INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Skeletal myogenesis is an excellent model system for the study of cell lineage determination, cell differentiation, and tissue-specific gene expression. These processes are regulated by a positive-feedback network of transcription factors comprised, at its core, of the myogenic basic helix-loop-helix factors (MyoD, Myf-5, myogenin, and MRF4) and members of the MEF-2 family (31, 33, 45). Myogenic basic helix-loop-helix and MEF-2 factors act individually and together to maintain their own expression, coordinate withdrawal from the cell cycle, and activate muscle-specific genes (31, 33, 45).

Despite the wealth of information about transcriptional control of myogenic differentiation, several observations suggest that myogenesis, and therefore presumably the core positive-feedback network, may be positively regulated by cell contact-mediated interactions between muscle precursor cells. For example, such interactions are required for myogenic determination and differentiation of explanted embryonic mesodermal cells of both frog and mouse origin, a phenomenon referred to as the community effect (11, 18, 19). Furthermore, differentiation of certain myogenic cell lines is dependent on cell aggregation (1, 37). Finally, it is widely appreciated that high cell density is a strong prodifferentiation condition for C2C12 and other well-studied myogenic cell lines in monolayer culture. The mechanisms by which cell surface proteins mediate such effects are not well understood. Proteins predicted to play a role include various cadherins and immunoglobulin (Ig) superfamily members (14, 20, 25, 37, 47). It seems likely, however, that additional, as yet unidentified proteins will also prove to be involved.

Sialomucins are a heterogeneous class of secreted and cell surface proteins characterized by one or more regions (mucin domains) that contain a high percentage of proline, threonine, and serine residues; examples include CD34, CD43, PSGL-1, GlyCAM-1, MAdCAM-1, and the MUC family (for a review, see reference 39). The threonine and serine residues are heavily O-glycosylated, and these O-linked glycans are very important to sialomucin function. For example, CD34 and other sialomucins serve as high-affinity ligands for selectins; selectins recognize sialylated carbohydrate structures on these ligands during the adhesion cascade by which leukocytes move from blood into tissues (38). Furthermore, sialomucins may serve as signaling receptors that regulate cell proliferation, differentiation, and apoptosis in the hematopoietic compartment, functions that are thought to be, at least in part, dependent on carbohydrate modification (3, 16, 30, 40, 41).

CD164 (also known as MGC-24v and endolyn) is a recently identified sialomucin implicated in hematopoiesis, although it is expressed at the mRNA level in most murine and human tissues, including skeletal muscle (21, 29, 41, 46). CD164 contains an extracellular region comprised of two mucin domains linked by a cysteine-rich motif that resembles a consensus pattern previously found in growth factor and cytokine receptors. CD164 also contains a single-pass transmembrane domain and a 13-amino-acid intracellular region that includes a C-terminal motif able to target the protein to endosomes and lysosomes (21). Like other sialomucins, CD164 is highly glycosylated, containing both O- and N-linked glycans (15). Furthermore, monoclonal antibodies that recognize carbohydrate-dependent epitopes on mucin domain I of CD164 affect adhesion and proliferation of hematopoietic precursor cells (15).

In order to identify cell surface proteins involved in cell contact-mediated regulation of myogenesis, we have used the yeast signal sequence trap, a cDNA library-based screening technique that permits the identification of genes encoding secreted and membrane-associated proteins (22, 23). Subsequent to isolation of positive clones, RNA expression analysis was used to further identify mRNAs whose expression is enhanced during differentiation of cultured myoblasts. We report here the identification of CD164 as a gene that is expressed in proliferating myoblasts, is upregulated during myogenic differentiation, and functions as a positive regulator of this process. Overexpression of CD164 enhanced differentiation of two independent myoblast cell lines, while expression of an antisense CD164 cDNA or soluble extracellular forms of CD164 inhibited differentiation. Finally, treatment of myoblasts with sialidase or O-sialoglycoprotease also inhibited differentiation. These data indicate that sialomucins represent a class of potential effectors of cell contact-mediated regulation of myogenic differentiation. The results also raise the possibility that carbohydrate-based cell recognition events play a role in this phenomenon.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

RNA isolation, library construction, and screening. RNA used for cDNA library construction and dot blot and Northern blot analyses was prepared with the Trizol reagent (Gibco-BRL, Gaithersburg, Md.). The polyadenylated fraction was obtained with the MessageMaker mRNA isolation system (Gibco-BRL). A random-primed, directional cDNA library was constructed in the signal sequence trap vector pSUC2T7M13ORI (23) with mRNA isolated from a mixture of proliferating, differentiating, and fully differentiated C2C12 cells. The library (>2 × 10⁷ independent clones) was constructed with the SuperScript Choice System kit from Gibco-BRL and amplified in Escherichia coli by the protocols of Jacobs et al. (22). The amplified cDNA library was transformed into the SUC2 Saccharomyces cerevisiae strain YTK12 with lithium acetate (5), and plasmids were isolated from colonies which survived invertase selection by growing on medium containing raffinose and antimycin A (22). The plasmid DNAs were then transformed into E. coli by electroporation and purified for sequencing, which was performed at the Mount Sinai DNA Sequencing Core Facility.

Cell culture. C2C12 cells (7) were cultured in Dulbecco's modified Eagle's medium (DMEM) plus 15% fetal bovine serum (FBS) (growth medium [GM]). The myoblast cell lines P2 and F3 (12), 10T1/2 cells, and 293T cells were all cultured in DMEM plus 10% FBS. Cells were induced to differentiate at 90 to 95% confluence by transferring them into DMEM plus 2% horse serum (differentiation medium [DM]).

Overexpression of CD164. A mouse CD164 cDNA (656 bp), including the entire open reading frame, 21 bp of the 5' untranslated region, and 41 bp of the 3' untranslated region, was isolated by reverse transcription (RT)-PCR on total RNA from C2C12 cells with primers 5'-CCGGAATTCGACGCCTGGGCTGAAGACACA-3' (sense) and 5'-CCCAAGCTTCACAAGTTAACTGCCAGTCCA-3' (antisense). The cDNA, whose sequence was identical to mouse CD164/MGC-24v (GenBank accession number AB014464), was ligated into the retroviral expression vector pMV7 (26). Production of recombinant retroviruses and infection of C2C12 and F3 cells were performed as previously described by Kang et al. (24). Infected cultures were selected in medium that contained appropriate antibiotics, and antibiotic-resistant colonies were pooled and analyzed as described in Results.

RNA and protein analyses. Northern blot analyses were performed by fractionating total cellular RNA through agarose-formaldehyde gels, blotting to nylon membranes, and hybridizing with DNA probes as described by Krauss et al. (28). Immunoblot analyses were performed essentially as described by Kang et al. (24). Cultures were harvested in lysis buffer (10 mM Tris-HCl, [pH 7.2]. 150 mM NaCl, 1% Triton X-100, 1% sodium deoxycholate, 0.1% sodium dodecyl sulfate [SDS], 1 mM EGTA) containing 1 mM phenylmethylsulfonyl fluoride (PMSF), 10 ng/ml leupeptin, 50 mM NaF, and 1 mM sodium orthovanadate. Cell lysates were then separated on SDS-polyacrylamide gels and transferred to nitrocellulose membranes (Amersham Life Science, Inc., Arlington Heights, Ill.), and the membranes were probed with one of the following antibodies from the indicated sources: anti-myosin heavy chain (MHC) (MF20 [2]); antitroponin T (TnT; Sigma Chemical Co., St. Louis, Mo.); anti-MyoD (Santa Cruz Biotechnology, Santa Cruz, Calif.); and antimyogenin (FD5) (44). After extensive washing with 40 mM Tris-HCl (pH 8.0)-50 mM NaCl-1 mM EDTA-0.1% Tween 20, the blots were reprobed with horseradish peroxidase-conjugated secondary antibody, and specific protein bands were visualized with the Lumi-Light chemiluminescent detection system (Roche).

Fluorescence microscopy. Immunocytochemical staining for MHC was performed with the monoclonal antibody MF20. Cells were washed with phosphate-buffered saline (PBS), fixed for 1 min with 100% methanol, and blocked with 5% horse serum-PBS for 1 h at room temperature. After incubating cells with MF20 antibody for 1 h at room temperature, cells were washed three times with PBS. Secondary detection was carried out by incubation for 1 h at room temperature with a 1:200 dilution of fluorescein-conjugated goat anti-mouse Ig antibody (Jackson ImmunoResearch Laboratories). Nuclei were counterstained by incubation for 5 min using DAPI (4',6-diamidino-2-phenylindole; 0.5 µg/ml). Fluorescence photomicroscopy was performed on a Nikon Eclipse TS100 microscope, and images were captured digitally with Spot TM 3.0.5 (Apple Event) software.

Myogenic colony assay. C2C12 cells were infected with pBabePuro-based retroviruses (34) harboring the mouse CD164 cDNA in the antisense orientation or, as a control, retroviruses lacking a cDNA insert. After a 2-week selection in GM containing 5 µg of puromycin per ml, macroscopic colonies were easily visible; the cultures were then switched to DM for 24 h and stained with a monoclonal antibody to MHC (MF20) as described by Bader et al. (2). After sequential incubation with biotinylated goat anti-mouse IgG and horseradish peroxidase (HRP)-conjugated streptavidin, cells were stained with 3,3'-diaminobenzidine (Sigma Chemical Co.), and colonies were analyzed by microscopy as described in Results.

CD164 fusion proteins. A PCR-derived fragment containing the signal sequence and entire extracellular region of mouse CD164 was subcloned into the Aptag-2 or Igtag vector (6) to produce CD164-alkaline phosphate (AP) and CD164-Fc, respectively. The following PCR primers were used to amplify the appropriate CD164 cDNA product: 5'-GGAAGATCTGAAGACACAATGTCGGGCTCC-3' (sense), and 5'-GGAAGATCTTGCATCAAAGGTCGACTTCCG-3' (antisense). These vectors, which place the fusion proteins under the transcriptional control of the cytomegalovirus promoter, were transfected into C2C12 cells with the Fugene reagent (Roche Diagnostic Corporation, Indianapolis, Ind.). Ten micrograms of each vector was cotransfected with 1 µg of pBabePuro, and cultures were selected in medium containing 5 µg of puromycin per ml. Aptag-4, which encodes a secreted, soluble form of AP itself, was used as a control. Secretion of CD164-AP and AP into the medium of selected cultures was determined by colorimetric assay of AP activity in conditioned medium (CM) as described by Flanagan and Leder (17). Production of CD164-Fc was determined by immunoprecipitation of CM with protein A-Sepharose, followed by immunoblot analysis of eluted proteins with HRP-coupled goat anti-human IgG.

Treatment of C2C12 cells with sialidase and O-sialoglycoprotease. Clostridium perfringens sialidase was purchased from Sigma, and Pasteurella haemolytica O-sialoglycoprotein endopeptidase (O-sialoglycoprotease) was purchased from Accurate Chemical & Scientific Corp. (Westbury, N.Y.). For treatment of C2C12 cells, cells were incubated with 0.025 U of sialidase or 15 µg of O-sialoglycoprotease per ml for 24 h in DM; fresh enzyme in DM was added to the cells every 24 h for 3 days. To test for the ability of CD164 to serve as a substrate for these enzymes, CM from 293T cells transiently transfected with the CD164-Fc expression vector was used. CM was treated overnight (16 h) with double the concentrations of sialidase or O-sialoglycoprotease stated above and then analyzed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and staining with Coomassie blue, Western blot analysis with detection by HRP-coupled goat anti-human IgG, and the Immun-Blot kit for glycoprotein detection (Bio-Rad, Hercules, Calif.). The kit was used according to the manufacturer's instructions for specific detection of terminal sialic acid residues on proteins bound to a nitrocellulose membrane. Briefly, terminal sialic acid residues on CD164-Fc were subjected to a specific oxidation reaction to label them with biotin; subsequent detection was done with streptavidin-AP and color development reagents.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Identification of CD164 as a gene whose expression is upregulated during C2C12 myoblast differentiation. To isolate cDNAs that encode secreted proteins whose expression is upregulated during myogenic differentiation, a two-step strategy was employed. First, a cDNA library from a mixture of proliferating, differentiating, and fully differentiated C2C12 cells was constructed in the signal sequence trap vector pSUC2T7M13ORI. As originally described by Jacobs et al. (23), this vector harbors a yeast invertase gene that lacks both a methionine to initiate translation and a signal sequence to direct the translated product into the secretory pathway. It is possible, therefore, to use an invertase selection protocol in yeast to isolate cDNAs that provide a start codon followed by a functional signal sequence when they are ligated in frame with the invertase gene (23). Second, cDNA inserts from plasmids that were isolated by such a selection protocol were used as probes on dot blots of total RNAs derived from C2C12 cells at various stages of differentiation, as well as several other myoblast and fibroblast cell lines (data not shown). One such cDNA fragment that hybridized to an mRNA whose expression was enhanced during C2C12 cell differentiation corresponded to the 5' end of murine CD164. A full-length CD164 cDNA was subsequently isolated from C2C12 cells by RT-PCR; this cDNA encodes the major CD164 isoform that includes all six exons (8, 29).

View larger version (48K): [in this window] [in a new window]   FIG. 1.   Expression of CD164 mRNA during myogenic differentiation in vitro. (A) C2C12 myoblasts were grown to near confluence in GM, shifted into DM, and harvested at the indicated time points. The upper panel shows a Northern blot analysis of CD164 mRNA expression. The ethidium bromide-stained gel displaying the 28S and 18S rRNA bands is shown in the lower panel as a loading control. (B) C2C12, F3, and P2 myoblasts and 10T1/2 fibroblasts were cultured during log-phase growth in GM (G) or at confluence for 3 days in DM (D) and harvested for Northern blot analysis as shown in the top panel. The ethidium bromide-stained gel displaying the 28S and 18S rRNA bands is shown in the lower panel as a loading control.

Overexpression of CD164 enhances differentiation of C2C12 and F3 myoblasts. To investigate the function of CD164 during myoblast differentiation, C2C12 and F3 cells that overexpress CD164 were generated. Cells were infected with control (pMV7) or CD164-expressing (pMV7-CD164) retroviruses and selected for G418 resistance; drug-resistant colonies were pooled and examined for expression of CD164 mRNA and their ability to differentiate. The CD164 virus-infected cells (designated C2C12/CD164 cells and F3/CD164 cells) displayed expression of both an endogenous 3.0-kb CD164 mRNA and a vector-derived 3.9-kb CD164 mRNA, while control vector-infected cells (designated C2C12/neo cells and F3/neo cells) expressed only the endogenous mRNA species (Fig. 2A). Overexpression of CD164 had no obvious effect on the morphology of C2C12 or F3 cells or on their ability to proliferate in GM (data not shown).

View larger version (147K): [in this window] [in a new window]   FIG. 2.   Overexpression of CD164 enhances differentiation of C2C12 and F3 myoblasts. C2C12 and F3 cells were stably infected with recombinant control (pMV7) or CD164-expressing (pMV7-CD164) retroviruses, selected with G418, and analyzed for expression of CD164 mRNA, myotube formation, and expression of muscle-specific proteins. (A) Northern blot analysis of CD164 mRNAs in infected cells. , infection with pMV7 virus; +, infection with pMV7-CD164 virus. (B) Immunofluorescence photomicrographs of infectants. C2C12 and F3 cells stably infected with the indicated viruses were cultured in DM for 3 days, fixed, and doubly stained with a monoclonal antibody to MHC (detected with a fluorescein-conjugated secondary antibody) and with DAPI for nuclear DNA. (C) Western blot analysis of muscle-specific proteins and, as a control, CDK2 expressed by C2C12 infectants cultured in DM for the indicated times. , infection with pMV7 virus; +, infection with pMV7-CD164 virus. (D) Western blot analysis of muscle-specific proteins and, as a control, CDK2 expressed by F3 infectants cultured in DM for the indicated times. , infection with pMV7 virus; +, infection with pMV7-CD164 virus. The numbers below the Western blots in C and D represent relative levels of each protein derived by densitometric analysis of the signal.

Expression of an antisense CD164 vector inhibits C2C12 cell differentiation. To determine whether inhibition of CD164 expression interferes with myogenic differentiation, the ability of an antisense CD164 expression vector to inhibit colonies of C2C12 cells from differentiating was assessed. C2C12 cells were infected with control (pBabePuro) or CD164 antisense-expressing (pBabePuro/asCD164) retroviruses and selected with puromycin. When puromycin-resistant colonies emerged, they were shifted into DM for 1 day, and the percentage of MHC-positive cells within the colonies was scored (Table 2). Compared to colonies infected with pBabePuro, colonies infected with pBabePuro/asCD164 displayed substantially fewer MHC-positive cells, suggesting that CD164 levels may be rate-limiting for myoblast differentiation. To test the effectiveness of the antisense vector, Northern blot analysis was performed on a polyclonal C2C12 line that had been stably propagated after infection with pBabePuro/asCD164 and selection in puromycin. Figure 3 shows that this line expressed exogenous vector-derived mRNA of the predicted size (3.6 kb) and displayed a concomitant loss of the endogenous CD164 mRNA (3.0 kb), presumably via degradation of RNA:RNA hybrids. As expected, these cells showed a reduced ability to differentiate in response to DM relative to control cells (data not shown).

View larger version (44K): [in this window] [in a new window]   FIG. 3.   C2C12 cells that express an antisense CD164 construct display a concomitant loss of endogenous CD164 mRNA. Northern blot analysis of CD164 mRNAs in infected cells is shown. , infection with control pBabePuro virus; +, infection with pBabePuro/asCD164 virus. Note that the cells infected with pBabePuro/asCD164 virus express a vector-derived band (3.6 kb) and have much reduced levels of endogenous CD164 mRNA (3.0 kb). A full-length, double-stranded CD164 probe was used to detect both mRNAs. The ethidium bromide-stained gel displaying the 28S and 18S rRNA bands is shown in the lower panel as a loading control.

Secreted, soluble extracellular regions of CD164 inhibit C2C12 cell differentiation. Many membrane-associated sialomucins play a role in cell-cell interactions by serving as ligands or receptors for selectins or other yet to be identified extracellular and membrane-associated factors (38-41). If CD164 functions to mediate interactions between myoblasts that contribute positively towards differentiation, it would be predicted that expression of soluble extracellular regions of CD164 by C2C12 cells might compete with endogenous membrane-bound CD164 and affect differentiation. To test this possibility, recombinant soluble fusion proteins that contain the entire CD164 extracellular region coupled to either AP or the Fc region of human IgG (CD164-AP and CD164-Fc, respectively; Fig. 4A) were constructed. C2C12 cells were stably transfected with expression vectors for CD164-AP, CD164-Fc, and, as a control, secreted AP alone. AP activity was easily detected in the CM of the CD164-AP and AP transfectants (data not shown), and CD164-Fc was detected immunologically in CM from the CD164-Fc transfectants (Fig. 4B).

View larger version (193K): [in this window] [in a new window]   FIG. 4.   Secreted, soluble extracellular regions of CD164 inhibit C2C12 cell differentiation. (A) Schematic representation of CD164 and the soluble fusion proteins CD164-AP and CD164-Fc. The pictured structure of two mucin domains linked by a cysteine-rich region containing putative disulfide bonds is based on data from reference 21. (B) Western blot analysis of CD164-Fc in CM from transiently transfected 293T cells and stably transfected C2C12 cells. (C) Left column, immunofluorescence photomicrographs of C2C12 cells stably transfected with expression vectors for the indicated proteins and cultured in DM for 3 days. Right column, immunofluorescence photomicrographs of C2C12 cells cultured for 3 days in a 1:1 mixture of DM and CM derived from 293T cells transiently transfected with expression vectors for the indicated proteins. Cultures were fixed and doubly stained with a monoclonal antibody to MHC (detected with a fluorescein-conjugated secondary antibody) and with DAPI for nuclear DNA. (D) Western blot analysis of muscle-specific proteins expressed by C2C12 cells stably transfected with expression vectors for the indicated proteins and cultured under differentiation-inducing conditions for the indicated times (see Materials and Methods for details). The numbers below the Western blots in D represent relative levels of each protein derived by densitometric analysis of the signal.

Treatment of C2C12 cells with sialidase or O-sialoglycoprotease inhibits differentiation. Sialomucins, including CD164, have the common characteristic of being highly glycosylated polypeptides, containing both O- and N-linked carbohydrate side chains (39, 41). Monoclonal antibodies against CD164 that alter the adhesive and proliferative properties of hematopoietic precursors recognize epitopes that are destroyed by treatment of cells with sialidase, which cleaves terminal sialic acid residues on O- or N-linked carbohydrates, or O-sialoglycoprotease, an enzyme that selectively degrades O-sialomucins (9, 32). It would be predicted, therefore, that treatment of C2C12 cells with these enzymes would inhibit differentiation.

View larger version (162K): [in this window] [in a new window]   FIG. 5.   CD164 is a substrate for sialidase and O-sialoglycoprotease, and treatment of C2C12 cells with these enzymes inhibits differentiation. (A) Western blot analysis of the effects of sialidase and O-sialoglycoprotease on CD164-Fc present in CM of transiently transfected 293T cells. (B) Detection of terminal sialic acid residues on CD164-Fc in 293T CM treated or not with sialidase (see Materials and Methods for details). (C) Assessment of nonspecific protease activity in preparations of sialidase and O-sialoglycoprotease. Duplicates of the samples in panel A were fractionated by SDS-PAGE, and the gel was stained with Coomassie blue. Lane M, protein markers. (D) Immunofluorescence photomicrographs of C2C12 cells cultured in DM for 3 days with and without the indicated enzymes. Cultures were fixed and doubly stained with a monoclonal antibody to MHC (detected with a fluorescein-conjugated secondary antibody) and with DAPI for nuclear DNA. (E) Western blot analysis of muscle-specific proteins and, as a control, CDK2 expressed by C2C12 cells cultured in DM for 3 days with or without the indicated enzymes. The numbers below the Western blots in E represent relative levels of each protein derived by densitometric analysis of the signal.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Determination and differentiation of skeletal muscle precursors requires cell-cell contact, and various cadherins and immunoglobulin superfamily members have been proposed to play a role in mediating this requirement (14, 20, 25, 37, 47). We report here that CD164, a cell surface sialomucin, has properties consistent with a role in mediating cell-cell interactions that contribute to myogenesis. CD164 was isolated in this study by signal sequence trapping in yeast, an efficient method for selecting genes that encode secreted proteins from complex libraries (22, 23). The additional step of rapidly screening positive clones for their expression pattern during myogenic differentiation permitted the identification of CD164 as a gene expressed in proliferating C2C12 myoblasts that is upregulated during differentiation. Subsequent functional analyses demonstrated that overexpression of CD164 in myoblast cell lines accelerated expression of biochemical markers of differentiation and enhanced formation of multinucleate myotubes, while expression of antisense CD164 or soluble extracellular regions of CD164 inhibited myogenesis. Finally, treatment of C2C12 myoblasts with sialidase or O-sialoglycoprotease, two enzymes that destroy functional epitopes on CD164 (15), also inhibited differentiation. Taken together, these data are consistent with the hypothesis that CD164 may play a rate-limiting role in myogenic differentiation in vitro.

While manipulation of CD164 levels or function could enhance or inhibit differentiation of myoblasts, as assessed by formation of multinucleate myotubes or by expression of the differentiation markers myogenin, MHC, and TnT, it did so without altering the levels of MyoD protein produced by these cells. This observation suggests that CD164 might somehow serve to increase MyoD activity at a posttranslational level, an event that precedes induction of myogenin and muscle structural proteins in myogenic differentiation. Very similar data were obtained in this lab with C2C12 and other myoblast cell lines by stable expression of wild-type or interfering mutants of CDO, a receptor-like protein that contains five Ig and three FNIII repeats (25). It will be interesting to determine whether CD164 and CDO exert their effects by a similar mechanism. However, whereas CDO has a long intracellular region that could easily function as a docking site for adapter proteins or enzymes involved in signaling, CD164's intracellular region is only 13 amino acids long, including 4 at the carboxy terminus that target the protein to lysosomes and endosomes (21).

It is not immediately obvious, therefore, how CD164 could serve as a signaling receptor in this context. One possibility is that CD164 may facilitate signaling by binding to other receptors or adhesion molecules as part of a complex. Alternatively, CD164 might function as a ligand for a signaling receptor present on the surface of adjacent myoblasts. It is also important to note that alteration of CD164 level or function appeared to have a more pronounced effect on morphological aspects of differentiation (i.e., formation of multinucleate myotubes) than on biochemical aspects of differentiation (i.e., expression of muscle-specific proteins). CD164 may therefore be more directly involved in regulating myoblast fusion than in coordinating the entire myogenic differentiation program. Further studies will be needed to address this point.

The most clearly documented functions for sialomucins pertain to the traffic signals that regulate leukocyte localization in the vasculature (for a review, see reference 38). Molecules like CD34, GlyCAM, and MAdCAM are expressed on high endothelial venules to provide sialylated carbohydrate ligands for selectin receptors expressed on various leukocytes. Interactions of the opposite "polarity," between specific sialomucins on leukocytes and selectins on activated endothelial cells, also occur. Selectin-sialomucin interactions are responsible for tethering flowing leukocytes to the vessel wall and for transient adhesive events involved in leukocyte rolling. These are required first steps in an adhesion cascade, with subsequent interactions between integrins and immunoglobulin superfamily members providing the more stable adhesive connections necessary for monocytes, neutrophils, and other cell types to exit the bloodstream and enter tissues. Because cell surface proteins of the cadherin, immunoglobulin, and, now, sialomucin families have all been implicated in myogenesis, it is tempting to speculate that, by analogy with the leukocyte traffic model, a series of adhesive and cell contact-mediated signaling interactions between myoblasts underlie phenomena such as the prodifferentiation effects of high cell density.

The function of sialomucins in leukocyte traffic depends on their specific glycosylation. Likewise, CD164 bears functional epitopes that are carbohydrate dependent (15). These observations raise the interesting possibility that carbohydrate-based cell recognition events may play a role in cell contact-mediated functions that are important in myogenesis. The fact that treatment of C2C12 cells with sialidase inhibited differentiation suggests that this may be true. It is important to note that many cell surface proteins in addition to CD164 are substrates for sialidase, and the most important substrates involved in mediating its inhibitory effects on myogenesis are not identified. Treatment of C2C12 cells with O-sialoglycoprotease also inhibited differentiation. This enzyme has a much more restricted set of substrates, apparently recognizing O-linked carbohydrates on mucin domain-containing proteins specifically and cleaving the polypeptide chain nearby (9, 32). CD164 may well be a major substrate of O-sialoglycoprotease on C2C12 cells, although the predicted cleavage and release of an amino-terminal fragment do not distinguish between the carbohydrate or peptide sequences lost as being of primary importance. Furthermore, the proteins to which the extracellular region of CD164 binds are not known; whether such interactions occur with lectin-like molecules via its extensive O- and N-linked sugars or with other types of receptors via its core protein sequence is also unknown. Nevertheless, expression of both lectin activity and specific lectin domain-containing proteins is regulated during myogenesis in vitro and in vivo (10, 13, 35, 36, 42), and the potential role of carbohydrate-based cell recognition events in differentiation is worthy of further exploration.

It is clear that sialomucins play roles other than those documented in leukocyte traffic as well. CD34 is a clinically useful marker of adult hematopoietic stem cells (27). Of relevance to the potential role of sialomucins in myogenesis, CD34 was also recently shown by Beauchamp et al. to be a marker for most, but not all, quiescent muscle satellite cells (4). CD34 mRNA expression is extinguished within 24 h of satellite cell activation, suggesting that it may play a role in maintaining the quiescent state of these cells (4). Additionally, CD34 is expressed in C2C12 myoblasts at very low levels in the great majority of cells in the culture; those few that express high levels of CD34 appear to represent a subset of cells that do not differentiate (4). These results suggest that CD34 and CD164 very likely perform different functions during myogenesis. Nevertheless, taking the results of Beauchamp et al. (4) together with the findings of this study, it can be concluded that sialomucins represent a family of cell surface proteins newly recognized as potentially important regulators of various aspects of muscle development.