Expression of Survivin and Bcl-2 in the Normal Human Endometrium

Molecular Human Reproduction vol.6 no.6 pp. 529–534, 2000

Expression of survivin and Bcl-2 in the normal human endometrium

R.Konno1, H.Yamakawa, H.Utsunomiya, K.Ito, S.Sato and A.Yajima Department of Obstetrics and Gynecology, Tohoku University School of Medicine, 1–1 Seiryo-machi, Aoba-ku, Sendai, 980-8574, Japan 1To whom correspondence should be addressed: Department of Obstetrics and Gynecology, Tohoku University School of Medicine, 1–1 Seiryo-machi, Aoba-ku, Sendai, 980-8574, Japan. E-mail: [email protected] Survivin is a novel inhibitor of apoptosis. It has been reported that survivin is expressed during fetal development and in cancer tissues, but its expression has not been reported in adult tissues. We investigated the expression of survivin in the endometria of women with regular menstrual cycles using reverse transcription–polymerase chain reaction (RT–PCR) and immunohistochemistry, and compared these ﬁndings with Bcl-2, an apoptosis inhibitor. Survivin mRNA was detected by RT–PCR in all samples (nine of nine) of endometrium during the secretory phase, but in only four out of seven samples from endometrium during the proliferative phase, and in none of the atrophic endometrium. Immunohistochemistry demonstrated a survivin protein expression that was strongest in the nuclei of glandular epithelial cells during the late secretory phase. In the proliferative phase, glandular epithelial cells were not stained for survivin. The cyclic changes of survivin and Bcl-2 showed an inverse relationship, with Bcl-2 expression being strongest in the proliferative phase and survivin expression being strongest in the secretory phase. The up-regulation of survivin expression may be due to the concurrent rise in progesterone concentrations during the normal menstrual cycle. Moreover, survivin could play an important role independent of Bcl-2 in physiological homeostasis in the normal endometrium. Key words: Bcl-2/endometrium/immunohistochemistry/RT–PCR/survivin

Introduction Menstruation has been recognized to be the result of Regulation of programmed cell death (apoptosis) preserves ischaemic necrosis in the endometrium (Speroff and Vande normal homeostasis, and tissue and organ morphogenesis Wiele, 1971). However, some authors (Hopwood and Levinson, (Vaux et al., 1994; Nagata, 1997). Aberrations of this process 1975) reported apoptotic bodies in the human endometrium participate in human diseases and may contribute to carcino- by light and electron microscopy. Recently, it has been reported genesis by prolonging cell viability resulting in the that apoptosis appears in the human endometrium during the accumulation of transforming mutations (Thompson, 1995). menstrual cycle by DNA gel fragmentation and in-situ Survivin is a new inhibitor of apoptosis protein (IAP) apoptosis analysis techniques (Kokawa et al., 1996). expressed during development and in human malignancy. The In addition, cyclic Bcl-2 expression in the human endomet- survivin gene was recently identiﬁed by hybridization screening rium has been demonstrated by several investigators (Gompel of human genomic libraries with the cDNA from a factor Xa et al, 1994; Otsuki et al., 1994; Tabibzadeh et al., 1995; Koh receptor, effector cell protease receptor-1 (EPR-1) (Altieri, et al., 1995; Jones et al., 1998). Endometrial glandular cells 1995; Ambrosini et al., 1997). express Bcl-2 during the proliferative phase, but not during Survivin can bind speciﬁcally to the terminal effector cell the late secretory phase. The disappearance of Bcl-2 expression death proteases, caspase-3 and -7 in vitro, and inhibit caspase during the late secretory phase is consistent with the appearance activity and apoptosis in cells exposed to diverse apoptotic of apoptotic cells during this same phase. Bcl-2 expression in stimuli (Tamm et al., 1998). Survivin has been shown to be glandular cells may also be regulated by steroid hormones expressed in 60 cancer cell lines, including breast, colon, (Gompel et al., 1994; Otsuki et al., 1994; Koh et al., 1995; brain, leukaemia/lymphoma, lung, melanoma, ovarian, prostate Tabibzadeh et al., 1995). and renal cancer (Tamm et al., 1998). To our knowledge, there has been no investigation of the Survivin expression has been shown to be present during expression of survivin in normal human endometrium during fetal development. In contrast, survivin transcripts were unde- the menstrual cycle. Thus, the role of survivin and its mechan- tectable in terminally differentiated adult tissues, including ism during menstruation in normal human endometrium is peripheral blood leukocytes, lymph node, spleen, pancreas, unknown. In this study, we examined the expression of survivin kidney, skeletal muscle, liver, lung, brain, and heart (Ambrosini and compared it with Bcl-2 in the normal human endometrium et al., 1997), reminiscent of the expression of another apoptosis during the menstrual cycle by immunohistochemistry and inhibitor, Bcl-2, in these adult organs (LeBurn et al., 1993). reverse transcription–polymerase chain reaction (RT–PCR).

Table I. Immunohistochemical staining scores of survivin and Bcl-2 in human endometrium

Age Survivin Bcl-2 (years) Score Positive Intensity Score Positive Intensity (PϫI) (P) (I) (PϫI) (P) (I)

Early proliferative phase 45 0 0 – 93 3 45 0 0 – 93 3 50 1 1 1 12 4 3 50 0 0 – 93 3 Mid-proliferative phase 44 1 1 1 9 3 3 4231 3 1243 41 0 0 – 12 4 3 Late proliferative phase 49 0 0 – 93 3 48 2 2 1 12 4 3 46 1 1 1 12 4 3 Early secretory phase 42 1 1 1 4 2 2 42 2 2 1 6 2 3 41 9 3 3 3 1 3 40 3 3 1 3 1 3 Mid-secretory phase 43 4 2 2 0 0 – 48 9 3 3 0 0 – 45 4 2 2 3 1 3 45 4 2 2 0 0 – Late secretory phase 43 9 3 3 0 0 – 47 12 4 3 0 0 –

Positivity (P) was scored as follows: 0 ϭϽ5%; 1 ϭ 5–25%; 2 ϭ 25–50%; 3 ϭ 50–75%; and 4 ϭϾ75%. The immunostaining intensity (I) was scored as follows: 1ϩ (weak); 2ϩ (moderate), or 3ϩ (intense).

Materials and methods at least five areas at ϫ400 magnification and assigned to one of the following categories: 0 ϭϽ5%; 1 ϭ 5–25%; 2 ϭ 25–50%; 3 ϭ 50– Immunohistochemistry 75%; and 4 ϭϾ75%. The immunostaining intensity of survivin and Tissue samples of endometria for immunohistochemistry were Bcl-2 were scored as 1ϩ (weak), 2ϩ (moderate) or 3ϩ (intense). Ϯ Ϯ obtained from 20 women (age, mean SD, 44.8 3.09, range 40– The relative intensity of survivin staining defined as 3ϩ was identical 50 years; four early proliferative, three mid-proliferative, three late to the intensity of Bcl-2 staining in stromal lymphocytes. The proliferative, four early secretory, four mid-secretory, two late secret- percentage of positive glandular epithelial cells and the staining ory phase) with regular menstruation (Table I). They underwent intensity were multiplied to produce a weighted score for each case. hysterectomy with the diagnosis of cervical intra-epithelial neoplasia Immunostaining findings of stromal cells, glandular epithelial cells at Tohoku University Hospital, Sendai, Japan, after informed consent obtained from basal layer, and superficial epithelial cells in the was obtained. They had received no hormone treatment prior to endometrium were described without scoring. surgery. Formalin-fixed and paraffin-embedded samples were prepared Statistical analyses were carried out using the Kruskal–Wallis test, for histological diagnosis and immunohistochemical analysis. There Scheffe´’s F test and Pearson’ s correlation coefficient test. P Ͻ 0.05 were no pathological findings related to the endometrium during post- was considered to be statistically significant. operative pathological examination of all samples used in the present study. Endometrial dating was performed according to Noyes’ criteria RT–PCR (Noyes et al., 1950). Immunohistochemistry was performed using the LSAB 2 Kit (Dako A total of 19 normal endometrial samples were used in this study, Japan, Kyoto, Japan) and an antigen retrieval method. The primary including 16 samples from women with regular menstrual cycles Ϯ Ϯ antibodies were polyclonal rabbit anti-human survivin (Surv 11; Alfa (age, mean SD, 44.6 5.21, range 33–51 years; proliferative Diagnostic International, San Antonio, TX, USA) and monoclonal phase, seven cases; secretory phase, nine cases) and three samples Ϯ Ϯ mouse anti-human Bcl-2 (Dako Japan). After deparaffinization, sec- from post-menopausal women (age, mean SD, 52.7 6.35, range tions were treated with methanol/hydrogen peroxide for 5 min to 49–60 years). These tissues were obtained after hysterectomy in block endogenous peroxidase. To retrieve masked antigens, the slides patients diagnosed with cervical intraepithelial neoplasia, following were immersed in citrate buffer (pH 6.0), and heated in an autoclave informed consent. They had normal regular menstrual cycles and had for 5 min at 120°C. The slides were incubated for 60 min with the received no hormone treatment prior to surgery. The endometrial primary antibody, followed by a 30 min incubation with the biotinyl- tissues collected were assessed for histopathological diagnosis accord- ated secondary antibody, and then with peroxidase-labelled streptavi- ing to previously described criteria (Noyes et al., 1950), and the din for 30 min. Negative control slides in the absence of primary remaining portions of the samples were frozen at –80°C until assay. antibody were included for each staining. Finally, 3,3Ј-diamino- benzidine was used to develop the colour, and haematoxylin was RNA extraction and first strand cDNA synthesis using used for counterstaining. oligo(dT) For the semi-quantitative evaluation of survivin and Bcl-2 expres- Total RNA was isolated from frozen human endometrial tissue using sion, we used a scoring method modified by Sinicrope and Lu (Lu a commercially available extraction method (Isogen, Nippon Gene et al., 1998). The mean percentage of positive glandular epithelial Inc, Tokyo, Japan). First-strand cDNA was synthesized from 3–4 µg cells in the functional layer of the endometrium was determined in of total RNA using an Oligo(dT)12–18 primer and the Superscript™ 530 Survivin and Bcl-2 expression in normal human endometrium

Figure 1. Expression of survivin in endometrial glandular epithelial cells throughout the menstrual cycle. P1 ϭ early proliferative phase; P2 ϭ mid-proliferative phase; P3 ϭ late proliferative phase; ϭ ϭ ϭ Figure 2. Expression of Bcl-2 in endometrial glandular epithelial S1 early secretory phase; S2 mid-secretory phase; S3 late ϭ secretory phase. Staining score was calculated as described in the cells throughout the menstrual cycle. P1 early proliferative Ϯ ϭ phase; P2 ϭ mid-proliferative phase; P3 ϭ late proliferative phase; text. Error bar SD. Kruskal–Wallis test, P 0.0096. P values of ϭ ϭ ϭ Scheffe’s F-test are indicated above. S1 early secretory phase; S2 mid-secretory phase; S3 late secretory phase. Staining score was calculated as described in the text. Error bar Ϯ SD. Kruskal–Wallis test, P ϭ 0.0047. P values of Preamplification System (Superscript™ II RT; 200 IU; Gibco BRL, Scheffe’s F-test are indicated above. Grand Island, NY, USA). Control RNA is included in this kit to verify system performance. One µl (50 ng) of control RNA was used as a template for first strand cDNA synthesis. Following treatment basal layer and the functional layer. Immunostaining was with RNase H, cDNA samples were stored at –20°C. nuclear in pattern within the glandular cells of the functional Amplification of the target cDNA layer (Figure 3C), while cytoplasmic in pattern in the basal layer cells (Figure 3D). The intensity was strongest in the Amplification of cDNA was performed with an Ambion Gene Specific Human Survivin Relative RT–PCR Kit (Ambion Inc, Austin, TX, lower portion of the functional layer, and peaked during the USA). Primer sequences for human survivin used in PCR amplification late secretory phase. Survivin immunostaining was weak in were: 5Ј GTGAATTTTTGAAACTGGACAG, 3Ј CCTTTCCTAAGA- the basal layer. Stromal cells were negative for survivin. CATTGCTAAG (Genbank Accession no. U75285). PCR was carried out in a reaction volume of 25 µl for 5 min at 94°C for initial Immunohistochemistry for Bcl-2 denaturing, followed by 35 cycles at 94°C for 1 min, 57°C for 1 min, Proliferative phase and 72°C for 1.5 min, on a TaKaRa PCR Thermal Cycler MP The staining pattern was uniformly intense in the cytoplasm (TaKaRa, Tokyo, Japan). Primers were confirmed in multiplex RT– of glandular epithelial cells (Figure 3E). Stromal cells were PCR on mouse and human RNA samples. Primers yield a gene- negative for Bcl-2, whereas rare positive-staining cells seemed specific product of the correct size and multiplex efficiently with the to be lymphocytes. Surface epithelial cells were mostly negative supplied 18S primers. Where possible, primers were designed to flank or weakly positive for immunoreactive Bcl-2 protein. intron sequences. PCR products were analysed on a 2% agarose gel and visualised by staining with ethidium bromide. Secretory phase Immunostaining for Bcl-2 disappeared from glandular epithe- Results lial cells as soon as secretory phase characteristics appeared (Figure 3F). In the premenstrual phase, the basal layer of the Results of immunohistochemistry are summarized in Table I glandular cells was positive for Bcl-2 immunoreactive protein, and Figures 1 and 2. as was found in the proliferative phase. Immunopositive stromal cells increased during the secretory phase, and peaked Immunohistochemistry for survivin in the late secretory phase. Of the predecidualized stromal Proliferative phase cells, some stromal granular lymphocytes were positive for Bcl- The glandular epithelium and stroma were negative or very 2, but predecidual cells were negative. The surface epithelium weakly positive for survivin immunostaining (Figure 3A). The stained heterogenously and weakly for Bcl-2 protein. surface epithelium (Figure 3B) and basal layer cells stained weakly and heterogenously in proliferative phase samples. Relationship between survivin and Bcl-2 Staining was localized to the cytoplasm. Immunohistochemistry data on survivin and Bcl-2 staining in Secretory phase the normal endometrium during the menstrual cycle are shown Immunolocalization of survivin was confined to the glandular in Figures 1 and 2. The cyclic changes of survivin and Bcl-2 epithelial cells. The staining pattern was different between the protein expression demonstrated an inverse relationship 531 R.Konno et al.

Figure 3. Immunohistochemical staining for (A–D) survivin and (E, F) Bcl-2 in the human endometrium. (A) Survivin expression was negative or weak in glandular epithelial cells and stromal cells during the proliferative phase. (B) The surface epithelium was stained weakly and heterogeneously in proliferative phase samples. Staining was localized to the cytoplasm. (C) Immunolocalization of survivin was restricted to the nuclei of glandular epithelial cells in the secretory phase. (D) Immunostaining for glandular cells in basal layer in the secretory phase was heterogenous and cytoplasmic in pattern. (E) Bcl-2 staining was observed in the cytoplasm of glandular epithelial cells and lymphocytes in the proliferative phase. (F) Expression of Bcl-2 was not observed in glandular epithelial cells, and positive immunostaining was localized to stromal lymphocytes in the late secretory phase. between the proliferative phase and the secretory phase (Figure observed to be down-regulated and/or undetectable in many 4). In the secretory phase, immunostaining was positive for normal adult tissues (Ambrosini et al., 1997), but present in survivin but negative for Bcl-2, while in the proliferative the thymus and placenta. The expression of survivin has not, phase, it was negative for survivin but positive for Bcl-2. however, been investigated in normal human endometrium. This is the first study investigating survivin expression in RT–PCR of survivin the normal endometrium as it relates to hormonal fluctuations In the normal endometrium, survivin mRNA expression was during the menstrual cycle. Prior to obtaining our results, detected using RT–PCR in four out of seven endometrial we expected that survivin might be expressed during the samples obtained during the proliferative phase, in all of nine proliferative phase, as was observed for Bcl-2 expression in samples from secretory phase endometria, and in none out of previous reports (Gompel et al, 1994; Otsuki 1994; Saitoh three inactive endometria (Figure 5). et al, 1999). In fact, the results from our experiments by immunohistochemistry and RT–PCR clearly show that survivin Discussion is expressed during the secretory phase, and is rare or absent It has been reported that survivin is expressed during embryonic in the proliferative phase. and fetal development. However, its expression has been Expression of Bcl-2 has been demonstrated in normal 532 Survivin and Bcl-2 expression in normal human endometrium endometrium during the menstrual cycle, being associated with the role of survivin in the mechanism of apoptosis in normally oestrogen and progesterone status. Our finding that Bcl-2 cycling endometrium. immunohistochemical staining predominated in glandular cells, We pointed out in the present study that the immunostaining peaking during the proliferative phase and disappearing at the pattern for survivin is different between glandular epithelial onset of the secretory phase, coincides with previous studies cells during the secretory phase and superficial cells during (Gompel et al, 1994; Otsuki et al., 1994; Koh et al., 1995). the proliferative phase. Nuclear staining was observed in Among the regulators of programmed cell death (apoptosis), glandular epithelial cells of the functional layer during the IAP have recently attracted considerable attention for their secretory phase. On the other hand, immunostaining was ability to suppress an evolutionarily conserved step in apoptosis cytoplasic in the proliferative phase, similar to that of the (Clem and Duckett, 1997), potentially involving direct caspase basal layer during the secretory phase. Furthermore, in previous inhibition (Deveraux et al., 1997). It has been reported that reports concerning cancer tissues, immunohistochemical stain- survivin binds specifically to terminal effector cell death ing was observed in the cytoplasm and not in nuclei (Ambrosini proteases, caspase-3 and -7 in vitro and inhibits caspase activity et al., 1997; Lu et al., 1998). It is probable that the difference of and cell death in cells exposed to diverse apoptotic stimuli immunolocalization for survivin is due to survivin–microtubule (Tamm et al., 1998). Moreover, survivin is expressed in the interactions. Since there is insufficient data to state a clear G2/M phase of the cell cycle in a cell-regulated manner. At relationship between the staining patterns and any G2/M the beginning of mitosis, survivin associates with microtubules checkpoints, further study is necessary. of the mitotic spindle in a specific and saturable reaction that In addition, we found that the expression of caspase-3 was is regulated by microtubule dynamics. Disruption of survivin- demonstrated immunohistochemically in endometrial glandular microtubule interactions results in the loss of survivin’s cells during the secretory phase, and not in the proliferative anti-apoptosis function and increased caspase-3 activity, a phase (R.Konno, unpublished data). It is likely that survivin mechanism involved in cell death, during mitosis. These results may be expressed antagonistically to caspase-3 expression in suggest that survivin may counteract a default induction of order to keep the balance between inhibition and induction of apoptosis in the G2/M phase of the cell cycle (Fengzhi et al., apoptosis in the secretory phase. It is reasonable to suggest 1998). The above information is seen to be of value considering that the up-regulation of survivin expression by progesterone could play a role in the physiological homeostasis in normal endometrium. In the present study, we found that immunohistochemical staining of survivin was restricted to endometrial glandular epithelium and was negative in endometrial stromal cells, while stromal cells were positive for Bcl-2. Previous studies (Koh et al., 1995; Jones et al., 1998; Konno et al., 1999) have demonstrated that most of the endometrial stromal granulocytes during the secretory phase were CD56-positive natural killer cells, containing perforin and granzyme B, which are cytotoxic granules inducing apoptosis (Konno et al., 1999), and revealed that endometrial stromal granulocytes were also expressing caspase-3 (R.Konno, unpublished data). Moreover, it has been suggested that endometrial stromal granulocytes expressing Bcl-2 and Ki67 simultaneously do not undergo apoptosis (Jones et al., 1998). We speculate that endometrial stromal cells Figure 4. Correlation between survivin and Bcl-2 staining scores on a per sample basis (r ϭ –0.790, P Ͻ 0.0001). Data points are during the secretory phase may antagonize caspase-3-mediated overlapped between 2 (#), 3 (§), and 4 (*) samples. apoptosis via the Bcl-2 pathway and/or other cell proliferative mechanism, but not using survivin.

Figure 5. Survivin expression in 16 menstruating women and three menopausal women, detected by reverse transcription–polymerase chain reaction (RT–PCR). M ϭ molecular weight marker; P ϭ positive control; survivin mRNA is observed as a 243 bp signal; 18S served as an internal control (495 bp). 533 R.Konno et al.

In summary, survivin expression was demonstrated in the endometrium during the secretory phase in the human adult. It is suggested that survivin may play a role in the mechanism of endometrial menstruation independent of Bcl-2 expression.

Acknowledgements This work was supported in part by Grants-in-Aid from the Hiromi Medical Research Foundation and the Ministry of Education, Science, Sports and Culture, Japan.

References Altieri, D.C. (1995) Xa receptor EPR-1. FASEB J., 9, 860–865. Ambrosini, G., Adida, C. and Altier, D.C. (1997) A novel-anti-apoptosis gene, Survivin, expressed in cancer and lymphoma. Nature Med., 3, 917–921. Clem, R.J and Duckett, C.S. (1997) The IAP genes – unique arbitrators of cell death. Trend Cell Biol., 7, 337–339. Deveraux, Q.L, Takahashi, R., Slavesen, G.S. and Reed, J. (1997) X-linked IAP is a direct inhibitor of cell-death proteases. Nature, 388, 300–304. Fengzhi, L., Ambrosini, G., Chu, E.Y. et al. (1998) Control of apoptosis and mitotic spindle checkpoint by Survivin. Nature, 396, 580–584. Gompel, A., Sabourin, J.C., Martin, A. et al. (1994) Bcl-2 expression in normal endometrium during the menstrual cycle. Am. J. Pathol., 144, 1195–1202. Hopwood, D. and Levison, D. (1975) Atrophy and apoptosis in the cyclical human endometrium. J. Pathol., 119, 159–166. Jones, R.K. Searle, R.F. Stewart, J.A. et al. (1998) Apoptosis, bcl-2 expression, and proliferative activity in human endometrial stroma and endometrial granulated lymphocytes. Biol. Reprod., 58, 995–1002. Koh, E.A., Illingworth, P.J., Duncan, W.C. and Critchley, H.O. (1995) Immunolocalization of bcl-2 protein in human endometrium in the menstrual cycle and simulated early pregnancy. Hum. Reprod., 10, 1557–1562. Kokawa, J., Shikone, T. and Nakano, R. (1996) Apoptosis in the human uterine endometrium during the menstrual cycle. J. Clin. Endocrinol. Metab., 81, 4111–4147. Konno, R., Igarashi, T., Okamoto, S. et al. (1999) Apoptosis of human endometrium mediated by perforin and granzyme B of NK cells and cytotoxic T lymphocytes. Tohoku J. Exp. Med., 187, 149–155. LeBurn, D.P., Warnke, R.A. and Cleary, M.L. (1993) Expression of Bcl-2 in fetal tissue sugests a role in morphogenesis. Am. J. Pathol., 142, 743–753. Lu, C.D., Altieri, D.C. and Tanigawa, N. (1998) Expression of a novel antiapoptosis gene, Survivin, correlated with tumor cell apoptosis and p53 accumulation in gastric carcinomas. Cancer Res., 58, 1808–1812. Nagata, S. (1997) Apoptosis by death factor. Cell, 88, 355–365. Noyes, R., Hertig, A. and Rock, J. (1950) Dating the endometrial biopsy. Fertil Steril., 1,3–25. Otsuki, Y., Misaki, O., Sugimoto, O. et al. (1994) Cyclic bcl-2 gene expression in human uterine endometrium during menstrual cycle. Lancet, 344,28–29. Saitoh, Y., Yaginuma, Y. and Ishikawa, M. (1999) Analysis of Bcl-2, Bax and Survivin genes in uterine cancer. Int. J. Oncol., 15, 137–141. Speroff, L. and Vande Wiele, R. (1971) Regulation of the human menstrual cycle. Am. J. Obstet. Gynecol., 109, 234–247. Tabibzadeh, S., Zupi, E., Babaknia, A. et al. (1995) Site and menstrual cycle- dependent expression of proteins of the tumour necrosis factor (TNF) receptor family, and BCL-2 oncoprotein and phase-speciﬁc production of TNF-α in human endometrium. Hum. Reprod., 10, 277–86. Tamm, I., Wang, Y., Sausville, E. et al. (1998) IAP-family protein Survivin inhibits caspase activity and apoptosis induced by Fas (CD95), Bax, Caspases, and anticancer drugs. Cancer Res., 58, 5315–5320. Thompson, C.B. (1995) Apoptosis in the pathogenesis and treatment of disease. Science, 267, 1456–1462. Vaux, D.L., Haecker, G. and Strasser, A. (1994) An evolutionary perspective on apoptosis. Cell, 76, 777–779.

Received on November 24, 1999; accepted on March 13, 2000

534