Characterization of a Human Antigen with Sera from Infertile Patients'

BIOLOGY OF REPRODUCTION 50, 1087-1093 (1994)

Characterization of a Human Antigen with Sera from Infertile Patients' ALAN B. DIEKMAN and ERWIN GOLDBERG 2 Departmentof Biochemistry, Molecular Biology and Cell Biology, Northwestern University Evanston, Illinois 60208-3500 ABSTRACT We report the cDNA cloning and subsequent characterization of a novel antigen implicated in antibody-mediated human infertility. This antigen, designated AgX (unknown antigen), was identified originally by screening a human testis Agtll cDNA expression library with infertile patients' sera known to contain anti-sperm antibodies. AgX cDNAs isolated from testis and placenta cDNA libraries (AgX-1 and AgX-2, respectively) differed by a 48-bp deletion in the open-reading frame (ORF). The AgX1 and AgX-2 ORFs encoded putative peptide chains of 505 and 521 amino acids (-55.5 and -57.3 kDa), respectively. The AgX amino acid sequences contained consensus motifs indicative of NTP binding. However, computer homology searches did not identify any significant similarity with known sequences. Quantitative analysis using the reverse transcriptase-polymerase chain reaction (RT-PCR) indicated that the AgX-1 mRNA was fiftyfold more abundant than AgX-2 in the testis, while AgX-2 was more abundant than AgX-1 in somatic tissues. An anti-AgX peptide antiserum identified two AgX isoforms on Western blots of human tissue extracts. An abundant 56-kDa isoform was detected only in testis and sperm. These data suggest that the 56- and 58-kDa isoforms are AgX-1 and AgX-2, respectively. AgX was localized by immunofluorescence to the principal piece of the sperm tail. Therefore, antibodies against an AgX isoform may reduce fertility by affecting sperm function.

INTRODUCTION The correlation of humoral antisperm antibodies with some cases of unexplained infertility in male and female patients suggests a role for these antibodies in blocking fertilization. Approximately 50% of vasectomized men also exhibit antisperm antibodies and upon vasectomy reversal, fertility is reduced among these individuals. Such antibodies may reduce fertility by retarding sperm motility or by blocking gamete interactions [1, 2]. Improved diagnosis and treatment, as well as a more complete understanding of the mechanism behind this antibody-mediated infertility, are dependent on identification of relevant antigens. However, candidate proteins have not been well characterized. Molecular mass ranges for candidate antigens have been defined by immunoprecipitation and immunoblotting using sera from infertile patients, but little additional information was provided by these analyses [3, 4].

sessed. While this method was successful for FA-1, testing the relevancy of every novel sperm protein to immunologic infertility would be impractical. In an alternate approach, a human testis gtll cDNA expression library was screened [9] with a pool of sera from male and female patients presenting at an infertility clinic. The individual sera contained antisperm antibodies as judged by clinical tests for immunobead binding, sperm immobilization, or sperm agglutination (Roberts TK, Shih S-C, Goldberg E, Kay D, Boettcher B; paper presented at the annual meeting of the Australian Society for Immunology; 1990). In the present study, we describe the characterization of AgX, a novel antigen detected by the infertile patients' sera. Two isoforms of AgX were identified, and the possible role of a sperm isoform in immunologic infertility is discussed. MATERIALS AND METHODS

Many sperm-specific proteins (e.g., LDH-C 4) do not react with sera from immunologically infertile patients [5]. Fertilization antigen-1 (FA-1) is an exception. FA-1 was found to be reactive with infertile patients' sera, and anti-FA-1 antibodies affected sperm function in various studies [6-8]. However, FA-1 was identified with monoclonal antibodies prepared against an extract of total human sperm-surface proteins, and then its involvement in infertility was asAccepted December 28, 1993. Received September 23, 1993. 'This work was supported in part by NIH Grant #HD23771 and in part by a subcontract with the Center for Recombinant Gamete Contraceptive Vaccinogens (U54 HD29099). A. Diekman is a predoctoral trainee supported by NICHD Training Grant #T32HD07068. c orrespondence: Erwin Goldberg, Ph.D., Department of Biochemistry, Molecular Biology and Cell Biology, Northwestern University, 2153 Sheridan Road, Evanston, IL 60208-3500. FAX: (708) 467-1380.

Isolation and Analysis of cDNA Human testis and placenta Xgtl 1 cDNA libraries (gifts from Dr. Jose Luis Millan, La Jolla Cancer Research Center, La Jolla, CA [10, 11]) were screened by the method of Benton and Davis [12]. All a-[ 32P]-dCTP-labeled cDNA probes were prepared using the Random Primed Labeling Kit (United States Biochemical, Cleveland, OH). Insert fragments produced by EcoRI (Promega, Madison, WI) restriction of plaquepurified, positive Xgtl 1 clones were isolated and subcloned into the plasmid vector, pBluescript (Stratagene, La Jolla, CA). Nested deletions of insert cDNA were prepared using the Exo/Mung Kit (Stratagene). Complementary DNA clones were sequenced with the Sequenase Kit (Stratagene) using a combination of nested deletions, restriction endonuclease deletions, and specific oligonucleotide primers [13].

1087

1088

DIEKMAN AND GOLDBERG

Nucleotide and putative amino acid sequences were analyzed with the University of Wisconsin Genetics Computer Group Software package [14]. The putative peptide chain encoded by the identified cDNAs was designated AgX. Northern and Reverse Transcription (RT)-Polymerase Chain Reaction (PCR) Analysis Total RNA from human tissues was isolated after the method of Chomczynski and Sacchi [15]. Human testis poly(A)+ RNA was a gift from John Herr, University of Virginia, Charlottesville, VA. Poly(A) + testis RNA (2 pxg) and total placenta RNA (20 ptg) were separated on 1% formaldehyde-agarose [16], blotted onto nitrocellulose, and fixed at 80°C in a vacuum oven. Filters were hybridized with an AgX random-primed cDNA probe, washed to a final stringency of 0.1-strength SSC (20strength SSC: 3 M NaCl, 0.3 M sodium citrate.2H 2O, pH 7.0), 0.1% SDS at 65°C, and exposed overnight to Kodak X-OMAT (Rochester, NY) x-ray film. Differential expression of AgX messages was assessed by RT-PCR [17]. Approximately 3 pxg total RNA with 1 pxg poly (T) 1 5-1 8 oligonucleotide in a volume of 10.5 l1was denatured for 5 min at 70°C. After cooling to room temperature, the RNA was reverse transcribed at 42°C for 1 h in a 25-jil reaction mix containing 25 U reverse transcriptase (Promega), single-strength RT buffer (50 mM Tris-HCl, pH 8.3, 75 mM KCl, 10 mM MgC12, 0.5 mM spermidine), 40 U RNase inhibitor (Promega), 1 mM each dNTP, and 10 mM dithiothreitol and then passed over a Sephadex G-50 spin column. The entire cDNA preparation was amplified in 100 1 containing 8 U Taq polymerase (Perkin-Elmer/Cetus, Norwalk, CT), single-strength PCR buffer (1.5 mM MgC12, 50 mM KCl, 10 mM Tris-HCI, pH 8.3), 25 pmol of each PCR primer, 32 and 5 pICi ao-[ P]-dCTP (Amersham, Arlington Hts., IL); it was then covered with two drops of light mineral oil. Reactions were incubated in a Perkin-Elmer/Cetus DNA thermal cycler for 2 min at 95°C and then cycled 20 times at 95°C for 1 min, 45°C for 2 min, and 72 0C for 1 min. Amplified cDNA was separated on a 6% polyacrylamide/8 M urea sequencing gel with wedged spacers. The dried gel was exposed to Hyperfilm-MP (Amersham) at -70°C with an intensifying screen. Bands were quantitated by scanning densitometry and analyzed by the Gel Scan software package (Pharmacia, Piscataway, NJ). RT- controls and PCR of genomic DNA confirmed that the amplified products were not a result of genomic DNA present in the RNA samples. Primers used in the PCR were as follows: AgX-sense 5'-CCAGTTTGCAAAGAAGT-3' and antisense 5'-GCTCTAGACATCATTGGCATCCITC-3', nucleotides 1390 to 1406 and 1611 to 1594, respectively, of the testis AgX cDNA (the antisense oligonucleotide contains the sequence of an Xba I site at the 5' end); human small ribosomal protein 14 (HS14)-sense 5 '-GGCAGACCGAGATGAATCCTCA-3' and antisense 5'-CCTATTTCCTCCTGTGGCCCGG-3', nucleotides 187 to 208 and 312 to 291, respectively, of the HS14

sequence [18]. HS14-specific primers were a gift from Dr. Mark Leonard, Northwestern University, Evanston, IL. Generation of Anti-AgX Antisera and Affinity Purification A synthetic peptide containing the carboxy-terminal 20 amino acids of the AgX peptide chain (gift from Dr. Pravin Kumaya, Ohio State University, Columbus, OH) was conjugated to diphtheria toxoid as a carrier protein [19] and used to immunize two rabbits. The rabbits were bled and the anti-AgX antibody titer was assessed by ELISA [20]. Antibodies against the AgX peptide were affinity purified by the method of Snyder et al. [21]. Briefly, antibodies were bound to and eluted from a 3-galactosidase fusion protein containing the 20 amino acid peptide as expressed by a Xgtll clone. Antibodies eluted from the protein expressed by the Xgtll vector were included as a negative control. Western Analysis Homogenates of human tissues were prepared by grinding frozen tissue samples to a powder with a mortar and pestle and homogenizing the samples with a Dounce homogenizer in 10% (w/v) 0.1 M Tris-HCl, pH 7.4. To prepare homogenates of human sperm, ejaculates were washed once with PBS (0.01 M P0 4, 0.15 M NaCI, pH 7.4), resuspended in 300 ul1 0.1 M Tris-HCI (pH 7.4), and homogenized by freeze-thaw in a MeOH-dry ice bath. Homogenates were spun at 12 000 rpm for 20 min to remove cell debris. The protein extracts were separated by reducing SDSPAGE [22], transferred to nitrocellulose [23], and blocked in 3% BSA/Tris-buffered saline (TBS: 50 mM Tris-HCI, 150 mM NaCl, pH 8.0). Filters were incubated with the affinitypurified antibodies overnight at 4C. Goat anti-rabbit alkaline phosphatase-conjugated secondary antibody (Bio-Rad, Richmond, CA) was diluted 1:1000 in 3% BSA/TBS and incubated with the Western blots for 4 h at 4C. The reaction product was developed with nitro blue tetrazolium and 5bromo4-chloro-3-indolyl phosphate in 100 mM Tris-HCl (pH 9.5), 100 mM NaCl, 5 mM MgC12. Between antibody incubations and before the color reaction, the blots were washed three times in 0.1% BSA/TBS with 0.05% Tween-20 in the second wash. Molecular masses of identified proteins were estimated by comparing relative mobility with a plot of molecular weight standards. Immunofluorescence Microscopy The anti-AgX peptide antiserum was diluted 1:150 in 1% BSA/PBS with or without 36 p.M AgX peptide. After overnight incubation at 4C, the samples were centrifuged at 14 000 rpm for 10 min. Motile human sperm were isolated by a swim-up procedure [24], air dried on glass slides, fixed with 2% paraformaldehyde, and blocked in 1% BSA/PBS. The slides were incubated with the prepared serum samples for 30 min at room temperature in a moist chamber. After three washes in PBS, the slides were incubated with

1089

CHARACTERIZATION OF A SPERM ANTIGEN

A GAATTCGGGGTGGCGAGAGGGGCGGGGGCCGGGOCTGTCTCCACTTGGCCCCGCTCCCGGCCGCCCCGCCGCCGCCC 80 CCCGGATGAGGGTATATATTCGGAGTGACGGGACCGATGGTGCCGCGCCGAAGGAGCTGGAGACGGTCGTACTG 160 CGTCGCCGAAAAGG'rTTACAGGTACATACA ACACCCTATCAAAGCTTTGCTATATACATVAAGAACAT 240 14 N I 3 TAATGACCTCAAACTCACGTTGTCCAAAGCTGGGCAAGGCACCTACTACGTTTCTGGAATGAGCTTGAGAAGCCCAAC 320 N D L K L T L S K A G Q E H L L R F # N E L E E Q Q 30 AGGTAGAACTTTATGCAGAGCTCCAGGCCATGAACTTTGAGGAGC GAACTTCTTTTTCCAAAGGCCATTGAAGGTTTT 400 V E L Y A E L A N N F E E L N F F F K A I E G F $6 AACCGTCTTCTCACC AAAAGAATG TGGAGCAC GAATGGAACCTGTGCC TCGAGAGGTATTAGGC AGTGCTACAAGGA 480 N Q S S H K N V D A R N E P V P R E V L G A T R D B3 TCAAGATCAGTCCAGGCCTGGGAAAGTGAAGGACTTTTCCAGATTTCTCAGAAAAAGTAGCAGTTCTTCTTCTAGCTG 560 Q D Q L Q A W E S E G L F Q I 0 N K V A V L L L A G 110 GTGGGCAGGGGACAAGAC TCGTCAACCTAAG TCGGCGTTGCATATAC G 0 G T R L G V A P K

G

TGTA TGAGTTGGT TTGCCATCCCGTAAGACAC TTTTT M D V G L P S R K T L F

640 136

CAGATTCAGOCAGAGCGTATCCTGAAGCTACAAGCAGGTTOCTGAAATATTATGGCAACA/ATGCATTATTCCATGGTA 720 Q I Q A E R I L L Q Q V & E K Y G N .K C I I P W * 163 TATAATGACCAGTGGCAGACAATGGAATCTACAAGGAGTTCTTCACCAAGCACAAGACTTTGGTTTAAAAAAAGAGA 000 I T S G R T E S T E F F T K H K F G L X K E N 190 ATGTAATC TTTTCAGCAAGGAATGC TCCCCGCCATGAGTTTTGATGOGA TTATTTTGAGAGAAGAACAAAGTT 880 V I F F Q Q G M L P A H S F D G K I I L E E K N K V 216 TCTTGGTCCCAGATGGAATGGGGCTAGOCACTVGCACCCAGAATATTGTCAGGATATGCAAAGAGG S H A P D G N G 0 L R A L A A Q N I V E D N E Q R G

960 243

CATTTGGAGCATTCATGTCTATTOTGTTGACAACATATTAGTATAAGGGTCAGACCCACGGTTCATTGATTTTATTC 1040 I S I H V C V D N I L V K V A D P R F I G F C I Q 270 AGARAGGAGCAGACVTGTGACAAAGGTGGTAGAGAAACGAACCCTACAGAACCAGTTGGAGTGGTTTGCCGAGTGGAT 1120 K G A D C G A K V V E K N P T E P V G V V C R V D 296 GGAGTT TACCAGGTGG TAGAATATAGAGAGATTTCC CTGGCAACAGCTC AAAAACGAAGC TC AGACGGACGAC TGCTGT T G V Q V V E E I S L A T A Q R S S D G R L L F

1200 323

CAATGCGGGGAACATTGCCAACCAT TCTTCACTGTACCATTTCTGAAGGATGT TGTCAATGTTTATGAACCTCAGTTGC 1280 N A G N I A N H F F T V P F L R V V N V Y E P L Q 350 AGCACCATGTGGC TCAAAAGAAATCCTTAGGGA A K K I P V D T N H V AA K

ACCCAAGGACGTTAATTACCAGACACCCAATGGAATA 1360 376 G 0 L I K P D K P N G

FIG. 2. Identification of a 2.3-kb AgX mRNA by Northern analysis. Two micrograms of poly(A) + human testis RNA and twenty micrograms of total human placenta RNA were separated by electrophoresis, blotted onto nitrocellulose, and hybridized with an AgX cDNA probe. Lane 1, testis mRNA; lane 2, total placenta RNA.

GATGGAAAATT TGTCTT TGACATC TTCCAGTTTGCAAAAGTTTGTGGTATATGAAGTATTGCGAGAAGATGAGTT 1440 E K F V F D F F A K F V V E V L R E D E F 403

TTCCCCACTAAGAATGCTGATAGCAGAATGGGAAGACAACCCTACTACTGCAAGGCATGCTTTGATGTCCCTTCATC 1520 S P L X N A D S Q N G K D N P T T A R H A L M S L H H 430 ATTGCTGGGTCCTCAATGCAGG CCATTTCATAGATGAAAATAGCTCTCGCCTTCCAGCAATTCCCGCTTGAAGGAT 1600 C V L N A G G F I D E N S S R L P A I P R L D 456 GCCAATGATGTACCAATCCAATGTGAAATCTCTCCTCTTATC TCCTATGCTGGAAGGATTAGAAAGTTATGTGGCAGA A N D V P I Q C E I S P L I S A G E G L S V A D

1680 3

TAAAGATTCCATGCACCTCTAATCATCGATGAGAATGGAGTTCATGAGCTGGTGAAAAATGGTATTTGAACCAGATACC1760 K E F H A P L I I D E N G V N E L V K N C I 505

and three overlapping clones were isolated. These cDNAs spanned 2279 bp and included a 1515-bp ORF (505 amino acids) (Fig. 1A). The cDNA nucleotide and derived amino acid sequences were not similar to any sequences in the Genbank, EMBL,

AAGTTTTG TTTGCCACGATAGGAAAGC TTTTATTTTTGATAGACCAACTGTGAACCA A TACAAGACGTCTTGGACAACTGA 1840 AGTTTAAATATCCACAGGGTTTTATTTGCTTGTTGAACTCTTAGAGCTATTGCAAACTTCCCAAGATCCAGATACTGA 1920 ATTTCAGATAGC ATTTTTATGATTCCCAAC TCATTGAAGGTC TTATTTATATAATTTTTTCC AAGCCAAGGAACCATTG 2000 GCCATCCAGGzAATTTCGTACACTGAAATATAGGCAGGATGTTCAACATCAGTTTACTGCAGCTGGAAGCA TTTGTTT 2080 TTGAAGTTGTACATAGTAATAATAVGTCATTGTACATGTTGAAAGG TTTCT TGGTACTAAAAGTTTGTTTTATTTTATC 2160 AAAATTAACTTTTTTAAAAATAATTCAGT TAAATGTATC TTCTTTCTGGAAAAAAAAAAAAAAAAA 2240 2279 _AA ~~~~~0CCGAATTC

A

B

AgX-1

e~~~~~

. ........ 7,, :

. ~ ....... ..... ................ '7 ,,,

.,

I

1515 bp

-

I.o

1563 bp

AgX-2 FIG. 1. A) Nucleotide sequence of the testis AgX-1 cDNA clone and deduced amino acid sequence. Sites indicative of NTP binding were identified on the basis of matches with the [G/A]XXXX[G/AIK (single underline) and DXXG (double underline) consensus elements. The 20 amino acids recognized by the human sera and later prepared as a synthetic peptide are denoted (dashed underline). Abbreviations for the amino acid residues are: A, Ala; C, Cys; D, Asp; E, Glu; F, Phe; G, Gly; H, His; I, lie; K, Lys; L, Leu; M, Met; N, Asn; P Pro; Q, Gn; R, Arg; S, Ser; T, Thr; V, Val; W, Trp; and Y, Tyr. B) Kyte-Doolittle [281 hydropathy plot of the deduced amino acid sequence.

1:50 goat anti-rabbit IgG-fluorescein isothiocyanate (Boehringer-Mannheim, Indianapolis, IN) for 15 min at room temperature in a moist chamber. The slides were washed in PBS, mounted in glycerol/25 mM N-propyl gallate (Sigma, St. Louis, MO), and visualized using a Zeiss (Thornwood, NY) phase microscope equipped for fluorescence microscopy [25, 26]. RESULTS Sera from infertile patients identified a 600-bp clone that contained 60 bp of open-reading frame (ORF) and a poly(A) tail evident at the 5' and 3' ends, respectively, of the cDNA. The testis gtl1 library was reprobed with the 600-bp clone,

I

}-

J......

U

48bp

B AgX-1 AgX-2

1586ccCo

.

.

.

.

.

.

.C.........................................T-I 1595

11111

11111

1586-C

1643

C AgX-1

447-RLPAIPR ................ LKDANDVP-461

AgX-2

447-RLPAIPRATNGKSETITADVNHNLKDANDVP-477

1111111

11111111

FIG. 3. Diagrammatic structure of two AgX isoforms. A human placenta gtl 1 library was screened with the testis AgX cDNA. The ORF of an identified placenta AgX cDNA contained an inframe 48-bp insertion; the remainder of the sequence was identical to the testis cDNA. Testis and placenta AgX cDNA clones encoded unique isoforms designated AgX-1 and AgX-2, respectively. A) Diagrammatic comparison of the AgX-1 and AgX-2 cDNA clones. B) Comparison of the nucleotide sequence from AgX-1 (1586 to 1595) and AgX-2 (1586 to 1643). C) Comparison of the deduced amino acid sequence from AgX-1 (447 to 461) and AgX-2 (447 to 477).

1090

DIEKMAN AND GOLDBERG

B IeI -17

10C

0 o

X V-

6C

X

4

20

~~m,,

Testis

Placenta

17·

Muscle

r

Liver

FIG. 4. Quantitation of AgX-1 and AgX-2 mRNA by RT-PCR. A) AgX-specific oligonucleotide primers that hybridized on either side of the site of insertion generated PCR products of 229 bp and 277 bp for AgX-1 and AgX-2, respectively. Specific oligonucleotides for the HS14 nucleotide sequence were included as an internal control and produced a 126-bp PCR product. Lane 1, testis; lane 2, placenta; lane 3, skeletal muscle; lane 4, liver; lane 5, AgX-1 cDNA control; lane 6, AgX-2 cDNA control; lane 7, liver with HS14-specific oligonucleotide primers only; and lane 8, no cDNA control. RT- controls and PCR of genomic DNA confirmed that the amplified products were not a result of genomic DNA present in the RNA samples. B) Relative expression of the AgX-1 and AgX-2 mRNAs. PCR products from two different amplifications were analyzed by electrophoresis in triplicate and scanning densitometry. AgX values were normalized by calculating the AgX/HS14 ratios for each tissue. Error bars represent standard error of the mean. Filled bars, mean AgX-1 mRNA level; open bars, mean AgX-2 mRNA level.

CHARACTERIZATION OF A SPERM ANTIGEN

FIG. 5. Identification of -56- and -58-kDa AgX isoforms by Western blot analysis. Antisera were raised in rabbits against a synthetic peptide (dashed underlined in Fig. 1) containing the carboxy-terminal 20 amino acids of the AgX peptide chain. Anti-peptide antibodies were affinity purified by the method of Snyder et al. [21] with a -galactosidase fusion protein containing the 20 amino acid peptide. Western blots of human tissue extracts separated by reducing SDS-PAGE were probed with the affinity-purified antibodies and alkaline phosphatase-conjugated secondary antibodies. Lane 1, seminal plasma; lane 2, sperm; lane 3, testis; lane 4, placenta; and lane 5, skeletal muscle. A Western blot of muscle protein incubated with antibodies that bound the -galactosidase fusion partner (lane 6) revealed that the reactivity with a lower-molecular-mass band was background.

Swiss-Protein, or PIR data libraries [14]. Therefore, the putative peptide was designated AgX for unknown antigen. The secondary structure of the AgX peptide chain was predicted by a Chou-Fasman computer analysis to have or-helical and ,-sheet contents characteristic of globular proteins [27]. Neither a hydrophobic signal peptide at the amino terminus nor a hydrophobic region indicative of a transmem-

1091

brane protein was identified by hydropathy plotting (Fig. 1B) [28]. Three matches with the consensus sequence for an ATP/GTP-binding site or P-loop ([Gly/Ala]-XXXX-[Gly/ Ala]-Lys) and one match with a second consensus element indicative of nucleotide binding (Asp-XX-Gly) were identified in the AgX peptide chain. Furthermore, the spacing between the first two putative P-loops and the second consensus element (155 and 84 amino acids) both correlated with consensus patterns (-130-170 and -40-80 amino acids) [29,30]. An AgX cDNA probe detected a -2.3-kb mRNA on a Northern blot of testis RNA and an mRNA of slightly higher molecular mass in the placenta (Fig. 2). A human placenta Xgtll library yielded a cDNA clone that contained an inframe, 48-bp insertion in the ORF (Fig. 3) with the remainder of the clone identical to the testis cDNA. The placenta cDNA's ORF encoded a putative peptide chain of 521 amino acids with amino acid residues 454 through 469 encoded by the 48-bp insertion. Computer analysis results for the placenta cDNA sequence were similar to those obtained with the testis cDNA sequence. For convenience, the putative isoforms encoded by the testis and placenta cDNAs were designated AgX-1 and AgX-2, respectively. AgX mRNAs were differentially expressed among the tissues studied. RT-PCR analysis of human RNA demonstrated that both AgX mRNAs were present in each tissue, but the relative abundance of each mRNA varied. The AgX-1 mRNA was fiftyfold more abundant than the AgX-2 mRNA in the testis and at least tenfold more abundant than either AgX mRNA elsewhere (Fig. 4). In the placenta, expression of the

FIG. 6. Immunofluorescence of human sperm (x3465). Motile human sperm were isolated and fixed as described. A) Sperm incubated with the anti-AgX peptide rabbit antiserum. Specific fluorescent staining was detected on the principal piece of the sperm tail, the neck region of the head, and, to a lesser extent, the midpiece of the tail. Competition with the AgX peptide removed specific binding to sperm. Preimmune serum had no specific binding activity. B) Phase contrast.

1092

DIEKMAN AND GOLDBERG

AgX-2 mRNA was twice that of AgX-1. The two mRNAs were present in low, but relatively equal, proportions in skeletal muscle and liver. Affinity-purified anti-AgX peptide antibodies identified two protein bands (-56 and -58 kDa) in human testis and sperm extracts by Western blot analysis (Fig. 5). These proteins were operationally designated p56 and p58. Sperm extracts contained the p56 isoform with trace amounts of p58. In the testis, p56 was slightly more abundant than p58. Only the p58 band was detected in placenta and skeletal muscle homogenates and in seminal plasma. Immunofluorescence analysis of fixed human sperm detected AgX in the principal piece of the tail (Fig. 6). Specific fluorescence was also detected in the neck region of the head and to a lesser extent in the midpiece of the tail. Competition with the AgX peptide prevented specific binding. Preimmune serum was included as a negative control. DISCUSSION

The human antigen AgX was originally identified by infertile patients' sera containing antisperm antibodies. The unique AgX-1 isoform identified in sperm and in the testis is potentially foreign to the immune system since it is expressed in an immune-privileged site and is expected to contain epitopes not present in the somatic AgX-2 isoform. The 16 amino acid insertion/deletion could create unique epitopes and/or affect the folding of the isoforms to provoke sperm-specific antibodies that reduce fertility. Furthermore, the reactivity of an anti-AgX peptide antiserum with human sperm suggests that anti-AgX antibodies could bind in vivo to affect sperm function. The unique sperm isoform p56 is apparently encoded by the AgX-1 mRNA. The predicted molecular masses of 55.5 and 57.3 kDa determined from the AgX-1 and AgX-2 cDNAs correlated with the measured molecular masses of the p56 and p58 isoforms, respectively. In addition, the abundance of p56 in testis and sperm extracts is consistent with the abundant expression of the AgX-1 mRNA in the testis. While both mRNAs were expressed in the somatic tissues examined, p56 was absent from these tissues. Either translation of the shorter mRNA was blocked or our assay was not sufficiently sensitive to detect p56 in tissues other than testis. Both AgX isoforms occur in testis, but p56 appears to be the only species present in mature spermatozoa; trace amounts of p58 might appear from contamination with seminal plasma. In the testis, AgX-1 may be expressed exclusively in the germ line and AgX-2 in somatic cells. Alternatively, a switch from AgX-2 to testis/sperm AgX-1 expression may occur during spermatogenesis. Alternative splicing is a likely explanation for the variant AgX mRNAs and for control of a switch in isoform production. Results from preliminary genomic Southern analysis indicate a single AgX gene (unpublished observation). The shorter mRNA may result from splicing out a "mini-exon"

as has been suggested for mRNAs that differ only by a small insertion, e.g., the C1 and C2 hnRNP proteins [31]. Alternative splicing of short exons has also been proposed as an "on/off switch" for the testis isoforms of other proteins [32]. This tight control of gene expression by alternative splicing frequently occurs in differentiating tissues [33]. Immunofluorescence analysis of human sperm detected AgX in regions important for locomotion. Anti-AgX antibodies that bind to these regions could agglutinate sperm or impair motility; this in turn would cause a reduction in fertility [1, 2]. However, on the basis of its amino acid sequence, AgX is an apparently globular, cytosolic protein. Localization of AgX on the sperm surface and the mechanism of this localization require further study. The original cDNA clone detected with infertile patients' sera expressed only the carboxy-terminal 20 amino acids of AgX. The presence of this peptide at the carboxy terminus of both the AgX-1 and AgX-2 putative peptide chains suggests that the patients' sera reacted with epitopes common to both testis/sperm and somatic AgX isoforms. Therefore, a general reaction against AgX may occur, in which reduced fertility is only one symptom of a more widespread autoimmune syndrome. This fertility reduction may be the most apparent symptom of the autoimmunity due to the high abundance of AgX-1 in the testis and sperm and/or the as yet undetermined biological function of AgX1. Furthermore, anti-AgX antibodies may compromise pregnancy by immunoreactivity with AgX-2 in the placenta. Formal demonstration that AgX-1 is involved in immunologic infertility requires that the expressed peptide induce antibodies detrimental to sperm function. Nevertheless, our initial characterization of this novel human antigen supports this hypothesis. Ideally, one could build on these observations to develop a tissue-specific contraceptive vaccine. Further characterization of AgX should lead to an improved understanding of antibody-mediated infertility. ACKNOWLEDGMENTS The authors wish to thank Daniel Linzer and Robert Holmgren for critical comments on the manuscript, James Bartles for advice on the immunofluorescence, and Mark Leonard for advice on the RT-PCR.

REFERENCES 1. Bronson R, Cooper G, Rosenfeld D. Sperm antibodies: their role in infertility. Fertil Steril 1984; 42:171-183. 2. Lehman D, Emmons LR. Immunological Phenomena observed in the testis and their possible role in infertility. Am J Reprod Immunol 1989; 19:43-52. 3. Lehmann D, Temminck B, Da Rugna D, Leibundgut B, Muller H. Blot-immunoblotting test for the detection of anti-sperm antibodies. J Reprod Immunol 1985; 8:329-336. 4. Primakoff P, Lathrop W, Bronson R. Identification of human sperm surface glycoproteins recognized by autoantisera from immune infertile men, women, and vasectomized men. Biol Reprod 1990; 42:929-942. 5. Shelton J, Goldberg E. Serum antibodies to LDH-C. J Reprod Immunol 1985; 8:321-327. 6. Naz RK, Phillips TM, Rosenblum BB. Characterization of the fertilization antigen 1 for the development of a contraceptive vaccine. Proc Natl Acad Sci USA 1986; 83:5713-5717.

CHARACTERIZATION OF A SPERM ANTIGEN 7. Naz RK, Alexander NJ, Isahakia M, Hamilton MS. Monoclonal antibodies to a human germ cell membrane glycoprotein that inhibits fertilization. Science 1984; 225:342-344. 8. Naz RK. Involvement of fertilization antigen (FA-1) in involuntary immunoinfertility in humans. J Clin Invest 1987; 80:1375-1383. 9. Young RA, Davis RW. Efficient isolation of genes by using antibody probes. Proc Nad Acad Sci USA 1983; 80:1194-1198. 10. MillAn JL. Molecular cloning and sequence analysis of human placental alkaline phosphatase. J Biol Chem 1986; 261:3112-3115. 11. MillAn J, Driscoll CE, LeVan KM, Goldberg E. Epitopes of human testis-specific lactate dehydrogenase deduced from a cDNA sequence. Proc Natl Acad Sci USA 1987; 84:5311-5315. 12. Benton WD, Davis RW. Screening Xgt recombinant clones by hybridization to single plaques in situ. Science 1977; 196:180-182. 13. Sanger F, Nicklen S, Coulson AR. DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci USA 1977; 74:5463-5467. 14. Devereux J, Heaberli P, Smithies O. A comprehensive set of sequence analysis programs for the VAX. Nucleic Acids Res 1984; 12:387-395. 15. Chomczynski P, Sacchi N. Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 1987; 162:156159. 16. Lehrach H, Diamond D, WozneyJM, Boedtker H. RNA molecular weight determinations by gel electrophoresis under denaturing conditions, a critical reexamination. Biochemistry 1977; 16:4743-4751. 17. Chelly J, Kaplan J-C, Maire P, Gautron S, Kahn A Transcription of the dystrophin gene in embryonic erythroid non-muscle tissues. Nature 1988; 333:858-860. 18. Chen I-T, Dixit A, Rhoads DD, Roufa DJ. Homologous ribosomal proteins in bacteria, yeast, and humans. Proc Natl Acad Sci USA 1986; 83:6907-6911. 19. Lee AC, Powell JE, Tregear GW, Nial HD, Stevens VC. A method for preparing {-hCG COOH peptide-carrier conjugates of predictable composition. Mol Immunol 1980; 17:749-756. 20. Engvall E, Perlmann P. Enzyme-linked immunosorbent assay, ELISA III. Quantitation of specific antibodies by enzyme labeled anti-immunoglobulin in antigen coated tubes. J Immunol Methods 1972; 109:129-135.

1093

21. Snyder M, Elledge S, Sweeter D, Young RA, Davis RW. Xgtll: gene isolation with antibody probes and applications. Methods Enzymol 1987; 154:107-128. 22. Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nat New Biol 1970; 227:680-685. 23. Towbin H, Staehelin T, Gordon J. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc Natl Acad Sci USA 1979; 76:4350-4354. 24. Windt M-L, Menkveld R, Kruger TF. Spermatozoa separation. In: Acosta AA, Swanson RJ, Ackerman SB, Kruger TF, van Zyl JA, Menkveld R (eds.), Human Spermatozoa in Assisted Reproduction. Baltimore: Williams & Wilkins; 1990: 85-89. 25. Wright RM, John E, Klotz K, Flickinger C, Herr J. Cloning and sequencing for the human intra-acrosomal antigen SP-10. Biol Reprod 1990; 42:693-701. 26. Giloh H, Sedat JW. Fluorescence microscopy: reduced photobleaching of rhodamine and fluorescein protein conjugates by N-propyl gallate. Science 1982; 217:1252-1255. 27. Chou PY, Fasman GD. Prediction of protein conformation. Biochemistry 1974; 13:222-245. 28. Kyte J, Doolittle RF. A simple method for displaying the hydropathic character of a protein. J Mol Biol 1982; 157:105-132. 29. Dever TE, Glynias MJ, Merrick WC. GTP-binding domain: three consensus sequence elements with distinct spacing. Proc Natl Acad Sci USA 1987; 84:18141818. 30. Saraste M, Sibbald PR, Wittinghofer. The P-loop-a common motif in ATP- and GTP-binding proteins. TIBS 1990; 15:430-434. 31. Nakagawa TY, Swanson MS, Wold BJ, Dreyfuss G. Molecular cloning of cDNA for the nuclear ribonuclear particle C proteins: a conserved gene family. Proc Natl Acad Sci USA 1986; 83:2007-2011. 32. Waeber G, Meyer TE, LeSieur M, Hermann HL, Gerard N, Habener JF. Developmental stage-specific expression of cyclic adenosine 3',5'-monophosphate response element-binding protein CREB during spermatogenesis involves alternative exon splicing. Mol Endocrinol 1991; 5:1418-1430. 33. Smith CWJ, Patton JG, Nadal-Ginard B. Alternative splicing in the control of gene expression. Annu Rev Genet 1989; 23:527-577.