TWO ISOFORMS OF TRPM8 IN RAT VAS DEFERENS

© 2019 K. Gulak et al.; Published by the Ivan Franko National University of Lviv on behalf of Біологічні Cтудії / Studia Biologica. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://www.budapestopenaccessinitiative.org/ and Creative Commons Attribution 4.0 License), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. UDC: 577.352.332

Trpm8 is a nonselective Ca 2+ -permeable ion channel activated by temperatures below 28 °C and cooling chemical compounds such as menthol and icilin. Trpm8 is expressed in sensory neurons where it functions for temperature detection. However, Trpm8 is also expressed in various internal organs where temperature is stably higher than 36 °C, that is much higher than cool temperature needed for Trpm8 activation. This opens possible makes roles for Trpm8, such as nociception. Trpm8 mRNA is expressed in vas deferens (VD), a smooth muscle organ of male reproductive system. However, no Trpm8-mediated currents were previously registered in the myocytes. VD is located outside of testes, and it consists of smooth muscle tube covered by the epithelium. It actively contracts and transfers sperm from testes to ejaculatory ducts prior to ejaculation. Temperature in VD is stably high, thus, Trpm8 role there could be outside of cold detection. The objective of this study was to analyze Trpm8 mRNA and protein expression in rat VD, as well as splice variant analysis. Trpm8 mRNA expression in VD was confirmed with RT-PCR, and Trpm8 protein was detected by the Western-blot analysis. Additionally, we found that both mRNA and protein of shorter non-classical isoform, as well as canonical isoform of Trpm8 in VD. We isolated smooth muscle cells from VD and performed a multi-cell PCR. This technique makes possible non-myocytic mRNA detection that could be isolated from e.g. sensory neurons termini where Trpm8 is expressed at much higher levels. Interestingly, in the isolated smooth muscle cells, no canonical Trpm8 transcript was found, though the non-classical isoform was present. We propose the shorter isoform could be formed as a result of alternative splicing. This would account for a difference in Trpm8 function in VD, i.e. no Trpm8-mediated currents registered in the myocytes. A shorter isoform could have a truncated N-terminal domain, that is consistent with known human Trpm8 isoforms sM8-6 and sM8-18.

INTRODUCTION
Trpm8 is a nonselective Ca 2+ -permeable cationic ion channel first disco ve red in 2001 as a prostate-specific transcript upregulated in prostate cancer [12]. Later, Trpm8 was classified as a TRP family member that is activated by cold temperatu res between 10-28 °C [7,9], and cooling compounds such as menthol [10] and icilin [2,13].
Trpm8 is expressed in sensory neurons where it functions for temperature dete c tion. Trpm8 -/-knock-out mice lose the ability to sense cold and develop cold allody nia [9]. Trpm8 is also expressed in various internal organs where temperature remains stable around 37 °C that is 10 °C higher than a threshold point for Trpm8 activation. This proposes an additional temperature unrelated functions for Trpm8 [8]. For example, Trpm8 role in a nociception is a subject of numtrous studies. Surprisingly, some works showed Trpm8 lowers pain sensitivity during trauma, while other works highlighted active role of Trpm8 in a nociception [13].
Previously, we showed that Trpm8 transcript is present in smooth muscle cells of rat vas deferens (VD) [6]. VD is a part of male reproductive system. It consists of smooth muscle tube covered epithelium, and it actively contracts and transfers sperm from testes to ejaculatory ducts prior to ejaculation. In addition, menthol inhibits contractile activity of VD smooth muscle strips, while the magnitude of that effect differs for strips taken from epididymal and prostate parts of VD [6]. Our previous work showed an unexpected absence of Trpm8 related currents in the isolated smooth muscle cells of VD [5].
The objective of this study was to elucidate if Trpm8 protein is present in rat VD, well as analyze possible splice isoforms of Trpm8 mRNA.

MATERIALS AND METHODS
All experiments were performed on 3-month-old male Wistar rats (250-270 g). Experimental procedures were carried out in accordance with the International Convention of animals and the Law of Ukraine "On protection of animals from cruelty", protocol #02/10 (08 February 2010) of the Bioethics Committee of Bogomoletz Institute of Physio logy NAS of Ukraine.
Western-blot analysis. VD tissue was isolated, disposed of epithelial layer, divided into epididymal and prostate parts, and homogenized in a liquid nitrogen. Tissue powder was incubated in RIPA buffer (25,0 mM Tris-HCl pH 7.6, 150 mM NaCl, 1% TX100, 1% Sodium deoxycholate, 0.1% SDS) during 45 min at 4 °C. Samples were spun at 16 000 g at 4 °C for 20 min. Supernatant was used for standard 7.5% SDS-polyacrylamide gel electrophoresis. Protein samples were transferred into nitrocellulose membrane using semi-dry bloting. Primary antibodies were rabbit polyclonal anti-TRPM8 (Abcam, Cat. number ab3243), and rabbit anti-beta Actin antibody (Abcam, Cat. number ab8227), secon dary antibody was Donkey anti-Rabbit IgG (H+L) Cross-Adsorbed (Thermo Fisher Scientific, Cat. number 31458). Signal was detected using standard electrochemiluminescence method on X-ray film. Membrane was additionally stained with Ponceau S in order to detect protein ladder PageRuler™ Unstained Broad Range Protein Ladder (Thermo Fisher Scientific).
Reverse transcription PCR. Total RNA from VD was isolated using RNAzol® RT Kit (Sigma-Aldrich) according to manufacturer's manual. RNA concentration was calculated at 260 and 280 nm absorption spectrum (NanoDrop). Only samples with 260/280>1.6 absorption ratio were used in further experiments. First cDNA strand synthesis was performed using MMLV-reverse transcriptase Kit (Thermo Fisher Scientific), adding oligo-dT(18) and random hexamer primer, 50 pmol/reaction (Thermo Fisher Scientific).
Multi-cell PCR. In order to exclude mRNA of non-myocytic origin, we performed the multi cell PCR. Cells were prepared as described previously [3]. Briefly, VD strips were cut into small segments, and treated with 0.2% Collagenase of Type XI (Sigma) in Ca 2+ -free physiological salt solution in the presence of 0.3% bovine serum albumin at 35 °C for 30-40 min. Single cells were obtained by gentle pipetting.
Freshly isolated VD myocytes (n = 50 cells) were identified under the microscope, and then captured one by one with a borosilicate glass micropipette (tip diameter 30-50 µm), and transferred into a tube containing reverse transcription reaction mix. RT-PCR was performed as previously described.
Restriction analysis. PCR products were cut from agarose gel. DNA of the isolated samples was purified using Silica Bead DNA Gel Extraction Kit (Thermo Fisher Scientific) and analyzed using HindIII restriction enzyme. The results of restriction digestion were visualized in 2.5% agarose gel.
Data analysis. To test selected primer quality PCR digital mock test was performed with FastPCR 4.0.13 (PCR Team). Primer dimer formation ability was tested using PCR Primer Stats (The Sequence Manipulation Suite) [11].
Protein sequence corresponding to an appropriate DNA fragment was identified with Translate (SIB ExPASy Bioformatics Resources Portal) [1] using information from UniProt database, #Q8R455 entry (rat Trpm8 protein).

RESULTS AND DISCUSSION
Trpm8 is expressed at protein level in myocytes isolated from rat VD. However, our previous work did not reveal typical for Trpm8 channel at detection using patch-clamp on VD myocytes [6]. The objective of this study was to elucidate protein and mRNA expression of Trpm8 in VD tissue, as well as a possible reason for absence of Trpm8 current in VD myocytes shown previously [5].
We found that although Trpm8 transcript was detected with RT-PCR, the resulting amplicons had two lengths: the expected full product with length of 213 bp and truncated variant with a length of approximately 170 bp (Fig. 1). Primers for PCR were selected in a way to exclude introns, specifically the reverse primer spans the junction between 11th and 12th exons of R. norvegicus Trpm8 mRNA. A search using as a reference RefSec mRNA database for R. norvegicus Primer-BLAST additionally showed that 100% nucleotide specificity corresponded only to Trpm8 transcript and its variants X2-X5 of the same size (213 bp). At unspecific PCR amplification and non-complementary base-pair formation, some additional amplicons could be observed: 22.5% unspecificity, i.e. formation of 5 of 20 non-complimentary base pairs on forward primer and 4/20 on reverse primer, amplification of Pou2f3 mRNA (296 bp) could be achieved; with 25% unspecificity (5/20 non-complimentary base pairs on both primers) products can be formed of Supt5h (761 bp), Nsun5 (648 bp), and Kcnk3 (598 bp) mRNAs. All suggested unspecific products were larger than the observed truncated amplicon, and were absent in our studies. Moreover, such non-canonical base pair formation in 22.5% is unlikely to happen often, yet the truncated amplicon was observed systematically in VD tissue. An additional digital mock PCR with PCR Primer Stats showed no primer-dimerisation or hairpin formation with the selected primers. Thus, a formation of truncated amplicon comes not from unspecific primer binding, but is synthesysed from Trpm8 mRNA.
To test this, we performed additional analysis of specificity. Amplicons form first PCR were cut and purified from agarose gel and used as templates for secondary PCR. Obtained results showed that synthesis of both fragments is specific, i.e. fragments of 213 bp and 170 bp give rise a to amplicons of their own corresponding length (Fig. 2). A restricion digestion of purified amplicons with HinDIII restrictase showed that 213 bp amplicon was cut into two expected fragments (132 bp and 81 bp), while 170 bp amplicon remainded undigested, thus, it does not have a restriction site in its nucleotide sequence (Fig. 3). Previous PCR were performed on VD tissue that contains not only myocytes but also additional cell types such as nerve fibers or blood vessels. We also performed multi-cell PCR exclusively on myocytes. This was done by sampling 50 myocytes into the micropipette using patch-clamp pumping set-up and transferring them into RT-PCR tube. Interes tingly, results showed absence of 213 bp product, but 170 bp amplicon was present (Fig. 4). Analysis based on the Western-blot hybridization of probes from VD tissue using Trpm8 antibodies showed presence of Trpm8 signal in selected tissue and in prostate positive control (Fig. 5). Interestingly, strips corresponding to VD tissue have two bands with Trmp8 antibody hybridization of different weight: 130.6 kDa and a smaller band. In contrast, prostate tissue has only one band corresponding to 130.6 kDa expected protein weight.

CONCLUSION
Our current results show that a selected section of Trpm8 mRNA could be a subject for alternative splicing, with a removal of a fragment of approximately 45 bp length in VD. This hypothesis of truncated mRNA resulting in a truncated Trpm8 protein could explain both the presence of immunohistochemical signal and Western-blot signal in VD rat tissue, as well as the absence of Trpm8 specific currents in isolated myocytes of rat VD observed previously [5]. The bioinformatic analysis of a selected sequence shows that the amplified fragment lies in the open frame region of Trpm8 mRNA, and corresponds to the 388-458 amino acid fragment of rat Trpm8 protein. UniProt entry Q8R455 for Trpm8 of R. norvegicus shows that the selected amino acid sequence is located within N-terminal domain of the Trpm8 protein. The N-terminal domain is the largest protein located in the cytoplasm. Studies done by Noyer and Bidaux teams show that Trpm8 mRNA isoforms such as Trpm8α and Trpm8β an be found in human prostate cancer cells. They encode proteins sM8-6 (6 kDa) and sM8-18 (18 kDa). The annotated protein weight of both isoforms differs from one we observed in this work. However, these proteins are translated from splice isoforms of N-terminal domain, possibly observed in this study [4,8]. Further work with this splice isoform of Trpm8 mRNA from rat VD is needed. This includes cloning and sequencing of the amplicon, and it will be a topic of our future research.