| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Department of Cell Biology and the Center for Recombinant Gamete Contraceptive Vaccinogens, University of Virginia School of Medicine, Charlottesville, Virginia 22908
b Ludwig Institute for Cancer Research, London W1W 7BS, United Kingdom
c BioQual, Inc., Rockville, Maryland 20850
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
reproductive immunology, sperm, sperm motility and transport
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and ß tubulin [3], dynein [4, 5], kinesin, tektin [6], and others [7]. The tektins are highly insoluble filamentous proteins found in both centrioles and the axoneme [8, 9]. Tektins were originally isolated from the flagella of sea urchin sperm, where 3 tektins of 55 (tektin A1), 51 (tektin B1), and 47 (tektin C) kDa have been defined [10, 11]. Complementary DNAs for tektins A1, B1, and C have been isolated from sea urchin, and the deduced amino acid sequences of these proteins have helical and nonhelical domains similar to those of intermediate-filament proteins [12–14]. More recently, tektins have been detected from a variety of sources such as mouse [15, 16] and dog [17]. Tektins have been studied developmentally in the sea urchin, in which both ciliogenesis and levels of tektin expression are correlated with the length of the microtubule doublets, suggesting coassembly from the monomeric pools of tubulin and tektin [18]. Further immunological work on sea urchin sperm flagella has demonstrated that tektins A1 and B1 form stable heterodimers [11] and that 1 outer doublet A-tubule protofilament is composed of tektin [19].
In invertebrates such as Drosophila melanogaster, tektin-like proteins have been found associated with the kl-3 loop of the Y chromosome [20]. Likewise, tektin from vertebrates has been found in the meiotic spindles of several dividing mammalian cell lines [8] and in murine neural tissue [21]. Monoclonal antibodies raised against sea urchin sperm flagellar tektin B1 stains a series of peptides of 48–50 kDa on Western blots of sea urchin sperm proteins and centrosomal material in mammalian Chinese hamster ovary cells where localization shows cell cycle dependence [7].
Our laboratory has been developing a comprehensive 2-dimensional (2D) gel protein database (Proteome) for human sperm by combining 2D gel electrophoresis, protein microsequencing, and immunoblotting [22, 23]. In the present study, novel peptide sequences were obtained by microsequencing a protein spot of 54 kDa with an isoelectric point (pI) of 5.3. These sequences were used to design optimized degenerate oligonucleotide primers to amplify and clone the cognate human testicular cDNA. We succeeded in cloning of a full-length tektin-B1 cDNA from human testis and in generating a monospecific polyclonal antibody to purified recombinant human tektin-B1 (rec-h-tekB1). Here, we report the coordinate position of the protein within our 2D sperm proteome, the expression and purification of the rec-h-tekB1 protein in bacteria, and immunological localization of the tektin-B1 in human sperm flagella.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Human sperm proteins for 2D gel electrophoresis were solubilized and separated as previously described by Naaby-Hansen et al. [22]. Fresh semen specimens were allowed to liquify for up to 3 h before centrifugation on 2-layer (80% and 55% in Ham F-10 medium) Percoll density gradients. The sperm pellet collected from the bottom of the 80% gradient was washed 3 times in Ham F-10 medium, and sperm were solubilized in lysis buffer consisting of 2% (v:v) NP-40, 9.8 M urea, 100 mM dithiothreitol (DTT), 2% ampholines (pH 3.5–10), and a cocktail of protease inhibitors. Isoelectric focusing was performed by loading 0.15 mg of sperm proteins on 15- x 0.15-cm acrylamide rods containing carrier ampholines (Pharmacia, Piscataway, NJ) composed of 28% (v:v) pH 3.5–5, 20% pH 5–7, 7% pH 7–9, and 45% pH 3.5–10 in a gel matrix [22]. Prior to electrophoresis, the sperm protein extracts in the rods were overlaid with a buffer containing 5% (v:v) NP-40, 1% ampholines (pH 3.5–10), 8 M urea, and 100 mM DTT. Focusing was conducted in steps: 2 h at 200 V, 5 h at 500 V, 4 h at 800 V, 6 h at 1200 V, and 3 h at 2000 V. Electrophoresis in the second dimension was performed by laying the isoelectrically focused tube gel onto a 0.15 x 20-cm linear (9%–15% acrylamide) slab gel in a Protean II xi Multi-Cell apparatus (Bio-Rad, Hercules, CA).
One-dimensional (1D) SDS-PAGE [24] of recombinant proteins and cell lysates was performed on a 16 x 18-cm gel electrophoresis apparatus (Bio-Rad) in 10%–15% polyacrylamide separating gels. Protein samples were suspended in Laemmli buffer [25] and boiled for 10 min. Escherichia coli lysates consisted of 1 OD-ml of bacterial culture pelleted before suspension in Laemmli buffer.
Electrophoretic transfer of separated proteins to nitrocellulose membranes was performed in transblot buffer (25 mM Tris, pH 8.3, 192 mM glycine, 20% methanol) at 1 amp for 1 h. The transblotted 2D gel membranes were utilized intact, and the 1D gel membranes were cut into strips, blocked with 5% fat-free dry milk in PBS-Tween (1x PBS, pH 7.4, 0.05% Tween), and incubated with rat antiserum in blocking buffer for 1 h. Immunodetection was performed with horseradish peroxidase-conjugated goat anti-rat IgG (Jackson ImmunoResearch, West Grove, PA) at a 1:5000 dilution in blocking buffer, and bands were visualized with either diaminobenzidine (DAB; Sigma, St. Louis, MO) in 0.1% H2O2 or TMB peroxidase substrate (Kierkegard and Perry Laboratories, Gaithersburg, MD).
Protein Microsequencing
Sequencing was performed by the W.M. Keck Foundation Center for Biomedical Mass Spectrometry at the University of Virginia. A protein spot of 54 kDa, pI 5.3, was cut from a polyacrylamide gel, minced, destained in 50% methanol, dehydrated in acetonitrile, and rehydrated in 50 µl of 10 mM DTT in 0.1 M ammonium bicarbonate for 1 h at 55°C. The sample was then alkylated in 50 µl 50 mM iodoacetamide and 0.1 M ammonium bicarbonate in the dark for 1 h at room temperature. The preparation was washed twice in 0.1 M ammonium bicarbonate and dehydrated in acetonitrile prior to lyophilization and digestion with 12.5 ng/µl trypsin in 50 mM ammonium bicarbonate on ice for 30 min. After removal of excess trypsin solution, 20 µl of ammonium bicarbonate was added and the sample was digested overnight at 37°C. Peptides were extracted from the gel in 50% acetonitrile/5% formic acid and were analyzed by liquid chromatography-mass spectrometry on an LC-Q apparatus (Finnigan, Austin, TX). Edman degradation was performed on a Biosystems Model 494 according to the manufacturer's instructions (Perkin Elmer Life Sciences, Boston, MA). Eight reliable peptide sequences were obtained: 1) SPXXSXK, 2) VATEFAFR, 3) GCLSLNLR, 4) SMMXXDTK, 5) MET(XorN)(EorQ)XDR, 6) FVPEVDTFTR, 7) TYRPNVELCR, and 8) LAQAQDALDALCK, where X could be either a leucine or an isoleucine residue.
Oligonucleotide Design and Reverse Transcription Polymerase Chain Reaction
Protein microsequences obtained by Edman and tandem mass spectrometry were compared, and oligonucleotides were designed for 2 sequences confirmed by both methods utilizing codon optimization tables [26]. Peptide 6 (FVPEVDTFTR) was reverse translated into oligonucleotides 6F (TTC/T-GTG-CCT/A-GAG-GTG-GAC/T-ACC-TTC/T-ACC-A/CG-3') and 6R (CG/T-GGT-G/AAA-GGT-G/ATC-CAC-CTC-A/TGG-CAC-A/GAA-3'), and peptide 8 (LAQAQDALDA) was reverse translated into oligonucleotides 8F (5'-CTG/T-GCC/T-CAG-GCC/T-CAG-GAT-GCC/T-CTC-GAC/T-GC-3') and 8R (5'-AG-G/AGC-G/ATC-GAG-G/AGC-ATC-CTG-G/AGC-CTG-G/AGC-3'). Oligonucleotide probes were manufactured by the University of Virginia Biomolecular Research Facility. For subsequent polymerase chain reactions (PCRs), human testicular RNA (Clontech, Palo Alto, CA) was reverse transcribed using 0.05 µg poly(A)+ RNA in a 20-µl reaction by combining 0.5 µg oligo-(dT)12–18 RNA and diethylpyrocarbonate (DEPC)-treated water prior to heating to 65°C for 10 min. Subsequently, 2 µl 10x reverse transcription (RT) buffer, 0.5 µl placental ribonuclease inhibitor (36 U/µl; Promega, San Luis Obispo, CA), 1 µl 10 mM solutions of the 4 dNTPs, and 1 µl AMV reverse transcriptase (23 U/µl; Stratagene, La Jolla, CA) were added, and the mixture was vortexed and incubated for 60 min at 42°C. After the addition of 80 µl DEPC-treated water, 1-µl aliquots of the cDNA solution were amplified for 40 cycles at various annealing temperatures (94°C for 30 sec; 60°C, 54°C, 48°C, or 42°C for 60 sec; and 68°C for 5 min) as specified by the manufacturer of the recombinant Thermophilus thermophilus (rTth) polymerase (Perkin Elmer). Separation and isolation of the PCR products were achieved by electrophoresis of reaction aliquots in 1.5% agarose gels made 1x in TAE (40 mM Tris-acetate, 1 mM EDTA) buffer followed by ethidium bromide staining, ultraviolet visualization, and photography. Collection of specific RT-PCR fragments was performed by electroelution from the agarose gel followed by precipitation, quantitation, and ligation into a pCR2.1-TOPO cloning vector according to the manufacturer's instructions (Invitrogen, Carlsbad, CA). This and all subsequent sequencing was performed by the University of Virginia Biomolecular Research Facility.
Screening of cDNA Library and DNA Sequencing
A 173-base pair (bp) insert obtained by RT-PCR was purified by treating the pCR2.1-TOPO cloning vector with EcoRI and collecting the released cDNA fragments by band excision after agarose gel separation. Fifty nanograms of the purified fragment was denatured by boiling and radiolabeled with [-32P]dCTP by using the 2 oligonucleotides described above with the random priming procedure of Feinberg and Vogelstein [27]. The radiolabeled fragment was purified on an Elutip-D column (Schleicher and Schuell, Keene, NH) and hybridized to plaque lifts (Magna Nylon Transfer Membranes; MSI, Westboro, MA) containing a total of 240 000 phage from a human testis 5'-stretch cDNA library (Clontech) in a solution containing 50% formamide, 5x standard sodium citrate (SSC), 5x Denhardt solution, 0.25 µg/ml yeast RNA, 0.5% SDS, and 0.05 M sodium phosphate (pH 7.0) at 42°C. After overnight hybridization, the filters were washed in a final solution of 0.2x SSC/0.2% SDS at 52°C before mounting, exposure to XAR-5 film (Kodak, Rochester, NY), and development. Twenty primary isolates were rescreened twice, and the remaining 8 positive isolates were converted from 
DR2 to pDR2 in AM1 cells according to the manufacturer's instructions. Sequencing was performed in both directions at the University of Virginia Biomolecular Research Facility, and the nucleotide and amino acid data were analyzed using the Genetics Computer Group and SEQWeb (Madison, WI) program packages.
Northern and Dot Blot Analyses
Fifty nanograms of the purified full-length 1.5-kilobase (kb) cDNA encoding h-tekB1 was radiolabeled by the random primer method [27], and the radiolabeled cDNA was isolated and hybridized in ExpressHyb (Clontech) to either a human multiple tissue Northern blot (Clontech) or a human RNA master blot (Clontech). All prehybridizations, hybridizations, and washings were performed according to the manufacturer's instructions. The human multiple tissue Northern blot contained 2 µg/lane of poly(A)+ RNA from spleen, thymus, prostate, testis, ovary, small intestine, mucosal lining of the colon, and peripheral blood leukocytes, and the human RNA master blot contained 100–500 ng poly(A)+ RNA from 50 different tissues normalized to various housekeeping genes.
Expression of rec-h-tekB1
The complete open reading frame (ORF) of the h-tekB1 cDNA was cloned into a pET-28b+ bacterial expression vector (Novagen, Madison, WI) by designing in-frame NcoI and XhoI sites in the 5' and 3' adapter primers, respectively. Engineering of a 3' XhoI site allowed for the addition of 6 histidine codons, contributed by the pET-28b+, onto the C-terminus of the h-tekB1 to facilitate isolation of the recombinant protein. The complete 1.5-kb cDNA containing both the 5' and 3' untranslated regions (UTRs) was amplified with rTth DNA polymerase using h-tekB1/pET-28/forward (5'-TCATGCCATGGCCACGCTGAGCGTCAAGCCAAG-3') and h-tekB1/pET28/reverse (5'-GCATCGCTCGAGGGCCAGCTCCAGCTGGCAGCT-3') primers. The resulting PCR products were separated on a 1.0% agarose gel, a 1.3-kb band was collected, and the pET-28 vector was digested with NcoI and XhoI. The band was reisolated on an agarose gel, ligated into the restricted pET-28 vector, and transformed into Novablue (DE3) host cells. Recombinants were screened by visualization of ethidium bromide-stained preparations, and a clone was chosen for sequencing by the University of Virginia Biomolecular Research Facility using primers surrounding the pET-28 vector cloning site to verify the correct insertion into the pET-28 expression vector and retention of the h-tekB1 reading frame.
Purified host cells containing the pET-28/h-tekB1 construct were grown in either 20-ml shake-flask cultures for preliminary examination or in a 10-liter fermentor (New Brunswick, Edison, NJ). Cultures were grown to an OD600 of between 0.5 and 0.7 before the addition of isopropyl ß-D-thiogalactopyranoside (IPTG) to a final concentration of 1 mM to induce protein expression. The cells were harvested by centrifugation after 3 h, lysed in Laemmli loading buffer, boiled for 5 min, and loaded onto standard 10% polyacrylamide separating gels as described above. Soluble and insoluble fractions of E. coli cells were obtained by resuspending cell pellets in 0.1x their original volume in 50 mM Tris-HCl (pH 8.0), 2 mM EDTA before addition of lysozyme to a final concentration of 100 µg/ml. After addition of 1% Triton X-100 at 0.1x the original cellular volume, the bacteria were incubated at 30°C for 15 min, sonicated, and centrifuged at 12 000 x g. The resulting supernatant contained the soluble protein fraction, and the pellet contained the insoluble cellular fraction. Fractionation of the E. coli cells into cytoplasmic and periplasmic fractions was performed by addition of CHCl3 at 0.01x the original volume of cells. After incubation for 15 min, 0.2x the original volume of 10 mM Tris-HCl (pH 8.0) was added, and the samples were then centrifuged for 15 min at 12 000 x g. The supernatant contained the periplasmic fraction of the bacterial cells, and the pellet consisted of the cytoplasmic fraction.
The rec-h-tekB1 containing a C-terminal (His)6-Tag was purified by resuspending the bacterial cell pellet in a binding buffer (5 mM imidazole, 0.5 M NaCl, 20 mM Tris, pH 7.9) prior to pulsed sonication at 4°C for 1–2 min and centrifugation at 12 000 x g. The resulting pellet containing the insoluble protein fraction and inclusion bodies was resuspended in binding buffer including 6 M urea, incubated on ice for 1 h, resonicated, and recentrifuged at 12 000 x g. The resulting supernatant was filtered through a 0.45-µm membrane. The filtered bacterial lysate was bound to a 5-ml bed-volume His-Bind Resin (Novagen) column prepared according to the manufacturer's instructions and washed with binding buffer containing 6 M urea until the OD280 approached baseline. The column was rewashed to baseline with wash buffer (40 mM imidazole, 0.5 M NaCl, 20 mM Tris, pH 7.9, 6 M urea) before the elution of the rec-h-tekB1 with elution buffer (300 mM imidazole, 0.5 M NaCl, 20 mM Tris, pH 7.9) and its dialyis against 1x PBS.
Generation of Rat Monospecific Polyclonal Serum
Young adult virgin female Lewis rats were immunized with 200 µg of purified rec-h-tekB1 in complete Freund adjuvant. The rats received 2 boosters of 200 µg rec-h-tekB1 in incomplete Freund adjuvantat at 2-wk intervals, and blood was collected 1 wk after each booster. Specificity of the rat antiserum for rec-h-tekB1 (-rec-h-tekB1) was tested by Western blotting against rec-h-tekB1 and an SDS lysate of human sperm proteins separated by 2D gel electrophoresis.
Affinity Purification by Olmsted Method
Purified rec-h-tekB1 (200 µg) was separated by standard SDS/PAGE and Western blotted as described above. After visualization by Ponceau S staining, a strip of the nitrocellulose membrane was cut and blocked by incubation in Blotto (PBS, Tween, 0.5% dry milk) before overnight incubation with 100 µl rat polyclonal -rec-h-tekB1 at 4°C in Blotto. After washing the strip 3 times in Blotto the specific h-tekB1 antibody was eluted in 0.4 ml 0.2 M glycine-HCl (pH 2.5) for 2 min before neutralization with 0.2 ml M KPO4 (pH 9.0) [28].
Immunofluorescent Microscopy
All incubations were performed at 22°C unless otherwise indicated. Immunofluorescence of swim-up sperm was performed by drying a human sperm preparation on poly-L-lysine-coated slides and fixing it with methanol for 10 min before blocking with 5% normal goat serum/PBS for 30 min. The sperm were next exposed to preimmune serum and antiserum (1:50) for 1 h before washing 3 times with 1% NGS/PBS/Tween and incubation with fluorescein isothiocyanate-conjugated goat anti-rat secondary antibody.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Our laboratory has utilized 2D electrophoresis to resolve silver-stainable proteins by isoelectric focusing (IEF) and nonequilibrium pH gradient electrophoresis (NEPHGE) from extracts of human sperm [22]. Using IEF, a single protein spot having a relative molecular mass of 54 kDa and an estimated pI of 5.3 (Fig. 6B) was chosen from a closely compacted area showing numerous 125I-labeled protein spots. The protein spot was excised from a Coomassie-stained 2D gel and was tryptic digested, and the resulting peptides were microsequenced yielding the peptide sequences listed in Materials and Methods. Analysis of these protein fragments by comparison with the nonredundant and expressed sequence tag (EST) National Center for Biotechnology Information (NCBI) databases did not reveal significant matches to any protein in the databases at that time.
|
Two peptides, 6 (FVPEVDTFTR) and 8 (LAQAQDALDA), were selected for the manufacture of oligonucleotides because they were at least 7 amino acids in length (the minimal length to synthesize oligonucleotides of sufficient Tm) and because their sequences were confirmed by both Edman and tandem mass spectrometry. Because the peptide sequences were obtained by sequencing tryptic fragments of the protein, the orientation of the peptides in the protein's primary sequence was unknown. To design a PCR strategy to overcome this difficulty, 2 sets of oligonucleotides were manufactured in both the forward and reverse directions corresponding to the alternatives of each peptide having a more N-terminal position in the primary sequence of the original protein. The number of sequences in the pool of oligonucleotides was reduced by employing those codons most common in mammals following codon bias data of Ausubel et al. [26].
The 2 sets of oligonucleotides were used for RT-PCR assay of human testicular poly(A)+ mRNA, and the RT-PCR products were separated on agarose gels, which were then stained with ethidium bromide (Fig. 1). The combination of oligonucleotides 6F and 8R failed to generate any discernable product, but oligonucleotides 6R and 8F produced 6 distinct bands (Fig. 1, lanes 1, 2, and 4) with various annealing temperatures utilized in the experiment. Degenerate calreticulin primers were utilized as a positive control to assure the viability of the reactions, and negative control reactions contained no input cDNA (data not shown). The 6 RT-PCR-generated DNA bands were excised from the gel and subcloned into the pCR2.1-TOPO cloning vector as described in Materials and Methods. Plasmid clones were verified, purified, and submitted for sequencing utilizing primers corresponding to vector sequences surrounding the insertion site. Upon examination, 5 of the clones represented PCR artifacts, including mispriming, out-of-frame amplimers with stop codons, and amplimers lacking a common register.
|
Clone 5 (Fig. 1, lane 2), consisting of a 173-bp insert, had both primers on either end of the RT-PCR-generated DNA segment and a productive reading frame that contained the 2 translated protein sequences from the microsequence data. A GenBank database search revealed the closest matches to be with a murine testicular EST DNA sequence (accession AA145596) and a Strongylocentrotus purpuratus (sea urchin) tektin B1 protein (48% homology over 43 amino acids; accession L21838). The 173-bp clone 5 insert was excised from the plasmid and utilized as a radiolabeled probe to obtain a full-length cDNA from a human testicular cDNA library in DR2. The 16 positive clones isolated were of 2 size classes (approximately 1.3 and 1.5 kb) as judged by agarose gel separation. One clone of each class was sequenced, confirming that the 2 clone classes contained the same cDNA sequence, except that the smaller was truncated at the 5' end.
Analysis of the h-tekB1 cDNA
Sequence analysis in both directions was performed on the larger clone, resulting in the nucleotide structure illustrated in Figure 2A. The clone possessed a 1290-bp ORF with a starting methionine codon at bp 128–130 (Fig. 2A) and a stop codon at bp 1418–1420. This ORF was preceded by a 127-bp 5' UTR containing an in-frame stop codon, no other apparent methionine codons, and a partial consensus Kozak sequence [29] at bp 122–127. The ORF was bracketed on the 3' end by a 90-bp 3' UTR that contained a polyadenylation signal (AATAAA) [30] at bp 1490–1495, 15 bp before the cDNA poly(A) tail. Translation of the putative ORF (Fig. 2A) revealed a 430-amino acid protein with a predicted molecular weight of 49.7 kDa. Computer analysis predicted a pI of 5.4, in precise agreement with the observed pI of the protein isolated from the 2D gel. Further analysis revealed numerous consensus serine/threonine protein kinase phosphorylation sites, 1 tyrosine kinase consensus site, and 5 N-glycosylation sites. Additionally, computer modeling predicted that the protein contained 64–68% alpha helical residues present in 5 stretches (amino acids 21–46, 62–107, 134–175, 218–320, and 339–381).
|
The ORF contained the 2 original peptides used for construction of the oligonucleotides employed in the RT-PCR cloning procedure (peptide 8, amino acids 356–368; peptide 6, amino acids 404–413) and a number of the other peptides obtained by microsequencing (peptide 5, amino acids 180–187; peptide 3, amino acids 188–195; peptide 1, amino acids 196–202; peptide 2, amino acids 259–266; peptide 7, amino acids 322–331; peptide 4, amino acids 383–389). A computer search of the GenBank database revealed (Fig. 3) a marked resemblance of the full-length clone to sea urchin tektin B1 except at the N- and C-termini, where the sequences diverged. The h-tekB1 cDNA had 55% identity (amino acids 58–398) with S. purpuratus tektin B1 (amino acids 31–371), 32% identity (amino acids 6–396) with S. purpuratus tektin C1 (amino acids 5–395), and 32% identity over a smaller region (amino acids 16–387) with S. purpuratus tektin A1 (amino acids 70–441). The human sequence also had 83% identity with a cloned murine tektin gene (tektin-t) distributed over its whole length [31]. A BLOCKS search revealed all 4 of the Tektin signatures (BLOCKS PR00511A-D) in h-tekB1 at levels of 41%, 81%, 35%, and 46% identity, respectively. Database searching (BLOCKS RPS-BLAST) also revealed a chaperonin signature [32], which appears over amino acids 253–304 in h-tekB1 with 47% homology (20% identity) with the bacterial Clp chaperonins (BLOCKS IBP001270).
|
Analysis of a genomic database (NCBI LocusLink) revealed that the h-tekB1 gene resides as 10 exons spanning 4.2 kb on chromosome 1 (Fig. 2B). The TEKT2 locus resides at 1pter-p32.3, 158 kb from the EIF2C1 locus and 115 kb from the BET3 locus.
Tissue Specificity of h-tekB1
To ascertain the pattern of tissue expression of the gene transcript, a Northern blot containing mRNA from various human tissues was probed with the radiolabeled h-tekB1 cDNA (Fig. 4A). A single prominent band of approximately 1.5 kb was observed only in the lane containing testicular mRNA. The size of the mRNA transcript in the autoradiogram also verified that a full-length cDNA had been isolated. An probe of the blot with radiolabeled ß-actin (Fig. 4A, lower panel) was performed as a quality control for the mRNA on the membrane. An array of 50 different human tissue mRNAs was then hybridized to a radiolabeled h-tekB1 on a dot blot (Fig. 4B). The results confirmed that although h-tekB1 is indeed transcribed in adult testis (D1), the gene is also expressed at similar levels in adult trachea (F3) and fetal lung (G7) and to a lesser degree in ovary (D2), pituitary (D4), adult lung (F2), fetal brain (G1), and fetal kidney (G3). Serial analysis of the gene expression tag to gene mapping (SAGEmap) also revealed expression of h-tekB1 in normal cerebellum. The message was also detected in mammary gland tumors.
|
Recombinant Expression and Immunological Analysis
To localize and characterize h-tekB1, a monospecific polyclonal serum was generated to rec-h-tekB1 expressed in the E. coli pET28 vector (Novagen). Figure 5 illustrates the expression of rec-h-tekB1 protein in bacteria on a Coomassie-stained SDS/polyacrylamide gel. The preinduction sample showed no protein band at the approximate size expected for the rec-h-tekB1 protein (Fig. 5, lane 1), but the 3-h postinduction culture demonstrated a prominent band at the expected electrophoretic mobility for rec-h-tekB1 (Fig. 5, lane 2). Fractionation of the bacterial cells sampled at the 3-h time point into soluble and insoluble fractions (Fig. 5, lanes 4 and 5, respectively) demonstrated the presence of the recombinant band in the insoluble cellular fraction (lane 5), and further fractionation (Fig. 5, lanes 6 and 7, respectively) revealed the recombinant band in the cytoplasmic compartment (lane 7).
|
Large-scale preparation of purified rec-h-tekB1 was performed in a 20-liter fermentor, and the cytoplasmic fraction was prepared and solubilized. The rec-h-tekB1 was bound and batch eluted from His-Bind (Novagen) resin, producing a fraction that was more than 99% pure (Fig. 5, lane 3). The yield of the purified rec-h-tekB1 was judged by SDS/PAGE to be approximately 1.7 mg of rec-h-tekB1/liter fermentate. This material was used to immunize virgin female Lewis rats. Western blotting of 1D gels with rat monospecific polyclonal antiserum stained the recombinant immunogen (Fig. 6A, lane 3). When an extract of human sperm proteins was analyzed on a 2D Western blot (Fig. 6B2), 3 h-tekB1 isoforms were specifically immunostained. These 3 h-tekB1 isoforms migrated at identical molecular masses (53.5 kDa) that varied slightly in pI from 5.25 to 5.35. This observation indicates the presence of modified forms of h-tekB1 in human sperm.
Immunofluorescence of Human Sperm
The rat -rec-h-tekB1 antibody was utilized to localize h-tekB1 in human sperm. Live human sperm did not stain with the rat -rec-h-tekB1 antibody (data not shown), indicating that the h-tekB1 gene product is not present on the sperm surface. When permeabilized human sperm were exposed to immune rat serum (Fig. 7), the antibody localized to the principal piece of human sperm flagellum, although the midpiece and end piece of human sperm did not react. These results, along with the negative live staining, suggest an interior or axonemal location for the h-tekB1. A small area at the base of the sperm head was also positive for h-tekB1 in a large percentage of sperm (Fig. 7D, arrows). This area corresponds to the location of the basal body region of human sperm. The control preimmune serum (Fig. 7B) produced no staining.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A punctate region at the base of the head of the sperm also was positive for h-tekB1. This region of the human sperm corresponds to the basal body, containing a ring of 9 adjoined triplet microtubules. The basal body is considered centriole-like and serves as a starting point for growth of the outer doublet microtubules during flagellar morphogenesis. As such, it is contiguous with the rest of the axoneme. This observation is in agreement with previous immunofluorescence studies showing reactivity of tektin antibodies with basal bodies in sea urchin sperm [34], immunoblots of isolated Chlamydomonas basal bodies [10], and preparations of basal bodies in both scallop gill and rabbit tracheal epithelial cells [35].
We used a comprehensive 2D map of human sperm proteins [22] to identify and clone a human sperm flagellar protein. The 2D sperm proteome was constructed to understand and resolve the complexity of sperm proteins and to select candidates for inclusion in a immunocontraceptive vaccine. Toward this end, we labeled human sperm with either 125I or biotin prior to solubilization and electrophoresis of the sperm extracts. Comparison of the array of labeled proteins with the original 2D map was posited to allow selection of those sperm protein components localized on the cell surface. The tektin protein spot was originally cored from an area of tightly compacted protein spots showing at least 6 125I-labeled proteins. The lack of surface localization of the h-tekB1 that we observed in our immunolocalization studies suggests that the spot cored may have been 125I labeled because of sperm membrane damage or the miscoring of an unlabeled spot in the cluster. Although the h-tekB1 cloned and examined in the present work is not a surface molecule, the work serves to define the h-tekB1 isoforms within the human sperm proteome. We are currently using biotin labeling and phase partitioning to enrich for vectorially labeled sperm surface proteins [23, 36].
The 2D Western blot data presented in this study indicate that there are 3 separate isoforms of h-tekB1 in human sperm. These immunoreactive forms of the protein differ not by molecular mass (they all migrate at 54 kDa) but by charge, indicating that some posttranslational modification has occurred. The observed mass of the h-tekB1 is nearly identical to the mass predicted from the cDNA sequence, consistent with a nonglycosylated cytoskeletal protein. It may be assumed that the 3 isoforms are all human tektin B, because tektins A, B, and C in other species have consistently been of different molecular masses. These results support previous observations of microheterogeneity in the patterns of tektins [37]. Shifts of charge such as that observed in this study are usually due to changes in levels of phosphorylation. Although the microsequenced peptide fragments utilized for the design of oligonucleotide primers did not demonstrate phosphorylation on any of the serine or threonine residues, the spot we cored for initial examination may have been an unphosphorylated form of tektin B1. However, phosphorylation may be implicated in this heterogenity of protein charge because kinases are known to be important in the posttranslational modification of other sperm proteins [38–40].
Examination of the protein sequence of h-tekB1 indicates that the highest degree of homology with a species in which all 3 tektins have been cloned is with sea urchin tektin B1. A BLOCKS search revealed a match to all 4 of the Tektin signature domains (BLOCKS PR00511A-D), providing further proof of the identity of the translated human cDNA clone. The differences in amino acid sequence between human and sea urchin tektin B1 are distributed evenly throughout the protein; however, both the N- and C-termini of the human protein revealed no homology with any of the sea urchin tektins. The S. purpuratus tektin A1 contained an extra 54 amino acids at the N-terminus and was shorter by 22 amino acids at the C-terminus, whereas sea urchin tektin C1 demonstrated only 15% homology over the first 40 amino acids with human tektin B1. In addition to sequence homology, a structural similarity to other tektins consisting of 5 helical segments [14] was found by analysis with secondary structure prediction programs. This secondary structure has been hypothesized to reflect the evolutionary relatedness of tektins to the intermediate filament family of proteins [12, 14]. A match to a chaperonin signature was discovered during database searching [32]. Over amino acids 253–304 in h-tekB1, there is 47% homology (20% identity) with the bacterial Clp chaperonin (BLOCKS IBP001270) subfamily, which is known to present and activate proteins for repair during stress conditions. Interesting functional correlations arise because mammalian chaperones are associated with the centrosome [41].
Although h-tekB1 was cloned and sequenced from testis mRNA, Northern blots revealed the presence of the transcript in pools of mRNA from trachea, lung, ovary, pituitary, brain, and kidney. Trachea, lung, and brain tissues are known from histological examination to contain many cilia. Expression of tektin B1 in tracheal tissue, lung, and neural tissue was noted previously in the mouse [21]. Because of the wide transcriptional distribution of this gene product and the lack of surface localization, h-tekB1 would not appear to be useful as a contraceptive immunogen. The broad tissue distribution also diminishes the likelihood that an antagonist might be found that selectively affected sperm motility by targeting tektin B1.
The single closest match in the databases was a murine tektin B1 [15] (accession AB027138) that showed 83% identity with the translated human clone. This murine tektin-t (testis) has been termed haploid germ cell specific based on Northern blot analysis, which demonstrated only testis transcription, and the cloning of the transcript from a subtracted murine testis cDNA library [15]. Our results for h-tekB1 differ with respect to tissue specificity because our cDNA, also cloned from a testicular source, is found in multiple tissues, and therefore the designation of h-tekB1 as testicular or germ cell specific is inappropriate. Differences in expression of tektin B1 in murine and human tissues may account for the discrepancies in our results and in those of Iguchi et al. [15]. In addition, on Western blots we observed a molecular mass of 54 kDa for h-tekB1 whereas Iguchi et al. [15] noted a mass of 85 kDa for murine tektin-t.
|
FOOTNOTES |
|---|
1 This work was supported by NIH U54 HD29099 and a grant from the Andrew W. Mellon Foundation. Note: The clone sequence from this study was assigned GenBank accession number AF054910. During the preparation of this paper, a human clone (accession AB033823) was submitted to GenBank on 12 January 2000 by Iguchi et al. AB033823 differs from our clone (submitted on 2 April 1998) by 66 of the 430 amino acids. 
2 Correspondence: John C. Herr, Department of Cell Biology, Box 439 Jordan Hall, University of Virginia, Charlottesville, VA 22908. FAX: 804 982 3912; jch7k{at}virginia.edu
Accepted: August 29, 2001.
Received: April 3, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. J. Wolkowicz, L. Digilio, K. Klotz, J. Shetty, C. J. Flickinger, and J. C. Herr Equatorial Segment Protein (ESP) Is a Human Alloantigen Involved in Sperm-Egg Binding and Fusion J Androl, May 1, 2008; 29(3): 272 - 282. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Zuccarello, A. Ferlin, A. Garolla, M. A. Pati, A. Moretti, C. Cazzadore, S. Francavilla, and C. Foresta A possible association of a human tektin-t gene mutation (A229V) with isolated non-syndromic asthenozoospermia: Case Report Hum. Reprod., April 1, 2008; 23(4): 996 - 1001. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Tanaka, N. Iguchi, Y. Toyama, K. Kitamura, T. Takahashi, K. Kaseda, M. Maekawa, and Y. Nishimune Mice Deficient in the Axonemal Protein Tektin-t Exhibit Male Infertility and Immotile-Cilium Syndrome Due to Impaired Inner Arm Dynein Function Mol. Cell. Biol., September 15, 2004; 24(18): 7958 - 7964. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. J. Wolkowicz, J. Shetty, A. Westbrook, K. Klotz, F. Jayes, A. Mandal, C. J. Flickinger, and J. C. Herr Equatorial Segment Protein Defines a Discrete Acrosomal Subcompartment Persisting Throughout Acrosomal Biogenesis Biol Reprod, September 1, 2003; 69(3): 735 - 745. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Shetty, M. J. Wolkowicz, L. C. Digilio, K. L. Klotz, F. L. Jayes, A. B. Diekman, V. A. Westbrook, E. M. Farris, Z. Hao, S. A. Coonrod, et al. SAMP14, a Novel, Acrosomal Membrane-associated, Glycosylphosphatidylinositol-anchored Member of the Ly-6/Urokinase-type Plasminogen Activator Receptor Superfamily with a Role in Sperm-Egg Interaction J. Biol. Chem., August 15, 2003; 278(33): 30506 - 30515. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
X174 markers. RT-PCR products were separated to demonstrate the 6 major products from the 8F/6R primer pair. Products were cloned into pTOPO vectors and sequenced. Only band 5 (180 bp) represented the legitimate PCR product corresponding to the original protein spot by virtue of an ORF and the presence of both primers in proper reading frame with respect to one another
