6 mouse

Journal of Reproductive Immunology, 15 (1989) ! 23-- 139 Elsevier Scientific Publishers Ireland Ltd.

123

JRI 00597

A testis-specific antigen of the C57BL/6 mouse

Osamu Mikami, Hiromi Sakamoto, Yoshihiro Yamamoto and

Jun-ichi Furuyama Department of Genetics, Hyogo Collegeof Medicine, Mukogawa-cho, Nishinomiya, Hyogo 663 (Japan) (Accepted for publication 6 March 1989)

Summary Spleen cells from female mice of the C57BL/6 strain iso-immunized with an homogenate from 3-week-old mice testis were fused with P3UI ceils. After cloning, two hybridomas, producing IgM-, were obtained: Tissue specificity of the monoclonal antibody (Moab) in ascitic fluid was investigated by indirect immunofluorescence. Moab 1A1 reacted specifically with the cytoplasm of spermatogenia, spermatocytes and spermatids, but not with spermatozoa. Testicular antigen recognized by Moab IA1 (MoablA1-TA) was prepared by tissue sonication and then subjected to gel filtration. MoablA1-TA detected in the void volume by ELISA was analyzed by SDS-PAGE and Western blotting. Immuno-staining of membrane filters revealed a broad area within the 45--205 kD range. MoablA1-TA was treated with proteolytic enzymes, but no changes were observed after Western blotting. Thus, MoablA1-TA was further digested by peptidases and glycolytic enzymes, electrophoresed using cellulose acetate membranes and immuno-stained with Moab 1A1. Evidence obtained from these experiments strongly suggests that MoablA1TA consists of an acidic peptide and a carbohydrate molecule. The antigenicity would be included in the carbohydrate epitope. Moreover, partial digestion of MoablA1-TA by keratanase indicates that the lacto-series structure is included in the antigenic carbohydrate moiety. Key words: Monoclonal antibodies; Mouse testis; Antigen; Cytoplasm. 0165-0378/89/$03.50 © 1989 Elsevier Scientific Publishers Ireland Ltd. Published and Printed in Ireland

124

Introduction

Mouse testis-specific antigens and their antigenic determinants have been the subject of investigations using monoclonal antibodies against testicular cells and sperm. Bechtol et al. (1979) were able-to obtain two monoclonal antibodies, XT-I and XT-II, which reacted specifically with mouse testicular cells, after iso-immunization of female mice of the 129 strain with testicular cells from 4-week-old male mice. Gaunt (1982) iso-immunized female mice of the C3H strain with sperm from the ductus deferens and obtained monoclonal antibody 1B3, which showed specificity for the surface membrane of testicular cells and sperm. Fenderson et al. (1984) reported the iso-immunization of C57BL/6 male mice with testicular cells and production of the monoclonal antibody Jl, which, except for spermatogonia, reacted specifically with the surface membrane of testicular cells and sperm. Solter and Knowles (1978) obtained a monoclonal antibody to the stage-specific mouse embryonal antigen 1 (SSEA-1) by iso-immunization of Balb/c mice with F9 embryonal carcinoma cells. This SSEA-1 is composed of giycoproteins and glycolipids, with the epitope being included in the carbohydrate moiety (Muramatsu et al., 1979; Gooi et al., 1981; Kannagi et al., 1982a, 1982b). In the experiments described here, iso-immunization of C57BL/6 female mice with a testis homogenate from 3-week-old males was performed and a testis-specific monoclonal antibody was obtained. The distinct properties of this antigen were investigated and are discussed. Materials and methods

Animals C57BL/6 mice were purchased from Charles River, Japan and CBF~ mice (a C57BL/6 and Balb/c hybrid) from the Shizuoka Laboratory Animal Center.

Iso-immunization procedure Testes from 3-week-old C57BL/6 strain male mice were homogenized by 30 strokes in a Dounce tissue grinder tight type B (Kontes Glass Co. or Wheaton Scientific). The homogenate was diluted 50% (v/v) in physicological saline, mixed with Freund's complete adjuvant (Difco) and used for immunization, by intraperitoneal injection (0.2 ml), of 5 to 7-week-old C57BL/6 female mice. At the same time, mice were injected subcutaneously with 0.2 ml of the 50°70 testis homogenate. A second immunization was carried out after 2 weeks and 4 weeks later a booster injection of 0.2 mi of 5 0 e testis homogenate was given intraperitoneally.

125

Cell fusion and cloning Three days after the booster injection, the immune splenic lymphocytes were fused with P3-X63-AgS-U1 (P3U1) cells CYelton et al., 1978), at a ratio of 10 : 1, in the presence of 50070 polyethylene glycolNo. 4000. Cells were incubated at 37°C and 5070 CO 2 in RPMI 1640 (GIBCO) containing HAT (Littlefield, 1964) and 15070 (v/v) fetal calf serum (FCS: FLOW or GIBCO). Cells growing in Limbro plates were screened for production of antibodies by a modified enzyme-linked immunosorbent assay (ELISA) (Engvall et al., 1971; Engvall and Perlmann, 1971) 2 weeks after culture was started. Antigen for ELISA was prepared as follows: testis homogenate from 3-week-old C57BL/6 mice (10---15 animals) was diluted with approximately 2 ml of distilled water and subjected to 5 cycles of freezing and thawing. It was then washed twice in 10 mM Tris--HC1 buffer, pH 7.5; the precipitate was resuspended in 1 ml of the same buffer and treated with 200/~g of DNase I (Sigma) for 30 rain at 37°C. After washing the precipitate three times with distilled water (DW), it was put through a stainless steel mesh to remove capsules and connective tissue debris and then fLxed in 10070 (v/v) neutral formalin for 30 min at 4°C, washed with DW, dispensed on 2m3 96-well microtitrationplates (Sanko) and allowed to dry before use. Dulbecco's phosphate buffer saline, pH 7.4 (PBS), containing 0.1 07o(v/v) Tween 40 in PBS (0.1 070 T-PBS), was used as washing and diluting buffer. The Ag-Ab reaction times were 1 h at room temperature (RT) with twice diluted cultured media and with peroxidase-conjugated rabbit anti-mouse immunoglobulins (DAKO) diluted 100 times. The coloring substrate [0.04070 (w/v) o-phenylenediamine, 0.01070 (v/v) H202, 0.05 M citrate/phosphate buffer, pH 5.0] was allowed to react for 30 rain. Results of the ELISA were judged by visual examination, cells in those wells showing a positive reaction were resuspended in RPMI 1640 containing HT and 15070 FCS and cloned three times using the limiting dilution method. Before cloning cells were screened by ELISA.

Determination of immunoglobulin class of monoclonal antibodies The subclass of the antibodies released by cloned cells was identified by the Ouchterlony method, using sheep anti-mouse IgM, IgA, IgG2a, IgG2b and IgG3 (Serotec) as reference standards.

Production of monoclonal antibodies from ascitic fluid CB F~ female mice were injected intraperitoneally with cloned hybridomas (3.0 x 10~--2.0 x 10v cells) for mass production of antibodies; the mice were injected intraperitoneally with 0.5 ml of pristane at least 7 days in advance. Ascitic fluid was drawn 2--3 weeks after injection and centrifuged against a 50070 saturation ammonium sulfate gradient to obtain the immu-

126

noglobulin fraction; the protein content was adjusted to 1--10 mg/ml and stored at - 80°C until use. The negative control was obtained from normal C57BL/6 mouse serum purified in the same manner. Monoclonal antibodies obtained from clones OM407-1A1 and OM410-6B1 are referred to as Moab 1AI and Moab 6B1, respectively.

Tissue specificity of monoclonal antibodies determined by indirect immunofluorescence Three- and 8-week-old mice were subjected to open thorax operation under ether anaesthesia. After 10 min irrigation with PBS from left ventricle to right ventricle they were irrigated with fixation solution (4070 (w/v) paraformaldehyde, 0.1 M phosphate buffer, pH 7.2) for 30 min. Then each organ was excised and washed for 3 days before preparation of frozen sections. Sperm preparations were made from sperm collected from the epididymides. Sperm were washed twice with 107o BSA-PBS (PBS containing 1070 (w/v) bovine serum albumin fraction V (BSA : sigma)), smeared over glass slides, fixed with fixation solution and washed with PBS. Indirect immunofluorescence staining was carried out in the usual manner with some modifications. Slides were immersed in PBS for 30 min, then overlaid with 10070 (v/v) normal rabbit serum in PBS and placed in a humidified chamber for 30 min. Moab 1A1 and Moab 6B1 (20/ag/ml) as well as negative control immunoglobulin (200/~g/ml) were diluted in 0.1 070 BSAPBS. The FITC-conjugated rabbit anti-mouse IgM (Litton) was diluted 20 times in the same buffer.

Solubilization and extraction of testicular antigen Testes from 10 C57BL/6 mice, 3-weeks-old, were excised, the surrounding fatty tissue was removed (wet weight approx. 0.7--0.8 g/10 animals) and pippetted to loosen cells in the seminiferous tubes. The testes were then washed three times in 10 mM Tris--HC1 buffer, pH 7.4, containing 0.25 M saccharose, 0.1 mM EDTA resuspended in 1.5 ml of the same buffer and homogenized by 20 strokes in a Dounce tissue grinder tight type B. The homogenate was centrifuged at 50 × g for 7 min to remove cell debris and connective tissue. The supernatant was carefully laid over an equal volume of 10 mM Tris--HCl buffer, pH 7.4, containing 0.34 M saccharose, 0.1 mM EDTA and centrifuged at 700 x g for 10 rain to remove the nuclei. The supernatant was treated in two separate ways: A, Penylmethyl sulfonyl fluoride (PMSF : Sigma) was added to a final concentration of 1 mM and then ultra-centrifuged at 105,000 × g for 90 min to remove insoluble components. NaC1 (5 M) aqueous solution was added to a final concentration of 0.5 M and centrifuged once more; B, NaCl (5 M) aqueous solution was added to a final concentration of 0.5 M, and sonicated at 20

127

kHz for 20 s in an ultrasonic sonicator 20 times. After solubilization, PMSF was added to a final concentration of 1 m M and then ultra-centrifuged to remove insoluble components. Solid a m m o n i u m sulfate was added to the supernatant and solubilized to reach 51 to 90070 saturation. Both the cytosol fraction (A) and a m m o n i u m precipitate (B) were recovered and dialyzed against 2 m M Tris--HC1 buffer, pH 7.4, containing 0.1 m M E D T A in a cellulose dialysis tubing, mol wt. cutoff 1000 (Spectrum), to remove salts, after which it was lyophilized. Unless stated, all experiments were carried out at 0---4°C.

Column chromatography For column chromatography (gel filtration) a Sephacryl S-200 Superfine (Pharmacia) chromatography tube, 9 m m diameter and 95 cm in length, was used. Tris--HC1 buffer (10 mM) pH 7.4, containing 0.1 M NaCI, 0.1 mM E D T A was employed to stabilize the column and also as elution buffer. The lyophilized material was solubilized in a small amount of the same buffer, overlaid and eluted at 1.9 m l / h and at 4°C using a peristaltic pump. The eluted solution was collected in a fraction collector in 1-ml aliquots. The optic absorption of each fraction was measured at 280 nm in a spectrophotometer and all fractions were reacted with Moab 1A1 (5/~g/ml) and negative control immunoglobulin (50/zg/ml) using the ELISA. As secondary antibody, 200 times diluted peroxidase-conjugated rabbit anti-mouse IgM was employed. The results of the reaction were measured in an Immuno-Reader at 490 nm. Those fractions which showed a strong positive reaction with Mab 1A1 were lyophilized for use in further experiments.

Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and immunoblotting After column chromatography, the sample material was solubilized in 200--400 ~1 of 10 m M T r i s - - H C l buffer, pH 7.4, dispensed into Eppendorf tubes in 35-/A aliquots and treated with proteolytic enzymes. Trypsin type I (from bovine pancreas: Sigma), 50/~g/5/~I and protease (from Streptomyses: Sigma), 50/~g/5 ~1, both solubilized in 10 mM Tris--HCl buffer, pH 7.4, were added to separate tubes. Control tubes contained 5/A of the buffer. Reaction mixtures were incubated at 3 7 o c for 2 h then 10/A of 300 mM Tris --HC1 buffer, p H 6.8, containing 10070 (w/v) SDS (Pierce), 2507o (v/v) mercaptoethanol, 25070 (v/v) glycerol and 0.002°70 (w/v) bromophenol blue were added to each tube and heated for 1 min at 100°C. Of each sample, 20/~i were used for SDS-PAGE. The gel employed was a 4--20070 gradient TEFCO-GEL (TEFCO), 1.5 m m thick and the electrophoresis was carried out at 25 mA. Molecular weight markers (Pharmacia) were electrophoresed at the same time.

128

Electrophoresed samples were transferred to a membrane for proteins, Durapore GVHP filter (Millipore), using a blotting apparatus, according to the Western blot technique (Towbin et al., 1979). Transfer was carried out at 20V for 12h. The membrane was washed in PBS, then the proteins which had not adhered to the membrane were blocked by incubating the membrane for 1 h in 1070BSA-PBS. The membrane was washed with 0.05070 (v/v) Tween 40 in PBS (0.05070 T-PBS) and incubated for 30 min in 0.0507o T-PBS containing 10070 (v/v) normal rabbit serum. T-PBS 0.05o70 was used as washing and diluting buffer. Incubation times were 2 h for either Moab 1A1 (3/~g/ml), negative control immunoglobulin (30 ~g/ml) or secondary antibody ( x 1/ 300), after which it was allowed to react for 30 min with the coloring substrate (0.05070 (w/v) 3-3' diaminobenzidine-4 HCI, 0.0107o (v/v) H202, 0.05 M Tris--HC1 buffer, pH 7.6).

Cellulose acetate membrane electrophoresis and immuno-staining (A) Lyophilized material was solubilized in 100--150 bd DW and treated with enzymes as follows: in an Eppendorf tube, 50 ~1 of this solution were allowed to react with 50 ~g of protease (solubilized in 10 ~1 of 10 mM Tris-HCI buffer, pH 7.4) for 24 h at 37°C; the reaction mixture was then placed at 100°C for 5 rain to stop the reaction. The protease treated material was dispensed into Eppendorf tubes in 10 ~1 aliquots and treated with keratanase (from Pseudomonas sp.), leucine aminopeptidase (from porcine kidney microsomes) or carboxy peptidase Y (from baker's yeast) for 24 h at 37°C; 0.25 units of each enyzme in 5/~1 of 10 mM Tris--HC1 buffer, pH 7.4, were used. All the enzymes were purchased from Sigma. (B) Lyophilized material was solubilized in 100--150 ~1 of 5 mM sodium acetate/acetate buffer, pH 4.5, dispensed into 10 ~1 aliquots and treated with 0.1 units of neuraminidase type X (from Ciostridium perfringens: Sigma) solubilized in 5 ~1 of sodium acetate/acetate buffer or with 50 ~g of mixed glycosidase (from Turbo cornutus: Seikagaku Kogyo Co.) solubilized in the same buffer. Reaction mixtures were allowed to react for 24 h at 37 °C. Samples (2--3 ~1) of treated material were spotted onto cellulose acetate membrane (Sartorius), previously soaked in 1 M acetate/pyridine buffer, pH 3.5, and electrophoresed at 0.5 m A / c m for 20 min. Toluidine blue 0 (Merk), 0.5070 (w/v) aqueous solution was used as staining solution. The membrane was first allowed to dry at RT, then soaked for a few seconds in 107o BSAPBS and fixed in 20070 (v/v) neutral formalin for 1 h at RT. After fixation the membrane was washed three times in PBS for 10 min and immunostained in the manner described above for membrane filters subjected to Western blotting.

129

Results

Production o f monoclonal antibodies to testicular cells Production of hybridomas was attempted four times, resulting in two positive clones: OM407-1AI and OM410-6B1. These clones, injected intraperitoneally into female mice for mass production of antibodies, resulted in two monoclonal antibodies Moab 1AI and Moab 6BI, both IgM type. Determination o f tissue specificity o f monoclonal antibodies by indirect immunofluorescence The results of indirect immuno-staining of frozen sections from ovaries, brain, liver, kidneys, adrenal glands, skeletal muscle and testes, as well as smeared preparations of sperm, from C57BL/6 mice with Moab 1A1 and Moab 6B1 are shown in Table I. With Moab IA1, specific staining was observed for testis but not for any other tissue. As shown in Fig. 1A,C specific staining was demonstrated in the cytoplasm but not in the nucleus of spermatogonia, spermatocytes and spermatids within seminiferous tubules. The cell membrane was apparently not stained. Moreover, using only the immunofluorescence method, staining of Sertoli cells or Leydig cells could not be demonstrated. No specific staining was observed in sperm from 8week-old mice treated with Moab IA1. When Moab 6B1 was used, specificity for testicular tissue could not be demonstrated (Table 1). Accordingly, the other experimer,ts were designed to extract the testicular antigen reacting with Moab 1A1 (Moab 1AI-TA). Solubilization and extraction o f Moab lA 1- TA As Moab 1A1 showed a strong specific reaction with the cytoplasm but not with the nucleus, preliminary experiments with this antibody were carded out allowing reaction with the cytosol fraction of testicular cells; however, no binding was detected (Fig. 2A). M o a b l A I - T A was therefore solubilized, as described above. A strong positive reaction was demonstrated by ELISA in the precipitate obtained by salt precipitation (51--90070 saturation). Column chromatography o f solubilized Moab lA 1- TA Solubilized MoablA1-TA was salt precipitated and concentrated by lyophilization before fractionation by column chromatography. A strong positive reaction was demonstrated in the void volume fraction (Fig. 2B). SDS-PA GE and Western blotting The molecular weight was estimated by immuno-staining after SDSPAGE and Western blotting of the purified M o a b l A I - T A . As shown in Fig.

.

-

IAI

6BI

.

.

-

-

Liver

.

.

-+

+

.

.

Kidney

.

.

+ Positive; :1: undetermined; - negative; *epididymal and seminiferous sperm.

8

.

-

1AI

6BI

3

Brain

Moab

Age of mice (weeks)

-

-

Adrenal gland

+-

-+

Skeletal muscle

TISSUE SPECIFICITY OF M O N O C L O N A L ANTIBODIES BY I N D I R E C T IMMUNOFLUORESCENCE

TABLE 1

-+ -+ -+ -+

Ovary

+ -+

+

+

Testis

Sperm*

ho

B

"ig. 1. Indirect immunofluorescence labelling of the cryostat-sectioned testis from 3-week-old C57BL/6 mouse using IAI monocional antibody (Moab IA1). A) Moab 1AI labels spermatogonia, spermatocytes and spermatids, within seminiferous tubules ( x 400). (B) Negative control (× 400) [see text]. (C) ,loablAl labels the cytoplasm of spermatogonium, spermatocyte and spermatid. Nuclei are not labeled ( × 1000).

A

132

AO.IO

0.6

-

E 0.05

< e~

ej O.OC 0

I 10

I

|

20

30

i

~ 10

i

0.3

!

E < i

50

I" I

60

I

70

0.0 80

FRACTION NUMBER

B o.2

2.0

~ 0,1

1.0

I I

t

< < 0.0 10

...." ~ - i ..... 20 30

ff

i- -'~ 40

50

60

--i 70

0.0 80

FRACTION NUMBER " A280nm,

----:

A490nmMoab

IAI,

.--.

: A4~lnmNegative

control

Fig. 2. Column chromatography of the testicul~ antigen. None of the fractions in the cytosol reacted with Moab IAI or with the negative control immunolOobulin (A). However, fractionation of the desalted components precipitating between 51 and 9 0 ~ saturation of ammonium sulfate revealed that the fraction showing the stronsest reaction to Moab IA! was consistent with the void volume 03) (Sephacryl S-200 9 mm X 95 cm).

3c, an unusually broad stained area going from 45--205 kD was observed. Treatment with protease or trypsin did not affect the eleetrophoresis pattern (Fig. 3a',b'). No specific band was observed with the negative control immunoglobulin (data not shown). The protein moiety in the eleetrophoresed material was too small to be determined. Moreover, the eleetrophoresis gel was directly stained with Coomassie Blue, but no clear band was observed (data not shown).

Cellulose acetate membrane electrophoresis and immuno-staining M o a b l A I - T A obtained by column chromatography was electrophoresed

133

205,000

=~,:!'~i~!~i~:'~; i!/~ii~i=i'~ ~ =~,~i :

~.

;:~:~':-~:

:~ ::

~:: ~

~.i!:~,~.~~,=,:~~: ~:~.,. :~.....

:i~: ~

:.;

.i:i;~,=:::.~:C. i:ii: " ::.~: :;ii

.:~.

~L~Jg:i:~::

:

::

116,000

.., ~i:

;:,)~:.: i..i.~:~:

45,0O0

i :i-/ii

i: !i" /::;:-~i: i i i . :

i.:::.:::.;i:: i)iii:

~ : ; : :i: i:=~: :

:i! i.::~

:

Fig. 3. SDS-PAGE of the void volume fraction (Fig. 2B) followed by immunolabeling of Western blots. Antigen was detected in a broad area within 45,000 and 205,000 daltons range (c). However, no differences in immunolabeling affinities or electrophoretic mobilities followed by the digestion of proteolytic enzymes could be detected (a',b'). a: Protease; a': sample treated with protease; b: trypsin; b': sample treated with trypsin; c: sample.

in cellulose acetate membrane. After staining with Toluidine blue 0, two spots were observed (Fig. 4Aa). In the immuno-stained membrane, only the spot showing less mobility was detected (Fig. 4Aa'). After treatment with protease only one spot was observed in both membranes (Fig. 4Ab,b'). Moreover, treatment with leucine amino-peptidase after digestion with protease resulted in decomposition o f MoablA1-TA, which showed very little

a

.,

b

.

c

:

d

i

e

~

a'

'

|

b'

.4

c'

d'

~.~.::~

•

.

..•.

e'

:

~::!:i~i •

: #~;::i

ing.

Fig. 4A. Detection of testicular cell antigen by immunostaining after cellulose acetate membrane electrophoresis of the void volume fraction in Fig. 2B. Changes of electrophoretic mobilities following the treatment of protease, peptidase and keratanase could be detected. But no differences in immunostaining ability following digestions with proteolytic enzymes and keratanase were observed, a,a': Sample; b,b': sample treated with protease; c,c': sample treated with protease, heat denatured and then treated with keratanase; d,d': sample treated with protease, heat denatured and then treated with leucine amino-peptidase; e,e': sample treated with protease, heat denatured and then treated with carboxy-peptidase, a--e: Toluidine blue 0 staining; a'--e': Moab I A 1 immunostain-

Origin

: .:.

•

t.o 4~

a

(

•

b

c

d

a'

b"

c'

d'

Fig. 4B. Detection of testicular cell antigen by immunostaining after cellulose acetate membrane electrophoresis of the void volume fraction in Fig. 2B. No changes in electrophoretic mobilities following the treatment of neuraminidase could be detected. Testicular cell antigen disappeared after digestion with mixed glycosidase, a,a': Sample; b,b': sample treated with neuraminidase; c,c': sample treated with mixed glycosidase; d,d': sample treated with neuraminidase and mixed glycosidase, a--d: Toluidine blue 0 staining; a'--d': Moab 1A I immunostaining.

Origin

-I-

136

mobility (Fig. 4Ad,d'). Digestion with carboxy peptidase also resulted in partial decomposition of the MoablA1-TA, showing less mobility towards the anode (Fig. 4Ae,e'). Similar results were obtained when MoablA1-TA was treated with keratanase (Fig. 4Ac,c'). However, in spite of the variation in mobility observed after enzymatic digestion, when investigated by immuno-staining techniques no major changes were observed (Fig. 4Ab',c',d',e'). Furthermore, MoablA1-TA was treated with neuraminidase, after which no variations in either electrophoretic mobility or immuno-staining were detected (Fig. 4Bb,b'). However, treatment with mixed glycosidase resulted in complete digestion of MoablA1-TA, no stained area was observed after staining with Toluidine blue 0 or after immuno-staining (Fig. 4Bc,c',d,d'). No spot was observed with the negative control immunoglobulin (data not shown). Discussion

Moab 1A1 reacted specifically with the cytoplasm of testicular spermatogonia, spermatocytes and spermatids, but not with seminiferous or epididymal sperm. This antibody must therefore be different from the monoclonal antibody 1B3 obtained by Gaunt (1982) after iso-immunization of C3H female mice with sperm from ductus deferens, which reacts specifically with the membrane of testicular cells and sperm. Moab 1A1 is also different from the J1 monoclonai antibody produced by Fenderson et al. (1984) in C57BL/6 male mice after syngeneic immunization with mechanically dissociated testicular cells, which reacts specifically with epididymai spermatozoa but not with spermatogonia. Moreover, Moab 1AI also differs from two monoclonal antibodies, XT-I and XT-II, obtained by Bechtol et al. (1979) after syngeneic immunization of strain 129 female mice with testicular cells from 4-week-old male mice. The tissue specificity of XT-I and XT-II was determined by a quantitative absorption test; XT-I reacted with antigen XT-1 on the surface of spermatocytes and spermatids, whilst XT-1 was present in very small amounts on the surface of epididymai sperm. XT-II, on the other hand, reacted with antigen XT-2 on the surface of spermatogonia, spermatocytes and spermatids, but XT-2 was not detected in epididymal sperm. Thus, although tissue specificity for XT-II and Moab 1AI is similar, they are different in that XT-2 was not clearly detected in the cytoplasm of these cells. Given that MoablA1-TA is not present in the cytosol fraction but within cellular organelles, except the nuclei, or attached to cellular organelles, and that it can be solubilized by sonication, we can infer that MoablA1-TA exists in a solubilized state. Additionally, as it can be detected in the void

137

volume after gel filtration, this antigen must be a very large molecule. Moreover, the results of SDS-PAGE followed by Western blotting and immunostaining, which revealed a broad stained area (45--205 kD), coupled with the fact that proteolytic enzymes did not cause any major changes in the molecular structure of MoablA1-TA, suggest that MoablA1-TA is not a simple protein, but probably having some carbohydrates or mucopolysaccharides, or it may be a complex compound of these elements. Judging from the results of the SDS-PAGE and the electrophoresis conditions it can be concluded that MoablA1-TA is different from the antigen recognized by Gaunt's monoclonal antibody 1B3 of 28 kD (estimated by SDS-PAGE) and from the antigen recognized by J l, reported by Fenderson et al. to be 300 kD and 200 kD (SDS-PAGE). Molecular weights of antigens XT-1 and XT-2 reported by Bechtol et al. were not determined by SDS-PAGE but both of them were digested by trypsin, therefore they are different from MoablA1TA reported here. Electrophoresis in cellulose acetate membrane shows that MoablA1-TA as a whole has the properties of mucopolysaccharides and could be stained with Toluidine blue 0. MoablA1-TA treated with leucine aminopeptidase was, however, digested and migration to the anode was less than that of untreated MoablA1-TA. On the other hand, treatment with neuraminidase did not result in mobility variations. Given that the substrate for neuraminidase is sialic acid, which is responsible for the negative electric charge of carbohydrates, these results suggest that the negative electric charge of MoablA1-TA is given by acidic peptide and not by sialic acid. As mentioned above, immuno-staining was preserved even after digestion with proteolytic enzymes and peptidase, however, MoablA1-TA treated with mixed glycosidase lost its immuno-staining ability, suggesting that the epitope is included in the carbohydrate moiety. MoablA1-TA probably comprises a carbohydrate of high molecular weight containing the epitope and a glycoprotein or glycopeptide consisting of acidic peptide chain. Moreover, a part of this carbohydrate structure possibly includes gaiactose, N-acetyl-glucosamine and the lacto-series structure with its/t-galactoside bonds but not sialic acid. The lacto-series structure is represented in the epitope of the determinants of the ABO(H) and Lewis blood systems (Lloyd and Kabat, 1968). The epitope of SSEA-1, characteristic of undifferentiated cells and early embryonic stages in mice, also includes lacto-series structures (Gooi et al., 1981; Kannagi et al., 1982a, 1982b). T h e antigen recognized by Jl monoclonal antibody (Fenderson et al., 1984) is digested by keratanase (endo-/3-galactosidase) and its epitope has been reported to belong to the facto-series structure (Symington et al., 1984). M o a b l A I - T A is cytoplasmic, its presence on the cell surface has not yet been confirmed. There are similarities between

138

the lacto-series structures of presently known cell surface antigens and the lacto-series structure in the carbohydrate moiety of MoablA1-TA. It is necessary, however, to determine the structure of the carbohydrate moeity in order to be able to establish differences and similarities between them. Cellulose acetate membrane electrophoresis is usually used to separate and identify proteins and mucopolysaccharides (Seno et al., 1970). Adsorption to the membrane is minimal and the molecules are separated on the basis only of their different electrical charge. Thus, if the cellulose acetate membrane were immuno-stained immediately after electrophoresis, either during incubation with the antibody or during washing, molecules on the membrane would detach and be washed away. In the experiments described here, electrophoresed material was fixed with BSA by formalin, which made immuno-staining possible. The physiological role and significance of MoablA1-TA remain to be established and will be the subject of further investigations. References Bechtol, K.B., Brown, S.C. and Kennett, R.H. (1979) Recognition of differentiation antigens of spermatogenesis in the mouse by using antibodies from spleen cell-myeloma hybrids after syngeneic immunization. Proc. Natl. Acad. Sci. USA 76, 363--367. Engvall, E., Jonsson, K. and Perlmann, P. (1971) Enzyme-linked immunosorbent assay. Biochim. Biophys. Acta 251,427--434. Engvall, E. and Perlmann, P. (1971) Enzyme-linked immunosorbent assay (ELISA): Quantitative assay of immunoglobulin G. Immunochemistry 8, 871--874. Fenderson, B.A., O'Brien, D.A., Millette, C.F. and Eddy, E.M. (1984) Stage-specific expression of three cell surface carbohydrate antigens during murine spermatogenesis detected with monoclonal antibodies. Dev. Biol. 103, 117--128. Gaunt, S.J. (1982) A 28K-Dalton cell surface autoantigen of spermatogenesis: Characterization using a monoclonal antibody. Dev. Biol. 89, 92--100. Gooi, H.C., Feizi, T., Kapadia, A., Knowles, B.B., Solter, D. and Evans, M.J. (1981) Stage-specific embryonic antigen involves al-3 fucosylated type 2 blood group chains. Nature 292, 156-- 158. Kannagi, R., Nudelman, E. and Hakomori, S. (1982a) Possible role of ceramide in defining structure and function of membrane glycolipids. Proc. Natl. Acad. Sci. USA 79, 3470--3474. Kannagi, R., Nudelman, E., Levery, S.B. and Hakomori, S. (1982b) A series of human erythrocyte glycosphingolipids reacting to the monoclonal antibody directed to a developmentally regulated antigen, SSEA-I. J. Biol. Chem. 257, 14865--14874. Littlefield, J.W. (1964) Selection of hybrids from matings of fibroblasts in vitro and their presumed recombinants. Science 145,709--710. Lloyd, K.O. and Kabat, E.A. (1968) Immunochemical studies on blood groups, XLI. Proposed structures for the carbohydrate portions of blood group A, B, H, Lewis I and Lewis b substances. Biochemistry 61, 1470-- 1477. Muramatsu, T., Gachelin, G., Damonneville, M., Delarbre, C. and Jacob, F. (1979) Cell surface carbohydrates of embryonai carcinoma cells: Polysaccharidic side chains of F9 antigens and of receptors to two lectins, FBP and PNA. Cell 18, 183--191. Seno, N., Anno, K., Kondo, K., Nagase, S. and Saito, S. (1970) Improved method for electrophoretic separation and rapid quantitation of isomeric chondroitin sulfates on cellulose acetate strips. Anal. Biochem. 37, 197--201. Solter, D. and Knowles, B.B. (1978) Monoclonal antibody defining a stage-specific mouse embryonic antigen (SSEA-I). Proc. Natl. Acad. Sci. USA 75, 5565--5569. Symington, F.W., Fenderson, B.A. and Hakomori, S. (1984) Fine specificity of a monoclonal anti-testicular cell antibody for glycolipids with terminal N-acetyl-D-glucosamine structure. Mol. lmmunol. 21. 877--882.

139 Towbin, H., Staehelin, T. and Gordon, J. (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: Procedure and some applications. Proc. Natl. Acad. Sci. USA 76, 4350--4354. Yelton, D.E., Diamond, B.A., Kwan, S.P. and Scharff, M.D. (1978) Fusion of mouse myeloma and spleen cells. C u r l Top. Microbiol. Immunol. 81, 1--7.