Inducible Guanylate Binding Protein 1 Is Induced in Human ...

Download PDF
49 downloads 122 Views 472KB Size Report
Comment
and after approval of the institutional review board at the Nassau. County Medical Center ... exposed to XAR-5 film (Eastman Kodak Co., Rochester, NY) for 72 h.
0021-972X/01/$03.00/0 The Journal of Clinical Endocrinology & Metabolism Copyright © 2001 by The Endocrine Society

Vol. 86, No. 6 Printed in U.S.A.

Messenger Ribonucleic Acid Encoding InterferonInducible Guanylate Binding Protein 1 Is Induced in Human Endometrium within the Putative Window of Implantation* SUSHMA KUMAR, QUANXI LI†, ANURADHA DUA, YU-KANG YING, MILAN K. BAGCHI‡, AND INDRANI C. BAGCHI Population Council and The Rockefeller University (Q.L., M.K.B., I.C.B.), New York, New York 10021; and Department of Obstetrics/Gynecology (S.K., A.D., Y.-K.Y.), Nassau University, East Meadow, New York 11554 ABSTRACT The putative window of embryo implantation in the human opens between days 19 –24 of the menstrual cycle. During this period, the endometrium undergoes distinctive structural and functional changes orchestrated by steroid hormones, growth factors, and cytokines to attain a receptive phase in which it acquires the ability to implant the developing embryo. A major challenge in the study of human reproduction is to identify the molecular signals that participate in the establishment of this critical receptive phase in the context of the natural cycle. Toward this goal, we analyzed human endometrial biopsies at various days of the menstrual cycle by employing messenger RNA (mRNA) differential display technique. We isolated several complementary DNAs representing genes that are either up- or down-regulated within the putative window of implantation. We identified one of these genes as that encoding interferon (IFN)-inducible guanylate-binding protein 1 (or GBP1), which possesses GTPase activity. Analysis of endometrial biopsies by North-

I

MPLANTATION OF THE blastocyst into the uterine wall is a feature unique to mammalian reproduction. The process begins with the initial adherence of the blastocyst to the uterine surface epithelium, followed by intimate interaction of the blastocyst trophectoderm with epithelial cells, leading to the progressive phases of implantation (1– 6). Studies by Psychoyos demonstrated that rat uterus can accept the blastocyst for implantation only for a brief period of time on day 5 of gestation, known as the receptive phase (7–9). Although previous research has indicated that multiple maternal factors, such as the steroid hormones, growth

Received September 25, 2000. Revision received December 28, 2000. Accepted January 7, 2001. Address all correspondence and requests for reprints to: Indrani C. Bagchi, Ph.D., Department of Veterinary Biosciences, University of Illinois at Urbana–Champaign, 2001 South Lincoln, Urbana, Illinois 61802. E-mail: [email protected]. * This work was supported by NIH Grants R01-HD-34527 (to I.C.B.) and National Cooperative Program on Markers of Uterine Receptivity for Blastocyst Implantation U01-HD-34760 (to I.C.B.). M.K.B. is supported by NIH Grants RO1-DK-50257 and U54-HD-13541. † Present address: Department of Veterinary Biosciences, University of Illinois at Urbana—Champaign, Urbana, Illinois 61801. ‡ Present address: Department of Molecular and Integrative Physiology, University of Illinois at Urbana—Champaign, Urbana, Illinois 61801.

ern blotting and RT-PCR demonstrated that GBP1 mRNA is specifically induced at the midsecretory phase of the menstrual cycle. In situ hybridization analysis revealed that GBP1 mRNA expression is localized in the glandular epithelial cells as well as in the stroma in the immediate vicinity of the glands. We observed that treatment of human endometrial adenocarcinoma cell, Ishikawa, with IFN-␥ or IFN-␣ markedly induced the expression of GBP1 mRNA. IFN-␥ was, however, a more potent inducer of GBP1 than IFN-␣. Consistent with this finding, the temporal profile of GBP1 expression during the menstrual cycle resembled that of IFN-␥ mRNA more closely than that of IFN-␣, predicting a regulatory role of IFN-␥ in GBP1 expression in midsecretory human endometrium. Although the precise function of GBP1 in the receptive human uterus remains unclear, its unique expression overlapping the putative window of implantation suggests that it might serve as a useful marker of uterine receptivity in the human. (J Clin Endocrinol Metab 86: 2420 –2427, 2001)

factors, and cytokines, regulate the events leading to implantation, relatively little is known of the molecular mechanisms by which these effectors promote uterine receptivity (2, 10). In humans, the ovum is fertilized in the fallopian tube, arrives in the uterine cavity around day 17 (day 14 is taken as the day of ovulation of a 28-day cycle), and remains there unattached until about day 18; implantation then occurs between days 18 –24 (11–15). The precise timing and molecular basis of the receptive window in the human remain undefined. To identify the molecular signals that participate in the establishment of a receptive human endometrium, we employed the messenger RNA (mRNA) differential display (DD) procedure to compare mRNAs obtained from the proliferative vs. midsecretory endometrium (16 –18). We isolated several complementary DNA (cDNA) clones representing genes that are either up- or down-regulated during the midsecretory phase of the cycle. We identified one of these genes as that encoding interferon (IFN)-inducible guanylate-binding protein 1 (or GBP1) (19 –20). In this report, we show that the expression of GBP1 is markedly induced in glandular epithelial and the surrounding stromal cells of human endometrium precisely at the midsecretory phase of the menstrual cycle, overlapping the putative window of implantation. The profile of GBP1 expression during the cycle closely resembled that of IFN-␥ mRNA, indicating that GBP1 is likely

2420

GENE REGULATION BY IFNs

to be an endogenous target gene of this cytokine in the periimplantation human endometrium. Although the precise functional role of GBP1 during implantation remains unclear, our studies revealed that GBP1 expression could be used as a potential marker of uterine receptivity in the human. Materials and Methods Endometrial tissues Human endometrial tissues were obtained as part of endometrial curettage from healthy, nonpregnant females between the ages of 25– 40, before elective sterilization with informed consent. These tissues were obtained in accordance with the rules and regulations of the institution and after approval of the institutional review board at the Nassau County Medical Center. Endometrial tissues were transported to the laboratory in HBSS on ice. Tissues were then snap-frozen in liquid nitrogen and stored at ⫺70 C until further use. Endometrial tissues were classified according to serum levels of estradiol and progesterone, and dating was performed based on the criteria of Noyes et al. (21).

DD Total RNAs were extracted from endometrial biopsies at different days of the menstrual cycle using a Trireagent isolation system (Molecular Research Center, Inc., Cincinnati, OH). RNA samples were freed of DNA after treatment with deoxyribonuclease I (Genehunter Corporation, Brookline, MA) and subjected to DD reactions as described previously (16 –18), with certain modifications. Briefly, 2 ␮g DNA-free total RNA were reverse-transcribed with 200 U MMLV reverse transcriptase (Promega Corp., Madison, WI) in the presence of 1 ␮mol/L T12 mA, T12 mC, or T12 mG primer (Genehunter Corporation), where M is a mixture containing dG, dA, and dC. The reaction was performed at 37 C for 1 h. One tenth of this reaction was then used in a PCR amplification reaction containing 2 ␮mol/L each of deoxynucleotide triphosphates, 10 mCi of [35S] deoxy-ATP (Amersham Pharmacia Biotech, Arlington Heights, IL), 2 primers: 1 mmol/L of a T12 oligonucleotide and 0.2 ␮mol/L of one of the five arbitrary decamers, AP-1 (5⬘-AGCCAGCGAA-3⬘); AP-2 (5⬘GACCGCTTGT-3⬘); AP-3 (5⬘-AGGTGACCGT-3⬘); AP-4 (5⬘-GGTACTCCAC-3⬘); AP-5 (5⬘-GTTGCGATCC-3⬘). These reactions also contained 1 U AmpliTaq DNA polymerase (Perkin-Elmer Cetus, Norwalk, CT). The cycling parameters for PCR were: 94 C for 30 sec, 40 C for 2 min, 72 C for 30 sec (for 40 cycles). After PCR amplification, samples were analyzed on a 6% polyacrylamide sequencing gel, dried without fixation, and exposed to XAR-5 film (Eastman Kodak Co., Rochester, NY) for 72 h. Bands exhibiting differential expression were cut out from the gel, and DNA was eluted by boiling as described before (16 –18). Eluted DNA samples were then reamplified by PCR using the corresponding pair of primers under the same conditions as described above, except that neither 25 mmol/L deoxynucleotide triphosphate nor radioisotope was used. The PCR products were cut from 1% agarose gels, subcloned into Pinpoint Vector (Promega Corp.), and subjected to nucleotide sequence analysis.

Northern blot analysis For Northern analysis, 30 ␮g total RNA was separated by formaldehyde agarose gel electrophoresis and transferred to Duralon membrane (Stratagene, La Jolla, CA). After transfer, the membranes were baked at 80 C for 2 h. Blots were prehybridized in 50 mmol/L NaPO4, pH 6.5/5⫻ SSC/5⫻ Denhardt’s/50% formamide/0.1% SDS and 100 ␮g/mL salmon sperm DNA for 4 h at 42 C. Hybridization was carried out overnight in the same buffer containing 106 cpm/mL of a 32P-labeled GBP1 cDNA fragment. The filters were washed twice for 15 min in 1⫻ SSC/0.1% SDS at room temperature, then twice for 20 min in 0.2⫻ SSC/0.1% SDS at 55 C, and the filters were exposed to x-ray films for 24 –72 h. The intensities of signals on the autoradiogram were estimated by densitometric scanning. To correct for RNA loading, the obtained signals were normalized with respect to the glyceraldehyde 3-phosphate dehydrogenase (GAPDH) signal in the same blot. For this, the filters were stripped of the radioactive probe by washing for 10 min in 0.5% SDS at 95 C. The blots were then reprobed with a 32P-labeled GAPDH probe as described above.

2421

RT-PCR reaction Endometrial RNA (0.4 ␮g) was subjected to RT reaction using an RT-PCR kit (Stratagene). Briefly, the RNA samples were mixed with oligo (dT) primer, incubated at 65 C for 5 min, and annealed at room temperature. First-strand cDNA was synthesized using MMLV reverse transcriptase at 37 C, and the reaction was stopped by heating the tubes at 95 C for 5 min. PCR reaction was then performed in 50 ␮L total vol using 35 ng GBP1-specific primer set; 200 ␮mol/L each of deoxy-ATP, dGTP, deoxycycidine triphosphate, and thymidine 5⬘-triphosphate; 1.5 mmol/L Mg2⫹; and 0.5 ␮L of Taq DNA polymerase (Perkin-Elmer Cetus). The conditions for PCR were 94 C, 30 sec (1 cycle) followed by 94 C, 30 sec; 65 C, 30 sec; and 68 C, 2 min (15– 40 cycles). PCR products were electrophoresed on agarose gels and processed for Southern blot analysis.

Southern blot analysis PCR products (2 ␮L each) were run on 1% agarose gel. After electrophoresis, the gel was transferred to a Duralon membrane (Stratagene). The membrane was prehybridized in 6⫻ SSC, 5⫻ Denhardt’s, 0.5% SDS, and 100 ␮g/mL salmon sperm DNA for 2 h at 68 C. Hybridization was performed in the same buffer containing 106cpm/mL of the 32P-labeled cDNA fragment of human GBP1 or GAPDH overnight at 68 C. The membrane was washed with 2⫻ SSC and 0.1% SDS for 15 min at room temperature, in 0.1⫻ SSC containing 0.5% SDS at 68 C for 45 min, and exposed to x-ray film for 12 h.

In situ hybridization Frozen endometrial tissues were cut at 8 ␮m and attached to 3-aminopropyl triethylsilane (Sigma, St. Louis, MO)-coated slides. In situ hybridization was then performed with digoxygenin (DIG)-labeled GBP1 antisense RNA probe. Prehybridization was carried out in a damp chamber at 37 C for 60 min in hybridization buffer (50% formamide, 5⫻ SSC, 2% blocking reagent, 0.02% SDS, 0.1% N-laurylsarcosine). Hybridization was carried out at 42 C overnight in a damp humidified chamber. To develop the substrate, sections were sequentially washed in 2⫻ SSC, 1⫻ SSC, and 0.1⫻ SSC for 15 min in each buffer at 37 C. Sections were then incubated with anti-DIG alkaline phosphatase-conjugated antibody. Excess antibody was washed away, and the color substrate (nitroblue tetrazolium salt and 5-bromo-4-chloro-3-indoylphosphate) was added. Slides were allowed to develop in the dark, and the color was visualized under light microscopy until maximum levels of staining were achieved. The reaction was stopped, and the slides were counterstained in Nuclear Fast Red for 5 min. The slides were washed in water, dehydrated, and coverslipped. Control incubations used a DIG-labeled RNA sense strand and were performed under identical conditions.

Ishikawa cell culture Ishikawa endometrial adenocarcinoma cells were maintained in DMEM (Life Technologies, Inc., Grand Island, NY) supplemented with 5% FBS (HyClone Laboratories, Inc., Logan, UT); 5 ⫻ 105 cells were plated on 10-cm tissue culture dishes in phenol red-free medium containing 5% charcoal-stripped serum. The cells were grown to 70% confluency, transferred to serum-free medium for 24 h before treatment with the IFNs. Cells were harvested after 24 h of IFN treatment, and RNA was isolated for RT-PCR analysis. The experiment was repeated at least three times.

Results Isolation of a cDNA encoding GBP1 by mRNA DD analysis

To isolate the genes that are differentially expressed during the menstrual cycle, we employed the mRNA DD method (16 –18). We compared RNA samples prepared from human endometrium at proliferative (P), midsecretory (S1), and late-secretory (S2) phases of the menstrual cycle. Our studies revealed several cDNA clones representing genes whose expression in the endometrium is up- or downregulated specifically at the midsecretory phase overlapping

2422

KUMAR ET AL.

the putative window of implantation (S. Kumar and I. Bagchi, unpublished observation). One of the differentially displayed bands, which was absent in the proliferative or latesecretory endometrium but present at the midsecretory phase (marked by an arrowhead in Fig. 1A, lane 2), was selected for further characterization. The band containing the cDNA of interest was recovered from the gel, and the cDNA was amplified by PCR (40-cycles). Nucleotide sequence analysis of the isolated cDNA and comparison with the GenBank

JCE & M • 2001 Vol. 86 • No. 6

database revealed a 100% identity with nucleotides 127– 427 of the gene encoding the guanylate-binding protein 1 (GBP1) (19). GBP1 contains GTPase activity (20). We next confirmed the differential expression of GBP1 mRNA in human endometrium during the menstrual cycle, by Northern blotting. A GBP1 cDNA fragment was labeled with 32P and used to probe blots of total RNA isolated from endometrial biopsies obtained at the proliferative, midsecretory, and late-secretory stages. As shown in Fig. 1B, an intense signal corresponding to the predicted size of the GBP1 transcript was observed in the midsecretory stage endometrium (lanes 2 and 3). No GBP1 mRNA signal was detectable in proliferative and late-secretory endometrium (lanes 1 and 4). Hybridization of the blot with a control probe (GAPDH) indicated that the difference in intensities of the signals did not arise because of unequal loading of mRNAs in these lanes. These results demonstrated that the GBP1 mRNA is indeed up-regulated in the midsecretory stage endometrium. GBP1 mRNA is expressed in human endometrium specifically within the window of implantation

To examine the profile of expression of GBP1 mRNA in the human endometrium during the menstrual cycle, we analyzed RNA isolated from human endometrial biopsies for the presence of GBP1, by the RT coupled PCR (RT-PCR). The RNA samples obtained from the endometrium of 57 patients, at different days of the menstrual cycle, were reverse-transcribed and amplified by PCR using GBP1-specific primers. The PCR-amplified products were then subjected to Southern blot analysis, employing a radiolabeled GBP1 cDNA fragment as a probe. The results depicted in Fig. 2 (upper) show that no GBP1 transcript was detected in the early (P1, days 3– 8) or late proliferative phase (P2, days 10 –14). The level of GBP1 mRNA increased dramatically during the early (S1E, days 16 –20) or late (S1L, days 21–24) midsecretory phase. Its level then declined by the late-secretory phase of the menstrual cycle (S2, days 25–28). The relative level of expression of GBP1 mRNA in the endometrium at different days of the cycle was estimated by densitometric scanning, followed by normalization with respect to the control GAPDH mRNA signal (Fig. 2, lower). A significant level of GBP1 mRNA was observed during the midsecretory phase, compared with other days of the cycle (Fig. 2). These results showed that GBP1 is expressed in human endometrium within a narrow window during the midsecretory phase of the menstrual cycle. This time frame overlaps the putative window of implantation. GBP1 mRNA is localized in the endometrial glands and stroma FIG. 1. Profiles of differentially expressed mRNAs in endometrial tissues at different stages of the menstrual cycle. A, Total RNA (2 ␮g) isolated from the endometrium at proliferative (lane 1), midsecretory (lane 2), and late-secretory stages (lane 3) of the cycle, was subjected to DD. A band representing mRNA that is up-regulated during the midsecretory phase is marked by an arrowhead (lane 2). B, RNA was analyzed by Northern blotting, followed by hybridization with a 32Plabeled GBP1 or GAPDH cDNA probe. Lane 1, RNA from proliferative phase; lanes 2 and 3, RNA from midsecretory phase; lane 4, RNA from late-secretory phase of the menstrual cycle.

To identify the site(s) of GBP1 mRNA expression in human endometrium, we performed in situ hybridization analysis with sections of endometrial specimens in the proliferative (day 7), midsecretory (day 20), and latesecretory (day 28) phases of the menstrual cycle. We used a 300-bp (nucleotides 127– 427) DIG-labeled antisense RNA probe containing sequences from the GBP1 cDNA. As shown in Fig. 3, a strong hybridization of the probe to the glandular epithelial cells was observed in the sections

GENE REGULATION BY IFNs

2423

FIG. 2. Profile of expression of GBP1 mRNA in human endometrium at different days of the menstrual cycle. A, Total RNA (1 ␮g) was isolated from 57 endometrial biopsies at different stages of the menstrual cycle and was subjected to RT-PCR using GBP1 (upper panel) or GAPDH (lower panel) specific primers, as described in Materials and Methods section. n, Number of independently isolated endometrial samples analyzed; P1, endometrial samples at the early proliferative stage (cycle days 3– 8, n ⫽ 12). Twelve samples were examined at this stage. Shown are results from 4 of these 12 samples. Similar results were obtained from the other 8 samples (data not shown). Similarly, P2 represents the late proliferative stage (days 10 –14, n ⫽ 9). Nine samples at P2 were analyzed, and the results from 6 of these samples are shown. S1E (days 16 –20, n ⫽ 11), and S1L (days 21–24, n ⫽ 10) represent samples from early and late midsecretory phases, respectively. The results from 3 samples at each of these stages are shown. S2 represents the late-secretory phase (days 25–28, n ⫽ 15). The results from 3 samples at this stage are shown. Similar results were obtained with the other 12 samples at S2 (data not shown). PCR reactions using both GBP1 and GAPDH primers were performed through multiple rounds, ranging from 15– 40 cycles, to ensure that the amplifications were in the linear range. The data shown represent 25 cycles of PCR amplification. The authenticity of the PCR products was confirmed by Southern blot analysis using GBP1 or GAPDH cDNA probes. B, Quantitation of GBP1 mRNA signals in RT-PCR reactions was performed by densitometry, followed by normalization with respect to corresponding GAPDH mRNA signals. For accurate densitometric measurements of the mRNA signals in the RT-PCR reactions, we used shorter exposures of the autoradiograms in which neither GBP1 nor GAPDH signal was saturating.

of the midsecretory phase endometrium (Fig. 3C). Specific hybridization signal was also observed in the stroma in the immediate vicinity of the glands. In contrast, only a low hybridization signal was present in the glandular epithelial or stromal cells of the proliferative or late-secretory phase endometrium (Fig. 3, A and B). Control uterine sections (day 20), hybridized with the corresponding sense RNA probe of equal length, did not exhibit any signal, demonstrating the specificity of the hybridization reaction (Fig. 3D). These results indicated that GBP1 mRNA is induced in the human endometrium around the midsecretory phase of the menstrual cycle, and it is predominantly localized in the glandular epithelial and stromal cells. Induction of GBP1 mRNA by IFN-␣ and -␥ in Ishikawa cells

Previous reports described the induction of GBP1 mRNA by IFN-␣ or IFN-␥ in cultured fibroblasts (19). We

therefore examined whether the expression of GBP1 mRNA is indeed induced by IFN-␣ or -␥ in cultured human cells of endometrial origin. For this purpose, we treated Ishikawa cells, which are transformed human endometrial adenocarcinoma cells, with these cytokines (22). These cells were grown in serum-free media and treated with either vehicle or increasing concentrations of IFN-␣ or IFN-␥. The GBP1 mRNA expression was monitored by RT-PCR analysis using total RNA isolated from these cells. As shown in Fig. 4, addition of increasing concentrations of either IFN-␣ (A) or IFN-␥ (B) to Ishikawa cells markedly enhanced GBP1 mRNA expression, suggesting that GBP1 expression in human endometrial cells is regulated by these IFNs. By our estimation, the level of GBP1 mRNA induced by 50 –100 U of IFN-␥ was 4 –5 times higher, compared with that induced by an equivalent amount of IFN-␣. IFN-␥ is therefore a more potent inducer of GBP1 mRNA than is IFN-␣.

2424

JCE & M • 2001 Vol. 86 • No. 6

KUMAR ET AL.

IFN-␥ were undetectable in the early (P1, days 5–9) or late (P2, days 10 –14) proliferative phase. However, a dramatic increase in the levels of both IFN-␣ and -␥ mRNAs was observed in the early half of midsecretory (S1E) phase between cycle days 16 –20. We estimated that, during this phase, the level of GBP1 mRNA was enhanced 20- to 25-fold, compared with the proliferative phase. The expression of IFN-␣ mRNA then declined only slightly during the later half (days 21–24) of the midsecretory phase, before dropping further in the late-secretory (S2) phase. In contrast, the level of IFN-␥ mRNA, which underwent a relatively steeper decline during the later half of the midsecretory phase, fell below detection limits during the late-secretory (S2, days 25–28) phase. However, the expression of type I IFN receptor (IFNR) mRNA remained unaltered throughout the cycle (Fig. 5A). Similarly, the expression of type II IFN receptor mRNA did not change during the menstrual cycle (data not shown). It is of interest to compare the profile of GBP1 mRNA with that of each IFN mRNA (Fig. 2 vs. Fig. 5) during the menstrual cycle. A rise in the level of either IFN-␥ or IFN-␣ mRNA during the midsecretory phase (Fig. 5, B and C, lanes 11–16) was accompanied by the marked induction of GBP1 mRNA (Fig. 2, lanes 11–16). Concomitant with a steep drop in IFN-␥ mRNA level in the late-secretory phase (Fig. 5B, lanes 17–19) we observed a sharp diminution in GBP1 expression (Fig. 2, lanes 17–19). In contrast, a substantial amount of IFN-␣ mRNA was found to be present in the late-secretory endometrium (Fig. 5C, lanes 17–19). These results indicated that, although both IFN mRNAs are simultaneously induced in the human uterus in the midsecretory phase, the temporal profile of GBP1 mRNA expression matches more closely with that of IFN-␥. This finding supports the concept that GBP1 is an endogenous target gene of IFN-␥ in the receptive human endometrium. FIG. 3. Localization of GBP1 mRNA in human endometrium, by in situ hybridization. Human endometrium specimens in the proliferative (day 7), midsecretory (day 20), late-secretory (day 28) phases of the menstrual cycle were subjected to in situ hybridization. The hybridization was performed employing a 300-bp (corresponding nucleotides 127– 427) DIG-labeled cRNA probe specific for GBP1 gene. A and B, Hybridization of day-7 and day-28 endometrial section with antisense RNA probe; C and D, hybridization of a day-20 endometrial section with antisense and sense RNA probe.

Temporal profiles of IFN-␣ and -␥ mRNA expression in human endometrium resemble that of GBP1 mRNA during the menstrual cycle

To test the possibility that IFN-␣ or IFN-␥ is the endogenous inducer of GBP1 expression in human endometrium during the menstrual cycle, we investigated the profiles of expression of these cytokines and their receptors during the cycle. It is known that IFN-␣ and -␥ act through the type I and type II IFN receptors, respectively. RNA samples, isolated from the endometrial biopsies obtained at various stages of the menstrual cycle, were reverse-transcribed and amplified using primers specific for IFN-␣, IFN-␥, and types I and II IFN receptors. The PCR-amplified products were then subjected to Southern blot analysis using IFN-␣, IFN-␥, and IFN receptor-specific probes. As shown in Fig. 5, the signals corresponding to IFN-␣ and

Discussion

The IFNs constitute a family of secreted proteins that play a leading role in the host defense against viruses and parasites (23, 24). They function as cytokines, and they control diverse biological activities ranging from inhibition of cell proliferation and induction of differentiation to modulation of the immune system (23). Previous studies have shown that the human endometrium contains a full range of immune cells, including macrophages, leukocytes, and lymphocytes, such as T and NK cells, which synthesize IFNs (25–29). It is generally believed that binding of IFNs to specific cellsurface receptors of responsive cells triggers intracellular signaling pathways, which ultimately lead to alteration in expression of many genes to orchestrate a cellular response (23, 24). A first step toward understanding the role of the IFNs in endometrial physiology would require identification of downstream target genes that are regulated by these cytokines. In this study, we report the isolation of a cDNA encoding guanylate-binding protein 1 (GBP1), which is expressed specifically in the midsecretory endometrium during the menstrual cycle. GBP1 is a GTP-binding protein with GTPase activity (30). It is one of at least two forms of GBPs that are induced in response to IFN-␣ or IFN-␥ in cultured human fibroblasts (31).

GENE REGULATION BY IFNs

2425

FIG. 4. Induction of GBP1 mRNA in response to IFN-␣ or IFN-␥ in Ishikawa cells. Ishikawa cells were cultured and treated with various doses of IFN-␣ (10 U, 100 U, and 1000 U/mL) (A; lanes 2, 3, and 4, respectively) and IFN-␥ (50 U, 500 U, and 5000 U/mL) (B; lanes 2, 3, and 4, respectively) as described in Materials and Methods. Lane 1 in A and B represents Ishikawa cells treated with vehicle only. Twenty-four hours after treatment with vehicle or IFNs, cells were harvested, and mRNA was isolated and subjected to RT-PCR analysis employing GBP1 or GAPDH-specific primers. PCR reactions, using both GBP1 and GAPDH primers, were performed through multiple rounds, ranging from 15– 40 cycles, to ensure that the amplifications were in the linear range. The data shown represent 25 cycles of PCR amplification. The authenticity of the PCR-products was confirmed by Southern blotting employing 32 P-labeled GBP1 and GAPDH cDNA probes.

The spatio-temporal expression of GBP1 in human endometrium is consistent with its regulation by the IFNs. The GBP1 mRNA expression in the glandular epithelium and the surrounding stromal cells is maximal between days 16 –24 of the menstrual cycle, coincident with the maximal induction of both IFN-␣ and -␥ mRNAs. Interestingly, the temporal expression of GBP1 seems to resemble the profile of IFN-␥ mRNA more closely than that of IFN-␣ mRNA. We found that both GBP1 and IFN-␥ mRNAs rose to their highest levels between days 16 –20 of the midsecretory phase of the cycle. Their levels also dropped simultaneously during the latesecretory phase (days 25–28) (Figs. 2 and 5). In contrast, the profile of IFN-␣ mRNA was altered much less dramatically during the late-secretory phase. A significant level of IFN-␣ mRNA expression was maintained during this phase, although the GBP1 mRNA expression sharply declined. It is therefore likely that GBP1 expression in the endometrium during the menstrual cycle is under regulation by IFN-␥ rather than IFN-␣. This concept received further support from our in vitro experiments in Ishikawa cells, where IFN-␥ was clearly more potent than IFN-␣ in inducing GBP1 mRNA (Fig. 4). Although our results showed a strong temporal correlation between the profiles of IFN-␥ and GBP1 mRNAs during the menstrual cycle, we do not provide direct evidence that IFN-␥ regulates GBP1 expression in normal en-

dometrium. Furthermore, it should be stated that we have not yet analyzed the expression and secretion of the GBP1 protein in the endometrium. The functional role of IFNs and their target genes during the menstrual cycle or pregnancy remains unclear. Tabibzadeh and co-workers (32–34) have suggested that lymphocytes play a role in endometrial maturation by modulating glandular proliferation. The close association of lymphoid aggregates with basal epithelium is consistent with the release of lymphoid aggregate-derived cytokines, including IFN-␥, with a local inhibitory effect on glandular proliferation (32–35). This hypothesis was supported by the observation that, during the menstrual cycle, IFN-␥ inhibits the proliferation of human endometrial epithelial cells cultured in vitro (34). Such inhibition of epithelial cell proliferation may have important regulatory consequences on the cyclical endometrial physiology during the menstrual cycle. Recent gene knockout studies suggested that IFN-␥ might play a critical role during the decidualization process (36). Mice lacking IFN-␥ or its receptor failed to initiate normal pregnancy-induced remodeling of decidual arteries at the implantation sites and displayed hypocellularity or necrosis of decidua (36). Administration of recombinant IFN-␥ alleviated these impairments in IFN-␥ knockout mice. It is interesting to note that, in the human endometrium, decidual-

2426

JCE & M • 2001 Vol. 86 • No. 6

KUMAR ET AL.

FIG. 5. Profile of expression of IFN-␣, IFN-␥, and IFNR mRNA in human endometrium at different days of the menstrual cycle. A, Total RNA (1 ␮g) was isolated from human endometrial biopsies at different stages of the menstrual cycle. As in Fig. 2: P1, early proliferative (days of cycle 3– 8, n ⫽ 12); P2, mid-late proliferative (days of cycle 10 –14, n ⫽ 9), S1E, early-midsecretory (days of cycle 16 –20, n ⫽ 11); S1L, late-midsecretory (days of cycle 21–24, n ⫽ 10); and S2, late-secretory (days of cycle 25–28, n ⫽ 15) phases of the menstrual cycle and subjected to RT-PCR using IFN␣⫺, IFN-␥⫺, IFNR-, and GAPDH-specific primers. n, Number of independently isolated endometrial samples analyzed. The data shown represent 30 cycles of PCR amplification. The authenticity of the PCR products was confirmed by Southern blot analysis using specific cDNA probes. B, Quantitation of IFN-␥ mRNA signals in RT-PCR reactions was performed by densitometry, followed by normalization with respect to corresponding GAPDH mRNA signals. C, Quantitation of IFN-␣ mRNA signals in RT-PCR reactions was performed by densitometry, followed by normalization with respect to corresponding GAPDH mRNA signals.

ization is triggered during the secretory phase in each menstrual cycle and is associated with proliferation of IFN␥-producing lymphocytes (29, 37). One can therefore speculate that IFN-␥ may also have a regulatory role during decidualization of human endometrium. Because GTP hydrolysis is a key process in intracellular signal transduction, it is conceivable that GBP1 is potentially involved in a variety of IFN-regulated physiological processes, including growth control and differentiation. Future studies will reveal whether the IFN-induced GBP1 expression has any role in human endometrium during implantation. A significant aspect of the present study is the emergence of GBP1 as a candidate marker to assess uterine receptivity in the context of the natural menstrual cycle. Although the putative window of implantation in the human is believed to open between days 18 –24, there is a serious dearth of a molecular marker(s) specific for the receptive endometrium. We detected very little GBP1 mRNA (or none) before day 16 or after day 24 of the menstrual cycle. The induction of GBP1 mRNA precisely at the midsecretory phase overlapping the

window of implantation qualifies it as an excellent marker of uterine receptivity. Furthermore, the abundant expression of GBP1 mRNA in the receptive endometrium may permit its detection by in situ hybridization or immunocytochemistry in sections of uterine biopsies. The development of such measurable markers may assist in the diagnosis of female infertility arising from a failure to acquire uterine receptivity in a timely manner, and this may facilitate management of clinical therapy for affected women. Acknowledgments We thank Bruce Lessey for Ishikawa cells. We also thank Evan Read for the artwork and Jean Schweis for carefully reading the manuscript.

References 1. Psychoyos A. 1973 Endocrine control of egg implantation. In: Greep RO, Astwood EG, eds. Handbook of physiology. Washington, DC: American Physiological Society; 187–215. 2. Carson DD, Bagchi I, Dey SK, et al. 2000 Embryo implantation. Dev Biol 223:217–237.

GENE REGULATION BY IFNs 3. Schlafke S, Enders AC. 1975 Cellular basis of interaction between trophoblast and uterus at implantation. Biol Reprod. 12:41– 65. 4. Yoshinaga K. 1988 Uterine receptivity for blastocyst implantation. Ann NY Acad Sci. 541:424 – 431. 5. Parr MB, Parr EL. 1989 The implantation reaction. In: Wynn RM, Jollie WP, eds. Biology of the uterus. New York: Plenum Press; 233–277. 6. Glasser SR. 1990 Biochemical and structural changes in uterine endometrial cell types following natural or artificial deciduogenic stimuli. In: Denker HW, Aplin JD, eds. Trophoblast research (trophoblast invasion and endometrial receptivity: novel aspects of the cell biology of embryo implantation). New York: Plenum Press; 377– 416. 7. Psychoyos A. 1973 Hormonal control of ovoimplantation. Vitam Horm. 31:205–255. 8. Psychoyos A. 1976 Hormonal control of uterine receptivity for nidation. J Reprod Fertil Suppl. 25:17–28. 9. Psychoyos A. 1986 Uterine receptivity for nidation. Ann NY Acad Sci. 476:36 – 42. 10. Sharkey A. 1998 Cytokines and implantation: reviews of reproduction. J Reprod Fertil. 3:52– 61. 11. Hertig AT, Rock J, Adams EC. 1956 A description of 34 human ova within the first 17 days of development. Am J Anat. 98:435– 493. 12. Formigli L, Formigli G, Roccio C. 1987 Donation of fertilized uterine ova to infertile women. Fertil Steril. 47:62– 65. 13. Rogers PAW, Murphy CR. 1989 Uterine receptivity for implantation: human studies. In: Yoshinaga K, ed. Blastocyst implantation. Norwell, MA: Serono Symposia; 231–238. 14. Navot DM, Anderson TL, Droesch K, et al. 1989 Hormonal manipulation of endometrial maturation. J Clin Endocrinol Metab. 68:801– 807. 15. Navot DM, Scott RT, Droesch K, Veeck LL, Liu HC, Rosenwaks Z. 1991 The window of embryo transfer and the efficiency of human conception in vitro. Fertil Steril. 55:114 –117. 16. Liang P, Pardee AB. 1992 Differential display of eukaryotic messenger RNA by means of polymerase chain reaction. Science. 257:967–971. 17. Utans U, Liang P, Wyner LR, Karnovsky MJ, Russell ME. 1994 Chronic cardiac rejection: identification of five up-regulated genes in transplanted hearts by differential mRNA display. Proc Natl Acad Sci USA. 91:6463– 6467. 18. Aiello LP, Robinson GS, Lin YW, Nishio Y, King GL. 1994 Identification of multiple genes in bovine retinal pericytes altered by exposure to elevated levels of glucose by using mRNA differential display. Proc Natl Acad Sci USA. 91:6231– 6235. 19. Cheng YS, Patterson CE, Staeheli P. 1991 Interferon-induced guanylate-binding proteins lack an N(T)KXD consensus motif and bind GMP in addition to GDP and GTP. Mol Cell Biol. 11:4717– 4725.

2427

20. Cheng YS, Colonno RJ, Yin FH. 1983 Interferon induction of fibroblast proteins with guanylate binding activity. J Biol Chem. 258:7746 –7750. 21. Noyes RW, Hertig AI, Rock J. 1950 Dating the endometrial biopsy. Fertil Steril. 1:3–25. 22. Nishida M, Kasahara K, Kaneko M, Iwasaki H. 1985 Establishment of a new human endometrial adenocarcinoma cell line, Ishikawa cells, containing estrogen and progesterone receptors. Acta Obstet Gynecol Jpn. 37:1103–1111. 23. Stark GR, Kerr IM, Williams BRG, Silverman RH, Schreiber RD. 1998 How cells respond to interferons. Annu Rev Biochem. 67:227–264. 24. Darnell Jr JE. 1997 STATs and gene regulation. Science. 277:1630 –1635. 25. Tabibzadeh S. 1991 Human endometrium: an active site of cytokine production and action. Endocr Rev. 12:272–290. 26. Bonatz G, Hansmann M, Buchholz L, Mettler L, Radzun H, Semm K. 1992 Macrophage- and lymphocyte-subtypes in the endometrium during different phases of the ovarian cycle. Int J Gynaecol Obstet. 37:29 –36. 27. Morris H, Edwards J, Tiltman A, Emms M. 1985 Endometrial lymphoid tissue: an immunohistological study. J Clin Pathol. 38:644 – 652. 28. Laguens G, Goni JJ, Laguens M, Goni JM, Laguens R. 1990 Demonstration and characterization of HLA-DR positive cells in the stroma of human endometrium. J Reprod Immunol. 18:179 –186. 29. Loke YW, King A. 2000 Decidual natural-killer-cell interaction with trophoblast: cytolysis or cytokine production? Biochem Soc Trans. 28:196 –198. 30. Schwemmle M, Staeheli P. 1994 The interferon-induced 67-kDa guanylatebinding protein (hGBP1) is a GTPase that converts GTP to GMP. J Biol Chem. 269:11299 –11305. 31. Neun R, Richter MF, Staeheli P, Schwemmle M. 1996 GTPase properties of the interferon-induced human guanylate-binding protein 2. FEBS Lett. 390:69 –72. 32. Tabibzadeh SS, Gerber MA, Satyaswaroop PG. 1986 Induction of HLA-DR antigen expression in human endometrial epithelial cells in vitro by recombinant gamma-interferon. Am J Pathol. 125:90 –96. 33. Tabibzadeh SS, Satyaswaroop PG. 1989 Sex steroid receptors in lymphoid cells of human endometrium. Am J Clin Pathol. 91:656 – 663. 34. Tabibzadeh SS, Satyaswaroop PG, Rao PN. 1986 Anti-proliferative effect of interferon-gamma in human endometrial epithelial cells in vitro: potential local growth modulatory role in endometrium. J Clin Endocrinol Metab. 67:131–138. 35. Stewart CJR, Farquaiarson A, Foulis AK. 1992 The distribution and possible function of gamma interferon-immunoreactive cells in normal endometrium and myometrium. Virchows Arch A Pathol Anat Histopathol. 420:419 – 424. 36. Ashkar AA, Di Santo JP, Croy BA. 2000 Interferon g contributes to initiation of uterine vascular modification, decidual integrity, and uterine natural killer cell maturation during normal murine pregnancy. J Exp Med. 192:259 –270. 37. King A. 2000 Uterine leukocytes and decidualization. Hum Reprod Update. 6:28 –36.