Avian neural crest cell migration is diversely regulated by the two major hyaluronan-binding proteoglycans PG-M/versican and aggrecan.

It has been proposed that hyaluronan-binding proteoglycans play an important role as guiding cues during neural crest (NC) cell migration, but their precise function has not been elucidated. In this study, we examine the distribution, structure and putative role of the two major hyaluronan-binding proteoglycans, PG-M/versicans and aggrecan, during the course of avian NC development. PG-M/versicans V0 and V1 are shown to be the prevalent isoforms at initial and advanced phases of NC cell movement, whereas the V2 and V3 transcripts are first detected following gangliogenesis. During NC cell dispersion, mRNAs for PG-M/versicans V0/V1 are transcribed by tissues lining the NC migratory pathways, as well as by tissues delimiting nonpermissive areas. Immunohistochemistry confirm the deposition of the macromolecules in these regions and highlight regional differences in the density of these proteoglycans. PG-M/versicans assembled within the sclerotome rearrange from an initially uniform distribution to a preferentially caudal localization, both at the mRNA and protein level. This reorganization is a direct consequence of the metameric NC cell migration through the rostral portion of the somites. As suggested by previous in situ hybridizations, aggrecan shows a virtually opposite distribution to PG-M/versicans being confined to the perinotochordal ECM and extending dorsolaterally in a segmentally organized manner eventually to the entire spinal cord at axial levels interspacing the ganglia. PG-M/versicans purified from the NC migratory routes are highly polydispersed, have an apparent M(r) of 1,200-2,000 kDa, are primarily substituted with chondroitin-6-sulfates and, upon chondroitinase ABC digestion, are found to be composed of core proteins with apparent M(r )of 360-530, 000. TEM/rotary shadowing analysis of the isolated PG-M/versicans confirmed that they exhibit the characteristic bi-globular shape, have core proteins with sizes predicted for the V0/V1 isoforms and carry relatively few extended glycosaminoglycan chains. Orthotopical implantation of PG-M/versicans immobilized onto transplantable micromembranes tend to 'attract' moving cells toward them, whereas similar implantations of a notochordal type-aggrecan retain both single and cohorts of moving NC cells in close proximity of the implant and thereby perturb their spatiotemporal migratory pattern. NC cells fail to migrate through three-dimensional collagen type I-aggrecan substrata in vitro, but locomote in a haptotactic manner through collagen type I-PG-M/versican V0 substrata via engagement of HNK-1 antigen-bearing cell surface components. The present data suggest that PG-M/versicans and notochordal aggrecan exert divergent guiding functions during NC cell dispersion, which are mediated by both their core proteins and glycosaminoglycan side chains and may involve 'haptotactic-like' motility phenomena. Whereas aggrecan defines strictly impenetrable embryonic areas, PG-M/versicans are central components of the NC migratory pathways favoring the directed movement of the cells.

Age-dependent inhibition of neural crest migration by the notochord correlates with alterations in the S103L chondroitin sulfate proteoglycan.

In avian embryos, the notochord inhibits neural crest migration, resulting in the absence of neural crest cells from the perinotochordal space. Here, we test whether temporal changes in the ability of the notochord to inhibit neural crest migration correlate with alterations in the S1O3L chondroitin sulfate proteoglycan (CSPG). Because CSPGs are abundant in the perinotochordal space and the inhibitory effects of the notochord are chondroitinase sensitive both in vivo and in vitro, we examined the distribution and biochemical nature of a large CSPG whose core protein is recognized by the S103L antibody. The S103L CSPG is specific to the perinotochordal space during the course of neural crest migration and codistributes with the HNK-1 carbohydrate. Biochemical characterization reveals that the S103L CSPG bears the HNK-1 epitope and is the only HNK-l immunoreactive proteoglycan present around the notochord at these stages. Following neural crest migration, the S103L CSPG staining is maintained in the perinotochordal region and also is expressed later in cartilage. In 4-day-old embryos, however, the S103L CSPG undergoes a reduction of HNK-1 immunoreactivity. To examine the temporal nature of the notochord's inhibitory ability, we assayed the effects, on neural crest migration of grafting notochords from 2- to 5-day-old donor quail embryos into 2-day-old host chick embryos. Donor notochords from 2- to 3-day-old embryos inhibit neural crest cell migration, whereas the degree of inhibition is reduced or absent when notochords are derived from > or = 4-day-old donors. This suggests that older notochords lose their inhibitory ability. Interestingly, preincubation of younger notochords with the HNK-1 antibody blocks the inhibitory effect, suggesting that glycosylation of the perinotochordal matrix may be important. The time when the notochord loses its inhibitory ability as assessed by our in vivo grafting assay correlates with the biochemical and immunocytochemical changes in the notochordal S103L antigen. These data suggest that a species of S103L CSPG, which is expressed by the early notochord and bears the HNK-1 epitope, may be important for the inhibition of neural crest migration.

Inhibitory effects of PG-H/aggrecan and PG-M/versican on avian neural crest cell migration.

Aggrecans and PG-M/versicans represent two newly defined families of hyaluronan-binding proteoglycans for which the function is still poorly understood. Using the avian neural crest as a model system, we have examined the molecular mechanisms entailed in the cell-proteoglycan interaction during embryonic cell motility. Both the primary cartilage aggrecan of the avian embryo (PG-H/aggrecan) and the largest variant of the avian mesenchymal versican (PG-M/versican VO) failed to support neural crest cell adhesion and migration when topographically immobilized onto the substrate. Conversely, solely the PG-H/aggrecan, and similar aggrecans from other species, counteracted the migration-promoting effect of a number of matrix molecules lacking proteoglycan affinity. This inhibitory effect was not reproduced by the isolated glycosaminoglycan chains, the isolated core protein, the reduced and alkylated macromolecule, or the aggrecan in which the G1 hyaluronan-binding domain had been inactivated with hyaluronan fragments or antibodies. Limited depolymerization of the side chains and preincubation of the PG-H/aggrecan with anti-glycosaminoglycan antibodies differentially reduced the inhibitory activity of the proteoglycan on cell motility. The results demonstrate a diverse inhibitory effect of aggrecans and PG-M/versicans on embryonic cell movement and show that the inhibitory action of aggrecans is independent of substrate binding, is dependent on a G1 domain-mediated association of the intact proteoglycan with cell surface-bound hyaluronan, and is differentially mediated by its glycosaminoglycan side chains.

Dose-response effects of glucose, insulin, and glucagon on mouse pre-embryo development.

The diabetic state, as well as elevated culture media glucose level (950 mg D-glucose/dL) per se, significantly retards in vitro development of mouse pre-implantation embryos from a two-cell stage to blastocyst stage; maternal insulin therapy to diabetic mice reverses this impairment. This study was undertaken to assess (1) whether less extreme elevation of the media glucose concentration would also impair development, and (2) whether elevated culture media insulin or glucagon levels would alter development. Two-cell pre-embryos were recovered from B6C3F1 mice that had been stimulated with pregnant mare serum gonadotropin (PMSG) and human chorionic gonadotropin (hGG), mated, and killed 48 hours later. Pre-embryos were observed in culture at 24-hour intervals for a total of 72 hours at four glucose levels: 110 (n = 108), 220 (n = 101), 440 (n = 65), and 950 (n = 106) mg D-glucose/dL. Impairment in progression of development was noted at each time period; compared with development in 110 mg glucose/dL, the distribution of development was significantly different at 24 hours (chi 2 = 60.1, P less than .001), at 48 hours (chi 2 = 36.7, P less than .001), and at 72 hours (chi 2 = 45.1, P less than .001). Rate of development as assessed by ANOVA was also significantly reduced at increasing glucose levels (P less than .0001), with Duncan Multiple Range test demonstrating differences between development at higher glucose levels in the comparison of development in 110 mg/dL versus 440 mg/dL and 950 mg/dL, and at 220 mg/dL versus 950 mg/dL.(ABSTRACT TRUNCATED AT 250 WORDS)

Absence of neural crest cells from the region surrounding implanted notochords in situ.

Avian neural crest cells migrating along the trunk ventral pathway are distributed throughout the rostral half of the sclerotome with the exception of a neural crest cell-free space of approximately 85 microns width surrounding the notochord. To determine if this neural crest cell-free space results from the notochord inhibiting neural crest cell migration, a length of quail notochord was implanted lateral to the neural tube along the neural crest ventral migratory pathway of 2-day chicken embryos. The subsequent distribution of neural crest cells was analyzed in embryos fixed 2 days after grafting. When the donor notochord was isolated using collagenase, neural crest cells avoided the ectopic notochord and were absent from the area immediately surrounding the implant (mean distance of 43 microns). The neural crest cell-free space was significantly less when notochords were isolated using trypsin or chondroitinase digestion and was completely eliminated when notochords were fixed with paraformaldehyde or methanol prior to implantation. The implanted notochords did not appear to affect the overall number of neural crest cells, and therefore were unlikely to exert this effect by altering their viability. These results suggest that the notochord produces a substance that can inhibit neural crest cell migration and that this substance is trypsin and chondroitinase labile.