Importin α mRNAs have distinct expression profiles during spermatogenesis

Authors

  • Cathryn A. Hogarth,

    1. The Monash Institute of Medical Research, Monash University, Clayton, Victoria, Australia
    2. The ARC Centre of Excellence in Biotechnology and Development, Monash University, Clayton, Victoria, Australia
    Search for more papers by this author
  • Sophina Calanni,

    1. Division for Biochemistry and Molecular Biology, Australian National University, Canberra, Australia
    Search for more papers by this author
  • David A. Jans,

    1. Department of Biochemistry and Molecular Biology, Monash University, Clayton, Victoria, Australia
    2. Division for Biochemistry and Molecular Biology, Australian National University, Canberra, Australia
    3. The ARC Centre of Excellence in Biotechnology and Development, Monash University, Clayton, Victoria, Australia
    Search for more papers by this author
  • Kate L. Loveland

    Corresponding author
    1. The Monash Institute of Medical Research, Monash University, Clayton, Victoria, Australia
    2. The ARC Centre of Excellence in Biotechnology and Development, Monash University, Clayton, Victoria, Australia
    • Monash Institute of Medical Research, 27-31 Wright Street, Clayton, Victoria 3168 Australia
    Search for more papers by this author

Abstract

Importin proteins control access to the cell nucleus by mediating the nuclear transport of specific cargoes. We hypothesized that developmental regulation of gene expression may be partially effected by changes in the nuclear transport machinery complement, manifested as regulated expression of importin α family genes. We first clarified the identity of the five known mouse importin α genes relative to those for human and then determined their expression throughout postnatal rodent testis using PCR and in situ hybridization. Distinct expression patterns were observed for each. At 10 dpp, all importin α mRNAs were detected in spermatogonia. In the adult mouse testis, importinsα1 and α3 were detected in spermatogonia and early pachytene spermatocytes. Importin α4 mRNA was identified in pachytene spermatocytes, α6 mRNA in round spermatids, and α2 mRNA in both of these. The distinct importin α expression patterns are consistent with their having specific roles and transport cargoes during spermatogenesis. Developmental Dynamics 235:253–262, 2006. © 2005 Wiley-Liss, Inc.

INTRODUCTION

To gain access to the nucleus, proteins >45 kDa in size must be actively transported through passageways in the nuclear envelope, the nuclear pore complexes. The importin α family of proteins mediates movement of specific cargo proteins when bound by importin β1, by recognizing and binding their nuclear localization sequences (NLSs) (Macara, 2001). This α/β/cargo trimeric complex moves into the nucleus via interactions between importin β1 and proteins lining the nuclear pore complex. Once inside the nucleus, the Ras family monomeric guanine nucleotide binding protein, Ran, in activated GTP-bound form, binds importin β1, dissociating the complex and thereby releasing importin α and its cargo.

In the mouse, there are five known importin α proteins arranged into 3 subgroups based on amino acid sequence identity (Table 1). Each importin α has distinct NLS binding specificities (Jans and Forwood, 2002), and is therefore assumed to transport different cargoes in vivo. Five of the human importin α proteins have each been shown to transport the two NLSs commonly used in import assays with different efficiencies: those of the simian virus 40 large-tumor antigen and the Xenopus laevis histone chaperone nucleoplasmin (Kohler et al., 1999). The transport of RCC1, a chromatin-bound Ran GTP/GDP exchange factor, is mediated by two importins of the same subclass, α3 and α4 (Kohler et al., 1999), whereas the nuclear import of lymphoid enhancer factor 1, a transcription factor, can be effected by importin α1 and α2, proteins from distinct subclasses (Prieve et al., 1998).

Table 1. The Importin α Family: Mouse and Human Comparisona
Subgroups of the importin α familyHuman importin (accession numbers and other names)Direct mouse importin homologue (accession numbers and other names)
  • a

    Accession numbers are italicized. Names used in this study are underlined. Names designated by Tsuji et al. (1997).

α-S, α1sImportin α5-NP 002255 Karyopherin α1, hSRP1Importin α1-Q60960 Srp1β, Rch2. NPI 1Importin α-S1
 Importin α7-AAC15233Importin α6-O35345 Importin α7Importin α-S2
 Importin α6-AAC51868No direct homologue, most similar to Importin α6
α-P, α2Importin α1-AAA69957 Srp1α, Karyopherin α2, Rch1Importin α2-P52293 Srp1α, Rch1, PendulinImportin α-P1
α-Q, α3sImportin α3-AAC2505 Qip1Importin α4-O35343 Qip1, Karyopherin α4Importin α-Q1
 Importin α4-NP_002258 hSRPIγImportin α3-O35344 Karyopherin α3Importin α-Q2

Studies of Drosophila melanogaster and Caenorhabditis elegans development have revealed vital roles for regulated importin expression in gametogenesis. Mutations in Drosophila importin α2 lead to a block in transport between nurse cells and the developing oocyte that causes female sterility (Gorjanacz et al., 2002). Onset of the heat shock response in very early Drosophila embryonic development (0–6 hr after egg deposition) is concordant with expression of importin α3, as this is required to transport the Drosophila heat shock factor into the nucleus (Fang et al., 2001). Importin α3 also plays a vital role in C. elegans gametogenesis, with mRNA knockdown studies demonstrating a requirement for this gene in mitotic germ cell development (Geles and Adam, 2001).

There is little published information about the developmental expression and function of importins in mammals. Messenger RNA and protein expression patterns for three members of the importin α family have been examined in a broad range of human and mouse tissues using Western and Northern blots and radioactive section in situ hybridization analysis (Kohler et al., 1997, 1999; Tsuji et al., 1997; Kamei et al., 1999). However, the diversity of names assigned to the same importin α family genes makes comparison of data unwieldy; to address this, we have compiled a list of these names, relating them to the current understanding of the human and mouse importin families.

Mammalian spermatogenesis provides an excellent opportunity to examine the regulation of importins in mammalian cellular differentiation and development. Germ cell development in the testis features three major processes: (1) mitotic divisions of spermatogonial stem cells and spermatogonia, (2) meiotic division of spermatocytes, and (3) morphological changes that transform the haploid, round spermatid into an elongated, motile mature spermatozoa, adapted to deliver its genetic material to the oocyte (de Kretser and Kerr, 1994).

This is the first study to provide a comprehensive analysis of all five known mouse importin α genes during cellular development, using spermatogenesis as the model system. The mRNA expression patterns were revealed using DIG-labeled riboprobes to precisely profile their cellular sites of synthesis from juvenile through adult mouse testis development, and this was further extended using RT-PCR to examine expression in male and female fetal gonads. The findings illustrate that individual importin α family members are present in male germ cells at distinct stages of maturation.

Abbreviations

ARM armadillo BHK baby hamster kidney cpp days post partum E embryonic day kb kilobases NLSs nuclear localization sequences RT reverse transcription

RESULTS

Nomenclature Clarification Within the Importin α Family

In mice and humans, multiple importin α genes encode functionally specialized family members (Kohler et al., 1999; Jans and Forwood, 2002). Malik and colleagues (1997) outlined the evolutionary relationships between importin αs across different species based on percentage sequence identity, defining two separate groups, the α1-Kaps and α2-Kaps. Another group separated the five mouse importin α genes into three subfamilies, α-P, α-Q, and α-S (Tsuji et al., 1997). Two more reports have assigned the six human importin α genes into three distinct subgroups: the α1s, α2s, and α3s (Kohler et al., 2002; Goldfarb et al., 2004).

What have not been delineated are the relationships between the mouse and human importin α homologues. The assignment of multiple names to each gene has hampered comparisons between studies in different systems. Human importin α1, for example, is also referred to as Srp1α and karyopherin α2; its direct sequence homologue in the mouse is importin α2. We applied the basic MGI Mouse Genome BLASTP tool (http://mouseblast.informatics.jax.org/) and the CLUSTALW tool (http://www.ebi.ac.uk/clustalw/) with a defined list of human importin α protein accession numbers to determine the direct homologue of each within the two species (Table 1). This table aligns different terminologies for the importin α proteins in mouse and human and, hence, creates a reference for researchers when comparing studies. Combining the approaches of Tsuji et al. (1997) and Goldfarb et al. (2004), we have organized the five known mouse importin proteins numbered 1 through 4 and 6 into the three groups based on the percentage sequence identity and identified their direct human homologue. Our intention in creating this table is not to propose a new importin α terminology, but to organize the multiple existing aliases mentioned in published studies and in various databases to enable the names of the mouse and human homologues to be paralleled.

All Importin α Genes Are Expressed in the Rodent Testis

Reverse transcription (RT) PCR was initially performed using cDNA from adult mouse testis to generate probes for subsequent mRNA analyses. Products from all 5 known genes were cloned and verified as identical to published sequences. Northern blot analysis was performed to determine whether all 5 known mouse importin genes could be readily detected in both the adult and juvenile testis. Membranes bearing adult rat, 10-dpp rat, and adult mouse total testis RNA samples were hybridized with each importin α probe. Transcripts encoding each of the 5 known importin α proteins were readily detected in the adult mouse testis (Fig. 1), while, in contrast, the relative expression levels in the 10-dpp rat testis appeared to be much lower for importin α2, α4, and α6. No apparent difference was observed in the RNA loading between samples using either GAPDH (Fig. 1) or ribosomal S26 (data not shown) as indicators.

Figure 1.

Northern blot analysis of importin α mRNAs in rodent testis. Total mRNA from adult rat, day-10 rat, and adult mouse testes was hybridized with importin α1, α2, α3, α4, and α6 cDNA/cRNA probes with GAPDH used as a loading control. Ad, Adult.

Several attempts were made to detect importin α1 and α3 transcripts by Northern blot in adult rat, 10-dpp rat, and adult mouse total testis RNA using [α-32P]dCTP-labeled probes. However, no signal could be visualized after exposure to Hyperfilm™ (Amersham Biosciences Pty. Ltd, Sydney, NSW, Australia). As an alternative approach, we used DIG-labeled cRNAs corresponding to these importins. In the adult mouse testis, importin α1 transcripts at 4.3 and 1.8 kb were detected (Fig. 1), the larger of these consistent with a previous report (Tsuji et al., 1997). We detected a 4.3-kb transcript for importin α3 in the adult mouse testis (Fig. 1), which agrees with previous data (Tsuji et al., 1997). No transcripts could be detected in the rat total testis RNA samples with the DIG-labeled cRNA probes (Fig. 1).

A specific band for importin α2 at a transcript size of 2.3 kb was detected in all samples (Fig. 1), which agrees with previous data (Tsuji et al., 1997). Northern analysis of importin α4 in the adult rodent testis detected a single transcript at 4.2 kb, matching one of the three transcripts reported by Tsuji et al., 1997. No importin α4 transcript could be detected in the 10-dpp rat total testis RNA sample (Fig. 1). A 2.8-kb band was identified for importin α6 in both the adult testis samples with a second band at 5.5 kb in the adult mouse testis sample (Fig. 1). No importin α6 transcripts were detected in the 10-dpp rat sample. The 5.5-kb transcript is consistent with previous data (Tsuji et al., 1997). However, this study also reported a 1.8-kb band, 1 kb smaller than the band we observed.

As we were able to detect transcripts for each known mouse importin α at sizes consistent with previous data in the rodent testis, we used these same probe sequences to generate in situ hybridization probes for all five known mouse importin α genes to assess the cellular expression patterns of their mRNAs in the mouse testis.

Importin α mRNA Expression Levels Vary Throughout the First Wave of Spermatogenesis

To examine the hypothesis that nuclear transport machinery is developmentally regulated in male germ cells, we first used approaches that examine the total testis RNA levels through the first wave of spermatogenesis, from the onset of spermatogenesis when the only germ cell type present is the quiescent gonocyte (0–0.5 dpp in the mouse; 0–5 dpp in the rat) through to adulthood, when all germ cell types are present from spermatogonial stem cells through to elongating spermatids (56 dpp in the mouse; 60–90 dpp in rat) (McCarrey, 1993). Changes in the level of a specific mRNA during the first wave of spermatogenesis can be related to the change in complement of germ cell types and to somatic cell maturation during this interval. To begin our survey of importin α expression in the rodent testis, we used real time PCR to investigate levels of mRNA production for two of the importin αs, α1 and α6, in total testis samples from rats of various ages. The analysis revealed that importin α1 mRNA levels were higher at 5 and 20 dpp than at 9 dpp and in adult samples, whereas importin α6 mRNA levels were higher at 9 dpp and in the adult (Fig. 2A). However, the number of importin α1 mRNA molecules detected in these rat samples was very low, at least 100-fold lower than the number of importin α6 molecules (Fig. 2A). This could explain our inability to detect transcripts in the 10-dpp and adult rat testis samples on the Northern blots for importin α1.

Figure 2.

Importin alpha expression throughout spermatogenesis. A: Real-time PCR used to determine the levels of importin α1 and α6 mRNAs in total testis RNA from Sprague-Dawley rats collected at 5 dpp (d5), 9 dpp (d9), 20 dpp (d20), and 60–90 dpp (adult; Ad). Data are presented as the number of mRNA copies (mean ± SD) per 108 copies of GAPDH mRNA. B: Comparison of importin α expression patterns throughout testis development. Arbitrary expression values recorded from the Affymetrix™ chip given on the Y axis, with the X axis depicting sample age. Asterisk (*), values for importin α2; filled circles, values for importin α4.

The disparate expression patterns exhibited by importin α1 and α6 are suggestive of distinct roles during spermatogenesis. These results imply importin α1 may play a role in pre-spermatogonial differentiation (4–6 dpp) and the late meiotic stages (15–26 dpp) with importin α6 important for early meiotic progression (9–12 dpp) and in the adult testis where all germ cell types are present.

For comparison with the mouse testis, we examined the GEOProfiles dataset GDS605, accessible via the NCBI website (http://www.ncbi.nlm.nih.gov/entrez/query.fcgi), for importin α gene expression patterns in mouse testis total RNA samples (normalized to internal negative controls) spanning from 0 dpp through to adult (Shima et al., 2004). These data offer the opportunity to identify trends in expression patterns and provide a comparison for data collected by other types of expression analysis. Using this GEOProfile dataset, we identified a dramatic change in expression levels for importin α2 throughout spermatogenesis. Between 0 and 10 dpp, importin α2 expression appears to remain relatively stable, increasing dramatically from 10 dpp through to adulthood, with Affymetrix values 3-fold higher at 56 than at 10 dpp (Fig. 2B). This pattern suggests that importin α2 is required for meiosis of the primary spermatocytes and in the haploid germ cells as these cell types begin to appear from this time point onwards. Importin α4 is also interesting as it appears to be expressed at very low levels until 20 dpp in the testis when expression increases sharply with increasing age (Fig. 2B). This expression pattern would be in accord with importin α4 being required in the late meiotic germ cells and the haploid spermatids.

Viewed together, these survey data that measure changes in the amounts of each mRNA indicate that each importin α would display specific expression patterns throughout spermatogenesis, and this could be dependent on the changing complement of cell types present in the testis as both the somatic and germ cells mature. To examine this, we performed in situ hybridization and RT-PCR on mouse testes from different ages to assemble expression patterns for all five known importin α genes throughout spermatogenesis. RT-PCR was performed on mouse testis total RNA from the time of gonad differentiation in the fetus, through adulthood. At E12.5, the expression of SRY, the sex determining gene, has established the gonad as a testis and by E15.5, the male gonocytes have multiplied and become quiescent. At 0 dpp, the quiescent gonocytes are about to resume mitosis, and by 5 dpp, these have been committed to differentiate, forming spermatogonial stem cells and differentiating spermatogonia. At 15 dpp, meiotic germ cells and spermatogonia are present, while in the adult, all germ cell types from spermatogonia through to spermatozoan are present and the sertoli cells are terminally differentiated. For in situ hybridization analysis, the 10-dpp time point was chosen based on the presence of spermatogonia and differentiating somatic cells.

Importin α1 and α6 Are Widely Expressed Throughout Testis Development

In situ hybridization was used to determine the cellular mRNA expression patterns of the importin αs in fixed sections of the juvenile and adult mouse testis and the results will be discussed in sections according to the 3 structural importin α Subgroups: the α1s (α1 and α6), α2, and the α3s (α3 and α4).

Importin α1 was expressed in spermatogonia through to the late meiotic pachytene spermatocytes of the adult mouse testis at all stages of the seminiferous tubules, indicating a widespread requirement for α1 protein in these cell types (Fig. 3A,H). There was no evidence to indicate Sertoli cell expression or expression outside of the seminiferous epithelium. The mRNA was also detected in spermatogonia at 10 dpp (Fig. 3D) but not in the Sertoli cells at this age.

Figure 3.

Differential expression of importin α1 and α6 in the juvenile and adult mouse testis. A,B: In situ hybridization of adult mouse testis with importin α1 (A) and α6 (B) antisense cRNAs. D,E: In situ hybridization of 10-dpp mouse testis with importin α1 (D) and α6 (E) antisense cRNAs. C,F:In situ hybridization of adult (C) and 10 dpp (F) with sense cRNA negative controls (importin α1 shown; identical results obtained for importin α6). sp, spermatogonia; ps, pachytene spermatocytes; rs, round spermatids. G: RT-PCR analysis for importin α1, α6 and β-actin (positive control) in E12.5 and E16.5 male (M) and female (F) gonads and testes from day 0, 5, 15, and Adult male Swiss mice. H: Adult mouse testis staging diagram, highlighted with colored bars behind and below cells in the relevant stages to indicate importin α1 and α6 mRNA expression patterns.

Importin α6, although structurally similar to importin α1, showed a very different expression pattern in the adult mouse testis. At this age, importin α6 mRNA was only detected in round spermatids and from Stages 1 through 8 (Fig. 3B,H). No importin α6 mRNA was detected in any of the diploid cell types, including the somatic cells, within the seminiferous epithelium at this age or in interstitial cells. At 10 dpp, it could only be detected as a weak signal in some spermatogonia (Fig. 3E) with no expression in the somatic cells. The results are in agreement with peaks of importin α6 expression at 9 and 56 dpp in the rat testis as determined by real-time PCR analysis (Fig. 2A) and are consistent with distinct roles for importin α6 during the first wave of spermatogenesis in the juvenile testis compared to the adult.

RT-PCR analysis of RNA from embryonic male and female gonads (E12.5 and E15.5) and from postnatal mouse testes at various ages (0 dpp, 5 dpp, 15 dpp, and adult) was performed, revealing the presence of transcripts at each age tested for both importin α1 and α6 (Fig. 3G).

Importin α2 mRNA Is Present in Both Diploid and Haploid Germ Cells

In situ hybridization revealed an mRNA expression pattern for importin α2 distinct from that of α1 and α6. Importin α2 mRNA first appeared at Stage 8 in the pachytene spermatocytes in the adult mouse testis, with expression extended through meiosis to haploid round spermatids in Stage 7 (Fig. 4A,E). There was no evidence of signal once the spermatids began to elongate or in somatic cells. At 10 dpp, importin α2 mRNA was detected in spermatogonia (Fig. 4C) but not in Sertoli cells. RT-PCR analysis with primers specific for importin α2 revealed transcript expression in the same samples that were tested for importin α1 and α6 (Fig. 4D).

Figure 4.

Differential expression of importin α2 in the juvenile and adult mouse testis. A–C: In situ hybridization of adult (A,B) and 10-dpp mouse testis (C) with importin α2 antisense and sense (as inset in A) cRNAs. sp, spermatogonia; ps, pachytene spermatocytes; rs, round spermatids. D: RT-PCR results for importin α2 and β-actin (positive control) in E12.5 and E16.5 male (M) and female gonads (F) and testes from day 0, 5, 15, and Adult male Swiss mice. E: Adult mouse testis staging diagram with highlighted importin α2 mRNA expression with a grey bar behind and beneath the cells in which it is expressed.

The Affymetrix™ GeneChip data displayed an interesting trend for importin α2 mRNA levels indicating a low level of expression in the juvenile testis, rising sharply in a linear fashion from 10 dpp through adulthood (56 dpp) (Fig. 2B). This suggests that the amount of importin α2 transcript in the testis begins to increase as the early spermatocytes appear and remains high as the haploid germ cells develop, in agreement with the in situ hybridization expression pattern. This indicates a specific production of importin α2 mRNA during the transition from diploid to haploid germ cells in both the juvenile and adult testis.

Importin α3 and α4 Expression in the Adult Mouse Testis

Importin α3 mRNA showed a cellular pattern of mRNA expression similar to that of importin α1 in the adult mouse testis with a faint signal apparent in the mitotic and early meiotic germ cells (Fig. 5A,H) that was not detectable in the Sertoli cells. At 10 dpp, both spermatogonia and Sertoli cells were weakly positive for importin α3 (Fig. 5D). Hence, importin α3 appears to be expressed in the mitotic germ cell populations throughout both the first wave of spermatogenesis and in the adult mouse testis. RT-PCR analysis detected importin α3 expression at each age tested in both the fetal gonad and postnatal testis (Fig. 5G).

Figure 5.

Differential expression of importin α3 and α4 in the juvenile and adult mouse testis. A,B: In situ hybridization of adult mouse testis with importin α3 (A) and α4 (B) antisense cRNAs. D,E: In situ hybridization of 10-dpp mouse testis with importin α3 (D) and α4 (E) antisense cRNAs. C,F: In situ hybridization of adult (C) and 10 dpp (F) with sense cRNA negative controls (importin α4 shown; identical results obtained for importin α3). sp, spermatogonia; ps, pachytene spermatocytes; rs, round spermatids. G: RT-PCR analysis for importin α3, α4 and β-actin (positive control) in E12.5 and E16.5 male (M) and female (F) gonads and testes from day 0, 5, 15, and adult male Swiss mice. H: Adult mouse testis staging diagram highlighted for importin α3 and α4 mRNA expression patterns using colored bars behind and below the cells in the relevant stages.

Importin α4 mRNA was readily detected by in situ hybridization in pachytene spermatocytes from stages 8–12, with very faint staining apparent in the early haploid germ cells in Stages 1–3 of the adult mouse testis (Fig. 5B,H). At 10 dpp, importin α4 mRNA was detected in spermatogonia (Fig. 5E). No importin α4 expression could be detected in the somatic cells at either age examined.

Comparison of the RT-PCR data for importin α4 (Fig. 5G) with the Affymetrix™ Gene Chip data (Fig. 2) revealed an excellent correlation. Both analyses show very low expression of importin α4 in the newborn testis, with signal increasing around 2 weeks post-partum and being most intense in the adult samples. The β-actin control for sample loading of all cDNA samples showed consistent loading between cDNA samples.

DISCUSSION

The importin α protein family mediates trafficking of nuclear factors from their site of production in the cytoplasm into the nucleus. Their essential roles in oogenesis, spindle assembly, and nuclear envelope formation is highlighted by findings in D. melanogaster and C. elegans (Askjaer et al., 2002; Gorjanacz et al., 2002; Mason et al., 2002), which also suggest a specificity of function and cargo transport amongst individual importin α proteins. The yeast genome contains only one importin α gene, SRP1, but in more complex organisms there are multiple family members, with three in Drosophila and C. elegans, five in the mouse, and six in the human. This diversity creates opportunities for both specificity and redundancy in cargo transport amongst members of the importin α family.

Until recently, numerous sequences have been recorded under various names for both mouse and human importin α family genes. A recent review by Goldfarb et al. (2004) included a simplified family tree and subdivided the importin αs into α1, α2, and α3 classes. The review also listed many of the names previously used for the different importin αs and roughly organized each of these names into one of the three groups. To extend this approach, we have taken the numerical naming system in the mouse for the importin α family and allocated each member to one of the three groups. In addition, we have aligned each human importin α with the mouse proteome to find its direct mouse homologue. Our results show that mouse α1 and α6 are the α1 subgroup, α3 and α4 are the α3 subgroup, and α2 is alone in the third group. This allocation can now be aligned with the three groups defined by Tsuji et al. (1997) as α-P, α-S, and α-Q as shown in Table 1. Importantly, this organization of the five known mouse importin α proteins (and their various aliases) into three defined groups has consolidated data from several reports and various databases into one simple nomenclature system that parallels the mouse and human importin α protein nomenclature.

Our analysis of the expression patterns of importin α mRNAs throughout spermatogenesis has revealed a stage- and cell-specific pattern for five in the adult mouse testis. This organ is comprised of one long convoluted tubule built from twelve repeating segments containing distinct cell type patterns (Russell et al., 1990). Therefore, any given cross-section of a whole testis will present tubules containing germ cells at varying degrees of maturity. Using multiple methods, we have demonstrated for the first time that, at any given stage of the seminiferous epithelium, there are different importin αs expressed in distinct germ cell types. Importin α1 and α3 are the most widely expressed, with mRNAs detected in spermatogonia through to pachytene spermatocytes in all tubules. This suggests a general requirement for these two proteins in mitotic and meiotic germ cell populations. Whether the expression of both these importins in the same cell types indicates they carry distinct or overlapping cargo remains to be assessed.

The unique cellular expression profiles of importins α2, α4, and α6 at discrete stages in the adult testis suggests there may be roles for these importin αs that are independent of classical nuclear import. All three of these mRNAs were expressed either in spermatocytes or in haploid germ cells and not in mitotic cells. At the beginning of meiosis, the nuclear envelope breaks down to enable chromosome condensation and spindle assembly, so that no facilitated nuclear transport is required. It has previously been shown that importin α2, in addition to its role in nuclear import, can inhibit mitotic spindle formation by binding to TPX2, an assembly factor (Gruss et al., 2001; Schatz et al., 2003), and the C. elegans germ-line-specific importin α, IMA-2, has been shown to participate in nuclear envelope assembly (Geles et al., 2002). It is conceivable that importin αs in meiotic germ cells participate in spindle assembly and nuclear envelope dynamics.

Consistent with roles for the importin αs in processes distinct from classical import are our own results showing a lack of importin β1 mRNA and protein in haploid germ cells (Loveland et al., unpublished data). The classical role of importin α in nuclear import involves binding to both a cargo protein and to importin β1. Low levels of importin β1 in the haploid germ cells implies that the importin αs expressed in these cells, as shown here, may either mediate an alternative nuclear transport process or possess a function completely distinct to that of import. An importin α-alone mode of nuclear transport has been described in plants (Hubner et al., 1999) and mammalian importin α has recently been shown to target a Ca2+/calmodulin-dependent protein kinase to the nucleus in the baby hamster kidney (BKH) cell line independent of either importin β or Ran (Kotera et al., 2005). It is possible that in haploid germ cells, importin α functions as a nuclear import protein but through a pathway that is not yet well understood.

The observations presented in this study demonstrate that the individual importin α mRNA levels and cellular expression patterns are unique with respect to each other throughout testicular development and spermatogenesis (Fig. 6). This is consistent with each having a specific role and distinct cargo. Thus, the regulated expression of importin α gene expression may underpin the transitions from one developmental stage to the next, at least partly by governing access to the nucleus for specific transcription factors and nuclear proteins.

Figure 6.

Summary of importin α expression in the adult mouse testis. The three importin α family subgroups are depicted in different shades: α1s, black; α2, white; α3s, grey.

EXPERIMENTAL PROCEDURES

Human and Mouse Nomenclature Analysis

Using a list of human importin α protein accession numbers kindly supplied by the Laboratory of Bostjan Kobe (University of Queensland, Queensland, Australia), the mouse importin homologues were identified using the basic MGI Mouse Genome BLASTP tool (http://mouseblast.informatics.jax.org/). Percentage identities to determine direct human and mouse homologues were assigned using the CLUSTALW tool (http://www.ebi.ac.uk/clustalw/). Alternative names for each importin alpha were collated from the NCBI Protein database (http://www.ncbi.nlm.nih.gov/entrez/), the MGI Mouse Genome database (http://mouseblast.informatics.jax.org/), and from references (Kohler et al., 1999, 2002).

Animals and Tissues

Swiss mice ranging from birth (0 dpp) to adulthood (35–56 dpp) and Sprague-Dawley adult male rats were obtained from Monash University Central Animal Services. The animals were killed by decapitation (fetuses and 0–10 dpp) or cervical dissociation (10 dpp to adult) and their testes dissected. Fetal gonad tissues were collected from Swiss mice embryos staged by fore- and hindlimb morphology. All investigations conformed to the NHMRC/CSIRO/AAC Code of Practice for the Care and Use of Animals for Experimental Purposes and were approved by the Monash University Standing Committee on Ethics in Animal Experimentation. Tissue samples for RNA preparation were snap frozen immediately after collection and stored at −70°C until use. Tissues for in situ hybridization were placed in Bouin's fixative for 5 hr immediately after collection, then dehydrated through a graded ethanol series and embedded in paraffin. Sections of 3–5 μm were placed on Superfrost® Plus slides (Menzel-Glaser, Braunschweig, Germany).

RT-PCR

Total RNA was extracted from testes and fetal gonads using the TRIzol™ reagent as per the manufacturer's instructions (Invitrogen, Mount Waverley, VIC, Australia). cDNA was prepared using SuperScript™ III Reverse Transcriptase as per the manufacturer's instructions (Invitrogen) with 500 ng of each total RNA (concentration calculated on an Eppendorf BioPhotometer, North Ryde, Australia) as template. Primers were designed using Primer3 (http://frodo.wi.mit.edu/cgi-bin/primer3/primer3_www.cgi), to amplify 200–400-bp regions of each of the five mouse importin α genes. The regions selected for α1, α2, α3, and α6 all crossed over introns to eliminate genomic DNA amplification. Primers used (529–3′): α1–CCTGAGGCTTGGAGAACAAG and GCTGCTGGCTAAGATCAACC, α2–GTGCATCTCGTTCATTGTGG and TCCATCCTGAACTTGGAAGG, α3–TCAGGAGATTTTGCCAGCTT and GCTCTGAGTGCTGCAGTTTG, α4–AAAGGGCTACAGGTCACAGG and CCACATACAGCACCACAACC, and α6–GAGAACATTCTTCGGCTTGG and CCGGAGGCAGACATTATAGC.

In Situ Hybridization

In situ hybridization was used to localize importin α mRNAs on mouse testis sections as previously described (Meehan et al., 2000). The PCR products derived using the primers outlined above were ligated into pGEMTEasy (Promega Corp, Madison, WI) and transformed into cells of the DH5α host strain. Positive colonies were identified via PCR, verified by sequencing, and the plasmids used as PCR templates to derive products using M13 forward and reverse primers (Sigma-Aldrich Pty. Ltd, Castle Hill, N.S.W, Australia) with the importin α probe region flanked by the T7 and Sp6 promoter sequences. These PCR products were used as the template for PCR-based production of DIG-cRNAs from the T7 and Sp6 promoters. Hybridization and washing were performed at 50°C and 55°C. Both antisense and sense (negative control) cRNAs were used on each sample, in every experiment, for each set of conditions tested.

Northern Blotting

Northern blots were prepared on Hybond XL membrane (Amersham Biosciences Pty. Ltd, Sydney, NSW, Australia) containing 20 μg of adult and 10 dpp rat and adult mouse total testis RNA according to the manufacturer's instructions. The PCR-derived cDNAs used to generate the DIG-cRNAs for importins α2, α4, and α6 were [α-32P]dCTP-labeled using the Megaprime DNA labeling system (Amersham Biosciences) as per the manufacturer's instructions, and hybridized to the membranes at 42°C overnight. The membranes were washed to a stringency of 0.1× SSC and 0.1% SDS up to 54°C and exposed to X-ray film (Hyperfilm™, Amersham Biosciences) overnight at −80°C in Hyperscreen™ Intensifying Screen cassettes (Amersham Biosciences). [α-32P]dCTP-labeled probes to GAPDH and ribosomal protein S26 were also hybridized to the same membranes and washed under the same conditions to illustrate relative RNA loading between samples.

Northern analysis for importins α1 and α3 was performed using DIG-labeled riboprobes. Blots were prepared as described above and prehybridized with 6 mL of ULTRAhyb™ (Ambion®, Austin, TX) at 68°C for 1 hr. DIG-labeled antisense riboprobe (100 ng/mL) was added and hybridized overnight at 68°C. Membranes were washed to a stringency of 0.1× SSC and 0.1% SDS at 68°C, then in maleic acid buffer (0.1M Maleic acid with 0.15M NaCl, pH 7.5, and 0.3% Tween). One percent (w/v) blocking reagent (Roche Molecular Biochemicals, Laval, QU, Canada) in maleic acid buffer was used to block the membranes and to dilute the anti-Digoxigenin-alkaline phosphatase conjugate (Roche Molecular Biochemicals) to 1:10,000. After exposure to the anti-DIG conjugate, the membranes were washed with the maleic acid buffer and then incubated with CDP-Star™ (Roche Molecular Biochemicals) detection reagent for 5 min. Chemiluminescent signal was detected with Hyperfilm™ (Amersham Biosciences) in Hyperscreen™ cassettes (Amersham Biosciences) for 1–3 hr.

GEOProfile Analysis

The expression graphs for each importin α were generated by downloading publicly available data records from the Entrez website. The GEO DataSet record downloaded was GDS605: spermatogenesis and testis developmental time course (MG-U74A) (http://www.ncbi.nlm.nih.gov/geo/gds/gds_browse.cgi?gds=605). From this record, the expression values for each mouse importin α was graphed using the ChartWizard in the Microsoft Excel software. The data were submitted by the Griswold Laboratory, Centre for Reproductive Biology, School of Molecular Biosciences, Washington State University, Pullman (Shima et al., 2004).

Real-Time PCR

Three micrograms of total RNA prepared from total testes of rats at 5, 9, 20, and 60–90 dpp (adult) was treated with DNAse I (IU/1 mg RNA) for 30 min at 37° C, and the enzyme inactivated by heating to 70°C for 5 min. The DNAseI-treated total RNA samples were then transcribed using 100 U Superscript II reverse transcriptase (Life Technologies, Grand Island, NY) and oligo DT primer according to the manufacturer's guidelines.

PCR samples were prepared in a final volume of 25 μl using the Applied Biosystems SYBR-Green PCR master mix, 300 nM F1 and R1 primers (Importin α1, Forward: 5′ caaagcaaaaccgcttgacca 3′; Importin α1, Reverse: 5′ aggtgggctcttccctcta 3′; Importin α6, Forward: 5′gactttgaagatgtacaagaacag 3′; Importin α6, Reverse: 5′tccctgcaaagcgagctatc 3′) and the reverse transcribed template. PCR was performed on the Applied Biosystems PRISM 7700 sequence detector (Perkin Elmer/PE Biosystem) using the following thermocycler conditions: Stage 1, 50°C for 2 min, 1 cycle; Stage 2, 95°C for 10 min, 1 cycle; Stage 3, 95°C for 15 s, and 60°C for 1 min, 40 cycles. Melting curve analysis and agarose gel electrophoresis were used to monitor the production of the appropriate PCR product.

Each PCR reaction was performed in duplicate with negative controls, where water was used in place of the reverse-transcribed template, included for each primer pair to exclude PCR amplification of any contaminating DNA. The amount of GAPDH mRNA was measured in each sample template to normalize between samples (GAPDH, Forward: 5′ acgtgccgcctggagaaacct3′; GAPDH, Reverse: 5′ tgcctgcttcaccaccttcttg3′).

To correlate the threshold values (Ct) from the sample amplification plot to copy number, a standard curve was produced for each product using plasmid DNA contain the target cDNA. The data were calculated as the mean ± standard deviation based on 3 separate experiments.

Acknowledgements

The authors acknowledge the assistance of Marilyn Bakker and Catherine Itman with the Northern blot analysis, the contribution of Dr. Mai Sarraj to the fetal gonad dissections, and comments on the manuscript by Dr. Julia Young. This work was supported in part by the NHMRC of Australia (143792 to K.L. and 334013 D.A.J.). C.H. is supported by an Australian Postgraduate Award and an ARC Centre of Excellence research scholarship.

Ancillary