Xing Fang, Zuo-Min Zhou, Li Lu, Lan-Lan Yin, Jian-Min Li, Yin Zhen, Hui Wang, Jia-Hao Sha

Key Laboratory of Reproductive Medicine, Nanjing Medical University, Nanjing 210029, China

Spermatogenesis involves a number of processes, including meiosis, haploid gene expression, formation of acrosome and flagellum, replacement of histones with protamines and nuclear condensation. Spermatogenesis commences in puberty and the expression of genes is different in different developmental stages of testis [1-3]. Study on genes expressed in the testis at different stages of development may identify new genes related to the function of testis [4]. A series of genes are involved in the control of spermatogenesis. Exploration of these genes could elucidate not only its physiology, but also the pathological mechanism of male infertility [5].

In the present study, the genes expressed in the human embryo and adult testes and spermatozoa were compared and a novel human pyridoxal kinase (PKH) gene with a specific expression in the testis (PKH-T) was identified.

Probes of human adult and fetal testes and spermatozoa were hybridized with human testis cDNA micro-array containing 9216 clones. Preparation of microarray and probes and hybridization was conducted according to Sha et al [6, 7].

Human ejaculates were obtained from 10 healthy volunteers of proven fertility and normal semen quality, as assessed by WHO criteria (1999). All the samples were obtained after the approval of the Ethical Committee of this University and the obtaining consent forms from all the participants. The semen samples were allowed to liquefy for 1 h at room temperature and then washed twice in phosphate-buffered saline (0.1 mol/L PBS). The sediments were stored in liquid nitrogen before RNA extraction.

Ten pooled sediments were homogenized and the total RNA was isolated with the trizol RNA isolation protocol (Gibco BRL, USA) and quantified with a UV spectrometer and electrophoresis. The testis RNA was purified using an affinity column filled with poly (dT) resin. These probes were prepared by the incorporation of ³³P-labeled dATP in a reverse transcription reaction using spermatozoa total RNA 50 g as the template with an oligo (dT) as the primer and M-MLV reverse transcriptase (Promega, USA). Each labeling reaction was carried out with 200 Ci of 33P-dATP following the attached instruction of the manufacturer (NEN Life Science, USA).

The differentially expressed genes in adult and fetal testes and spermatozoa were purified using mini-preps (QIAprep Spin Miniprep Kit, Qiagen, Germany). They were then sequenced by an ABI377 automatic sequencer (Applied Biosystems, USA). DNA sequences of the differentially expressed genes were inputted into the database for the search of homologous proteins using the BLAST and SMART programs. The nucleotide sequence and the putative protein structure were analyzed by the GenRunner program.

Hybridization of cDNA microarray containing 9216 human clones with adult and fetal testis and spermatozoa probes identified a spermatogenesis-related clone, PKH-T. This gene was highly expressed in adult testis and spermatozoa, but lower in fetal testis; its hybridization signal intensity in adult and fetal testes and spermatozoa were 259.66, 52.45 and 149.23, respectively (Figure 1).

The cDNA full-length is 1069 bp and contains an open reading frame (nt 190-909) that encodes a protein with 239 amino acids (Figure 2). The methionine at nt 190-192 was the initiation site, because there was an upstream stop code TGA at nt 76-78. The cDNA sequence of this clone was deposited in the GenBank under the accession number of AY303972.

Blast search in the human genome database localized the PKH-T gene at the chromosome 21. NT_011515.101/HS21_11672. One clone in chromosome 21 (length= 340000 bp) contains the genomic sequence of PKH. The PKH gene is spliced by 10 exons (with lengths ranging from 47 to 265 bp) and 9 introns (with lengths ranging from 89 to 2869bp), encompassing 14 749 bp genomic DNA in NT_011515.101. Blast search of the contig map showed that all exons were located within chromosome 21q22.3, so this PKH-T gene was mapped to chromosome 21q22.3.

Blast searches revealed that PKH-T was highly homologous to 4 other genes, all of which belong to the PKH (human pyridoxal kinase) gene family and have been identified in the human (form 110-1057 bp of PKH-T, identity rate=100 %) as NM003681, BC00123, BC005825 and U89606 (Figure 3).

Splicing comparison of PKH-T with homologous genes indicates that 5 homologues originate from one gene and consist of 12 exons (Figure 3). Exon 3 is present only in PKH-T and PKH-T lacks exons 1 and 2, thus the difference between PKH-T and other PKH genes may be caused by alternative splicing in different tissues.

PKH-T coded a 239 amino acid protein with predicted molecular weight of 26.83 kD. Its peptide sequence was 73 a.a, being shorter than that of PKH (U89606 was a representative and NM003681, BC00123, etc. also encoded 312 a.a). The sequence of PKH-T was the same as that from 74th amino acid to the end of PKH and the identity rate was 100 %.

Blast conserved domain search for PKH-T protein showed a Pdxk (PN/PL/PM kinase) domain (3-239 a.a); furthermore, the SMART software (http://www.smart.embl-heidelberg.de/) had identified one PKH domain (1-211 a.a).

To compare PKH-T and PKH with other B6-vitamin kinases, we present an alignment that includes E.coli PdxK, PdxY, T.brucei PdxK, C.elegans PdxK, R.norvegicus Plk and human PKH (PKH_human) (Figure 4). Activities have been demonstrated directly only for E.coli PdxK and PdxY [8, 9], human PKH [10] and T.brucei PdxK [11]; the other sequences are putative homologs identified by blast searches. This alignment suggests that the conserved motifs may be involved in substrate binding or catalysis, including a candidate tyrosine residue (marked 1), which cannot be modified following PL binding [12], a dissociable P-loop motif (regions 3), which may be involved in ATP binding [11], candidate aspartic and glutamic acid residues (marked 2), which may act as the general bases in phosphate transfer[11]and a region (marked 3) that is affinity labeled by the bisubstrate analog adenosine tetraphosphate in the PL kinase isolated from the sheep brain [13].

PCR and electrophoresis showed that this novel PKH gene was particularly and strongly expressed in human testis (Figure 5) and therefore was named PKH-T. It was not expressed in heart, brain, placenta, lung, liver, skeletal muscle, kidney, pancreas, spleen, thymus, prostate gland, ovary, small intestine, colon and peripheral leukocytes.

A variety of approaches have been developed to obtain information on gene expression levels, including in situ hybridization [14], differential display technology [15] and cDNA microarray [16]. In this study, cDNA microarray was used and a testis specific protein gene PKH-T was found, which possesses the same structural characteristics necessary for the activation of pyridoxal kinase [11-13, 17-19], i.e., a nearly identical PKH domain and very conserved motifs of PKH, possibly involved in substrate binding or catalysis (Figure 4). For the above reasons, we considered that PKH-T is a splice variant of PKH in the testis.

Michael CH et al [10]used 32P-labelled ORF PKH cDNA as a probe for the Northern analysis of tissue distribution of PKH and showed that a major 1.5-kb PKH transcript was expressed in all tissues tested and the hybridizing mRNA had an exceptional high expression in the testis. PKH-T is highly homologous with PKH gene (from 323-1272 bp of PKH, identity rate =100 %). In view of the facts that Michael CH used the complete PKH ORF cDNA as a probe and that both PKH and PKH-T are expressed in the testis with a higher expression level of PKH-T, we believe that this probe hybridized with the PKH mRNA in testis and that the blot signals are the simultaneous expression of the PKH and PKH-T gene.

Several lines of evidence indicated the relation between PKH-T and spermatogenesis: 1) PKH-T is highly expressed only in the testis; 2) It is highly expressed in the primer of life testes, in human spermatozoa, but not in fetal and aged testis; 3) The expression loss of PKH-T in testis of SCOS patients indicates that it is expressed not in somatic cells, including the Sertoli cells and Leydig cells, but only in the spermatogenic cells. These findings specifically linked PKH-T to the spermatogenic cell function, but it is not clear how the gene participates in this function.

Recently, ubiquitin dependence of gametogenesis has been gradually perceived. Ubiquitin ligases E1, E2, E3 and UBC4 are active in the testis. Ubiquitin and proteosomal subunits can be detected in human sperm centrosome undergoing dramatic reduction during spermatid elongation. Spermatid histones are ubiquitinated as they are being transiently replaced by transitional proteins and permanently by protamines. The defective spermatozoa become surface-ubiquitinated during their descent down the epididymis [20, 21]. Since ubiquitin-mediated degradation is extensively involved in spermatogenic process, the degradation products, small peptides and individual amino acids, must be produced in large quantity. Most amino acids cannot be deaminated directly and must first pass through transamination and pyridoxal-5'-phosphate (PLP) is a coenzyme of all transaminases. Function of PKH-T gene in testis should be similar to that of PKH in somatic cells. We may assume that PKH-T may play a role in ubiquitin-mediated degradation, then there is no surprise that PKH-T gene is expressed at such a high level in testis and spermatogenic cells.

The present study suggests that abnormal expression of PKH-T may be correlated with infertility. Expression of PKH-T is lost in SCOS and partially in spermatogenic arrest patients, which may be one of the causes of these disorders. PKH-T loss might affect the process of replacement of spermatid histones by protamines, thus interfering in spermatogenesis. In conclusion, the specific expression of the new gene PKH-T in the human spermatogenic cells may be correlated with spermatogenesis and involved in infertility.

The work was supported by grants from National Project 973, China (No. G1999055901) and Chinese Natural Science Funds (No. 30170485).