| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Regular Article |
a Center for Reproductive Biology, School of Molecular Biosciences, Washington State University, Vancouver, Washington 98686
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
spermatid, spermatogenesis, testis
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Transferrins are iron-binding proteins belonging to a family of glycoproteins that includes serotransferrin in blood (often referred to simply as transferrin), ovotransferrin (conalbumin) in bird egg whites, lactoferrin in mammalian biological fluids, and melanotransferrin in membranes of melanocarcinoma cells [2]. Each of these proteins share sequence homology, including highly conserved cysteine residues. The approximate N-terminal half is homologous to the C-terminal half of each protein. X-ray analysis of the structure of rabbit serotransferrin revealed that each half forms a lobe of approximately 330 amino acids composing two domains, which contribute binding pockets for ferric ion and an accessory anion [3]. The cysteine residues, which are highly conserved among the transferrins and among species, are also conserved in hemiferrin, suggesting that the protein could fold somewhat like the C-terminal lobe of transferrins.
The transferrins are believed to have originated from a gene duplication event, with each half of the molecule then evolving further [4, 5]. However, an ancestral single gene has not been conclusively demonstrated in primitive organisms. Hemiferrin could represent a candidate ancestral gene or perhaps an derived form of that gene. A small transferrin pseudogene reported in human genomic DNA [6] has been proposed as the human counterpart of rat hemiferrin [5]. However, only one copy of the transferrin gene was observed in rat genomic DNA [7].
Overlapping 5' and 3' portions of the bovine transferrin cDNA were cloned from a liver cDNA library after all attempts to select a full-length transferrin clone failed [8]. The sequence of the putative 3' portion of bovine liver transferrin shares >96% identity with the reported sequence of rat hemiferrin [8, 9]. If this substantial identity were correct, it would have significant implications in the understanding of the evolution of transferrins. Alternatively, hemiferrin may be a highly conserved sequence expressed in bovine liver, whereas no detectable expression of a shorter transferrin transcript was observed in rat liver [1]. Initially, we directed our efforts toward understanding more about the rat hemiferrin gene and protein; however, repeated difficulties with molecular techniques led us to question the reported sequence. We found evidence that the 0.9-kb transcript in rat germ cells is a truncation of the rat transferrin message and is not a unique sequence. Hereinafter, the term hemiferrin refers to a cDNA with the sequence reported by Stallard et al. [1].
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Adult male rats were purchased from B&K Universal (Fremont, CA) and maintained with ad libitum food and water until they were incorporated into treatment groups or killed to harvest tissues. Fresh bovine liver and testis tissue samples from a 3-yr-old bull were acquired at slaughter (Dr. Jerry Reeves, Washington State University, Pullman, WA). The bovine tissues were placed in RNA-later (Ambion, Austin, TX) and stored at -80°C until used. Germ cells were isolated as described by Bellvé et al. [10]. All animal protocols were conducted in accordance with the Guide for Care and Use of Laboratory Animals and approved in advance by the Institutional Animal Care and Use Committee of Washington State University.
Northern Blot Analysis
Total RNA was isolated from fresh tissues or from tissues frozen in RNA-later using Trizol reagent (Gibco-BRL, Grand Island, NY) according to the manufacturer's instructions. RNA samples were separated on 1% agarose gels in the presence of formaldehyde, wick blotted onto nylon membranes (Osmonics, Westborough, MA), and cross-linked to the membrane with ultraviolet light. The 842-base pair (bp) hemiferrin cDNA in pTZ18U and the 687-bp rat 3' transferrin cDNA (3'rTf) in pSP65 were a gift from Mike Griswold (Washington State University, Pullman, WA). The 3'rTf insert was identical to residues 1569–2093 of the rat transferrin sequence (GenBank no. NM017055) [11].
Random primed 32P-labeled cDNA probes were generated using the DECAprime DNA Labeling Kit (Ambion). The membranes were prehybridized in 50% formamide, 4x saline sodium citrate (SSC), 5x Denhardt solution, 0.2% SDS, and 100 µg/ml salmon sperm DNA at 42°C for 6 h in a roller oven. Hybridization with the selected 32P-labeled probes was accomplished under the same conditions for 10 h. The membranes were washed at 65°C for 15 min with 2x SSC and 1% SDS and then 0.2x SSC and 0.1% SDS for 15 min. The membranes were then exposed to x-ray film in the presence of an intensifying screen.
Ribonuclease Protection Assay
The hemiferrin-containing plasmid was cut with the BamHI endonuclease, and the rat 3' transferrin-containing vector was cut with HindIII. Radioactive antisense RNA probes were synthesized from the linearized templates using 32P-UTP (NEN, Boston, MA) and the Maxiscript transcription kit (Ambion) according to manufacturer's instructions. Total RNA was isolated from fresh rat livers and testes or bovine tissues frozen in RNA-later as described above. The RPA-III ribonuclease protection assay kit (Ambion) was used according to the manufacturer's instructions, with actin as a positive control for the kit reagents and rat cyclophilin as a positive internal control for tissue RNA. All probes were gel purified as recommended. The products of ribonuclease digestion were separated by electrophoresis in 5% polyacrylamide gels in the presence of 8 M urea. The gels were covered with plastic wrap and exposed to x-ray film with an intensifying screen overnight at -20°C.
Complementary DNA Library Screening
A high-quality rat testis cDNA library in 
gt11 enriched in cDNA 5' end regions (5'-Stretch Plus) was purchased from Clontech (Palo Alto, CA). Approximately 5 x 106 plaque-forming units was plated and hybridized with probe generated from the gel-purified 3'rTf insert sequence. Many positive clones were plaque purified, and the clone with an insert of the appropriate size was submitted for automated DNA sequencing (Laboratory for Biotechnology and Bioanalysis, Washington State University, Pullman, WA) using primer sequences from the vector arms.
Bacterial Expression of Hemiferrin
To gain optimal restriction sites, an EcoRI/HindIII fragment of approximately 700 bp of the hemiferrin sequence was subcloned into the same sites of the pCR-Script plasmid (Stratagene, La Jolla, CA). The KpnI/BamHI fragment containing the insert was then subcloned into the same sites of pQE-30, -31, and -32 vectors of the QIA-express Kit Type IV (Qiagen, Valencia, CA). Colonies were lifted onto nitrocellulose, induced with isopropyl-thiogalactoside for 4 h, and then lysed as described by the manufacturer. The filters were washed, blocked with 5% dry milk in Tris-buffered saline (TBS), and then incubated with 40 µl INDIA HisProbe-HRP (Pierce Chemical Co., Rockford, IL) in 50 ml TBS for 1 h at room temperature. After two 10-min washes with TBS, the horseradish peroxidase was detected with 4-chloronaphthol. The positive pQE-30 clones expressed a protein of the appropriate molecular weight that was largely contained in inclusion bodies. Automated DNA sequencing (Laboratory for Biotechnology and Bioanalysis) confirmed the presence of the hemiferrin cDNA insert in the pQE-30 vector.
Production of Hemiferrin Antibody
Large batches of bacteria expressing His-tagged hemiferrin were extracted with B-PER reagent (Pierce) to release soluble protein, and the inclusion bodies were further extracted with 8 M urea. The combined extracts were passed through a nickel-based affinity column (Qiagen), and the His-tagged protein and some contaminating proteins were eluted with 150 mM imidazole (pH 4.5). The eluted protein was dialyzed overnight against 33.3-mM 2-(4-morpholino)-ethane-sulfonic acid, 33.3 mM Hepes, and 33.3 mM sodium acetate (pH 4.5) and loaded onto a 4.6- x 100-mm POROS 50 HS cation exchange column (Perseptive Biosystems, Framingham, MA) in the same buffer. The column was washed and then eluted with a gradient of 0–1.5 M NaCl in the same buffer over 15 column volumes at a flow rate of 3.5 ml/min. Hemiferrin was eluted as a single INDIA HisProbe-HRP-positive peak and was judged 98% pure by silver stains of SDS-polyacrylamide gels. A total of 2.4 mg of this purified protein was accumulated and submitted for commercial antibody production in chickens (Strategic Biosolutions, Ramona, CA).
Western Blot Analysis
Approximately 50 mg of tissue or isolated cells was mixed with 2x Laemli SDS sample buffer, homogenized with a microfuge pestle, heated to 95°C for 5 min, and centrifuged at 14 000 x g to remove debris. Proteins in the supernatant were separated in 12% polyacrylamide gels and transferred to nitrocellulose as previously described [12]. After blocking with 5% dry milk in PBS, blots were treated with rabbit anti-rat transferrin (Cappel, Durham, NC) or chicken anti-hemiferrin in PBS with Tween 20 (PBS-T) for 3 h at room temperature. After three 10-min washes in PBS-T, the blots were treated with peroxidase-conjugated goat anti-rabbit IgG (Sigma, St. Louis, MO) or peroxidase-conjugated rabbit anti-chicken IgY (Cappel). After three washes, the peroxidase was detected by enzyme chemiluminescence utilizing the Supersignal West Pico substrate (Pierce) and Sterling High Speed x-ray film (Midwest Scientific, St. Louis, MO).
Metabolic Labeling of Germ Cell Proteins
Rat germ cells were isolated from a 400-g rat and prepared for cell culture as described by Bellvé et al. [10] with the following modifications. The isolated germ cells were washed twice with Dulbecco modified Eagle medium (DMEM) lacking methionine, resuspended, and counted in a hemacytometer. Approximately 3 x 107 cells were incubated in 2.5 ml of DMEM lacking methionine and supplemented with 100 µCi 35S-methionine (New England Nuclear, Boston, MA) for 18 h at 32°C under 5% CO2. Proteins from both the cells and spent medium were analyzed by immunoprecipitation and SDS-PAGE. The procedure was repeated with two additional rats.
Germ cell proteins were also labeled metabolically in vivo. A 500-g rat was anesthetized, and the left testis was exposed through an incision in the scrotum. The testis was injected centrally with 500 µCi of 35S-methionine in 100 µl of PBS slowly over a 20-sec period. The rat was kept warm and remained anesthetized for 1 h. Germ cells were then isolated from the injected testis as described above and analyzed by immunoprecipitation and SDS-PAGE without being placed in culture. The procedure was repeated with a second rat.
Immunoprecipitation
Spent medium was centrifuged at 14 000 x g for 15 min at 4°C, and 1 ml of the supernatant was immunoprecipitated. Approximately 50 µl of packed cells from germ cell isolations or cell cultures was lysed in five volumes of 1% Triton X-100 in 0.15 M NaCl, 0.005 M EDTA, and 0.01 M Tris, pH 7.4 (NETT) for 5 min on ice with occasional mixing. Zysorbin (10% solution; Zymed, South San Francisco, CA) was washed with five volumes of NETT, collected by centrifugation at 14 000 x g for 20 sec, and resuspended in NETT to the original volume. Spent medium or cells in lysis solution was incubated with 25 µl of the washed Zysorbin for 1 h at 4°C with gentle rocking and then centrifuged at 14 000 x g for 20 sec. The supernatants were transferred to new tubes and incubated with rabbit anti-rat transferrin and 25 µl washed Zysorbin for 1 h at 4°C. The Zysorbin was then collected by centrifugation at 14 000 x g for 20 sec and resuspended in 100 µl NETT. The suspension was then layered onto 1 ml of 1% Triton X-100, 1% sodium deoxycholate, 0.1% SDS, and 15% sucrose in 0.15 M NaCl, 0.003 M sodium azide, and 0.05 M Tris (pH 8.5) in a microfuge tube and centrifuged at 14 000 x g for 20 sec. The supernatant was discarded, and the pelleted Zysorbin was resuspended in 30 µl of 4% SDS, 20% glycerol, and 10% 2-mercaptoethanol in 0.125 M Tris (pH 6.8). After heating to 95°C for 5 min, the tubes were centrifuged again, and the upper 20 µl was submitted to electrophoresis on a 12% polyacrylamide gel. The gels containing 35S-labeled proteins were dried in cellophane and subjected to PhosphorImager analysis (Molecular Dynamics, Sunnyvale, CA).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
When rat testis RNA was analyzed by Northern blot with the 3'rTf probe, two distinct bands appeared, transferrin (2.6 kb) and a 0.9-kb transcript, whereas when RNA from isolated germ cells was analyzed only the lower band appeared (Fig. 1). The 0.9-kb transcript is more abundant than the 2.6-kb transferrin mRNA in rat testis. Similar results were observed when the blots were probed with hemiferrin, although the bands were routinely weaker (data not shown). Under routine conditions, no smaller hemiferrin-like transcript was observed in Northern analysis of bovine liver or testis. However, after prolonged exposure of blots, a weak band of 0.9 kb was observed in bovine testis RNA using the 3'rTf probe but not the hemiferrin probe (Fig. 2). Transferrin mRNA levels are routinely lower in bull testes than in rat [9]. Despite the streaking observed in the bovine liver lanes with overexposure, no recognizable band appeared near 0.9 kb in liver samples with either probe, as assessed by digital scanning of the film.
|
|
Ribonuclease Protection Assay
Antisense cRNA probes were synthesized and used in a ribonuclease protection assay of rat and bovine testis and liver RNA. The 3'rTf antisense probe was protected by transcripts present in both rat liver and rat testis but was not protected by bovine transcripts (Fig. 3). Conversely, the hemiferrin antisense probe was protected by bovine liver transcripts but not by transcripts in rat tissues. Prolonged exposure of this gel revealed a weak band indicating hemiferrin probe protection by transcripts present in bovine testis similar in size to that in the bovine liver sample (data not shown). The intensity difference in rat liver and testis lanes is likely due to a difference in transcript copy number per cell [7].
|
Rat Testis cDNA Library Screening
One of the clones selected from the 5'-Stretch Plus cDNA library with the 3'rTF probe was of the approximate size expected for hemiferrin or a truncated 3' transferrin, less a poly(A) sequence. DNA sequencing of this clone revealed a fragment of 730 bp sharing identity with bases 1541–2271 of the rat transferrin cDNA (GenBank no. NM017055) [11]. This sequence would code for the C-terminal 187 amino acids of transferrin if it were translated but did not include any methionine codons. The nearest methionine in the open reading frame of transferrin that could serve as a potential start codon was upstream of the sequence and would give a 213-amino acid peptide of 24 kDa.
Western Analysis
The chicken anti-hemiferrin was of very high titer when tested against the expressed hemiferrin protein. It cross-reacted with commercial rat transferrin and with a 78-kDa protein in rat testis on Western blots but with no proteins in germ cell extracts (Fig. 4). Rabbit anti-rat transferrin (Cappel) did not cross-react with bacterially expressed hemiferrin nor did it detect any bands in rat germ cell extracts.
|
Immunoprecipitation of Metabolically Labeled Proteins
In an effort to detect a smaller transferrin-like protein, rat germ cells were isolated and placed in culture in the presence of 35S-methionine. PAGE of anti-transferrin immunoprecipitations of the spent medium and extracted germ cells revealed a number of minor bands but none in the area of a potential translation product from a 0.9-kb transcript. To assure that culture conditions did not have an effect on translation of the transcript, metabolic labeling was conducted in vivo in anesthetized rats. One hour after injecting a testis with 35S-methionine, the germ cells were isolated and extracted. PAGE of anti-transferrin immunoprecipitates of the extracts revealed only very weak bands that migrated in the approximate area of a predicted protein (Fig. 5). The dark band below 21 kDa was deemed to be free methionine, because more extensive washes of the immunoprecipitate on Zysorbin eliminated this band (data not shown).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The ribonuclease protection assays showed that a hemiferrin sequence was not detectable in rat tissues, whereas the 3'rTf probe was protected by sequences present in rat testis and liver. The fact that the 3'rTf probe was both protected in ribonuclease protection assays of testis transcripts and bound strongly to the 0.9-kb band in Northern analysis of rat testis and isolated germ cells suggests that the 0.9-kb band is a 3'-truncated form of transferrin. In support of this contention was the selection of a 730-bp clone from the rat testis library that was identical to the 3' region of rat transferrin. Because the 0.9-kb transcript is more abundant than the 2.6-kb transcript in whole testis RNA, this clone is a likely candidate. However, the clone could have arisen from a full-length transferrin transcript that was shortened during production of the library rather than occurring naturally. Maguire and coworkers [13] found that a 5' rat transferrin cDNA probe did not hybridize with the 0.9-kb band in Northern analysis of whole testis. All of the evidence presented here suggests that the orignal cloning of rat hemiferrin was in error and probably the result of contamination with bovine transcripts. The slight deviation from identity between the bovine 3' transferrin and hemiferrrin (sequenced manually) was probably due to sequencing errors.
Stallard and coworkers [1] presented evidence that a small proportion of the 0.9-kb transcript was associated with macromolecular structures that behaved like polysomes. This finding was consistent with what is known of many germ cell messages that are stored for a time before being recruited to polysomes [14]. However, no shortening of the hemiferrin message was noted when it was associated with the putative polysomes, as is noted with some other haploid expressed genes [15]. Despite this apparent association with translational complexes, we were unable to detect a protein product of the predicted molecular weight that could be identified by antibodies to rat transferrin (the whole molecule) or by antibodies generated against the bacterially expressed 3' bovine sequence (hemiferrin). We were also unable to detect an in vitro translation product of germ cell mRNA (data not shown).
In a previous study, 55Fe bound to human transferrin and incubated with rat seminiferous tubules became associated with a single band in urea/polyacrylamide gel analysis of anti-rat transferrin immunoprecipitates of whole testis homogenates [16]. This band was consistent with a full-length transferrin protein. However, because migration of proteins in urea/polyacrylamide is somewhat unpredictable, we cannot state conclusively that a half-transferrin would not comigrate. With an earlier lot of rabbit anti-rat transferrin, immunoprecipitates of 35S testicular proteins labeled in vivo produced a strong band for transferrin but negligible bands below 40 kDa [16]. These results suggest that little or no half-transferrin protein is made in germ cells.
There are examples of foreshortened haploid transcripts wherein a protein has been identified (see [17] and references therein). However, others have noted the appearance of foreshortened transcripts in the testis and have negated any useful contribution because of the lack of evidence for the corresponding translated protein [18, 19]. Until such time as a half-transferrin-like protein is demonstrated, the contribution of the 0.9-kb transcript to spermatogenesis will remain unknown.
|
FOOTNOTES |
|---|
1 This work was supported by NIH grant HD37180. 
2 Correspondence: Steven R. Sylvester, Molecular Biosciences, Washington State University, 14204 NE Salmon Creek Ave., Vancouver, WA 98686. FAX: 360-546-9064; sylveste{at}vancouver.wsu.edu
Accepted: April 15, 2002.
Received: July 30, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
