J. Cell Sri. 65, 233-248 (1984)
233
Printed in Great Britain © The Company of Biologists Limited 1984
IDENTIFICATION OF SPERMATOGENIC CELL
PLASMA MEMBRANE GLYCOPROTEINS BY TWODIMENSIONAL ELECTROPHORESIS AND LECTIN
BLOTTING
CLARKE F. MILLETTE* AND B. KEYES SCOTT
Department of Anatomy and Laboratory of Human Reproduction and Reproductive
Biology, Harvard Medicat School, 45 Shattuck Street, Boston, MA 02115, U.SA.
SUMMARY
Plasma membrane glycoproteins present in purified mouse spermatogenic cells have been identified by two-dimensional polyacrylamide gel electrophoresis and lectin blotting techniques. Four
membrane glycoproteins labelled with Bandeiraea simplicifolia lectin (I) have been detected, ranging in M, from 55 000 to 76000 and in pi from 6-0 to 6-3. Only one of these proteins, p76/6-3, is
synthesized by short-term in vitro cultures of spermatogenic cells, as determined by the incorporation of [35S]methionine. Approximately 20 surface glycoproteins labelled with concanavalin A have
been identified, ranging in M, from 50000 to 151000 and in pi from 5-7 to 7-0. None of the
components detected with B. simplicifolia lectin (I) are labelled significantly with concanavalin A.
A major concanavalin A-binding protein in the membranes of both pachytene spermatocytes and
round spermatids is pl5l/6-0. This glycoprotein has been previously shown to be exposed on the
outer surface of spermatogenic cell membranes and may represent a mediator of germ cell-Sertoli
cell interactions. Furthermore, two constituents identified in the present study represent stagespecific markers. Component p73/5-7 is detected with concanavalin A only in the membranes of
pachytene spermatocytes. Conversely, p84/6-3 is found only in round spermatid membranes.
These results, then have: (a) provided a map of membrane glycoproteins in developing mouse male
germ cells; (b) identified pl5l/6-0 as a membrane constituent of possible functional significance;
and (c) identified the first reported glycoprotein surface differentiation markers for mouse spermatogenesis.
INTRODUCTION
Plasma membrane glycoproteins have been implicated as determinants of cellular
differentiation in many diverse biological systems including myogenesis (Marino,
Cossu, Neri & Molinaro, 1980; Walsh & Phillips, 1981), histogenesis of the neural
retina (Edelman, 1983), haematopoiesis (Fehlman, Laufler & Marceu, 1976; Ackerman & Freeman, 1979), and differentiation of embryonal carcinoma cells (Muramatsu et al. 1978). In addition, cell surface glycoproteins are thought to mediate specific
cell-cell interactive events as varied as the self-association of slime moulds (West,
McMahon & Molday, 1978; Ivatt, Prem Das, Henderson & Robbins, 1981) and the
metastasis of malignancies (Greig & Jones, 1977; Atkinson & Hakimi, 1980). Other
cellular functions such as pinocytosis (Neufeld & Ashwell, 1980) and proliferation
(Glaser, 1980) may also involve cell surface glycoproteins. It is likely, therefore, that
•Author for correspondence.
234
C. F. Millette and B. K. Scott
plasma membrane glycoproteins are important determinants of differentiation and
interaction during mammalian spermatogenesis, which entails mitotic proliferation,
meiosis, remarkable alterations in cell shape and extensive cellular adhesions between
germ cells and Sertoli cells within the seminiferous epithelium (Bellve, 1979; Ewing,
David & Zirkin, 1980).
Furthermore, the mammalian testis and particularly late spermatogenic cells exhibit extraordinarily high rates of acetate incorporation into the isoprenoid lipid
dolichol (Rupar & Carroll, 1978; James & Kandutsch, 1980; Potter, Millette, James
& Kandutsch, 1981). Dolichol derivatives are well-known intermediates in the
glycosylation of ./V-asparagine-linked cell surface glycoproteins (Waechter & Lennarz, 1976; Hubbard & Ivatt, 1981), but many aspects of dolichol metabolism during
spermatogenesis and in the male reproductive tract (Wenstrom & Hamilton, 1980)
remain largely unexplained. It is of particular interest that mouse pachytene spermatocytes incorporate acetate into dolichol at a rate fivefold higher than that noted
in preleptotene spermatocytes (Potter et al. 1981). Moreover, the ratio of acetate
incorporation into dolichol compared to incorporation into cholesterol increases
dramatically in both pachytene spermatocytes and round spermatids (Potter et al.
1981). The high rate of dolichol synthesis as well as the observed temporal changes
in dolichol metabolism, therefore, suggest strongly that membrane glycoproteins and
alterations thereof may be significant regulators of sperm differentiation within the
testis.
Few investigators have analysed the overall glycoprotein composition of developing
spermatogenic cells or have compared glycoproteins expressed at precise stages of,
spermatogenesis. Stage-specific plasma membrane markers have been identified both
serologically and biochemically in the mouse (Millette & Bellve, 1977, 1980; Millette
& Moulding, 1981a), rabbit (Romrell, O'Rand, Sandow & Porter, 1982), guinea pig
(Tung, Han & Evan, 1979) and rat (Tung & Fritz, 1978). Recent studies have also
made such analysis feasible for the human (Shepard, Millette & DeWolf, 1981;
Narayan, Scott, Millette & DeWolf, 1983). With few exceptions, however, it is not
known whether the cell surface constituents already described are glycosylated. Surface glycoproteins present on spermatogenic cells (O'Rand, 1981; Grootegoed et al.
1982a) and on spermatozoa (Olson & Hamilton, 1978) have been reported, but none
of these components are known to be unique to individual stages of cell differentiation
in the testis. The physiological functions of these glycoproteins are presumed to be
operative during sperm maturation in the epididymis and/or fertilization. Glycoprotein membrane constituents with functions manifest during spermatogenesis itself
have not yet been described.
The present experiments have been designed to analyse the membrane glycoprotein
composition of purified mouse pachytene spermatocytes and round spermatids in
order to facilitate future investigations of dolichol-mediated glycosylation in the testis.
In addition, comparisons of the glycoprotein maps obtained have allowed the identification of a particular surface component unique to spermatocytes and another constituent unique to spermatids. The results will allow studies of functional roles for
spermatogenic cell surface glycoproteins.
Spermatogenic cell surface glycoproteins
MATERIALS
235
AND METHODS
Materials
The following compounds were obtained from Sigma Chemical Co.: trypsin, bovine pancreas,
type I I I ; bovine serum albumin, fraction V; pyruvate, Na + salt, type I I ; DL-lactate, Na + salt,
~ 6 0 % syrup; Concanavalin A (ConA), type IV; dithiothreitol; Coomassie Brilliant Blue R; horseradish peroxidase, type VI; 4-chloro-l-naphthol. [35S] methionine, SJ.204, 600-1300 Ci/mmol and
125
I-labelled protein A, IM.112, 100 Ci/ml were obtained from Amersham Corp. Collagenase, CLS
grade, was obtained from Worthington Corp. Minimum essential medium Eagle (modified) with
Earle's salts, without methionine or glutamine, was obtained from Flow Laboratories. Penicillin/
streptomycin solution, no. 600-5145, and dialysed foetal bovine serum were purchased from Grand
Island Biochemical Co. Nitrocellulose paper, BA-85, was purchased from Schleicher and Schuell.
Bandeiraea simplicifolia lectin (I) (BSL), with specificity for a-D-galactosyl groups, was prepared
(Baues & Gray, 1977) using seeds from Calbiochem. Antiserum to BSL was prepared by immunizing female New Zealand White rabbits (White Pine Rabbitry) intramuscularly with 5 mg purified
lectin emulsified in complete Freund's adjuvant (DIFCO). One month later a booster injection of
2mg lectin emulsified in incomplete Freund's adjuvant (DIFCO) was administered both
intramuscularly and subcutaneously. Animals were bled from the marginal ear vein 1 week following
the booster injection. Antiserum was heat-inactivated at 56°C for 1 h before use.
Animals and cell preparation
Adult CD-I mice aged 60-120 days were obtained from Charles River Breeding Laboratories.
Adult Tac: (SW)fBR mice of the same age were obtained fromTaconic Farms, Inc. No differences
were detected in the analysis of samples from either of these related mouse types. Mixed
seminiferous cell suspensions were prepared according to methods previously described (Romrell,
Bellvfi & Fawcett, 1976; Bellvfi, Millette, Bhatnager & O'Brien, 1977). These cell suspensions
consist predominantly of pachytene spermatocytes, round spermatids and residual bodies.
Contaminating Sertoli cells, Leydig cells or other interstitial somatic cell types constitute less than
5 % of the total suspension. Purified populations of pachytene spermatocytes and round spermatids
(steps 1-8 of spermiogenesis) were obtained from the mixed germ cell suspensions using unit gravity
sedimentation as already detailed (Romrell et al. 1976). Cell purity was assayed by Nomarski
interference microscopy using established criteria (Romrell etal. 1976; Bellve'e/a/. 1977). Purities
of isolated pachytene spermatocytes and round spermatids were greater than 90%. Pachytene
spermatocytes were usually contaminated with low numbers (<5 % total cells) of binucleated round
spermatids. Round spermatids were contaminated by residual bodies (<5 % total cells), but not by
significant numbers of pachytene spermatocytes. All cell preparations were washed at least three
times by centrifugation at 200£ for 5 min at 4°C (Romrell et al. 1976) before polyacrylamide gel
electrophoresis or the preparation of plasma membrane fractions.
Plasma membrane preparation
Purified plasma membranes were prepared from mixed seminiferous cells, from isolated
pachytene spermatocytes or from isolated round spermatids using methods previously reported
(Millette, O'Brien & Moulding, 1980). Biochemical and ultrastructural studies have established that
these surface membrane fractions are not contaminated significantly by intracellular membranes,
organelles or soluble cytoplasmic components (Millette et al. 1980).
In vitro cell culture
Mixed seminiferous cell suspensions were cultured for 26 h at 33 °C in minimal essential medium,
Earle's salts, without methionine. The culture medium was supplemented with 10 % (v/v) dialysed
foetal bovine serum, 100/ig/ml streptomycin sulphate, 60/ig/ml penicillin G (K + salt), 2mM-Lglutamine, 1 mM-pyruvate and 6mM-lactate. Lactate was included in these cultures since other
investigators have demonstrated increased survival and metabolic activity by short-term in vitro
cultures of rat spermatogenic cells with added lactate (Jutte, Grootegoed, Rommerts & van der
Molen, 1981; Mita & Hall, 1982). Atotal of 8-5 X 107 cells were cultured in a final volume of 2-2 ml.
236
C. F. Millette and B. K. Scott
Cultures were maintained under 5 % CO2: 95 % air. The medium also contained [35S]methionine
(0-35 mCi). Cells were harvested by simple agitation and washed by centrifugation at 200^ before
plasma membranes were isolated. Mouse spermatogenic cells do not adhere to plastic cultureware.
Polyacrylamide gel electrophoresis
Two-dimensional gel electrophoresis was conducted according to O'Farrell (1975) using 20%
glycerol and 40 mM-dithiothreitol in the sample buffer. Stacking gels containing 5 % acrylamide and
linear acrylamide gradients (7 % to 12 %) were used for all second-dimension gels. Gels were fixed
and stained with Coomassie Brilliant Blue R, 0-1 % in 45 % methanol, 10% acetic acid. Gels were
destained in 7-5 % acetic in 5 % methanol and photographed using Ektapan film. In some experiments gels were stained using silver nitrate with a procedure modified from those of Morrisey (1981)
and Wray, Boulikas, Wray & Hancock (1981). No significant qualitative differences have been
detected using either staining procedure with the protein loads (100-1200/*g) analysed in these
experiments. For autoradiography gels were transferred to nitrocellulose sheets, which were, in
turn, air-dried and exposed to Kodak X-Omat AR film. Exposure times were as indicated in the
figure legends.
Lectin blotting
Proteins were transferred electrophoretically from two-dimensional polyacrylamide gels to
nitrocellulose sheets according to the procedure of Towbin, Staehelin & Gordon (1979) with slight
modification. Gels were first incubated in three changes of transfer buffer (25mM-Tris, 192 mMglycine (pH8-3) containing 25 % (v/v) methanol) for a total of 90min. Transfer was effected for
15—17 h in an Electroblot apparatus (E-C Apparatus Corp.) at 3-9 V. Lectin blots were then
prepared and developed using a procedure devised by E. M. Eddy and colleagues, similar to that
recently reported by Hawkes (1982). The exact experimental protocols differed for blots stained
with ConA and BSL.
ConA staining was conducted according to the following protocol. Blots were first incubated in
blocking solution for at least 1 h at 37-40°C. Blocking solution consisted of 0-9 % (w/v) NaCl, 3 %
(w/v) bovine serum albumin, 10min-NaN3 in 10mM-Tris-HCl (pH7-4). Blots were then rinsed
in the Tris-HCl buffer containing 0-9% (w/v) NaCl (Tris-saline) before their incubation with
lOOg/ml ConA in blocking solution for 1 h at room temperature. Excess lectin was removed by
multiple washes in Tris-saline for a total of 1 h. Rinsed blots were then incubated with 100/ig/ml
horseradish peroxidase in blocking solution for 1 h at room temperature. Excess peroxidase was
removed by sequential washes with Tris-saline until the rinse solution failed to react with 4-chloro1-naphthol. Total rinse time for this step was 30-60min. Each wash was 5-10 min. Finally, blots
were developed with 0-06% (w/v) 4-chloro-l-naphthol according to Hawkes (1982), except that the
final H2O2 concentration was 0-025 % (v/v). Development time was usually 30s to 5 min.
Staining for BSL-reactive proteins was conducted according to the following protocol. Blots were
incubated in blocking solution and rinsed as already described for ConA. After rinsing blots were
incubated with 100^g/ml BSL in blocking solution for 1 h at room temperature. Excess lectin was
removed by rinsing as previously detailed. Labelled blots were then incubated with rabbit antiserum, diluted 1 part in 10 parts with blocking solution, for 1 h at room temperature. Excess
antiserum was removed by rinsing in Tris-saline. Blots were finally incubated with 12SI-labelled
Fig. 1. Identification of ConA-binding glycoproteins in whole cell homogenates of mouse
pachytene spermatocytes. A. Two-dimensional electrophoresis of pachytene cell
homogenate stained with Coomassie Blue (CB). The circle indicates p93/5-7, a major
constituent stained with CB and ConA. Arrows indicate major CB staining proteins that
are not ConA-binding glycoproteins. The bracket indicates the position of p73/5-7 which
is not stained by CB. Note that pl51/6-0 represents a minor component in the total
homogenate. B. Corresponding nitrocellulose blot stained with ConA. As before, p93/5-7
is circled and the arrows indicate the positions of CB staining proteins not labelled by
ConA. Note that p73/5 -7 is labelled with ConA and that pi 51/6-0 represents an important
lectin-binding constituent. Total protein loaded was 1170/ig. PS, pachytene spermatocytes.
Spermatogenic cell surface glycoproteins
6-2
6;0
A/frx10- 3
57
237
-Pi
P151/60
94-
V
68-
45-
1A
PS CB
3151/60
1073/57
68-
B
PS ConA
Fig. 1
238
C. F. Millette and B. K. Scott
protein A (total of 1-73 X lO^c.p.m. added) for 1 h at room temperature, air-dried and
autoradiographed. This procedure is based on that of Legocki & Verma (1981).
RESULTS
Homogenates of intact pachytene spermatocytes (Fig. 1) and intact round spermatids (Fig. 2) were first analysed for the expression of ConA-reactive glycoproteins.
Blots of each cell type revealed approximately 20 ConA-binding components, with
MT 50000-151000 and pi 5-7-7-0. Deposition of the 4-chloro-l-naphthol reaction
product on these gels was prevented completely by including lOOmM-a-methyl-Dmannoside in the blocking solution during the labelling procedure, demonstrating
that the observed proteins were recognized as a consequence of lectin affinity. Comparison of the whole cell homogenate blots from both cell types revealed overall
similarity. The majority of proteins detected with Coomassie Blue were not labelled
by the lectin (arrows, Figs 1, 2), as might be expected if the identified glycoproteins
represented plasma membrane components each constituting a small fraction of total
cellular protein. Conversely, most of the ConA binding proteins were not major
quantitative components in the cell homogenates as determined by staining with
Coomassie Blue or silver nitrate. For example, pl51/6-0 was a minor constituent in
whole pachytene spermatocytes and round spermatids, but represented one of the two
major lectin binding proteins detected in each cell type (Figs 1, 2). These data
suggested, therefore, that the reactive glycoproteins were perhaps localized at the cell
surface.
The other major ConA-binding protein, p93/5-7, was noted as an important quantitative constituent using both Coomassie Blue and ConA in spermatogenic cells
(circles, Figs 1, 2). Previous investigations from this laboratory have identified
p93/5-7 as a component of purified plasma membranes from spermatocytes and
spermatids (Millette & Moulding, 1981a). In addition, p93/5-7 has been identified
in the membranes of both cell types as demonstrated by vectorial iodination studies
(Millette & Moulding, 19816). The reason for the apparent abundance of p93/5-7 in
whole cell homogenates is not yet clear.
Although overall similarity was observed between the lectin blots of homogenized
spermatocytes and spermatids some striking differences were noted. For example,
constituent p73/5-7 was detected only in preparations of pachytene spermatocytes
(Fig. 1). Another glycoprotein, p84/6-3, was labelled by ConA only in preparations
Fig. 2. Identification of ConA-binding glycoproteins in whole cell homogenatea of mouse
round spermatids. A. Two-dimensional electrophoresis of spermatid cell homogenate
stained with Coomassie Blue (CB). For comparison with Fig. 1, p93/5-7 is circled and
examples of shared constituents are indicated by arrows. The bracket indicates the position
of p84/6-3, which does not stain using CB. Note that pIS 1/6-0 again represents a relatively minor CB staining component, B. Corresponding nitrocellulose blot stained with ConA.
The circle indicates p93/5-7 and the arrows indicate the positions of major CB-binding
proteins not stained by ConA. Important ConA-binding constituents include p84/6-3,
p93/S-7 and pl5l/6"0. Comparison with Fig. 1 reveals that p73/5-7 is detected only in
pachytene spermatocytes and p84/6 - 3 is detected only in round spermatids. Total protein
loaded was 600 ng. RS, round spermatids.
239
Spermatogenic cell surface glycoproteins
6;2
60
57
-3
-Pi
Mrx10
i
Jg15i/60
^
94-
68-
45-
2A
RSCB
151/6-0
94-
J384/6-3
68-
45-
B
RS ConA
Fig. 2
240
C. F. Millette and B. K. Scott
of round spermatids (Fig. 2). These two proteins, therefore, represent stage-specific
markers of mouse spermatogenesis.
To demonstrate the plasma membrane localization of glycoproteins identified in the
intact cell studies blots were prepared using purified surface membranes from mixed
populations of seminiferous cells. For comparison, plasma membranes were also
isolated for electrophoretic analysis after short-term in vitro culture of the mixed cells
in medium containing [ 35 S]methionine. The resultant two-dimensional gels were
transferred to nitrocellulose and the blots subsequently autoradiographed. Companion blots were stained with ConA to allow direct comparison with the earlier
experiments. Finally, additional blots were stained with BSL in order to determine
whether non-ConA-binding cell surface glycoproteins were present. Results of these
studies are presented in Fig. 3.
The autoradiographic pattern of [ 35 S]methionine-labelled plasma membranes
resembled closely the electrophoretic profiles detected using Coomassie Blue or silver
nitrate (Fig. 3A). Blots stained with BSL, which binds to a-D-galactopyranose
glycosides but not mannose derivatives (Delmotte & Goldstein, 1980), revealed four
reactive glycoproteins, p76/6-3, p70/6-2, p70/6-0 and p55/6-0 (Fig. 3B). Only one
of these proteins, p76/6 - 3, was also synthesized in vitrv, as determined by the incorporation of labelled methionine (compare Fig. 3A, B). None of the BSL-binding
proteins was significantly labelled using ConA (compare Fig. 3B, c). Control blotting
experiments using normal rabbit serum were negative, indicating that the observed
staining was specific to BSL. In addition, BSL binding could be inhibited competitively using D-galactose.
Blots of mixed membranes stained with ConA revealed glycoprotein patterns virtually identical to those seen earlier in whole cell homogenates. Approximately 20
reactive components were noted reproducibly, including pl51/6-0 and the stagespecific markers p73/5-7 and p84/6-3. The latter two glycoproteins should have coexisted in these preparations since both pachytene spermatocytes and round spermatids were present in the original cell suspensions. Comparison of the ConA blot
pattern and the autoradiographic pattern indicated that although most of the lectin
binding constituents were synthesized in vitrv, neither p73/5-7 nor p84/6 - 3 incorporated significant methionine under the culture conditions employed in these experiments (compare Fig. 3A, C). The blot analyses of mixed germ cell membranes, then,
demonstrated: (a) that the ConA-binding proteins detected in whole cell homogenates
were in fact cell surface components; and (b) that additional membrane glycoproteins
Fig. 3. Identification of glycoproteins in purified spermatogenic cell plasma membranes.
A. [35S]methionine incorporation into purified membranes after 26 h in culture. Brackets
indicate the relative positions of p84/6-3 and p73/5-7, neither of which exhibits substantial labelling. The arrow indicates p76/6-3 for comparison with B. A total of 5-8 X 10s
c.p.m. was loaded onto thefirst-dimensiongel for isoelectric focusing. Exposure time was
24 h. B. Corresponding nitrocellulose blot stained with BSL. Of four labelled components,
p76/6-3, p70/6-2, p70/6-0 and p55/6-0, only one, p76/6-3, is also detected using [35S]methionine (compare A), C. Corresponding nitrocellulose blot stained with ConA. None
of the BSL-binding constituents label with ConA. Both p73/5-7 and p84/6'3 are labelled,
as is pl5l/60. A total of ~200^g protein was loaded onto the gels for B,C.
241
Spermatogenic cell surface glycoproteins
6-3
60
57
P151/60
i
63
60
Mrx10' 3
94—
( )
( )
68—
45—
3A
u
C35S]Met
6 3
B
BSL
60
5-7
p151/6-0
94—
p73/57
68—
45—
Con A
Fig. 3
242
C. F. Millette and B. K. Scott
65
6-3
6 0 5;7
-pi
•151/6-0
68-
45-
PS membranes
4A
94—
68-
45-
B
RS membranes
Fig. 4
Spermatogenic cell surface glycoproteins
243
existed and could be identified using lectins of varying saccharide specificities. These
experiments did not, however, address the stage-specific expression of p73/5-7 and
P 84/6-3.
Formal demonstration of the surface membrane localization and specific temporal
appearance of marker glycoproteins p73/5-7 and p84/6-3 was completed using
purified plasma membrane fractions from isolated populations of pachytene spermatocytes and round spermatids. Results of ConA blotting experiments conducted on
these samples are compared in Fig. 4.
As expected, there was overall similarity between the patterns from spermatocyte
and spermatid membranes. Surface glycoproteins detected in each cell type were
those constituents already identified in whole cell homogenates (Figs 1, 2) or in
samples of mixed spermatogenic cell membranes (Fig. 3c). Stage-specific expression
of p73/5-7 on the surface membranes of pachytene spermatocytes (Fig. 4A) and of
p84/6-3 on the surface membranes of round spermatids (Fig. 4B) was confirmed. In
addition, components p93/5-7 and pl51/6-0 weje the most evident ConA-binding
glycoproteins in the membranes of both cell types, as previously noted.
DISCUSSION
Recent studies indicate that the mammalian testis exhibits unusually active rates of
dolichol synthesis, which could relate to high rates of plasma membrane glycoprotein
turnover, to temporally regulated synthesis of acrosomal enzymes in late pachytene
spermatocytes or early spermatids, and/or to the secretion of specific glycoproteins
by the epididymal epithelium (James & Kandutsch, 1980; Wenstrom & Hamilton,
1980; Potter et al. 1981). The data presented here provide an initial biochemical
characterization of cell surface glycoproteins expressed during the late stages of spermatogenesis in the mouse; information necessary for the future identification of those
particular membrane constituents glycosylated via dolichol intermediates.
Acrosomal enzymes may also represent end products of dolichol-mediated
glycosylation in the testis since many of these constituents are glycoproteins (Fldchon,
1979; Mukerji & Meizel, 1979). The precise stages of spermatogenesis during which
individual acrosomal enzymes are first synthesized and subsequently packaged
intracellularly have not been determined, but recent biochemical studies (Ji, Yoo &
Ji, 1981) suggest that some fucosylated components of the acrosome may be
Fig. 4. Identification of ConA-binding glycoproteins in purified plasma membranes from
isolated pachytene spermatocytes and round spermatids. A. ConA blot of purified spermatocyte plasma membranes. The two major constituents are p93/5-7 and pl51/6-0, as
noted earlier in whole cell homogenates and mixed spermatogenic cell membranes. The
pachytene marker protein p73/5-7 is labelled, but the spermatid marker protein p84/6-3
(bracket) represents a minor constituent whose presence is due to a 5 % contamination of
the pachytene population by round spermatids. B. ConA blot of purified spermatid plasma
membranes. The two major constituents are again p93/5-7 and pl5l/6-0. The spermatid
marker protein p84/6-3 is strongly labelled, but the spermatocyte marker, p73/5-7, is
absent (bracket). Total protein loaded onto the gels was 510/ig for pachytene membranes
and 270 //g for spermatid membranes.
244
C. F. Millette and B. K. Scott
synthesized as early as meiotic prophase. Cytochemical analysis of fucose incorporation (Kopecny & Fle"chon, 1981; Tang, Lalli & Clermont, 1982) and of phosphotungstic acid staining (Courtens, 1978; Holt, 1979; Sinowatz & Wrobel, 1981) indicates
that continued synthesis of acrosomal proteins occurs throughout most of spermiogenesis, including steps 9—17 in the rat when the nuclear chromatin condenses.
ConA-binding acrosomal proteins have not been detected in the current studies utilizing whole cell homogenates from either spermatocytes or early (step 1—8) spermatids,
although other workers have demonstrated that some acrosomal constituents such as
hyaluronidase (Yang & Srivastava, 1975) react with this lectin. The failure to detect
additional ConA-binding proteins in cell homogenates as compared to purified plasma
membranes presumably reflects a low relative concentration of these components or
their synthesis and glycosylation at later stages of spermatid differentiation. Previous
investigations (Millette et al. 1980) have established that acrosomal enzymes do not
significantly contaminate isolated plasma membrane fractions from mouse round
spermatids.
The stage-specific plasma membrane protein markers p73/5-7 (spermatocytes) and
p84/6 - 3 (spermatids) described here represent the first such glycosylated constituents
reported during mouse spermatogenesis. Stage-specific markers identified previously
by this laboratory (Millette & Moulding, 1981a) were not labelled by either ConA or
BSL in either whole cell homogenates or purified plasma membranes. Other membrane components earlier recognized on restricted classes of mammalian spermatogenic cells are not specific to any individual stage of differentiation, nor is it
known whether they are glycosylated. O'Rand and colleagues have described RSA-1,
a membrane glycoprotein from the rabbit testis (RomrelleJ al. 1982), but this protein
first appears on pachytene spermatocytes, is expressed throughout spermiogenesis,
and has been implicated in sperm-egg interaction at fertilization. In a similar manner,
Gaunt (1982) has prepared a monoclonal antibody that recognizes autoantigen 1B3,
a cell surface constituent that first appears at pachynema and is retained on epididymal
spermatozoa. No data concerning the possible glycosylation of 1B3 have been reported. Unlike RSA-1 and 1B3 the newly identified glycoprotein markers p73/5-7 and
p84/6-3 may function primarily within the seminiferous epithelium and not during
sperm maturation or at fertilization. Further characterization of these membrane
proteins will be required before functional assays can be designed.
The lack of [35S]methionine incorporation by either p73/5-7 or p84/6-3 in vitro
need not be particularly noteworthy when the complex cellular differentiative
processes involved in spermatogenesis (Bellve', 1979; Ewing et al. 1980) are considered. It should be remembered that isolated pachytene spermatocytes and round
spermatids are not mitotically active and, in fact, that the last round of DNA synthesis
occurred days earlier in vivo during the preleptotene stage. These cells, therefore, do
not necessarily synthesize all of their constituent RNAs or polypeptides in vitro.
Furthermore, round spermatids are haploid cells and are presumed to synthesize only
a limited number of proteins, some of which may be translated from exceedingly longlived mRNA (Stern, Kleene, Gold & Hecht, 1983). Certainly, the lack of [3SS]methionine incorporation by p73/5-7 or p84/6-3 should not be taken as an a priori
Spermatogenic cell surface glycoproteins
245
indication that these glycoproteins are not components of the plasma membrane.
Purely technical considerations may account for the lack of [35S]methionine in vitro.
Our results may, for example, simply demonstrate a low methionine content present
in P73/5-7 and p84/6-3.
In contrast to p73/5-7 and p84/6'3, functional assays of pl51/6-0, one of the two
most evident ConA-binding proteins identified in the present experiments, may now
be contemplated. Component pIS 1/6-0 has previously been shown to be an important
quantitative constituent of both spermatocyte and spermatid membranes (Millette &
Moulding, 1981a) and to be oriented extracellularly as determined by vectorial
iodination studies (Millette & Moulding, 19816). Very recently, it has also been
determined that pl51/6-0 is synthesized in vitro by isolated pachytene spermatocytes,
but not by isolated round spermatids as determined by the incorporation of [35S]methionine (Gerton & Millette, 1983). All of these data together suggest that pl5l/
6-0 may be involved in the regulation of germ cell-Sertoli cell interactions. Numerous
morphologically defined junctional specializations have been described, between
developing germ cells and Sertoli cells (Russell, 1980). The cellular associations
between rat Sertoli cells and spermatids are sensitive to trypsin (Romrell & Ross,
1979), and could conceivably involve glycoprotein interactions. Recently, Cohen,
Gorbsky & Steinberg (1983) identified glycosylated components of bovine epidermal
desmosomes. These desmosomal core glycoproteins bind ConA and are of Mr 150 000.
Experiments are now in progress to examine possible structural homologies between
pl51/6-0 and the desmosomal glycoproteins.
Grootegoed et al. (19826) have described a differential in vitro adhesiveness of
pachytene spermatocytes and round spermatids to monolayer cultures of mouse Sertoli cells. Palombi et al. (1980) have also reported a germ cell-Sertoli cell co-culture
system, which maintains germ cell viability for longer periods than possible using
isolated single cell populations from the testis. Using these or similar in vitro bioassay
systems and appropriate monoclonal antibodies directed against pl51/6 - 0, it should
be feasible to examine directly the role of this plasma membrane glycoprotein in germ
cell-Sertoli cell interaction. This experimental approach may be logically extended
to other defined cell surface components, including p73/5-7 and p84/6-3, and should
allow the analysis of spermatogenic regulation within the seminiferous epithelium at
the level of the plasma membrane.
We thank Mr Steven Borack for excellent photographic assistance, Mrs Barbara Lewis for
preparation of the manuscript and Dr George L. Gerton for assistance with them vitro experiments.
Animals used in this study were maintained in accordance with the guidelines of the Committee
on Animals of the Harvard Medical School and those prepared by the Committee on Care and Use
of Laboratory Animals of the Institute of Laboratory Animal Resources, National Research Council
(DHEW publication no. (NIH) 78-23, revised 1978).
REFERENCES
ACKERMAN, G. A. & FREEMAN, W. H. (1979). Membrane differentiation of developing hemic cells
of the bone marrow demonstrated by changes in concanavalin A surface labeling. J. Histochem.
Cytochem. 27, 1413-1423.
246
C. F. Millette and B. K. Scott
ATKINSON, P. H. & HAKIMI, J. (1980). Alterations in glycoproteins of the cell surface. In The
Biochemistry of Glycoproteins and Proteoglycans (ed. W. J. Lennarz), pp. 191-239. New York:
Plenum Press.
BAUES, R. J. & GRAY, G. R. (1977). Lectin purification on affinity columns containing reductively
aminated disaccharides. J. biol. Chem. 252, 57—60.
BELLVE, A. R. (1979). The molecular biology of mammalian spermatogenesis. In Oxford Reviews
in Reproductive Biology, vol. 1 (ed. C. A. Finn), pp. 159-261. Oxford: Clarendon Press.
BELLVE, A. R., MILLETTE, C. F., BHATNAGAR, Y. M. & O'BRIEN, D. A. (1977). Dissociation of
the mouse testis and characterization of isolated spermatogenic cells. J. Histochem. Cytochem. 25,
480-494.
COHEN, S. M., GORBSKY, G. & STEINBERG, M. S. (1983). Immunochemical characterization of
related families of glycoproteins in desmosomes. J. biol. Chem. 258, 2621-2627.
COURTENS, J. L. (1978). Cytochemical localization of glycoproteins in the developing acrosome of
ram spermatids.J. Ultrastruct. Res. 65, 173-181.
DELMOTTE, F. M. & GOLDSTEIN, I. J. (1980). Improved procedures for purification of the Bandeiraea simplicifolia I isolectins and Bandeiraea simplidfolia II lectin by affinity
chromatography. Eur.J. Biochem. 112, 219-223.
EDELMAN, G. M. (1983). Cell adhesion molecules. Science, N.Y. 219, 450-457.
EWING, L. L., DAVID, J. C. & ZIRXIN, B. R. (1980). Regulation of testicular function: A spatial
and temporal view. In Int. Rev. Physiol. vol. 22 (ed. R. O. Greep), pp. 41-115. Baltimore:
University Park Press.
FEHLMAN, M., LAUFLER, L. & MARCEU, N. (1976). Surface membrane differentiation of
hemopoietic cells as observed by radioactive labeling..?, cell. Physiol. 90, 455-464.
FLSCHON, J.-E. (1979). Sperm glycoproteins of the boar, bull, rabbit and ram. I. Acrosomal
glycoproteins. Gamete Res. 2, 43-51.
GAUNT, S. (1982). A 28K-dalton cell surface autoantigen of spermatogenesis: characterization
using a monoclonal antibody. Devi Biol. 89, 92-100.
GERTON, G. L. & MILLETTE, C. F. (1983). In vitro synthesis of plasma membrane proteins by
isolated mouse pachytenc spermatocytes and round spermatids. Fedn Proc. FednAm. Socs exp.
Biol. 42, 2051.
GLASER, L. (1980). From cell adhesion to growth control. In The Cell Surface: Mediator of
Developmental Processes (ed. S. Subtelny & N. K. Wessels), pp. 79-97. New York: Academic
Press.
GREIG, R. G. & JONES, M. N. (1977). Mechanisms of intercellular adhesion. Biosystems 9, 43-55.
GROOTEGOED, J. A., JUTTE, N. H. P. M., ROMMERTS, F. F. G., VAN DER MOLEN, H. J. & OHNO,
S. (1982a). Concanavalin A-induced attachment of spermatogenic cells to Sertoli cells in vitro.
Expl Cell Res. 139, 472-475.
GROOTEGOED, J. A., KRUGER-SEWNARGIN, B. C , JUTTE, N. H. P. M., ROMMERTS, F. F. G. &
VAN DER MOLEN, H. J. (19826). Fucosylation of glycoproteins in rat spermatocytes and spermatids. Gamete Res. 5, 303-315.
HAWKES, R. (1982). Identification of concanavalin A-binding proteins after sodium dodecyl sulfategel electrophoresis and protein blotting. Analyt. Biochem. 123, 143-146.
HOLT, W. V. (1979). Development and maturation of the mammalian acrosome. A cytochemical
study using phosphotungstic acid staining. J. Ultrastruct. Res. 68, 58—71.
HUBBARD, S. C. & IVATT, R. J. (1981). Synthesis and processing of asparagine-linked oligosaccharides. A. Rev. Biochem. 50, 555-583.
IVATT, R. J., PREM DAS, O., HENDERSON, E. J. & ROBBINS, P. W. (1981). Developmental
regulation of glycoprotein biosynthesis inDictyostelium.jf. supramolec. Struct. CellBiochem. 17,
359-368.
JAMES, M. J. & KANDUTSCH, A. A. (1980). Elevated dolichol synthesis in mouse testes during
spermatogenesis. J. biol. Chem. 255, 16-19.
Ji, I., Yoo, B. Y. & Ji, T . H. (1981). Surface proteins and glycoproteins of ejaculated bovine
spermatozoa. II. Molecular composition of the midpiece and mainpiece. Biol. Reprod. 24,
627-636.
JUTTE, N. H. P. M., GROOTEGOED, J. A., ROMMERTS, F. F. G. & VAN DER MOLEN, H. J. (1981).
Exogenous lactate is essential for metabolic activities in isolated rat spermatocytes and spermatids. J. Reprod. Fertil. 62, 399-405.
Spermatogenic cell surface glycoproteins
247
KOPECNY, V. & FL£CHON, J.-E. (1981). Fate of acrosomal glycoproteins during the acrosomal
reaction and fertilization: A light and electron microscope autoradiographic study. Biol. Reprvd.
24, 201-216.
LEGOCKI, R. P. & VERMA, D. P. S. (1981). Multiple immunoreplica technique: Screening for
specific proteins with a series of different antibodies using one polyacrylamide gel. Analyt.
Biochem. I l l , 385-392.
MARINO, M., COSSU, G., NERI, G. & MOLINARO, M. (1980). Appearance of a class of cell-surface
fucosyl-glycopeptides in differentiated muscle cells in culture. Devi Biol. 78, 258-267.
MILLETTE, C. F. & BELLVE\ A. R. (1977). Temporal expression of membrane antigens during
mouse spermatogenesis. J. Cell Biol. 74, 86-97.
MILLETTE, C. F. & BELLV£, A. R. (1980). Selective partitioning of plasma membrane antigens
during mouse spermatogenesis. Devi Biol. 79, 309—324.
MILLETTE, C. F. & MOULDING, C. T. (1981a). Cell surface marker proteins during mouse spermatogenesis: Two-dimensional electrophoretic analysis. J. Cell Sci. 48, 367-382.
MILLETTE, C. F. & MOULDING, C. T. (19816). Radio-iodination of plasma membrane polypeptides
from isolated mouse spermatogenic cells. Gamete Res. 4, 317-331.
MILLETTE, C. F., O'BRIEN, D. A. & MOULDING, C. T. (1980). Isolation of plasma membranes
from purified mouse spermatogenic cells. J. Cell Sci. 43, 279-299.
MITA, M. & HALL, P. F. (1982). Metabolism of round spermatids from rats: Lactate as the
preferred substrate. Biol. Reprod. 26, 445—455.
MORRISSEY, J. H. (1981). Silver stain for proteins in polyacrylamide gels: A modified procedure
with enhanced uniform sensitivity. Analyt. Biochem. 117, 307-310.
MUKERJI, S. K. & MEIZEL, S. (1979). Rabbit testis proacrosin. Purification, molecular weight
estimation and amino acid and carbohydrate composition of the molecule. .7. biol. Chem. 254,
11721-11728.
MURAMATSU, T., GACHELIN, G., NICOLAS, J. F., CONDAMINE, H. & JAKOB, H. (1978). Carbo-
hydrate structure and cell differentiation: Unique properties of fucOsyl-glycopeptides isolated
from embryonal carcinoma cells. Proc. nattt. Acad. Sci. U.SA. 75, 2315-2319.
NARAVAN, P., SCOTT, B. K., MILLETTE, C. F. & DEWOLF, W. C. (1983). Human spermatogenic
cell marker proteins detected by two-dimensional electrophoresis. Gamete Res. 7, 227-239.
NEUFELD, E. F. & ASHWELL, G. (1980). Carbohydrate recognition systems for receptor-mediated
pinocytosis. In The Biochemistry of Glycoproteins and Proteogly cans (ed. W. J. Lennarz), pp.
241-266. New York: Plenum Press.
O'FARRELL, P. H. (1975). High resolution, two-dimensional electrophoresis of proteins. J. biol.
Chem.ZSO, 4007-4021.
OLSON, G. E. & HAMILTON, D. W. (1978). Characterization of the surface glycoproteins of rat
spermatozoa. Biol. Reprod. 19, 26-35.
O'RAND, M. G. (1981). Inhibition of fertility and sperm-zona binding by antiserum to the rabbit
sperm membrane autoantigen RSA-1. Biol. Reprod. 25, 621-628.
PALOMBI, F., ZIPARO, E., GALDIERI, M., RUSSO, M. A. & STEFANINI, M. (1980). Ultrastructural
evidence for germ cells-Sertoli cells interaction in cultures of rat seminiferous epithelium. Biol.
Cell 39, 249-252.
POTTER, J. E. R., MILLETTE, C. F., JAMES, M. J. & KANDUTSCH, A. A. (1981). Elevated choles-
terol and dolichol synthesis in mouse pachytene spermatocytes. J. biol. Chem. 256, 7150-7154.
ROMRELL, L. J., BELLV£, A. R. & FAWCETT, D. W. (1976). Separation of mouse spermatogenic
cells by sedimentation velocity. A morphological characterization. Devi Biol. 49, 119-131.
ROMRELL, L. J., O'RAND, M. J., SANDOW, P. R. & PORTER, J. P. (1982). Identification of surface
autoantigens which appear during spermatogenesis. Gamete Res. 5, 35-48.
ROMRELL, L. J. & Ross, M. H. (1979). Characterization of Sertoli cell-germ cell junctional
specializations in dissociated testicular cells. Anat. Rec. 193, 23-42.
RUPAR, C. A. & CARROLL, K. K. (1978). Occurrence of dolichol in human tissues. LJpids 13,
291-293.
RUSSELL, L. D. (1980). Sertoli-germ cell interrelations: A review. Gamete Res. 3, 179-202.
SHEPARD, R. W., MILLETTE, C. F. & DEWOLF, W. C. (1981). Enrichment of primary pachytene
spermatocytes from the human testis. Gamete Res. 4, 487-498.
SINOWATZ, F. & WROBEL, K.-H. (1981). Development of the bovine acrosome. An ultrastructural
and cytochemical study. Cell Tiss. Res. 219, 511-524.
248
C. F. Millette and B. K. Scott
STERN, L., KLEENE, K. C , GOLD, B. & HECHT, N. (1983). Gene expression during mammalian
spermatogenesis. III. Changes in populations of mRNA during spermiogenesis. Expl Cell Res.
143, 247-255.
TANG, X. M., LALLI, M. F. & CLERMONT, Y. (1982). A cytochemical study of the Golgi apparatus
of the spermatid during spermiogenesis in the rat. Ain.J. Anat. 163, 283-294.
TOWBIN, H., STAEHELIN, T. & GORDON, J. (1979). Electrophoretic transfer of proteins from
polyacrylamide gels to nitrocellulose sheets: Procedure and some applications. Proc. natn. Acad.
Set. U.SA. 76, 4350-4354.
TUNG, K. S. K., HAN, L. B. & EVAN, A. P. (1979). Differentiation autoantigen of testicular cells
and spermatozoa in the guinea pig. Devi Biol. 68, 224-238.
TUNG, P. S. & FRITZ, I. B. (1978). Specific surface antigens on rat pachytene spermatocytes and
successive classes of germinal cells. Devi Biol. 64, 297-315.
WAECHTER, C. J. & LENNARZ, W. J. (1976). The role of polyprenol-linked sugars in glycoprotein
synthesis./!. Rev. Biochem. 45, 95-113.
WALSH, F. S. & PHILLIPS, E. (1981). Specific changes in cellular glycoproteins and surface proteins
during myogenesis in clonal muscle cells. Devi Biol. 81, 229-237.
WENSTROM, J. C. & HAMILTON, D. W. (1980). Dolichol concentration and biosynthesis in rat testis
and epididymis. Biol. Reprod. 23, 1054-1059.
WEST, C. M., MCMAHON, D. & MOLDAY, R. S. (1978). Identification of glycoproteins, using
lectins as probes, in plasma membranes from Dictyostelium discoideum and human erythrocytes.
J. biol. Chem. 253, 1717-1724.
WRAY, W., BOULIKAS, T., WRAY, V. P. & HANCOCK, R. (1981). Silver staining of proteins in
polyacrylamide gels. Analyt. Biochem. 118, 197-203.
YANG, C. H. & SRIVASTAVA, P. N. (1975). Purification of bull sperm hyaluronidase by concanavalin A affinity chromatography. Biochim. biophys. Ada 391, 382-387.
{Received 5 July 1983 -Accepted 8 August 1983)