(C) 1999 Wiley-Liss, Inc. Induction of Heat Shock Protein Expression in

OBJECTIVE: The 70kD heat shock protein (Hsp70), induced when cells are subjected to environmental stress, prevents the denaturation and incorrect folding of polypeptides and may expedite replication and transmission of DNA and RNA viruses. We analyzed whether messenger RNA (mRNA) for Hsp70 was expressed following exposure of a cultured human cervical cell line (HeLa cells) to human semen or in cervical cells from sexually active women. STUDY DESIGN: HeLa cells were co-cultured with a 1:50 dilution of semen from four men or with purified spermatozoa or cell-free seminal fluid. Endocervical swabs were acquired at mid-cycle from 53 women. Heat shock protein 70 mRNA was detected by a reverse transcriptase-polymerase chain reaction utilizing specific primer pairs and analysis on agarose gels. In cervical cells Hsp70 mRNA was measured identically followed by hybridization with an Hsp70-specific internal probe and detection by enzyme-linked immunosorbent assay (ELISA). Cervical immunoglobulin A (IgA) antibodies to the human Hsp70 were determined by ELISA. RESULTS: HeLa cell-semen co-culture resulted in the induction of Hsp70 mRNA. In addition, cell-free seminal plasma and motile sperm incubated individually with HeLa cells also induced this mRNA. Heat shock protein 70 mRNA was detected in 28 (52.8%) of 53 endocervical samples obtained from women at various time points following intercourse. The percentage of samples expressing this mRNA was 37.5% at less than 10 hours, 64.3% at 10 hours, 70% at 11 hours, and between 36% and 50% at later times after semen exposure. The detection of cervical IgA antibodies to the Hsp70 was highly associated with Hsp70 gene transcription. CONCLUSION: Human semen induces transcription of Hsp70 in cervical epithelial cells.

also be induced following exposure to a wide variety of physio-chemical insults: temperature variation, steroids, amino acid analogues, heavy metals, and oxidative damage, e Mammalian cells produce Hsps in response to infection by viral or bacterial pathogens, and this response has been indicated to exert a cytoprotective effect during disease pathogenesis. A number of investigators have demonstrated that induction of transcription of the gene coding for the inducible heat shock protein 70 (Hsp70) led to an inhibition of transcription of genes coding for the pro-inflammatory cytokines interleukin-1 and tumor necrosis factor-c. [3][4][5][6] Tran-*Correspondence to: Jan C. Jeremias MSc, Department of Obstetrics and Gynecology, Weill Medical College of Cornell University, 515 East 71st Street, New York, NY 10021. E-mail: jaj2003@mail.med.cornell.edu scriptional induction of Hsps requires activation of a heat shock factor, a transregulatory protein that attaches to specific promoter elements located upstream of several Hsp genes.
Seminal fluid is rich in prostaglandins, polyamines, zinc, proteases, and other enzymes, which may initiate a stress response and thereby induce transcription of HspT0 in endocervical cells. Contact between sperm and endocervical cells may also bring about a stress response. Spermatozoa are capable of associating with, and even penetrating into, somatic cells. 7 During coitus, the pH in the vagina changes from 4.0-4.5 to approximately 7.0.
Since an acidic pH suppresses the synthesis of HspT0, 8 neutralization of the pH in the vagina by semen would create a milieu, at this site, acceptable for the synthesis of HspT0. Therefore, it was of interest to determine whether semen induced HspT0 messenger RNA (mRNA) in cells of the female genital tract. The actuality of such induction would suggest an additional mechanism limiting immunity to spermatozoa: prevention of proinflammatory cytokine synthesis and T-lymphocyte activation secondary to initiation of HspT0 gene transcription. In this communication we demonstrate that human semen, as well as cell-free seminal fluid and motile spermatozoa, induced HspT0 mRNA transcription in a cell line derived from human cervical cells. The HspT0 mRNA was also identified in cells of the endocervix from sexually active women after intercourse.

Processing of Semen
Semen samples were obtained by masturbation. Following liquefaction, motile spermatozoa were isolated by overlaying the semen samples with an equal volume of sterile warm (37C) phosphate buffered saline (PBS), incubated at 37C for 60 minutes, and collecting the upper PBS layer. The remaining semen was centrifuged, and cell-free seminal fluid was sterilized with a 0.2 pm syringe filter.
HeLa/Semen Experiments Cells were grown in 24 well flat bottom plates until confluent, after which the monolayers were washed with Hank's Balanced Salt Solution (HBSS). Unfractionated semen, seminal fluid, or motile sperm, diluted as indicated in the individual experiments, was added to each well in duplicate. After an overnight incubation in a 37C, 5% CO z incubator, 0.5 mL of 0.25% Trypsin-lmM edetic acid (EDTA) was added to each culture and like wells were pooled and washed twice with HBSS. The collected cells were pelleted by centrifugation at 6500 rpm and RNA extracted as described below. Viability was determined on an aliquot of each cell culture by vital dye exclusion (0.4% trypan blue).

Subjects
The study population was comprised of 53 married women of reproductive age. All subjects had been tested for aerobic and anaerobic organisms within the past 12 months and treated if culture positive. All were tested for Chlamydia trachomatis by the Amplicor polymerase chain reaction (PCR; Roche Diagnostics, Branchberg, NJ) at the time of sample collection, and none were positive. At the time of the study, none of the women were using contraception.

Cervical Sample Preparation
Cervical samples were obtained from all women at mid-cycle by inserting a Dacron swab into the endocervix, twirling the swab, and then removing it into a tube containing 0.5 mL PBS. Samples were vortexed, and liquid was extracted from the swab using a sterile Pasteur pipette. Cells were pelleted by centrifugation and the supernatant removed and frozen at -80C until used for the determination of anti-Hsp70 antibodies. The cell pellet was washed three times with PBS, and RNA was immediately extracted as described below.

RNA Isolation
Cell pellets were resuspended in 0.2 mL 10-mM Tris-HC1 pH 7.5, containing 0.15 M NaC1, 1.5 mM MgC1 z, 0.65% Nonidet P-40 detergent, and pL RNAguard ribonuclease inhibitor (Pharmacia, Piskataway, NJ), to lyse the cells and release the nucleic acid. Following centrifugation for 5 rain. at 6500 rpm, the supernatant was removed, admixed with an equal volume of 7 M urea, 1% sodium dodecyl sulfate, 0.35 M NaC1, 10 mM EDTA, and 10 mM Tris-HC1. RNA was extracted by addition of an equal volume of 50:50:1 phenol:chloroform:isoamyl alcohol. The samples were vortexed and centrifuged at 12,500 rpm for 4 min. at 4C. The RNA underwent a double extraction procedure to ensure RNA purity. RNA was precipitated at -20C overnight in 100% ethanol and'3 M sodium acetate, washed with -20C 70% ethanol, dried by desiccation, and resuspended in diethyl pyrocarbonate-treated water. Samples were stored at -80C until reverse transcription (RT) and PCR were performed.

Deoxyribonuclease Treatment and Reverse
Transcription of RNA Prior to RT of the RNA, samples were treated for 15 rain. with unit of amplification grade Deoxyribonuclease (DNase 1, Gibco B RL Life Technologies, Grand Island, NY) to destroy any contaminating DNA. DNase activity was removed by the addition of 2 mM EDTA and heating for 10 rain. at 65C.
As a control to guarantee that cDNA synthesis occurred in our samples, each sample was examined for the presence of [3-actin cDNA. Samples of cDNA were amplified, using the same reaction conditions as above, in the presence of 6 pmol of human [3-actin specific primers: reverse 5'-CGT CAT ACT CCT GCT TGC TGA TCC AAT CTGC-3' and forward 5'-ATC TGG CAC CAC ACC TTC TAC AAT GAG CTG CG-3' (Oswel DNA Service, Southhampton, England).

Quantitation of Amplified PCR Products by
Enzyme-Linked Immunosorbent Assay Amplified PCR products were detected by enzyme-linked immunosorbent assay (ELISA) following hybridization with an Hsp70-specific internal probe, to increase sensitivity and specificity of mRNA detection from the human endocervicalderived samples. Aliquots (10pL) of digoxigeninamplified PCR products were denatured at 25C for 10 rain. and hybridized with 7.5 pmol/mL of a biotinylated Hsp70-specific oligonucleotide probe: biotin 5'-GCA AGG TGG AGA TCA TCG CCA ACG ACC AGG-3' (Oswel DNA Service). Digoxigenin-labeled biotin PCR complexes were then incubated in a streptavidin-coated microtitre plate for 3 hr. at 51.1C, and the bound product was detected using peroxidase conjugated anti-digoxigenin antibody (Bochringer Mannhcim) and the colorimetric substrate ABTS (2,2'-azino-di-[3ethylbenzthiazoline sulfate (6)    The presence of cervical anti-Hsp70 IgA correlated with detection of cervical Hsp70 mRNA (P < 0.0001); this antibody was present in eight (80%) of 10 women with detectable Hsp70 mRNA and in only two (7.7%) of 26 women without detectable Hsp70 mRNA ( Table 2). The prevalence of IgA antibodies to Hsp70 increased with the time since last semen exposure (P 0.008). Immunoglobulin A anti-Hsp70 was identified in the cervices of five (71.4%) of seven women who had sexual intercourse greater than 48 hours prior to sample collection, in four (17.4%) of 23 women exposed to semen from 10 to 16 hours earlier, and in one (12.5%) of eight women who had intercourse less than 10 hours before sample collection (Table 3). An increased production of antibodies in women previously sensitized to Hsp70 following re-exposure to newly synthesized Hsp70 protein after intercourse is consistent with the induction of a secondary immune response.

DISCUSSION
Hsp 70 gene transcription was induced in vivo and in vitro in endocervix-derived cells following exposure to semen. Cell-free seminal fluid and motile sperm were also capable of initiating this response.
The cells that actually produce Hsp70 mRNA have not yet been identified.
The presence of HspT0 has been identified in human sperm and seminal fluid. However, in all our experiments, all nonadherent semen components were removed by washing, and we were unable to detect HspT0 mRNA in spermatozoa.
Therefore, we are confident that the source of HspT0 was the HeLa cells. Identification of Hsp70 mRNA in cervical cells obtained from women after sexual intercourse parallels the in vitro observations with cultured cells. The correlation between HspT0 mRNA and cervical IgA antibodies to HspT0 in cervical cells following exposure to semen further suggests that Hsp70 gene transcription P 0.008. and the subsequent production of Hsp70 protein stimulated a localized immune response.
There is a possibility that the HspT0 in semen could have contributed to the cervical anti-Hsp70 IgA response. However, none of the semen samples used in the HeLa cell experiments contained HspT0 mRNA. Notwithstanding, the semen samples from the male partners of the women in our study were not available and, therefore, were not assayed for Hsp70 protein.
Heat shock proteins accumulate within virally infected cells and appear to play a significant role in some viral infections. Hsp 70 specifically is intimately associated with the life cycles of numerous RNA and DNA viruses. Hsp 70 may be required to promote the transcription of viral genes, assembly of viral polypeptides, and/or virus infectivity. 9,1 During vaccinia infection, Hsp70 mRNA directly associates with intracellular vaccinia proteins and may act as a chaperonin in the assembly of this virus. 11 Induction of Hsps by elevated temperature upregulates virus production from cell lines chronically infected with human immunodeficiency virus (HIV). lz Hsp induction by heat or chemicals activated the long terminal repeat (LTR) of HIV. In HeLa cells transfected with the HIV-1 LTR, Hsp induction by heat induced the transcription of the HIV-1 LTR and the Hsp70 heat shock promoter. 13 Elevations in HIV reverse transcriptase were observed for 72 hours following heat shock in infected lymphocytic and monocytic cell lines. Cells allowed to recover after heat shock first showed a lag in release of viral particles, followed by accelerated viral release in stressed cells. Physiological levels of heat shock were not able to induce virus production from cells in vitro. However, these temperatures were able to act synergistically with interleukin-6 and granulocyte macrophage-colony stimulating factor to enhance virus production. 3 Viral activation in response to environmental stress has also been identified for cytomegalovirus (CMV). 14 A DNA sequence present in the enhancer element of CMV, similar to a decameric sequence in the NF-KB binding site of the HIV LTR, resembled the heat shock element core consensus regulatory sequence. 14,1 This suggested that either Hsps may activate CMV genes or that Hsp genes and CMV genes respond to similar stimuli. Initiation of Hsp70 transcription thus might facilitate the reactivation of latent viral infections, as well as augment viral replication and transmission "from an infected donor to a noninfected recipient by promoting enhanced viral production.
Since viruses such as HIV and CMV are sexually transmitted, it was of interest to determine whether semen and its components could activate Hsp transcription in cervical cells. Such expression might facilitate the reactivation of latent viral infections as well as augment viral transmission from an infected donor to a noninfected recipient.
Whether semen-induced Hsp70 is beneficial in inhibiting immune responses to spermatozoa, harmful in facilitating viral infectivity, or inhibits immune activation in response to other sexually transmitted infections may depend on a number of factors. The capacity of individual women's cells to transcribe, translate, and degrade HspT0 mRNA is under genetic control. In addition, the components of individual semen samples to induce Hsp70 expression may differ. The presence of virally infected cells or microbial pathogens in the female genital tract or in the ejaculate may influence the consequences of Hsp70 transcription. The observed difference between women at various times after intercourse may be related to these factors.
The effect of semen on Hsp70 transcription in additional epithelial and nonepithelial cell lines is currently under investigation. The effect of sexual intercourse on cytokine expression and whether Hsp70 expression differs in women using various forms of birth control is also being explored.