Molecular Cloning and Immunologic Characterization of a Novel cDNA Coding for Progesterone-Induced Blocking Factor

Beata Polgar, Gyula Kispal, Margit Lachmann, Gabriella Paar, Eszter Nagy, Peter Csere, Eva Miko, Laszlo Szereday, Peter Varga and Julia Szekeres-Bartho
J Immunol December 1, 2003, 171 (11) 5956-5963; DOI: https://doi.org/10.4049/jimmunol.171.11.5956

This article has a correction. Please see:

Abstract

Previous studies from our laboratory showed that the immunomodulatory effects of progesterone are mediated by a 34-kDa protein, named the progesterone-induced blocking factor (PIBF). Lymphocytes of women with threatened abortion fail to produce this factor. Via inducing a Th2 biased cytokine production and blocking of NK activity, PIBF prevents induced pregnancy loss in mice, suggesting that substitution therapy with PIBF could be useful as an alternative treatment of certain forms of recurrent spontaneous abortions. Our study was aimed at mapping the sequence and structure of PIBF coding cDNA and characterizing the encoded protein product. Screening of a human liver cDNA library revealed a 2765-bp clone with a 2271-bp open reading frame. The PIBF1 cDNA encodes a protein of 757 amino acid residues with an 89-kDa predicted molecular mass, which shows no significant amino acid sequence homology with any known protein. PIBF produced via recombinant technique is recognized by the Ab specific for the secreted lymphocyte PIBF Ab, and possesses the biological activities of the secreted lymphocyte PIBF. The full-length PIBF is associated with the nucleus, whereas secretion of shorter forms, such a 34-kDa protein is induced by activation of the cell. The 48-kDa N-terminal part of PIBF is biologically active, and the part of the molecule, responsible for modulating NK activity is encoded by exons 2–4. These data provide an initial step for exploiting the possible diagnostic and therapeutic potential of this immunomodulatory molecule.

Peripheral lymphocytes from healthy pregnant women express progesterone receptors (PR) 3 (1, 2, 3, 4, 5). In the presence of this hormone, PR-positive lymphocytes mainly γδ TCR-positive cells (6) produce a mediator protein named the progesterone-induced blocking factor (PIBF) (7, 8). PIBF was first isolated from the supernatants of progesterone-treated pregnancy lymphocytes by biochemical methods, and polyclonal Abs recognizing this molecule were raised in rabbits.

Lymphocytes from women with threatened abortion fail to produce this factor (9). One of the main mechanisms by which NK cells kill their target cells is the exocytosis of perforin and serine esterase-containing granules. In the presence of PIBF, activated NK cells fail to release perforin from the storage granules, and as a result, they do not lyse target cells (10). Another mechanism, by which PIBF exerts its anti-NK effect, is through altered cytokine production (11).

Earlier we showed that activation of γδ T cells via the Vγ1.4Vδ1 TCR results in a significantly increased IL-10 expression, whereas activation of human γδ T cells via the Vγ9Vδ2 receptor leads to a significantly reduced IL-10 expression and a slightly increased IL-12 expression (12), suggesting the presence of two functionally distinct γδ subpopulations in peripheral blood of pregnant women.

Normal human pregnancy is associated with a Th2 biased peripheral cytokine profile. Makhseed et al. (13) found significantly higher levels of the type 1 and lower levels of the type 2 cytokines in supernatants of activated peripheral lymphocytes from women undergoing preterm delivery, than in those from healthy pregnant women. Furthermore, abortion-prone women who proceeded to have successful pregnancy were more Th2 biased than abortion-prone women who aborted (14).

Neutralization of the biological effects of PIBF in vivo results in a Th1 shift (15). Recent data from our laboratory show, that PIBF stabilizes the mRNA of IL-10, without any effect on IL-2 mRNA (T. Henics and C. Zimmer, unpublished observation).

In mice PIBF is anti-abortive, which is partly due to a direct anti-NK effect. Neutralization of endogenous PIBF activity in pregnant mice results in a 70% reduction in the number of viable fetuses per mother, and this is associated with an increased splenic NK activity (16). Some 90% of pregnancy loss is corrected by treatment of the pregnant animals with anti-NK Abs (16). These data suggest that at least in mice PIBF contributes to the success of pregnancy and that the major part of its pregnancy-protective effect is due to its NK inhibitory activity.

In humans although there is no direct proof for the need for PIBF in the maintenance of pregnancy, the percentage of PIBF-positive lymphocytes in peripheral blood of healthy pregnant women is significantly higher than in that of women at risk for premature pregnancy termination. In peripheral blood of patients undergoing spontaneous pregnancy termination at the time of sampling and in those of women showing symptoms of premature pregnancy termination, we found a lower than normal percentage of PIBF-positive cells (9). PIBF expression of the lymphocytes shows an inverse correlation with NK activity and a positive relationship with the outcome of pregnancy (9, 17). This experimental evidence implies that PIBF is a player in immunomodulation during normal human pregnancy.

For human application a molecule with antiabortive potential and the least possible side effects needs to be designed. Based on the nucleotide and amino acid sequence, the secondary and tertiary structure of the protein is predictable, and analysis of a structure function relationship might enable us to design a smaller, biologically active molecule. Because in mice, exogenous PIBF prevented abortions due to PR block, or those induced by elevated NK activity, it is possible that substitution therapy with PIBF could be useful as an alternative treatment in certain forms of recurrent spontaneous abortions. We report on the molecular and structural characteristics of the cDNA coding for PIBF and on immunobiological effects of PIBF produced by recombinant techniques.

Materials and Methods

Cloning, restriction mapping, and sequence analysis of PIBF cDNA

Strains and vectors.

Escherichia coli DH-5α, XL1-Blue MRF, BL21, and BL21 pLys were used as host cells for cloning, sequencing end expression studies. pBluescript SK− phagemid was used as a plasmid for ligation during restriction mapping and sequencing, pCR-Script Amp SK+ vector for the cloning of and pGex-4T-1 and pGex-4T-3 for the expression of GST-PIBF proteins.

Enzymes kits and nucleotides.

Human liver Uni-λ ZAP XR expression cDNA library and the in vitro packaging system were purchased from Stratagene (La Jolla, CA), restriction endonucleases and T4 DNA ligase from MBI-Fermentas (Biocenter, Temesvari, Hungary). We used T7 DNA Sequenase kit version 2.0 (Amersham Life Science, Budapest, Hungary), TaqDNA Polymerase and Ultrapure dNTP-Set (Pharmacia Biotech, Budapest, Hungary), [α-35S]dATP and [α-35P]dCTP (Izotóp Intézet, Budapest, Hungary), and Random Primers DNA Labeling System (Sigma-Aldrich, Budapest, Hungary). We applied universal T7 and T3 primers and synthetic oligonucleotides synthesized by The Great American Gene Company (Newington, NH) for sequencing and Takara reverse transcriptase for the PCR.

For identification of cDNAs encoding PIBF, a normal human liver cDNA library (in Uni-ZAP XR 937241 vector) was plated on 150-mm petri dishes at 100,000 plaques/plate and replicas were transferred onto nitrocellulose membrane (Sartorius-Membran, Budakeszi, Hungary) from all 20 plates. Filters were probed with a polyclonal rabbit anti-PIBF IgG, which was raised by immunizing rabbits with the 34-kDa band obtained from nitrocellulose blots of SDS-PAGE separated supernatants of progesterone-treated human pregnancy lymphocytes. The primary immunoreaction was detected by alkaline-phosphatase-conjugated goat anti-rabbit IgG (Sigma-Aldrich) and developed by 5-bromo-4-chloro-3-indolyl phosphate-nitroblue tetrazolium phosphatase reaction system. The positive lambda phage cDNA inserts were converted into pBluescript SK− phagemids with R408 helper phage according to the method of Short and Sorge (18). Cultivation of bacteria, preparation of plasmid DNA, characterization of cDNA inserts by restriction enzyme analysis, and subcloning of insert fragments were as described in Sambrook et al. (19). The nucleotide sequences of the inserts were determined from both strands in both directions by the dideoxy-chain termination method (20).

Expression of the human recombinant PIBF protein in pGex expression vector.

Several GST fusion proteins containing PIBF sequences were generated. First the full-length PIBF1 cDNA and a 1255 bp (N-terminal 48-kDa protein coding) segment were ligated into a pGex-4T-1 expression vector.

Based on the previously described exon-intron structure of PIBF (21), sequences containing exons 2–4 (PN1), 5–7 (PN2), and 8 and 9 (PN3) were amplified by PCR. The cycling parameters were as follows: 94°C for 2 min, then 94°C for 30 s, 50°C for 30 s, and 68°C for 1 min for 25 cycles followed by a final extension at 68°C for 10 min. Platinum Pfx polymerase was purchased from Invitrogen (Csertex, Budapest, Hungary). For generation of the construct the expression vector pGex-4T3 (Promega, Bio-Science, Budapest, Hungary) that already contained the GST was used. The PCR product as well as the vector was digested with BamH1 and XhoI (MBI-Fermentas) to insert the PIBF cDNA downstream from the GST DNA.

The primers used for amplification were the following: PN111 pGex-4T-3 forward BamHI exons 2, 3, and 4, AAA GGA TCC ATG TCT CGA AAA ATT TCA AAG; PN121 Reverse XhoI stop, AAA CTC GAG TTA AGA TAC ATA TTC AGG AAT AG; PN211 pGex-4T-3 forward BamHI exons 5, 6, and 7, AAA GGA TCC GTT CGC TTC TAT GAG CTA G; PN221 Reverse XhoI stop, AAA CTC GAG TTA CTC TTT TGA TAA TTC ACT TC; PN311 pGex-4T-3 forward BamHI exons 8 and 9, AAA GGA TCC GTA GTC ACC TTA GAG CAA AC; PN321 Reverse XhoI stop, AAA CTC GAG TTA GTT TTC TCG TTC ATA CAT TTC.

The amplified fragments were cloned into pGex-4T-3 expression vector and the selected clones were transformed into E. coli BL-21 and BL-21 pLys host strains. GST-PIBF fusion protein expression and affinity purification were performed as described by the manufacturer (Pharmacia Biotech). Recombinant proteins were analyzed by SDS-PAGE and Western blotting.

Epitope mapping was done by mAbs reacting with the N-terminal 48-kDa part of the recombinant PIBF. ELISA plates were coated with the recombinant proteins (48-kDa N-terminal part, as well as PN1, PN2, and PN3) at concentrations of 1 μg/ml overnight at 4°C and then reacted with 1/1000 dilution of mAbs raised by immunizing with the 48-kDa N-terminal fragment and finally, with anti-mouse-IgG-biotin (1:5000) anti-mouse-streptavidin-HRP 1:1000 (Amersham Biosciences, Budapest, Hungary).

For identifying conformational epitopes, the recombinant protein constructs were heated to 100°C for 20 min.

Subcellular localization of PIBF

Three approaches were used for investigating the subcellular localization of PIBF: 1) PIBF containing cells as well as Western blots from their supernatants were reacted with an anti-PIBF Ab; 2) MCF-7 cells were transfected with a green fluorescence protein (GFP) containing vector coding for the full-length recombinant PIBF, and GFP localization was followed with fluorescent microscopy; and 3) PIBF was detected in immunoblots from cytosolic and nuclear cell fractions of MCF-7 cells.

Construction of vector for eukaryotic expression of PIBF-eGFP fusion protein.

To generate the PIBF-eGFP fusion construct the full-length PIBF cDNA was amplified by PCR. The upstream primer 5′-CCAGAATTCATGTCTCGAAAAATTTCAAAGGAGTC introduced an EcoRI restriction site while the downstream primer 5′-AAGGTCGACAAGGTCTTCATCTTTTGTTTCTTAGACC contained a SalI site. Expanded High Fidelity PCR System (Roche, Basel, Switzerland) was used for amplification according to the manufacturers instructions and the cycling parameters were as follows: 94°C for 1 min, then 94°C for 20 s, 55°C for 30 s, and 72°C for 2 min for 30 cycles, followed by another extension at 72°C for 10 min. For generation of the construct, the mammalian expression vector pCIneo (Promega) that already contained the eGFP was used. The PCR product as well as the vector was digested with EcoRI and SalI (both Invitrogen) to insert the PIBF cDNA upstream of the eGFP cDNA.

Cell culture and transfection.

The human MCF-7 cell line was cultured in RPMI 1640 medium supplemented with 10% heat inactivated FCS (Sigma-Aldrich), 100 U/ml penicillin, 100 μg/ml streptomycin, and 1% l-glutamine at 37°C in a humidified atmosphere of 5% CO2.

Cells were plated at a concentration of 5 × 104 cells per well in tissue culture chamber slides (Nalge Nunc International, Rochester, IL) the day before transfection. Lipofectamine 2000 (Invitrogen) was used to transfect either the PIBF-eGFP or eGFP constructs according to the manufacturer’s instruction.

Detection of eGFP and eGFP-fusion proteins.

Transfected cells were washed with PBS and fixed with 3.7% formaldehyde for 20 min at room temperature. Nuclei were counterstained with 0.1 μg/ml 4′,6′-diamidino-2-phenylindole (DAPI) in PBS. Slides were examined with a fluorescence-equipped microscope (Zeiss, Oberkochen, Germany). Images for illustration purposes were obtained using a digital camera (SPOT; Diagnostic Instruments, Sterling Heights MI). The images of the eGFP and eGFP-fusion protein-expressing cells were superimposed on those stained with DAPI.

Cell fractionation.

MCF-7 cells (∼5 × 106 cells) were trypsinized and washed in PBS. All subsequent steps were performed at 4°C. The pellet was resuspended in 1 ml of hypotonic lysis buffer (10 mM HEPES/NaOH, pH 7.4, 10 mM NaCl, 3 mM MgCl2, 0.05% Nonidet P-40, 1 mM EGTA, 1 mM PMSF, and complete EDTA-free tablets) and incubated 10 min before addition of 125 μl of 2M sucrose, 3 μl of 1M MgCl2, and 17 μl of 20% Triton. After another incubation for 1 min the nuclei were separated from the cytosolic fraction by density gradient centrifugation at 400 × g for 5 min. The pellet containing nuclei and insoluble material is washed once with lysis buffer before it was resuspended in 100 μl of high-salt extraction buffer (50 mM HEPES/NaOH, pH 7.4, 500 mM NaCl, 3 mM MgCl2, 0.5% Nonidet P-40, 1 mM EGTA, 1 mM PMSF, and complete EDTA-free tablets), incubated for 5 min and centrifuged at 10,000 × g for 5 min. The cytosolic and the high-salt soluble nuclear fractions were further analyzed by Western blot.

Western blot analysis.

For the detection of PIBF in cytosolic and nuclear fractions of MCF-7 cells, 25 μg of total protein was separated by 10% SDS-PAGE, transferred to a Hybond ECL membrane (Amersham Biosciences), blocked with 5% nonfat dry milk in PBS for 1 h, and incubated with anti-recombinant PIBF 48-kDa Ab (1:500) in PBS containing 1% nonfat dry milk. After the membrane was washed three times with PBS, HRP-coupled anti-rabbit IgGs (1:5000; Amersham Biosciences) were added and the signal was detected with the ECL Plus Western blotting detection system (Amersham Biosciences) according to the manufacturer’s instructions.

Functional tests

Treatment of lymphocytes.

Ficoll-Paque (Pharmacia Biotech) gradient isolated peripheral lymphocytes from nonpregnant humans at a cell count of 106/ml were incubated at 37°C with different concentrations of recombinant human PIBF 48-kDa or PN1, PN2, and PN3 (in the range from 0 to 15 μg/ml) for 24 h in the presence or absence of anti-human recombinant PIBF IgG (5 μg/ml). Lymphocytes incubated with GST or progesterone under the same conditions served as controls. Lymphocytes from healthy pregnant women were incubated with different concentrations of anti-recombinant human PIBF 48-kDa IgG for 3 h.

NK cytotoxicity test.

K562 target cells were labeled with the green fluorescent dye PKH-67 (Sigma-Aldrich). One million lymphocytes were treated with 20 U of recombinant IL-2 (Sigma-Aldrich) for 1 h, together with recombinant PIBF or anti-PIBF Abs. Target cells and lymphocytes (at a ratio of 1 to 25) were centrifuged at 500 rpm for 5 min for allowing close cell contact and incubated at 37°C in CO2 for 3 h. Target cells were stained with propidium iodide (Sigma-Aldrich), and the percentage of dead target cells was determined by flow cytometry.

Cytokine and PIBF determination by immunocytochemistry.

Lymphocytes were cytocentrifuged on microscope slides, fixed with acetone, and checked for expression of IL-10 and IL-12. The same cells were tested for NK activity. Immunocytochemistry was performed as follows: Endogenous peroxidase activity of the cells was blocked with 1% H2O2. Nonspecific protein binding sites were blocked by 1% BSA (Sigma-Aldrich). All incubations were conducted at room temperature in a humid chamber. The primary Abs (monoclonal anti-human IL-10 and anti-human IL-12 both from R&D Systems (Biomedica, Budapest, Hungary), or monoclonal anti-PIBF Abs were diluted 1/50 in TBS containing 0.5% BSA. After 1 h of incubation with the primary Ab, 1/100 diluted HRP-labeled anti-mouse IgG (DAKO, Frank Diagnosztika, Budapest, Hungary) was added as a secondary Ab for 30 min. The slides were washed three times in TBS and the reaction was developed by diaminobenzidine and intensified with silver staining. Control slides without the primary Abs or with isotype Ig were included.

Determination of cytokine expression by flow cytometry

The following mAbs were used: FITC-conjugated mouse anti-human CD3 (clone HIT3a; BD PharMingen, Soft Flow Hungary, Budapest, Hungary), allophycocyanin-conjugated anti-human IL-10 (clone JES3-19F1; BD PharMingen), PE-conjugated anti-human IFN-γ (clone B27; BD PharMingen).

For activation, 500 μl of heparinized peripheral blood diluted with an equal volume of RPMI 1640 was activated with PMA (25 ng/ml final concentration) and ionomycin (1 μg/ml final concentration) in the presence of brefeldin A (10 μg/ml final concentration). The samples incubated for 4 h at 37°C.

The cells were labeled with anti-CD3-FITC for 30 min at room temperature in the dark. Then RBC were lysed by treatment with 2 ml of FACS lysing solution (BD Pharmingen) for 10 min in the dark. The surface-stained cells were permeabilized with 500 μl of FACS permeabilizing solution (BD Pharmingen) for 10 min and stained for intracellular cytokines by incubating with fluorochrome-conjugated anti-cytokine mAbs for 30 min at room temperature in the dark. Stained cells were washed and fixed with 1% paraformaldehyde solution. Cells were analyzed for fluorescence using a FACSCaliber flow cytometer (BD Immunocytometry Systems; BD Biosciences, San Jose, CA) equipped with the CellQuest software program (BD Biosciences) for data acquisition and analysis.

Abs recognizing different forms of the recombinant human PIBF

Polyclonal Abs were generated in our laboratory by immunizing rabbits with a secreted 34-kDa lymphocyte PIBF, 89-kDa recombinant human PIBF-GST, a 48-kDa N-terminal recombinant human PIBF-GST (with and without GST). The Ab titers were determined by ELISA using the recombinant human PIBF protein as the Ag, and the immune sera were affinity purified with protein A or protein G column to generate IgG. mAbs were generated by immunizing with the 48-kDa N-terminal part of the recombinant molecule.

Statistics

The two-tailed Student’s t test was used for statistical evaluation of the data. Differences were considered significant at p ≤ 0.05.

Results

Characterization of the cDNA encoding PIBF

Screening of a human liver cDNA expression library with polyclonal (rabbit) anti-human PIBF IgG revealed a positive clone of 2765 bp (GenBank accession number Y09631), displaying a strong immunoreactivity. DNA sequence analysis revealed an open reading frame of 2271 bp between a 347 bp of 5′-untranslated and a 146 bp of the 3′-untranslated region. At a position of 348 bp there is a translation initiation codon (ATG) indicating that this is not a partial cDNA clone.

Analysis of the PIBF1 cDNA encoded protein with a National Center for Biotechnology Information Nucleotide BLAST program did not show sequence homology with any known protein. The coded protein contains 757 amino acids with a predicted average molecular mass of 89,724.5 Da and with a theoretical isoelectric point 5.73. The amino acid composition showed a relatively higher percentage of Glu (14.4%), Leu (13.6%), and Lys (10.2%) and a lower percentage of Trp (0.1%) and Cys (0.8%). The hydrophobicity analysis according to Kyte and Doolittle (Expasy Molecular Biology Server Prot Scale) indicated that PIBF is a hydrophilic protein.

The secondary structure analysis revealed an “all α helical” general structure. The molecule contains leucine-zipper (Expasy Molecular Biology Server PSORT II), nuclear localization signal (NLS, Expasy Molecular Biology Server PSORT II) and basic zipper characteristics of DNA-binding proteins. Search for possible posttranslational modification sites revealed several potential phosphorylation sites for tyrosine kinase and glycosylation sites for N-glycosylation (Expasy Molecular Biology Server ScanProsite) (Fig. 1⇓). The molecule contains a putative PEST sequence (Expasy Molecular Biology Server EMBnet Austria) suggesting a short half-life. Furthermore, a putative short N-terminal signal sequence and endoplasmic reticule membrane retention signal were predicted (Expasy Molecular Biology Server PSORT II).

FIGURE 1.
FIGURE 1.

Amino acid sequence analysis of the protein coded by PIBF1 cDNA.

Reactivity of the recombinant PIBF with an Ab specific for the secreted lymphocyte PIBF

To verify that the product of the isolated clone was related to the previously described PIBF of lymphocyte origin (10), recombinant PIBF was subjected to immunoblot analysis. The blot was developed with a polyclonal Ab that had been generated by immunizing with the 34-kDa PIBF, 34-kDa band obtained from nitrocellulose blots of SDS PAGE separated supernatants of progesterone-treated human pregnancy lymphocytes. The Ab reacted with the recombinant PIBF, whereas no reactivity was seen with the bacterial lysate (Fig. 2⇓).

FIGURE 2.
FIGURE 2.

Reactivity of the recombinant PIBF with an Ab specific for the secreted PIBF of lymphocyte origin (A) and of secreted PIBF with anti-recombinant PIBF Ab (B). A, Recombinant PIBF was subjected to 10% SDS-PAGE. The immunoblot was developed by a polyclonal (rabbit) Ab, reacting with the secreted lymphocyte PIBF. Recombinant human PIBF (1) and lysate of E. coli (2) are shown. B, PIBF of lymphocyte origin subjected to 10% SDS-PAGE. The immunoblot was developed by a polyclonal (rabbit) Ab, to the 48-kDa recombinant PIBF.

Intracellular localization and secretion of PIBF

Polyclonal Ab specific for the full-length recombinant PIBF labels the nucleus of MCF-7 mammary carcinoma cells. Immunoblot analysis of the supernatant of these cells does not show reactivity with the anti-PIBF Ab (Fig. 3⇓A). However, upon insulin treatment of the cells, PIBF secretion is induced. PIBF containing secretorygranules are seen in the cytoplasm of the cells, and anti-PIBF reactive bands appear in the supernatant (Fig. 3⇓B).

FIGURE 3.
FIGURE 3.

Subcellular localization of PIBF by immunocytochemistry. MCF-7 mammary carcinoma cells cultured with 0.8 μg/ml insulin or without insulin were stained with a polyclonal Ab specific for the full-length recombinant PIBF. The supernatants of the cells were analyzed by Western blot for PIBF reactivity. A, MCF-7 cells cultured without insulin. B, MCF-7 cells cultured with insulin. Magnification was ×1000.

Transfection of MCF-7 cells with a vector coding for the full-length recombinant PIBF-eGFP resulted in a nucleus-associated PIBF reactivity (Fig. 4⇓) with a patchy appearance. Moreover, newly synthesized PIBF seemed to be associated with the nuclear membrane from the cytoplasmic surface. In addition, Western blot analysis of the nuclear extracts of untransfected MCF-7 cells revealed an ∼90-kDa protein band corresponding to the full-length PIBF (Fig. 5⇓), which was missing from the cytoplasmic fraction.

FIGURE 4.
FIGURE 4.

Localization of PIBF-GFP fusion protein. MCF-7 cells were transiently transfected with plasmid DNA encoding for PIBF-eGFP fusion protein (1) or eGFP fusion partner (2), and visualized by fluorescence microscopy 6 h posttransfection. Merged images of DAPI stained nuclei and GFP are shown.

FIGURE 5.
FIGURE 5.

Localization of endogenous PIBF in fractionated MCF-7 cells. MCF-7 cells were grown and fractionated as described in Materials and Methods. A total of 25 μg of total protein of cytosolic (1) and nuclear (2) fractions were separated by 10% SDS-PAGE, blotted, and probed with rabbit polyclonal anti-48-kDa recombinant PIBF antiserum.

These data suggest that the full-length PIBF is associated with the nucleus and upon activation of the cell the smaller m.w. forms might enter the secretory pathway.

Functional characterization of the recombinant human PIBF protein and the effect of the recombinant PIBF on NK activity and cytokine production

Our previous data revealed that PIBF prevents pregnancy termination in mice due to its NK inhibitory effect (16). Based on these results we examined the effect of the recombinant human PIBF on NK activity. Lymphocytes from five nonpregnant individuals were treated with the N-terminal 48-kDa part of the recombinant human PIBF for 24 h and the cells were used as effectors in an NK assay.

Treatment of lymphocytes with the N-terminal 48-kDa part of the recombinant human PIBF, or with secreted PIBF of lymphocyte origin, resulted in a significant (p < 0.01) inhibition of NK activity. Abs to the the 48-kDa N-terminal part of the recombinant PIBF neutralized the NK blocking activity of PIBF secreted by pregnancy lymphocytes (Fig. 6⇓).

FIGURE 6.
FIGURE 6.

The effect of human recombinant PIBF on NK activity. Lymphocytes from nonpregnant individuals treated with secreted PIBF of lymphocyte origin (10 μg/ml), or with human recombinant PIBF (10 μg/ml), acted as effectors in cytotoxicity assay. Ab specific for the N-terminal part of the human recombinant PIBF (5 μg/ml) was used for neutralizing the effect of “secreted” PIBF. Error bars represent the mean ± SEM of five individual tests, ∗, p < 0.01.

Treatment of nonpregnancy lymphocytes with recombinant human PIBF 48-kDa increased the percentage of IL-10-positive and decreased that of IL-12-positive lymphocytes in a concentration dependent fashion (Fig. 7⇓). Lymphocytes from healthy pregnant women were incubated with a polyclonal (rabbit) antirecombinant human PIBF 48-kDa IgG at 37°C for 3 h and cytokine production of these cells was determined by immunocytochemistry. Treatment of pregnancy lymphocytes with an anti-PIBF 48-kDa IgG caused a 50% elevation in the percentage of IL-12-positive cells and decreased the percentage of IL-10-positive cells by 49% compared with untreated control cells (Fig. 8⇓). These data suggest that the recombinant human PIBF is an immunologically active protein and exerts similar effects to that secreted by pregnancy lymphocytes.

FIGURE 7.
FIGURE 7.

The concentration-dependent effect of recombinant human PIBF on IL-10 and IL-12 expression of peripheral lymphocytes. Each point represents the mean ± SE of 26 determinations.

FIGURE 8.
FIGURE 8.

The effect of anti-human recombinant PIBF IgG on IL-10 and IL-12 production. Lymphocytes from eight pregnant women were treated with anti-48-kDa human recombinant PIBF IgG. The percentage of IL-10- and IL-12-positive cells was determined by immunocytochemistry. Error bars represent the mean ± SEM of eight determinations; ∗, p < 0.05, ∗∗, p < 0.01.

Localization of the functional sites

The 48-kDa part of the recombinant molecule (exons 2–9) affected the cytokine secretion pattern and NK activity. For further localization of the region responsible for the biological activity, we amplified sequences containing exons 2–4 (PN1), 5–7 (PN2), and 8 and 9 (PN3) by PCR (Fig. 9⇓) and expressed them in E. coli. The mAbs that had been raised by immunizing mice with the N-terminal 48-kDa part of the recombinant protein showed three types of reactivity with the constructs.

FIGURE 9.
FIGURE 9.

Recombinant human PIBF constructs. GST fusion proteins were generated by cloning PCR-amplified PIBF sequences into pGex-4T-1 and pGex 4T-3 expression vector.

M2, M3, M4, M6, M8, and M9 reacted with the polypeptide encoded by exons 2–4, M7 reacted with the polypeptide coded for by exons 5–7, whereas M1 and M5 were specific for the polypeptide encoded by exons 8 and 9 (Fig. 10⇓). M4, M7, and M8, recognize conformational epitopes, based on analysis of heat inactivation of proteins (Fig. 11⇓).

FIGURE 10.
FIGURE 10.

Reactivity of anti-48-kDa recombinant PIBF mAbs with the PN1, PN2, and PN3. ELISA plates were coated with the constructs and reacted with 1/1000 dilution of nine mAbs. The reaction was developed with anti-mouse IgG-biotin-streptavidin-HRP and o-phenylenediamine.

FIGURE 11.
FIGURE 11.

Identification of sequential and conformational epitopes. ELISA plates were coated with the 48-kDa N-terminal part of the recombinant PIBF that had been heated to 100°C for 20 min. The plates were then reacted with the mAbs and developed with anti-mouse IgG-HRP and o-phenylenediamine. The reduction of ODs compared with those obtained with the native protein is indicated.

Lymphocytes from 14 nonpregnant individuals were pretreated for 60 min with the PIBF constructs and tested for NK activity (Fig. 12⇓). PN1 and the 48-kDa N-terminal part of the recombinant PIBF significantly inhibited NK activity, whereas PN2 and PN3 had no effect (Fig. 12⇓).

FIGURE 12.
FIGURE 12.

The effect of PN1, PN2, and PN3 on NK activity. Lymphocytes from 14 nonpregnant individuals were treated with the constructs and tested for NK activity. Error bars indicate the mean ± SEM of 14 tests, ∗, p < 0.05.

Cytokine expression in the same cells was determined by flow cytometry. Neither of the constructs induced significant alterations in expression of IL-10 or IFN-γ (data not shown).

The Abs gave different staining patterns on MCF-7 cells by immunocytochemistry. All Abs showed some nucleus-associated reactivity, together with a granular staining of the cytoplasm. Staining with an Ab specific for the secreted “lymphocyte” PIBF as well as with the Ab specific for the 48-kDa N-terminal peptide (exons 2–9) reacts with the nuclear membranes as well as with the cytoplasmic granules. All mAbs, except for M7, gave a similar pattern (Fig. 13⇓A). The mAb M7, which recognizes a conformational epitope in the region encoded by exons 5–7, showed a nuclear staining pattern (Fig. 13⇓B), suggesting that the latter Ab does not recognize the smaller secreted forms of the molecule.

FIGURE 13.
FIGURE 13.

Immunocytochemical reactivity of anti-PIBF Abs. Magnification was ×1000. C, Isotype control is shown.

Discussion

In this study, we have shown that the PIBF cDNA is 2765-bp long, with an open reading frame of 2271 bp. The nucleotide and the coded amino acid sequences show no significant homology with any of the known proteins; thus it is difficult to associate the encoded protein with a functionally defined molecular family. On the other hand, the nucleic acid sequence is 88% homologous with mouse PIBF (GenBank accession number AX452710).

The recombinant protein reacts with Abs raised by immunization with the secreted PIBF of lymphocyte origin, suggesting that the recombinant PIBF is related to that, secreted by the lymphocytes (22, 23). The recombinant human PIBF proved to be immunologically active, exerting similar effects to PIBF, secreted by pregnancy lymphocytes. The NK inhibitory activity is associated with the sequence encoded by exons 2–4, whereas the region responsible for modulating cytokine production could not be localized within the 48-kDa N-terminal part of the molecule.

When PIBF was originally discovered, the biological functions were identified with a 34–36 kDa secreted protein (7). In this study, we showed that the cDNA of PIBF encodes a 90-kDa molecule. However, the recombinant protein reacts with an Ab that was raised by immunizing with the 34-kDa PIBF of lymphocyte origin. This suggests that the latter corresponds to a part of the full-length recombinant molecule.

In resting cells the full-length PIBF is associated with the nucleus and secretion of different protein forms that are synthesized due to alternative splicing begins upon activation of the cell. 4

Computer analysis of the PIBF1 sequence indicated a 94% probability of a nuclear localization (Fig. 1⇑). Transfection of PR-positive mammary carcinoma (MCF-7) cells with a plasmid coding for the full-length human PIBF resulted in a nucleus-associated localization as detected by immunofluorescence. Western blot analysis of the nuclear extract from untransfected MCF-7 cells revealed a 90-kDa protein, corresponding to the full-length PIBF. These data suggest that after translation, the full-length protein is associated with the nucleus. The presence of the basic zipper sequence suggests that it might act as a transcription factor. Smaller forms, which do not contain NLS or basic zipper sequences, might represent the secreted molecule. Insulin, which stimulates proliferation of MCF-7 cells (24) and activates the cells via different signal pathways (25), treatment of MCF-7 cells results in the appearance of PIBF containing cytoplasmic secretory granules, and smaller m.w. proteins appear in the culture medium of the cells. Based on these findings it cannot be excluded that the product of the same gene acts as a transcription factor and as a cytokine. Cytokines are known to signal by interactions with cell surface receptors coupled to cytoplasmic components that transmit the signal. A growing body of evidence indicates that some cytokines also must act within the cell to exert their complete effect (26). Several cytokines or their receptors contain NLSs, and after internalization these cytokine-receptor complexes may move into the nucleus (27). A protein binds via its NLS to a complex called importin (28), which mediates docking to the nuclear pore complex and translocation across the nuclear envelope (29). A number of cytokines or their receptors contain putative NLSs. Nuclear translocation has been observed for several of these cytokines, including insulin, IFN-γ, IL-1, IL-5, PDGFs, growth hormone, and members of the fibroblast growth factor family (30, 31, 32, 33). In some cases the NLS function has been confirmed by demonstration of the ability to target a heterologous protein to the nucleus (33). Based on the nucleus-associated localization of the full-length PIBF together with the presence of the NLS, PIBF might act in a similar way.

Although the proteins expressed in prokaryotic vectors do not carry posttranslational modifications, the lack of these modifications did not influence the biological effects of the recombinant human PIBF. The biological effects of the recombinant PIBF are similar to those described earlier for the lymphocyte-produced PIBF, suggesting that glycosylation of the molecule is not indispensable for the biological effect.

The N-terminal 48-kDa part of PIBF inhibits NK activity to the same extent as the secreted lymphocyte PIBF does and the latter effect is neutralized with Abs specific for the recombinant PIBF. These data suggest that the active site, responsible for modulating NK activity lies within the 48-kDa N-terminal part of the molecule.

Similarly to the secreted PIBF the recombinant molecule increases IL-10 production and inhibits IL-12 production by peripheral lymphocytes, both in a concentration-dependent manner. The low percentage of cytokine-expressing cells is in line with that found in the literature (34).

To localize the sites responsible for the biological effects, three constructs representing exons 2–4 (PN1), 5–7 (PN2), and 8 and 9 (PN3) were produced. The mAbs reacting with these peptides were used for epitope mapping. The mAbs M2, M3, M6, and M9 recognized sequential epitopes on the peptide encoded by exons 2–4, whereas M4 and M8 recognize conformational epitopes in the same area. M7 recognizes a conformational epitope on the peptide coded for by exons 5–7. M1 and M5 recognize sequential epitopes encoded by exons 8 and 9. PN1 significantly inhibited NK activity of nonpregnancy lymphocytes, whereas PN2 and PN3 had no effect. Although the 48-kDa N-terminal part of the recombinant PIBF induced a Th2 dominant cytokine production, neither of the constructs affected cytokine expression of nonpregnancy lymphocytes in a significant manner. The failure to associate cytokine effects with either of the constructs might be due to the fact that the regions responsible for modulating cytokine production are localized elsewhere, or alternatively, that the constructs were designed in a way that they do not contain intact and functional receptor binding sites. Our findings suggest that the 48-kDa N-terminal part of the recombinant PIBF is biologically active; no posttranslational modifications are needed for the biological effect and the part responsible for modulating NK activity is encoded by exons 2–4.

PIBF was originally described as a progesterone-induced secreted protein. However, in this study we have shown that it is present in MCF-7 mammary carcinoma cells in absence of progesterone. A search in cDNA databases revealed that PIBF cDNA is a characteristic of undifferentiated/proliferating cells, and recently we have demonstrated PIBF in a variety of malignant tumors. 5 Furthermore, Rozenblum et al. (21) identified the PIBF1 gene on the chromosomal region 13q21-q22, which has been implicated as a common site for somatic deletions in a variety of malignant tumors. The relationship between progesterone-induced and progesterone-independent PIBF production as well as the conditions that result in translation are still to be clarified, but it cannot be excluded that the PIBF is of more general importance than shown earlier.

Footnotes

  • ↵1 This work was supported by Grants from the Hungarian National Research Fund (OTKA T031737), the Hungarian Ministry of Health (ETT 347/2000), and the Hungarian Academy of Sciences.

  • ↵2 Address correspondence and reprint requests to Dr. Julia Szekeres-Bartho, Department of Medical Microbiology and Immunology, Medical School, Pecs University, 12 Szigeti Str, H-7643 Pecs, Hungary. E-mail address: szjuli{at}main.pote.hu

  • ↵3 Abbreviations used in this paper: PR, progesterone receptor; PIBF, progesterone-induced blocking factor; GFP, green fluorescence protein; DAPI, 4′,6′-diamidino-2-phenylindole; NLS, nuclear localization signal.

  • ↵4 M. Lachmann, D. Gelbmann, C. Paar, E. Kalman, M. Buschle, B. Polgar, J. Szekeres-Bartho, and E. Nagy. Expression and cellular localization of the progesterone induced blocking factor (PIBF) in malignant cells. Submitted for publication.

  • ↵5 B. Polgar, E. Miko, L. Szereday, I. Monstad, I. Vona, G. Farkas, Z. Baliko, M. Schmelczer, L. Farkas, M. Lachmann, E. Nagy, and J. Szerkeres-Bartho. PIBF (progesterone induced blocking factor) is a tumor marker and has immunomodulatory activity relevant to cancer therapy. Submitted for publication.

  • Received June 2, 2003.
  • Accepted September 9, 2003.

References

  1. ↵
    Szekeres-Bartho, J., M. F. Reznikoff-Etievant, P. Varga, Z. Varga, G. Chaouat. 1989. Lymphocytic progesterone receptors in human pregnancy. J. Reprod. Immunol. 16:239.
    OpenUrlCrossRefPubMed
  2. ↵
    Szekeres-Bartho, J., Gy. Szekeres, P. Debre, B. Autran, G. Chaouat. 1990. Reactivity of lymphocytes to a progesterone receptor-specific monoclonal antibody. Cell. Immunol. 125:273.
    OpenUrlCrossRefPubMed
  3. ↵
    Chiu, L., M. Nishimure, Y. Ishii, M. Nieda, M. Maeshima, Y. Takedani, Y. Shibata, K. Tadokoro, T. Juji. 1996. Enhancement of the expression of progesterone receptor on progesterone-treated lymphocytes after immuntherapy in unexplained spontaneous abortion. Am. J. Reprod. Immunol. 35:552.
    OpenUrlCrossRefPubMed
  4. ↵
    Van den Heuvel, M., B. A. McBey, A. C. Hahnel, B. A. Croy. 1996. An analysis of the uterine lymphocyte-derived hybridoma cell line GWM 1–2 for expression of receptors for estrogen, progesterone and interleukin 2. J. Reprod. Immunol. 31:37.
    OpenUrlCrossRefPubMed
  5. ↵
    Roussev, R. G., N. G. Higgins, J. A. McIntyre. 1993. Phenotypic characterization of normal human placental mononuclear cells. J. Reprod. Immunol. 25:15.
    OpenUrlCrossRefPubMed
  6. ↵
    Polgar, B., A. Barakonyi, I. Xynos, J. Szekeres-Bartho. 1999. The role of γ/δ T cell receptor positive cells in pregnancy. Am. J. Reprod. Immunol. 41:239.
    OpenUrlCrossRefPubMed
  7. ↵
    Szekeres-Bartho, J., F. Kilar, G. Falkay, V. Csernus, A. Torok, A. S. Pacsa. 1985. Progesterone-treated lymphocytes of healthy pregnant women release a factor inhibiting cytotoxicity and prostaglandin synthesis. Am. J. Reprod. Immunol. Microbiol. 9:15.
    OpenUrlPubMed
  8. ↵
    Szekeres-Bartho, J., B. Autran, P. Debre, G. Andreu, L. Denver, G. Chaouat. 1989. Immunoregulatory effects of a suppressor factor from healthy pregnant women’s lymphocytes after progesterone induction. Cell. Immunol. 122:2.
    OpenUrl
  9. ↵
    Szekeres-Bartho, J., Zs. Faust, P. Varga. 1995. Progesterone-induced blocking factor (PIBF) in normal and pathological pregnancy. Am. J. Reprod. Immunol. 34:342.
    OpenUrlCrossRefPubMed
  10. ↵
    Faust, Z., G. Laškarin, D. Rukavina, J. Szekeres-Bartho. 1999. Progesterone induced blocking factor inhibits degranulation of NK cells. Am. J. Reprod. Immunol. 42:71.
    OpenUrlPubMed
  11. ↵
    Szekeres-Bartho, J., Z. Faust, P. Varga, L. Szereday, K. Kelemen. 1996. The immunological pregnancy protective effect of progesterone is manifested via controlling cytokine production. Am. J. Reprod. Immunol. 35:348.
    OpenUrlCrossRefPubMed
  12. ↵
    Barakonyi, A., B. Polgar, J. Szekeres-Bartho. 1999. The role of γ/δ TCR positive cells in pregnancy: part II. Am. J. Reprod. Immunol. 42:83.
    OpenUrlPubMed
  13. ↵
    Makhseed, M., R. Raghupathy, S. El-Shazly, F. Azizieh, J. A. Al-Harmi, M. M. Al-Azemi. 2003. Pro-inflammatory maternal cytokine profile in preterm delivery. Am. J. Reprod. Immunol. 49:308.
    OpenUrlCrossRefPubMed
  14. ↵
    Makhseed, M., R. Raghupathy, F. Azizieh, A. Omu, E. Al-Shamali, L. Ashkanani. 2001. Th1 and Th2 cytokine profiles in recurrent aborters with successful pregnancy and with subsequent abortions. Hum. Reprod. 16:2219.
    OpenUrlAbstract/FREE Full Text
  15. ↵
    Szekeres-Bartho, J., G. Par, L. Szereday, C. Y. Smart, I. Achatz. 1997. Progesterone and non-specific immunological mechanisms in pregnancy. Am. J. Reprod. Immunol. 38:176.
    OpenUrlCrossRefPubMed
  16. ↵
    Szekeres-Bartho, J., G. Par, G. Dombay, C. Y. Smart, Z. Volgyi. 1997. The anti-abortive effect of PIBF in mice is manifested by modulating NK activity. Cell. Immunol. 177:194.
    OpenUrlCrossRefPubMed
  17. ↵
    Check, J. H., M. Arwitz, J. Gross, M. Meymer, J. Szekeres-Bartho. 1997. Lymphocyte immunotherapy increases serum levels of progesterone-induced blocking factor (PIBF). J. Reprod. Immunol. 37:1.
    OpenUrlCrossRefPubMed
  18. ↵
    Short, J. M., J. A. Sorge. 1992. In vivo excision properties of bacteriophage lambda ZAP expression vectors. Methods Enzymol. 216:495.
    OpenUrlCrossRefPubMed
  19. ↵
    Sambrook, J., E. F. Fritsch, T. Maniatis. 1989. Molecular Cloning: A Laboratory Manual 2nd Ed.63. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
  20. ↵
    Sanger, F., S. Nicklen, A. R. Coulson. 1977. DNA sequencing with chain terminating inhibitors. Proc. Natl. Acad. Sci. USA 74:5463.
    OpenUrlAbstract/FREE Full Text
  21. ↵
    Rozenblum, E., P. Vahteristo, T. Sandberg, J. T. Bergthorsson, K. Syrjakoski, D. Weaver, K. Haraldsson, H. K. Johannsdottir, P. Vehmanen, S. Nigam, et al 2002. A genomic map of a 6-MB region at 13q21–q22 implicated in cancer development: identification and characterization of candidate genes. Hum. Genet. 110:111.
    OpenUrlCrossRefPubMed
  22. ↵
    Szekeres-Bartho, J., R. Kinsky, G. Chaouat. 1990. The effect of a progesterone induced immunologic blocking factor on NK-mediated resorption. Am. J. Reprod. Immunol. 24:105.
    OpenUrlCrossRefPubMed
  23. ↵
    Szekeres-Bartho, J., R. Kinsky, G. Chaouat. 1990. A progesterone-induced immunologic blocking factor corrects high resorption rate in mice treated with antiprogesterone. Am. J. Obstet. Gynecol. 163:1320.
    OpenUrlCrossRefPubMed
  24. ↵
    Hailey, J., E. Maxwell, K. Koukouras, W. R. Bishop, J. A. Pachter, Y. Wang. 2002. Neutralizing anti-insulin-like growth factor receptor 1 antibodies inhibit receptor function and induce receptor degradation in tumor cells. Mol. Cancer Ther. 1:1349.
    OpenUrlAbstract/FREE Full Text
  25. ↵
    Pawelczyk, T., M. Sakowicz, M. Podgorska, M. Szczepanska-Konkel. 2003. Insulin induces expression of adenosine kinase gene in rat lymphocytes by signaling through the mitogen-activated protein kinase pathway. Exp. Cell Res. 286:152.
    OpenUrlCrossRefPubMed
  26. ↵
    Leaman, D. W., S. Leung, X. Li, G. R. Stark. 1996. Regulation of STAT-dependent pathways by growth factors and cytokines. FASEB J. 10:1578.
    OpenUrlAbstract
  27. ↵
    Gorlich, D., I. W. Mattaj. 1996. Nucleocytoplasmic transport. Science 271:1513.
    OpenUrlAbstract
  28. ↵
    Dingwall, C., R. A. Laskey. 1991. Nuclear targeting sequences: a consensus?. Trends Biochem. Sci. 16:478.
    OpenUrlCrossRefPubMed
  29. ↵
    Johnson, H. M., B. A. Torres, M. M. Green, B. E. Szente, K. I. Siler, J. Larkin, P. S. Subramaniam. 1998. Cytokine-receptor complexes as chaperones for nuclear translocation of signal transducers. Biochem. Biophys. Res. Commun. 244:607.
    OpenUrlCrossRefPubMed
  30. ↵
    Lobie, P. E., H. Mertni, G. Morel, O. Morales-Bustos, G. Norstedt, M. J. Waters. 1994. Receptor-mediated nuclear translocation of growth hormone. J. Biol. Chem. 269:21330.
    OpenUrlAbstract/FREE Full Text
  31. ↵
    Podlecki, D. A., R. M. Smith, M. Kao, P. Tsai, T. Huecksteadt, D. Brandenburg, R. S. Lasher, L. Jarett, J. M. Olefski. 1987. Nuclear translocation of the insulin receptor: a possible mediator of insulin’s long term effects. J. Biol. Chem. 262:3362.
    OpenUrlAbstract/FREE Full Text
  32. ↵
    Grenfell, S., N. Smithers, K. Miller, R. Solari. 1989. Receptor-mediated endocytosis and nuclear transport of human interleukin 1α. J. Biochem. 264:813.
    OpenUrl
  33. ↵
    Jans, D. A.. 1994. Nuclear signaling pathways for polypeptide ligands and their membrane receptors. FASEB J. 8:841.
    OpenUrlAbstract
  34. ↵
    Szegedi, A., M. Aleksza, A. Gonda, B. Irinyi, S. Sipka, J. Hunyadi, P. Antal-Szalmás. 2003. Elevated rate of T helper1 (TH1) lymphocytes and serum IFN-γ levels in psoriatic patients. Immunol. Lett. 86:277.
    OpenUrlCrossRefPubMed
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section