Cells and at Immune-Privileged Sites Dendritic Inhibitor 9, Is Mainly

The Granzyme B Inhibitor, Protease Inhibitor 9, Is Mainly Expressed by Dendritic Cells and at Immune-Privileged Sites1

Bellinda A. Bladergroen,* Merel C. M. Strik,† Niels Bovenschen,* Oskar van Berkum,* George L. Scheffer,* Chris J. L. M. Meijer,* C. Erik Hack,†‡ and J. Alain Kummer2*

Granzyme B is released from CTLs and NK cells and an important mediator of CTL/NK-induced apoptosis in target cells. The human intracellular serpin proteinase inhibitor (PI)9 is the only human protein able to inhibit the activity of granzyme B. As a first step to elucidate the physiological role of PI9, PI9 protein expression in various human tissues was studied. A mAb directed against human PI9 was developed, which specifically stained PI9-transfected COS-7 cells, and was used for immunohistochemistry. Both in primary lymphoid organs and in inflammatory infiltrates, PI9 was present in different subsets of dendritic cells. Also T-lymphocytes in primary and organ-associated lymphoid tissues were PI9 positive. Endothelial cells of small vessels in most organs tested as well as the endothelial layer of large veins and arteries showed strong PI9 staining. Surprisingly, high PI9 protein expression was also found at immune-privileged sites like the placenta, the testis, the ovary, and the eye. These data fit with the hypothesis that PI9 is expressed at sites where degranulation of CTL or NK cells is potentially deleterious. The Journal of Immunology, 2001, 166: 0000–0000.

ytotoxic cells such as NK cells and CTL form an impor- phocytes, and Fas-associated death domain-like IL-1-converting tant line of defense against virally infected cells and tu- enzyme-inhibitory protein (3), which directly inhibits the Fas-me- C mor cells. CTL and NK cells kill target cells by inducing diated signaling pathway (6). Fas ligand is expressed by target apoptosis in the latter. This mainly occurs via two pathways, one cells in immune-privileged sites like testis and the eye (7, 8). involving the death receptor induced apoptosis (mainly the Fas As the apoptotic machinery is mediated by activation of a cas- ligand/Fas pathway), the other being dependent on the exocytosis cade of proteases, cells could also escape apoptosis by expressing of cytotoxic granules from the effector cell. Cytotoxic granules intracellular protease inhibitors. Indeed, inhibition of apoptosis has contain perforin, which forms pores in the membrane of the target been shown for the poxvirus serine protease inhibitor (serpin) cell, and several serine proteases, termed granzymes. Studies with crmA, which mainly inhibits caspase 1, but may also inhibit Fas- purified components (1), transfected cells (2), and knock-out mice mediated apoptosis by inhibition of caspase 8 (reviewed in Ref. 9). (3, 4) have established perforin to be essential for membrane lysis, Furthermore, the human intracellular serpin plasminogen activator and granzyme B to be necessary for rapid target cell DNA frag- inhibitor 2 (PAI-2)3 can protect cells from TNF-induced apoptosis mentation and apoptosis. The signaling pathway involving Fas li- (10). CrmA and PAI-2 both belong to the subfamily of OVA ser- gand and Fas is an important complementary route, as evidenced pins that are usually present as intracellular cytoplasmic proteins by the residual, slow cytotoxic activity in perforin-deficient mice, (9). The function of these serpins is largely unknown. Recently, a which is abrogated using target cells from Fas-deficient lpr/lpr novel human intracellular serpin, proteinase inhibitor (PI) 9, was mice (5). Importantly, the granzyme B- as well as Fas-mediated described to be expressed by CTLs. This serpin efficiently inhibits apoptosis is finally executed by unique cysteine proteases granzyme B in vitro and in vivo and, hence, PI9 transfected cells (caspases). Granzyme B shares its unusual substrate specificity are protected against granzyme B mediated apoptosis (11, 12). with these caspases. Consequently, this mediator of cytotoxicity Using RT-PCR and Northern blotting it was shown that PI9 can directly trigger the endogenous apoptotic machinery. mRNA is mainly expressed by B and T cell lines and, therefore, it Recent studies indicate that normal and pathological cells can was proposed that PI9 protects CTLs against death induced by protect themselves against apoptotic stimuli at different levels. their own misdirected granzyme B (11). Cells of the immune system as well as certain tumor cells can At the moment PI9 is the only human inhibitor known to inter- express several proteins that confer resistance to Fas-mediated ap- fere with granzyme B mediated apoptosis. As a first step to resolve optosis, such as Fas ligand, which kills attacking cytotoxic lym- the (patho)physiological function of this inhibitor, we decided to study the expression of this inhibitor in various human tissues.

Departments of *Pathology and †Clinical Chemistry, Free University Hospital, Am- sterdam, the Netherlands; and ‡Department of Pathophysiology of Plasma Proteins, Materials and Methods Central Laboratory of The Netherlands Red Cross Blood Transfusion Service, Am- sterdam, The Netherlands Materials and cells Received for publication August 9, 2000. Accepted for publication December COS-7 cells (CRL-1651; American Type Culture Collection, Manassas, 11, 2000. VA) were described in Ref. 13. YT-Indy cells were kindly provided by Dr. The costs of publication of this article were defrayed in part by the payment of page C.J. Froelich (Department of Research, Evanston Hospital, Northwestern charges. This article must therefore be hereby marked advertisement in accordance University, Evanston, IL). DMEM and IMDM were obtained from Bio- with 18 U.S.C. Section 1734 solely to indicate this fact. Whittaker Europe (Verviers, Belgium). FBS was purchased from Life 1 This study was supported by the Dutch Cancer Foundation (Grant VU-98-1718). 2 Address correspondence and reprint requests to Dr. J. Alain Kummer, Department of Pathology, Free University Hospital, P.O. Box 7057, 1007 MB Amsterdam, The 3 Abbreviations used in this paper: PAI-2, plasminogen activator inhibitor 2; PI, pro- Netherlands. E-mail address: [email protected] teinase inhibitor; sABC, avidine-biotine-HRP complex.

Technologies (Rockville, MD). All primers were synthesized by Eurogen- bodies, containing the rPI9, were pelleted by centrifugation at 14,000 rpm tec (Brussels, Belgium). Qiaex II Gel Extraction Kit and Qiagen Plasmid for 15 min and washed twice with 0.5% (w/v) Triton X-100 in TE buffer. Maxi Kit were obtained from Qiagen (Hilden, Germany). The pCR-Script Purified rPI9 inclusion bodies were resuspended in PBS to a final concen- Amp SKϩ cloning vector was obtained from Stratagene (La Jolla, CA) and tration of ϳ1 mg/ml, shortly sonicated, and used for immunization. pcDNA3.1/Hygroϩ from Invitrogen (Carlsbad, CA). Epicurian Coli XL1- Blue MRF’Kan super competent cells were obtained from Stratagene, and Production and purification of anti-PI9 mAbs JM109 high efficiency competent cells were obtained from Promega (Mad- Female BALB/c mice were immunized both s.c. and in the footpads ison, WI). The Thermo Sequenase DNA sequencing kit was purchased with 50 ␮g of purified rPI9 inclusion bodies suspended in CFA. This from Amersham (Arlington Heights, IL). mAb GB7 (subtype IgG2a) rec- was followed by one or two subsequent booster injections of 50 ␮g ognizing granzyme B was produced and purified as described previously inclusion bodies suspended in IFA at 10-day intervals. Three days after (14). Anti-CD21 mAb 1F8 (subtype IgG1), anti-S100 polyclonal Ab, bi- the last booster injection, popliteal lymph nodes were collected. Lym- Ј otinylated rabbit-anti-mouse F(ab )2 Ig, biotinylated swine-anti-rabbit phocytes were isolated and fused with mouse Sp2/0 myeloma cells as Ј F(ab )2, HRP-conjugated rabbit-anti-mouse Ig, avidine-biotine-HRP com- described previously (16). Hybridomas producing immunoreactive anti- plex (sABC), biotinylated tyramine, and streptavidine-FITC were obtained PI9 mAbs were identified by screening culture medium of the hybrid- from Dako (Glostrup, Denmark). Biotin-labeled goat-anti-mouse IgG1 and omas with ELISA on immobilized, rPI9 containing bacterial E. coli HRP-labeled goat-anti-mouse IgG2a Abs were obtained from Southern BL21 (DE3)pLysE lysate. Selected anti-PI9-producing hybridomas Biotechnology Associates (Birmingham, AL) and tyramine-rhodamine were subcloned by three cycles of limiting dilution. Anti-PI9 mAbs from DuPont Pharmaceuticals (Wilmington, DE). The constructs PI6- were purified from conditioned medium by protein A-Sepharose affinity pcDNA3 and PI8-pcDNA3 were a kind gift from Dr. W. Kisiel (Depart- chromatography. ment of Pathology, University of New Mexico, Albuquerque, NM). PAI- 2-pCI-neo was kindly provided by Dr. E.K. Kruithof (Division of Immunohisto/cytochemistry Angiology and Hemostasis, University Hospital Geneva, Geneva, Switzer- land). FuGENE 6 transfection reagent was obtained from Roche Molecular Slides with sections of formalin-fixed, paraffin-embedded normal human Biochemicals (Indianapolis, IN). 4-(2-aminoethyl)-benzenesulfonylfluo- tissues were obtained from the tissue bank of the Department of Pathology ride.HCl was obtained from A.G. Scientific (San Diego, CA) and apo- at the Free University Hospital (Amsterdam, the Netherlands). All tissues transferrine from Sigma (St. Louis, MO). The Micro BCA Protein Assay were sampled from surgical specimens within2hofresection. Tissues was obtained from Pierce (Rockford, IL). were processed routinely by fixation in 10% formalin for 18 h and subsequently embedded in paraffin. Three-micrometer thick sections were Cell culture mounted on poly-L-lysine-coated slides. Cytospins of serpin-transfected COS-7 cells (see below) were fixed in 10% formalin for 1 h. COS-7 cells were grown in DMEM supplemented with 10% (v/v) heat- Endogenous peroxidase activity was blocked by incubation for 30 min inactivated FBS, 2 mM L-glutamine and penicillin/streptomycin at final with 0.3% (v/v) H2O2 in methanol. Unless stated otherwise, tissue sections concentrations of 50 IU/ml and 50 ␮g/ml, respectively. YT-Indy cells were and cytospins were subjected to Ag retrieval by boiling in 0.01 M sodium- cultured in IMDM containing 10% (v/v) heat-inactivated FBS, 2 mM citrate (pH 6) for 10 min in a microwave oven. In case sections were L-glutamine, 50 IU/ml penicillin/50 ␮g/ml streptomycin, and 20 ␮g/ml stained for the dendritic cell marker S100, reactivity was enhanced by apo-transferrine. All cells were maintained under an atmosphere of 5% incubation with 0.5% (w/v) trypsin in 0.5% (w/v) CaCl2 for 30 min at CO2/95% air in a humidified incubator at 37°C. 37°C. Tissue sections and cytospins were stained for PI9 with purified To obtain cell lysates, cells were pelleted, washed twice with PBS (pH mAb 17 at 2.5 and 10 ␮g/ml, respectively. To identify specific cell pop- 7.4), and resuspended in PBS with 1% (w/v) Nonidet P-40 at a concen- ulations, sequential sections of several tissues were incubated with the tration of 40 ϫ 106 cells/ml. YT-Indy cells were also lysed in PBS/1% appropriate Abs against various cell markers. In the figures, staining with Nonidet P-40 supplemented with a protease inhibitor mixture consisting of the following Abs is shown: anti-granzyme B mAb GB7 1:500 (1.2 ␮g/ml), 1 mM 4-(2-aminoethyl)-benzenesulfonylfluoride⅐HCl, 10 mM EDTA, 10 anti-CD21 mAb 1F8 1:25 and anti-S100 polyclonal Ab 1:400. The sec- mM benzamidine, 5 mM N-ethylmaleimide, 1 ␮g/ml pepstatin and 50 ␮M Ј ondary Ab was biotinylated rabbit-anti-mouse F(ab )2 Ig 1:500 diluted or iodoacetamide. Cells were lysed for 30 min on ice, whereafter the lysate Ј biotinylated swine-anti-rabbit F(ab )2 1:300 in case the first Ab was a was centrifuged for 10 min at 3000 rpm to remove cell debris and DNA. mouse monoclonal or a rabbit polyclonal Ab, respectively. Biotinylated The supernatant (cell lysate) was stored at Ϫ20°C until further use. secondary Abs were detected with sABC. On tissue sections stained for cell markers, bound Abs were visualized by incubation with diaminobenzidine/ Cloning and production of rPI9 H2O2. However, PI9 stained tissue sections were further incubated with Construction and amplification of the cDNA coding for the entire PI9 pro- biotinylated tyramine. After a second incubation with sABC PI9 was vi- tein was achieved by PCR with primers 5Ј-GGATTCCCATATG sualized with 3-amino-9-ethylcarbazole. Cytospins of transfected COS-7 GAAACTCTTTCTAA-3Ј (sense) and 5Ј-CGGAATTCTATGGCGAT cells were incubated with 3-amino-9-ethylcarbazole directly after the first GAGAACCTGC-3Ј (antisense). The primers were deduced from the N- incubation with sABC. Slides were counterstained with hematoxylin and and C-terminal ends of the published cDNA sequence for human PI9 (15) mounted. Negative control slides were stained with mouse IgG of the ap- and NdeI and EcoRI restriction sites were introduced in the sense and propriate subclass. antisense primers, respectively. PCR was performed on first strand cDNA For colocalization of PI9 and granzyme B double immunofluorescence prepared with random hexamer primers from RNA extracted from a BJAB staining was performed on sections of tonsil tissue as follows. After Ag ␮ cell line. The PCR product was purified by agarose gel electrophoresis and retrieval, sections were stained with mAb 17 at 2.5 g/ml as well as with ␮ isolated from the gel with Qiaex II Gel Extraction Kit. The purified cDNA GB7 at 12 g/ml for 1 h. PI9 was detected by subsequent incubation with was ligated blunt end into the SrfI site of the pCR-Script Amp SKϩ Clon- biotin-labeled goat-anti-mouse IgG1, sABC, biotinylated tyramine, and ing Vector and transformed to Epicurian Coli XL1-Blue MRF’Kan super streptavidine-FITC. After blocking the remaining peroxidase activity with competent cells according to the manufacturer’s instructions. The authen- 0.3% (v/v) H2O2 in methanol, sections were incubated with HRP-labeled ticity of the cloned cDNAs was confirmed by nucleotide sequence analysis goat-anti-mouse IgG2a followed by tyramine-rhodamine to detect gran- Ј Ј according to the Thermo-Sequenase protocol. zyme B. Sections were counterstained with 4 ,6 -diamidino-2-phenylin- The PI9 cDNA was cut from PI9-pCR-Script by digestion with NdeI dole and mounted. and EcoRI and subcloned into the expression vector pRSET6A yielding Transfections expression under the control of the T7 RNA polymerase promoter. For the expression of rPI9, the plasmid PI9-pRSET6A was transformed to the The PI9-pCR-Script construct was used as a template in a PCR with prim- Escherichia coli expression host BL21(DE3)pLysE, which expresses T7 ers containing either BamHI or EcoRI restriction sites (sense primer 5Ј- RNA polymerase under the control of the isopropyl-␤-D-thio-galactopyr- TCGGATCCCATATGGAAACTCTTTCT-3Јand antisense primer 5Ј-AC anoside-inducible lac UV 5 promoter. Several colonies were used to inoc- GAATTCTTATGGCGATGAGAACCT-3Ј). The resulting PCR product ulate LB medium supplemented with ampicillin and chloramphenicol and was ligated between the BamHI and EcoRI sites of the cloning vector the bacteria were grown at 37°C until the absorbance at 600 nm was be- pBluescript. Subsequently, a cDNA fragment harboring the region coding tween 0.3 and 0.5. To induce rPI9 expression, isopropyl-␤-D-thio-galacto- for the entire PI9 protein was excised from the pBluescript vector with pyranoside was added to a final concentration of 1 mM and the cells were BamHI and ApaI. The excised fragment was ligated into the mammalian grown for an additional 4 h. The cells were harvested by centrifugation at expression vector pcDNA3.1/Hygro. The constructs PI8-pcDNA3 and PI6- 4,000 rpm for 15 min and resuspended in TE buffer (10 mM Tris-HCl (pH pcDNA3 comprised the cDNA sequence coding for either the full-length 8.0), 1 mM EDTA) containing 1 mM phenylmethylsulfonyl fluoride. Cells PI8 or PI6 protein as an insert in the expression vector pcDNA3. A cDNA were lysed by three freeze/thaw cycles and sonication. Insoluble inclusion fragment coding for the PAI-2 protein was present as an insert in the 3220 EXPRESSION OF PROTEINASE INHIBITOR 9 IN NORMAL HUMAN TISSUES

expression vector pCI-neo. Each of these expression vectors contains an mouse Ig. Bound Abs were visualized with a chemiluminescence devel- SV40 origin of replication for amplification to high copy number in COS opment reagent (ECL system; Amersham) according to the manufacturer’s monkey cells and a cytomegalovirus promoter for transcription of inserted instructions. cDNA. Plasmid DNA was amplified by transformation to JM109 cells and isolated with Qiagen plasmid maxi kit. A total of 2 ϫ 105 COS-7 cells were grown in 25 cm2 culture flasks Results for 24 h in DMEM with 10% heat-inactivated FBS, antibiotics, and Characterization of mAbs directed against human PI9 L-glutamine. Medium was refreshed just before transfection with 3.8 ␮g isolated PI9-pcDNA3.1/Hygro, PI8-pcDNA3, PI6-pcDNA3, or PAI-2- From a fusion experiment of a mouse immunized with recombi- pCI-neo. Control cells were transfected with empty pcDNA3.1/Hygro nant human PI9 isolated from inclusion bodies, 29 mAbs were vector. Transfection was performed with FuGENE 6 transfection re- obtained. In screening experiments, one of these mAbs, mAb 17 agent. FuGENE 6 transfection reagent (microliters) to plasmid DNA (micrograms) was used in a ratio of 5:1. At 48 h after transfection, cells (subtype IgG1), appeared to be specific for PI9, whereas another were trypsinized and washed twice with PBS. Part of the cells were mAb, mAb 1 (subtype IgG2b), turned out to cross-react with sev- resuspended in PBS plus 1% (w/v) BSA and used to prepare cytospins. eral other intracellular serpins. These two mAbs were used in ad- The remaining cells were lysed in lysis buffer (PBS, 1% (w/v) Nonidet ditional experiments. Immunohistochemistry of serpin-transfected ϫ 6 P-40) at a concentration of 3 10 cells/ml lysis buffer. After lysis for COS-7 cells with mAb 17 supported the specificity of this mAb; it 30 min on ice, the lysate was cleared by centrifugation and stored at Ϫ20°C until further use. only stained PI9-transfected COS-7 cells (Fig. 1A, d), but not COS-7 cells transfected with the two most homologous serpins PI6 Immunoblotting and PI8 (Fig. 1A, b and c, respectively) or with the related serpin Lysates of tissue sections were prepared by dissolving five frozen tissue PAI-2 (Fig. 1A, a). Furthermore, in accordance with the intracel- sections of 5-␮m thick in 50 ␮l PBS with 1% (w/v) Nonidet P-40. Protein lular localization of PI9, a strong cytoplasmic staining was seen was measured according to the procedure of the Micro BCA Protein Assay with mAb 17. ␮ with BSA as a standard. About 5 g of tissue section lysate protein or 15 On Western blot, mAb 17 only recognized a 42-kDa band in ␮l of cell lysate from transfected COS-7 cells was resolved by electrophoresis on a 10% SDS-polyacrylamide gel under reducing conditions. lysates of PI9 transfected COS-7 cells (Fig. 1B). This band is con- From YT-Indy cell lysates only 0.5–5 ␮l 10-times-diluted cell lysate was sistent with the expected size of PI9. YT-Indy cells, a NK leuke- loaded. After electrophoresis, proteins were transferred to nitrocellulose mia cell line, also contained substantial levels of PI9. However, membranes by electrophoretic blotting. The membranes were then blocked most PI9 in YT-Indy cells was present either in the form of a for1hinblocking buffer (5% (w/v) skim milk powder, 0.5% (w/v) BSA, and 0.1% (v/v) Tween 20 in PBS). Subsequently, membranes were incu- 67-kDa band or a 38-kDa band (Fig. 1B, lane 7). The 67 kDa form bated for 2.5 h or overnight with either 3.6 ␮g/ml mAb 17 or 8.5 ␮g/ml appeared to be the SDS-resistant complex of PI9 and the serine mAb 1 diluted in blocking buffer, followed by HRP-conjugated rabbit anti- protease granzyme B, because this band was also recognized by an

FIGURE 1. Anti-PI9 mAb 17 is specific for PI9. A, Cytospins of COS-7 cells transfected with PAI-2-pCI-neo (a), PI6-pcDNA3 (b), PI8-pcDNA3 (c), and PI9-pcDNA3.1/Hygro (d) were stained with 10 ␮g/ml mAb 17 as described in Materials and Methods. B, COS-7 cells were transfected with the above- mentioned constructs and lysed. Control cells were transfected with empty plasmid pcDNA3.1/Hygro. Also YT-Indy cells were lysed in the absence (YT-Indy Ϫ) or presence (YT-Indy ϩ) of protease inhibitor mixture. A total of 15 ␮l COS-7 cell lysate and 5 ␮l (YT- Indy Ϫ)or2.5␮l (YT-Indy ϩ) 10-times-diluted YT-Indy cell lysate were analyzed by 10% SDS-PAGE and immunoblotting. Blots were incubated for 2.5 h with mAb 17 (left)or overnight with mAb 1 (right). The molecular mass of marker bands is indicated on the left and right. The Journal of Immunology 3221 anti-granzyme B Ab (data not shown). The 38-kDa band presum- are derived from CTL or NK cells as confirmed by the coexpres- ably represented PI9 cleaved in the reactive site loop by granzyme sion of granzyme B (Fig. 2n). In contrast, only few T cells present B. Both the 67-kDa complex and the 38-kDa cleaved form of PI9 in the thymic cortex and medulla expressed PI9, whereas most disappeared at the expense of a 42-kDa band representing un- lymphocytes in this organ were negative (Fig. 2, g and h). cleaved serpin, when YT-Indy cells were lysed in the presence of protease inhibitors (Fig. 1B, lane 6). PI9 is also expressed by endothelial cells and cells present in In contrast to mAb 17, mAb 1, though raised against rPI9, also immune-privileged sites ϳ recognized a band of 42 kDa in COS-7 cells transfected with PI6 In most nonlymphoid organs tested, like the gastrointestinal tract, or PI8. PAI-2 was not recognized although it was present in an breast, prostate, liver, kidney, lung, brain, and heart (Fig. 3, a–h), active form in the COS-7 cells because lysates of PAI-2 trans- the epithelial and mesanchymal structures were PI9 negative. fected cells inhibited urokinase, a target protease of PAI-2 (results However, the endothelial cells of the small vessels present in these not shown). Notably, both mAbs also recognized a second, higher organs were strongly positive (arrows, Fig. 3, d–h) and actually molecular mass protein band in lysates of all serpin-transfected could be used as a positive internal control. Also in large veins and COS-7 cells, of which the identity is unknown yet. Because this arteries (Fig. 3,i and j, respectively), the endothelial layer was protein was not detected on the cytospins of serpin-transfected positive, while the smooth muscle cells present in the medium of COS-7 cells (Fig. 1A), mAb 17 was considered to be specific for the larger vessel wall were negative. PI9 and it was used for further immunohistochemical studies. Surprisingly, high PI9 protein expression was found in immune- Dendritic cells express high levels of PI9 privileged sites such as the eye (lens), the testis, the ovary, and the placenta (Fig. 3, k–o). In the testis, PI9 was mainly localized in the Analysis of PI9 tissue distribution (Fig. 2 and 3) by immunohis- Sertoli cells (Fig. 3l, arrow) but low expression was also found in tochemistry showed that in the lymphoid organs like tonsil (Fig. 2, the spermatogonia and the primary spermatocytes. However, the a–f), lymph node (not shown), thymus (Fig. 2, g–i), and spleen Leydig cells were negative. In the ovary, the granulosa cells sur- (Fig. 2, j–l) a strong cytoplasmic staining for PI9 was present in rounding the primordial follicle and primary follicle (Fig. 3m) different subsets of dendritic cells. Langerhans cells in the epithe- were PI9 positive while the ovarial stroma was negative. The gran- lial layers (Fig. 2b), follicular dendritic cells in the follicle center ulosa cell layer and theca interna in maturing follicles were neg- of tonsil and spleen (Fig. 2, d and j, respectively), interdigiting ative (not shown). Fig. 3, n and o, show placental tissue from an cells in the mantle zone (Fig. 2c, arrowhead), and dendritic cells in extra-uterine, tubal pregnancy in the first trimester. A strong PI9 the medulla of the thymus (Fig. 2h) showed a strong cytoplasmic expression was seen predominantly in the cytotrophoblast but staining. The identity of the PI9-positive cells was confirmed by hardly in the syncytiotrophoblast (Fig. 3n). In Fig. 3o the inter- staining with several markers like CD21 expressed in the tonsil mediate trophoblast (X cells) infiltrating the tubal wall is shown. follicular center (Fig. 2e) and S100 in the thymic medulla (Fig. 2i). These cells expressed high levels of PI9 as was also found in the PI9-positive dendritic cells were not only detected in the primary chorion present in the placental membrane (not shown). The am- lymphoid organs but also in inflammatory infiltrates present in niotic layer was negative. The same results were found in placental several other organs such as the prostate (Fig. 3c, arrow). These tissue from the second and third trimester (data not shown). No cells also expressed S100 (result not shown). Several isotype con- staining was observed using isotype control Abs (results not trol mAbs (Fig. 2a) did not show any background staining in any shown). Ј of the organs tested. In addition, F(ab )2 fragments of mAb17 showed an identical immunohistochemical staining pattern com- Confirmation of PI9 tissue distribution by Western blot pared with the intact mAb, indicating that there was no specific interaction of the Fc portion of the Ab with the dendritic cells The distribution pattern of PI9 among the various tissues as seen (results not shown). In contrast to dendritic cells, different macro- by immunohistochemical staining, as well as the identity of the phage subsets present in lymphoid organs, including the tingible protein, was confirmed by Western blotting (Fig. 4). Consistent body macrophages (see Fig. 2, c and d, asterisk) or the sinus mac- with immunohistochemistry results high amounts of PI9 were rophages, or in nonlymphoid organs, like Kupffer cells in liver and found in lymphoid tissues like tonsil and spleen and the immune- alveolar macrophages in lung (Fig. 3, d and f), were clearly privileged sites placenta and testis. Very faint PI9 bands, probably negative. due to PI9-positive endothelial cells present in these tissues, could Various types of B cells present in the follicle center (centro- be detected in lysates of kidney and lung. Lysates of heart and cytes, centroblasts) in lymphoid organs were PI9 negative (Fig. brain were negative for PI9. 2d). However, in the spleen several B cells, especially those In placenta, testis, and tonsil, mAb 17 detected a band of 42 present in the marginal zone, showed a strong cytoplasmic PI9 kDa, consistent with the size of uncleaved PI9 protein. In the expression (Fig. 2l), whereas B cells in the mantle zone showed spleen, mAb 17 recognized a band of 38 kDa, indicating that PI9 only weak staining (Fig. 2k). Mature plasma cells, present in a was completely cleaved in this tissue by a yet unknown protease. chronic inflammatory reaction in a patient with ulcerative colitis, did not express PI9 (Fig. 2o, arrowhead). Discussion T cells present in the paracortex in lymphoid tissue and in the Immunostaining and Western blot results demonstrated that the organ-associated lymphoid tissues like mucosa-associated lym- granzyme B inhibitor, PI9, is present in many normal human tis- phoid tissue were PI9 positive. Fig. 3a shows PI9-positive intra- sues. However, immunostaining showed that within these tissues epithelial lymphocytes present in the villi of the small intestine. PI9 expression is restricted to only certain cell types. Both in lym- Double staining of tonsil with anti-granzyme B and PI9 mAbs phoid and nonlymphoid tissues high levels of PI9 were found in showed PI9 expression in granzyme B positive cells (Fig. 2f, ar- dendritic cells, T-lymphocytes, and endothelial cells. Other impor- row) indicating that CTL express PI9. This was also the case in the tant PI9-producing cells appeared to be the intermediate tropho- endometrial mucosa where numerous PI9-positive lymphocytes blastic cells in placental tissue, Sertoli cells in testis, granulosa were detected (Fig. 2m). These cells represented the endometrial cells in ovary, and lens cells in the eye, i.e., cells of immune- stromal granulocytes, also called the decidual lymphocytes, that privileged sites. 3222 EXPRESSION OF PROTEINASE INHIBITOR 9 IN NORMAL HUMAN TISSUES

FIGURE 2. Immunohistochemical detection of PI9 in lymphoid tissues. Tissue sections of tonsil (a–f), thymus (g–i), spleen (j–l), endometrium (m and n) and colon (o) were stained with a nonspecific isotype control mAb (a), anti-PI9 mAb 17 (b–d, g–h, j–m, o), anti-CD21 mAb 1F8 (e), anti-S100 polyclonal Ab (i), or anti-granzyme B mAb GB7 (n). Section f was double stained with mAb 17 and GB7. Magnification, ϫ400 (a, g, j), ϫ630 (c and f). Higher magnification (ϫ1000) showing PI9 staining in Langerhans cells present in tonsil epithelium (b), dendritic cells in tonsil follicular center (d) and thymic medulla (h), B-lymphocytes in the mantle zone (k) and marginal zone (l) of the splenic white pulp, lymphocytes in endometrium (m), and a chronic Tingible body macrophage PI9 negative; ‹, interdigiting cell PI9 positive; 3, CTL expressing both ,ء .(inflammatory infiltrate in ulcerative colon (o granzyme B (red) and PI9 (green) in yellow; ‹, plasma cell PI9 negative.

It was previously reported that in addition to CTLs PI9 is par- spleen are highly positive for PI9. PI9 was absent in macrophages, ticularly produced in B-lymphocytes (12). These conclusions were plasma cells, epithelial cells, mesanchymal cells, and muscle cells. based on analysis of PI9 protein expression in EBV-transformed B These results show that the expression pattern of PI9 protein cell lines. However, immunostaining results from the present study differs from that of other OVA serpins. PI6 is predominantly syn- revealed that different B cell types in lymphoid organs are negative thesized in epithelial cells and capillary endothelial cells (17) that for PI9. In contrast, B-cells present in the marginal zone in the are negative and positive for PI9, respectively. In peripheral blood The Journal of Immunology 3223

FIGURE 3. Immunohistochemical detection of PI9 in normal human tissues. The following tissues were stained for PI9: small intestine (a); breast (b); prostate (c); liver (d); kidney (e); lung (f); brain (g); heart (h); artery (i); vein (j); lens (k); testis (l); ovary (m); placenta villus (n), and intermediate trophoblast (o). Tissue sections were stained with anti-PI9 mAb 17 as described in Materials and Methods and counterstained with hematoxylin. Mag- nification, ϫ400 (b–f); ϫ630 (a, g–k, n, o); ϫ1000 (l, m). leukocytes, PI6 is present in monocytes and granulocytes, but macrophages (19). In contrast, both PI9 and PAI-2 are highly not in lymphocytes (18). In contrast, in the present study PI9 synthesized by trophoblast cells. Also maspin is present in the was observed in T-lymphocytes in various tissues, while mac- placenta where it is localized to cytotrophoblast cells (20). rophages and neutrophils were negative for PI9. Although Maspin is associated with epithelial cells in many organs, Northern blot analysis revealed that PI8 mRNA is widely dis- whereas PI9 is not detected in epithelia. tributed in a variety of tissues (15), the cells that produce PI8 The detection of PI9 in T-lymphocytes is in agreement with protein are not identified yet due to a lack of specific anti-PI8 earlier reports from studies with human leukocyte cell lines (11, Abs. Unlike PI9, PAI-2 is mainly expressed in monocytes and 12). The function of PI9 in these cells is presumed to provide 3224 EXPRESSION OF PROTEINASE INHIBITOR 9 IN NORMAL HUMAN TISSUES

blast cells (27) express Fas-L on their cell surface, that may function as an immunosuppressive agent by eliminating infiltrating Fas-expressing lymphocytes. So, on the one hand tissues produce different molecules to de- fend themselves against CTL-induced apoptosis. On the other hand, many tissues also express more than one serpin. For in- stance, in the placenta various serpins including PI9, PI6 (17), PI8 (15), PAI-2 (19), and maspin (20) are expressed. Because all these serpins contain different P1 residues in their unique reactive site loop, they will specifically inhibit different proteases. Therefore, the expression of a variety of serpins in one tissue will provide additional protection against the effects of various proteases. To FIGURE 4. PI9 tissue distribution on Western blot. Five 5-␮m thick fully understand the precise role of the various serpins in each sections of the indicated tissues were dissolved in 50 ␮l PBS plus 1% tissue, further research will be necessary to determine the identity Nonidet P-40. Equal amounts of lysate protein (5 ␮g) were separated by of the physiological targets of the different serpins. 10% SDS-PAGE and analyzed by immunoblotting with mAb 17. For con- In conclusion, despite the widespread distribution of PI9 in trol purposes, 0.5 ␮l of 10-times-diluted lysate of YT-Indy cells, lysed in many tissues, PI9 expression appears to be restricted to cell types the presence of protease inhibitor mixture, were applied. The position of that can meet substantial levels of granzyme B released from the molecular mass marker bands is indicated on the right. neighboring or circulating CTLs. Furthermore, resistance to granzyme B-induced apoptosis is absolutely essential for these cell types to keep their function. Therefore, based on its distribution, protection against their own misdirected granzyme B (11). Gran- PI9 can be proposed to play a role in vivo in such diverse processes zyme B is the predominant mediator of early induction of DNA as initiation of an immune response, preserving the integrity of fragmentation and CTL-induced apoptosis of target cells, particu- vessels and maintenance of immune privilege. larly tumor cells, as has been shown in studies with knockout mice (3). PI9 efficiently inhibits granzyme B in vitro and in vivo (11, Acknowledgments 12). This suggests that by expressing PI9 cells may become pro- We thank Vanessa Oude Engberink for the support of the PI9 cDNA clon- tected against a cytotoxic attack. Indeed, PI9-transfected cells re- ing and Hans Wiessenberg for additional technical support. sist granzyme B-induced but not Fas-induced apoptosis (11). In addition to granzyme B, also other proteases can be inhibited by References PI9. PI9 binds to and inhibits the activity of the serine protease 1. Young, J. D., H. Hengartner, E. R. Podack, and Z. A. Cohn. 1986. Purification elastase (21) although with less efficiency than granzyme B (as- and characterization of a cytolytic pore-forming protein from granules of cloned ϫ 6 Ϫ1 Ϫ1 ϫ 5 lymphocytes with natural killer activity. Cell 44:849. sociation rate constant Ka is 1.7 10 M s and 1.5 10 Ϫ Ϫ 2. Nakajima, H., H. L. Park, and P. A. Henkart. 1995. Synergistic roles of gran- M 1s 1 for complex formation of PI9 with granzyme B (12) and zymes A and B in mediating target cell death by rat basophilic leukemia mast cell elastase (21), respectively). Although it is not an effective caspase tumors also expressing cytolysin/perforin. J. Exp. Med. 181:1037. 3. Heusel, J. W., R. L. Wesselschmidt, S. Shresta, J. H. Russell, and T. J. Ley. 1994. inhibitor (11), PI9 has been reported to inhibit caspase 1 (22, 23) Cytotoxic lymphocytes require granzyme B for the rapid induction of DNA frag- and caspase 4 (11) in vitro. mentation and apoptosis in allogeneic target cells. Cell 76:977. Because it is a specific inhibitor of granzyme B, PI9 present in 4. Kagi, D., B. Ledermann, K. Burki, P. Seiler, B. Odermatt, K. J. Olsen, E. R. Podack, R. M. Zinkernagel, and H. Hengartner. 1994. Cytotoxicity medi- dendritic cells can be postulated to protect these cells against gran- ated by T cells and natural killer cells is greatly impaired in perforin-deficient zyme B released from neighboring CTLs and NK cells. Dendritic mice. Nature 369:31. 5. Lowin, B., M. Hahne, C. Mattmann, and J. Tschopp. 1994. Cytolytic T-cell cells are also resistant to apoptosis through the Fas ligand/Fas cytotoxicity is mediated through perforin and Fas lytic pathways. Nature 370: pathway (24), the other important pathway via which apoptosis is 650. mediated. Thus, dendritic cells seem to have more than one de- 6. Irmler, M., M. Thome, M. Hahne, P. Schneider, K. Hofmann, V. Steiner, J. L. Bodmer, M. Schroter, K. Burns, C. Mattmann, et al. 1997. Inhibition of fense mechanism against apoptosis. The resistance of dendritic death receptor signals by cellular FLIP. Nature 388:190. cells to apoptosis might be essential for their function as APCs and 7. Sugihara, A., S. Saiki, M. Tsuji, T. Tsujimura, Y. Nakata, A. Kubota, T. Kotake, their ability to initiate an immune response. Obviously, more stud- and N. Terada. 1997. Expression of Fas and Fas ligand in the testes and testicular germ cell tumors: an immunohistochemical study. Anticancer Res. 17:3861. ies need to be done to unravel the role of PI9 in dendritic cells. 8. Griffith, T. S., T. Brunner, S. M. Fletcher, D. R. Green, and T. A. Ferguson. 1995. Also endothelial cells have to be protected against circulating Fas ligand-induced apoptosis as a mechanism of immune privilege. Science 270: granzymes, which under certain circumstances can be present in 1189. 9. Bird, P. I. 1998. Serpins and regulation of cell death. Results Probl. Cell Differ. the blood at high concentrations (25). High levels of PI9 in endo- 24:63. thelial cells may account for the protection against granzyme B, 10. Dickinson, J. L., E. J. Bates, A. Ferrante, and T. M. Antalis. 1995. Plasminogen activator inhibitor type 2 inhibits tumor necrosis factor ␣-induced apoptosis: ev- while other serpins might interact with granzyme A. The ability of idence for an alternate biological function. J. Biol. Chem. 270:27894. PI9 to inhibit elastase activity has been suggested to protect en- 11. Bird, C. H., V. R. Sutton, J. Sun, C. E. Hirst, A. Novak, S. Kumar, J. A. Trapani, dothelial cells from elastase-mediated degradation of the extracel- and P. I. Bird. 1998. Selective regulation of apoptosis: the cytotoxic lymphocyte serpin proteinase inhibitor 9 protects against granzyme B-mediated apoptosis lular matrix during neutrophil extravasation (21). without perturbing the Fas cell death pathway. Mol. Cell Biol. 18:6387. At immune-privileged sites like placenta and testis, PI9 may be 12. Sun, J., C. H. Bird, V. Sutton, L. McDonald, P. B. Coughlin, T. A. De Jong, an important means for cells to escape from CTL-induced apopto- J. A. Trapani, and P. I. Bird. 1996. A cytosolic granzyme B inhibitor related to the viral apoptotic regulator cytokine response modifier A is present in cytotoxic sis. Trophoblastic cells invading endometrium meet many gran- lymphocytes. J. Biol. Chem. 271:27802. zyme B-producing cells (Ref. 26 and Fig. 2n). In testis, cells have 13. Gluzman, Y. 1981. SV40-transformed simian cells support the replication of early SV40 mutants. Cell 23:175. developed mechanisms to evade immune responses. In addition to 14. Kummer, J. A., A. M. Kamp, T. M. Tadema, W. Vos, C. J. Meijer, and PI9 as an inhibitor of granzyme B, these organs display other strat- C. E. Hack. 1995. Localization and identification of granzymes A and B-express- egies to maintain immune privilege and escape from unwanted ing cells in normal human lymphoid tissue and peripheral blood. Clin. Exp. Im- munol. 100:164. apoptosis induced by activated lymphocytes. Sertoli cells, Leydig 15. Sprecher, C. A., K. A. Morgenstern, S. Mathewes, J. R. Dahlen, S. K. Schrader, cells, and germ cells in the testis (7) as well as placental tropho- D. C. Foster, and W. Kisiel. 1995. Molecular cloning, expression, and partial The Journal of Immunology 3225

characterization of two novel members of the ovalbumin family of serine pro- 22. Annand, R. R., J. R. Dahlen, C. A. Sprecher, P. De Dreu, D. C. Foster, teinase inhibitors. J. Biol. Chem. 270:29854. J. A. Mankovich, R. V. Talanian, W. Kisiel, and D. A. Giegel. 1999. Caspase-1 16. Flens, M. J., M. A. Izquierdo, G. L. Scheffer, J. M. Fritz, C. J. Meijer, (interleukin-1␤-converting enzyme) is inhibited by the human serpin analogue R. J. Scheper, and G. J. Zaman. 1994. Immunochemical detection of the multi- proteinase inhibitor 9. Biochem. J. 342:655. drug resistance-associated protein MRP in human multidrug-resistant tumor cells 23. Young, J. L., G. K. Sukhova, D. Foster, W. Kisiel, P. Libby, and U. Schonbeck. by monoclonal antibodies. Cancer Res. 54:4557. 2000. The serpin proteinase inhibitor 9 is an endogenous inhibitor of interleukin 17. Scott, F. L., P. B. Coughlin, C. Bird, L. Cerruti, J. A. Hayman, and P. Bird. 1996. 1␤-converting enzyme (caspase-1) activity in human vascular smooth muscle Proteinase inhibitor 6 cannot be secreted, which suggests it is a new type of cells. J. Exp. Med. 191:1535. cellular serpin. J. Biol. Chem. 271:1605. 24. Ashany, D., A. Savir, N. Bhardwaj, and K. B. Elkon. 1999. Dendritic cells are 18. Scott, F. L., C. E. Hirst, J. Sun, C. H. Bird, S. P. Bottomley, and P. I. Bird. 1999. resistant to apoptosis through the Fas (CD95/APO-1) pathway. J. Immunol. The intracellular serpin proteinase inhibitor 6 is expressed in monocytes and 163:5303. granulocytes and is a potent inhibitor of the azurophilic granule protease, cathep- sin G. Blood 93:2089. 25. Spaeny-Dekking, E. H., W. L. Hanna, A. M. Wolbink, P. C. Wever, 19. Jensen, P. H. 1997. Structure and function of plasminogen activator inhibitor-2: A. J. Kummer, A. J. Swaak, J. M. Middeldorp, H. G. Huisman, C. J. Froelich, and an intracellular serine proteinase inhibitor modulating apoptosis. Int. J. Oncol. C. E. Hack. 1998. Extracellular granzymes A and B in humans: detection of 11:557. native species during CTL responses in vitro and in vivo. J. Immunol. 160:3610. 20. Pemberton, P. A., A. R. Tipton, N. Pavloff, J. Smith, J. R. Erickson, 26. Konno, R., T. Igarashi, S. Okamoto, S. Sato, T. Moriya, H. Sasano, and Z. M. Mouchabeck, and M. C. Kiefer. 1997. Maspin is an intracellular serpin that A. Yajima. 1999. Apoptosis of human endometrium mediated by perforin and partitions into secretory vesicles and is present at the cell surface. J. Histochem. granzyme B of NK cells and cytotoxic T lymphocytes. Tohoku J. Exp. Med. Cytochem. 45:1697. 187:149. 21. Dahlen, J. R., D. C. Foster, and W. Kisiel. 1999. Inhibition of neutrophil elastase 27. Hammer, A., and G. Dohr. 2000. Expression of Fas-ligand in ﬁrst trimester and by recombinant human proteinase inhibitor 9. Biochim. Biophys. Acta 1451:233. term human placental villi. J. Reprod. Immunol. 46:83.