Dependence of sea urchin primary mesenchyme cell migration on xyloside- and sulfate-sensitive cell surface-associated components

DEVELOPMENTAL

127, 78-87 (1988)

BIOLOGY

Dependence of Sea Urchin Primary Mesenchyme Cell Migration on Xyloside- and Sulfate-Sensitive Cell Surface-Associated Components MARYCONSTANCELANEANDMICHAEL Depurfment

of Biology, University

SOLURSH

of Iowa, Iowa City, Iowa 52242

Accepted January

8, 1988

The migration of sea urchin primary mesenchyme cells (PMC) is inhibited in embryos cultured in sulfate-free seawater and in seawater containing exogenous xylosides. In the present study, primary mesenchyme cells and extracellular matrix have been isolated from normal and treated Lytechinus pictus and Strongylocentrotus purpuratus embryos and recombined in an in vitro migration assay to determine whether the cells or the matrix are migration defective. Normal cells were found to migrate on either normal or treated matrix, whereas sulfate-deprived and xyloside-treated PMC failed to migrate in vitro on normal and treated substrata. Migratory ability can be restored to defective cells by returning the PMC to normal seawater, or by exposing the defective cells to materials removed from the surface of normal cells with 1 M urea. The similarity of the results obtained with sulfate-deprived and xylosidetreated PMC suggested that a common molecule may be affected by the two treatments. As a first test of this S. purpuratus PMC were given the urea extract prepared from sulfate-deprived S. purpurpossibility, xyloside-treated atus PMC, and this extract did not restore migratory ability. These findings indicate that PMC normally synthesize a surface-associated molecule that is involved in cell migration, and the sensitivity to exogenous xylosides and sulfate deprivation suggests that a sulfated proteoglycan may be involved in primary mesenchyme cell migration. c 1988 Academic

Press, Inc.

INTRODUCTION

In the sea urchin, PMC migration can be inhibited by a number of environmental perturbations during development. Although there may be some variation in species responses, sulfate deprivation (Herbst, 1904; Karp and Solursh, 1974) and exposure to fi-D-xyloside (Akasaka et al., 1980; Solursh et al., 1986) block PMC migration. Under either manipulation, in affected species, ingression of the cells appears normal (Karp and Solursh, 1974; Solursh et ab, 1986). Microcinematography reveals that PMC are active on the vegetal plate, and scanning electron microscopy detects special cellular processes between cells and between cells and the basal lamina (Akasaka et al., 1980; Katow and Solursh, 1981). The first visible effect of sulfate deprivation in the species Lytechinus pictus is the absence of PMC migration; however, this treatment is known to have a variety of biochemical effects on embryos. Reduction in [3H]thymidine and [3H]uridine incorporation (Runnstrom et al., 1964) and nuclear retention of newly synthesized RNA (Gezelius, 1976) have been reported in other species, but these effects seem greatest in postmigratory stages. At the mesenchyme blastula stage, when migration should normally occur, the sulfated glycan content of the blastocoel is reduced (Immers, 1961; Sugiyama, 1972; Karp and Solursh, 1974); by scanning electron microscopy, the normally rough surface of the PMC appears smooth (Karp and Solursh, 1974), and at the ultrastructural level, the blastocoel contains few of the 15- to 30-nm diameter granules (Katow and Solursh,

The molecular mechanisms operating at the interface of the extracellular matrix and the plasma membrane of a migrating cell remain virtually unknown. As a model system, the primary mesenchyme cells (PMC) of the developing sea urchin embryo provide an excellent opportunity to study the interaction between a moving cell and its environment (Solursh, 1986). In the course of normal development, these cells ingress into the blastocoel via an epithelial-mesenchymal transition, actively migrate as individual cells along the basal lamina that lines the blastocoel, and ultimately reaggregate as two ventrolateral clumps joined by a subequatorial ring of cells to produce the larval skeleton. The sea urchin system offers a number of distinct advantages over vertebrate systems: the embryos are transparent and relatively light-insensitive, which has allowed extensive descriptions of in viva cell behavior (Gustafson and Kinnander, 1956; Gustafson and Wolpert, 1961, 1963); large numbers of synchronized embryos can be cultivated in the laboratory under conditions that yield normal embryos; the embryos can be manipulated to give migration-defective embryos (discussed below); and, finally, techniques are available to isolate the migrating cells (Venkatasubramanian and Solursh, 1984) or the extracellular matrix (ECM) that serves as the migratory substratum for these cells (Harkey and Whiteley, 1980; Solursh and Katow, 1982). 00121606/88 Copyright All rkhts

$3.00

0 1988 by Academic Press, Inc. of reproduction in any form reserved.

78

LANEAND~OLURSH

Primary

1979), believed to be proteoglycans (Solursh et al., 1986), that are associated with the kinetic appendages of migrating PMC in normal embryos (Katow and Solursh, 1981). 1n vitro, Venkatasubramanian and Solursh (1984) found that while normal L. pi&us PMC could migrate on a substratum of human plasma fibronectin, sulfate-deprived PMC showed a greatly reduced capacity, and this condition was ameliorated 6 hr after the cells were returned to normal seawater. In Strongylocentrotus purpwatus, the first visible effect of /3-D-xyloside exposure is also the absence of PMC migration along the wall of the blastocoel. At the ultrastructural level, the defect caused by xyloside exposure is similar to that of sulfate deprivation; again, there is a paucity of the 15- to 30-nm diameter granules associated with the PMC and in the extracellular matrix of the blastocoel (Solursh et al., 1986). Biochemically, there is a reduction in the sulfated blastocoelic material voided on a Sepharose CL-2B column concomitant with an increase in included material. The majority of this included material is sensitive to chondroitinase ABC treatment (Solursh et al., 1986). These results are consistent with the known effects of xylosides in other systems-namely, that xylosides substitute for the core protein to initiate the synthesis of free glycosaminoglycan chains, thereby disrupting the production of intact proteoglycans. Specifically, xylosides interfere with the production of proteoglycans containing the serine-xylose protein-glyeosaminoglycan linkage. In S, puqrurutus, several classes of proteoglycans have been reported (Solursh and Katow, 1982; Solursh et al., 1986). The disruption of PMC migration by fi-D-xyloside treatment and sulfate deprivation suggests that proteoglycans and sulfated molecules are part of the machinery required for migration by sea urchin primary mesenchyme cells. Both treatments have been shown at the ultrastructural and biochemical levels to affect the extracellular matrix, and in the case of sulfate deprivation, to cause a defect in the PMC themselves which decreases their migration on fibronectin (Venkatasubramanian and Solursh, 1984). In the present study, we have examined the interaction in vitro between the PMC and the extracellular matrix of the blastocoel. Normal and treated (xyloside-exposed or sulfate-deprived) PMC were recombined with normal and treated matrices to determine whether treated cells, matrix, or both are migration defective. METHODS AND MATERIALS Gametes were obtained from adult L. pictus and S. (Pacific Biomarine, Venice, CA) by intracoelomic injection of 0.5 MKCl. Eggs were washed twice with artificial seawater (ASW) and fertilized with a

purpuratus

Mesenchyme

79

Cell Migration

dilute sperm suspension. L. pictus embryos were cultured at 17°C and S. purpuratus at 13°C in ASW, sulfate-free seawater (SFSW), or seawater containing 2 mM p-nitrophenyl-P-D-xylopyranoside (XASW) until the mesenchyme blastula stage, as described previously (Solursh et at,, 1986). Dissociatiow qf Embryos and Isolation Blastocoel Contents

qf

Embryos were washed twice with cold calcium, magnesium-free seawater (CMFSW) and then incubated twice for 5 min each in cold CMFSW containing 100 PM EDTA. The embryos were then washed twice in cold calcium-free seawater (CFSW), resuspended in CFSW (1 ml of embryos + 7 ml of CFSW), and dissociated by six to eight passes in a 7-ml Wheaton Dounce homogenizer, using a pestle with a clearance between 63 and 89 pm. Dissociation was monitored by light microscopy, and the resultant single-cell suspension was centrifuged at 12008, at 4°C in a clinical centrifuge for 7 min. The supernatant was removed, checked for the absence of cells, aliquoted, and either used fresh to coat coverslips or stored frozen at -20°C. Aliquots were used to determine the protein content of the supernatant (BioRad, Richmond, CA). The contents of this supernatant were ethanol precipitated and analyzed on 7.5% polyacrylamide slab gels (Laemmeli, 1970). Gels were either silver stained (BioRad, Richmond, CA) or stained by the concanavalin Aperoxidase method of Parish et al, (1977). Isolation

qf Primary

Mesench yme Cells

Primary mesenchyme cells (PMC) were isolated as described by Venkatasubramanian and Solursh (1984) with the following exceptions: for P-D-xyloside-treated PMC, all solutions contained 2 mM p-nitrophenyl-P-Dxylopyranoside (Sigma, St. Louis, MO); for sulfate-deficient PMC, all dissociation steps were carried out as rapidly as possible since CFSW and CMFSW contain sulfate. The dissociated cells were resuspended in SFSW plus 2% dialyzed horse serum; and finally, all plates were washed five times with the appropriate seawater before removal of the attached cells. The identification of the attached cells as PMC (>99% for all treatments) was established by staining with FITCwheat germ agglutinin (Ettensohn and McClay, 1987). The attached cells were removed from the plates, centrifuged at 800g for 5 min at 4°C in a silanized tube, and resuspended in 0.5 ml of ASW, SFSW, or XASW. Migration

Assa,y

Coverslips (40 mm) for Skyes-Moore chambers were treated with concanavalin A (Sigma, St. Louis, MO), 2.5

80

DEVELOPMENTALBIOLOGYVOLUME127,1988

mg/ml in 0.02 M phosphate-buffered saline (PBS), pH ‘7.2,for 2 hr at 37°C. The excess solution was removed, and the coverslip was washed once with PBS and then treated with the blastocoel material (adjusted to 30-60 pg protein/ml) described above at room temperature. After 1 hr the excess suspension was removed, the substratum was washed with the appropriate seawater, and the coverslip was either assembled into a SykesMoore chamber or used to visualize the substratum by scanning electron microscopy (SEM, see below). Isolated, suspended PMC were added to the chamber and allowed to settle and attach for 45 min. The chamber was inverted, viewed, and recorded using a 20x longworking distance objective on a Leitz SM-Lux compound microscope or Leitz Diavert microscope connected to a Dage-MT1 video camera, Panasonic video recorder, and Audiotronic monitor. Cell behavior was monitored for 90 min by tracing cell perimeters intermittently on a transparent plastic sheet mounted on the monitor. Net movement was determined by connecting cell centers at the initial and final time points. Alternatively, behavior was photographed using Kodak Panatomic-X film and a Wild MPS12 35-mm camera mounted to the microscope, or a Nikon 35-mm camera directed at the video monitor. After 90 min of observation, the behavior of individual cells in the field was scored as “no net movement,” “net movement less than one cell diameter,” or “net movement greater than one cell diameter.” These categories were chosen to simplify a variety of cell behaviors into a concise table that reflected observations. “No net movement” includes cells that did not change position and cells that underwent extensive shape changes but still maintained a visible attachment to the cell’s original location. Examples in Fig. 4 are cells 4 and 5, respectively. The categories “movement less than one cell diameter” and “movement greater than one cell diameter” were chosen to distinguish between cells that may be simply settling onto the substratum and cells actually translocating. It should be noted that the cells in the final category are traveling many cell diameters (e.g., cells 1, 2, and 3 in Fig. 4). Scanning Electron Microscopy

Coverslips for SEM analysis were fixed on ice for 15 min in 1% glutaraldehyde in ASW, rinsed for 2 min in ASW, and subsequently postfixed in 1% 0~0, in 0.1 M cacodylate buffer, pH 7.2, for 15 min. The coverslips were then dehydrated in a graded ethanol series and critical point dried in an EM Scope CPD 500, sputter coated in an EM Scope SC 500, and viewed in an Hitachi S-570 scanning electron microscope.

Urea Extraction

of Primary

Mesenchyme

Cells

After washing the cells previously attached to 12 tissue culture dishes (150 X 25 mm Integrid Tissue Culture Dishes, Falcon No. 3025) five times with the appropriate seawater, PMC were exposed to 50 ml of 1 M urea (Ultra Pure, Schwarz/Mann, Spring Valley, NY) in 0.05 M Tris-buffered ASW, pH 8.1, at 4°C for 1 hr. Cells that detached from the plate were removed by centrifugation at 12OOg,7 min, 4”C, and the resulting supernatant was transferred to dialysis tubing (Spectra/Par membranes No. 1 or 3, MWt cutoffs of 3.5 and 6-8 kDa, respectively; Spectrum Medical Industries, Los Angeles, CA). The dialysis tubing was surrounded with solid pellets of polyethylene glycol (MW, 15-20 kDa, Sigma, St. Louis, MO) and kept at 4°C until the volume was reduced to 1 ml. The resulting solution was dialyzed against SFSW or XASW (depending on the cell type to be treated with the extract) until there was no further change in volume. Isolated sulfate-deprived or xyloside-treated PMC were resuspended in the conditioned seawater diluted 1:l with the appropriate seawater for 20-30 min and then added to a coverslip coated with normal ECM. Cell migration was monitored as above. RESULTS

The extracellular materials isolated following extensive washing of the embryos in CMFSW, CMFSW + EDTA, and CFSW and dissociation of the cells were analyzed by silver and eoncanavalin A-peroxidase staining of SDS-polyacrylamide gels. Silver staining (Fig. 1, lane A) reveals a large number of bands and large amounts of material that remain in the stacking gel, and Con A-peroxidase staining (Fig. 1, lane B) reveals that a number of molecules present in the preparation are glycoconjugates that bind Con A. Although there is some variation in the band pattern between L. picks (Fig. 1) and S. purpuratus (not shown), both species show numerous bands under both staining protocols. The complex band pattern obtained has been compared with an earlier SDS-PAGE analysis of intact blastocoelic bags isolated by the method of Harkey and Whiteley (Norbeck, 1983). There are differences in the intensity of staining of a few bands, but the pattern of bands is the same and the intensity of the staining is similar for most bands. An initial characterization of the major sulfated components and a description of the morphology was the subject of a previous report (Solursh and Katow, 1982). These observations lead us to believe that the ECM isolated represents the contents of the blastocoel. Migratory substrata were prepared by treating glass coverslips with concanavalin A and subsequently with blastocoelic ECM. The observations by Iwata and Na-

LANE

A

AND SOLURSH

Primary

B

Mesenchyme

in situ. The cell body is raised off the substratum, numerous filopodia contact matrix material, and the broad lamellipodia seen in cells on Con A alone are not present. L. pictus PMC on Mesenchyme Blastula

11697.4-

29+

FIG. 1. SDS-PAGE analysis of L. pi&us extracellular matrix from the blastocoel, isolated as described under Methods and Materials. Lane A is stained using silver (Bio-Rad) and lane B is stained using concanavalin A. The positions of the molecular weight markers are shown on the left. The ECM contains numerous molecular species that stain positively with silver, and many of those also bind Con A.

kano (1984) that dissociated embryonic cells attached to Con A coated dishes were surrounded by extracellular material suggested that Con A could retain matrix material, and this might prove a suitable migratory substratum for PMC. SEM observation of coated coverslips (Fig. 2) shows that networks of fibrillar and granular materials have been retained by the Con A, and the morphological appearance of these materials is similar to the in situ ultrastructural appearance of the blastocoel ECM (see Katow and Solursh, 19’79) and the appearance of isolated, intact blastocoelic bags (see Solursh and Katow, 1982). Con A coated coverslips bind isolated PMC, but the cells do not migrate on coverslips coated with Con A alone. Fig. 3A depicts normal PMC after 4 hr on a Con A coated coverslip. The extensive spreading of the cells and the presence of “webbing” between filopodia are atypical of PMC, and suggest that the PMC may be unable to utilize Con A as a migratory substratum because the cells adhere strongly to the lectin. PMC on coverslips coated with Con A and subsequently with ECM (Fig. 3B) display a morphology that bears much great.er resemblance to that of normal PMC

81

Cell Migration

Stage EC44

The inhibition of PMC migration in sulfate-deprived embryos could be the result of defects in the PMC, the ECM, or in both the cells and the matrix. To distinguish between these possibilities, normal or sulfate-deprived PMC were seeded onto coverslips coated with normal or sulfate-deficient ECM and the resulting interaction was monitored by microcinematography. During a 90-min observation period, 26% of normal cells were found to migrate on normal matrix (Fig. 4 and Table 1) and 26Yo on sulfate-deficient matrix. The behavior and morphology of the migrating cells were similar to those observed in vivo; cells formed filopodia, underwent frequent shape changes, and moved at varying rates across the substratum. Sulfate-deprived cells were not observed to migrate on either normal (Fig. 5 and Table 1) or sulfate-deficient matrix (0% on either substrata). The failure of sulfate-deprived cells to migrate on either substratum suggests that there must be a defect in the PMC themselves, and the defect is not corrected by the availability of normal ECM.

FIG. 2. SEM of a migratory substratum prepared as described under Methods and Materials. Fibers and granules, similar to those seen in intact, isolated blastocoelic bags (Solursh and Katow, 1982), are retained on coverslips first treated with Con A. Scale bar = 15 pm.

82

DEVELOPMENTAL BIOLOGY

VOLUME 127,1988

Venkatasubramanian and Solursh (1984) reported that sulfate-deprived PMC were capable of migrating on fibronectin after 6 hr in normal ASW. They concluded that the ability of sulfate-deprived PMC to recover in normal seawater indicates that the PMC synthesize a sulfate-dependent component that is required for migration. We find by continuous monitoring of sulfate-deprived cells and embryos returned to normal ASW that PMC recovery takes ca 90 min, both in vitro and in Vito at 18°C. Karp and Solursh (1974) had previously suggested that a sulfated component on the surface of PMC might be required for migration. In order to test, this hypothesis directly, normal PMC were treated with 1 Murea to remove peripheral membrane-associated materials. After this extraction, the cells were still viable although many had assumed round morphologies or detached from the dish. If the cells are removed from the dish, concentrated, and recultured in 2% horse serum, normal in vitro morphology and behavior are restored, and after ca 24 hr, spicules form (not shown). Most important, if sulfate-deprived PMC were treated with the materials removed from the surface of normal PMC, in vitro migratory capacity was restored (27% migrated; Table 1 and Fig. 6). These results provide evidence to support the suggestion that PMC migration is dependent on a sulfated surface component. Sulfate deprivation could affect many metabolic pathways in the embryo. To test whether the sulfatedeprivation defect could be rectified by the addition of sulfur-containing amino acids, L. pi&s embryos were cultured in SFSW containing 7.5 mg/liter L-methionine and 12.0 mg/liter L-cystine. Addition of the amino acids did not alleviate the migration block, and the embryos also failed to gastrulate and form spicules. This result suggests that the migration defect is not caused by the depletion of sulfur-containing amino acids. S. purpwratus

FI G. 3. SEMs of S. mrpuratus PMC attached to coated glass coverSlip ; after 4 hr. (A) Normal PMC on Con A. Broad lamellipodia (short arro w) are often present between adjacent filopodia, and the cell body spre ads and becomes somewhat flattened (long arrows) on the Con A subs .tratum. PMC seeded onto coverslips coated with Con A and subsequlently with embryonic ECM (B) display a normal bipolar morphology. The cell body is raised from the substratum and filopodia interact extensively with the ECM. Scale bar = 3 pm.

PMC 0’11Mesenchyme Blusfdu

Stage ECM

To determine whether the migration defect caused by /3-D-xyloside exposure resides in the ECM or the PMC, normal or xyloside-treated PMC were seeded onto normal or xyloside-treated ECM. Table 2 summarizes the results of the i?z vitro recombination assays for S. peerpuratus; during a 90-min interval, normal PMC were able to migrate on either normal (21%) or xylosidetreated ECM (29%, which is not statistically different, by Tukey’s HSD test, p < 0.05), and the behavior of these cells in z&o again resembled in Guo observations. Xyloside-treated cells were unable to migrate on both substrata (1% migrated on normal and 0% on xylosidetreated matrix), indicating that xyloside exposure is also directly affecting the PMC.

83

A

.

c

c

I

FIG. 4. L. piths migrate

normal PMC on normal matrix (A) at 0 min, (B) at 13 min, (C) at 46 min, and (D) at 91 min of observation. on sulfate-deficient matrix. Bar = 50 pm.

Migratory ability can be restored in two ways. Cells returned to normal ASW will resume migrating in ca 2 hr, both in viva and in vitro; alternatively, cells in the presence of /3-D-xyloside will resume migrating (2795, on normal ECM) when treated with a crude extract obtained by treating normal PMC with 1 M urea in Tris-buffered ASW. A similar extract prepared from xyloside-treated PMC does not alleviate the migratory block (0% migrated), but subsequent culture in ASW for 2 hr yields a population of actively migrating cells (28% migrated). These results demonstrate that the PMC synthesize a xyloside-sensitive surface component that is required for migration. The migration-promoting activity of the mesenchyma1 extract was tested by exposing xyloside-treated PMC with an analogous extract prepared by treating epithelial cells with 1 M urea in Tris-buffered ASW.

Normal

cells also

The extract derived from epithelial cells did not restore migratory ability to the xyloside-treated PMC, although it did have a pronounced effect on PMC morphology. Cells exposed to the epithelial extract showed larger, more numerous, and more complex filopodia that interacted more extensively with other cells and the substratum (not shown) than did cells given the PMC extract. These results indicate the urea extracts from epithelial and mesenehymal cells differ significantly in their biological properties and in their ability to correct the xyloside defect that blocks PMC migration.

vitro

The similarity of the pattern of results obtained in using sulfate deprivation in L. pictus and P-D-xy-

84

DEVELOPMENTAL BIOLOGY TABLE

Normal PMC on normal matrix Normal PMC on SF matrix SF PMC on normal matrix SF PMC on SF matrix SF PMC on normal matrix plus urea extract SF PMC on normal matrix plus S. ptqznLratus urea extract

than in L. pictus (e.g., we observe delayed hatching), it also inhibits PMC migration at the mesenthyme blastula stage. To begin to determine whether sulfate deprivation and xyloside exposure might affect the same molecule(s), S. purpuratus xyloside-generated PMC were given an extract made from S. purpurutus sulfate-deprived PMC; as in the case of extract isolated from xyloside-generated PMC, the sulfate-deficient extract did not restore migratory ability to the cells (0% migrated, Table 2). In the interspecies recovery assay, sulfate-deprived L. pictus PMC were treated with materials extracted from the surface of normal S. purpuratus PMC. This extract promoted cell migration in the sulfate-deprived cells (22% migrated, Table 1).

purpuratus

1

L. p&us IN VITRO MIGRATION ASSAYS

Cell-matrix combination’

VOLUME 127, 1988

No net movement*

Moved ~1 cell diameterb

Moved >l cell diameter’

NC

25 43 93 100

49 31 7 0

26d 26 0 0

53 61 44 22

20

65

15

69

68

10

22

84

“Two to six trials were performed for each cell-matrix combination used. b Movement categories are discussed under Methods and Materials. ‘All figures stated are percentages, except N, which was the number of cells observed. d By Tukey’s HSD test, there is no effect of sulfate-deficient or normal matrix on the migratory behavior of normal or sulfate-deprived cells (p < 0.05). However, sulfate-deprived differed significantly from normal cells on either matrix (p < 0.05).

loside exposure in S. purpuratus suggested that a similar molecule on the surface of PMC in both species may be affected by the two treatments. To examine this hypothesis, a cross-treatment and an interspecies recovery assay were performed. Although sulfate-deprivation appears to affect earlier developmental events in S.

DISCUSSION

Sulfate deprivation and P-D-xyloside exposure have been found to have effects on the primary mesenchyme cells that limit migratory activity in an in vitro migration assay. Normal L. pictus PMC migrated in vitro on normal and on sulfate-deficient ECM, but sulfate-deficient PMC did not migrate on either substratum. This inhibition could be alleviated by treating the migration-defective PMC with cell surface-associated materials removed from normal PMC by treatment with 1 M urea, or by returning the cells to normal ASW. This suggests that the PMC normally synthesize a sulfated molecule (or molecules) that is associated with the ex-

_’

FIG. 5. L. p&us sulfate-deprived PMC on normal matrix. At 0 min, (B) at 80 min. Bar = 50 lm.

Sulfate-deprived

1

PMC do not migrate on normal nor sulfate-deficient

matrices.

(A)

LANEANDSOLURSH

A FIG. 6. L. pictus sulfate-deprived ASW, on normal matrix. Treated

Primary

Mesenchyme

c

6 PMC, treated with materials extracted cells are able to migrate. Bar = 50 pm.

tracellular surface of the plasma membrane, and that this molecule is required for migration. Exposure to P-D-xyloside has effects on S. purpuratus PMC that block in vitro migration. Normal cells migrate on either normal or treated matrix, but xylosideexposed PMC migrate on neither. The ability to migrate can be restored either by removal of the @-D-xyloside, or by treatment with materials extracted from the surface of normal PMC by 1 Murea. These findings suggest that P-D-xyloside interferes with the synthesis of a cell surface-associated molecule that is required for PMC migration, and implicates a role for serine-xylose-linked proteoglycans in cell migration. These findings in vitro may correlate with the in situ observations on sulfate-deprived (Karp and Solursh, 1974) and xyloside-exposed (Solursh et al., 1986) embryos. Under both conditions, there is a reduction in the 15 to 30-nm granules seen in the blastocoel associated with the kinetic appendages of PMC. These granules are believed to be proteoglycans, and could be susceptible to both the presence of xyloside and the absence of sulfate. In 5’. pwpuratus, PMC migration is inhibited by both treatments, and although the materials extracted from the surface of normal PMC permit xyloside-treated PMC to migrate, materials from the surface of sulfatedeprived PMC do not. In an interspecies migration assay, sulfate-deprived L. pictus PMC were given materials extracted from the surface of normal S. purpuratus PMC, and migration was subsequently observed. This raises the possibility that a common molecule may be affected by sulfate deprivation and xyloside-exposure, and suggests that sulfated proteoglycans may play

85

Cell Migration

from the surface of normal

PMC with 1 M urea in Tris-buffered

a role in in vitro cell migration. The urea extract is presently being analyzed, with particular attention given to sulfated proteoglycans. In the present in vitro migration assay, both normal and abnormal matrices supported migration of normal PMC. As the cells were not capable of migrating on a substratum of Con A or on untreated glass (unpublished data), materials present in the ECM preparation, which are apparently not susceptible to either sulfate deprivation or xyloside-exposure, are required for PMC migration. It is clear that these treatments have both morphological and biochemical effects on the ECM of the blastocoel (Katow and Solursh, 1979; Solursh and Katow, 1982; Solursh et al., 1986), but how these matrix defects influence PMC migration is unclear. Initial results indicate that the PMC do exhibit altered spreading and migratory behavior in the presence of defective matrix, and these effects are currently under investigation. The only conclusion pertaining to the role of the matrix that can be drawn from the in vitro migration assay results is that the availability of normal matrix does not compensate for the defects in the PMC caused by exogenous xylosides and sulfate deprivation. Roles for proteoglycans in cell-matrix interactions, including migration, have been implicated in a number of systems. Lark and Culp (1983) found that the initial attachment site of murine fibroblasts to fibronectincoated dishes was enriched in a large heparan sulfate proteoglycan, while the “old” footpad adhesion site left at the posterior end of the cell consisted of smaller heparan sulfate-containing molecules. The composition of footpads had been previously reported to change with

86

DEVELOPMENTAL BIOLOGY TABLE2 S pqnwatus

Cell-matrix combination’

Benson suggested the experiment with sulfur-containing amino acids, and C. Ettensohn the identification of PMC utilizing FITC-WGA. This work was supported by NIH Grants HD16549 and GM07228, and by the Graduate College of the University of Iowa.

IN VITROMIGRATION ASSAYS No net movement*

Moved
Moved >l cell diameter*

VOLUME 127,1988

NC REFERENCES

Normal PMC on normal matrix Normal PMC on xyloside matrix Xyloside PMC on normal matrix Xyloside PMC on xyloside matrix Xyloside PMC on normal matrix plus urea extract of normal PMC Xyloside PMC on normal matrix plus urea extract of xyloside PMC Xyloside PMC on normal matrix plus urea extract of xyloside PMC, returned to ASW for 2 hr Xyloside PMC on normal matrix plus urea extract of sulfatedeprived PMC

51

28

21d

129

12

59

29

34

6

93

1

70

6

94

0

36

14

59

27

22

12

88

0

25

9

63

28

57

0

100

0

20

‘Two to six trials were performed for each cell-matrix combination used, except for xyloside PMC + xyloside extract. *Movement categories are discussed under Methods and Materials. ‘All figures stated are percentages, except N, which was the number of cells observed. dBy Tukey’s HSD test, there is no effect of xyloside-generated or normal matrix on the migratory behavior of either normal or xyloside-treated cells (p < 0.05). However, xyloside-treated cells differed significantly from normal cells (p < 0.05).

time, from high levels of heparan sulfate and low levels of chondroitin sulfate during the early attachment phase, to lower levels of heparan sulfate and higher chondroitin sulfate (Rollins and Culp, 1979). Studying chick cardiac mesenchyme, Funderburg and Markwald (1986) found that heparan sulfate proteoglycan was distributed pericellularly on cells migrating in collagen lattices, and that the migrating mesenchyme left behind a mesenchyme-specific chondroitin sulfate proteoglycan that conditioned the substratum. Additional studies, particularly biochemical characterization of the important molecules, are needed to clarify the role of surface-associated proteoglycans in the migratory behavior of embryonic cells. The authors thank S. Mitchell, K. Venkatasubramanian, and L. Peterson for their assistance and encouragement, and C. K. Brown for statistical analysis. We also acknowledge the contributions made by the anonymous reviewers, whose suggestions significantly improved the manuscript. Very special thanks to S. K. Frost. These results were first presented at the Twenty-sixth Annual Meetings of the American Society for Cell Biology, December 7-11,1986, in Washington, D.C. S.

AKASAKA, K., AMEMIYA, S., and TERAYAMA, H. (1980). Scanning electron microscopical study of the inside of sea urchin embryos (Pseudocentrotus depressus). Effects of aryl @-xyloside, tunicamycin and deprivation of sulfate ions. Exp. Cell Res. 129, l-13. ETTENSOHN, C. A., and MCCLAY, D. R. (1987). A new method for isolating primary mesenchyme cells of the sea urchin embryo. Panning on wheat germ agglutinin-coated dishes. Ezp. Cell Res. 168,431-438. FUNDERBURG, F. M., and MARKWALD, R. R. (1986). Conditioning of native substrates by chondroitin sulfate proteoglycans during cardiac mesenchymal migration. J. Cell Biol. 103,24’75-2487. GEZELIUS, G. (1976). Effects of sulphate deficiency on the RNA synthesis of sea urchin larvae. Zoon 4,43-46. GUSTAFSON, T., and KINNANDER, H. (1956). Microaquaria for timelapse cinematography studies of morphogenesis in swimming larvae and observations on sea urchin gastrulae. Exp. Cell Res. 11, 36-51. GUSTAFSON, T., and WOLPERT, L. (1961). Studies on the cellular basis of morphogenesis in the sea urchin embryo. Directed movements of primary mesenchyme cells in normal and vegetalized larvae. Exp. Cell Res. 24, 64-79. GUSTAFSON, T., and WOLPERT, L. (1963). The cellular basis of morphogenesis and sea urchin development. In “International Review of Cytology” (G. H. Bourne and J. K. Danielli, Eds.), Vol. 15, pp. 139-214. Academic Press, New York. HARKEY, M. A., and WHITELEY, A. H. (1980). Isolation, culture and differentiation of echinoid primary mesenchyme cells. Wilhelm RQUX’S Arch. Dev. Biol. 189,111-122. HERBST, C. (1904). Uber die zur entwicklung des seeigellarven notwendigen anorganischen stoffe, ihre rolle und vertretbarkeit. II. Teil. die Rolle der notwendigen anorganischen stoffe. Wilhelm Roux’s Arch. Dev. Biol. 17, 306-520. IMMERS, J. (1961). Comparative study of the localization of incorporated 14C-labelled amino acids and %04 in the sea urchin ovary, egg, and embryo. Exp. Cell Res. 24,356-378. IMMERS, J., and RUNNSTROM, J. (1965). Further studies of the effects of deprivation of sulfate on the early development of the sea urchin Paracentrotus lividus. J. Embryol. Exp. Morphol. 14,289-305. IWATA, M., and NAKANO, E. (1984). Cell-to-substratum adhesion of dissociated embryonic cells of the sea urchin, Pseudocentrotus depressus. Wilhelm Roux’s Arch. Dev. Biol. 193, 71-77. KARP, G. C., and SOLURSH, M. (1974). Acid mucopolysaccharide metabolism, the cell surface, and primary mesenchyme cell activity in the sea urchin embryo. Dev. Biol. 41,110-123. KATOW, H., and SOLURSH, M. (1979). Ultrastructure of blastocoel material in blastulae and gastrulae of the sea urchin Lytechinuspictus. J Exp. Zoot. 210,561-567. KATOW, H., and SOLURSH, M. (1981). Ultrastructural and time-lapse studies of primary mesenchyme cell behavior in normal and sulfate-deprived sea urchin embryos. Exp. Cell Res. 136,233-245. LAEMMLI, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 227, 680-685. LARK, M. W., and CULP, L. A. (1983). Turnover

of heparan

sulfate

LANE AND SOLURSH

Primary

proteoglycans from substratum adhesion sites of murine fibroblasts. J. Biol. Chem. 259, 212-217. NORBECK, B. S. (1983). “Composition of the Blastocoel Contents of the Sea Urchin Lytechinus p&us.” M. S. Thesis, University of Iowa. PARISH, R. W., SCHMIDLIN, S., and MULLER, U. (1977). The effects of proteases on proteins and glycoproteins of Dictyostelium discoi&urn plasma membranes. Exp. Cell Res. 110,267-276. ROLLINS, B. J., and CULP, L. A. (1979). Glycosaminoglycans in the substrate adhesion sites of normal and virus-transformed murine cells. Biochemistry 18, 141-148. RTJNNSTR~M, J., H~RSTADIUS, S., IMMERS, J., and FUDGE-MASTRENGELO, M. (1964). An analysis of the role of sulfate in the embryonic differentiation of the sea urchin (Paracentrotus lividus). Rezlue Suisse 2001. 71, 21-54. SOLURSH, M. (1986). Migration of sea urchin primary mesenchyme

Mesenchyme

Cell Migration

87

cells. In “Developmental Biology: A Comprehensive Synthesis,” Vol. 2, pp. 391-431. Plenum, New York. SOLURSH, M., and KATOW, H. (1982). Initial characterization of sulfated macromolecules in the blastocoels of mesenchyme blastulae of Strcmgylocentrotua purpuratzis and Lytechinus pi&us. Dev. Biol. 94,326-336. SOLURSH, M., MITCHELL, S. L., and KATOW, H. (1986). Inhibition of cell migration in sea urchin embryos by fi-D-xyloside. Dev. Biol. 118, 325-332. SUGIYAMA, K. (1972). Occurrence of mucopolysaccharides in the early development of the sea urchin and its role in gastrulation. Dezl. Growth Difler. 14, 63-73. VENKATASUBRAMANIAN, K., and SOLURSH, M. (1984). Adhesive and migratory behavior of normal and sulfate-deficient sea urchin cells in vitro. Exp. Cell Res. 154, 421-431.