Endocytosis in primary mesenchyme cells during sea urchin larval skeletogenesis

Author’s Accepted Manuscript Endocytosis in Primary Mesenchyme Cells during Sea Urchin Larval Skeletogenesis Christopher E. Killian, Fred H. Wilt

www.elsevier.com/locate/yexcr

PII: DOI: Reference:

S0014-4827(17)30400-7 http://dx.doi.org/10.1016/j.yexcr.2017.07.028 YEXCR10675

To appear in: Experimental Cell Research Received date: 4 April 2017 Revised date: 26 June 2017 Accepted date: 22 July 2017 Cite this article as: Christopher E. Killian and Fred H. Wilt, Endocytosis in Primary Mesenchyme Cells during Sea Urchin Larval Skeletogenesis, Experimental Cell Research, http://dx.doi.org/10.1016/j.yexcr.2017.07.028 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Endocytosis in Primary Mesenchyme Cells during Sea Urchin Larval Skeletogenesis

Christopher E. Killian, Fred H. Wilt* Department of Molecular and Cell Biology, University of California, Berkeley, 395 Life Sciences Addition, M/C#3200, Berkeley, California 94720-3200, U. S. A. [email protected] [email protected] *Corresponding

author: Prof. Fred H. Wilt. Tel.: 510-642-2807

ABSTRACT The sea urchin larval embryo elaborates two calcitic endoskeletal elements called spicules. Spicules are synthesized by the primary mesenchyme cells (PMCs) and begin to form at early gastrula stage. It is known that the calcium comprising the spicules comes from the seawater and we wish to further consider the mode of calcium transport from the extracellular seawater to the PMCs and then onto the forming spicules. We used PMC in vitro cultures, calcein, fluorescently labeled dextran, and fluorescently labeled Wheat Germ Agglutinin (WGA) to track calcium transport from the seawater into PMCs and spicules and to determine how molecules from the surface of PMCs interact with the incoming calcium. Labeling of PMC endocytic vesicles and forming spicules by both calcein and fluorescently tagged dextran indicate that calcium is taken up from the seawater by endocytosis and directly incorporated into spicules. Calcein labeling studies also indicate that calcium from the extracellular seawater begins to be incorporated into spicules within 30 minutes of uptake. In addition, we demonstrate that fluorescently labeled WGA and calcein are taken up by many of the same endocytic vesicles and are

incorporated into growing spicules. These findings suggest that PMC specific surface molecules accompany calcium ions as they enter PMCs via endocytosis and are incorporated together in the growing spicule. Using anti-spicule matrix protein antibodies, we pinpoint a subset of spicule matrix proteins that may accompany calcium ions from the surface of the PMCs until they are incorporated into spicules. Msp130 is identified as one of these spicule matrix proteins. Abbreviations: PMC, primary mesenchyme cell; SMP, spicule matrix protein; WGA, wheat germ agglutinin; FNS, filtered natural seawater; NGS, normal goat serum; GPI, glycosylphosphatidylinositol

Key words: calcium, endocytosis, calcein, msp130, endoskeleton, sea urchin

INTRODUCTION Despite intriguing and enthralling scientists for more than a century, the sea urchin larva constructing its pair of calcitic endoskeletal spicules still presents unanswered and compelling biological questions. Our studies put forth here address a couple such questions by examining endocytosis in the primary mesenchyme cells (PMCs). The PMCs are the cells that construct the sea urchin larval endoskeletal spicules. We wish to consider how calcium ions enter the complex cellular machinery of the PMCs. It has been known for some time that calcium ions in the seawater are the ultimate source of the mineral in spicules [1]. Our studies examine how PMCs acquire and process seawater calcium ions that are ultimately incorporated into spicules. Bentov et al. [2] identified the source of calcium in forming calcareous foraminifera tests by following the uptake of calcein dissolved in the surrounding seawater. Calcein is an anionic dye that has a marked preference for binding to calcium [3]. From these studies, Bentov et al. [2] recognized that endocytosis of seawater was the source of calcium for the mineralized tests of these marine invertebrates. Following from this study, three recently published papers [4-6] examine how sea urchin larvae take up calcium ions from seawater and incorporate them into growing spicules. In brief, these researchers find calcein labeled intracellular vesicles in the epithelial cells covering of the sea urchin larva as well as the internal PMCs, and they also detect them in filopodia that traverse the blastocoel. Importantly they showed clear evidence that the calcein labeled vesicles, in a number of these instances, contained mineral. Additionally, they find that treatment with verapamil, an inhibitor of L-type calcium channels, causes PMCs to

only form rudimentary spicules. Earlier studies, by our lab, found that calcein staining granules in PMCs were direct precursors of the forming spicules [7]. Another interesting recent study that examines spicule formation in sea urchin larvae was published by Mozingo [8]. Her studies suggest a link between molecules on the surface of PMCs and the synthesis of spicules. She observed that fluorescently labeled wheat germ agglutinin (WGA), which binds specifically to PMC surfaces, is subsequently taken up by PMCs via endocytosis. A significant portion of this vesiculated WGA is then incorporated into growing spicules. We wondered if this observation by Mozingo [8] had any bearing on the transport of calcium from the seawater to the forming spicule. Thus, we undertook the present studies to re-evaluate how PMCs take up calcium ions from the extracellular environment and incorporate them into growing spicules. We also examined if PMC surface molecules are involved in the process of taking up calcium. All of our experiments use in vitro cultures of PMCs that differentiate into spicule making cells. This simple system enjoys robust spicule formation, and allows unfettered access to the differentiating cells. Our studies presented here provide evidence that calcium ions destined for spicule incorporation are taken up directly via endocytosis in PMCs. We find good evidence that primary mesenchyme cell surface specific molecules do accompany calcium ions into endocytic vesicles as well as during processing and incorporation into spicules. We pinpoint a subset of spicule matrix proteins that may accompany the calcium from the surface of the PMC until incorporation into spicules. Msp130 is identified as one of these spicule matrix proteins.

Materials and Methods Sea urchin embryo and in vitro primary mesenchyme cell cultures. Adult Strongylocentrotus purpuratus sea urchins were obtained from the University of California, Davis, Bodega Marine Laboratory, Bodega Bay, California. Sea urchin gametes were collected and sea urchin embryo cultures generated following the methods described by Foltz et al. [9]. All sea urchin embryos were cultured in filtered natural seawater (FNS) at 15°C. In vitro primary mesenchyme cell (PMC) cultures were obtained by isolating micromeres from 16-cell embryos and inducing their differentiation into PMCs synthesizing spicules as described by Wilt and Benson [10]. The differentiation of cultured micromeres into spicule making PMCs follows closely the timing of the differentiation of these same cells of intact embryos [10, 11]. Clean Fisherbrand glass coverslips, 18mm x 18mm, no. 1, were placed in the petri dishes just prior to the plating of the isolated micromeres. The cultured micromeres adhered to and differentiated on the coverslips. These glass coverslips with differentiated PMCs synthesizing spicules on them were then subsequently transferred to petri dishes for the various treatments with calcein, WGA-rhodamine, and/or dextran-rhodamine described below.

Calcein, WGA-rhodamine, and dextran-rhodamine labeling. PMCs growing attached to glass coverslips were labeled with fluorescent agents by incubating the coverslips in polystyrene petri dishes (35-100mm) containing FNS with dissolved 100-150µg/ml calcein (Sigma Chemical Company), and/or 0.5-25µg/ml WGA-rhodamine (Vector Laboratories, Inc.), and/or 50µM dextran-tetramethylrhodamine (10,000MW; Molecular Probes). The cultured PMCs

were kept in the dark and at 15°C during the treatment with each of these agents. Seawater chases were done for various lengths of time by transferring the coverslips into petri dishes containing FNS after washing them for a few minutes in another petri dish containing FNS. Following their labeling treatment, cells were fixed in ice-cold 10% formalin in FNS containing 10mM Tris pH 8.0 for 10-15 minutes and then washed with ice-cold FNS. Some samples colabeled with calcein and dextran-rhodamine were additionally treated briefly with ice-cold 1.5% sodium hypochlorite followed by 3 gentle washes with ice-cold FNS. This gentle bleach treatment was done to expose the spicules from some of the encasing cellular material for better observation of labeled spicules. All samples were mounted on glass slides. Eight µl of Fluoromount-G (SouthernBiotech) anti-quenching reagent was spotted on the sample cells and a coverslip was placed over the sample with layers of double sticky tape as a support to prevent crushing the cells. The samples were then examined using a Zeiss LSM510 META/NLO Axioimager laser scanning confocal microscope to localize the presence of fluorescent probes. The images captured were projections of stacks of optical z-section micrographs. Calcein fluorescence was obtained with excitation with 488nm illumination and capture of BD 500nm-550nm filtered emissions. Rhodamine fluorescence was obtained with 543nm illumination and capture of BD 650nm-710nm emissions. Projections were generated using the Zeiss LSM Image Browser version 4.2.0.121. The slit lengths and photomultiplier gain were adjusted following imaging of control samples using the individual fluorescent molecules (calcein and rhodamine) to assure there was minimal crosstalk between detection channels.

Antibody staining. An anti-msp130 mouse monoclonal antibody, designated 6a9, was used for immunolocalization studies in fixed cultured primary mesenchyme cells that had been labeled with WGA-rhodamine prior to fixation. The fluorescent WGA was used to indicate endocytic vesicles. The 6a9 monoclonal antibody was a generous gift from Dr. Charles Ettensohn, Carnegie Mellon University and the generation and characterization of this antibody was described previously [12]. Cultured PMCs synthesizing spicules adhering to glass coverslips were exposed to 5µg/ml WGArhodamine for 30 minutes followed by 30 min incubation in FNS. These cells were then fixed by placing the coverslips in ice-cold 10% formalin in FNS containing 10mM Tris, pH 8.0 for 15 minutes. After fixation, the cells were washed gently 4x with FNS containing 5% normal goat serum (NGS). The PMCs labeled with WGArhodamine, fixed and then washed were then exposed to primary antibodies. The mouse monoclonal antibody cell supernatant generated against msp130, designated 6a9, was diluted 1/10 in FNS+5%NGS. The primary antibody solution was gently added to 35mm petri dishes containing coverslips with attached fixed cells and spicules. The primary antibody was left on the samples overnight at 4°C. The primary antibody solution was then removed carefully and the samples washed gently 4x with FNS+5%NGS. The samples were then treated with goat, anti-mouse Alexa488 fluorescent secondary antibody (Molecular Probes) diluted 1/1000 in FNS+5%NGS for 2h. After the samples were exposed to the secondary antibody, the samples were then washed gently 4x with FNS+5%NGS. The coverslips were then placed face up on glass slides. Three layers of double sticky tape were used to surround the coverslip. Ten µl of 35% glycerol dissolved in FNS+5%NGS was

spotted on the center of the sample and a 40mm x 22mm #1 glass coverslip was used to cover the sample and adhere to the surrounding tape. A Zeiss LSM510 META/NLO Axioimager laser scanning confocal microscope was used to localize the presence of fluorescent tags. The images captured were projections of stacks of optical z-section micrographs. Projections were generated using the Zeiss LSM Image Browser version 4.2.0.121. Alexa488 fluorescence was obtained with excitation with 488nm illumination and capture of BD 500nm550nm filtered emissions. Rhodamine fluorescence was obtained with 543nm illumination and capture of BD 650nm-710nm emissions. The slit lengths and photomultiplier gain were adjusted following imaging of control samples using the individual fluorophores (Alexa488 and rhodamine) to assure there was minimal crosstalk between detection channels. Control immunostaining of samples without primary antibody showed little or no background staining of the cells and spicules by the fluorescent secondary antibody.

RESULTS Primary mesenchyme cells take up extracellular calcein, dextran and WGA via endocytosis. We use calcein as a marker for calcium in a number of studies we present here. Calcein is a fluorescent anionic derivative of fluorescein that binds to calcium [3] and it has been utilized often to label and track calcium as it is incorporated into mineralized structures [2, 7, 13-15]. Since calcein does not cross phospholipid membranes, if cells do take it up intracellularly, it must be through vacuolization and/or vesiculation of the extracellular environment—presumably via some form of endocytosis [2, 13]. Fluorescent puncta of calcein label observed in our studies described here we assume are contained within intracellular vesicles. Primary mesenchyme cells differentiated from isolated micromeres and actively synthesizing spicules were exposed to calcein dissolved in filtered natural seawater (FNS). Within 30 minutes of being exposed to calcein, fluorescently labeled vesicles are seen within cultured PMCs. One also sees calcein beginning to be incorporated at the tip of the growing spicule (see figures 1A-D). The distribution of the calcein label after washing away the extracellular calcein and letting the PMCs continue to synthesize spicules reveals how quickly the calcein labeled calcium is utilized to construct spicules. One sees distinct regions of calcein labeling as well as terminal regions of unlabeled spicule. In figures 1E and 1F, one sees the calcein labeled regions at both ends of this linear spicule. One also sees a small region of unlabeled spicule at the very ends of the spicule. Since spicules grow from their ends [16], this unlabeled end of the spicule reflects when the calcein loaded into the PMCs was depleted and no longer available for incorporation into the spicule. Our

labeling studies indicate that most of the vesicularized calcein is incorporated into the spicule by these cultured PMCs within 4 to 6 hours following calcein labeling. It has been assumed that dissolved calcein and calcium ions incorporated into spicules are taken up by PMCs utilizing the same cellular mechanisms [5]. Is the uptake of dissolved calcein and calcium ions into PMCs via a shared general mechanism of endocytosis directly from the seawater or are they taken up by PMCs by different cellular mechanisms? Fluorescently labeled dextrans do not bind calcium and are often used as markers for endocytosis (i.e. [17-19]). To investigate the mechanism of calcium uptake by PMCs, we examined the labeling of endocytic vesicles of cultured PMCs that were exposed fluorescently tagged dextran and calcein. We also followed these molecules as they are incorporated into growing spicules. Figures 2 A-D show intracellular endocytic vesicles labeled with dextranrhodamine and calcein in PMCs treated with these molecules. Many of the intracellular vesicles are labeled with both molecules (see arrows). A number of other vesicles are labeled with just one or the other. We did not extensively quantitate what proportion of vesicles are labeled by calcein and dextranrhodamine, but we do see about two-thirds of the labeled vesicles are labeled by both calcein and dextran-rhodamine. The rest of the labeled vesicles are approximately evenly labeled by either calcein or dextran-rhodamine. These observations in S. purpuratus cultured PMCs are concordant with the observations Vidavsky et al. [5] reported for whole P. lividus larvae continuously labeled with fluorescent dextran and calcein. PMCs do appear to take up dextran and calcein via many of the same endocytic vesicles. These findings suggest that calcium ions and calcein enter PMCs via the same endocytic vesicles.

We also examined if fluorescent dextran and calcein are incorporated into spicules with the same or similar pattern, as might be expected if both molecules were taken up from the seawater and processed by the same cellular processes. Vidavsky et al. [5] reported that the fluorescently labeled dextran they used to colabel with calcein in whole sea urchin embryos was excluded from spicules. We, on the other hand, do see spicules labeling with fluorescent dextran. Figures 2 E-H show a typical fluorescent labeling pattern we see in spicules following a pulse labeling of cultured PMCs with dextran-rhodamine and calcein. The fluorescently tagged dextran incorporated into the spicule is presumable precipitated along with, or encased by, the precipitated calcium carbonate mineral. Our experiments show that the spatial labeling patterns for both these fluorescent molecules in growing spicules are nearly the same. We are not sure why we see different results from those reported by Vidavsky et al. [5]. Perhaps because we gently treated the spicules after labeling with dilute bleach to better see the labeling of the spicules, or because we labeled in vitro PMC cultures while they labeled whole embryos, may account for our differences. Vidavsky et al. (2016) also used a smaller molecular weight fluorescently tagged dextran (3 kDa molecular weight) to label the larvae and perhaps it is less likely to precipitate or be encased in the precipitated mineral. Despite these discrepancies, our results do suggest that calcein and dextranrhodamine are transported and incorporated by PMCs into spicules via the same mechanism. Following from these findings, we believe extracellular dissolved calcein and calcium ions that are ultimately incorporated into sea urchin larval spicules are brought into PMCs by the same endocytic vesicles.

Processing of vesicles. As a first step to further understand what is involved in the processing of PMC endocytic vesicles once they enter the cell, we wondered if molecules from the surface of PMCs might have a role. A few decades ago Spiegel and Burger [20] as well as Desimone and Spiegel [21] reported that the lectin, wheat germ agglutinin (WGA), binds preferentially to PMC cell surfaces in sea urchin larvae. Since these observations were published, WGA binding has been used as a marker for PMC differentiation, as well for mass isolation of PMCs for in vitro culture [22, 23]. More recently, Mozingo [8] observed that PMCs labeled with fluorescently tagged WGA transport the surface-bound WGA into the PMCs via endocytosis. The PMCs then transport many of these WGA-labeled vesicles to the growing spicule and WGA is incorporated into the spicule. These observations raise the interesting question of whether the same cellular transport system of WGA observed by Mozingo [8] is also transporting calcein into the PMC, and then subsequently into the forming spicule. We examined treatment of cultured primary mesenchyme cells with fluorescently tagged WGA and calcein, and followed the localization of these molecules within the PMCs as the PMCs synthesized the spicules. WGA-rhodamine brightly labels the surface of cultured PMCs synthesizing spicules shortly after the treatment begins (figure 3). One sees that a 30 min treatment with WGA-rhodamine completely labels the surface of the PMCs with the cell body, syncytial sheath around the spicule as well as the filopodia all labeled (figures 3A and B), including a few endocytic vesicles labeled at this stage. However, most of the WGA-rhodamine label is on the surface of the PMCs. By 4 hr after a 30 min pulse labeling with WGA-rhodamine, most of the

label from the surface has been taken up by endocytosis. A number of large vesicles within the PMCs and the ends of the spicules are labeled (figures 3C and D). These observations are similar to what Mozingo [8] observed when she labeled PMCs with fluorescent WGA by transiently permeabilizing whole sea urchin embryos. We also observe that some of the fluorescent WGA tag is not transported to and incorporated into the growing spicule and remains in large intracellular vesicles. These labeled intracellular vesicles remained for as long as we examined the PMC cultures after initial labeling (>24hr). Mozingo [8] observed a similar phenomenon in her studies. Cultured PMCs treated with WGA-rhodamine and calcein in the extracellular seawater resulted in many endocytic vesicles being colabeled by both fluorescent molecules (see figures 4A-H). We see this colabeling of vesicles in the cell bodies of PMCs, the syncytial sheath that wraps around the spicule as well as within the filopodia extending from the PMCs (see arrows in figures 4A-H). Most of the calcein labeled vesicles are also labeled with WGA-rhodamine during these sorts of experiments. The fraction of calcein labeled vesicles also being labeled with WGArhodamine during similar experiments ranged from a little more than half to threequarters. These observations suggest that the same, or many of the same vesicles, transport calcein and WGA-binding molecules from the surface of the PMCs. These vesicles presumably then transport the calcein and WGA-rhodamine out of the PMCs and deposit them at the growing endoskeletal tip. Close examination of the labeling pattern of calcein and WGA-rhodamine incorporated into spicules reveals that there are some differences in the processing dynamics of these two molecules. Figure 5 illustrates the differences we observe

(figures 5A-D). In the experiment presented in Figure 5, we pulse labeled cultured PMCs with calcein and WGA-rhodamine for 2hr. The label was then washed out and the labeled cells were then cultured for another 4hr in FNS before we observed the labeling patterns in the growing spicules. From this experiment, one sees distinct band of fluorescent calcein accumulation near the end of the spicule. However, the very end of the spicule is unlabeled with calcein. This lack of calcein label at the tip of the spicule indicates the point during spicule synthesis the PMCs were depleted of intracellular calcein labeled calcium. The spicule-labeling pattern for WGArhodamine, while similar to calcein, it is a bit different. The WGA-rhodamine label is present everywhere the calcein label is present, but it also extends to the tip of the spicule. Presumably the WGA-rhodamine available for deposit in spicules was not depleted by 4 hr after the wash out of the label. We believe this difference in labeling patterns may reflect the different ways these two molecules label the PMCs. When PMCs are treated with calcein, the calcein enters by endocytosis directly from the seawater. After the extracellular calcein is washed away, little or no additional calcein is taken up by PMCs. On the other hand, when PMCs are treated with extracellular WGA-rhodamine, the WGA-rhodamine binds to PMC specific sugar moieties and/or sugar moieties attached to PMC specific cell surface molecules. After extracellular WGA-rhodamine is washed out, even though the WGA-rhodamine dissolved in the seawater is removed and not available for endocytosis, fluorescently labeled WGA is nevertheless still present on the cell surface and acts as an additional source of WGA available to be taken up by endocytosis and eventually incorporated into spicules. We believe this explains why one does not see the sharp termination of fluorescent staining after a wash out of the extracellular

WGA-rhodamine in Figure 5 that one sees after a wash out of the extracellular calcein. Despite calcein and WGA treatments causing modestly different labeling patterns of growing spicules, our studies nevertheless indicate that calcein and WGA are taken up and transported in PMCs by the same or many of the same endocytic vesicles. Cell surface molecules and calcein are transported together and eventually incorporated into growing spicules via the same vesicles. These studies imply that PMC specific surface molecules are taken up and are present in the same endocytic vesicles as the calcium ions that are incorporated in growing spicules.

Spicule matrix proteins first come into close contact with calcium during endocytosis by primary mesenchyme cells. Since our experiments suggest that calcium and PMC specific surface molecules are transported together from the surface to the growing endoskeletal spicule, we considered the identity of some of these PMC specific cell surface molecules. Proteomic analysis of the proteins comprising the integral spicule matrix reveals that a number of prominent spicule matrix proteins (SMPs) are PMC surface proteins [24]. These prominent cell surface proteins include msp130, the second most prominent SMP, and the 5 members of the msp130 related family of proteins. Other prominent putative PMC surface SMPs are a PMC specific carbonic anhydrase (SPU_012518) [25], p58-a and p58-b [26], and P16 [27]. Previously published experimental evidence for each of these SMPs indicates that they all play a role in sea urchin spicule biomineralization [26-33].

To examine whether spicule matrix proteins are accompanying calcium ions in endocytic vesicles during PMC spicule formation, we utilized immunological tools to localize SMPs in spicule forming cells. We initially used a rabbit polyclonal antiserum known to interact with all of the SMPs [34, 35] and tried to determine if it bound endocytic vesicles that are ultimately transported to the forming spicule. For these experiments, we were unable to successfully use calcein or dextranrhodamine labeling to identify endocytic vesicles while simultaneous using antibody immunostaining procedures. The calcein and dextran-rhodamine label would not persist through fixation and washing steps of the immunofluorescent labeling procedures we employed. However, WGA-rhodamine staining did persist nicely through the anti-spicule matrix protein immunostaining procedure and was a convenient marker for identifying endocytic vesicles during these experiments. The preliminary results with this antiserum showed that there is significant staining for SMPs on the surface of PMCs as well as in vesicles inside the PMCs. There were also many vesicles that are labeled by both WGA-rhodamine and the total-SMP antiserum (see figures S1 A-C in supplementary material). These interesting findings lead us to wonder if we could identify any of the individual PMC surface SMPs we previously identified as possible candidates for being present in endocytic vesicles. Msp130 was one of these candidate SMPs and we had access to a mouse monoclonal antibody that binds msp130. We therefore assayed for the presence of msp130 in PMC endocytic vesicles by immunostaining WGA-rhodamine labeled PMCs with a mouse monoclonal antibody, designated 6a9 [12]. This anti-msp130 antibody extensively labeled the PMC cell surface and many intracellular endocytic vesicles that were also labeled with WGA-rhodamine (see

figures 6A-E). Many of these labeled endocytic vesicles tended to be labeled by 6a9 at the periphery of the vesicles while WGA-rhodamine tended to label the middle of the vesicle (see figure 6 D and E). These colabeled vesicles are located in the cell bodies, the syncytial sheath surrounding the spicule, as well as in filopodia extending out from the cell bodies (see arrows in figure 6). This staining pattern is very similar to the pattern observed in the PMCs stained with anti-total SMP antiserum and WGA-rhodamine (figures S1 A-C). The monoclonal antibody 6a9 does bind msp130 and stains PMC cell surfaces specifically [12]. However, it has not been directly examined for cross-reaction with other members of the msp130-like family of proteins (Charles Ettensohn, personal communication). We, therefore, cannot exclude the possibility that some of the immunostaining signal we observe with 6a9 might be from binding other members of the msp130 family of proteins. However, since msp130 is much greater in prevalence than the other members of the msp130related family of proteins [24] and the staining pattern of 6a9 is so similar to the anti-SMP antiserum pattern, we believe it is most likely the vast majority of the 6a9 staining signal is provided by binding to msp130. Our studies indicate that SMPs and WGA are taken up from the surface in the same or many of the same endocytic vesicles. We have also shown that the same, or many of the same, endocytic vesicles in PMCs transport WGA and calcein together from the cell surface to the growing spicule. These findings together suggest that calcium ions and certain SMPs first interact at the surface of the PMCs as they are localized together and transported into PMCs via endocytic vesicles. We identify msp130 as a likely SMP to be transported with calcium ions in these endocytic vesicles.

DISCUSSION Vidavsky et al. [4-6] have published three recent papers examining calcium uptake in developing sea urchin larvae using calcein labeling as a marker for calcium. After continuously labeling whole embryos with calcein, they observe calcein labeled endocytic vesicles throughout the larva. They provide evidence that calcein labeled vesicles in the PMCs and in the PMCs’ filopodia contain calcium mineral [4, 6]. We show that many of these calcein labeled endocytic vesicles observed by Vidavsky et al. [4, 6] also contain PMC specific cell surface molecules. Our calcein labeling studies suggest that calcium ions taken up by endocytosis begins to be incorporated into spicules within 30 minutes of entering the PMCs. Our studies also show that calcein and fluorescent dextran colabel many of the same endocytic vesicles in PMCs indicating that calcein, and presumably calcium ions, is taken up from the extracellular seawater by endocytosis. This observation is consistent with the findings and conclusions of Vidavsky et al. [5]. Vidavsky et al. [5] also impressively revealed, by utilizing cryogenic focused ion beam scanning electron microscopy 3D imaging, the extensive network of connected endocytic vesicles and vacuoles at the surface of PMCs. Unlike Vidavsky et al. [5], however, we observe that a pulse treatment of PMCs with calcein and dextranrhodamine results in both fluorescent molecules being incorporated into spicules with essentially the same spatial pattern. Vidavsky et al. [5] do not see labeling of spicules by fluorescently tagged dextran taken up by PMCs. We discussed earlier in the results section our possible explanation for the discrepancy in our observations. We do note that Bentov et al. [2] does see fluorescent dextran endocytosis and incorporation into Foraminifera mineralized tests. Our results lead us to conclude

that calcein and dextran-rhodamine are taken up and processed by the same, or very similar, endocytic and secretion mechanisms. These findings indicate that calcium ions enter PMCs by endocytosis from the extracellular seawater and are then transported and incorporated into spicules. Studies by Vidavsky et al. [6] indicate that the calcium ions taken up by endocytosis are converted into amorphous calcium carbonate shortly after they are taken up by PMCs or PMC filopodia. Vidavsky et al [4, 6] also observed calcein labeled endocytic vesicles in other sea urchin larval tissues, particularly in the larval outer ectodermal epithelial cell bodies and in the epithelial filopodia. We also see this pattern of calcein labeling in whole S. purpuratus larvae (data not shown). However, Vidavsky et al. [4, 6] suggest that these epithelial cells may provide some mineral to PMCs via filopodia to help construct spicules. Our studies presented here, as well as a number of other previously published studies (e.g. [10, 11, 36]), demonstrate that PMCs are capable of making spicules robustly without the presence of the rest of the tissues of the sea urchin larva. We believe the contribution of mineral to spicule construction from the sea urchin larval ectodermal epithelium is probably not significant. Mozingo [8] first observed that WGA, which specifically binds to the surface of PMCs, is taken up by endocytosis and a significant amount of the vesicularized WGA is incorporated into growing spicules. Our studies presented here show that many of the same endocytic vesicles that take up calcein also transport WGA into the PMCs and into the growing spicule. From these findings, we conclude that PMCspecific surface molecules accompany calcium ions as they enter the same endocytic

vesicles, or many of the same vesicles, and are also incorporated into growing spicules. Proteomic analysis of sea urchin larval integral SMPs reveals that a number of the prevalent SMPs are cell surface proteins [24]. This led us to examine if PMC endocytic vesicles contain any spicule matrix proteins. We do in fact see anti-spicule matrix antibody staining of endocytic vesicles that also label with fluorescent WGA. We also use a monoclonal antibody to identify msp130 as at least one of the SMPs that may first interact with calcium ions as they enter the PMCs. Quite a bit of experimental information is already known about msp130. Msp130 was one of the first sea urchin PMC specific proteins cloned and identified [37, 38]. Msp130 mRNA expression pattern in PMCs throughout larval spicule growth is also well characterized [39]. Msp130 is a cell surface protein that attaches to the surface of PMCs by GPI linkages [40] and proteomic analyses find msp130 is the second most prevalent protein present within spicules, only behind SM50, in the S. purpuratus larval spicules [24]. The Lennarz research group, more than three decades ago, showed that a monoclonal antibody that binds msp130 also arrests spicule formation and 45Ca2+ uptake into cultures of PMCs without adversely affecting cell viability [29]. Farach-Carson et al. [41] also observed that msp130 binds calcium and hypothesized that msp130 acts to bind calcium divalent ions at the surface of PMCs so that it may be internalized for eventual deposition in the growing spicule. Our results, together with these previous characterizations of msp130, certainly suggest that may very well be the case. Cameron and Bishop [28] have also reported that msp130, a few of the msp130 related proteins, and a carbonic anhydrase similar to the sea urchin PMC specific carbonic anhydrase are

highly conserved in the hemichordate, S. kowalevskii, and are presumably involved in biomineralization of S. kowalevskii adult mineralized ossicles. Hemichordates and echinoderms diverged from each other more than 500 million years ago [42], so the conservation of these proteins’ amino acid sequences in S. purpuratus and S. kowalevskii certainly suggests that these proteins have important conserved functions in these two organisms. Our studies presented here clarify our thinking about how SMPs interact with the mineral phase of growing spicules. Our studies suggest that a subset of spicule matrix proteins expressed on the surface of PMCs, such as msp130, first interact with calcium ions at the surface. They then accompany the calcium ions and subsequently precipitated calcium carbonate mineral as they pass through the PMC to the site of the growing spicule. Other spicule matrix proteins, such as SM50 and the SM30 family of proteins, are vectorially secreted in separate vesicles [43, 44] and are embedded in and around the growing mineralized spicule. Examination of the contents of PMC endocytic vesicles, as we have presented here, should enable one to identify additional proteins involved in the uptake of calcium ions from the extracellular environment and eventual incorporation into spicules. A number of other identified SMPs, besides msp130, are thought to be PMC cell surface molecules. These include five members of the msp130 related family of proteins [25, 45], a PMC specific carbonic anhydrase (SPU_012518) [25], p58-a and p58-b [26], and P16 [27]. These SMPs are all good candidates for initial studies examining if they are included in the same endocytic vesicles as the incoming calcium ions and msp130. For each of these SMPs, there is either direct experimental evidence that they participate in biomineralization and/or there is phylogenetic evidence that

these proteins are conserved in distantly related biomineralizing organisms [26-28, 31, 33, 45, 46]. These proteins may well be involved in the processing and delivery of mineral to growing spicules. It will be particularly interesting to see if carbonic anhydrase, SPU_012518 [25], is localized in these endocytic vesicles given its amino acid sequence conservation in S. kowalevskii. First it would help explain how calcium and carbonate ions come together in the same endocytic vesicles shortly after the vesicles enter PMCs or PMC filopodia and, second, it would strengthen the notion that this mechanism of mineral precipitation is shared among echinoderms and hemichordates. Future experiments should be able to determine this readily. An important but as yet unanswered question is what proportion of the calcium incorporated into spicules is directly provided by PMC endocytosis. From our studies and the studies of Vidavsky et al. [4-6], it can be concluded that PMC endocytosis provides a significant amount of calcium to growing sea urchin larval spicules. Msp130 appears to be involved in calcium ion endocytosis and msp130 gene expression is regulated by a gene regulatory network believed to regulate sea urchin skeletogenesis [32, 33, 47]. These findings strengthen our supposition that calcium uptake via PMC endocytosis is an important source of calcium ions for larval spicule formation. However, we remain unsure what amount of spicule calcium might come from cellular mechanisms other than endocytosis. Vidavsky et al. [5] looked at this problem and observed that sea urchin larvae treated with verapamil, a drug known to block L-type calcium channels, arrests spicule formation at the early tri-radiate spicule stage. They suggested that some unknown amount of calcium ions transported by L-type is incorporated into forming spicules. Based on the verapamil studies published to date [5, 48-50], we

concur that these studies indicate calcium channel activity is needed for spicule formation by PMCs. However, we do not believe that it can be concluded that L-type calcium channels are contributing any calcium ions directly for spicule formation. Blocking extracellular calcium ions from entering the cytoplasm of PMCs would be expected to have effects on a number of intracellular mechanisms and not limited to providing calcium ions for spicules synthesis. It cannot be excluded that disrupted cellular processes are the root cause of the cessation of spicule synthesis. We note that the treatment of sea urchin larvae with batimastat (a.k.a. BB-94), an inhibitor of matrix metalloproteinases, arrests larval spicule formation at the very same stage as the verapamil treated larvae. It has been shown that this arrest of spiculogenesis is caused by a disruption of secretion in PMCs [43, 51]. Hwang and Lennarz [49] also examined the effects of calcium channel antagonists, including verapamil, and other agents that disrupt membrane traffic on spicule formation by in vitro cultured sea urchin PMCs. They concluded that calcium channel antagonist arrest of spiculogenesis might well be caused by a general blockage of PMC secretion mechanisms. Given all these previously findings, we do not believe L-type calcium channels have been demonstrated experimentally to provide calcium ions directly to forming spicules. It remains an open question whether calcium in larval spicules comes from cellular mechanisms other than endocytosis.

CONCLUSIONS Our examination of endocytosis in the sea urchin larva’s PMCs leads us to conclude that PMC take up calcium ions from seawater by endocytosis and begin to be incorporated into forming spicule within 30 minutes of entering the cell. PMC specific surface molecules escort calcium ions from the cell surface via the same endocytic vesicles and eventually into the growing endoskeleton. We identify msp130 as one of the PMC surface molecules that may accompany calcium ions as they enter PMCs. Our observations provide a richer understanding of PMC endocytosis during spiculogenesis and supply an experimental procedure to identify other PMC surface molecules that may accompany calcium ions as they enter PMCs.

ACKNOWLEDGEMENTS Experiments presented here were supported by grants to FHW from the U. C. Berkeley Emeritus Research Enabling grant, and from NSF Award 0444724. We thank Dr. Charles Ettensohn, Carnegie Mellon University, for generously providing us a sample of the 6a9 monoclonal antibody supernatant. We thank Dr. Stephen Weiner (Weizmann Institute of Science), Dr. Lia Addadi (Weizmann Institute of Science) and Dr. Stephen Benson (California State University, East Bay) for their helpful criticisms of our manuscript. This work was performed in part at the CRL Molecular Imaging Center at U. C. Berkeley, supported by NIH 3R01EY015514-01S1. We also thank Holly Aaron and Jen-Yi Lee from the CRL Molecular Imaging Center for their invaluable assistance.

REFERENCES [1]

E. Nakano, O. K, T. Iwamatsu, Accumulation of radioactive calcium in larvae of the sea urchin Pseudocentrotus depressus, Biol. Bull. 125 (1963) 125-132.

[2]

S. Bentov, C. Brownlee, J. Erez, The role of seawater endocytosis in the biomineralization process in calcareous foraminifera, Proc. Natl. Acad. Sci. U. S. A. 106 (2009) 21500-21504.

[3]

D.F.H. Wallach, D.M. Surgenor, J. Soderberg, Preparation and properties of 3, 6-dihydroxy-2, 4-bis [N, N'-di-(carboxymethyl)-aminomethyl] fluoran. Utilization for the ultramicrodetermination of calcium, Anal. Chem. 31 (1959) 456-460.

[4]

N. Vidavsky, S. Addadi, J. Mahamid, E. Shimoni, D. Ben-Ezra, M. Shpigel, S. Weiner, L. Addadi, Initial stages of calcium uptake and mineral deposition in sea urchin embryos, Proc. Natl. Acad. Sci. U. S. A., 2014, Vol. 111, pp. 39–44.

[5]

N. Vidavsky, S. Addadi, A. Schertel, D. Ben-Ezra, M. Shpigel, L. Addadi, S. Weiner, Calcium transport into the cells of the sea urchin larva in relation to spicule formation, Proc. Natl. Acad. Sci. U. S. A. 113 (2016) 12637-12642.

[6]

N. Vidavsky, A. Masic, A. Schertel, S. Weiner, L. Addadi, Mineral-bearing vesicle transport in sea urchin embryos, J. Struct. Biol. 192 (2015) 358-365.

[7]

F.H. Wilt, C.E. Killian, P. Hamilton, L. Croker, The dynamics of secretion during sea urchin embryonic skeleton formation, Exp. Cell Res. 314 (2008) 1744-1752.

[8]

N.M. Mozingo, Lectin uptake and incorporation into the calcitic spicule of sea urchin embryos, Zygote (2014) 1-7.

[9]

K.R. Foltz, N.L. Adams, L.L. Runft, Echinoderm eggs and embryos: procurement and culture, Meth. Cell Biol. 74 (2004) 39-74.

[10]

F.H. Wilt, S.C. Benson, Isolation and culture of micromeres and primary mesenchyme cells, Meth. Cell Biol. 74 (2004) 273-285.

[11]

K. Okazaki, Spicule formation by isolated micromeres of the sea urchin embryo, Am. Zool. 15 (1975) 567-581.

[12]

C.A. Ettensohn, D.R. McClay, Cell lineage conversion in the sea-urchin embryo, Dev. Biol. 125 (1988) 396-409.

[13]

J.M. Bernhard, J.K. Blanks, C.J. Hintz, G.T. Chandler, Use of the fluorescent calcite marker calcein to label foraminiferal tests, J. Foraminiferal Res. 34 (2004) 96-101.

[14]

S.J. Du, V. Frenkel, G. Kindschi, Y. Zohar, Visualizing normal and defective bone development in zebrafish embryos using the fluorescent chromophore calcein, Dev. Biol. 238 (2001) 239-246.

[15]

K.A. Guss, C.A. Ettensohn, Skeletal morphogenesis in the sea urchin embryo: regulation of primary mesenchyme gene expression and skeletal rod growth by ectoderm-derived cues, Development 124 (1997) 1899-1908.

[16]

G.L. Decker, W.J. Lennarz, Growth of linear spicules in cultured primary mesenchymal cells of sea urchin embryos is bidirectional, Dev. Biol. 126 (1988) 433-436.

[17]

S. Elmquist, R. Libelius, G. Lawoko, S. Tågerud, Dextrans as markers for endocytosis in innervated and denervated skeletal muscle, Muscle Nerve 15 (1992) 876-884.

[18]

J.M. Oliver, R.D. Berlin, B.H. Davis, Use of horseradish peroxidase and fluorescent dextrans to study fluid pinocytosis in leukocytes, Meth. Enzymol. 108 (1984) 336-347.

[19]

J. Swanson, Fluorescent labeling of endocytic compartments, Meth. Cell Biol. 29 (1989.) 137-151.

[20]

M. Spiegel, M.M. Burger, Cell adhesion during gastrulation: A new approach, Exp. Cell Res. 139 (1982) 377-382.

[21]

D.W. DeSimone, M. Spiegel, Wheat germ agglutinin binding to the micromeres and primary mesenchyme cells of sea urchin embryos, Dev. Biol. 114 (1986) 336-346.

[22]

C.A. Ettensohn, D.R. McClay, A new method for isolating primary mesenchyme cells of the sea urchin embryo: Panning on wheat germ agglutinin-coated dishes, Exp. Cell Res. 168 (1987) 431-438.

[23]

M. Kiyomoto, F. Zito, S. Sciarrino, V. Matranga, Commitment and response to inductive signals of primary mesenchyme cells of the sea urchin embryo, Dev., Growth Differ. 46 (2004) 107-114.

[24]

K. Mann, F.H. Wilt, A.J. Poustka, Proteomic analysis of sea urchin(Strongylocentrotus purpuratus) spicule matrix, Proteome Sci. 8 (2010) 33.

[25]

B.T. Livingston, C.E. Killian, F.H. Wilt, A. Cameron, M.J. Landrum, O. Ermolaeva, V. Sapojnikov, D.R. Maglott, A.M. Buchanan, C.A. Ettensohn, A genome-wide analysis of biomineralization-related proteins in the sea urchin Strongylocentrotus purpuratus, Dev. Biol. 300 (2006) 335-348.

[26]

A. Adomako-Ankomah, C.A. Ettensohn, P58-A and P58-B: novel proteins that mediate skeletogenesis in the sea urchin embryo, Dev. Biol. 353 (2011) 8193.

[27]

M.S. Cheers, C.A. Ettensohn, P16 is an essential regulator of skeletogenesis in the sea urchin embryo, Dev. Biol. 283 (2005) 384-396.

[28]

C.B. Cameron, C.D. Bishop, Biomineral ultrastructure, elemental constitution and genomic analysis of biomineralization-related proteins in hemichordates., Proc. R. Soc. B 279 (2012) 3041-3048.

[29]

D. Carson, M.C. Farach, D.S. Earles, G.L. Decker, W.J. Lennarz, A monoclonal antibody inhibits calcium accumulation and skeleton formation in cultured embryonic cells of the sea urchin, Cell 41 (1985) 639-648.

[30]

K. Mitsunaga, K. Akasaka, H. Shimada, Y. Fujino, Carbonic anhydrase activity in developing sea urchin embryos with special reference to calcification of spicules, Cell Differ. 18 (1986) 257-262.

[31]

F. Zito, D. Koop, M. Byrne, V. Matranga, Carbonic anhydrase inhibition blocks skeletogenesis and echinochrome production in Paracentrotus lividus and Heliocidaris tuberculata embryos and larvae, Develop. Growth Differ. 57 (2015) 507-514.

[32]

K. Rafiq, M.S. Cheers, C.A. Ettensohn, The genomic regulatory control of skeletal morphogenesis in the sea urchin, Development 139 (2012) 579–590.

[33]

K. Rafiq, T. Shashikant, C.J. McManus, C.A. Ettensohn, Genome-wide analysis of the skeletogenic gene regulatory network of sea urchins, Development 141 (2014) 2542-2542.

[34]

S. Benson, N. Benson, F. Wilt, The organic matrix of the skeletal spicule of sea urchin embryos, J. Cell Biol. 102 (1986) 1878-1886.

[35]

C.E. Killian, F.H. Wilt, Characterization of the proteins comprising the integral matrix of Strongylocentrotus purpuratus embryonic spicules, J. Biol. Chem. 271 (1996) 9150-9159.

[36]

R.T. Knapp, C.-H. Wu, K.C. Mobilia, D. Joester, Recombinant sea urchin vascular endothelial growth factor directs single-crystal growth and branching in vitro, J. Am. Chem. Soc. 134 (2012) 17908-17911.

[37]

J.A. Angstrom, J.E. Chin, D.S. Leaf, A.L. Parks, R.A. Raff, Localization and expression of msp130, a primary mesenchyme lineage-specific cell-surface protein of the sea-urchin embryo, Development 101 (1987) 255-265.

[38]

D.S. Leaf, J.A. Anstrom, J.E. Chin, M.A. Harkey, R.M. Showman, R.A. Raff, Antibodies to a fusion protein identify a cDNA clone encoding msp130, a primary mesenchyme-specific cell surface protein of the sea urchin embryo, Dev. Biol. 121 (1987) 29-40.

[39]

Z. Sun, C.A. Ettensohn, Signal-dependent reulation of the sea urchin skeletogenic gene regulatory network, Gene Expr. Patterns 16 (2014) 93103.

[40]

B.A. Parr, A.L. Parks, R.A. Raff, Promoter structure and protein sequence of msp130, a lipid-anchored sea urchin glycoprotein, J. Biol. Chem. 265 (1990) 1408-1413.

[41]

M.C. Farach-Carson, D.D. Carson, J.L. Collier, W.J. Lennarz, H.R. Park, G.C. Wright, A calcium-binding, asparagine-linked oligosaccharide is involved in

skeleton formation in the sea urchin embryo, J. Cell Biol. 109 (1989) 12891299. [42]

O. Simakov, T. Kawashima, F. Marlétaz, J. Jenkins, R. Koyanagi, T. Mitros, K. Hisata, J. Bredeson, E. Shoguchi, F. Gyoja, J.-X. Yue, Y.-C. Chen, R.M. Freeman, A. Sasaki, T. Hikosaka-Katayama, A. Sato, M. Fujie, K.W. Baughman, J. Levine, P. Gonzalez, C. Cameron, J.H. Fritzenwanker, A.M. Pani, H. Goto, M. Kanda, N. Arakaki, S. Yamasaki, J. Qu, A. Cree, Y. Ding, H.H. Dinh, S. Dugan, M. Holder, S.N. Jhangiani, C.L. Kovar, S.L. Lee, L.R. Lewis, D. Morton, L.V. Nazareth, G. Okwuonu, J. Santibanez, R. Chen, S. Richards, D.M. Muzny, A. Gillis, L. Peshkin, M. Wu, T. Humphreys, Y.-H. Su, N.H. Putnam, J. Schmutz, A. Fujiyama, J.-K. Yu, K. Tagawa, K.C. Worley, R.A. Gibbs, M.W. Kirschner, C.J. Lowe, N. Satoh, D.S. Rokhsar, J. Gerhart, Hemichordate genomes and deuterostome origins, Nature 527 (2015) 459-465.

[43]

E.P. Ingersoll, K.L. McDonald, F.H. Wilt, Ultrastructural localization of spicule matrix proteins in normal and metalloproteinase inhibitor-treated sea urchin primary mesenchyme cells, J. Exp. Zool., Part A 300A (2003) 101-112.

[44]

C.E. Killian, F.H. Wilt, Molecular aspects of biomineralization of the echinoderm endoskeleton, Chem. Rev. 108 (2008) 4463-4474.

[45]

C.A. Ettensohn, Horizontal transfer of the msp130 gene supported the evolution of metazoan biomineralization, Evol. Dev. 16 (2014) 139-148.

[46]

N. Le Roy, D.J. Jackson, B. Marie, P. Ramos-Silva, F.d.r. Marin, The evolution of metazoan a-carbonic anhydrases and their roles in calcium carbonate biomineralization, Front. Zool. 11 (2014) 75.

[47]

V.F. Hinman, A.M. Cheatle Jarvela, Developmental gene regulatory network evolution: Insights from comparative studies in echinoderms, Genesis 52 (2014) 193–207.

[48]

Y. Fujino, K. Mitsunaga, A. Fujiwara, I. Yasumasu, Inhibition of 45Ca2+ uptake in the eggs and embryos of the sea urchin, Anthocidaris crassispina, by several calcium antagonists, anion transport inhibitor, and chloride transport inhibitors, J. Exp. Zool. 235 (1985) 281-288.

[49]

S.P. Hwang, W.J. Lennarz, Studies on the cellular pathway involved in assembly of the embryonic sea urchin spicule, Exp. Cell Res. 205 (1993) 383387.

[50]

I. Yasumasu, K. Mitsunaga, Y. Fujino, Mechanism for electrosilent Ca 2+ transport to cause calcification of spicules in sea urchin embryos, Exp. Cell Res. 159 (1985) 80-90.

[51]

E.P. Ingersoll, F.H. Wilt, Matrix metalloproteinase inhibitors disrupt spicule formation by primary mesenchyme cells in the sea urchin embryo, Dev. Biol. 196 (1998) 95-106.

Figure 1 – Calcein uptake and spicule labeling kinetics. [A, B, C, D] Cultured PMCs exposed to 150µg/ml calcein in filtered natural seawater (FNS) for 30 min and then fixed (bars = 5µm). Arrows point to calcein labeling at the tip of the growing spicule demonstrating that it takes less than 30 minutes for calcein to be taken up by endocytosis and incorporated into a growing spicule. [E, F] Cultured PMCs were exposed to 150µg/ml calcein in seawater for 30 minutes and then subsequently washed and cultured in FNS for 6 hrs before fixation (bars = 20µm). Boxed in area shows a close up of one end of the labeled spicule. bf=bright field.

Figure 2 – [I] [A, B, C, D] Calcein and dextran-rhodamine localization in PMCs. Cultured PMCs were exposed to 150µg/ calcein and 50µM dextran-rhodamine (MW 10,000) in seawater for 2hr and then fixed (bars = 5µm). Calcein (green) and dextran-rhodamine (red) localization in PMCs were then examined by fluorescence confocal microscopy. The arrows indicate endocytic vesicles containing both calcein and dextran-rhodamine. [II] [E, F, G, H] Calcein and dextran-rhodamine labeling of spicules. Cultured PMCs exposed to 150µg/ calcein and 50µM dextran-rhodamine (MW 10,000) in FNS for 2hr and then washed and cultured for another 6hr in FNS. The PMCs and spicules were then briefly exposed to dilute bleach to remove some cellular material and expose the spicules for clearer observation. The samples were then examined by fluorescence confocal microscopy to reveal the localization of calcein (green) and dextran-rhodamine (red) within the growing spicule (bars = 10µm). bf=bright field.

Figure 3 – WGA-rhodamine labeling of cultured PMCs. [A/B] Cultured PMCs were exposed to 5µg/ml WGA-rhodamine in FNS for 30 min and then fixed (bars = 5µm). [C/D] Cultured PMCs were exposed to 5µg/ml WGA-rhodamine in FNS for 30 min, and subsequently washed and cultured in FNS for an additional 4 hr before fixation and examination by fluorescence confocal microscopy (bars = 5µm). bf=bright field.

Figure 4 – Colocalization of calcein and WGA-rhodamine in PMC cell bodies and filopodia. [A-D] Cultured PMCs synthesizing spicules exposed to 5µg/ml WGArhodamine plus 100µg/ml calcein in FNS for 30 minutes followed by 2hr of 100µg/ml calcein in FNS followed by fixation. Calcein (green) and WGA-rhodamine (red) localization were then examined by fluorescence confocal microscopy. (bars = 5µm). White arrows indicate some of the vesicles labeled with calcein and WGArhodamine found in the PMC cell bodies and blue arrows indicate some of the vesicles labeled with calcein and WGA-rhodamine found in filopodia. [E-H] Cultured PMCs were exposed to 2µg/ml WGA-rhodamine plus 100µg/ml calcein in seawater for 30 minutes followed by 2hr of 100µg/ml calcein in seawater followed by fixation and examination by fluorescence confocal microscopy (bars = 5µm). White arrows indicate some of the vesicles labeled with calcein and WGA-rhodamine found in the PMC cell bodies and blue arrows indicate some of the vesicles labeled with calcein and WGA-rhodamine found in filopodia. bf=bright field.

Figure 5 – Calcein and WGA-rhodamine labeling of spicules. [A-D] Cultured PMCs were exposed to 5µg/ml WGA-rhodamine plus 150µg/ml calcein in seawater for 2hr followed by 4hr in seawater. The PMCs were then fixed. Calcein (green) and WGArhodamine (red) localization in spicules were then examined by fluorescence confocal microscopy. (bars = 5µm). bf=bright field.

Figure 6- Localization of msp130 and WGA-rhodamine in cultured PMCs. [A-E] Cultured PMCs synthesizing spicules were labeled with WGA-rhodamine and then fixed. Msp130 protein was localized in these cells by immunofluorescent staining using the anti-msp130 monoclonal antibody 6a9. Intracellular vesicles were identified as endocytic vesicles by WGA-rhodamine fluorescent labeling. Boxed in areas of these images show a close up of vesicles labeled by 6a9 and WGArhodamine. Blue arrows indicate some of the vesicles located within filopodia that react with the 6a9 antibody (green) and labeled with WGA-rhodamine (red). White arrows indicate some of the vesicles located within PMC cell bodies or the syncytial envelope surrounding the spicule that react with the 6a9 antibody (green) and label with WGA-rhodamine (red). [D, E] Boxed in area of these images show a close up of vesicles labeled by 6a9 (green) and WGA-rhodamine (red). Bars = 5µm.

HIGHLIGHTS

•

Primary mesenchyme cell (PMC) labeling experiments using calcein indicate calcium ions destine for incorporation into sea urchin larval endoskeletal spicules are taken up by larval PMCs via endocytosis. Within 30 minutes of uptake by PMCs, calcium ions begin to be incorporated into growing spicules.

•

PMC labeling experiments using calcein and fluorescently tagged wheat germ agglutinin suggest PMC specific surface molecules accompany calcium ions as they enter PMC endocytic vesicles and as they are incorporated into growing spicules.



Using anti-spicule matrix protein antibodies, we pinpoint a subset of spicule matrix proteins that may accompany calcium ions from the surface of PMCs until incorporation into spicules. We identify msp130, a prominent spicule matrix protein, as one of the PMC specific surface molecules that may accompany calcium ions from the PMC surface to incorporation into the growing spicule.