Molecular Human Reproduction Vol.10, No.6 pp. 433±444, 2004 DOI: 10.1093/molehr/gah052 Advance Access publication March 25, 2004

Expression analysis of the human testis-speci®c serine/ threonine (TSSK) homologues. A TSSK member is present in the equatorial segment of human sperm

Zhonglin Hao1, Kula N.Jha1, Young-Hwan Kim1, Soumya Vemuganti1, V.Anne Westbrook1, Olga Chertihin1, Karin Markgraf1, Charles J.Flickinger1, Michael Coppola1, John C.Herr1 and Pablo E.Visconti1,2,3 Downloaded from https://academic.oup.com/molehr/article/10/6/433/1119808 by guest on 04 October 2021 1Center for Research in Contraception and Reproductive Health (CRCRH), Department of Cell Biology, University of Virginia, Charlottesville, VA 22908 and 2Department of Veterinary and Animal Sciences, University of Massachusetts, 208 Paige Laboratories, Amherst, MA 01003, USA 3To whom correspondence should be addressed at: Department of Veterinary and Animal Sciences, University of Massachusetts, 208 Paige Laboratories, Amherst, MA 01003, USA. E-mail: [email protected]

Two members of the human testis-speci®c serine/threonine (Ser/Thr) kinase family, TSSK 1 and TSSK 2, were cloned and sequenced from a human testis adaptor-ligated cDNA library using a PCR strategy. Within the cDNA, open reading frames (ORF) were de®ned encoding proteins of 367 and 358 amino acids respectively, as well as conserved kinase domains typical of the superfamily of Ser/Thr . Both were intronless and mapped to 5 and 22 respectively. The human and mouse homologues of TSSK 1 and TSSK 2, together with TSSK 3 and SSTK/FKSG82, constitute a kinase subfamily closely related to the calmodulin kinases and SNF/nim 1 kinase subfamilies. Similar to the mouse, tissue expression by northern and dot blot analysis revealed that human TSSK 1 and 2 messages are expressed exclusively in the testis. However, mRNA for these kinases can be detected in other tissues using real-time PCR. In addition, TSKS, the human homologue of a putative substrate of TSSK 1 and 2, was cloned. TSKS had an ORF of 592 amino acids and was also expressed exclusively in the testis as demonstrated by northern and dot blot analyses; however, lower levels of expression in other tissues were detected using real- time PCR. Human TSSK 2 and TSKS interacted in a yeast two-hybrid system and also co-immunoprecipitated after in vitro translation. TSSK 2 expressed in yeast and bacteria was able to autophosphorylate and also phosphorylated recombinant TSKS in vitro. Antibodies against recombinant TSSK 2 demonstrated that a member of the TSSK family was present in human testis and localized to the equatorial segment of ejaculated human sperm. In contrast, TSKS was only found in the testis. The ®nding of a TSSK family member in mature sperm suggests that this family of kinases might play a role in sperm function.

Key words: serine/threonine kinase/testis-speci®c/spermatogenesis/TSKS/TSSK

Introduction reorganization of the chromatin, the elongation and species-speci®c Spermatogenesis, the process in which functional sperm are produced reshaping of the cell, and the assembly of the ¯agellum (Sharpe, in the testis, involves speci®c interactions between the developing 1994). These events result from changes in both transcription germ cells and their supporting Sertoli cells as well as hormonal (Hecht, 1988) and protein translation (Hake et al., 1990) that occur regulation by the androgen-producing Leydig cells. The general during this developmental period. Many proteins translated in the organization of spermatogenesis is essentially the same in all haploid spermatid will remain incorporated in the cytoarchitecture of mammals and can be divided into three distinct phases: (i) the initial the mature sperm after it leaves the testis, while others may be shed phase is the proliferative or spermatogonial phase during which with the cytoplasmic droplet. Given these considerations, proteins that spermatogonia undergo mitotic division and generate a pool of are synthesized during spermiogenesis might play necessary roles in spermatocytes; (ii) the meiotic phase, which yields the haploid spermatid differentiation and/or sperm functions during epididymal spermatids; and (iii) spermiogenesis whereby each round spermatid maturation, capacitation, sperm transport and fertilization. differentiates into a spermatozoon (Sharpe, 1994). Although the Our group is interested in the study of signalling events that are molecular mechanisms regulating the ®rst two phases have been involved in spermatogenesis and in sperm function. One particular relatively well characterized (Sassone-Corsi, 1997), many of the area of interest focuses on signal transduction processes that modulate molecules involved in spermiogenesis remain uncharacterized. the acquisition of sperm fertilizing capacity. After ejaculation, sperm Mammalian spermiogenesis, the post-meiotic phase of spermatogen- are able to move actively but lack fertilizing competence. They esis, is characterized by considerable morphological changes that acquire the ability to fertilize in the female genital tract in a time- occur in the haploid spermatid. Some of these changes include the dependent process called capacitation (for review see Visconti and formation of the acrosome and its contents, the condensation and Kopf, 1998). We have recently demonstrated that capacitation is

Molecular Human Reproduction vol. 10 no. 6 ã European Society of Human Reproduction and Embryology 2004; all rights reserved 433 Z.Hao et al.

Table I. Proposed name Alternative names Introns

Human Mouse Human Mouse Predicted Kinase Gene name

TSSK 1 SPOGA4 5q22.2 16 10.40 cM No No Yes STK22D FKSG81 TSSK 1b SPOGA 1 22q11.21 ± No No STK22A TSSK 1 TSSK 2 SPOGA2 22q11.21 16 10.40 cM No No Yes STK22B FLJ38613 DGS-G TSSK2 TSSK 3 SPOGA3 1p34.3 4D2.2 One No Yes STK22C TSSK3 TSSK 4 FKSG82/SSTK 19p13.11 8B3.3 One No Yes SSTK Downloaded from https://academic.oup.com/molehr/article/10/6/433/1119808 by guest on 04 October 2021 accompanied by the increase in protein phosphorylation of a subset of against the database on the National Center for Biotechnology proteins on tyrosine residues. This increase in protein tyrosine Information (NCBI) BLAST server. Mouse TSSK 1 matched an intronless phosphorylation is regulated by a cAMP/ A (PKA) sequence on chromosome 5 (Ref | NT_006899.7). This sequence was used to pathway (Visconti et al., 1995a,b, 1997). With the exception of PKA, design PCR primers Hao73 (TCTATCCAGGATGTAAATGAGCACACT, other kinase(s) involved in the regulation of capacitation remain sense), Hao74 (GAGAAGAAGCTGATGAAAATAGAGGCT, antisense), Hao75 (CTGTAGAGGGCAGCCTCAGAGGCACTG, sense), and Hao76 unde®ned. (GAACCATGTGGGTTACCA AGTTCAAGG, antisense). These primers Post-translational modi®cations of proteins through phosphoryl- were used to amplify human TSSK1 from an adaptor-ligated human testicular ation play a role in many cellular processes such as the transduction of Marathon ready cDNA (library) (Clontech, USA). Amplimers from Hao73/ extracellular signals, intracellular transport, and cell cycle progres- Hao76 were puri®ed using a 1% agarose gel, subcloned into a pTOPO cloning sion. Protein kinases make up a large family of related vector (Invitrogen, USA) and both strands were sequenced. (Hanks et al., 1988) characterized by a homologous region of ~300 To clone human TSSK 2 (also known as stk22b; Table I), mouse TSSK 2 amino acids. Considering the importance of phosphorylation events in nucleotide sequence was used to retrieve the human genomic sequence using the regulation of cellular mechanisms, it is not surprising that several BLAST search. Mouse TSSK 2 matched an intronless sequence on protein kinases have been shown to be involved in spermatogenesis chromosome 16 (AC004471). This sequence was used to design the PCR (Sassone-Corsi, 1997). However, only a few of them are exclusively primers Hao57 (antisense, CATGGAGCTGCAGACCATGATGTA) and Hao58 (sense, GACGTCAGCCAGCGGCTGCACATC). These primers were expressed in germ cells or in the testis (Jinno et al., 1993; Walden and used in combination with adaptor primers from the adaptor-ligated Marathon Cowan, 1993; Nayak et al., 1998; Toshima et al., 1998, 1999; Tseng ready testicular cDNA library (Clontech) to obtain the full length human TSSK et al., 1998; Shalom and Don, 1999). Examples of testis-speci®c 2 sequence using 5¢ and 3¢ rapid ampli®cation of cDNA end (RACE) as kinases include the recently described testis-speci®c serine/threonine described by the manufacturer. The amplimers were cloned into the pTOPO kinases (TSSK) 1, 2 and 3 (Bielke et al., 1994; Kueng et al., 1997; plasmid after gel puri®cation and sequenced. Finally, two primers Hao68 Zuercher et al., 2000). These kinases are expressed postmeiotically (antisense, TGTGCCTTCTTCAGATCCTACCTGCCT) and Hao67 (sense, during spermiogenesis. Although the mechanism of action of the TTGAGGACAATGCCTGCTGGCCCACAT), designed according to the two TSSK is unknown, the importance of phosphorylation events in RACE fragments, were used to amplify the longest cDNA. The cloned cDNA signalling processes suggests that this kinase family might have a role was sequenced in both strands. in sperm differentiation and/or subsequent sperm functions during Human TSKS was cloned from Marathon ready human testis cDNA library (Clontech) in a two step PCR reaction. First, the 5¢ and 3¢ regions were maturation, capacitation and fertilization. ampli®ed separately by PCR, using primers Hao89 (5¢-ATGGCGAGCG- To continue our studies on the TSSK family, initiated with the TGGTGGTGAAGACGATC-3¢) and Hao90 (5¢GAACAGCTTCTGCAA- cloning of the mouse and human homologues of TSSK 3 (Visconti TTCCTGCTC-3¢), or Hao65-1 (GAGGCCATTTATTGTTCAGGGGCT- et al., 2001), we have now cloned the human homologues of TSSK 1 GAG) and Hao93 (5¢-CAGCAGCAGCTGCAGGATGAGACG) pairs respect- and TSSK 2. In addition, we have cloned the human homologue of a ively. The 5¢ and 3¢ regions overlapped each other in the middle section by 100 putative substrate of the TSSK (TSKS) (Kueng et al., 1997) and shown bp. The amplimers were then puri®ed by gel electrophoresis and full length that this protein is abundantly expressed in testis. Interestingly, TSKS was ampli®ed using oligo primers Hao89 and Hao65-1 and a mixture of recombinant TSSK 2 is able to interact and to phosphorylate TSKS the 5¢ and 3¢ PCR products as template. The full length TSKS was puri®ed by in vitro. In addition, these two proteins interact in a yeast two-hybrid gel electrophoresis, subcloned into pTOPO cloning vector and sequenced. assay. Antibodies made against recombinant TSSK 2 and TSKS showed that TSSK 2 is present in human testis and in ejaculated Plasmid construction sperm, while the TSKS protein was only found in the testis. These To construct pGBKTSSK2, oligonucleotide primers Hao100 (5¢- results are the ®rst to demonstrate that a member of the TSSK family is GATGAATTCGACGATGCCACAGTCCTAAGGAAG-3¢) and Hao101 (5¢- CGAGGATCCCTAGGTGCTTGCTTTCCCCACCTC-3¢) were used to amp- present in mature sperm. lify the TSSK2 open reading frame (ORF) by PCR. The PCR fragment was then subcloned into the pGBK vector after cleavage of both the PCR amplimer and the pGBK vector with EcoRI and BamHI restriction enzymes. Similarly, the Materials and methods TSKS ORF was ®rst ampli®ed using Hao99 (5¢-TCTCTCGAGTGA- GGCCATTTATTGTTCAGG-3¢) and Hao107 (5¢-GTCAAGATCTTGG- Cloning of the human homologues of TSSK1, TSSK2 and their CGAGCGTGGTGGTGAAGACG-3¢). The amplimer cleaved by BglII and putative substrate TSKS XhoI was subcloned into the pGAD between BamHI and XhoI sites. Mouse TSSK 1 (also known as stk22a and FSKG81, Table I) nucleotide TSSK 2 and TSKS expression plasmids pESCtrpTSSK2 and pESCleuTSKS sequence was initially used to perform a basic local alignment search (BLAST) were constructed as follows. PCR ampli®cation of TSSK 2 with an N-terminal 434 Sperm and testis-speci®c kinases

Myc tag was performed using pGBKTSSK2 as a template and oligo primers reverse transcriptase (Qiagen, USA) in a 100 ml reaction using the reaction Hao136 (5¢-ATAGCGGCCGCGAATTTGTAATACGACTCACTATA-3¢) buffer supplied by the manufacturer. Real-time RT±PCR was performed using and Hao101 (5¢-CGAGGATCCCTAGGTGCTTGCTTTCCCCACCTC-3¢). a Bio-Rad (USA) i-cycler IQ system. Primer pairs were validated in 20 ml PCR The PCR fragment was then cleaved by NotI and BamHI and inserted into reactions containing 2 ml of testis cDNA, 10 ml of IQ SYBR Green supermix the pESCtrp vector following cleavage by NotI and BglII. The pGADTSKS (BioRad), and 0.6 mlofa10mmol/l stock of each primer (0.3 mmol/l ®nal insert was cleaved by BglII and XhoI and inserted into the BamHI, XhoI concentration). Cycle conditions were 95°C for 3 min followed by 45 cycles of cleaved pESCleu vector to construct pESCleuTSKS. All plasmid constructs 95°C for 10 s and 60°C for 30 s. The ampli®cation was followed by melting were sequenced before transformation to con®rm the correct reading frame and curve analysis in which the PCR products were denatured at 95°C for 1 min and to ensure that no mutations had been accidentally introduced. annealed at 72°C for 10 s. The temperature was then increased in 0.1°C increments while monitoring the loss of SYBR green ¯uorescence. Each primer Multi-sequence alignment and phylogenetic analysis pair ampli®ed a single product with a sharp melting peak at a temperature Multiple sequence alignment of human TSSK 1, TSSK 2, TSSK 3 and TSSK 4 consistent with the calculated Tm of the predicted PCR product. Agarose gel was performed using Clustal W (Thompson et al., 1994) on the EMBL/EBI electrophoresis con®rmed that the ampli®ed products had the expected sizes. server and a consensus was determined. To construct a phylogenetic tree, an No products were obtained when reverse transcriptase was omitted from the amino acid BLAST search of TSSK 1 was conducted ®rst to retrieve those cDNA synthesis reactions. kinases that showed homology to human TSSK 1. A group of 28 kinases was The oligonucleotide primers were all purchased from Qiagen Operon then chosen for multiple sequence alignment using the pile-up program in the (Alameda, USA). Where possible, the primers were selected to span a splice Downloaded from https://academic.oup.com/molehr/article/10/6/433/1119808 by guest on 04 October 2021 GCG package. The pile-up ®le was subsequently used to construct a junction. Obviously, this could not be done in cases of intronless genes such as phylogenetic tree using the PHYLIP program in the GCG software package. TSSK 1 and TSSK 2. Primer pairs for each gene were: TSSK 1, 5¢-GCC- The kinases used for phylogenetic tree analysis included Abl (P00519), Src1 CCTAGGTGGATGAGG-3¢ (forward) and 5¢-TCACGCTCTGGGGGAGTA- (P12931), PKC (P17252), PKG (P00516), PKA (P17612), tpk2 (P06245), 3¢ (reverse); for TSSK 2, 5¢-GGGTTCCTACGCAAAAGTCA-3¢ (forward) and (P06244), ark2 (P35626), ark1 (P21146), nim1 (P07334), snf1 (P06782), kin1 5¢-GTTTTCTTGCGGTCGATGAT3¢ (reverse); for TSSK 3, 5¢-GGG- (P13185), kin2 (P13186), CaMKII1A (P11275), CaMKIIB (P11276), FKSG82 GAAGGGACCTACTCAAA-3¢ (forward) and 5¢-GTCCAGGGTACG (AF348077), TSSK 3 (AF296450), TSSK 2 (AF362953), TSSK 1 (AY028964), GACGATTT-3¢ (reverse); for TSSK 4, 5¢-TACGCGTCACCCGAGTGCT-3¢ tesk1 (NM_006285) (Toshima et al., 1999), tesk2 (NM_007170) (Rosok et al., (forward) and 5¢-ACGCCCATGCTCCACACA (reverse); for TSKS, 5¢- 1999), nek1 (S25284), ERK2 (M64300), ERK1 (AF155236), cdc28 (P00546), GCTGAGCGAGAATCTGGAG-3¢ and 5¢-TTCAGCATCTTCCACAGACC- Cdc2 (P06493), MCK1 (P21965)1, GSK3 (NM_017344). 3¢; for TSKS-1 5¢-GGATTCAAATGAGGCTCCAAC-3¢ (forward) and 5¢- TGGAGGTAGCGCAGCTTCT-3¢ (reverse); for glyceraldehyde-3-phosphate Northern blots and RNA dot blot analyses dehydrogenase (GADPH), 5¢-ATCATCAGCAATGCCTCCTG-3¢ (forward) Northern and dot blot analyses were performed as previously described (Hao and 5¢-ATGGCATGGACTGTGGTCAT-3¢ (reverse). Relative levels of mRNA expression in human tissues were determined using et al., 2002). Brie¯y, human multiple tissue northern membranes (MTN) multiple tissue cDNA (MTC) panels from BD Biosciences (USA) in real-time containing eight human tissues with 1 mg of poly A+ RNA loaded in each lane and multi-tissue array RNA dots containing 76 tissues were purchased from PCR reactions as described above. The point at which the PCR product is ®rst Clontech. Double strand DNA of 1.35 kb (TSSK1), 1.25 kb (TSSK2) and 1.8 kb detected above a ®xed threshold, termed cycle threshold (Ct), was determined for each sample. Melting curve analysis con®rmed the ampli®cation of only the (TSKS) containing the entire ORF of the three genes were excised from pTOPO clones and used as hybridization probes for northern and dot blot hybridization expected product. To determine the quantity of gene-speci®c transcripts present analyses. The hybridization probes were prepared by the random prime in each tissue cDNA relative to testis, their respective Ct values were ®rst normalized by subtracting the Ct value obtained from the GADPH control (e.g. labelling method using [a32P]dCTP and Klenow DNA polymerase. ± DCt = Ct TSSK 3 ± Ct GADPH). The concentration of gene-speci®c mRNA Hybridizations were carried out in ExpressHyb (Clontech) at 68°C for 1 h followed by two washes in 23standard salt citrate (SSC), 0.1% sodium dodecyl in a given tissue relative to testis was then calculated by subtracting the sulphate (SDS) for 20 min each at 22°C and three washes in 0.1% SSC, 0.1% normalized Ct values obtained with each tissue from that obtained with testis (e.g. ± DDCt = DCt of brain ± DCt of testis), and the relative concentration was SDS for 20 min, each at 65°C. Films were exposed 24±96 h at ±80°C with two determined (relative concentration = 2±AÈ AÈ Ct). intensifying screens. After the probes were stripped, the same membranes were probed with b-actin as the loading control. For hybridization with multi-tissue RNA dot blots, the same probes were Expression of TSSK 2 and TSKS and antibody production used. Hybridizations were performed at 68°C overnight in ExpressHyb solution TSSK 2 and TSKS were subcloned into a pET28b expression vector (Hao et al., containing single-stranded salmon sperm DNA and Cot-1 DNA following the 2002) and the respective recombinant proteins expressed in Escherichia coli. manufacturer's instructions. The array contained 76 human tissues including The recombinant TSSK 2 contained the entire ORF of human TSSK 2 plus a whole brain, cerebral cortex, frontal lobe, parietal lobe, occipital lobe, temporal histidine (his) tag in its amino terminus, whereas the recombinant TSKS lobe, paracentral gyrus of cerebral cortex, pons, left cerebellum, right contained a differentially spliced version of TSKS and a his tag in its amino cerebellum, corpus callosum, amygdala, caudate nucleus, hippocampus, terminus. This version of differentially spliced TSKS is otherwise 100% medulla oblongata, putamen, substantia nigra, accumbens nucleus, thalamus, identical to the testis transcript (from amino acids 20 to 386) except the unique pituitary gland, spinal cord, heart, aorta, left atrium, right atrium, left ventricle, leader sequence (amino acid 1±19). The recombinant proteins were puri®ed right ventricle, interventricular septum, apex of the heart, oesophagus, stomach, using a Ni-NTA af®nity column and subsequently resolved by preparative duodenum, jejunum, ileum, ilocaecum, appendix, ascending colon, transverse SDS±PAGE using a Prep-Cell apparatus (Model 491; Bio-Rad, USA). The colon, descending colon, rectum, kidney, skeletal muscle, spleen, thymus, puri®ed proteins, which appeared as single bands on SDS±PAGE after silver peripheral blood lymphocytes, lymph node, bone marrow, trachea, lung, staining, were used for injection into rats following the schedule previously placenta, bladder, uterus, prostate, testis, ovary, liver, pancreas, adrenal gland, described (Hao et al., 2002). thyroid gland, salivary gland, mammary gland, leukaemia HL-60, Hela S3, leukaemia K-562, leukaemia MOLT-4, Burkkitt's lymphoma Raji, Burkkitt's Preparation of sperm lymphoma Daudi, colorectal adenocarcinoma SW480, lung carcinoma A549, Semen specimens were obtained from volunteers by masturbation. The use of fetal brain, fetal heart, fetal kidney, fetal liver, fetal spleen, fetal thymus, and semen samples was approved by the UVA Human Investigation Committee. fetal lung. The membranes were washed twice in 23SSC, 0.1% SDS each at Individual semen samples were allowed to liquefy at room temperature (0.5±3 22°C for 10 min and twice in 0.1% SSC, 0.1% SDS each at 68°C for 20 min and h) and mature sperm were puri®ed by Percoll (Pharmacia Biotech, Sweden) exposed for 96 h at ±80°C. density gradient centrifugation as previously described (Naaby-Hansen et al., 1997). The basic medium used for all experiments was modi®ed human tubal Real-time RT±PCR analysis ¯uid (mHTF; Irvine Scienti®c, USA). Sperm presenting >90% motility were First strand cDNA was prepared from 5 mg of total RNA from human testis prepared immediately for either western blots or immuno¯uorescence (Ambion, USA) using 1 mmol/l oligo d(T) primer (Ambion) and Omniscript experiments. 435 Z.Hao et al.

Immuno¯uorescence microscopy was centrifuged in a 1.5 ml centrifuge tube at 10,000 3g for 2 min, at 22°C. The Sperm were air-dried onto slides, washed three times with PBS, permeabilized supernatants were carefully transferred to fresh tubes on ice. Aliquots of 120 ml with methanol, washed with PBS and then blocked with 10% normal goat of culture supernatant were mixed with 360 ml of assay buffer [prepared by serum (NGS) in PBS. Incubations were then carried out with a TSSK 2 mixing two volumes of 0.5 mol/l sodium acetate pH 4.5 and 1 vol of antibodies (1:250) diluted in PBS with 1% NGS (PBS-NS), washed and 100 mmol/l of p-nitrophenyl a-D-galactopyranoside (PNP-a-Gal)]. The reaction mixes were incubated at 30°C for 5 h before stopping the reaction incubated with FITC-conjugated F(Ab)2 fragments of donkey a rat IgG (1:200) (Jackson ImmunoResearch) in PBS-NS. Slides were washed with PBS and with the addition of 520 ml of stop buffer (1 mol/l sodium carbonate). The mounted with Slow-Fade Light (Molecular Probes, USA). Sperm were OD410 of each reaction was recorded by a spectrophotometer using a reaction observed by differential interference contrast (DIC) and epi¯uorescence mix containing SCM-leu-trp medium in the place of culture supernatant as the microscopy using a Zeiss axiophot microscope (Carl Zeiss, Inc., USA). blank. The activity was expressed either as arbitrary units or percentage of wild type gene (as a way to record interaction strength between TSSK 2 and TSKS deletion mutants). Each assay was repeated three times and the average was SDS±PAGE and immunoblotting reported. After incubation under different experimental conditions, the sperm were concentrated by centrifugation 10,000 3g 2 min at room temperature, washed Expression of TSSK 2 and TSKS in yeast, preparation of protein in 1 ml of phosphate-buffered saline (PBS) containing 1 mmol/l sodium extracts and in vitro kinase assay orthovanadate, resuspended in sample buffer (Laemmli, 1970) without Downloaded from https://academic.oup.com/molehr/article/10/6/433/1119808 by guest on 04 October 2021 To express human TSSK 2 and TSKS in yeast for in vitro kinase assay, the 2-mercaptoethanol and boiled for 5 min. After centrifuging, the supernatants pESC-trp-TSSK 2 with an N-terminal myc tag and pESC-leu-TSKS with an were saved and 2-mercaptoethanol was added to a ®nal concentration of 5%, N-terminal HA tag were constructed as above. Yeast host strain YPH500 the sample was boiled for 5 min, and then subjected to 10% SDS± (Genotype: Mat-a ura3-52, lys2-801amber, ade2-201ochre, trp1-D63, his3-D200, polyacrylamide gel electrophoresis (PAGE) (Laemmli, 1970). Testis extracts leu2-D1) was transformed either with pESC-trp-TSSK 2 alone or together with were obtained from Clontech. Electrophoretic transfer of proteins to pESC-leu-TSKS. To make yeast protein extracts, yeast transformants carrying Immobilon P (Bio-Rad) and immunodetection were carried out using a appropriate plasmid(s) were cultured in selective media to mid-log phase TSSK 2 rat polyclonal antibodies as previously described (Kalab et al., 1994). (OD <1.0). Cells were harvested by centrifugation at 4000 g for 15 min, Immunoblots were developed with the appropriate secondary antibody 600 washed twice in ice-cold water and suspended in appropriate volume of ice- conjugated to horseradish peroxidase (Sigma Chemical Co., USA) and an cold HB buffer with protease inhibitors (see above) to make non-denaturing ECL kit (Amersham Corp., USA) according to the manufacturer's instructions. protein extracts using the beads-beating method. Alternatively, denaturing protein extracts were made by mixing a 1:1 ratio of 20% trichloroacetic acid In vitro transcription/translation and immunoprecipitation (TCA), yeast cell suspension in TCA buffer (Tris 20 mmol/l, EDTA 2 mmol/l, In vitro transcription/translation was performed using the TNT Quick Coupled ammonium acetate 50 mmol/l pH 8.0 plus protease inhibitor cocktail (Roche) Transcription/Translation System (Promega, USA) following the manufactur- and 425±600 mm diameter glass beads (Sigma, USA). The cells were er's instruction. Plasmids with TSSK 2 (tagged with myc in the N terminus) and dissociated in a screw-capped tube for 2330 s using a bead-beater (Biospec, TSKS (tagged with HA in the N-terminus) inserts immediately downstream of USA) at the highest setting. The supernatant was collected and the remaining a T7 promoter in pGBK and pGAD respectively were used as templates. In beads were washed again by suspending in appropriate volumes of a 1:1 mix of vitro transcription and translation were started with the addition of the TNT 20% TCA and TCA buffer and dissociated for another 30 s. The proteins were quick master mix, methionine and the plasmid templates. The mixture was then collected by centrifugation at 10,000 3g. In all cases we used a incubated at 30°C for 60±90 min and stopped by addition of 23SDS loading microcentrifuge at maximum setting for 10 min at 4°C. The tagged proteins buffer for western blot analysis using monoclonal anti-myc antibody (mouse; were immunoprecipitated individually with myc (9E10) or HA high af®nity Clontech) or monoclonal anti-HA high af®nity antibody (3F10; Roche, USA). monoclonal antibody (3F10) coupled to agarose beads and washed thoroughly. For co-immunoprecipitation assay, TSSK 2 and TSKS were transcribed and In vitro kinase assay was performed as described previously (Visconti et al., translated in the same reaction as above. The reactions were stopped by 1997). Brie¯y, recombinant TSSK 2 bacterial soluble fraction before and after addition of 23kinase buffer (see below for composition). To immunopreci- induction with isopropyl-b-D-thiogalactopyranoside (IPTG) were used as the pitate the tagged proteins, the in vitro-translated proteins were incubated with source of and incubated at 30°C for 10 min in a buffer containing either myc antibody coupled to agarose beads (Clontech) or HA antibody 10 mmol/l Tris±HCl pH 7.4, 10 mmol/l MgCl2, 10 mmol/l MnCl2,10mmol/l coupled to agarose beads (Roche) for 1 h at 4°C. Following three washes with aprotinin, 10 mmol/l leupeptin, 100 mmol/l Na3VO4, 40 mmol/l p-nitrophenyl 13kinase buffer, the beads were boiled in SDS loading buffer, and analysed by phosphate, 40 mmol/l b-glycerol phosphate and 5 mCi [32P] gATP. In some western blot. The reaction mixture without plasmid template and treated cases, to evaluate the ability of TSSK 2 to phosphorylate TSKS, the puri®ed identically served as the negative control. C-terminal domain of recombinant TSKS was added to the reaction. The phosphorylation assay was then stopped by boiling in sample buffer and the 32P Two-hybrid analysis incorporation was analyzed using 10 % PAGE and autoradiography. The same The Match-Maker two-hybrid System 3 (Clontech) was used to assess protein± buffer was used to evaluate whether immunoprecipitates contained yeast protein interactions. To detect interaction between human TSSK 2 and TSKS, recombinant TSSK 2 kinase activity. In this case, immunoprecipitation was TSSK 2 was fused with Gal4 DBD in pGBKT7 whereas TSKS was fused with performed after immunoprecipitation with a-Myc antibodies. After the 32 Gal4 AD in pGADT7 (see above). The two plasmids were co-transformed into phosphorylation reaction, the beads were boiled in sample buffer and the P the AH109 host strain (Genotype: Mat a, trp1-901, leu2-3, 112, ura3-52, his3- incorporation evaluated by PAGE and autoradiography.

200, gal4D, gal80D, LYS2::GAL1UAS-GAL1TATA-HIS3, MEL1 GAL2UAS- GAL2TATA-ADE2, URA3::MEL1UAS-MEL1TATA-lacZ) and analysed for expres- sion of both hybrid proteins (DBD-TSSK 2, AD-TSKS) by western blot of Results myc-tagged and HA-tagged protein respectively. The AH109 transformants harboring both pGBKT7-TSSK 2 and pGADT7-TSKS were streaked out in Nomenclature, chromosomal localization and homology of complete drop-out medium (SCM) lacking tryptophan, leucine and histidine to the TSSK family test for histidine prototrophy. Two plasmids containing simian virus (SV) 40 Since the ®rst member of the TSSK family was identi®ed and cloned large T antigen (LgT) fused with GALAD in pGADT7 and p53 fused with in 1993 (Bielke et al., 1994), several groups have cloned other GALDBD in pGBKT7 were co-transformed into AH109 and used as positive controls. a-Galactosidase reporter gene activity was measured following members of this family, and as a consequence, several names have instructions provided in the kit (Match-Maker two-hybrid System 3; Clontech) been assigned to these kinases, lending a measure of confusion to the with minor modi®cation. Brie¯y, AH109 clones harbouring both pGBK TSSK ®eld. In this section, a nomenclature for the TSSK family is 2 and pGAD TSKS were cultured in SCM-leu-trp medium to optical density introduced. The original name TSSK given by Bielke et al. (1994)

(OD)600 nm ~1.0. One millilitre of culture from each inoculation (in triplicate) shall be adhered to for reasons of historical accuracy and the fact that 436 Sperm and testis-speci®c kinases Downloaded from https://academic.oup.com/molehr/article/10/6/433/1119808 by guest on 04 October 2021

Figure 1. Alignment of human TSSK 1±4 amino acid sequences. Amino acids found in two of the three aligned sequences are shaded to show identity. The highly conserved signature sequence that ®ts the consensus `DLKXXN' for serine/threonine kinases is underlined. The 12 kinase subdomains were marked over the alignment with roman numerals. The complementary DNA sequences of TSSK1 and TSSK2 were deposited into the GenBank database under the accession numbers AY028964 and AF362953 respectively.

TSSK stands for testis-speci®c serine kinase, a term that re¯ects both proposed particularly to differentiate this pseudogene from the the function of these enzymes and the abundant (although not authentic human homologue of TSSK 1 located on chromosome 5. exclusive) expression in the testis (see results below). Other names The second TSSK gene located on human is 87% given to this kinase family are presented in Table I, including the identical to mouse Tssk 2 and has been designated TSSK 2 in our genome-annotated designations of the four TSSK members. nomenclature while its gene symbol is STK22B. Human TSSK 2 and In the mouse, two homologous Tssk family members, Tssk1 and TSSK1b are localized in a region deleted in DiGeorge's syndrome. Tssk2, are closely linked on chromosome 16 A3. This region of the Human TSSK 1 is 85% identical to mouse Tssk1 and maps to mouse chromosome is syntenic to human chromosome 22q11.21 human chromosome 5q22.2 in a region without synteny to mouse where two human homologues of mouse Tssk are closely mapped. chromosome 16, the locus where mouse TSSK1 is found. The One of these human sequences on 22q11, which we designate evolutionary signi®cance of this lack of synteny is not apparent at TSSK1b, appears to represent a truncated, non-transcribed pseudo- present. As described before, the human TSSK 3 gene maps to gene related to mouse Tssk 1. The longest possible open reading frame chromosome 1p34.3 and is syntenic with the mouse Tssk 3 gene on for TSSK1b encodes a potential protein fragment corresponding to the chromosome 4 D2.2 (Visconti et al., 2001). Another predicted gene C terminus of TSSK1. This fragment spans amino acids 196±358 of with high homology to TSSK is SSTK or FKSG82. The SSTK gene TSSK1 and lacks a complete kinase domain. Hence, TSSK 1b is maps to chromosome 19p13.11 and has a mouse homologue, Sstk, in predicted to be inactive as a kinase. This human pseudogene appears the syntenic region of mouse chromosome 8 B3.3. Expression analysis as TSSK 1 or SPOGA 1 in databases. The term TSSK 1b (Table I) is of SSTK mRNA (see below) and expressed sequence tags (EST) 437 Z.Hao et al.

49% identity with TSSK 1 and TSSK 2 respectively. Unlike TSSK 3, TSSK 1 and TSSK 2 have an ~100 amino acid C-terminal extension located outside the kinase domain. A phylogenetic tree was constructed as indicated in Materials and methods using sequences from 28 human kinases with high homology to TSSK 2 (Figure 2). This study showed that the TSSK 1 and TSSK 2 genes constitute a subfamily within the TSSK kinase family, which, in turn, lies within the superfamily of Ser/Thr kinases. Other related kinases, in order of decreased similarity to TSSK 1 and TSSK 2, include TSSK 3 and TSSK 4. In addition, CaMK IIA, CaMK IIB, KIN1, KIN2 and nim1, SNF have signi®cant homology with the TSSK family.

Tissue expression of the TSSK family by northern and dot blots

In mouse, the expression of TSSK 1, TSSK 2 and TSSK 3 is reported Downloaded from https://academic.oup.com/molehr/article/10/6/433/1119808 by guest on 04 October 2021 to be limited to testicular germ cells (Kueng et al., 1997; Visconti et al., 2001). Expression of TSSK 3 is also testis speci®c in humans using similar methods (Visconti et al., 2001). To evaluate the tissue distribution of TSSK 1 and TSSK 2 in humans, the human homologues of TSSK 1 and TSSK 2 were cloned. The expression pattern of TSSK 1 and TSSK 2 was analysed by northern and dot blots of multiple human tissues as described above (Figure 3B and C). Similar to results reported in the mouse, these techniques showed that mRNA corres- ponding to TSSK 1 and TSSK 2 were detected exclusively in the testis. Although it is not possible to discount lower levels of expression of these genes in other tissues, these northern analyses suggest that TSSK 1 and TSSK 2 are abundant within and predominantly expressed in the testis. Northern blots corresponding to TSSK 1 showed two bands of 1.5 and 2.7 kb (Figure 3B). Since TSSK 1 and TSSK 1b are highly homologous, and TSSK 1 b predicts a truncated form of TSSK 1, it is Figure 2. Phylogenetic tree of the catalytic domains of protein kinases. To construct the phylogenetic tree, an amino acid basic local alignment search possible that the 1.5 kb message represents expression of the truncated (BLAST) search of TSSK 1 was conducted ®rst to retrieve kinases that TSSK 1b gene whose message is predicted to be 1.4 kb. TSSK 2 gave showed homology to human TSSK 1. A group of 28 kinases was then chosen only a single band of 1.7 kb. for multiple sequence alignment using the pile-up program in the GCG The tissue distribution of a putative human testicular substrate of package. The pile-up ®le was subsequently used in construction of a phylogenetic tree using PHYLIP program in the GCG software package. The TSSK 1 and TSSK 2, named TSKS (testis-speci®c kinase substrate), kinases used for phylogenetic tree analysis and their respective accession previously discovered by the Ziemiecki group in the mouse (Kueng numbers are given in the Materials and methods section. et al., 1997), was also explored by northern and dot blot analyses of human tissues. In mouse, TSKS interacts with TSSK 2 and with TSSK 1 in a yeast two-hybrid system and can be phosphorylated in vitro by TSSK 2 and TSSK 1. To investigate whether TSKS is highly sequences suggest that this gene is also abundantly expressed in the expressed in the human testis, the human TSKS homologue was testis. Due to the high homology between SSTK and the other cloned as described in Materials and methods. Similar to the mouse, members of the TSSK family, we propose the name TSSK 4 for the analysis of TSKS tissue distribution by northern and dot blot analysis protein encoded by the gene SSTK. The gene symbols for the TSSK showed a 2.0 kb message that was exclusively expressed in the testis. family are currently STK22A-D for TSSK 1±3 and SSTK for TSSK 4. Since it is now clear that these genes are distributed on chromosomes Real-time PCR analysis of TSSK tissue distribution 5, 22, 1 and 19 (humans) and 16, 4 and 8 (mice), a renaming of these genes is warranted, since the 22 chromosomal designation no longer Although northern and dot blot techniques indicated that TSSK 1, applies to all members. Table I summarizes a proposed nomenclature TSSK 2 and TSSK 3 were only expressed in the testis, to evaluate the for the members of the TSSK family including alternative names, possibility of lower mRNA expression levels in other tissues, real-time chromosomal localizations and current gene symbols. We hope this PCR was used with probes speci®c for each of the TSSK kinases, nomenclature is considered by the Human Genome Nomenclature including TSSK 4. Although TSSK 1 and TSSK 2 were abundantly Committee. expressed in the testis (Figure 4), several other tissues showed lower It is noteworthy that the genes encoding human TSSK 1 and TSSK 2 levels of expression. TSSK1 was noted in pancreas, while TSSK2 encode proteins of 367 and 358 amino acids respectively, and are expression was noted in heart, brain and placenta. By contrast, TSSK 3 intronless, as are their corresponding mouse homologues. The human and TSSK 4 were only found to be expressed in the testis (Figure 4). It TSSK 3 and TSSK 4 genes, by comparison, each have two exons that is not clear at present whether the low expression of TSSK 1 and encode proteins of 268 and 273 amino acids respectively. TSSK 2 in other tissues has a physiological signi®cance. The human TSSK (Figure 1) show very high homology at the amino acid level. The identity between TSSK 1 and TSSK 2 is 83% in the Recombinant human TSSK 2 and TSKS interact kinase region. This identity, however, decreases to 72% across the Two different approaches were used to investigate whether the human entire ORF, re¯ecting a divergence of the C-terminal amino acid homologues of TSSK 2 and TSKS interact. First, the interaction sequence (Figure 1). In comparison, the TSSK 3 protein has 47.5 and between TSSK 2 and TSKS was analysed using a yeast two-hybrid 438 Sperm and testis-speci®c kinases Downloaded from https://academic.oup.com/molehr/article/10/6/433/1119808 by guest on 04 October 2021

Figure 3. Multi-tissue northern (A) and dot blot (B) analyses of TSKS, TSSK 1 and TSSK 2 and TSKS mRNA expression. Left panels: northern blots containing 1 mg of poly-A(+) mRNA in each lane were hybridized to a probe labelled with 32P using 1.25 kb of the respective cDNA as template. (A) TSKS, (B) TSSK 1 and (C) TSSK 2. The same membranes were stripped of the respective probe and hybridized to a b-actin probe labelled in the same way. The ®lms were exposed for 6±96 h according to need. Right panels: membranes dotted with RNA from 76 human tissues (for details of the tissues, see Materials and methods) were hybridized to the same TSKS, TSSK 1 and TSSK 2 probes as above. The membranes were exposed to the ®lm for 96 h before development. system (Clontech) as described in the Materials and methods section. + GalAD-LgT (Figure 5C). a-Galactosidase activity was signi®cantly The TSSK 2 ORF was fused with Gal4 DNA binding domain (DBD) higher when GalDBD-p53 and GalAD-LgT (8-fold higher) and when (Myc tag) and the TSKS ORF was fused with Gal4 DNA activation GalDBD-TSSK 2 and GalAD-TSKS (12-fold higher) were expressed domain (AD) (HA tag). As a positive control, p53 was fused with Gal4 together in comparison with a-galactosidase activity when GalDBD + DBD, and simian virus SV40 large T (LgT) was fused with Gal4 AD. GalAD-TSKS were co-expressed. The S. cerevisiae host strain AH109 was transformed with four pairs As a second approach, an in vitro transcription/translation system of plasmids, which contained GalDBD + GalAD-TSKS, GalDBD- coupled with a co-immunoprecipitation approach was used to further TSSK 2 + GalAD-TSKS, GalDBD + GalAD-LgT and GalDBD-p53 + investigate whether human TSSK 2 and human TSKS are able to form GalAD-LgT; as expected, they all grew on SCM plates lacking both a complex when co-expressed. Expression plasmids PGBK-TSSK2 leucine and tryptophan (Figure 5A). Western blot analysis showed that (TSSK 2 tagged with Myc) and pGAD-TSKS (TSKS tagged with HA) all proteins were expressed (Figure 5B). On re-streaking the same were transcribed and translated together in a rabbit reticulocyte-based transformed colonies onto SCM plates lacking leucine, tryptophan and expression system. A mouse monoclonal antibody against the myc tag histidine, only the GalDBD-TSSK 2 + GalAD-TSKS and GalDBD- coupled to agarose beads (Clontech) was used to immunoprecipitate p53 + GalAD-LgT pair supported growth (Figure 5 A). This TSSK 2, whereas an anti-HA tag rat monoclonal antibody (3F10) experiment indicated that TSSK 2 and TSKS are able to interact, coupled to agarose beads was used to precipitate TSKS (Figure 5D). bring together GalDBD and GalAD and cause activation of histidine As predicted, the TSKS-HA was precipitated by anti-HA immunor- amino acid synthesis in the host strain similar to the positive control eagents and TSSK 2-myc was immunoprecipitated by anti-myc pair GalDBD-p53 + GalAD-LgT. To con®rm these data and to semi- antibodies, validating that each tagged protein was appropriately quantify the strength of the interaction, a-galactosidase activity was immunoreactive. Importantly, anti-HA precipitated TSSK 2-myc, and measured in culture supernatants of yeast clones harbouring GalDBD conversely, anti-myc also precipitated TSKS-HA. The control anti- + GalAD-TSKS, GalDBD-TSSK 2 + GalAD-TSKS, and GalDBD-p53 HA antibodies did not precipitate TSSK 2-Myc and conversely, anti- 439 Z.Hao et al. Downloaded from https://academic.oup.com/molehr/article/10/6/433/1119808 by guest on 04 October 2021

Figure 4. Real-time PCR analyses of TSSK 1±4 expression. Relative levels of mRNA expression in human tissues were determined using multiple tissue cDNA (MTC) panels from BD Biosciences (USA) in real-time PCR reactions as described in Materials and methods. The point at which the PCR product is ®rst detected above a ®xed threshold, termed cycle threshold (Ct), was determined for each sample. Melting curve analysis con®rmed the ampli®cation of only the expected product. To determine the quantity of gene-speci®c transcripts present in each tissue relative to testis, their respective Ct values were ®rst normalized by subtracting the Ct value obtained from the GADPH control (e.g. ± DCt = Ct TSSK 3 ± Ct GADPH). The concentration of gene-speci®c mRNA in a given tissue relative to testis was then calculated by subtracting the normalized Ct values obtained with each tissue from that obtained with testis (e.g. ± DDCt = DCt of brain ± DCt of testis), and the relative concentration was determined (relative concentration = 2±DDCt).

Myc antibodies were not able to precipitate TSKS-HA (Figure 5E). assay (Figure 6A, right panel). Parallel samples were probed with anti- These cross-immunoprecipitation results using co-expression in the Myc antibodies and showed the presence of TSSK 2-Myc in the reticulocyte transcription/translation system reinforce the yeast two- immunoprecipitates (Figure 6B). The robust phosphorylation signal hybrid data and together offer evidence in support of the hypothesis observed in bacteria extract in the absence of induction is likely to be that human TSSK 2 and human TSKS interact. an endogenous bacterial substrate for an endogenous bacterial kinase or alternatively the autophosphorylation of an endogenous kinase. Recombinant human TSSK2 phosphorylates TSSKS author Although reduced, this signal is also visible in the induced population. correct? To analyse TSSK 2 kinase activity and its ability to phosphorylate Western blot and immuno¯uorescence analyses of TSSK 2 TSKS, TSSK 2 was expressed in bacteria. The soluble recombinant Puri®ed TSSK 2 was used to produce rat polyclonal antibodies that TSSK 2 was assayed for kinase activity in the presence or in the were used for western blots analysis of human sperm and testis absence of recombinant TSKS. As shown in Figure 6A (left panel), (Figure 7A). Anti-TSSK 2 recognized the recombinant protein and recombinant TSSK 2 incorporated 32Pinanin vitro phosphorylation also a protein in sperm and testes with the predicted mol. wt of 40 kDa, assay indicating autophosphorylation. In addition, when TSKS was suggesting that at least one member of the TSSK family was present in present in the incubation assay, TSKS was also phosphorylated. To ejaculated sperm. On the other hand, anti-TSKS only recognized a enrich both TSSK and TSKS, immunoprecipitation after phosphoryl- protein in testis but not in ejaculated sperm. Anti-TSSK 2 was then ation was employed. Anti-Myc immunoprecipitates of yeast extracts used to study the intracellular localization of this protein in human containing TSSK 2-Myc also incorporated 32P in the phosphorylation sperm. Figure 7B shows that TSSK 2 localized to the sperm head, with 440 Sperm and testis-speci®c kinases Downloaded from https://academic.oup.com/molehr/article/10/6/433/1119808 by guest on 04 October 2021

Figure 5. TSSK 2 and TSKS interaction. (A) Yeast host strain AH109 was transformed with either a pair of plasmids or a single plasmid: pGAD, pGBK- TSSK2 (1); pGAD-TSKS, pGBK-TSSK2 (2); pGAD, pGBKp53 (3); pGAD-lgT, pGBK-p53 (4); pGAD (5); pGAD-TSKS (6); and pGAD-LgT (7). The transformants were cultured in appropriate selection medium and streaked out on complete drop-out media lacking both leucine and tryptophan (SCM-L-T), or leucine, tryptophan and histidine (SCM-L-T-H) to test for histidine prototrophy. (B) Expression of the respective proteins was con®rmed by western blot using antibodies against the Myc tag (mouse monoclonal 9E10) and the HA tag (rat monoclonal 3F10). The theoretical molecular weight for GBK-TSSK2-Myc is 59 kDa and for GAD-TSKS-HA is 80 kDa. (C) a-Galactosidase reporter gene assay. AH109 host strain was transformed with the indicated pairs of plasmids. Yeast clones harbouring each pair of plasmids were cultured in complete drop-out medium lacking both leucine and tryptophan to OD600 ~1.0. Culture supernatants were processed for assaying a-galactosidase activity using PNP-a-Gal as substrate. The activities were expressed as arbitrary units after calibration with optical density (OD)600 of the culture. (D) TSSK 2 co-immunoprecipitates with TSKS. HA-tagged human TSKS and Myc-tagged human TSSK 2 were transcribed together and translated in vitro in a rabbit reticulocyte lysate-based system as described in the Materials and methods. The reaction mix was separated into three aliquots. One of them was not treated and the other two immunoprecipitated with rat monoclonal antibody against HA (3F10) coupledto agarose beads or with mouse monoclonal antibody against Myc (9E10) coupled to agarose beads. The reaction mix, the a-HA and the a-Myc immunoprecipitates were then resolved on 10% SDS±PAGE and detected with a-HA (left) or a-Myc antibody (right) as indicated in the ®gure. +/+ indicates in vitro cotranslation, ±/± indicates that no plasmids were added to the in vitro translation mix. (E) HA-tagged human TSKS and Myc-tagged human TSSK2 were transcribed in separate reactions and translated in vitro in a rabbit reticulocyte lysate-based system as described in the Materials and methods. Each reaction mix was separated into three aliquots. One of them was not treated and the other two immunoprecipitated with rat monoclonal antibody against HA (3F10) coupled to agarose beads or with mouse monoclonal antibody against Myc (9E10) coupled to agarose beads. The reaction mix, the a-HA and the a- Myc immunoprecipitates were then resolved on 10% SDS±PAGE and detected with a-HA (left)ora-Myc antibody (right) as indicated in the ®gure. The bands marked with * are IgG detected with the second antibody. The theoretical mol. wts for TSSK2-Myc and TSKS-HA are 45 and 75 kDa respectively. 441 Z.Hao et al. Downloaded from https://academic.oup.com/molehr/article/10/6/433/1119808 by guest on 04 October 2021

Figure 6. In vitro kinase assay of recombinant TSSK 2. (A, left panel) Autophosphorylation of TSSK2 and phosphorylation of TSKS. The soluble bacterial recombinant TSSK 2 was assayed for kinase activity in the presence or in the absence of recombinant TSKS (C-terminal; theoretical mol. wt: 50 kDa) as described in the Materials and methods. (A, right panel) Autophosphorylation of TSSK-2. Yeast extracts containing TSSK 2-Myc were immunoprecipitated using a-Myc antibodies and a kinase assay was performed. The phosphorylation reaction mixture was then separated by SDS±PAGE, the gels dried and the 32P incorporation detected by autoradiography after overnight exposure. (B) Western blot using a-Myc antibodies demonstrates yeast expression of myc-tagged TSSK2. TSSK 2-Myc is present in yeast extract only after induction and can be recovered in the immunoprecipitate after immunoprecipitation with a-Myc. The bands marked with * are IgG detected with the second antibody. The theoretical molecular weight for TSSK2-Myc is 45 kDa. staining most intense in the equatorial segment. Control experiments other protein kinases. TSSK 1 displays high homology to a group of were performed using rat preimmune serum and both western blot and yeast Ser/Thr kinases encoded by SNF-1, nim-1, KIN-1 and KIN-2 immuno¯uorescence results were negative (Figure 7A, plus data not (Bielke et al., 1994). Tssk 2, on the other hand, is the mouse shown). Although the antibody against TSSK 2 was produced against homologue of the human DiGeorge syndrome gene (DGSG). This recombinant TSSK 2, we cannot discard the possibility of cross- human gene has been characterized as one of 11 putative transcription reactivity with other members of the TSSK family since their units encoded in the minimal DiGeorge critical region of 250 kb, sequences have high homology. located in the proximal arm of human chromosome 22 (Gong et al., 1996; Galili et al., 1997; Kueng et al., 1997). Of these genes, the transcription factor tbx1 is the gene responsible for developmental Discussion disorders observed in Di George's syndrome (Epstein, 2001; Lindsay et al., 2001; Merscher et al., 2001). However, the localization of TSSK Protein kinases play a pivotal role in intracellular signal transduction 2 in the same region of chromosome 22 has been responsible for systems involved in the regulation of cell proliferation, differentiation, assignment of the name stk22 b to this gene. Since other members of metabolism, and other activities. An increasing number of genes encoding putative protein kinases have been isolated by cDNA this kinase family have a different chromosomal localization (Table I), cloning (Lindberg and Hunter, 1990; Cance et al., 1993). Some of a nomenclature change of these kinases from stk22 to TSSK and stk22 these newly identi®ed protein kinases contain catalytic and non- a, b, c and d to TSSK 1, 2, 3 and 4 may both simplify and clarify this catalytic domains unrelated to known members of the protein kinase family of genes. Reasons for assigning the name TSSK (testis speci®c family. These novel members are assumed to be involved in signalling serine/threonine kinases) to these kinase are: (i) a short sequence motif pathways for which the substrates remain unde®ned. One classical in the kinase subdomain VIB (DKCEN) (Figure 1) diagnostic for ser/ example where a new kinase was identi®ed before its substrates were thr kinases is present in all the members of this family; (ii) northern, de®ned is in the JAK family of protein kinases. This protein kinase dot blot and real-time PCR analyses indicate that these kinases are family was identi®ed as a solitary class of protein kinases not related mainly expressed in the testis; (iii) the ®rst cloned member of this to any other known kinases (Cance et al., 1993) and were later family was originally named TSSK. demonstrated to be involved in signalling pathways mediating The human homologues of TSSK 1 and 2 cloned in the present interferon and other cytokine effects (Silvennoinen et al., 1993; study have a high percentage identity with mouse TSSK 1 and TSSK 2 Watling et al., 1993). (84 and 92% respectively), suggesting that they are the authentic The TSSK family of kinases are at present hypothesized to be human orthologues of the mouse genes. The four members of the involved in spermatogenesis and/or sperm function, although the TSSK family have high homologies in their kinase domains with precise pathways are unclear. The ®rst member of this family (TSSK1) TSSK 1 and 2 presenting the highest homology, followed closely by was cloned by Bielke et al. (1994) using degenerate oligonucleotides TSSK 3 and TSSK 4. Similar to results in the mouse, northern and dot corresponding to two highly conserved motifs within the protein blot analyses of human TSSK 1, 2 and 3 indicate that they are kinase catalytic domain and a PCR-based cloning strategy. Another exclusively expressed in the testis. However, real-time PCR analysis two members were subsequently cloned by the same group and/or by revealed lower levels of expression in other tissues. Whether the low others using different methodologies (Kueng et al., 1997; Zuercher expression of TSSK kinases in other tissues has physiological et al., 2000; Visconti et al., 2001). The sequence of a fourth member of relevance is still not known. Although the TSSK 4 cDNA has yet to the TSSK family, SSTK/FKGS82 (TSSK 4), has been deduced using be cloned, Unigene analysis showed several EST corresponding to this bioinformatics. The TSSK family is closely related to a variety of gene that are expressed in the testis, suggesting that TSSK 4 is also a 442 Sperm and testis-speci®c kinases Downloaded from https://academic.oup.com/molehr/article/10/6/433/1119808 by guest on 04 October 2021

Figure 7. Analyses of the presence of TSSK 2 and TSKS in human testis and in ejaculated human sperm. (A) The respective recombinant protein (TSSK 2 or TSKS with expected mol. wts of 43 and 46 kDa respectively), or native proteins (TSSK 2 or TSKS with expected mol. wts of 39 or 65 kDa respectively) from sperm and testis protein extracts were analysed by western blots using anti-TSSK 2 and anti-TSKS rat polyclonal antibodies or the respective preimmune serum as indicated in the ®gure. Human testis extract was purchased from Clontech and the sperm extract prepared as described in the Materials and methods. R = recombinant protein; T = testis protein extracts; S = sperm protein extracts. (B) Human sperm immuno¯uorescence was performed using anti-TSSK 2 antibodies as indicated in Materials and methods. testis-abundant message. This hypothesis is supported by real-time antibody. However, the presence of at least one TSSK kinase in PCR experiments shown in this manuscript. sperm suggests that this kinase family might have a role in sperm In the mouse, Tssk 1, 2 and 3 have been found expressed function. postmeiotically in germ cells. This expression pattern suggests a role Several different protein phosphorylation events regulate sperm in spermiogenesis or in sperm function. A putative substrate of Tssk 1 motility (Eddy and O'Brien, 1994) and capacitation (Visconti et al., and 2, Tsks, has an expression pattern similar to these kinases (Kueng 2002); one of the protein kinases playing a role in both processes is et al., 1997) and was shown to interact with these kinases using a yeast PKA. However, other protein kinases involved in these processes are two-hybrid approach. After immunoprecipitation of Tssk 1 and Tssk 2 not known. The presence of at least one member of the TSSK family in from testicular extracts, Tsks remained associated with these kinases the equatorial segment of human sperm suggests that this kinase(s) and became phosphorylated in an in vitro assay (Kueng et al., 1997). could have a role in some of the properties attributed to this sperm In the present report, the human homologue of TSKS was cloned and structure. The equatorial segment is the place where the fusion events two approaches, a yeast two-hybrid system and co-immunoprecipita- accompanying the acrosome reaction begin. After the acrosome tion, demonstrated that TSKS is able to interact with TSSK 2. Similar reaction is completed, the equatorial segment remains attached to the to the TSSK family, human TSKS is also a testis-abundant message as sperm and becomes the ®rst sperm structure to interact with the oocyte shown by northern and dot blots. The role of TSKS has yet to be oolema. It is believed that most of the proteins needed for the fusion established. between the sperm and the oocyte reside in this subcellular Recombinant human TSSK 2 and TSKS were used to obtain rat compartment. Proteins in this region are also the ®rst to go inside polyclonal antibodies and these antibodies recognized the recombi- the oocyte and could be involved in oocyte activation as sperm factors. nant proteins by western blots. Anti-TSSK 2 antibodies demonstrate In addition, the equatorial segment remains intact after the principal the presence of a 40 kDa band in sperm and in testicular extracts. In segment is lost during the acrosome reaction and proteins in this contrast, TSKS protein was only recognized in the testis. Due to the domain may play a role in equatorial segment structural integrity. high homology between TSSK 1 and TSSK 2, it is not possible to Lastly, the equatorial segment is the region where breakdown of the know which of these kinases are recognized by the anti-TSSK 2 nuclear envelope is initiated (Wolkowicz et al., 2003). 443 Z.Hao et al.

In sum, although the mechanism of action of the TSSK family is Naaby-Hansen S, Flickinger CJ and Herr JC (1997) Two-dimensional gel unknown, the importance of phosphorylation events in signalling electrophoretic analysis of vectorially labeled surface proteins of human spermatozoa. Biol Reprod 56,771±787. processes, the conservation of the kinase domains in the three human Nayak S, Galili N and Buck CA (1998) Immunohistochemical analysis of the TSSK, and the TSSK(s) expression patterns suggest that TSSK might expression of two serine-threonine kinases in the maturing mouse testis. have a role in mammalian germ cell differentiation and/or sperm Mech Dev 74,171±174. function. The TSSK sequences present several differences compared Rosok O, Pedeutour F, Ree AH and Aasheim HC (1999) Identi®cation and to other known members of the kinase family, which suggests that characterization of TESK2, a novel member of the LIMK/TESK family of speci®c inhibitors toward this kinase family could be obtained, raising protein kinases, predominantly expressed in testis. Genomics 61,44±54. interest in these molecules as contraceptive targets. Sassone-Corsi P (1997) Transcriptional checkpoints determining the fate of male germ cells. Cell 88,163±166. Acknowledgements Shalom S and Don J (1999) Tlk, a novel evolutionarily conserved murine serine threonine kinase, encodes multiple testis transcripts. Mol Reprod Dev This study was supported by NIH HD38082 (to P.E.V.), by U54HD29099 and a 52,392±405. grant from Schering AG (to J.C.H.) and by the Andrew W.Mellon Foundation. Sharpe RM (1994) Regulation of spermatogenesis. In Knobil E and Neill JD Z.H. and K.J. were supported by D43TW/HD00654 from the Fogarty (eds) The Physiology of Reproduction, vol 1 Raven Press, New York, pp International Center. 1363±1434. Silvennoinen O, Ihle JN, Schlessinger J and Levy DE (1993) Interferon- Downloaded from https://academic.oup.com/molehr/article/10/6/433/1119808 by guest on 04 October 2021 References induced nuclear signalling by Jak protein tyrosine kinases. Nature 366,583± Bielke W, Blaschke RJ, Miescher GC, Zurcher G, Andres AC and Ziemiecki 585. A (1994) Characterization of a novel murine testis-speci®c serine/threonine Thompson JD, Higgins DG and Gibson TJ (1994) CLUSTAL W: improving kinase. Gene 139,235±239. the sensitivity of progressive multiple sequence alignment through sequence Cance WG, Craven RJ, Weiner TM and Liu ET (1993) Novel protein kinases weighting, position-speci®c gap penalties and weight matrix choice. Nucleic expressed in human breast cancer. Int J Cancer 54,571±577. Acids Res 22,4673±4680. Eddy EM and O'Brien DA (1994) The Spermatozoon. In Knobil E and Neill Toshima J, Koji T and Mizuno K (1998) Stage-speci®c expression of testis- JD (eds) The Physiology of Reproduction, vol 1 Raven Press, New York, pp speci®c protein kinase 1 (TESK1) in rat spermatogenic cells. Biochem 29±77. Biophys Res Commun 249,107±112. Epstein JA (2001) Developing models of DiGeorge syndrome. Trends Genet Toshima J, Tanaka T and Mizuno K (1999) Dual speci®city protein kinase 17,S13±17. activity of testis-speci®c protein kinase 1 and its regulation by Galili N, Baldwin HS, Lund J, Reeves R, Gong W, Wang Z, Roe BA, Emanuel autophosphorylation of serine-215 within the activation loop. J Biol BS, Nayak S, Mickanin C et al (1997) A region of mouse chromosome 16 is Chem 274,12171±12176. syntenic to the DiGeorge, velocardiofacial syndrome minimal critical region Tseng TC, Chen SH, Hsu YP and Tang TK (1998) Protein kinase pro®le of [published erratum appears in Genome Res (1997) 7(4), 399]. Genome Res sperm and eggs: cloning and characterization of two novel testis-speci®c 7,17±26. protein kinases (AIE1, AIE2) related to yeast and ¯y chromosome Gong W, Emanuel BS, Collins J, Kim DH, Wang Z, Chen F, Zhang G, Roe B segregation regulators. DNA Cell Biol 17,823±833. and Budarf ML (1996) A transcription map of the DiGeorge and velo- Visconti PE and Kopf GS (1998) Regulation of protein phosphorylation during cardio-facial syndrome minimal critical region on 22q11. Hum Mol Genet sperm capacitation. Biol Reprod 59,1±6. 5,789±800. Visconti PE, Bailey JL, Moore GD, Pan D, Olds-Clarke P and Kopf GS Hake LE, Alcivar, AA and Hecht, NB (1990) Changes in mRNA length (1995a) Capacitation of mouse spermatozoa I Correlation between the accompany translational regulation of the somatic and testis-speci®c capacitation state and protein tyrosine phosphorylation. Development cytochrome c genes during spermatogenesis in the mouse. Development 121,1129±1137. 110,249±257. Visconti PE, Moore GD, Bailey JL, Leclerc P, Connors SA, Pan D, Olds- Hanks SK, Quinn AM and Hunter T (1988) The protein kinase family: Clarke P and Kopf GS (1995b) Capacitation of mouse spermatozoa II conserved features and deduced phylogeny of the catalytic domains. Protein tyrosine phosphorylation and capacitation are regulated by a cAMP- Science 241, 42±52. dependent pathway. Development 121,1139±1150. Hao Z, Wolkowicz MJ, Shetty J, Klotz K, Bolling L, Sen B, Westbrook VA, Visconti PE, Johnson LR, Oyaski M, Fornes M, Moss SB, Gerton GL and Coonrod S, Flickinger CJ and Herr JC (2002) SAMP32, a testis-speci®c, Kopf GS (1997) Regulation, localization and anchoring of isoantigenic sperm acrosomal membrane-associated protein. Biol Reprod subunits during mouse sperm capacitation. Dev Biol 192,351±363. 66,735±744. Visconti PE, Hao Z, Purdon MA, Stein P, Balsara BR, Testa JR, Herr JC, Moss Hecht NB (1988) Post-meiotic gene expression during spermatogenesis. Prog SB and Kopf GS (2001) Cloning and chromosomal localization of a gene Clin Biol Res 267,291±313. encoding a novel serine/threonine kinase belonging to the subfamily of Jinno A, Tanaka K, Matsushime H, Haneji T and Shibuya M (1993) Testis- testis-speci®c kinases. Genomics 77,163±170. speci®c mak protein kinase is expressed speci®cally in the meiotic phase in spermatogenesis and is associated with a 210-kilodalton cellular Visconti PE, Westbrook VA, Chertihin O, Demarco I, Sleight S and Diekman phosphoprotein. Mol Cell Biol 13,4146±4156. AB (2002) Novel signaling pathways involved in sperm acquisition of fertilizing capacity. J Reprod Immunol 53,133±150. Kalab P, Visconti P, Leclerc P and Kopf GS (1994) p95, the major phosphotyrosine-containing protein in mouse spermatozoa, is a hexokinase Walden PD and Cowan NJ (1993) A novel 205-kilodalton testis-speci®c with unique properties. J Biol Chem 269,3810±3817. serine/threonine protein kinase associated with microtubules of the Kueng P, Nikolova Z, Djonov V, Hemphill A, Rohrbach V, Boehlen D, Zuercher spermatid manchette. Mol Cell Biol 13,7625±7635. G, Andres AC and Ziemiecki A (1997) A novel family of serine/threonine Watling D, Guschin D, Muller M, Silvennoinen O, Witthuhn BA, Quelle FW, kinases participating in spermiogenesis. J Cell Biol 139,1851±1859. Rogers NC, Schindler C, Stark GR, Ihle JN et al (1993) Complementation Laemmli UK (1970) Cleavage of structural proteins during the assembly of the by the protein tyrosine kinase JAK2 of a mutant cell line defective in the head of bacteriophage T4. Nature 227,680±685. interferon-gamma signal transduction pathway [see comments]. Nature Lindberg RA and Hunter T (1990) cDNA cloning and characterization of eck, 366,166±170. an epithelial cell receptor protein-tyrosine kinase in the eph/elk family of Wolkowicz MJ, Shetty J, Westbrook A, Klotz K, Jayes F, Mandal A, protein kinases. Mol Cell Biol 10,6316±6324. Flickinger CJ, Herr JC (2003) Equatorial segment protein de®nes a discrete Lindsay EA, Vitelli F, Su H, Morishima M, Huynh T, Pramparo T, Jurecic V, acrosomal subcompartment persisting throughout acrosomal biogenesis. Ogunrinu G, Sutherland HF, Scambler PJ et al (2001) Tbx1 Biol Reprod 69(3),735±745. haploinsuf®cieny in the DiGeorge syndrome region causes aortic arch Zuercher G, Rohrbach V, Andres AC and Ziemiecki A (2000) A novel defects in mice. Nature 410,97±101. member of the testis speci®c serine kinase family, tssk-3, expressed in the Merscher S, Funke B, Epstein JA, Heyer J, Puech A, Lu MM, Xavier RJ, Leydig cells of sexually mature mice. Mech Dev 93,175±177. Demay MB, Russell RG, Factor S et al (2001) TBX1 is responsible for cardiovascular defects in velo-cardio-facial/DiGeorge syndrome. Cell Submitted on December 23, 2003; resubmitted on February 13, 2004; 104,619±629. accepted on February 20, 2004 444