The program of protein synthesis during the development of the micromere-primary mesenchyme cell line in the sea urchin embryo

DEVELOPMENTAL

BIOLOGY

100,

12-28 (1983)

The Program of Protein Synthesis during the Development of the Micromere-Primary Mesenchyme Cell Line in the Sea Urchin Embryo MICHAEL A. HARKEY AND ARTHUR H. WHITELEY Department of Zoology and The Friday Harbor Laboratories, University of Washington, Seattle, Washington 98195 Received September 2, 1982;accepted in revised form May 31, 1983 Changes in the pattern of protein synthesis were analyzed during the in vitro development of the micromere-primary mesenchyme cell line of the sea urchin embryo. Micromeres were isolated and cultured from 16-cell stage embryos, and primary mesenchyme cells were isolated and cultured from early gastrulae. Both cell isolates developed normally in culture with about the same timing as their in situ counterparts in control embryos. Newly synthesized proteins were labeled with rH]valine at several stages of development and were analyzed by two-dimensional polyacrylamide gel electrophoresis and fluorography. The electrophoretic pattern of labeled proteins changed dramatically during development. More than half of the analyzed proteins underwent qualitative or quantitative changes in their relative rates of valine incorporation and these changes were highly specific to this cell line. Almost all of the changes were initiated prior to gastrulation and many prior to hatching. The highest frequency of changes in the micromere pattern of protein synthesis occurred between hatching and the start of gastrulation. This peak of activity coincided with the normal time of ingression of the primary mesenchyme and preceded the differentiation of spicules by more than 30 hr. Most of the observed changes were characterized as either decreases in the synthesis of proteins that showed maximum incorporation at the 16-cell stage or increases in the synthesis of proteins that showed maxima in the fully differentiated cells. Very few proteins exhibited transient synthetic maxima at intermediate stages. Thus, the program of protein synthesis associated with the development of micromeres consists largely of a switch in emphasis from early to late proteins, with the primary time of switching being between hatching and the onset of gastrulation.

that is essentially identical to that synthesized by other cells in the embryo (Chamberlain, 1977; Tufaro and Brandhorst, 1979; Harkey and Whiteley, 1982a) and very similar to that of unfertilized eggs (Brandhorst, 1976; Tufaro and Brandhorst, 1979). By the early gastrula stage (about 20 hr prior to spicule differentiation), the primary mesenchyme synthesizes a specialized population of proteins, including some that are totally absent from the synthetic programs of other embryonic cells (Harkey and Whiteley, 198213).Thus, the development of the micromere-primary mesenchyme cell line must involve cell-specific changes in protein synthesis. Furthermore, at least some of these changes must occur during a period that is distinct from both the time of commitment (16-cell stage) and the time of terminal differentiation (early prism) of these cells. In this report we present detailed examination of the program of protein synthesis associated with the development of isolated micromeres and primary mesenchyme cells, using two-dimensional electrophoresis and fluorography of [3H]valine-labeled proteins. We describe the extent, timing and cell specificity of the changes in protein synthesis that occur during the entire developmental sequence of the micromere-primary mesenchyme cell line, from the 16-cell stage to the differentiation of spicules.

INTRODUCTION

The micromeres of sea urchin embryos provide a particularly useful system for the study of gene expression in development. They are distinguished at the 16-cell stage as four small cells at the vegetal pole of the embryo. They are the progenitors of the primary mesenchyme of posthatching embryos, and ultimately they produce the magnesium calcite skeleton of the pluteus larva. Micromeres are fully committed to this pathway at the 16-cell stage; they can be isolated en musse and they will differentiate in vitro according to their normal developmental program (Okazaki, 1975a). At the gastrula stage, the primary mesenchyme cells can be isolated en masse, and these cells also differentiate normally in vitro (Harkey and Whiteley, 1980). In addition, differentiated primary mesenchyme cells can be partially purified from more advanced embryos. The micromere-primary mesenchyme cell line, therefore, offers a unique opportunity to study the sequence of molecular events occurring within a single cell type throughout its entire development. Protein synthesis has been examined at two stages in the development of this cell line using two-dimensional polyacrylamide gel electrophoresis. At the 16-cell stage, micromeres synthesize a population of proteins 12 0012-1606/83 $3.00 Copyright All rights

0 1983 by Academic Press, Inc. of reproduction in any form reserved

HARKEY AND WHITELEY MATERIALS

AND

13

Protein Synthesis in Micromeres

METHODS

Embryos of Strongylocentrotus purpuratus (Stimpson) were used in all experiments. Rearing of embryos and isolation and culture of primary mesenchyme cells from early gastrulae were carried out as described by Harkey and Whiteley (1980). All culturing was done at 9-10°C. Isolation and culture of micromeres. Fertilization membranes were removed from S-cell embryos by the method of Showman and Foerder (1979). When embryos reached the 16-cell stage (about 7.5-8.0 hr) they were chilled to 4°C to prevent further cleavages. Some of the embryos were set aside as controls and the rest were dissociated into single cells by four washes in calciumand magnesium-free artificial seawater (CMFSW, Tyler, 1953), followed by several passages through 54 pm Nitex netting. The cells from 20 ml packed embryos were suspended in 200 ml CMFSW and layered over a prechilled (6°C) 600-ml linear sucrose gradient composed of 1 m sucrose and calcium-free seawater (Tyler, 1953) in the ratios of 1:20 at the top and 3:7 at the bottom. After 1.5 hr of sedimentation at lg, micromeres were harvested by gentle aspiration from a band just below the top of the gradient and were collected by centrifugation at 600g for 5 min over a 1 m sucrose cushion. Typically, 20 ml of packed embryos yielded 0.15 ml of packed, 98% pure micromeres by this procedure. The micromeres were returned to seawater gradually since many of the cells lysed upon direct resuspension in seawater. They were washed sequentially in 3:1, l:l, and 1:3 mixtures of CFSW and Millipore-filtered (45 pm) seawater containing 250 kg/ml streptomycin sulfate (MFSW + S), then resuspended in MFSW + S containing 2% horse serum (Okazaki, 1975a). The cells were collected from each wash by sedimentation at SOOgfor 2 min. Micromeres and control embryos, which were cultured in MFSW + S and subjected to identical temperature regimens throughout the procedure in order to maximize synchrony, were returned to 9°C for culturing. Micromeres were maintained in 60-mm plastic petri dishes at a density of about 15 ~1 of packed cells per dish and embryos were maintained in MFSW + S in loo-mm dishes at 100 ~1 packed embryos per dish. Partial embryos, lacking viable ectoderm and enriched for primary mesenchyme cells, were prepared from prism and pluteus stage larvae in two ways. Glycine procedure: ectoderm was dissociated in 1 Mglycine, 100 PLMEDTA, pH 8.0, and the remaining partial embryos were purified by several cycles of differential centrifugation (Harkey and Whiteley, 1980). SDS procedure: in order to kill ectodermal cells selectively, embryos were suspended for 30 set in seawater containing 0.033% (w/v) SDS, then quickly hand centrifuged through a layer of 15% 1 m sucrose, 85% seawater onto a cushion

of 1 m sucrose and washed five times in 40 volumes of MFSW + S. Both types of partial embryo preparations were suspended at a concentration of 10% (v/v) in MFSW + S containing 2% horse serum and cultured as l-ml drop cultures for immediate labeling. Labeling,

extraction,

and electrophoresis

of proteins.

Cultured cells and embryos were prepared for labeling by two washes in their respective culture media, followed by reduction of the total volumes to l-2 ml/dish. Proteins of cells and embryos were labeled at various times after isolation with [3H]valine for 5-hr pulses at 200 &i/ml, extracted in SDS buffer, separated by two-dimensional polyacrylamide gel electrophoresis, and detected by fluorography (Harkey and Whiteley, 1982b). Pluteus epithelium protein synthesis was monitored by labeling and extracting proteins in intact pluteus embryos. These embryos incorporate [3H]valine preferentially into ectodermal and, to a lesser extent, into endodermal cell proteins with almost no incorporation by mesenchyme cells (Karp and Weems, 1975; Harkey and Whiteley, 1982; unpublished data). Therefore, these labeled wholeembryo preparations represent synthetic activity specifically of the ecto-endodermal epithelium. Unless stated otherwise, the fluorographs shown represent about 1 X 10” total disintegrations of [3H]valine considering both the number of incorporated counts loaded onto the gels, and the exposure time of the fluorographs. Analysis of developmental changes in the pattern of newly synthesized proteins. Fluorographs of two-dimen-

sional gels of newly synthesized proteins were prepared from eight stages of developing micromere cultures (see Fig. 1E). A population of spots was selected for analysis from the developmental series of fluorographs in the following manner in an attempt to include all the major spots from all stages. For each of the eight stages, 100 of the most intense and clearly resolvable spots were selected. Each of the spots selected at a given stage was screened for analyzability at the other stages. Those spots whose presence or absence was equivocal at some stages were not considered further. By this process, 161 analyzable and electrophoretically distinct proteins were selected from the total of eight labeled stages. Each spot was then followed through the developmental sequence of fluorographs and a decision was made at each interval in the sequence as to whether that spot exhibited an increase, a decrease, or no change in relative intensity. These decisions were made subjectively as described by Harkey and Whiteley (198213).Relative intensity changes were recorded only if: (1) they were reproducible between separate experiments and between separate analysis of the fluorographs, and (2) they appeared to represent a relative change of twofold or greater. In cases of gradual changes extending over several intervals, detectable subthreshold increments of

14

DEVELOPMENTAL BIOLOGY

VOLUME 100, 1983

that change observed at each interval were also recorded. This method of analysis was designed to indicate general trends in the program of changing gene expression. Analysis of changes in protein synthesis during development of primary mesenchyme cells or epithelial cells from early gastrulae was performed in a similar manner. Comparison of primary mesenchymal and epithelial patterns of protein synthesis was made as described elsewhere (Harkey and Whiteley, 1982b); the same criteria for determining a diflerence in relative intensity of a spot were applied in all comparisons.

species (Okazaki, 19’75a), isolated micromeres from 5’. 16-cell stage embryos developed in culture with about the same schedule as in control embryos (Figs. lA, C). Initially the isolated cells formed loose aggregates (Fig. 2A) that drifted in the medium or settled but did not attach to the surfaces of the plastic petri dishes. The cells within these aggregates cleaved during the normal cleavage period. At 43-55 hr, the aggregates attached to the bottoms of the dishes. The individual cells became flattened, mobile, and widely dispersed, but maintained a network of intercellular connections derived from filopodial activity (Figs. ZB, C). The timing and character of this acquisition of mesenchymal behavior were very similar to those of the normal processes of ingression and migration of primary

purpuratus

RESULTS

Development of isolated micromeres and primary mesenchyme cells in culture. As previously reported for other

START OF GASIRULAIION HATCHING DEVELOPMENT

‘:‘;““‘I

1

INGRE:;‘;;

lMTATl:I

“““‘:::::lE

GROWTH

OF MICROMERES IN vgyg

l

I

DEVELOPMENT

(hours)

0

DEVELOPMENT OF

MICROMERES &4

VU SPICULE

RUDIMENTS

ATTACHMENT

@ DEVELOPMENT

OF

PRIMARY IS01 A!:7

MESENCHYME IN

0

0

MIGRATION

T SPICULE RUDIMENTS

V/

STAGES (labeling

‘7’

I

I

1 I

ml 1

1

I

1

SPICULL

I

I

I

I

1

1 I

I I

I 8

1 I

1 I

I 1

DE

EF

GROWIH

,

ANALYZED pulses

)

DEVELOPMENTAL INTERVALS

STUDIED

m

Ellm

AB

cam

BC CD

ml

FG

I

I I

GH I

FIG. 1. Schedules of development and of labeling of embryos and isolated cells. (A) Normal development of micromeres within the intact embryo. (B) Time of development. Time zero was the time of fertilization. Time was not advanced during the isolations of micromeres or primary mesenchyme cells since both the cells and the control embryos were developmentally arrested during these procedures by low temperature (see Materials and Methods). (C) Development of isolated micromeres in culture. (D) Development in culture of primary mesenchyme cells isolated from early gastrulae. (E) Stages analyzed. Each box represents a 5-hr period during which cells or embryos were labeled for two-dimensional gel analysis of newly synthesized proteins. (F) Developmental intervals during which changes in protein synthesis were observed. Each interval represents the time between the midpoints of two consecutive labeling periods.

HARKEY AND WHITELEY

Protein Synthesis

in Micromeres

15

FIG. 2. Development of isolated micromeres (A-D) and isolated primary mesenchyme cells (E, F) in culture. (A) Cluster of micromeres at 13 hr of development. (B) Migration of micromere-derived cells from a cluster at 48 hr. (C) Formation of typical mesenchymal associations at 58 hr. (D) Spicule growth at 100 hr. (E) Primary mesenchyme cells immediately after isolation at the early gastrula stage (48 hr). Note the bag of ectodermal basal lamina (bl) surrounding each cluster of cells. (F) Spicule growth by isolated primary mesenchyme cells at 98 hr. Note that each “bag” of cells, derived from one embryo, is producing a spicule. Bars represent 20 pm for A-D and 40 pm for E and F.

mesenchyme cells in control embryos. The isolated cells began spicule formation between 65 and 78 hr, also in approximate synchrony with control embryos, and the spicules continued to grow (Fig. 2D). An important deviation from the normal developmental program was that spicule formation did not occur uniformly throughout the culture. In many cells and cell clusters birefringent granules appeared but soon

disappeared or failed to grow into spicules. Other cells showed no sign at all of spicule formation. Generally lo-20% of the micromeres became associated with growing spicules. Thus, in order to corroborate findings from late micromere cultures, primary mesenchyme cells were isolated and cultured from early gastrulae. As described elsewhere (Harkey and Whiteley, 1980), 95-100s of these cells differentiates spicules on schedule with

FIG. 3. Patterns of protein synthesis during the development of isolated micromeres. Panels A-H, corresponding to the labeling periods indicated in Fig. lE, are identified according to the stage of control embryos at the time of labeling. Each fluorograph represents approximately 1 X 10” total disintegrations of tritium. A positional reference grid, produced by connecting a set of reference proteins (Harkey and Whiteley, 198213) is displayed on each fluorograph. Certain specific proteins are identified as pm (specific to or enriched in primary mesencbyme cells), c (common to primary mesenchyme and epithelial cells of gastrulae or older embryos), or E (“early proteins” which are specific to or enriched in the earliest stages of developing micromeres). These designations are followed by the approximate molecular weight in kilodaltons. Actin (act), (Y tubulin ((YT), and @ tubulin (pT) have been identified previously (Harkey and Whiteley, 1982b). (A) The undifferentiated pattern immediately after isolation. Circles indicate 25 of the most intensely labeled spots at this stage, chosen on the basis of reproducibility, resolvability, and intensity. Arrowheads indicate all of the observed “early proteins,” those that decrease in relative intensity during development. (D) Arrowheads indicate all of the observed “intermediate proteins,” those that undergo transient changes during development. (H) The differentiated pattern. Circles indicate the 17 major labeled proteins of this stage. Arrowheads indicate all of the observed “late proteins,” those that increase in relative intensity during development.

16

DEVELOPMENTALBIOLOG

HARKEY AND WHITELEY

PT-ottin

Synthesis

in Micromeres

17

18

DEVELOPMENTALBIOLOGYVOLUMEloo,1983

control embryos (see Figs. 2E, F, and schedule in Fig. lD), thereby providing homogeneous populations of micromere-line cells during postgastrula stages of development. The extent of change in protein synthesis during micrmere development. Figure 3 is a series of fluorographs of two-dimensional polyacrylamide gels representing the newly synthesized proteins at eight stages in the development of cultured micromeres. The isolated cells were labeled with [3H]valine according to the schedule illustrated in Fig. 1E and processed for electrophoretic analysis as described under Materials and Methods. By comparison of the fluorographs from freshly isolated (Fig. 3A) and differentiated (Fig. 3H) cultures, it is clear that the pattern of protein synthesis changed dramatically during the development of these cells. An examination of the behaviors of 161 individual spots in these fluorographs (detailed below) revealed that over 50% of the detectable proteins changed in relative labeling intensity during development. Of the 83 observed changes, 53 were quantitative and 30 were qualitative with respect to the sensitivity range of the fluorographs. When this analysis was limited to the most intensely labeled proteins, the proportion that changed was much higher. Of the 17 most intensely labeled proteins from differentiated cultures (Fig. 3H; circles), 9 were undetectable and 6 more were labeled at much reduced intensities in the freshly isolated cells. Conversely, of the 25 major proteins indicated in the “early” pattern of synthesis (Fig. 3A; circles), 9 were undetectable and 11 were of reduced relative intensity in the “late” or differentiated pattern. Another important difference between the early and late patterns involved the distribution of label between major and minor spots. The label in the early pattern was relatively dispersed among a large number of moderately intense spots, while in the late pattern, it was heavily concentrated in a few major spots. The 17 major spots indicated by circles in Fig. 3H contained the vast majority of incorporated label in the late pattern. Significantly, 13 of these 17 spots (labeled “PM”) represented proteins whose synthesis is specific to or greatly enriched in primary mesenchyme cells as opposed to the other cells of the embryo (Harkey and Whiteley, 1982b; also see below). Thus, an important feature of the development of the micromere-primary mesenchyme cell line is a focusing of protein synthetic activity on the production of a few cell- and stage-specific proteins. The schedule of changes in protein synthesis. A detailed analysis of the changing pattern of protein synthesis during micromere development was carried out by following the relative intensities of 161 individual spots through the developmental series of fluorographs. These spots were selected, as described under Materials and

Methods, such that they represented the major, wellresolved, and reproducible species from all eight stages. For each interval in the series, each spot was scored as having increased, decreased, or remained unchanged in relative intensity. Only those changes that were reproducible in separate experiments and that exceeded a subjectively determined threshold of 2X were considered real. As an example of a minimal detectable change, we considered the apparent increase in intensity of PM32 between the prehatched (Fig. 3B) and hatched (Fig. 3C) blastula stages to be 2X or greater. Of the 161 analyzed proteins, 83 underwent changes in relative labeling intensity during micromere development. These modulated gene products fell into three major categories as illustrated in Fig. 4 which shows several major examples. The first category contained proteins that showed their highest relative intensities at the 16-cell stage and underwent decreases during development (Fig. 4A). These proteins, designated as “early” proteins are identified in Fig. 3A. The second category, “late” proteins (Fig. 4C), exhibited increases during development and maximum relative intensities in the fully differentiated cells (identified in Fig. 3H). Ninety percent of the modulated proteins were either of the early (56%) or late (34%) types. Thus, the vast majority of modulated proteins underwent simple increases or decreases that were never reversed. The third category, “intermediate” proteins (Fig. 4B) underwent transient fluctuations during intermediate stages. Only nine proteins (identified in Fig. 3D) belonged to the intermediate group. Six of these exhibited maximum relative intensities, two showed minima, and one showed both during intermediate stages (see examples in Fig. 4B). Most prominent among the transient proteins was PM65, which labeled intensely only at the mesenchyme blastula stage (Fig. 3D). The major actin variant of these embryos exhibited the most complex sequence of changes of all the proteins studied, including a transient rise in relative intensity at about the time of hatching, followed by a decrease to almost zero at the early gastrula stage and a second rise during later development. The modulated proteins exhibited a variety of individual schedules of change within each of the three major categories (Fig. 4). Some changed abruptly (e.g., E50, PM36, C33), while others changed gradually over several intervals (e.g., PMlOO, E75). In many cases a change was initiated early, followed by a period of no detectable change, and finally, a late continuation of the original change (e.g., PM45, PM52). More than one-third of the observed changes occurred over extended periods of three or more intervals. Some changes began as early as the first interval (e.g., E45, PM28, PM32), while a few did not occur until the last interval (e.g., PM29). The

HARKEY

A.

EARLY

AND WHITELEY

Protein

PROTEINS Stage

Synthesis

c.

19

in Micromeres

LATE

PROTEINS stage

ABCDEFGH

ABCDEFGH

PMlOd

PM100

PM94 PM50 PM35

PM 52

PM49

B.

INTERMEDIATE

PROTEINS

PM45 c55 PM36 c33 P~32 PM28

PM29

FIG. 4. The variety of schedules of change in the synthesis of individual proteins during the development of cultured micromeres. Examples of the three major classes of modulated proteins are shown. Individual proteins are represented by horizontal bands extending through the eight stages (A-H) of development indicated in Fig. 1F. An increase or decrease in the thickness of the band indicates an increase or decrease, respectively, in the relative rate of synthesis of the protein. No quantitative relationship between synthesis rate and band thickness is intended. The first observed instance of an increase (A) or decrease (7) defines the initiation of that change. (A) Early proteins are those that only decrease during development. Example proteins indicated on the right are identified in Fig. 3A. (B) Intermediate proteins are those that undergo transient changes. Example proteins are identified in Fig. 3D. (C) Late proteins are those that only increase during development. Example proteins, which are identified in Fig. 3H, include all of the major (circled) late proteins.

majority of changes began somewhere between these extremes (e.g., PM36, PM49). These changes are summarized in Fig. 5. The upper curve represents the total numbers of proteins that underwent changes in relative labeling intensity (both initial and continuing changes) during each interval, normalized by the length of that interval. This provides an indicator of the frequency of changes throughout development. The data clearly show that a major peak in the frequency of modulating activity occurred between hatching and the start of gastrulation. Initiations of changes (lower curve) were even more sharply restricted to this period of development. Of all of the changes observed, 85% were initiated prior to gastrulation and over 50% were initiated between hatching and gastrulation.

The major proteins that dominated the labeling pattern of fully differentiated micromeres (Fig. 3H; circles) were of particular interest since their intense labeling was largely cell- and stage-specific. Fourteen of these 17 major proteins were classified as late proteins (arrowheads). Three general observations were made concerning the schedules of change among this group of proteins. First, some major late proteins were synthesized prior to hatching. For example, PM32, PM94, and PM28, which could not be detected at the 16-cell stage (Fig. 3A), were clearly labeled 12 hr later at the prehatched blastula stage (Fig. 3B). At this time PM94 and PM28 were already among the most intensely labeled proteins in the cells. Second, all of the major late proteins exhibited de nova or increased labeling before the start of gastrulation (see Fig. 4C). In this way they resembled

20

DEVELOPMENTALBIOLOGYV0~~~~100,1983 HOURS

OF

DEVELOPMENT

I

I

I

I

20

40

60

80

HATCHING

+

SPICULES +

I A8

I

I

RC CD INTERVAL

I

I

DE

EC

I

CC

I

GH

OF DEVELOPMENT

FIG. 5. The frequency of changes in the relative rates of synthesis of proteins during the development of cultured micromeres. (A) Frequency of total observed changes. For each interval of development defined in Fig. lF, the total number of observed changes was divided by the length of the interval. Total changes included both initiations (the first observed increase or decrease of a given labeled protein) and continuations (increases or decreases that follow similar changes in previous intervals). (A) Frequency of initiations of changes.

the general population of proteins. Third, most of them continued to increase in relative intensity during the later stages of development. This is clearly seen by comparison of Figs. 3F, G, and H; the major late proteins become increasingly prominent throughout this postgastrula period of development. In summary: (1) the vast majority of proteins that changed in relative labeling intensity during the development of micromeres exhibited maximum intensities at either the start or the end of the process; (2) the pattern of protein synthesis began to change prior to hatching; (3) the majority of changes were initiated prior

to gastrulation; (4) the frequency of changes exhibited a sharp peak between hatching and the start of gastrulation; (5) during postgastrula development, a largely cell-specific group of major proteins continued to increase in relative labeling intensity as compared to the general population ultimately dominating the pattern of protein synthesis; and (6) in contrast to the general trend, a small number of minor proteins initiated changes late in development, well after the major peak of modulating activity. Authenticity of the program of protein synthesis in cultured micromeres. The low efficiency of spicule for-

HARKEY AND WHITELEY

Protein Synthesis in Micrcmeres

mation by cultured micromeres suggested that a large fraction of these cells might not differentiate the normal skeletogenic pattern of protein synthesis. The observed pattern may represent an aberrant mode of gene expression unrelated to skeletogenesis. In order to test this possibility, we conducted a similar analysis of protein synthesis during the differentiation of primary mesenchyme cells isolated from early gastrulae. Since these later cells differentiate spicules synchronously in culture with nearly 100% efficiency, they provide a more reliable test for the later stages of development of this cell type. Figures 6A and B compare the patterns of protein synthesis in differentiated cultures of (a) micromeres isolated at the 16-cell stage and cultured for 90 hr and (b) primary mesenchyme cells isolated at the early gastrula stage and cultured for 50 hr. Both cultures had a developmental age of 98 hr (see Fig. 1) comparable to the early pluteus stage of whole embryos and both were engaged in spicule production. As the figure shows, the patterns of protein synthesis in these two types of cultures were remarkably similar. Only a single difference (M) was found to be reproducible in this comparison. Of particular importance, all of the major cell- and stage-specific proteins identified in differentiated micromere cultures were synthesized at similarly high relative rates in the uniformly differentiated primary mesenchyme cultures. In addition, the changes in protein synthesis that we have described for the postgastrula portion of development of micromere cultures were also observed during the development of primary mesenthyme cultures (data not shown). Thus the program of protein synthesis observed in cultured micromeres does not deviate significantly from that of the morphologically more normal cells despite their failure to initiate spicule deposition uniformly. In order to determine whether or not the changes observed in the cultured cells occur also during normal embryo development, methods were devised for labeling and analyzing proteins from primary mesenchyme cells that had differentiated in situ. Since these cells do not label efficiently in situ (Harkey and Whiteley, 1982b) and since this is thought to reflect the fact that the ectoderm restricts diffusion of valine into the blastocoel (Karp and Weems, 19X5), the ectoderm was either removed from or killed in advanced embryos just prior to labeling. Removal of the ectoderm from prisms and plutei with 1 Mglycine resulted in partial embryos composed of dissociated endoderm and mesoderm cells enclosed in a bag of basal lamina (Fig. 7A). Since the glycine procedure disrupted the normal shapes and associations of primary mesenchyme cells with the spicules, a second procedure was also used, one that resulted in a normal primary mesenchyme morphology. Embryos

21

were exposed briefly to dilute SDS under conditions that lysed most of the ectodermal cells without visibly damaging or dissociating internal structures (Figs. 7B-D). The gut of these partial embryos remained actively ciliated and continued to pass material for several days in culture. Likewise, the primary mesenchyme remained associated with the spicules and continued its normal behavior of extending and contracting filopodia. Partial embryos of both types were produced from early prism and pluteus embryos, labeled immediately with [3H]valine, and the labeled proteins analyzed by twodimensional electrophoresis and fluorography. The SDS- and glycine-treated partial embryos yielded essentially identical electrophoretic patterns. Figure 6C shows a typical fluorograph derived from SDS-treated partial plutei. Many of the major labeled mesenchymespecific proteins described above were identifiable in these heterogeneous cell preparations (arrows)., In most cases, these labeled proteins were visibly enriched in the partial plutei as compared to whole plutei (Fig. 6D), indicating that they were synthesized preferentially or exclusively by internal cells. The minor protein PM29, which appeared de nova at the pluteus stage in cultured micromeres and primary mesenchyme cells (Fig. 3H), showed the same stage-specificity in partial embryos: while partial prisms prior to spicule initiation did not label PM29 detectably (data not shown), partial plutei did (Fig. 6C, inset). These data are consistent with the conclusion that the program of protein synthesis observed in cultured micromeres cells is also carried out by these cells within the normally developing embryo. Early specialization of micromeres. When the pattern of newly synthesized proteins from whole 16-cell embryos was compared to that of freshly isolated micromeres from this stage, no reproducible differences were detectable (compare Figs. 8A and 3A). This result corroborates the report of Tufaro and Brandhorst (1979) that all three cell types of 16-cell embryos synthesize the same patterns of proteins. Since micromeres are fully determined at the 16-cell stage, it is of interest to know how early in development they begin to diverge from the other cells with respect to protein synthesis. Therefore, control embryos, composed mostly of nonmicromere cells, were reared and analyzed in parallel with isolated micromeres. Figures 8B and C show the respective electrophoretic patterns from prehatched blastulae and early mesenthyme blastulae. By comparison with the pattern from 16-cell stage embryos, it can be seen that major early increases occurred in whole embryos (vertical arrows), some before hatching, and that these changes were maintained or augmented with further development. Prominent examples were the dramatic early increases in labeling of the major electrophoretic variants of LY

22

DEVELOPMENTALBIOLC GY

VOLUMEloo,1983

HARKEY AND WHITELEY

Protein

Synthesis

in Micmmeres

23

FIG. ‘7. Partial pluteus embryos in which the ectoderm has been removed or destroyed. (A) Partial pluteus after treatment of the embryo with 1 M glycine. The ectoderm cells have been dissociated and dispersed, leaving a basal laminar bag containing skeletal spicules and a mixture of dissociated endoderm and mesoderm cells. (B-D) Partial plutei in which the ectoderm cells have been lysed by brief exposure to SDS. (B) Focus is on primary mesenchyme cells (PM) attached to a spicule. Note filopodia (fil) which were not disturbed by the treatment. (C) Focus is on the ectoderm, showing disruption of the cells. (D) Focus is on the gut, showing undisturbed tripartite structure with welldefined epithelial borders. Bar = 25 pm.

and p tubulin, and actin. Comparison of these patterns with those of micromere cultures (Figs. 3A-D) shows that most of these major changes did not occur in micromeres. For example, actin labeling increased only slightly and tubulin labeling decreased beyond detectability by mesenchyme blastula. Conversely, major increases or de nova appearances that were observed in the early development of micromeres were seen only as minor increases in whole embryos (Figs. 8B, C; horizontal arrows). For example, the major primary mesenchymal proteins PM28 and PM65 were minor proteins in the whole-embryo pattern at the mesenchyme blastula stage while PM32 was undetectable (Fig. 8C). As micromeres contain only l/30 of the volume of the entire embryo (based on measurements of cell diameters), these results are consistent with the conclusion that such proteins were synthesized exclusively or largely by the micromeres. When embryos were dissociated into single cells and cultured in the same way as micromeres, the patterns of protein synthesis were essentially of the whole-embryo type rather than the micromere type (data not shown). Therefore, the specificity of protein synthesis observed in cultured micromeres cannot be attributed to a general response by embryonic cells to dissociation

or culture. We conclude that the pattern of protein synthesis in micromeres begins to diverge from that of other embryo cells sometime between early cleavage and hatching. Continued divergence of the primary mesenchyme and epithelial programs. By the early gastrula stage, con-

siderable cell specificity is detectable between the derivative of the micromeres (the primary mesenchyme) and that of the mesomeres and macromeres (the gastrula epithelium). About 25% of the observed proteins exhibit quantitative or qualitative differences in relative labeling between these cell types (Harkey and Whiteley, 1982b). In order to determine whether or not the patterns of protein synthesis in these cells continue to diverge during later development, a similar analysis was performed at the pluteus stage. Using protein preparations from differentiated cultures of primary mesenchyme cells (e.g., Fig. 6A) and from pluteus epithelium of the same developmental age (e.g., Fig. SD), 209 labeled proteins were classified as: (1) cell-specific (labeled in only one of the two cell types), (2) cell-enriched (labeled more intensely in one cell type than the other), or (3) cellcommon (labeled equally in both cell types). The criteria for this classification are described elsewhere (Harkey and Whiteley, 1982b). The electrophoretic positions and

FIG. 6. The synthesis of primary mesenchymal proteins under various conditions of cell isolation and culture. Each fluorograph represents 1 X 10” total disintegrations. (A) Primary mesenchyme cells isolated from early gastrulae at 48 hr and labeled in culture at 98 hr (during spicule growth). (B) Micromeres isolated from 16-cell embryos at 8 hr and labeled in culture at 98 hr (during spicule growth). A single spot (M) is reproducibly observed in micromere but not primary mesenchyme isolates at this stage. (C) Partial 100-hr plutei. Primary mesenchyme cells were allowed to differentiate spicules within the intact embryo, then were enriched by selectively killing the ectoderm with SDS, and labeled immediately. The inset represents a threefold longer fluorographic exposure of the region around PM29 taken from glycine-treated partial plutei of the same age. (D) Intact lOO-hr plutei. Specific proteins are identified as in Fig. 3.

24

DEVELOPMENTALBIOLOGY

VOLUME 100, 1983

classifications of the analyzed proteins are indicated in Fig. 9. Over half of the proteins analyzed in pluteus cells belonged to the first two categories and thus exhibited cell-specific regulation. However, qualitative differences observed between the two cell populations at the pluteus stage (12% of analyzed proteins) were no more frequent than those observed at the early gastrula stage (Harkey and Whiteley, 1982b). Therefore, the primary mesenchyme and epithelial programs of protein synthesis continued to diverge between the early gastrula and pluteus stages, and most of this divergence was due to quantitative changes in the synthesis of individual proteins. DISCUSSION

FIG. 8. Patterns of protein synthesis of intact early embryos. (A) 16-cell stage embryos. (B) Prehatched blastulae. (C) Early mesenchyme blastulae. Vertical arrows indicate proteins that exhibited early increases in labeling in whole embryos but not in isolated micromeres. Horizontal arrows indicate the positions of proteins that exhibited early increases specifically in isolated micromeres. Specific proteins are identified as in Fig. 3.

In this report we have examined the program of protein synthesis associated with the development of micromeres into the skeleton-forming cells of the sea urchin embryo. This analysis was limited to the most intense, well-resolved spots on fluorographs of two-dimensional gels. Proteins such as histones with isoelectric points outside the pH 4.5-6.5 resolving range of the gels (Harkey and Whiteley, 1982b) and proteins synthesized at relatively low rates were not included. As the development of micromeres was studied in vitro, it was necessary to verify that they develop normally under these conditions. As Okazaki (1975a) has reported for several species of sea urchins, and as we have verified here for S. purpuratus, isolated micromeres develop normally and synchronously to the point of spicule formation. Subsequently, some of the cells engage in spicule formation but others do not. This lack of uniform spicule formation does not necessarily indicate abnormal development of the capacity to form spicules. Rather, it may simply reflect a nonpermissive aspect of the culture environment with respect to the deposition of skeletal materials. In this regard, Okazaki (1971a,b; 1975a) has shown that horse serum is required in the culture medium for the growth of spicules by isolated micromeres but not for the development of spiculeforming capacity. At the level of protein synthesis micromeres appear to differentiate normally in culture. They exhibit similar changes in protein synthesis during postgastrula development and the same terminally differentiated pattern of protein synthesis as cultured primary mesenchyme cells, which undergo uniform spiculogenesis. In addition, many stage- and cell-specific properties of the program of protein synthesis of these cells have been corroborated in viva (Harkey and Whiteley, 1982b; this report). We conclude that isolated micromeres develop in culture according to their normal program of gene expression. The 16-cell stage pattern of protein synthesis. We have corroborated the finding of Tufaro and Brandhorst (1979)

HARKEY AND WHITELEY

Protein Synthesis in Micromeres

25

pm 100 1

\

FIG. 9. Comparison of the proteins synthesized by primary mesenchyme cells and epithelial cells of the pluteus larva. The electrophoretic positions of the labeled spots included in the analysis are indicated. Each spot is classified as described in the text as: mesenchyme-specific (O), mesenchyme-enriched (0), epithelium-specific (A), epithelium-enriched (A), or cell-common (*).

that 16-cell stage micromeres synthesize the same array of proteins, in the same relative proportions, as do the other cells of the embryo. This pattern of protein synthesis is very similar to that of zygotes and unfertilized eggs (Brandhorst, 1976; Tufaro and Brandhorst, 1979). Thus, the formation of micromeres, an event that results in the strict determination of these cells to differentiate eventually both a highly specialized pattern of protein synthesis (Harkey and Whiteley, 1982a,) and skeletal spicules (Okazaki, 1975a), does not involve an immediate specialization of the pattern of synthesis of detectable proteins. Micromeres do differ from the other blastomeres of 16-cell stage embryos in several ways including: (1) the transcription of reiterated DNA sequences (Mizuno et al, 1974), (2) the complexities of nuclear and cytoplasmic RNAs (Rodgers and Gross, 1978; Ernst et al, 1980), and (3) the relative rates of synthesis of various histones (Senger and Gross, 1978; Senger et aL, 1978). The particular roles that these and other cell-specific properties may play in the determination of micromeres are not understood. Specialization

of the micromere-primary

mesenchyme

program More than 50% of the proteins analyzed during

the development of micromeres exhibit detectable changes in their apparent rates of synthesis relative to other proteins. This is a much more extensive change than that reported for whole embryos during the same period of development (Brandhorst, 1976). It is at least partially due to the fact that we have singled out a cell type that achieves a high degree of specialization early in the development of the embryo. More than 90% of the specific changes in protein synthesis that occur in primary mesenchyme cells during development do not occur in other embryonic cells (unpublished data). As a result, the cell types exhibit increasingly divergent patterns of protein synthesis. By the early gastrula stage, 28% of the analyzable proteins detected on twodimensional gels from ecto-endoderm and primary mesenchyme cell preparations show cell-specific labeling intensities (Harkey and Whiteley, 1982b). As shown here, this value rises to more than 50% by the pluteus stage. One aspect of this specialization is a focusing of protein synthesis on major late proteins. At the 16-cell stage, a large number of proteins are synthesized at moderate relative rates, producing a complex electrophoretic pattern. In contrast the fully differentiated pattern is dominated by a few major proteins, most of

26

DEVELOPMENTALBIOLOGY

which are both cell- and stage-specific, and whose rapid synthesis appears to engage most of the synthetic machinery of the cells. This focus of synthetic activity is probably a general phenomenon of differentiation reflecting the specialization of cells. Laskey et al (1980) have shown that a superabundant class of mRNA appears for the first time in the sea urchin embryo at the pluteus @age,when the embryo is organizing specialized tissues and organs. In mammals, terminal differentiation of myoblasts (Garrels, 1979), myeloblasts (Liebermann et aZ., 1980), erythroblasts (Affara et aZ., 1979), and skin basal cells (Balmain et ak, 1979) all involve the initiation of massive synthesis of a few major proteins that are of importance in the specialized functions of those cells. By analogy with these systems, it is probable that most of the major late proteins synthesized by differentiated micromeres are important in skeletogenesis, perhaps as components of a skeletal matrix or sheath (Okazaki, 1960), or as catalysts of the accumulation and deposition of calcium carbonate. The contributions of quantitative and qualitative changes. Over two-thirds of the changes in protein synthesis that occurred during the development of micromere-primary mesenchyme cell line were quantitative. Over four-fifths of the differences between primary mesenchyme and epithelial cells at the pluteus stage were also quantitative (see Fig. 9). The quantitative nature of these differences was not an artifact of asynchronous development or of impurity of the cultured cells since distinct qualitative differences were observed in the same preparations. Quantitative stage- and cellspecific differences in protein synthesis are encountered in a variety of other embryos including those of mammals (Garrels, 1979; Affara et ah, 1979; Klose and von Wallenberg-Pachaly, 1976), amphibians (Brock and Reeves, 1978; Bravo and Knowland, 1979), insects (Rodgers and Shearn, 1977; Sakoyama and Okubo, 1981), and slime molds (Alton and Lodish, 1977; Alton and Brenner, 1979; Coloma and Lodish, 1981). Taken together, these findings suggest that quantitative regulation of the synthesis of individual proteins is of general importance in determining both the cell-specific and stage-specific aspects of developing embryos. The micromere program of protein synthesis also exhibited several apparently qualitative cell- and stagespecific changes. It is interesting that qualitative changes were particularly prevalent among the major late proteins. Seventy percent of these proteins appeared de nova during micromere development, accounting for 75% of the total observed de nova appearances of all late proteins. Thus, it appears that while the majority of changes in protein synthesis in micromeres are regulated quantitatively, the major proteins that characterize the fully differentiated state of these cells are regulated in a predominantly qualitative fashion.

VOLUME loo,1983

The bimodal character of the program. An interesting aspect of the program of protein synthesis in developing micromeres was the predominance of early and late proteins as compared to intermediate ones. Early proteins can be thought of as characterizing the undifferentiated state of micromeres since they are synthesized ubiquitously among the blastomeres of 16cell embryos as well as in zygotes (Tufaro and Brandhorst, 1979). This nonspecific synthesis, taken together with the decline in relative rates of synthesis of early proteins during development, suggests that these proteins serve mostly general functions in early blastomeres rather than specialized functions related to the differentiation of micromeres. On the other hand, the fact that late proteins exhibited their highest relative rates of synthesis in differentiated micromeres strongly suggests that the primary roles of these proteins are related to the differentiated activity of spicule deposition. The seven intermediate proteins that exhibited maximal synthesis during intermediate stages are the only strong candidates for newly synthesized proteins whose primary functions are likely to be related specifically to the transitional period of micromere development. Therefore, the micromere program of protein synthesis consists mainly of two modes, undifferentiated and differentiated, with an intermediate period of transition. This result is interesting since the morphological development of these cells is not simply bimodal. Micromeres proceed through a series of distinct events including several cleavages, incorporation into the blastula epithelium, ingression, migration, ring formation, and spicule deposition (for reviews see Okazaki, 1975a,b; Harkey, 1983). With the possible exception of the few transient proteins, this series of events does not appear to be associated with a parallel sequence of discrete, stage-specific patterns of synthesis of detectable proteins. Spieth and Whiteley (1980) have demonstrated that the progression of morphological events in primary mesenchyme development requires a parallel program of processing of nuclear RNA into polyadenylated mRNA, suggesting that the translation of new mRNAs is necessary for each event. The execution of the intermediate developmental events may be regulated by the synthesis of minor or basic proteins, not detectable on two-dimensional gels, by the few intermediate proteins observed here, or by events other than protein synthesis. However, it would appear that the vast majority of changes in the synthesis of major proteins in developing micromeres is related, not to these intermediate events but to the differentiated product of that development. Temporal aspects of the program. Major changes in protein synthesis occur in the micromeres before the onset of hatching, and these changes are distinct from the equally dramatic changes that occur in the other blastomeres. These early mesenchyme-specific changes

HARKEY AND WHITELEY

Protein

include the appearances of proteins such as PM28 and PM32 that will be major constituents of the fully differentiated pattern of protein synthesis. At the time these major proteins appear, the cells are morphologically similar to all the other cells of the embryo: lo20 hr will pass before they emerge into the blastocoel as the primary mesenchyme, and an additional 30 hr are required for the initiation of spicule differentiation. Most of the changes are concentrated in the period of development prior to gastrulation, with a peak of activity between hatching and the start of gastrulation. This peak correlates with the time of ingression of primary mesenchyme in the normal embryo. As a result, all of the major proteins and most of the minor ones that are synthesized during the formation and growth of spicules are already being synthesized at the early gastrula stage. The peak of changes in protein synthesis in micromeres coincides with a peak of changes in gene expression in the embryo as a whole. A large body of evidence indicates that the period between hatching and gastrulation involves a shift from the expression of maternally derived mRNA to the expression of mRNA transcribed from the embryonic genome (for reviews see Whiteley and Whiteley, 1975;Davidson, 1976). Qualitative changes in protein synthesis are most frequent during this period (e.g., Terman, 1970; Westin et al, 1967). Two-dimensional electrophoretic analysis of several hundred major proteins in developing sea urchin embryos (Brandhorst, 1976; Bedard and Brandhorst, 1983) indicates that this shift in gene expression in the embryo occurs at or before the mesenchyme blastula stage. During this period between hatching and gastrulation, a family of ectodermspecific mRNAs increases by more than loo-fold (Klein et al., 1983), and the protein products of these messages accumulate rapidly in the ectoderm (Bruskin et ab, 1981, 1982). Thus, the peak of changes observed in developing micromeres may be related to a major, embryo-wide change in gene expression. During the period between hatching and gastrulation, the micromere-primary mesenchyme cell line undergoes dramatic morphological and physiological changes (for review see Harkey, 1983). At hatching, these cells are integrated members of the blastula epithelium. Shortly thereafter, they release their attachments to the hyalin layer and to neighboring cells, penetrate the basal lamina of the blastocoel wall, migrate into the blastocoel, and assume the characteristic morphology and behavior of the primary mesenchyme. Certain enzyme activities that have been histochemically localized in primary mesenchyme cells such as alkaline phosphatase (Hsiao and Fujii, 1963; Pfohl and Giudice, 1967) and acetyl cholinesterase (Ozaki, 1974) undergo initial increases at about the time of ingression. These cells also begin, exclusively, to secrete sulfated mucopoly-

Synthesis

in Micromeres

27

saccharides (Motomura, 1960; Immeres, 1961; Sugiyama, 1972) and to express cell-specific surface antigens (M&lay et aZ., 1983) at this time. Other activities of probable functional importance to skeleton formation such as collagen deposition (Pucci-Minafra et aL, 1972, 1975, 1980; Golob et aL, 1974), calcium uptake (Nakano et al., 1967), and carbonic anhydrase activity (Chow and Benson, 1979) also show rises during this period or shortly thereafter. The temporal correlation between the peak of changes in protein synthesis described here with these morphological and biochemical changes suggests a functional relationship. It appears that a massive coordinated switch in gene expression occurs at ingression of the primary mesenchyme and that this event is reflected immediately as a fundamental change in the physiology of these cells. We thank Dr. Merrill Hille, Dr. Gerald Schubiger, and Dr. Helen R. Whiteley for their critical readings and helpful discussions during the development of this manuscript. This work was supported in part by the following grants from the National Institutes of Health: Grant HL-I0312 to A.H.W., Grant GM-20784 to Dr. Helen R. Whiteley, Grant ES-02190 to N. Karle Mottet, M.D., and Training Grants ES-07032 and HD-00266. REFERENCES AFFARA,N., and DAUBAS,P. (1979). Regulation of a group of abundant mRNA sequences during Friend cell differentiation. Dew. BioL 72, 110-125. ALTON, T. H., and BRENNER,M. (1979). Comparison of proteins synthesized by anterior and posterior regions of Dictyostelium discoideum pseudoplasmodia. Dev. BioL 71, l-7. ALTON, T. H., and LODISH, H. F. (1977). Developmental changes in messenger RNAs and protein synthesis in Dictyostelium diswideum Dev. BioL 60,180-206.

BALMAIN,A., LOEHREN,D., FISCHER,J., and ALONSO,A. (1977). Protein synthesis during fetal development of mouse epidermis. I. The appearance of “histidine-rich protein.” Dev. BioL 60, 442-452. BALMAIN, A., LOEHREN,D., ALONSO,A., and GOERTTLER,K. (1979). Protein synthesis during fetal development of mouse epidermis. II. Biosynthesis of histidine-rich and cystine-rich proteins in vitro and in vivo. Dev. BioL 73,338-344.

BCDARD, P.-A., and BRANDHORST,B. P. (1983). Patterns of protein synthesis and metabolism during sea urchin embryogenesis. Dew. BioL 96, 74-83.

BRANDHORST,B. P. (1976). Two-dimensional gel patterns of protein synthesis before and after fertilization of sea urchin eggs. Dev. BioL 52, 310-317. BRAVO,R., and KNOWLAND,J. (1979). Classes of proteins synthesized in oocytes, eggs, embryos, and differentiated tissues of Xenopus 13, 101-108. laevis. D#krentiation BROCK,H. W., and REEVES, R. (1978). An investigation of de nova protein synthesis in the South African clawed frog, Xenopus Levis. Dev. BioL 66,128-141.

BRUSKIN,A. M., TYNER, A. L., WELLS, D. E., SHOWMAN,R. M., and KLEIN, W. H. (1981). Accumulation in embryogenesis of five mRNAs enriched in the ectoderm of the sea urchin pluteus. Dev. BioL 87, 308-318. BRUSKIN, A. M., B~DARD, P.-A., TYNER, A. L., SHOWMAN,R. M., BRANDHORST,B. P., and KLEIN, W. H. (1982). A family of proteins accumulating in ectoderm of sea urchin embryos specified by two related cDNA clones. Dev. BioL 91, 317-324.

28

DEVELOPMENTALBIOLOG,Y

CHAMBERLAIN,J. P. (1977). Protein synthesis by separated blastomeres of sixteen-cell sea urchin embryos. .I Cell Biol 75, 33a (abstract). CHOW,G., and BENSON,S. C. (1979). Carbonic anhydrase activity in developing sea urchin embryos. Exp. Cell Res. 124,451-453. COLOMA,A., and LODISH, H. F. (1981). Synthesis of spore- and stalkspecific proteins during differentiation of Dictyostelium discoideum. Dev. BioL 81,238-244.

DAVIDSON,E. H. (1976). “Gene Activity in Early Development,” 2nd ed. Academic Press, New York. ERNST, S. G., HOUGH-EVANS,B. R., BRITPEN, R. J., and DAVIDSON, E. H. (1980). Limited complexity of the RNA in micromeres of sixteen-cell sea urchin embryos. Dev. BioL 79,119-127. GARRELS,J. I. (1979). Changes in protein synthesis during myogenesis in a clonal cell line. Dev. Biol. 73, 134-152. GOLOB,R., CHETSANGA,C. J., and DOTY,P. (1974). The onset of collagen synthesis in sea urchin embryos. Biochim. Biophys. Acta 349,135141.

HARKEY,M. A. (1983).Determination and differentiation of micromeres in the sea urchin embryo. In “Time, Space, and Patterns in Embryonic Development” (W. R. Jeffery and R. A. Raff, eds.), pp. 131155. Liss, New York. HARKEY, M. A., and WHITELEY, A. H. (1980). Isolation, culture, and differentiation of echinoid primary mesenchyme cells. Wilhelm Roux’s Arch. 189,111-122.

HARKEY, M. A., and WHITELEY, A. H. (1982a). The translational program during the differentiation of isolated primary mesenchyme cells. Cell D@er. 11, 325-329 (extended abstract). HARKEY, M. A., and WHITELEY, A. H. (1982b). Cell-specific regulation of protein synthesis in the sea urchin gastrula: A two-dimensional electrophoretic study. Dev. BioL 93, 453-462. HSIAO, S. C., and FLJJII,W. K. (1963). Early ontogenic changes in the concentration of alkaline phosphatase in Hawaiian sea urchins. Exp. Cell Res. 32, 21’7-231. IMMERS,J. (1961). Comparative study of the localization of incorporated “C-labeled amino acids and %01 in the sea urchin ovary, egg, and embryo. Exp. Cell Res. 24, 356-378. KARP, G. C., and WEEMS, M. D. (1975). ‘H-amino acid uptake and incorporation in sea urchin gastrulae and exogastrulae: An autoradiographic study. J. Exp. Zoo1 194,535-546. KLEIN, W. H., CARPENTER,C. D., TYNER, A. L., SPAIN, L. M., ELDON, E. D., and BRUSKIN,A. M. (1983). A family of genes expressed in the embryonic ectoderm of sea urchins. In “Time, Space, and Patterns in Embryonic Development” (W. R. Jeffery and R. A. Raff, eds.), pp. 87-100. Liss, New York. KLOSE,J., and VON WALLENBERG-PACHALY,H. (1976). Changes of soluble protein populations during organogenesis of mouse embryos as revealed by protein mapping. Dev. BioL 51,324-331. LASKEY, L. A., LEV, Z., XIN, J., BRITTEN, R. J., and DAVIDSON, E. H. (1980). Messenger RNA prevalence in sea urchin embryos measured with cloned cDNAs. Proc. NatL Acad. Sci. USA 77,5317-5321. LIEBERMANN,D., HOFFMAN-LIEBERMAN,B., and SACHS,L. (1980). Molecular dissection of differentiation in normal and leukemic myeloblasts: Separately programmed pathways of gene expression. Dev. BioL 79, 46-63.

MCCLAY,D. R., CANNON,G. W., WESSEL,G. M., FINK, R. D., and MARCHASE,R. B. (1983). Patterns of antigenic expression in early sea urchin development. In “Time, Space, and Patterns in Embryonic Development” (W. R. Jeffery and R. A. Raff, eds.), pp. 157-169. Liss, New York. MIZUNO, S., LEE, Y. R., WHITELEY, A. H., and WHITELEY, H. R. (1974). Cellular distribution of RNA populations in 16-cell stage embryos of the sand dollar Dendraster excentricus. De-v. BioL 37, 18-27. MOTOMURA,J. (1960). Secretion of mucosubstance in the gastrula of the sea urchin embryo. Bull. BioL Stn Asamushi 10, 165-169. NAKANO, E., OKAZAKI, K., and IWAMATSU,T. (1963). Accumulation of radioactive calcium in larvae of the sea urchin Pseudocentrotus depressus. BioL Bull. 125, 125-133.

VOLUME 100, 1983

OKAZAKI,K. (1960).Skeleton formation of sea urchin larvae. II. Organic matrix of the spicule. Embrgologia 5.283-320. OKAZAKI, K. (1971a). In vitro culture of the micromeres and primary mesenchyme cells isolated from sea urchin embryos and larvae. In “Cells in Early Stages of Development” (edited by Japan. Sot. Develop. Biol.). Iwanami Shoten, Tokyo. [in Japanese] OKAZAKI, K. (1971b). Spicule formation in sea urchin larvae; observations in tivo and in vitro. Symp. Cell BioL 22,X3-171. [in Japanese] OKAZAKI, K. (1975a). Spicule formation by isolated micromeres of the sea urchin embryo. Amer. ZooL 15, 567-581. OKAZAKI, K. (1975b). Normal development to metamorphosis. In “The Sea Urchin Embryo” (G. Czihak, ed.), pp. 175-232. Springer-Verlag, Berlin. OZAKI, H. (1974). Localization and multiple forms of acetylcholinesterase in sea urchin embryos. Dev. Growth wer. 16,267-279. PFOHL, R. J., and GIUDICE, G. (1967). The role of cell interaction in the control of enzyme activity during embryogenesis. B&him. Bi+ phys. Acta 142,263-266.

PUCCI-MINAFRA,I., CASANO,C., and LA ROSA,C. (1972). Collagen synthesis and spicule formation in sea urchin embryos. Cell Dz$k-r. 1, 157-165.

PUCCI-MINAFRA,I., FANARA, M., and MINAFRA, S. (1980). Chemical and physical changes in the organic matrix of mineralized tissues from embryo and adult Paracentrotus lividus. J. Submicrosc. CytoL 12, 267-273.

PUCCI-MINAFRA,I., MINAFRA,S., GIANGUZZA,F., and CASANO,C. (1975). Amino acid composition of collagen extracted from the spicules of sea-urchin embryos (Paracentrotus lividus). Boll. ZooL 42,201-204. RODGERS,M. E., and SHEARN,A. (1977). Patterns of protein synthesis in imaginal discs of Drosophila melanogaster. Cell 12, 915-921. RODGERS,W. H., and GROSS,P. R. (1978). Inhomogeneous distribution of egg RNA sequences in the early embryo. Cell 14, 179-288. SAKOYAMA,Y., and OKUBO,S. (1981). Two-dimensional gel patterns of protein species during development of Drosophila embryos. Dev. BioL 81, 361-365.

SENGER,D. R., and GROSS,P. R. (1978). Macromolecule synthesis and determination in sea urchin blastomeres at the sixteen-cell stage. Dev. BioL 65, 404-415.

SENGER,D. R., ARCECI,R. J., and GROSS,P. R. (1978). Histones of sea urchin embryos. Transients in transcription, translation, and the composition of chromatin. Dev. BioL 65, 416-425. SHOWMAN,R. M., and FOERDER,C. A. (1979).Removal of the fertilization membrane of sea urchin embryos employing aminotriazole. Exp. Cell Res. 120, 253-256.

SPIETH, J., and WHITELEY, A. H. (1980). Effect of 3’-deoxyadenosine (cordycepin) on the early development of the sand dollar, Den&aster excentricus.

Dev. BioL 79, 95-106.

K. (1972). Occurrence of mucopolysaccharides in the early development of the sea urchin embryo and its role in gastrulation.

SUGIYAMA,

Dev. Growth mer.

14, 63-73.

TERMAN,S. A. (1970). Relative effects of transcription-level and translation-level control of protein synthesis during early development of the sea urchin. Proc NatL Acad Sci 65,985-992. TUFARO,F., and BRANDHORST, B. P. (1979). Similarity of proteins synthesized by isolated blastomeres of early sea urchin embryos. Dev. BioL 72, 390-397.

TYLER,A. (1953). Prolongation of life-span of sea urchin spermatozoa, and improvement of spermatozoa and eggs with metal-chelating agents (amino acids, versene, DEDTC, oxine, cupron). BioL Bull. 104, 224-239.

WESTIN,M., PERLMANN,H., and PERLMANN,P. (1967). Immunological studies of protein synthesis during sea urchin development. J. Exp. ZooL 166, 331-346.

H. R., and WHITELEY, A. H. (1975). Changing populations of reiterated DNA transcripts during early echinoderm development. “Current Topics in Developmental Biology” (A. A. Moscona and A. Monroy, eds.), Vol. 9, pp. 39-88. Academic Press.

WHITELEY,