Patterns of expression of sperm flagellar genes: early expression of genes encoding axonemal proteins during the spermatogenic cycle and shared features of promoters of genes encoding central apparatus proteins*
1Center for Research on Reproduction and Women's Health and 2Center for Clinical Epidemiology and Biostatistics, University of Pennsylvania Medical Center, Philadelphia, PA 19104, USA
3 To whom correspondence should be addressed at: Center for Research on Reproduction and Women's Health, 1354 BRB II/III, 421 Curie Blvd, Philadelphia, PA 19104, USA. E-mail: jfs3{at}mail.med.upenn.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Sperm are motile cells. Thus, a significant component of the spermatogenic cycle is devoted to the formation of flagellum, a process that must be coordinated to insure proper construction. To document the temporal pattern of flagellar gene expression, we employed real-time PCR to assess changes in accumulation of a cohort of genes encoding axoneme, outer dense fibre (ODF) and fibrous sheath (FS) proteins during the first wave of spermatogenesis in the mouse. Axoneme genes were expressed first at the pachytene spermatocyte stage, followed by expression of transcripts encoding ODF and FS components. However, there were differences among these families with respect to the time of initial expression and the rate of mRNA accumulation. To gain understanding of factors that determine these patterns of expression, we cloned the promoters of three axoneme central apparatus genes (Pf6, Spag6 and Pf20). These promoters shared common features including the absence of a TATA box, and putative binding sites for several factors implicated in spermatogenesis (CREB/CREM, SOX17 and SPZ1) as well as ciliogenesis (FOXJ1). Collectively, our findings demonstrate a sequential pattern of expression of flagellar component genes, differential times of expression or rates of transcript accumulation within each class and shared promoter features within a class.
Key words: Mouse/PF6/PF20/SPAG6
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Spermatogenesis encompasses a complex and tightly coordinated differentiation process leading to the generation of specialized motile haploid cells. Sperm motility is essential for fertilization in vivo, and the later stages of the spermatogenic cycle are devoted in large part to the remodelling of the germ cell cytoplasm and the formation of the flagellum, which is composed of a number of cytoskeletal elements whose proper assembly is critical for subsequent motility. Abnormalities in genes encoding flagellar proteins can cause infertility (e.g. primary ciliary dyskinesia) and it is, therefore, of interest to understand the regulation of their expression (Cowan et al., 2001
; Chodhari et al., 2004).
The axoneme, the motor driving the flagellar beat, extends continuously throughout the length of the flagellum and is composed of a ‘9+2’ array of microtubules and associated proteins (Fawcett, 1975; Oko et al., 1990). The axoneme is surrounded by accessory structures in discrete subregions of the sperm tail. Nine outer dense fibres (ODF) surround the axoneme in the midpiece of the flagellum and are paired with the nine peripheral doublet microtubules, forming a ‘9+9+2’ pattern. The mitochondrial sheath is helically wound around the ODF under the plasma membrane. Although the ODF extend through the principle piece of the tail, fibres 3 and 8 are replaced by the dorsal and ventral longitudinal columns of the fibrous sheath (FS) in this region. The columns are connected by numerous regularly spaced transverse ribs (Fawcett, 1975; Oko et al., 1990). Toward the distal end of the flagellum, the ODF and the FS taper off, demarcating the endpiece of the tail. Proteins comprising both the ODF and the FS have been identified.
We have been interested in the genes that encode axonemal proteins, particularly those in the central apparatus, which is essential for normal flagellar function. The central apparatus is composed of two microtubules with attached projections and bridges that link the two microtubules (Goodenough and Heuser, 1985; Smith and Lefebvre, 1997). Mutagenesis studies revealed that several genes (e.g. PF6, PF15, PF16 and PF20) encoding central apparatus proteins in the green algae Chlamydomonas reinhardtii are essential for flagellar motility, since their inactivation results in flagellar paralysis. It is notable that the two central tubules are structurally and biochemically dimorphic and by convention are given separate designations (e.g. C1 and C2). At least 10 different polypeptides are uniquely associated with the C1 microtubule (e.g. PF6 and PF16) and seven are unique to the C2 microtubule (e.g. PF20) (Dutcher et al., 1984). This biochemical and structural asymmetry is believed to have functional significance with respect to the flagellar beat and waveform (Wargo and Smith, 2003).
We have identified the mammalian orthologues of Chlamydomonas PF6, PF16 (SPAG6) and PF20, and discovered that both SPAG6 and PF20 are essential for normal sperm motility (Spag6) and spermatogenesis (Pf20) in the mouse through targeted mutagenesis (Sapiro et al., 2002; Zhang et al., 2002). Moreover, we have also found that the PF6, SPAG6 and PF20 proteins form an interaction network, which presumably links the central apparatus microtubules (C1 and C2) and their projections (the 1a projection) into a functional unit (Z. Zhang and J.F. Strauss III, unpublished observations). In order to place the patterns of expression of these genes into the larger context of assembly of the mammalian sperm flagellum, we performed a quantitative analysis of expression of these axonemal protein genes and the ODF and FS genes in the mouse testis during the first wave of spermatogenesis using real-time quantitative PCR (QPCR). We also isolated the promoters of the murine Pf6, Spag6 and Pf20 genes and examined them for shared features, which might account for the coordinated patterns of expression revealed in our quantitative analysis of transcript abundance.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Animals
Testes, lung, oviduct, uterus, brain, heart, liver, spleen, kidney and seminal vesicle were collected from C57Bl6/J mice (Charles River Laboratories, Wilmington, MA). Likewise, day 6, 8, 12, 16, 20, 30 and 42 (adult) testes were collected from C57B16/J male mice. CREM-deficient and Spag6-deficient mice and their wild-type littermates have been previously described (Blendy et al., 1996
; Sapiro et al., 2002).
RNA extraction, RT and QPCR
Total RNA was isolated from whole tissues (testes, lung, oviduct, uterus, brain, heart, liver, spleen, kidney and seminal vesicle) using TRIzol reagent (Invitrogen, Carlsbad, CA). Round spermatids were isolated from mixed germ cells using the Staput method as previously described (Travis et al., 1998) for RNA extraction.
RT and QPCR were carried out as previously described (Wood et al., 2003). Briefly, total RNA (3 µg) was pretreated with DNase I (Promega, Madison, WI) and reverse transcribed with Moloney murine leukaemia virus (Promega) in the presence of random primers (Roche Diagnostics Corporation, Indianapolis, IN). Primers were designed for detection of axonemal, ODF and FS genes (Table I) using the Primer Express 2.0 software (PE Applied Biosystems, Foster City, CA) and tested empirically for the optimal primer concentration for detection of specific amplicons. QPCR reactions were carried out using equivalent dilutions of each cDNA sample and the 2x SYBR green master mix (PE Applied Biosystems). To account for differences in starting material, QPCR was also carried out for each cDNA sample using the Applied Biosystems 18S rRNA 20x primer and probe reagent (PE Applied Biosystems). In order to define the relative abundance of each transcript in each experimental group, analysis of the resulting QPCR reactions was carried out as previously described (Wood et al., 2003). Briefly, the threshold cycle for each gene target and 18S rRNA in each cDNA sample was converted to an arbitrary value using a standard curve generated from serial dilutions of a sample containing high levels of the target transcript. The relative abundance of the target was divided by the relative abundance of 18S rRNA in each sample to generate a normalized abundance for each transcript in each sample.
|
Northern blot analysis
Total testicular RNA (20 µg) from wild-type and CREM-deficient mice was separated on a denaturing agarose gel transferred to Hyclone N+ membrane (Amersham Biosciences, Piscataway, NJ), and probed for CREM, ODF1, ODF2 and ODF3 RNA as previously described (Zhang et al., 2004
). The cDNA probes for CREM (5'-GAGGACAAATGTAAGGCAAAT-3' and 5'-GAGGGCCTTGAGTTCCTCAAT-3'), ODF1 (5'-GGTACTCACAGAACAATAAGG-3' and 5'-CCTACAGGAGAATCGGCTTCC-3'), ODF2 (5'-CAGAGCCTGCCGACCCAATTGCCC-3' and 5'-CTGCCTTGTTAAGGTGTTGATGTC-3') and ODF3 (5'-GCTGCTCTAAGCCCATTCCTG-3' and 5'-CCACATCGACAACCAAGG GTG-3') were generated by PCR using the indicated primers.
Statistical analysis
Linear regression models were fit for each of the genes investigated as a function of mouse age and the relationship between relative abundance and day of sacrifice was assessed. A cubic model in time (day) was used for all models for comparability across models. The model for expression was:
 |
Model fit was assessed by graphic visualization and comparison of model adjusted R2 values. Ninety-five percent confidence boundaries about the predicted curves (dashed lines) were estimated for each predicted curve (solid line). Given the relative abundance that was 50% of the predicted maximum (horizontal line on graph), Y50, an estimate of the day this occurred, day50, was determined using Proc Model in SAS version 8.2 (SAS Institute Inc., Cary, NC, 2003). The solution was restricted to lie within the range of the observed data, 0
day42. The confidence interval (CI) about day50 was computed in a similar fashion using the lower and upper boundaries about the Y50 value (dashed lines).
Analysis of variance (ANOVA) methods were also employed to compare average expression levels at different ages. Pair wise comparisons between the groups were carried out using the Duncan's multiple range test. All expression levels were transformed to log(Y+1), where log signifies the natural log transformation. This was done to equalize the estimated variances across the groups.
Differences in transcript abundance between the CREM-deficient mice and the wild-type mouse testis were evaluated using t-test. We considered a P value of <0.05 to be significant.
5'Rapid amplification of cDNA ends (5'-RACE)
5'-RACE was carried out to define the 5'untranslated region sequence and transcriptional start site of the mouse PF20, PF6 and SPAG6 mRNAs using mouse testis poly (A)+ RNA and the Marathon cDNA amplification kit (Clontech, Palo Alto, CA) according to the manufacturers' instructions. Briefly, a primer was designed within the coding sequences of PF20 (5'-AGAAGCCACGAAGTCACCACAGGAGT-3'), PF6 (5'-ATGACACAATACTGAAGAGTTTCCGC-3'), or SPAG6 (5'-GCTGCAGTCTGTTGAATTGTTGGGAC-3') and used together with the Marathon cDNA adaptor primer to generate 5'-RACE products. The 5'-RACE products were cloned into the pCR2.1-TOPO TA vector and subjected to DNA sequence analysis.
Promoter constructs and transient transfections
Varying lengths of the Pf6 (2, 1.4 and 0.5 kb), Pf20 (2, 1.2 and 0.5 kb) and Spag6 (2, 1 and 0.5 kb) promoters identified based on the 5'-RACE data and genomic sequences in public databases were amplified by PCR and cloned into the pGL3 basic vector (Promega). Putative transcription binding sites in these promoter sequences were identified using the ConSite program (http://mordor.cgb.ki.se/cgi-bin/CONSITE/consite/).
The activity of each promoter sequence was assessed using Madin Darby canine kidney (MDCK) cells (ATCC, Manassas, VA) and BEAS-2B human transformed bronchial epithelial cells (ATCC), which are cilia-bearing cells. Cells were maintained in Dulbecco's modified Eagle's medium (Invitrogen, Carlsbad, CA) with 10% bovine serum. About 24 h prior to transfection, 30 000 cells were plated in 12 well plates. Triplicate wells were transfected with 500 ng of the Pf6-2.0, Pf6-1.4, Pf6-0.5, Pf20-2.0, Pf20-1.2, Pf20-0.5, Spag6-2.0, Spag6-1.0, or Spag6-0.5 vector using Fugene 6 (Roche Diagnostics Corporation, Indianapolis, IN). In addition, 25 ng of the Renilla luciferase plasmid was transfected into each well to evaluate transfection efficiency. After 48 h, the cells were harvested and the promoter activity was measured with the Dual Luciferase Reporter Assay System (Promega, Madison, WI).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Temporal pattern of expression of genes encoding flagellar proteins
Genes encoding axoneme and accessory structure proteins are expressed in ciliated cells including sperm. Thus, to define the tissue specific expression of several of these genes including SPAG6, PF6, PF20, kinesin-like protein 1 (KLP1), dynein heavy chain 7 (DNAH7), ODF protein (ODF) 1, ODF2, ODF3, protein kinase A associated protein (AKAP) 3, and AKAP4 in germ cells (round spermatids) and testis, QPCR was carried out. In addition, the expression of these genes in the organs containing cilia-bearing cells (e.g. lung, oviduct, uterus and brain) was also measured. The mRNA levels of the axonemal central apparatus proteins PF6, SPAG6, PF20 and DNAH7, which is associated with the inner dynein arms of the nine outer doublet microtubules, were most prominent in round spermatids and testis with lesser expression in lung, oviduct, uterus, brain and in few cases liver (Figure 1A). Transcript levels were undetectable in spleen, kidney and seminal vesicle. The expression of the central apparatus protein KLP1 (Yokoyama et al., 2004
) was also high in the testis and round spermatids (Figure 1A). However, negligible levels of KLP1 mRNA were found in lung, oviduct and uterus while gene expression was detected in brain, liver and kidney. The mRNAs for the sperm flagellum accessory proteins (ODF1, ODF2, ODF3, AKAP3 and AKAP4) were primarily detected in the testis and round spermatids (Figure 1B and C).
|
Temporal expression of the genes encoding the axoneme and accessory structure proteins is likely coordinated to ensure the appropriate construction of a functional flagellum sperm. To test this hypothesis, the temporal pattern of flagellar gene expression during spermatogenesis was determined using QPCR. Total RNA was collected from mouse testis on day 6 (seminiferous epithelium containing only type A spermatogonia), day 8 (type B spermatogonia appear), day 10 (meiotic prophase starts), day 14 (early pachytene spermatocytes), day 18 (late pachytene spermatocytes), day 20 (haploid round spermatids), day 30 (condensing spermatids) and day 42 (adult testis) (Steiner et al., 1999
). The RNA was reverse transcribed and subjected to QPCR for the PF6, PF20, SPAG6, KLP1, DNAH7, ODF1, ODF2, ODF3, AKAP3 and AKAP4 mRNAs. The first day that a statistically significant increase in transcript level was detected was determined by ANOVA, and the day at which transcript abundance reached 50% of the adult level was determined by using a cubic model in time (day) for each gene (Figure 2). This provides a measure of the initiation of gene expression in the spermatogenic cycle and an index of how rapidly transcripts accumulate relative to the initiation of gene expression, reflecting both the rate of transcription and the rate of transcript degradation. A measure of variation among the triplicate samples for each time point was also obtained, which provides an index of how tightly the accumulation of transcripts is controlled.
|
Transcripts encoding the axonemal proteins, DNAH7 and KLP1, exhibited a significant increase in abundance on day 12, which is prior to the formation of pachytene spermatocytes while the mRNA levels of PF6, PF20 and SPAG6 increased on day 16 when pachytene spermatocytes are present (Figure 2A, Table II). These patterns of axonemal gene expression were consistent with initiation of transcription during meiosis. The mRNA levels of ODF1 and ODF3 and the FS proteins AKAP3 and AKAP4 increased on day 20 when round spermatids are present (Figure 2B and C, Table II). This pattern of expression is consistent with post-meiotic gene expression. Interestingly, ODF2 transcript levels increased on day 16, suggesting that the Odf2 gene is expressed either very early after meiosis is completed or at the end of meiosis (Figure 2B, Table II). When the day of half maximal transcript abundance Y50) was examined for each gene, there was a direct relation between the Y50 and the first day of elevated transcript levels (Table II). For example, the Y50 for DNAH7 was 20.32 days while the Y50 for ODF1 was 29.42 days. Furthermore, there was little variation in the transcript levels detected between triplicate samples indicating that the expression of these genes is tightly regulated (Figure 2). However, there were significant differences in the Y50 for genes that exhibited the same day of first transcript increase. For example, the Y50 for DNAH7 was 20.32 while the Y50 for KLP1 was 24.25 and the Y50 for SPAG6 (21.75 days) was less than the Y50 for PF6 and PF20 (25.55 and 23.88, respectively) suggesting differences in the accumulation rate of these transcripts.
|
To validate the temporal pattern of gene expression of the axonemal, ODF and FS proteins that was determined by QPCR, the mRNA abundance of PF6, PF20, SPAG6, KLP1, DNAH7, ODF1, ODF2, ODF3, AKAP3 and AKAP4 was assessed in mice with a stage-specific arrest in spermatogenesis. Male mice nullizygous for the transcription factor CREM exhibit normal spermatogenesis until the early round spermatid stage, at which time the spermatids undergo extensive apoptosis (Blendy et al., 1996
). Thus, we compared the transcript levels of the flagellar genes in the testis of CREM-deficient and wild-type mice using QPCR. The mRNA abundance of the axonemal genes was not significantly different between CREM knockout and wild-type mice (Table III). Conversely, ODF2 and ODF3 transcript levels were reduced approximately 50 and 33%, respectively, in the CREM-deficient mice while the ODF1, AKAP3 and AKAP4 mRNAs were reduced to negligible levels in the CREM knockout compared to wild-type mice. The differential expression of ODF1, ODF2 and ODF3 in the testis of CREM-deficient compared to wild-type mice was confirmed by Northern blot analysis (Figure 3). Collectively, these data suggested that Odf1, Akap3 and Akap4 gene expression is initiated after the germ cells are depleted in the CREM-deficient mice, while Odf2 and Odf3 gene expression takes place during the time of germ cell loss in the CREM knockout mouse.
|
|
To begin to assess the role of CREM in the regulation of flagellar protein gene expression, the mRNA abundance of activator of CREM in testis (ACT) that specifically regulates CREM activity in the testis was determined. ACT mRNA abundance was significantly decreased in the testis of CREM-deficient mice compared to their wild-type littermates. Conversely, transcript levels for the transcription factor FOXJ1, also known as HFH-4, which is essential for ciliogenesis (Chen et al., 1998
; Brody et al., 2000), was not different in the CREM knockout mice. When the mRNA abundance of the axonemal, ODF and FS genes was compared between the SPAG6-deficient mice and their wild-type littermates, only SPAG6 mRNA abundance was decreased (Table II) indicating that this axonemal protein does not influence the expression of other genes encoding flagellar proteins.
Transcriptional activity of the Pf6, Pf20 and Spag6 promoters
The transcript levels of PF6, PF20 and SPAG6 begin to increase on day 16 suggesting that the transcription of these three genes might be commonly regulated. To test this hypothesis, the sequences of the proximal promoters of the Pf6, Spag6 and Pf20 genes were cloned and their DNA sequences and compared using ConSite, a program that determines high probability binding sites for known transcription factors based on sequence analysis. Although there are two transcripts for the mouse Pf20 gene (PF20-long and PF20-short) (Zhang et al., 2004), the promoter for the PF20-long transcript, which encodes the protein that localizes to the central apparatus was analysed in this study. The Pf6, Spag6 and Pf20 promoters lacked a canonical TATA box and had multiple transcriptional start sites. Each promoter had one potential CREB/CREM binding site near the transcriptional start sites, and multiple binding sites for the testis-specific transcription factor spermatogenic leucine zipper protein (SPZ1), SOX17 and the winged helix/forkhead homologue family member, HNF-3ß (Figures 4–6). The presence of putative cis elements capable of binding CREB/CREM, SPZ1, SOX17 and A/T-rich sequences that coincide with binding sites for members of the winged helix/forkhead family of transcription factors is consistent with the fact that these transcription factor families are implicated in spermatogenesis (CREM, SPZ1 and SOX17) and ciliogenesis (FOXJ1). However, the fact that Pf6, Spag6 and Pf20 gene expression was not altered in the testis of CREM-deficient mice indicates that CREM is not an essential transcription factor governing these genes. In addition to the shared repertoire of putative transcription factor binding sites, there were transcription factor binding sites specific to each promoter that could confer transcriptional specificity to each promoter. For example, the promoters for Spag6 and Pf20 but not Pf6 contained binding sites for Hunchback, which is a transcription factor that controls morphogenesis in Drosophilia melanogaster. Likewise, only the Pf6 promoter has a putative androgen receptor binding site, while only the Spag6 promoter contains a putative NF-
B binding site.
|
|
|
The promoter sequences for Pf6, Pf20 and Spag6 were cloned into the pGL3 reporter vector and transfected into cilia-bearing cells. There are no male germ cell lines that could be used for these assays. Thus, two somatic cell lines, MDCK cells and BEAS-2B human transformed bronchial epithelial cells, which are both cilia-bearing cells and, therefore, likely contain the repertoire of transcription factors required to activate transcription of the Pf6, Pf20 and Spag6 promoters were used for the transfection experiments. When the reporter vectors containing a portion of the Pf6, Pf20, or Spag6 promoter sequence were transfected into the MDCK and BEAS-2B cells, the firefly luciferase activity was increased compared to the empty vector (pGL3) (Figure 7). While the activity of the shorter fragments of each promoters exhibited increased activity compared to the longer promoter fragments in BEAS-2B cells, the different promoter fragments had similar activity levels in the MDCK cells. When the promoter fragments were cloned in the opposite orientation there was no increase in the firefly luciferase activity compared to the empty vector (data not shown). Because the MDCK and BEAS-2B cells were grown under standard culture conditions (i.e. the cells were not polarized), the level of promoter activity observed may well be underestimated since transcription factors driving axoneme gene expression may not have been expressed at optimal levels.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Spermatozoa are produced in a highly ordered sequence of events, which include mitotic, meiotic and post-meiotic events (Shima et al., 2004
). The assembly of the mammalian sperm flagellum is a multi-step process that has been defined at the ultrastructural level. An axoneme forms in round spermatids lacking accessory structures. The longitudinal columns of the FS first appear during step 2 and continue to be assembled in a distal-to-proximal direction relative to the sperm head over an extended period (through step 17 in the rat) (Irons and Clermont, 1982b). In contrast, the ribs appear during a shorter period of time in later stages of spermatogenesis (between steps 11 and 15). ODF formation begins later than FS assembly, starting at step 8 in the rat (Irons and Clermont, 1982b). Fibres first develop in association with the microtubular doublets and then thicken and enlarge. Moreover, unlike the FS, the ODFs are assembled in a proximal-to-distal direction.
This study is the first to report the temporal expression of a collection of genes involved in the formation of the sperm flagellum during spermatogenesis utilizing QPCR. This sensitive and quantitative technique allowed us to identify the time at which a significant increase in the expression occurred. Our findings demonstrate that the genes encoding axonemal proteins are generally expressed before the genes encoding accessory structure proteins which is consistent with the timing of assembly and previous reports of expression, which was assessed by Northern blot analysis (Zhang et al., 2002). However, there were some differences in the time of transcription increase and the rate of transcript accumulation within a cohort of genes encoding proteins incorporated into the specific flagellar structures. For example, transcripts for DNAH7 accumulated prior to those encoding the central apparatus proteins. Likewise, SPAG6 mRNA accumulated at a faster rate than transcripts for PF6 and PF20. These temporal differences in expression and rate of transcript accumulation may have relevance to the timing of assembly of flagellar structures. For example, SPAG6 interacts with both PF6 and PF20 (Zhang et al., 2002). Thus, the assembly of SPAG6 into the central apparatus may be necessary for the subsequent organization of PF6 and PF20. However, it should be recognized that there is not always a direct temporal relationship between transcription and translation of germ cell genes.
Conversely, we detected a significant rise in transcript abundance of AKAP3 and AKAP4 on day 20 without noteworthy differences between the two. However, AKAP3 is synthesized early in round spermatids and incorporated into the FS concomitantly with the rib precursor. AKAP4, which makes up almost 50% of the protein in the FS extracted from mouse sperm, is synthesized and incorporated into the nascent FS late in spermatid development (Brown et al., 2003). It is possible that both genes are expressed at the same time (early round spermatid stage), but that translation of AKAP3 mRNA occurs right after transcription while translation of AKAP4 message takes place at a later time (Brown et al., 2003).
Expression of the genes encoding ODF proteins was previously determined to be during the post-meiotic phase of spermatogenesis (Hoyer-Fender et al., 1998; Beissbarth et al., 2003). However, the increased sensitivity of the QPCR data revealed differential timing of ODF1, ODF2 and ODF3 expression. Specifically, transcription of the Odf2 gene was initiated by day 16, indicating expression at the pachytene spermatocyte stage while Odf1 and Odf3 gene expression was not initiated until day 20. Spermatogenesis in CREM-deficient mice terminates at the early round spermatid stage. Analysis of transcript abundance by QPCR and Northern blot analysis in the testis of wild-type and CREM-deficient mice suggested that the Odf2 gene is transcribed first followed by expression of the Odf3 and Odf1 genes at a later time point, which is consistent with the temporal pattern of expression determined by the original QPCR data. Taken together, these data suggest that the earlier timing of ODF2 expression may relate to a possible role in meiosis where its capacity to bind to centrosomal microtubules could be exploited (Donkor et al., 2004).
The genes encoding the axonemal central apparatus proteins PF20, PF6 and SPAG6 exhibited the same temporal pattern of gene expression. When the proximal promoters of these three genes were compared in silico, common transcription factor binding sites were defined. Furthermore, all three promoters were functionally active upon transient transfection into the cilia-bearing cell lines, MDCK and BEAS-2B. The fact that the Pf6, Spag6 and Pf20 promoters were functional in MDCK cells that produce a primary cilium lacking a central apparatus indicates that the transcriptional machinery, which drives expression of genes encoding the nine outer doublets and their associated proteins, can also contribute to expression of the genes encoding central apparatus components.
The promoters for Pf6, Pf20 and Spag6 contain consensus binding sites for transcription factors that are implicated in spermatogenesis including the CREB/CREM, SOX17, SPZ1 and HNF-3ß. Cyclic AMP response elements respond to members of basic leucine zipper transcription family, which includes CREM. However, the normal levels of PF6, SPAG6 and PF20 mRNAs in the testis of CREM-deficient mice rules out a major role for CREM in the transcriptional regulation of these genes. SPZ1 is a basic helix-loop-helix-leucine zipper transcription factor that binds to the consensus bHLH binding site (Hsu et al., 2001) and has been implicated in spermatogenesis from misexpression studies in mice (Hsu et al., 2004). SOX17, is an SRY-related protein that is highly expressed in testis and is thought to be a transcriptional activator in meiotic germ cells (Kanai et al., 1996). HNF-3ß is a member of the winged helix/forkhead gene family that includes the transcription factor FOXJ1 and, therefore, some of the putative HNF-3ß binding sites are similar to the motifs identified for FOXJ1 responsive promoters (Lim et al., 1997). Mice lacking FOXJ1 fail to develop cilia in all organs (Chen et al., 1998; Brody et al., 2000). The target genes for many of these transcription factors (e.g. SPZ1, SOX17 and FOXJ1) in male germ cells remain largely unknown. Our discovery of potential SPZ1, SOX17 and FOXJ1 response elements in genes encoding central apparatus protein provides a set of candidate genes that may be regulated by these transcription factors. Although the elements identified are only putative cis elements, the shared repertoire of putative transcription binding sites suggests that there may be a combinatorial code of factors that regulates the timing and rate of transcription of these central apparatus genes.
|
Acknowledgements |
|---|
This research was supported by NIH grants HD037416 and HD06724. We thank Dr. Julie Blendy for the CREM-deficient and wild-type littermate mouse testis.
|
Notes |
|---|
* DNA sequences described in this report have been deposited in GenBank (accession nos. AY742711 [GenBank] , AY742709 [GenBank] , AY792595 [GenBank] ).
Authors Eran Horowitz and Zhibing Zhang contributed equally to this work. 
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Beissbarth T, Borisevich I et al. (2003) Analysis of CREM-dependent gene expression during mouse spermatogenesis. Mol Cell Endocrinol 212(1–2), 29–39.[CrossRef][Web of Science][Medline]
Blendy JA, Kaestner KH et al. (1996) Severe impairment of spermato genesis in mice lacking the CREM gene. Nature 380(6570), 162–165.[CrossRef][Medline]
Brody SL, Yan XH et al. (2000) Ciliogenesis and left-right axis defects in forkhead factor HFH-4-null mice. Am J Respir Cell Mol Biol 23(1), 45–51.
Brown PR, Miki K et al. (2003) A-kinase anchoring protein 4 binding proteins in the fibrous sheath of the sperm flagellum. Biol Reprod 68(6), 2241–2248.
Chen J, Knowles HJ et al. (1998) Mutation of the mouse hepatocyte nuclear factor/forkhead homologue 4 gene results in an absence of cilia and random left–right asymmetry. J Clin Invest 102(6), 1077–1082.[Web of Science][Medline]
Chodhari R, Mitchison HM et al. (2004) Cilia, primary ciliary dyskinesia and molecular genetics. Paediatr Respir Rev 5(1), 69–76.[CrossRef][Medline]
Cowan MJ, Gladwin MT et al. (2001) Disorders of ciliary motility. Am J Med Sci 321(1), 3–10.[CrossRef][Web of Science][Medline]
Donkor FF, Monnich M et al. (2004) Outer dense fibre protein 2 (ODF2) is a self-interacting centrosomal protein with affinity for microtubules. J Cell Sci 117(Pt 20), 4643–4651.
Dutcher SK, Huang B et al. (1984) Genetic dissection of the central pair microtubules of the flagella of Chlamydomonas reinhardtii. J Cell Biol 98(1), 229–236.
Fawcett DW (1975) Gametogenesis in the male: prospects for its control. Symp Soc Dev Biol (33), 25–53.
Goodenough UW and Heuser JE (1985) Outer and inner dynein arms of cilia and flagella. Cell 41(2), 341–342.[Web of Science][Medline]
Hoyer-Fender S, Petersen C et al. (1998) Mouse Odf2 cDNAs consist of evolutionary conserved as well as highly variable sequences and encode outer dense fiber proteins of the sperm tail. Mol Reprod Dev 51(2), 167–175.[CrossRef][Web of Science][Medline]
Hsu SH, Shyu HW et al. (2001) Spz1, a novel bHLH-Zip protein, is specifically expressed in testis. Mech Dev 100(2), 177–187.[CrossRef][Web of Science][Medline]
Hsu SH, Hsieh-Li HM et al. (2004) Dysfunctional spermatogenesis in transgenic mice overexpressing bHLH-Zip transcription factor, Spz1. Exp Cell Res 294(1), 185–198.[CrossRef][Web of Science][Medline]
Irons MJ and Clermont Y (1982a) Formation of the outer dense fibers during spermiogenesis in the rat. Anat Rec 202(4), 463–471.[CrossRef][Medline]
Irons MJ and Clermont Y (1982b) Kinetics of fibrous sheath formation in the rat spermatid. Am J Anat 165(2), 121–130.[CrossRef][Web of Science][Medline]
Kanai Y, Kanai-Azuma M et al. (1996) Identification of two Sox17 messenger RNA isoforms, with and without the high mobility group box region, and their differential expression in mouse spermatogenesis. J Cell Biol 133(3), 667–681.
Lim L, Zhou H et al. (1997) The winged helix transcription factor HFH-4 is expressed during choroid plexus epithelial development in the mouse embryo. Proc Natl Acad Sci USA 94(7), 3094–3099.
Oko R, Moussakova L et al. (1990) Regional differences in composition of the perforatorium and outer periacrosomal layer of the rat spermatozoon as revealed by immunocytochemistry. Am J Anat 188(1), 64–73.[CrossRef][Web of Science][Medline]
Sapiro R, Kostetskii I et al. (2002) Male infertility, impaired sperm motility, and hydrocephalus in mice deficient in sperm-associated antigen 6. Mol Cell Biol 22(17), 6298–6305.
Shima JE, McLean DJ et al. (2004) The murine testicular transcriptome: characterizing gene expression in the testis during the progression of spermatogenesis. Biol Reprod 71(1), 319–330.
Smith EF and Lefebvre PA (1997) The role of central apparatus components in flagellar motility and microtubule assembly. Cell Motil Cytoskeleton 38(1), 1–8.[CrossRef][Web of Science][Medline]
Steiner R, Ever L et al. (1999) MEIG1 localizes to the nucleus and binds to meiotic chromosomes of spermatocytes as they initiate meiosis. Dev Biol 216(2), 635–645.[CrossRef][Web of Science][Medline]
Travis AJ, Foster JA et al. (1998) Targeting of a germ cell-specific type 1 hexokinase lacking a porin-binding domain to the mitochondria as well as to the head and fibrous sheath of murine spermatozoa. Mol Biol Cell 9(2), 263–276.
Wargo MJ and Smith EF (2003) Asymmetry of the central apparatus defines the location of active microtubule sliding in Chlamydomonas flagella. Proc Natl Acad Sci USA 100(1), 137–142.
Wood JR, Nelson VL et al. (2003) The molecular phenotype of polycystic ovary syndrome (PCOS) theca cells and new candidate PCOS genes defined by microarray analysis. J Biol Chem 278(29), 26380–26390.
Yokoyama R, O'Toole E et al. (2004) Regulation of flagellar dynein activity by a central pair kinesin. Proc Natl Acad Sci USA 101(50), 17398–17403.
Zhang Z, Sapiro R et al. (2002) A sperm-associated WD repeat protein orthologous to Chlamydomonas PF20 associates with Spag6, the mammalian orthologue of Chlamydomonas PF16. Mol Cell Biol 22(22), 7993–8004.
Zhang Z, Kostetskii I et al. (2004) Haploinsufficiency for the murine orthologue of Chlamydomonas PF20 disrupts spermatogenesis. Proc Natl Acad Sci USA 101(35), 12946–12951.
Submitted on December 10, 2004; accepted on February 7, 2005.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  Z. Zhang, X. Shen, D. R. Gude, B. M. Wilkinson, M. J. Justice, C. J. Flickinger, J. C. Herr, E. M. Eddy, and J. F. Strauss III MEIG1 is essential for spermiogenesis in mice PNAS, October 6, 2009; 106(40): 17055 - 17060. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Z. Zhang, X. Shen, B. H Jones, B. Xu, J. C Herr, and J. F Strauss III Phosphorylation of Mouse Sperm Axoneme Central Apparatus Protein SPAG16L by a Testis-Specific Kinase, TSSK2 Biol Reprod, July 1, 2008; 79(1): 75 - 83. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Assou, T. Anahory, V. Pantesco, T. Le Carrour, F. Pellestor, B. Klein, L. Reyftmann, H. Dechaud, J. De Vos, and S. Hamamah The human cumulus-oocyte complex gene-expression profile Hum. Reprod., July 1, 2006; 21(7): 1705 - 1719. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Z. Zhang, I. Kostetskii, W. Tang, L. Haig-Ladewig, R. Sapiro, Z. Wei, A. M. Patel, J. Bennett, G. L. Gerton, S. B. Moss, et al. Deficiency of SPAG16L Causes Male Infertility Associated with Impaired Sperm Motility Biol Reprod, April 1, 2006; 74(4): 751 - 759. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Sironen, B. Thomsen, M. Andersson, V. Ahola, and J. Vilkki An intronic insertion in KPL2 results in aberrant splicing and causes the immotile short-tail sperm defect in the pig PNAS, March 28, 2006; 103(13): 5006 - 5011. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. M. Bieber, L. Marcon, B. F. Hales, and B. Robaire Effects of Chemotherapeutic Agents for Testicular Cancer on the Male Rat Reproductive System, Spermatozoa, and Fertility J Androl, March 1, 2006; 27(2): 189 - 200. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||