explants. Neural tubes were suspended in the matrigel with the dorsal surface upwards and the dishes were immediately placed into the incubator to accelerate the gelling process. Once the matrix had gelled completely, unsupplemented DMEM was carefully dispensed into each well, and the neural tubes cultured for 72 hours. Visualization of mitotic figures A separate group of splotch and splotchdelayed explants, which had been cultured for 24 or 48 hours on gelatin-coated tissue culture dishes, were carefully rinsed with phosphate buffered saline, fixed in ethanol/
acetic acid (3:l) (Maxwell, '761, and processed by using the Feulgen staining reaction (Humason, '79). An ocular grid was used to determine the percentage of mitotic figures per given area encompassed by the grid. At least three separate regions around each explant were counted and these were subsequently averaged. All counts were performed before the embryos were genotyped. No attempt was made to assess the mitotic indices of 3-dimensional cultures.
Data collection and analysis Each explant was examined and photographed daily over the 72 hour period by using either Zeiss or Olympus phase contrast optics at 100 or 200 x . The front of maximum NCC outgrowth was marked directly onto contact prints of phase-contrast images from 24, 48, and 72 hour cultures. The average distance from the edge of the explant to the maximum front of migrating NCCs was calculated for each culture from 2-3 measurements at each time period. In addition, for explants cultured on flat substrates, the average number of cells within the regions of NCC migration was determined a t each time point by counting cells with a n ocular grid. The genotype of the explant was not revealed until after all measurements had been completed. Statistical analyses were carried out by using the Student's t-test, and the G-test of independence was used to analyze mitotic indices (Sokal and Rohlf, '81). RESULTS
NCC Morphology in planar cultures After 24 hours, early NCC outgrowth from control (wild-type) neural tubes explanted on a flat substrate consists of confluent monolayers with refractile halos surrounding each cell. Cells a t the periphery of the outgrowth are slightly larger. They begin to send out cellular projections and flatten against the substrate as they commence separation from one another. In 48 hour cultures, the NCCs continue to move outward. They become larger and more defined a s they flatten themselves against the substrate, and increase their multipolar projections in both number and length. By 72 hours, the rate of outgrowth begins to decrease and the outermost NCCs have reached their maximum size. At this point, the majority of cells have separated from one another. However, they still remain in
DELAYED EMIGRATION OF MUTANT NCCs
24hr
SPd/+
+ I+
175
cells derived from all types of cultures showed multipolar extensions, particularly those at the periphery of the outgrowth. At 48 and 72 hours, cultures derived from Spd mutant, heterozygous, and wild-type explants were morphologically indistinguishable from each other. Neural tube explants obtained from Sp mutant embryos exhibited more obvious differences. After 24 hours in culture, it was readily apparent that substantially fewer NCCs were being released from mutant neural tubes (Fig. 3A) than from non-mutant explants (Fig. 3B,C). Figure 3A is representative of 5 of the 8 mutant cultures, all of which showed similar patchy outgrowths of NCCs. The remaining Sp mutant cultures, although less strikingly different, were noticably deficient in initial (24 hour) cell outgrowth as well. These cells were often surrounded by larger halos than were non-mutant cells, indicating less adhesiveness to the substrate. By 48 hours, the number of NCCs released from Sp mutant explants had increased (Fig. 3D). These cells had begun to flatten against the substrate, possessing multiple extensions similar to non-mutant NCCs (Fig. 3E,F). By 72 hours the three types of cultures were virtually indistinguishable, with the mutant cultures having apparently caught up to heterozygote and wild-type cultures (see outgrowth data below).
Fig. 2. Comparison of NCC morphology on gelatincoated dishes in Spd mutant, heterozygote, and wildtype cultures after 24 hours. A: NCCS emigrating from the neural tube (NT) of an Sp" mutant. Note the difference in cell density which results in the slightly larger appearance of these NCCs compared to NCCs in (B)Spd heterozygote and (C) wild-type cultures. Bar -= 50 km.
NCC morphology in 3 -dimensional cultures In contrast t o the flattened, multipolar appearance of NCCs in planar cultures, NCCs surrounded by basement membrane matrigel were round and lacked cellular projections (Fig. 4C,D). No visible differences in NCC morphology were noted between Sp mutant and non-mutant cultures.
contact via their cellular projections. There was no correlation between the somite number of the embryo from which the neural tube was excised and the extent of NCC outgrowth at any time during the culture period. In 24 hour Spd mutant cultures (Fig. ZA), the cells surrounding the explants were often less confluent than those observed in the controls (Fig. 2B,C). This allowed the cells to lie flatter against the substrate, which consequently resulted in a greater proportion of mutant NCCs appearing larger in size than controls. Neural crest
Measurement of NCC outgrowth in planar cultures Comparison of Spd mutant t o non-mutant cultures showed significant differences between mean outgrowths of NCCs from neural tube explants (Table 1). The average NCC outgrowth from Spd mutant explants was less than that of either Spd heterozygous or wild-type explants in both 24 hour (P < 0.001) and 48 hour (P < 0.05, P < 0.01) cultures. This was due to failure of the outgrowth t o spread between 24 and 48 hours in mutant cultures. However, by 72 hours,
176
C.E. MOASE AND D.G. TRASLER
+/+
Fig. 3. Comparison of NCC outgrowth and morphology on gelatin-coated dishes in Sp mutant, heterozygote, and wild-type cultures after 24 and 48 hours. A: NCCs emigrating from the Sp mutant explant. Note that few NCCs are released after 24 hours and are surrounded by large refractile halos, unlike NCCs in (B)Sp
heterozygote and (C) wild-type cultures. D: NCC migration in the Sp mutant culture after 48 hours. Note: NCCs have increased in number and lie flatter against the substrate, similar to (E)Sp heterozygote and (F) wild-type cultures. Bar = 50.
mutant cultures had compensated for this initial delay and the difference in outgrowth from explants of the three genotypes was not significant (see Discussion). Similar results were obtained for Sp mutant explants. The extent of NCC outgrowth was less than that of either Sp heterozygous (P < 0.001) or wild-type (P < 0.01) explants after 24 hours in culture (Table 2). After 48 hours, the extent of outgrowth from mutant explants continued to be significantly less than that of the wild-type explants (P
Delayed Neural Crest Cell Emigration From Sp and Spd Mouse Neural Tube Explants CONNIE E. MOASE AND DAPHNE G. TRASLER Department of Biology and Centre for Human Genetics, McGill University, Montreal, Quebec, Canada H3A 1B1
ABSTRACT Splotch (Sp) and splotch-delayed (Spd)are allelic mutations on chromosome 1 of the mouse. Embryos homozygous for either allele have neural tube defects (NTDs) and deficiencies in neural crest cell (NCC) derived structures. The fact that Spd mouse mutants sometimes have deficiencies in NCC derivatives in the absence of an NTD led to the hypothesis that neurulation and the release of NCCs may depend on a regulatory event that is common to both processes. Therefore, it may be possible to understand the cause of NTDs in these mutants by examining the basis of aberrant NCC derivatives. Caudal neural tubes were excised from day 9 Sp and Spd embryos and placed into gelatin-coated tissue culture dishes, or 3-dimensional basement membrane matrigel, and cultured for 72 hours. A cytogenetic marker was used to genotype the embryos, In planar cultures, no morphological differences were observed between NCCs from neural tube explants of Spd mutants compared to those from heterozygous or wild-type embryos. However, there appeared to be a delay in the release of NCCs from the neural tube in both Sp and Spd mutants, which was particularly evident in Sp. After 24 hours in culture, the extent of NCC outgrowth, as well as the number of NCCs emigrating from explanted neural tubes, was significantly lower in Sp and Spd mutant cultures than in controls. No differences were observed in the mitotic indices among cells which had emigrated. By 72 hours, mutant cultures and their non-mutant counterparts were similar in terms of outgrowth, cell number, and migratory capability. After 24 hours in 3-dimensional basement membrane matrigel, cell outgrowth from Sp explants was also significantly less than controls. The pattern of NCC outgrowth in both types of culture conditions indicates a 24 hour delay in mutant cultures compared to controls. This stems from a delay in the release of NCCs from the neural tube, suggesting that the defect lies within the neuroeptihelium with respect to the release of NCCs. Neurulation and the release of neural crest cells (NCCs) from the neural tube, are two fundamental processes which occur either concomitantly or in close succession during the earlier stages of vertebrate development. Examination of mutants with both neural tube closure and NCC emigration defects may reveal mechanisms on which both these processes depend for normal development. Splotch (Sp) and splotchdelayed (Spd) are allelic mutations located on chromosome 1 in the mouse. Embryos that are homozygous for either of these alleles have neural tube defects (NTDs), as well as deficiencies in specific NCC deriva0 1990 WILEY-LISS, INC
tives including spinal ganglia and melanocytes (Auerbach, '54; Moase and Trasler, '89). However, these defects are more severe in Sp than in Spd mutants. Heterozygotes rarely develop a n NTD and have deficiencies in NCC derivatives that are less severe. These include a shortage of melanocytes that results in a white belly patch and white extremities in the adult, as well as reduced spinal ganglia volume in the lumbosacral region of heterozygous embryos (Moase and "rasler, '89). Received October 20, 1989; accepted January 30, 1990.
172
C.E. MOASE AND D.G. TRASLER
The use of a biochemical marker to distinguish homozygous Sp or Spd mutants from their heterozygous and wild-type littermates has revealed the presence of a few embryos (11%) with the mutant genotype that fail to display an NTD (Moase and Trasler, '89). However, deficiencies in NCC derived structures were present in all mutants. This, in conjunction with other evidence (Kapron-Bras and Trasler, '88),led to the hypothesis that there is a regulatory event during neurulation that is necessary for both neural tube closure and the release of NCCs from the neural tube. Elucidation of the mechanism responsible for deficient NCC derivatives could, in turn, reveal the cause of NTDs in these mutants. Histological studies have shown that in Sp homozygotes, a t the time of NCC emigration from the neural tube, the surface ectoderm is in tight apposition with the underlying neural tube in the trunk region of the embryo where the neural tube has managed to close (Kapron-Bras and Trasler, '88). This was in contrast to the normal situation where a n extracellular matrix (ECMI-filled space separated the two tissues, thus providing room for the emigrating NCCs to move out from the neural tube and commence migration. In addition, Kapron-Bras and Trasler observed a significant reduction in the number of NCCs that migrated from the Sp mutant neural tube. However, it was unclear whether the NCCs themselves were at fault; whether the abnormality resided in the neuroepithelium from which they arose, or in the environment in which they were released and migrated. It is unknown what triggers the release and subsequent migration of NCCs, although a specified set of conditions must exist in order for this to occur (Newgreen and Gibbins, '82; Thiery et al., '85). Individual analyses of some of these factors have added to the confusion. Lofberg et al. ('85), using ECM-adsorbed microcarriers, have demonstrated that NCCs can be stimulated to migrate prematurely. However, Morriss-Kay et al. ('86) report th at NCC migration still occurs in areas where the ECM has been depleted. Others maintain that the cellular adhesion system plays a key role (Innes, '85; Martins-Green and Erickson, '87). This study was designed to monitor the release and subsequent migration of NCCs from Sp and Spd mutant and non-mutant (i.e., heterozygous and wild-type) neural
tubes in vitro. Our aim was to determine whether the effect of the mutant gene was intrinsic or extrinsic to the neural tube from which NCCs are derived. It was also of interest to determine whether both Sp and Spd mutants would provide similar results in vitro, or differ in accordance with the differences they exhibit in vivo. The results indicate that there is a n initial delay in the release of NCCs from mutant neural tubes explanted onto a flat gelatin-coated substrate or into a 3-dimensional matrix containing basement membrane components, with Sp being more severely affected than Spd.However, once mutant NCCs have been released, their subsequent migration appears to occur normally in vitro. This suggests that the faulty mechanism resides within the neuroepithelium with respect to its ability to release NCCs at the appropriate time in development, rather than in the environment external to the neural tube. MATERIALS AND METHODS
Mice The splotch and splotch-delayed stocks were obtained in 1979 (Jackson Laboratory, Bar Harbor, ME), and have since been maintained in our laboratory by brothersister matings. The inversion marker stock (In(1)lRk) was obtained in 1984 from Dr. T.H. Roderick and is currently maintained through random mating. The breeding scheme used to incorporate the In(1)lRk marker into the mutant lines has been published elsewhere (Moase and Trasler, '87). Briefly, mice heterozygous for the mutant allele were crossed to In(1)lRk mice, which are homozygous for a n inverted segment on chromosome 1. This inverted segment encompasses the Sp (or Spd) locus but carries the wild-type allele. Double heterozygotes (carrying the mutant allele and the inversion) were then intercrossed to produce a n F, with three possible genotypes, which are distinguishable cytogenetically (see below). Mice were housed in plastic cages with wood-chip bedding and fed Purina Lab Chow and tap water ad libitum. A 12 hour darkllight cycle was maintained with a con1°C. One stant room temperature of 22 male was placed in a cage overnight with several females and, in the presence of a vaginal plug, the following day was designated as day 0 of gestation. Females were sacrificed on day 9.5 of gestation, and the uterus transferred to sterile Hanks' bal-
*
DELAYED EMIGRATION OF MUTANT NCCs
173
culture 1 % trypsin
4 OC
20 min
day 9 embryo
remove: -ectoderm -mesoderm -somites
\
\
cytogenetics
\
Fig. 1. Schematic representation of caudal neural tube isolation from day 9.5 embryos, as well as a chromosomal spread prepared from a n Spd heterozygote. G-
anced salt solution (HBSS). Further manipulations were conducted in a sterile flow cabinet. Embryos were removed from the uterus with watchmaker's forceps and transferred to fresh HBSS for the preparations described below.
banded chromosomes show the pattern associated with Sp or Spd gene carriers (a), or carriers of the wild-type allele on the inverted chromosomal segment (b).
dried slides (aged 4-7 days) were treated with 0.02% trypsin in 0.15M NaCl for 12-16 sec at 37°C and processed according to standard G-banding techniques (Davisson and Roderick, '73). Slides were scanned on a Leitz Laborlux K microscope, and a minimum of three metaphase spreads showing a conCytogenetics sistent genotype were scored per embryo. A Cytogenetic preparations were based on particular G-banding pattern on chromothe methods described by Evans et al. ('72). some 1 indicates the presence or absence of The anterior portion of each day 9 embryo, the Sp (or Spd) allele. Chromosome 1 is the including the amnion and embryonic yolk largest chromosome in the mouse karyotype sac, was placed in a separate 35 mm petri and, with the aid of G-banding, is easily disdish containing RPMI 1640 (Flow Laborato- tinguishable in a chromosomal spread. The ries, McLean, VA) with 5%fetal calf serum two heavy bands close together on the distal (FCS) plus 1%colcimid (GIBCO Laborato- part of chromosome 1 indicate that the Sp ries, Grand Island, NY), incubated for 1.5 allele is present (Fig. la). However, a 180" hours a t 37"C, transferred to 0.5% KC1 (15 rotation of the proximal band indicates that min at 37"C), and then fixed (3:l methanol: the wild-type allele from the inverted segacetic acid) for a minimum of 45 min a t 4°C. ment is present (Fig. lb). Thus the presence The tissue was dissociated in 60% acetic of both banding patterns in a single chroacid for 4 min and cell suspensions were mosomal spread typifies a heterozygous emdropped onto glass slides heated to 37°C. Air bryo.
174
C.E. MOASE AND D.G. TRASLER
Neural tube explants Primary neural tube explants were prepared and cultured according to the methods outlined by Ito and Takeuchi ('84) and Jaenisch ('851, slightly modified. The caudal portion of day 9 embryos (14-28 somite pairs) containing the pre-migratory NCC region was excised and placed in 1% trypsin (GIBCO, 1:250) in HEPES buffer (Flow) at 4°C for 20 min, a s illustrated in Figure 1. (Each genotype was represented by a n even distribution of embryos at various developmental stages.) The tissue was rinsed twice in HEPES containing 10%FCS, transferred to Dulbecco's modified Eagle's medium (DMEM),and dissected free of the surrounding tissues by using sharpened tungsten needles. A Wild M7A stereomicroscope was used to ascertain the complete removal of somites, notochord, mesoderm, and ectoderma1 tissues. The neural tubes were transferred to 35 mm plastic tissue culture dishes that had been coated with 0.5% gelatin, and cultured in 1.5 ml of DMEM supplemented with 5% FCS and 5% horse serum for 72 hours a t 37°C in a humidified atmosphere of 5% co,. Neural tubes for explantation into a 3dimensional basement membrane matrix were prepared a s above, with the addition of a 1 hour pre-incubation of the explanted neural tubes in DMEM containing 10% FCS prior to transfer into gelling matrix (Bilozur and Hay, '88). For this experiment, only Sp mutant cultures, and not Spd, were compared to their heterozygous and wild-type counter arts. Basement membrane matrigelT' (Collaborative Research, Bedford, MA) was thawed on ice, diluted 2:l with DMEM, and dispensed into microtiter plates just prior to the addition of individual neural tube explants. Neural tubes were suspended in the matrigel with the dorsal surface upwards and the dishes were immediately placed into the incubator to accelerate the gelling process. Once the matrix had gelled completely, unsupplemented DMEM was carefully dispensed into each well, and the neural tubes cultured for 72 hours. Visualization of mitotic figures A separate group of splotch and splotchdelayed explants, which had been cultured for 24 or 48 hours on gelatin-coated tissue culture dishes, were carefully rinsed with phosphate buffered saline, fixed in ethanol/
acetic acid (3:l) (Maxwell, '761, and processed by using the Feulgen staining reaction (Humason, '79). An ocular grid was used to determine the percentage of mitotic figures per given area encompassed by the grid. At least three separate regions around each explant were counted and these were subsequently averaged. All counts were performed before the embryos were genotyped. No attempt was made to assess the mitotic indices of 3-dimensional cultures.
Data collection and analysis Each explant was examined and photographed daily over the 72 hour period by using either Zeiss or Olympus phase contrast optics at 100 or 200 x . The front of maximum NCC outgrowth was marked directly onto contact prints of phase-contrast images from 24, 48, and 72 hour cultures. The average distance from the edge of the explant to the maximum front of migrating NCCs was calculated for each culture from 2-3 measurements at each time period. In addition, for explants cultured on flat substrates, the average number of cells within the regions of NCC migration was determined a t each time point by counting cells with a n ocular grid. The genotype of the explant was not revealed until after all measurements had been completed. Statistical analyses were carried out by using the Student's t-test, and the G-test of independence was used to analyze mitotic indices (Sokal and Rohlf, '81). RESULTS
NCC Morphology in planar cultures After 24 hours, early NCC outgrowth from control (wild-type) neural tubes explanted on a flat substrate consists of confluent monolayers with refractile halos surrounding each cell. Cells a t the periphery of the outgrowth are slightly larger. They begin to send out cellular projections and flatten against the substrate as they commence separation from one another. In 48 hour cultures, the NCCs continue to move outward. They become larger and more defined a s they flatten themselves against the substrate, and increase their multipolar projections in both number and length. By 72 hours, the rate of outgrowth begins to decrease and the outermost NCCs have reached their maximum size. At this point, the majority of cells have separated from one another. However, they still remain in
DELAYED EMIGRATION OF MUTANT NCCs
24hr
SPd/+
+ I+
175
cells derived from all types of cultures showed multipolar extensions, particularly those at the periphery of the outgrowth. At 48 and 72 hours, cultures derived from Spd mutant, heterozygous, and wild-type explants were morphologically indistinguishable from each other. Neural tube explants obtained from Sp mutant embryos exhibited more obvious differences. After 24 hours in culture, it was readily apparent that substantially fewer NCCs were being released from mutant neural tubes (Fig. 3A) than from non-mutant explants (Fig. 3B,C). Figure 3A is representative of 5 of the 8 mutant cultures, all of which showed similar patchy outgrowths of NCCs. The remaining Sp mutant cultures, although less strikingly different, were noticably deficient in initial (24 hour) cell outgrowth as well. These cells were often surrounded by larger halos than were non-mutant cells, indicating less adhesiveness to the substrate. By 48 hours, the number of NCCs released from Sp mutant explants had increased (Fig. 3D). These cells had begun to flatten against the substrate, possessing multiple extensions similar to non-mutant NCCs (Fig. 3E,F). By 72 hours the three types of cultures were virtually indistinguishable, with the mutant cultures having apparently caught up to heterozygote and wild-type cultures (see outgrowth data below).
Fig. 2. Comparison of NCC morphology on gelatincoated dishes in Spd mutant, heterozygote, and wildtype cultures after 24 hours. A: NCCS emigrating from the neural tube (NT) of an Sp" mutant. Note the difference in cell density which results in the slightly larger appearance of these NCCs compared to NCCs in (B)Spd heterozygote and (C) wild-type cultures. Bar -= 50 km.
NCC morphology in 3 -dimensional cultures In contrast t o the flattened, multipolar appearance of NCCs in planar cultures, NCCs surrounded by basement membrane matrigel were round and lacked cellular projections (Fig. 4C,D). No visible differences in NCC morphology were noted between Sp mutant and non-mutant cultures.
contact via their cellular projections. There was no correlation between the somite number of the embryo from which the neural tube was excised and the extent of NCC outgrowth at any time during the culture period. In 24 hour Spd mutant cultures (Fig. ZA), the cells surrounding the explants were often less confluent than those observed in the controls (Fig. 2B,C). This allowed the cells to lie flatter against the substrate, which consequently resulted in a greater proportion of mutant NCCs appearing larger in size than controls. Neural crest
Measurement of NCC outgrowth in planar cultures Comparison of Spd mutant t o non-mutant cultures showed significant differences between mean outgrowths of NCCs from neural tube explants (Table 1). The average NCC outgrowth from Spd mutant explants was less than that of either Spd heterozygous or wild-type explants in both 24 hour (P < 0.001) and 48 hour (P < 0.05, P < 0.01) cultures. This was due to failure of the outgrowth t o spread between 24 and 48 hours in mutant cultures. However, by 72 hours,
176
C.E. MOASE AND D.G. TRASLER
+/+
Fig. 3. Comparison of NCC outgrowth and morphology on gelatin-coated dishes in Sp mutant, heterozygote, and wild-type cultures after 24 and 48 hours. A: NCCs emigrating from the Sp mutant explant. Note that few NCCs are released after 24 hours and are surrounded by large refractile halos, unlike NCCs in (B)Sp
heterozygote and (C) wild-type cultures. D: NCC migration in the Sp mutant culture after 48 hours. Note: NCCs have increased in number and lie flatter against the substrate, similar to (E)Sp heterozygote and (F) wild-type cultures. Bar = 50.
mutant cultures had compensated for this initial delay and the difference in outgrowth from explants of the three genotypes was not significant (see Discussion). Similar results were obtained for Sp mutant explants. The extent of NCC outgrowth was less than that of either Sp heterozygous (P < 0.001) or wild-type (P < 0.01) explants after 24 hours in culture (Table 2). After 48 hours, the extent of outgrowth from mutant explants continued to be significantly less than that of the wild-type explants (P
