- Email: [email protected]
Generation and characterization of a monoclonal antibody against canine tissue factor
G Model
ARTICLE IN PRESS
VETIMM-9366; No. of Pages 7
Veterinary Immunology and Immunopathology xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Veterinary Immunology and Immunopathology journal homepage: www.elsevier.com/locate/vetimm
Short communication
Generation and characterization of a monoclonal antibody against canine tissue factor Tracy Stokol a,∗ , Janelle Daddona a,2 , Christine DeLeonardis a,3 , Lynn Dong b,4 , Bettina Wagner a,1 a b
Departments of Population Medicine and Diagnostic Sciences, College of Veterinary Medicine, Cornell University, Ithaca, NY 14853, United States Biomedical Sciences, College of Veterinary Medicine, Cornell University, Ithaca, NY 14853, United States
a r t i c l e
i n f o
Article history: Received 20 February 2015 Received in revised form 12 June 2015 Accepted 1 July 2015 Keywords: Antibodies Monoclonal Flow cytometry Coagulation factor III Hemostasis
a b s t r a c t Tissue factor (TF, coagulation factor III) has recently identified roles in innate immunity and cancer. We generated a murine mAb against canine TF (cTF) cloned from Madin-Darby canine kidney cells and expressed in Chinese Hamster Ovarian (CHO) cells, with an equine IL-4 tag. One clone was selected for purification based on initial screening of CHO cell supernatants. The mAb was further characterized with flow cytometry, immunofluorescent microscopy, immunoblotting and immunohistochemical staining of normal and neoplastic canine tissue. The mAb labeled high, but not low, TF-expressing canine breast cancer (CMT25) and osteosarcoma (HMPOS) cells with flow cytometry and immunofluorescent microscopy. Immunoblotting revealed a 42 kDa protein with homogenized canine brain and CMT25, but not HMPOS, lysates. The mAb labeled renal tubules and glomeruli, intestinal and dermal epithelium, and arteriolar adventitial cells in frozen tissues. Using immunofluorescent microscopy, increased numbers of labeled PBMCs were observed after LPS stimulation. Our results indicate that the anti-cTF mAb detects a protein with the expected tissue distribution and molecular weight of TF in normal, LPS-stimulated and neoplastic canine cells. This mAb may prove useful for exploring the role of TF in neoplastic and infectious disorders in dogs. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Tissue factor (TF, also known as coagulation factor III or tissue thromboplastin) is a protein best known for its role in hemostasis, where it is the main trigger of thrombin generation (Mackman, 2009). Tissue factor is normally expressed on perivascular fibroblasts in the adventitia of arteries and activates coagulation on vessel injury. It is also richly expressed in brain parenchyma and
∗ Corresponding author at: S1-058 Schurman Hall, College of Veterinary Medicine, Upper Tower Road, Ithaca, NY 14583, USA. Tel.: +1 607 253 3255; fax: +1 607 253 3711. E-mail addresses: [email protected] (T. Stokol), [email protected] (J. Daddona), [email protected] (C. DeLeonardis), [email protected] (L. Dong), [email protected] (B. Wagner). 1 S1-068 Schurman Hall, College of Veterinary Medicine, Upper Tower Road, Ithaca, NY 14583, USA. 2 3792 Riverside Drive, Sayre, PA 18840, USA. 3 S1-091 Schurman Hall, College of Veterinary Medicine, Upper Tower Road, Ithaca, NY 14583, USA. 4 Veterinary Research Tower, College of Veterinary Medicine, Upper Tower Road, Ithaca, NY 14583, USA.
various epithelia in human and murine tissues (Drake et al., 1989; Flossel et al., 1994; Luther et al., 1996). In pathological states, inflammatory cytokines and LPS induce TF in monocytes, both on the cell surface and on membrane-derived microvesicles. This induced expression is thought to be the main initiator of disseminated intravascular coagulation in sepsis (Pawlinski and Mackman, 2010). Tumors can also express TF on their membranes or shed microparticles (Andreasen et al., 2014; de la Fuente et al., 2014; Stokol et al., 2011). Tissue factor has been implicated in tumor growth, angiogenesis, metastasis and paraneoplastic thrombosis (Andreasen et al., 2014; Wang et al., 2012). Recent studies have also uncovered roles for TF and hemostasis in immune responses to infectious agents (Antoniak and Mackman, 2014; Armstrong et al., 2013). Murine mAbs against human TF have been used to detect Ag expression in histological sections of normal and diseased tissues (Drake et al., 1989; Flossel et al., 1994; Luther et al., 1996), capture and quantify monocyte- or tumor-derived microparticles (Lee et al., 2012; Tatsumi et al., 2014), and inhibit TF procoagulant activity (Key and Mackman, 2010). High levels of TF Ag have been correlated with tumor grade, incidence of thrombotic events and prognosis in cancer patients (Geddings and Mackman, 2013).
http://dx.doi.org/10.1016/j.vetimm.2015.07.001 0165-2427/© 2015 Elsevier B.V. All rights reserved.
Please cite this article in press as: Stokol, T., et al., Generation and characterization of a monoclonal antibody against canine tissue factor. Vet. Immunol. Immunopathol. (2015), http://dx.doi.org/10.1016/j.vetimm.2015.07.001
G Model VETIMM-9366; No. of Pages 7
ARTICLE IN PRESS T. Stokol et al. / Veterinary Immunology and Immunopathology xxx (2015) xxx–xxx
2
Similarly, microparticle TF activity correlates to bacterial LPS load (Hellum et al., 2014) and is associated with coagulation activation in endotoxemic mice (Wang et al., 2009). This indicates that TF antigen measurement could serve as a diagnostic and prognostic biomarker in patients with cancer or bacterial infections. Unfortunately, murine mAbs raised against human TF do not appear to crossreact with the canine protein (Morrissey et al., 1988). Recently, a rabbit anti-TF mAb was used to detect TF in canine gliomas in formalin-fixed tissue (de la Fuente et al., 2014). We and others have also used polyclonal anti-human or anti-canine TF Ab to detect TF in canine cancer cells (Stokol et al., 2011) or tissues (Andreasen et al., 2014), however, polyclonal Ab are not readily adaptable for capture-based ELISAs used for TF measurement in plasma (Key and Mackman, 2010). Here, we describe the generation and characterization of a murine mAb raised against the canine TF (cTF) ectodomain for potential future test development for TF detection in native and neoplastic canine blood, tissues and cells or further characterizing TF expression in normal and diseased canine tissue.
2. Materials and methods Tissue factor was cloned from Madin-Darby canine kidney (MDCK) cells, as we have previously described (Stokol et al., 2011). Methods for protein expression and mAb production have been described previously in detail (Wagner et al., 2003; Wagner et al., 2005). In brief, the insert was sub-cloned into the mammalian expression vector (pcDNA3.1 (−)/Myc-His, version B, Invitrogen, Carlsbad, CA, USA) containing equine IL-4 (eIL-4) (Wagner et al., 2012). Recombinant cTF was expressed as an eIL-4 fusion protein (rcTF-eIL-4) in Chinese Hamster Ovarian (CHO) cells and purified from serum-free media using an anti-eIL-4 affinity column (Wagner et al., 2012). A BALB/C mouse was immunized as previously described in detail (Wagner et al., 2003), using 50 g purified rcTF-eIL4 protein initially followed by 12.5 g protein booster injections, with an adjuvant (Adjuvant MM, Gerbu, Heidelberg, Germany), for the first 3 immunizations. MAbs were generated by fusion of splenic B cells from the immunized mouse and murine myeloma cells (Wagner et al., 2003). Increased serum anti-fusion product titers were confirmed with an anti-eIL-4 ELISA. Spleen cells were fused to X63-Ag8.653 myeloma cells, then culture supernatants were tested for mAb to the rcTF-eIL4 fusion protein by ELISAs using the purified fusion protein or purified recombinant eIL-4 as Ag in parallel, as previously described (Wagner et al., 2012). Single cell clones detecting rcTF-eIL4 but not eIL-4 were transferred into individual wells of 96 well plates. Supernatants were screened using flow cytometry with 0.5% saponin-permeabilized CHO cells expressing rcTF-eIL-4 or eIL-4 and immunohistochemical staining of tubules in frozen canine kidney sections. A single mAb, #133-2 (isotyped as IgG1) was purified with a protein G column and characterized by reactivity with canine cancer cells, normal canine tissues (kidney, intestine, skin and brain) and LPS-stimulated PBMCs, using immunologic techniques as outlined below. All reagents were from Sigma-Aldrich (St Louis, MO), unless specified. For canine cancer cell lines, we used CMT25 mammary tumor and HMPOS osteosarcoma cells that express high and low TF, based on labeling with a rabbit polyclonal anti-human TF Ab (Stokol et al., 2011). Cells were cultured in L-15 (CMT25) or RPMI-1640 (HMPOS) media, supplemented with 10 mM HEPES (CMT25), 10% FBS (20% for CMT25), sodium pyruvate (1 mM), L-glutamine (2 mM), penicillin (100 U), and streptomycin (100 g). Cells were detached with 0.25% trypsin-EDTA and viability was >90% with trypan blue exclusion (Stokol et al., 2011). Tissue culture reagents were from Life Technologies (Grand Island, NY), except for FBS (Thermo-Scientific, Rockford, IL). PBMCs were isolated from the blood of clinically healthy dogs using double-density gradient centrifugation
(Histopaque 1.070 and 1.119), as described (Ogasawara et al., 2012). Cells were plated in 48-well culture plates (1 × 106 /well) in RPMI1640 with 10% FBS, l-glutamine and penicillin-streptomycin. After 1 h, wells were washed with PBS to remove loosely adherent cells and cells were cultured in media overnight. The following day, cells were stimulated for 4 h with 1 g/mL LPS, using PBS as a vehicle control. For flow cytometry, detached cancer cells (5 × 105 cells/ reaction) were incubated with 20 g/mL mAb #133-2 or isotype control (Jackson ImmunoResearch Laboratories Inc., West Grove, PA), followed by an Alexa488 (A488)-conjugated donkey antimouse IgG (1:200, Life Technologies), both for 15 min on ice. We have previously used a rabbit polyclonal anti-human TF Ab to describe TF expression on canine cancer cells (Stokol et al., 2011), thus we used this same polyclonal Ab (20 g/mL, Sekisui Diagnostics, Exton, PA) as a positive control for TF expression with a rabbit IgG negative control (Jackson ImmunoResearch) followed by an A488 donkey-anti-rabbit IgG (1:200, Life Technologies). Analysis was done on 10,000 cells (FACSCaliburTM , BD Biosciences, Franklin Lakes, NJ) as described (Stokol et al., 2011). For immunohistochemical staining of canine tissues, instructions of a commercial kit were followed (ImmPRESS Peroxidase anti-mouse polymer detection kit, Vector Laboratories, Burlingame, CA). In brief, cryostat frozen sections were fixed in ice-cold acetone, washed, then endogenous peroxidase was quenched with 0.5% hydrogen peroxide. Sections were incubated with 20 g/mL #133-2 or isotype control overnight at 4 ◦ C, followed by 1 h at 37 ◦ C. Ag localization was visualized with AEC substrate (Life Technologies), with light hematoxylin counterstaining. Immunofluorescent microscopy was performed on cancer cells (1 × 105 on fibronectin-coated coverslips) and PBMCs (in 48 well plates, washed after LPS or PBS exposure) as previously described (Stokol et al., 2011), using 20 g/mL mAb #133-2 or isotype control. For immunoblotting, cultured cells were lysed in RIPA buffer (50 mm Tris–HCl, 150 mm sodium chloride, 1% Nonidet P40, 0.5% deoxycholate, 0.1% sodium dodecyl sulfate, 1 mM EDTA) containing protease inhibitors (2 M leupeptin, 10 M aprotonin, 1 mM phenylmethylsulfonyl fluoride). Snap-frozen canine brain was ground with a mortar and pestle in PBS with 2 mM EDTA and protease inhibitors (Halt protease inhibitor cocktail, Thermo-Scientific, Rockford, IL). Ground tissue was centrifuged at 30,000 × g for 20 min at 4 ◦ C and the pellet was sonicated in PBS with 1% sodiumdodecyl-sulfate. Protein concentrations were determined with a commercial kit (DC protein assay, Thermo-Scientific). Separation of proteins was performed with a 10% SDS-PAGE gel and mAb #133-2 (10–15 g/mL) was applied overnight at 4 ◦ C. Signal was detected with a horseradish-peroxidase-based chemiluminescent substrate (Supersignal West Pico, Thermo-Scientific) and autoradiographic or chemiluminescent (ChemiDocTM MP system, Bio-Rad) exposure. To test if the mAb was inhibitory for TF procoagulant activity, we incubated CMT25 cancer cells with mAb #133-2 and performed a two-stage amidolytic assay for generation of activated coagulation factor X, as described (Ogasawara et al., 2012; Stokol et al., 2011). Experiments were repeated a minimum of 3 separate times. Animal use protocols (blood collection for PBMCs, tissue collection for immunohistochemical staining) were approved by the Institutional Animal Care and Use Committee at Cornell University.
3. Results and discussion Based on initial screening of cultured hybridoma supernatants, using purified fusion protein and purified eIL-4 as positive and negative selection markers, flow cytometry on permeabilized CHO cells transiently transfected with cTF-eIL4 or eIL-4, and immunohistochemical staining of renal tubules in frozen canine kidney sections,
Please cite this article in press as: Stokol, T., et al., Generation and characterization of a monoclonal antibody against canine tissue factor. Vet. Immunol. Immunopathol. (2015), http://dx.doi.org/10.1016/j.vetimm.2015.07.001
G Model VETIMM-9366; No. of Pages 7
ARTICLE IN PRESS T. Stokol et al. / Veterinary Immunology and Immunopathology xxx (2015) xxx–xxx
we selected one clone (#133-2) which yielded the strongest results for purification and further characterization. Kidney sections were chosen for this initial screening because the MDCK cells used as a source of cloned cTF were originally derived from canine renal tubular epithelium. The purified mAb was first characterized with flow cytometry on high TF-expressing CMT25 mammary cancer and low TFexpressing HMPOS osteosarcoma cells. The mAb strongly labeled 100% of the high TF-expressing CMT25 breast cancer cells, with more intense labeling than the rabbit polyclonal anti-human TF Ab (Fig. 1A), previously used to characterize TF expression in these cells
3
with flow cytometry and immunofluorescent microscopy (Stokol et al., 2011). In contrast, neither mAb #133-2 nor the polyclonal Ab labeled HMPOS osteosarcoma cells (Fig. 1A), which express little surface TF (Stokol et al., 2011). The mAb results were corroborated by immunofluorescent microscopy staining of fixed cells (Fig. 1B). We also performed immunoblotting on lysed cancer cells with mAb #133-2. A protein was detected at approximately 42 kDa in lysates from CMT25 cells, but not in lysates from HMPOS cells or with lysates from either cell line with the isotype control (Fig. 2A). We also saw a single band of higher molecular weight (approximately 49 kDa) in lysates from CHO cells containing the rcTF-eIL-4-IgG4
Fig. 1. Flow cytometric and immunofluorescent microscopic labeling of high tissue factor (TF)-expressing canine mammary tumor cells (CMT25) and low TF-expressing canine osteosarcoma cells (HMPOS) with anti-canine TF mAb #133-2. (A) With flow cytometry, the mAb labeled CMT25, but not HMPOS, cells. Based on the median fluorescent intensity (MFI), the mAb labeled CMT25 more intensely than the positive Ab control, a rabbit polyclonal anti-human TF Ab, used previously to show TF expression in these two cancer cell lines (Stokol et al., 2011). Solid purple curve and M1 region represent positive cells (>101 arbitrary fluorescence units); green open curves represent negative controls (<101 arbitrary fluorescence units). (B) With immunofluorescent microscopy, only CMT25 cells show strong membrane staining with the mAb (green fluorescence with blue DAPI nuclear counterstaining). Bar = 10 m. These findings recapitulate our previously reported observations with the rabbit polyclonal anti-human TF Ab in these 2 tumor cell lines (Stokol et al., 2011) (For interpretation of the color information in this figure legend, the reader is referred to the web version of the article.).
Please cite this article in press as: Stokol, T., et al., Generation and characterization of a monoclonal antibody against canine tissue factor. Vet. Immunol. Immunopathol. (2015), http://dx.doi.org/10.1016/j.vetimm.2015.07.001
G Model VETIMM-9366; No. of Pages 7 4
ARTICLE IN PRESS T. Stokol et al. / Veterinary Immunology and Immunopathology xxx (2015) xxx–xxx
fusion product but not from CHO cells containing eIL-4 (data not shown). The predicted molecular weight of cTF from the amino acid sequence is 32 kDa, slightly lower than that predicted from the human TF sequence (ExPasy Bioinformatics Research Portal, IB Swiss Institute of Bioinformatics) (Artimo et al., 2012). Differences in post-translational modification likely account for the lower observed molecular weight of cTF compared to human TF, which is a 47 kDa protein on immunoblot analysis (Camerer et al., 1996). For instance, cTF has two predicted N-glycosylation sequences versus 3 such sequences in the human protein (NetNGlyc 1.0 Server, www. cbs.dtu.dk/services/NetNGlyc/; accessed January 2015). Titration of the protein concentration in CMT25 lysates showed that the mAb could detect 5 g of protein, although the band was weak at this low concentration (Fig. 2B). We chose high and low TF-expressing canine cancer cells as model cells for demonstrating mAb reactivity. Since the cTF construct used as the Ag to generate the mAb was derived from the
Fig. 2. Immunoblots of canine (CMT25 breast cancer) and (HMPOS osteosarcoma) cells and homogenized normal canine brain with anti-canine TF mAb #133-2. (A) A protein at approximately 42 kDa is seen with mAb #133-2 (10 g/mL) in lysates (150 g protein, with 5% -mercaptoethanol [BME]) from CMT25, but not HMPOS, cancer cell lines. No bands were seen at this molecular weight in either cell line when blots were labeled with a mouse IgG1 isotype control as the primary antibody. Staining of the membrane with Ponceau S prior to labeling showed equal protein loading. (B) Titration of the protein concentration in CMT25 lysates revealed that mAb #1332 (10 g/mL) could detect 5 g of protein, although the band was faint. For A and B, the secondary horseradish peroxidase (HRP)-conjugated antibody was used at a 1:15,000 dilution and exposure was done with chemiluminescent imaging. (C) A single band of approximately 42 kDa was observed with mAb #133-2 (15 g/mL) in homogenized canine brain lysate (100 g protein) with or without 5% BME. The secondary HRP-conjugated Ab was used at a dilution of 1:8000 and exposure was done with autoradiography. * molecular weight marker.
MDCK tissue culture cell line, which could potentially have transformed in culture, it is essential to show that the mAb can detect TF in non-neoplastic canine tissues. We thus used immunohistochemical staining of snap-frozen canine tissue (kidney, intestine and skin) and immunoblotting of homogenized brain tissue to test whether the mAb could detect constitutively expressed TF in normal canine tissue. We found that mAb #133-2 strongly labeled renal tubular epithelium and glomeruli, intestinal mucosal epithelial cells, hair follicles (particularly the outer part) and stratum spinosum with slightly more intense staining of the stratum granulosum of the skin, and perivascular interstitial cells in the adventitia of arterioles in all three organs (Fig. 3). Reactivity was diffuse in individual cells and was cell-associated, with no staining observed in the extracellular matrix. The peri-arteriolar staining pattern, presumably in perivascular fibroblasts, and staining of intestinal mucosa, stratum granulosum of skin and glomeruli mirrors that seen in frozen and formalin-fixed human tissues after staining with anti-human TF mAbs (Drake et al., 1989; Flossel et al., 1994). However some differences in reactivity were noted as compared to human tissue. We observed staining of hair follicles and stratum spinosum in the skin and renal tubular epithelial cells, which lack TF expression reactivity in human tissue. A similar expression pattern in the skin was seen with a goat anti-rabbit TF Ab (a gift from Dr. U. Pendurthi, University of Texas H.S.C. at Tyler, data not shown), suggesting these species-differences are not Ab-related. We did not see staining of the intestinal subepithelial fibroblastic sheath, as reported for the human intestine (Drake et al., 1989; Flossel et al., 1994). Positive TF staining in renal tubules, however, has been noted in adult mice and embryos (Luther et al., 1996). These data indicate that there are some species differences in the pattern of TF expression in different organs; however, the acrossspecies preservation of TF expression in perivascular interstitial cells exemplifies the importance of TF in this site for triggering hemostasis. Brain contains abundant TF, with staining observed in astrocytes and neuronal bodies in frozen and formalin-fixed human tissues, respectively (Drake et al., 1989; Flossel et al., 1994). Indeed, brain is the source of TF (tissue thromboplastin) and phospholipid that is used as the trigger reagent for the prothrombin time in clotting assays. We thus performed immunoblot analysis with mAb #1332 on homogenized brain from a normal dog and observed a single band at approximately 42 kDa in non-reduced samples (Fig. 2C). Human TF has two disulfide bonds in the ectodomain (Rehemtulla et al., 1991), which are conserved in the canine amino acid sequence (alignment performed by ClustalW, http://www.genome.jp/tools/ clustalw/; accessed January 2015). A slightly higher band was observed in reduced samples, suggesting that the disulfide bonds compact the protein facilitating its diffusion through the gel. Taken together, the immunohistochemical and immunoblotting results indicate that mAb #133-2 can detect native TF that is constitutively expressed in canine tissues. We then tested whether TF was upregulated on the surface of LPS-stimulated monocytes. LPS is a known stimulant of TF in monocytes (Egorina et al., 2005; Ogasawara et al., 2012) and we used adherent PBMCs cultured overnight in plastic tissue culture wells as the source of monocytes. Using immunofluorescent microscopy on non-permeabilized cells, we found that more LPS-stimulated cells expressed TF compared to PBS-treated controls, although only a subset of adherent cells were labeled with the mAb (Fig. 4). Although we washed the wells to remove loosely adherent cells, such as lymphocytes, the retained population likely still consists of a mixture of lymphocytes and monocytes. Also, flow cytometric labeling of LPS-stimulated human PBMCs has shown that only a subset of monocytes upregulate TF after LPS stimulation (Egorina et al., 2005). Our PBMC data shows that the mAb can detect TF induced in monocytes by a pathological stimulus.
Please cite this article in press as: Stokol, T., et al., Generation and characterization of a monoclonal antibody against canine tissue factor. Vet. Immunol. Immunopathol. (2015), http://dx.doi.org/10.1016/j.vetimm.2015.07.001
G Model VETIMM-9366; No. of Pages 7
ARTICLE IN PRESS T. Stokol et al. / Veterinary Immunology and Immunopathology xxx (2015) xxx–xxx
5
Fig. 3. Immunohistochemical staining of snap-frozen cryostat sections of canine kidney, intestine and skin with anti-cTF mAb #133-2. The antibody was applied at 20 g/mL and was detected with an ImmPRESS Peroxidase anti-mouse polymer detection kit (Vector Laboratories). Renal tubular epithelium and glomeruli stained intensely (bar = 100 M). Intestinal epithelium was also strongly stained (bar = 100 M). The stratum spinosum and granulosum and hair follicular epithelium (arrow) were labeled (bar = 100 M), with the inset showing more intense staining in the stratum granulosum (arrowhead, bar = 25 M). Interstitial cells in the adventia of arterioles showed diffuse labeling (bar = 50 M). These blood vessels were in the skin, but similar staining was seen in some arterioles in the other two examined organs. No staining was observed with the mouse IgG1 negative control.
Please cite this article in press as: Stokol, T., et al., Generation and characterization of a monoclonal antibody against canine tissue factor. Vet. Immunol. Immunopathol. (2015), http://dx.doi.org/10.1016/j.vetimm.2015.07.001
G Model VETIMM-9366; No. of Pages 7
ARTICLE IN PRESS T. Stokol et al. / Veterinary Immunology and Immunopathology xxx (2015) xxx–xxx
6
Fig. 4. Immunofluorescent microscopic staining of canine PBMCs cultured overnight then stimulated for 4 h with 1 g/mL LPS or a PBS vehicle control. Cells were labeled with the anti-cTF mAb #133-2 or isotype control (both at 20 g/mL), followed by a secondary A488-conjugated donkey anti-mouse IgG (1:200, green), with DAPI nuclear counterstaining (blue). Bar = 10 m. More cells show membrane staining for TF with mAb #133-2 after stimulation with LPS than the vehicle control (arrows), compatible with LPS-stimulated induction of TF expression. No staining was seen with the negative control (For interpretation of the color information in this figure legend, the reader is referred to the web version of the article.).
Finally, we tested if the mAb could inhibit the generation of activated factor X by high TF-expressing CMT25 cancer cells. No inhibition of activity was observed with the mAb at concentrations between 10 and 200 g/mL (data not shown). In the study herein, we describe a combination of immunologicbased techniques to demonstrate the reactivity of a murine anti-cTF mAb that was raised against the cloned cTF ectodomain expressed in and purified from CHO cells. Our results indicate that the purified mAb shows expected reactivity in canine cancer cells and non-neoplastic cells and tissues, but unfortunately does not inhibit TF-based procoagulant activity in cancer cells. Nevertheless, this mAb may prove useful in characterizing cTF in canine tissue and cells, including in situ expression in various canine neoplasms, and potentially for developing assays to measure TF Ag, in the form of microparticles, in whole blood or plasma. Conflict of interest The authors have no conflicts of interest. Acknowledgement This study was funded by a Morris Animal Foundation (grant #D07CA-047). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.vetimm.2015.07. 001 References Andreasen, E.B., Nielsen, O.L., Tranholm, M., Knudsen, T., Kristensen, A.T., 2014. Expression of tissue factor in canine mammary tumours and correlation with
grade, stage and markers of haemostasis and inflammation. Vet. Comp. Oncol., Epub ahead of print. Antoniak, S., Mackman, N., 2014. Multiple roles of the coagulation protease cascade during virus infection. Blood 123, 2605–2613. Armstrong, M.T., Rickles, F.R., Armstrong, P.B., 2013. Capture of lipopolysaccharide (endotoxin) by the blood clot: a comparative study. PLoS One 8, e80192. Artimo, P., Jonnalagedda, M., Arnold, K., Baratin, D., Csardi, G., de Castro, E., Duvaud, S., Flegel, V., Fortier, A., Gasteiger, E., Grosdidier, A., Hernandez, C., Ioannidis, V., Kuznetsov, D., Liechti, R., Moretti, S., Mostaguir, K., Redaschi, N., Rossier, G., Xenarios, I., Stockinger, H., 2012. ExPASy: SIB bioinformatics resource portal. Nucleic Acids Res. 40, W597–W603. Camerer, E., Pringle, S., Skartlien, A.H., Wiiger, M., Prydz, K., Kolsto, A.B., Prydz, H., 1996. Opposite sorting of tissue factor in human umbilical vein endothelial cells and Madin-Darby canine kidney epithelial cells. Blood 88, 1339–1349. de la Fuente, C., Pumarola, M., Blasco, E., Fernandez, F., Viu, J., Anor, S., 2014. Immunohistochemical evaluation of tissue factor, fibrin/fibrinogen and D-dimers in canine gliomas. Vet. J. 200, 387–392. Drake, T.A., Morrissey, J.H., Edgington, T.S., 1989. Selective cellular expression of tissue factor in human tissues. Implications for disorders of hemostasis and thrombosis. Am. J. Pathol. 134, 1087–1097. Egorina, E.M., Sovershaev, M.A., Bjorkoy, G., Gruber, F.X., Olsen, J.O., Parhami-Seren, B., Mann, K.G., Osterud, B., 2005. Intracellular and surface distribution of monocyte tissue factor: application to intersubject variability. Arterio. Vasc. Biol. 25, 1493–1498. Flossel, C., Luther, T., Muller, M., Albrecht, S., Kasper, M., 1994. Immunohistochemical detection of tissue factor (TF) on paraffin sections of routinely fixed human tissue. Histochemistry 101, 449–453. Geddings, J.E., Mackman, N., 2013. Tumor-derived tissue factor-positive microparticles and venous thrombosis in cancer patients. Blood 122, 1873–1880. Hellum, M., Ovstebo, R., Brusletto, B.S., Berg, J.P., Brandtzaeg, P., Henriksson, C.E., 2014. Microparticle-associated tissue factor activity correlates with plasma levels of bacterial lipopolysaccharides in meningococcal septic shock. Thromb. Res. 133, 507–514. Key, N.S., Mackman, N., 2010. Tissue factor and its measurement in whole blood, plasma, and microparticles. Semin. Thromb. Hemost. 36, 865–875. Lee, R.D., Barcel, D.A., Williams, J.C., Wang, J.G., Boles, J.C., Manly, D.A., Key, N.S., Mackman, N., 2012. Pre-analytical and analytical variables affecting the measurement of plasma-derived microparticle tissue factor activity. Thromb. Res. 29, 80–85. Luther, T., Flossel, C., Mackman, N., Bierhaus, A., Kasper, M., Albrecht, S., Sage, E.H., Iruela-Arispe, L., Grossmann, H., Strohlein, A., Zhang, Y., Nawroth, P.P., Carmeliet, P., Loskutoff, D.J., Muller, M., 1996. Tissue factor expression during human and mouse development. Am. J. Pathol. 149, 101–113. Mackman, N., 2009. The role of tissue factor and factor VIIa in hemostasis. Anesth. Analg. 108, 1447–1452.
Please cite this article in press as: Stokol, T., et al., Generation and characterization of a monoclonal antibody against canine tissue factor. Vet. Immunol. Immunopathol. (2015), http://dx.doi.org/10.1016/j.vetimm.2015.07.001
G Model VETIMM-9366; No. of Pages 7
ARTICLE IN PRESS T. Stokol et al. / Veterinary Immunology and Immunopathology xxx (2015) xxx–xxx
Morrissey, J.H., Fair, D.S., Edgington, T.S., 1988. Monoclonal antibody analysis of purified and cell-associated tissue factor. Thromb. Res. 52, 247–261. Ogasawara, S., Daddona, J., Trimpert, J., Stokol, T., 2012. The effect of recombinant canine interleukin-6 and interleukin-8 on tissue factor procoagulant activity on canine peripheral blood mononuclear cells and purified canine monocytes. Vet. Clin. Pathol. 41, 325–335. Pawlinski, R., Mackman, N., 2010. Cellular sources of tissue factor in endotoxemia and sepsis. Thromb. Res. 125 (Suppl. 1), S70–S73. Rehemtulla, A., Ruf, W., Edgington, T.S., 1991. The integrity of the cysteine 186-cysteine 209 bond of the second disulfide loop of tissue factor is required for binding of factor VII. J. Biol. Chem. 266, 10294–10299. Stokol, T., Daddona, J.L., Mubayed, L.S., Trimpert, J., Kang, S., 2011. Evaluation of tissue factor expression in canine tumor cells. Am. J. Vet. Res. 72, 1097–1106. Tatsumi, K., Antoniak, S., Monroe 3rd, D.M., Khorana, A.A., Mackman, N., 2014. Subcommittee on H. malignancy evaluation of a new commercial assay to measure microparticle tissue factor activity in plasma: communication from the SSC of the ISTH. J. Thromb. Haemost. 12, 1932–1934.
7
Wagner, B., Hillegas, J.M., Babasyan, S., 2012. Monoclonal antibodies to equine CD23 identify the low-affinity receptor for IgE on subpopulations of IgM+ and IgG1+ B-cells in horses. Vet. Immunol. Immunopathol. 146, 125–134. Wagner, B., Radbruch, A., Rohwer, J., Leibold, W., 2003. Monoclonal anti-equine IgE antibodies with specificity for different epitopes on the immunoglobulin heavy chain of native IgE. Vet. Immunol. Immunopathol. 92, 45–60. Wagner, B., Robeson, J., McCracken, M., Wattrang, E., Antczak, D.F., 2005. Horse cytokine/IgG fusion proteins—mammalian expression of biologically active cytokines and a system to verify antibody specificity to equine cytokines. Vet. Immunol. Immunopathol. 105, 1–14. Wang, J.G., Geddings, J.E., Aleman, M.M., Cardenas, J.C., Chantrathammachart, P., Williams, J.C., Kirchhofer, D., Bogdanov, V.Y., Bach, R.R., Rak, J., Church, F.C., Wolberg, A.S., Pawlinski, R., Key, N.S., Yeh, J.J., Mackman, N., 2012. Tumor-derived tissue factor activates coagulation and enhances thrombosis in a mouse xenograft model of human pancreatic cancer. Blood 119, 5543–5552. Wang, J.G., Manly, D., Kirchhofer, D., Pawlinski, R., Mackman, N., 2009. Levels of microparticle tissue factor activity correlate with coagulation activation in endotoxemic mice. J. Thromb. Haemost. 7, 1092–1098.
Please cite this article in press as: Stokol, T., et al., Generation and characterization of a monoclonal antibody against canine tissue factor. Vet. Immunol. Immunopathol. (2015), http://dx.doi.org/10.1016/j.vetimm.2015.07.001
ARTICLE IN PRESS
VETIMM-9366; No. of Pages 7
Veterinary Immunology and Immunopathology xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Veterinary Immunology and Immunopathology journal homepage: www.elsevier.com/locate/vetimm
Short communication
Generation and characterization of a monoclonal antibody against canine tissue factor Tracy Stokol a,∗ , Janelle Daddona a,2 , Christine DeLeonardis a,3 , Lynn Dong b,4 , Bettina Wagner a,1 a b
Departments of Population Medicine and Diagnostic Sciences, College of Veterinary Medicine, Cornell University, Ithaca, NY 14853, United States Biomedical Sciences, College of Veterinary Medicine, Cornell University, Ithaca, NY 14853, United States
a r t i c l e
i n f o
Article history: Received 20 February 2015 Received in revised form 12 June 2015 Accepted 1 July 2015 Keywords: Antibodies Monoclonal Flow cytometry Coagulation factor III Hemostasis
a b s t r a c t Tissue factor (TF, coagulation factor III) has recently identified roles in innate immunity and cancer. We generated a murine mAb against canine TF (cTF) cloned from Madin-Darby canine kidney cells and expressed in Chinese Hamster Ovarian (CHO) cells, with an equine IL-4 tag. One clone was selected for purification based on initial screening of CHO cell supernatants. The mAb was further characterized with flow cytometry, immunofluorescent microscopy, immunoblotting and immunohistochemical staining of normal and neoplastic canine tissue. The mAb labeled high, but not low, TF-expressing canine breast cancer (CMT25) and osteosarcoma (HMPOS) cells with flow cytometry and immunofluorescent microscopy. Immunoblotting revealed a 42 kDa protein with homogenized canine brain and CMT25, but not HMPOS, lysates. The mAb labeled renal tubules and glomeruli, intestinal and dermal epithelium, and arteriolar adventitial cells in frozen tissues. Using immunofluorescent microscopy, increased numbers of labeled PBMCs were observed after LPS stimulation. Our results indicate that the anti-cTF mAb detects a protein with the expected tissue distribution and molecular weight of TF in normal, LPS-stimulated and neoplastic canine cells. This mAb may prove useful for exploring the role of TF in neoplastic and infectious disorders in dogs. © 2015 Elsevier B.V. All rights reserved.
1. Introduction Tissue factor (TF, also known as coagulation factor III or tissue thromboplastin) is a protein best known for its role in hemostasis, where it is the main trigger of thrombin generation (Mackman, 2009). Tissue factor is normally expressed on perivascular fibroblasts in the adventitia of arteries and activates coagulation on vessel injury. It is also richly expressed in brain parenchyma and
∗ Corresponding author at: S1-058 Schurman Hall, College of Veterinary Medicine, Upper Tower Road, Ithaca, NY 14583, USA. Tel.: +1 607 253 3255; fax: +1 607 253 3711. E-mail addresses: [email protected] (T. Stokol), [email protected] (J. Daddona), [email protected] (C. DeLeonardis), [email protected] (L. Dong), [email protected] (B. Wagner). 1 S1-068 Schurman Hall, College of Veterinary Medicine, Upper Tower Road, Ithaca, NY 14583, USA. 2 3792 Riverside Drive, Sayre, PA 18840, USA. 3 S1-091 Schurman Hall, College of Veterinary Medicine, Upper Tower Road, Ithaca, NY 14583, USA. 4 Veterinary Research Tower, College of Veterinary Medicine, Upper Tower Road, Ithaca, NY 14583, USA.
various epithelia in human and murine tissues (Drake et al., 1989; Flossel et al., 1994; Luther et al., 1996). In pathological states, inflammatory cytokines and LPS induce TF in monocytes, both on the cell surface and on membrane-derived microvesicles. This induced expression is thought to be the main initiator of disseminated intravascular coagulation in sepsis (Pawlinski and Mackman, 2010). Tumors can also express TF on their membranes or shed microparticles (Andreasen et al., 2014; de la Fuente et al., 2014; Stokol et al., 2011). Tissue factor has been implicated in tumor growth, angiogenesis, metastasis and paraneoplastic thrombosis (Andreasen et al., 2014; Wang et al., 2012). Recent studies have also uncovered roles for TF and hemostasis in immune responses to infectious agents (Antoniak and Mackman, 2014; Armstrong et al., 2013). Murine mAbs against human TF have been used to detect Ag expression in histological sections of normal and diseased tissues (Drake et al., 1989; Flossel et al., 1994; Luther et al., 1996), capture and quantify monocyte- or tumor-derived microparticles (Lee et al., 2012; Tatsumi et al., 2014), and inhibit TF procoagulant activity (Key and Mackman, 2010). High levels of TF Ag have been correlated with tumor grade, incidence of thrombotic events and prognosis in cancer patients (Geddings and Mackman, 2013).
http://dx.doi.org/10.1016/j.vetimm.2015.07.001 0165-2427/© 2015 Elsevier B.V. All rights reserved.
Please cite this article in press as: Stokol, T., et al., Generation and characterization of a monoclonal antibody against canine tissue factor. Vet. Immunol. Immunopathol. (2015), http://dx.doi.org/10.1016/j.vetimm.2015.07.001
G Model VETIMM-9366; No. of Pages 7
ARTICLE IN PRESS T. Stokol et al. / Veterinary Immunology and Immunopathology xxx (2015) xxx–xxx
2
Similarly, microparticle TF activity correlates to bacterial LPS load (Hellum et al., 2014) and is associated with coagulation activation in endotoxemic mice (Wang et al., 2009). This indicates that TF antigen measurement could serve as a diagnostic and prognostic biomarker in patients with cancer or bacterial infections. Unfortunately, murine mAbs raised against human TF do not appear to crossreact with the canine protein (Morrissey et al., 1988). Recently, a rabbit anti-TF mAb was used to detect TF in canine gliomas in formalin-fixed tissue (de la Fuente et al., 2014). We and others have also used polyclonal anti-human or anti-canine TF Ab to detect TF in canine cancer cells (Stokol et al., 2011) or tissues (Andreasen et al., 2014), however, polyclonal Ab are not readily adaptable for capture-based ELISAs used for TF measurement in plasma (Key and Mackman, 2010). Here, we describe the generation and characterization of a murine mAb raised against the canine TF (cTF) ectodomain for potential future test development for TF detection in native and neoplastic canine blood, tissues and cells or further characterizing TF expression in normal and diseased canine tissue.
2. Materials and methods Tissue factor was cloned from Madin-Darby canine kidney (MDCK) cells, as we have previously described (Stokol et al., 2011). Methods for protein expression and mAb production have been described previously in detail (Wagner et al., 2003; Wagner et al., 2005). In brief, the insert was sub-cloned into the mammalian expression vector (pcDNA3.1 (−)/Myc-His, version B, Invitrogen, Carlsbad, CA, USA) containing equine IL-4 (eIL-4) (Wagner et al., 2012). Recombinant cTF was expressed as an eIL-4 fusion protein (rcTF-eIL-4) in Chinese Hamster Ovarian (CHO) cells and purified from serum-free media using an anti-eIL-4 affinity column (Wagner et al., 2012). A BALB/C mouse was immunized as previously described in detail (Wagner et al., 2003), using 50 g purified rcTF-eIL4 protein initially followed by 12.5 g protein booster injections, with an adjuvant (Adjuvant MM, Gerbu, Heidelberg, Germany), for the first 3 immunizations. MAbs were generated by fusion of splenic B cells from the immunized mouse and murine myeloma cells (Wagner et al., 2003). Increased serum anti-fusion product titers were confirmed with an anti-eIL-4 ELISA. Spleen cells were fused to X63-Ag8.653 myeloma cells, then culture supernatants were tested for mAb to the rcTF-eIL4 fusion protein by ELISAs using the purified fusion protein or purified recombinant eIL-4 as Ag in parallel, as previously described (Wagner et al., 2012). Single cell clones detecting rcTF-eIL4 but not eIL-4 were transferred into individual wells of 96 well plates. Supernatants were screened using flow cytometry with 0.5% saponin-permeabilized CHO cells expressing rcTF-eIL-4 or eIL-4 and immunohistochemical staining of tubules in frozen canine kidney sections. A single mAb, #133-2 (isotyped as IgG1) was purified with a protein G column and characterized by reactivity with canine cancer cells, normal canine tissues (kidney, intestine, skin and brain) and LPS-stimulated PBMCs, using immunologic techniques as outlined below. All reagents were from Sigma-Aldrich (St Louis, MO), unless specified. For canine cancer cell lines, we used CMT25 mammary tumor and HMPOS osteosarcoma cells that express high and low TF, based on labeling with a rabbit polyclonal anti-human TF Ab (Stokol et al., 2011). Cells were cultured in L-15 (CMT25) or RPMI-1640 (HMPOS) media, supplemented with 10 mM HEPES (CMT25), 10% FBS (20% for CMT25), sodium pyruvate (1 mM), L-glutamine (2 mM), penicillin (100 U), and streptomycin (100 g). Cells were detached with 0.25% trypsin-EDTA and viability was >90% with trypan blue exclusion (Stokol et al., 2011). Tissue culture reagents were from Life Technologies (Grand Island, NY), except for FBS (Thermo-Scientific, Rockford, IL). PBMCs were isolated from the blood of clinically healthy dogs using double-density gradient centrifugation
(Histopaque 1.070 and 1.119), as described (Ogasawara et al., 2012). Cells were plated in 48-well culture plates (1 × 106 /well) in RPMI1640 with 10% FBS, l-glutamine and penicillin-streptomycin. After 1 h, wells were washed with PBS to remove loosely adherent cells and cells were cultured in media overnight. The following day, cells were stimulated for 4 h with 1 g/mL LPS, using PBS as a vehicle control. For flow cytometry, detached cancer cells (5 × 105 cells/ reaction) were incubated with 20 g/mL mAb #133-2 or isotype control (Jackson ImmunoResearch Laboratories Inc., West Grove, PA), followed by an Alexa488 (A488)-conjugated donkey antimouse IgG (1:200, Life Technologies), both for 15 min on ice. We have previously used a rabbit polyclonal anti-human TF Ab to describe TF expression on canine cancer cells (Stokol et al., 2011), thus we used this same polyclonal Ab (20 g/mL, Sekisui Diagnostics, Exton, PA) as a positive control for TF expression with a rabbit IgG negative control (Jackson ImmunoResearch) followed by an A488 donkey-anti-rabbit IgG (1:200, Life Technologies). Analysis was done on 10,000 cells (FACSCaliburTM , BD Biosciences, Franklin Lakes, NJ) as described (Stokol et al., 2011). For immunohistochemical staining of canine tissues, instructions of a commercial kit were followed (ImmPRESS Peroxidase anti-mouse polymer detection kit, Vector Laboratories, Burlingame, CA). In brief, cryostat frozen sections were fixed in ice-cold acetone, washed, then endogenous peroxidase was quenched with 0.5% hydrogen peroxide. Sections were incubated with 20 g/mL #133-2 or isotype control overnight at 4 ◦ C, followed by 1 h at 37 ◦ C. Ag localization was visualized with AEC substrate (Life Technologies), with light hematoxylin counterstaining. Immunofluorescent microscopy was performed on cancer cells (1 × 105 on fibronectin-coated coverslips) and PBMCs (in 48 well plates, washed after LPS or PBS exposure) as previously described (Stokol et al., 2011), using 20 g/mL mAb #133-2 or isotype control. For immunoblotting, cultured cells were lysed in RIPA buffer (50 mm Tris–HCl, 150 mm sodium chloride, 1% Nonidet P40, 0.5% deoxycholate, 0.1% sodium dodecyl sulfate, 1 mM EDTA) containing protease inhibitors (2 M leupeptin, 10 M aprotonin, 1 mM phenylmethylsulfonyl fluoride). Snap-frozen canine brain was ground with a mortar and pestle in PBS with 2 mM EDTA and protease inhibitors (Halt protease inhibitor cocktail, Thermo-Scientific, Rockford, IL). Ground tissue was centrifuged at 30,000 × g for 20 min at 4 ◦ C and the pellet was sonicated in PBS with 1% sodiumdodecyl-sulfate. Protein concentrations were determined with a commercial kit (DC protein assay, Thermo-Scientific). Separation of proteins was performed with a 10% SDS-PAGE gel and mAb #133-2 (10–15 g/mL) was applied overnight at 4 ◦ C. Signal was detected with a horseradish-peroxidase-based chemiluminescent substrate (Supersignal West Pico, Thermo-Scientific) and autoradiographic or chemiluminescent (ChemiDocTM MP system, Bio-Rad) exposure. To test if the mAb was inhibitory for TF procoagulant activity, we incubated CMT25 cancer cells with mAb #133-2 and performed a two-stage amidolytic assay for generation of activated coagulation factor X, as described (Ogasawara et al., 2012; Stokol et al., 2011). Experiments were repeated a minimum of 3 separate times. Animal use protocols (blood collection for PBMCs, tissue collection for immunohistochemical staining) were approved by the Institutional Animal Care and Use Committee at Cornell University.
3. Results and discussion Based on initial screening of cultured hybridoma supernatants, using purified fusion protein and purified eIL-4 as positive and negative selection markers, flow cytometry on permeabilized CHO cells transiently transfected with cTF-eIL4 or eIL-4, and immunohistochemical staining of renal tubules in frozen canine kidney sections,
Please cite this article in press as: Stokol, T., et al., Generation and characterization of a monoclonal antibody against canine tissue factor. Vet. Immunol. Immunopathol. (2015), http://dx.doi.org/10.1016/j.vetimm.2015.07.001
G Model VETIMM-9366; No. of Pages 7
ARTICLE IN PRESS T. Stokol et al. / Veterinary Immunology and Immunopathology xxx (2015) xxx–xxx
we selected one clone (#133-2) which yielded the strongest results for purification and further characterization. Kidney sections were chosen for this initial screening because the MDCK cells used as a source of cloned cTF were originally derived from canine renal tubular epithelium. The purified mAb was first characterized with flow cytometry on high TF-expressing CMT25 mammary cancer and low TFexpressing HMPOS osteosarcoma cells. The mAb strongly labeled 100% of the high TF-expressing CMT25 breast cancer cells, with more intense labeling than the rabbit polyclonal anti-human TF Ab (Fig. 1A), previously used to characterize TF expression in these cells
3
with flow cytometry and immunofluorescent microscopy (Stokol et al., 2011). In contrast, neither mAb #133-2 nor the polyclonal Ab labeled HMPOS osteosarcoma cells (Fig. 1A), which express little surface TF (Stokol et al., 2011). The mAb results were corroborated by immunofluorescent microscopy staining of fixed cells (Fig. 1B). We also performed immunoblotting on lysed cancer cells with mAb #133-2. A protein was detected at approximately 42 kDa in lysates from CMT25 cells, but not in lysates from HMPOS cells or with lysates from either cell line with the isotype control (Fig. 2A). We also saw a single band of higher molecular weight (approximately 49 kDa) in lysates from CHO cells containing the rcTF-eIL-4-IgG4
Fig. 1. Flow cytometric and immunofluorescent microscopic labeling of high tissue factor (TF)-expressing canine mammary tumor cells (CMT25) and low TF-expressing canine osteosarcoma cells (HMPOS) with anti-canine TF mAb #133-2. (A) With flow cytometry, the mAb labeled CMT25, but not HMPOS, cells. Based on the median fluorescent intensity (MFI), the mAb labeled CMT25 more intensely than the positive Ab control, a rabbit polyclonal anti-human TF Ab, used previously to show TF expression in these two cancer cell lines (Stokol et al., 2011). Solid purple curve and M1 region represent positive cells (>101 arbitrary fluorescence units); green open curves represent negative controls (<101 arbitrary fluorescence units). (B) With immunofluorescent microscopy, only CMT25 cells show strong membrane staining with the mAb (green fluorescence with blue DAPI nuclear counterstaining). Bar = 10 m. These findings recapitulate our previously reported observations with the rabbit polyclonal anti-human TF Ab in these 2 tumor cell lines (Stokol et al., 2011) (For interpretation of the color information in this figure legend, the reader is referred to the web version of the article.).
Please cite this article in press as: Stokol, T., et al., Generation and characterization of a monoclonal antibody against canine tissue factor. Vet. Immunol. Immunopathol. (2015), http://dx.doi.org/10.1016/j.vetimm.2015.07.001
G Model VETIMM-9366; No. of Pages 7 4
ARTICLE IN PRESS T. Stokol et al. / Veterinary Immunology and Immunopathology xxx (2015) xxx–xxx
fusion product but not from CHO cells containing eIL-4 (data not shown). The predicted molecular weight of cTF from the amino acid sequence is 32 kDa, slightly lower than that predicted from the human TF sequence (ExPasy Bioinformatics Research Portal, IB Swiss Institute of Bioinformatics) (Artimo et al., 2012). Differences in post-translational modification likely account for the lower observed molecular weight of cTF compared to human TF, which is a 47 kDa protein on immunoblot analysis (Camerer et al., 1996). For instance, cTF has two predicted N-glycosylation sequences versus 3 such sequences in the human protein (NetNGlyc 1.0 Server, www. cbs.dtu.dk/services/NetNGlyc/; accessed January 2015). Titration of the protein concentration in CMT25 lysates showed that the mAb could detect 5 g of protein, although the band was weak at this low concentration (Fig. 2B). We chose high and low TF-expressing canine cancer cells as model cells for demonstrating mAb reactivity. Since the cTF construct used as the Ag to generate the mAb was derived from the
Fig. 2. Immunoblots of canine (CMT25 breast cancer) and (HMPOS osteosarcoma) cells and homogenized normal canine brain with anti-canine TF mAb #133-2. (A) A protein at approximately 42 kDa is seen with mAb #133-2 (10 g/mL) in lysates (150 g protein, with 5% -mercaptoethanol [BME]) from CMT25, but not HMPOS, cancer cell lines. No bands were seen at this molecular weight in either cell line when blots were labeled with a mouse IgG1 isotype control as the primary antibody. Staining of the membrane with Ponceau S prior to labeling showed equal protein loading. (B) Titration of the protein concentration in CMT25 lysates revealed that mAb #1332 (10 g/mL) could detect 5 g of protein, although the band was faint. For A and B, the secondary horseradish peroxidase (HRP)-conjugated antibody was used at a 1:15,000 dilution and exposure was done with chemiluminescent imaging. (C) A single band of approximately 42 kDa was observed with mAb #133-2 (15 g/mL) in homogenized canine brain lysate (100 g protein) with or without 5% BME. The secondary HRP-conjugated Ab was used at a dilution of 1:8000 and exposure was done with autoradiography. * molecular weight marker.
MDCK tissue culture cell line, which could potentially have transformed in culture, it is essential to show that the mAb can detect TF in non-neoplastic canine tissues. We thus used immunohistochemical staining of snap-frozen canine tissue (kidney, intestine and skin) and immunoblotting of homogenized brain tissue to test whether the mAb could detect constitutively expressed TF in normal canine tissue. We found that mAb #133-2 strongly labeled renal tubular epithelium and glomeruli, intestinal mucosal epithelial cells, hair follicles (particularly the outer part) and stratum spinosum with slightly more intense staining of the stratum granulosum of the skin, and perivascular interstitial cells in the adventitia of arterioles in all three organs (Fig. 3). Reactivity was diffuse in individual cells and was cell-associated, with no staining observed in the extracellular matrix. The peri-arteriolar staining pattern, presumably in perivascular fibroblasts, and staining of intestinal mucosa, stratum granulosum of skin and glomeruli mirrors that seen in frozen and formalin-fixed human tissues after staining with anti-human TF mAbs (Drake et al., 1989; Flossel et al., 1994). However some differences in reactivity were noted as compared to human tissue. We observed staining of hair follicles and stratum spinosum in the skin and renal tubular epithelial cells, which lack TF expression reactivity in human tissue. A similar expression pattern in the skin was seen with a goat anti-rabbit TF Ab (a gift from Dr. U. Pendurthi, University of Texas H.S.C. at Tyler, data not shown), suggesting these species-differences are not Ab-related. We did not see staining of the intestinal subepithelial fibroblastic sheath, as reported for the human intestine (Drake et al., 1989; Flossel et al., 1994). Positive TF staining in renal tubules, however, has been noted in adult mice and embryos (Luther et al., 1996). These data indicate that there are some species differences in the pattern of TF expression in different organs; however, the acrossspecies preservation of TF expression in perivascular interstitial cells exemplifies the importance of TF in this site for triggering hemostasis. Brain contains abundant TF, with staining observed in astrocytes and neuronal bodies in frozen and formalin-fixed human tissues, respectively (Drake et al., 1989; Flossel et al., 1994). Indeed, brain is the source of TF (tissue thromboplastin) and phospholipid that is used as the trigger reagent for the prothrombin time in clotting assays. We thus performed immunoblot analysis with mAb #1332 on homogenized brain from a normal dog and observed a single band at approximately 42 kDa in non-reduced samples (Fig. 2C). Human TF has two disulfide bonds in the ectodomain (Rehemtulla et al., 1991), which are conserved in the canine amino acid sequence (alignment performed by ClustalW, http://www.genome.jp/tools/ clustalw/; accessed January 2015). A slightly higher band was observed in reduced samples, suggesting that the disulfide bonds compact the protein facilitating its diffusion through the gel. Taken together, the immunohistochemical and immunoblotting results indicate that mAb #133-2 can detect native TF that is constitutively expressed in canine tissues. We then tested whether TF was upregulated on the surface of LPS-stimulated monocytes. LPS is a known stimulant of TF in monocytes (Egorina et al., 2005; Ogasawara et al., 2012) and we used adherent PBMCs cultured overnight in plastic tissue culture wells as the source of monocytes. Using immunofluorescent microscopy on non-permeabilized cells, we found that more LPS-stimulated cells expressed TF compared to PBS-treated controls, although only a subset of adherent cells were labeled with the mAb (Fig. 4). Although we washed the wells to remove loosely adherent cells, such as lymphocytes, the retained population likely still consists of a mixture of lymphocytes and monocytes. Also, flow cytometric labeling of LPS-stimulated human PBMCs has shown that only a subset of monocytes upregulate TF after LPS stimulation (Egorina et al., 2005). Our PBMC data shows that the mAb can detect TF induced in monocytes by a pathological stimulus.
Please cite this article in press as: Stokol, T., et al., Generation and characterization of a monoclonal antibody against canine tissue factor. Vet. Immunol. Immunopathol. (2015), http://dx.doi.org/10.1016/j.vetimm.2015.07.001
G Model VETIMM-9366; No. of Pages 7
ARTICLE IN PRESS T. Stokol et al. / Veterinary Immunology and Immunopathology xxx (2015) xxx–xxx
5
Fig. 3. Immunohistochemical staining of snap-frozen cryostat sections of canine kidney, intestine and skin with anti-cTF mAb #133-2. The antibody was applied at 20 g/mL and was detected with an ImmPRESS Peroxidase anti-mouse polymer detection kit (Vector Laboratories). Renal tubular epithelium and glomeruli stained intensely (bar = 100 M). Intestinal epithelium was also strongly stained (bar = 100 M). The stratum spinosum and granulosum and hair follicular epithelium (arrow) were labeled (bar = 100 M), with the inset showing more intense staining in the stratum granulosum (arrowhead, bar = 25 M). Interstitial cells in the adventia of arterioles showed diffuse labeling (bar = 50 M). These blood vessels were in the skin, but similar staining was seen in some arterioles in the other two examined organs. No staining was observed with the mouse IgG1 negative control.
Please cite this article in press as: Stokol, T., et al., Generation and characterization of a monoclonal antibody against canine tissue factor. Vet. Immunol. Immunopathol. (2015), http://dx.doi.org/10.1016/j.vetimm.2015.07.001
G Model VETIMM-9366; No. of Pages 7
ARTICLE IN PRESS T. Stokol et al. / Veterinary Immunology and Immunopathology xxx (2015) xxx–xxx
6
Fig. 4. Immunofluorescent microscopic staining of canine PBMCs cultured overnight then stimulated for 4 h with 1 g/mL LPS or a PBS vehicle control. Cells were labeled with the anti-cTF mAb #133-2 or isotype control (both at 20 g/mL), followed by a secondary A488-conjugated donkey anti-mouse IgG (1:200, green), with DAPI nuclear counterstaining (blue). Bar = 10 m. More cells show membrane staining for TF with mAb #133-2 after stimulation with LPS than the vehicle control (arrows), compatible with LPS-stimulated induction of TF expression. No staining was seen with the negative control (For interpretation of the color information in this figure legend, the reader is referred to the web version of the article.).
Finally, we tested if the mAb could inhibit the generation of activated factor X by high TF-expressing CMT25 cancer cells. No inhibition of activity was observed with the mAb at concentrations between 10 and 200 g/mL (data not shown). In the study herein, we describe a combination of immunologicbased techniques to demonstrate the reactivity of a murine anti-cTF mAb that was raised against the cloned cTF ectodomain expressed in and purified from CHO cells. Our results indicate that the purified mAb shows expected reactivity in canine cancer cells and non-neoplastic cells and tissues, but unfortunately does not inhibit TF-based procoagulant activity in cancer cells. Nevertheless, this mAb may prove useful in characterizing cTF in canine tissue and cells, including in situ expression in various canine neoplasms, and potentially for developing assays to measure TF Ag, in the form of microparticles, in whole blood or plasma. Conflict of interest The authors have no conflicts of interest. Acknowledgement This study was funded by a Morris Animal Foundation (grant #D07CA-047). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.vetimm.2015.07. 001 References Andreasen, E.B., Nielsen, O.L., Tranholm, M., Knudsen, T., Kristensen, A.T., 2014. Expression of tissue factor in canine mammary tumours and correlation with
grade, stage and markers of haemostasis and inflammation. Vet. Comp. Oncol., Epub ahead of print. Antoniak, S., Mackman, N., 2014. Multiple roles of the coagulation protease cascade during virus infection. Blood 123, 2605–2613. Armstrong, M.T., Rickles, F.R., Armstrong, P.B., 2013. Capture of lipopolysaccharide (endotoxin) by the blood clot: a comparative study. PLoS One 8, e80192. Artimo, P., Jonnalagedda, M., Arnold, K., Baratin, D., Csardi, G., de Castro, E., Duvaud, S., Flegel, V., Fortier, A., Gasteiger, E., Grosdidier, A., Hernandez, C., Ioannidis, V., Kuznetsov, D., Liechti, R., Moretti, S., Mostaguir, K., Redaschi, N., Rossier, G., Xenarios, I., Stockinger, H., 2012. ExPASy: SIB bioinformatics resource portal. Nucleic Acids Res. 40, W597–W603. Camerer, E., Pringle, S., Skartlien, A.H., Wiiger, M., Prydz, K., Kolsto, A.B., Prydz, H., 1996. Opposite sorting of tissue factor in human umbilical vein endothelial cells and Madin-Darby canine kidney epithelial cells. Blood 88, 1339–1349. de la Fuente, C., Pumarola, M., Blasco, E., Fernandez, F., Viu, J., Anor, S., 2014. Immunohistochemical evaluation of tissue factor, fibrin/fibrinogen and D-dimers in canine gliomas. Vet. J. 200, 387–392. Drake, T.A., Morrissey, J.H., Edgington, T.S., 1989. Selective cellular expression of tissue factor in human tissues. Implications for disorders of hemostasis and thrombosis. Am. J. Pathol. 134, 1087–1097. Egorina, E.M., Sovershaev, M.A., Bjorkoy, G., Gruber, F.X., Olsen, J.O., Parhami-Seren, B., Mann, K.G., Osterud, B., 2005. Intracellular and surface distribution of monocyte tissue factor: application to intersubject variability. Arterio. Vasc. Biol. 25, 1493–1498. Flossel, C., Luther, T., Muller, M., Albrecht, S., Kasper, M., 1994. Immunohistochemical detection of tissue factor (TF) on paraffin sections of routinely fixed human tissue. Histochemistry 101, 449–453. Geddings, J.E., Mackman, N., 2013. Tumor-derived tissue factor-positive microparticles and venous thrombosis in cancer patients. Blood 122, 1873–1880. Hellum, M., Ovstebo, R., Brusletto, B.S., Berg, J.P., Brandtzaeg, P., Henriksson, C.E., 2014. Microparticle-associated tissue factor activity correlates with plasma levels of bacterial lipopolysaccharides in meningococcal septic shock. Thromb. Res. 133, 507–514. Key, N.S., Mackman, N., 2010. Tissue factor and its measurement in whole blood, plasma, and microparticles. Semin. Thromb. Hemost. 36, 865–875. Lee, R.D., Barcel, D.A., Williams, J.C., Wang, J.G., Boles, J.C., Manly, D.A., Key, N.S., Mackman, N., 2012. Pre-analytical and analytical variables affecting the measurement of plasma-derived microparticle tissue factor activity. Thromb. Res. 29, 80–85. Luther, T., Flossel, C., Mackman, N., Bierhaus, A., Kasper, M., Albrecht, S., Sage, E.H., Iruela-Arispe, L., Grossmann, H., Strohlein, A., Zhang, Y., Nawroth, P.P., Carmeliet, P., Loskutoff, D.J., Muller, M., 1996. Tissue factor expression during human and mouse development. Am. J. Pathol. 149, 101–113. Mackman, N., 2009. The role of tissue factor and factor VIIa in hemostasis. Anesth. Analg. 108, 1447–1452.
Please cite this article in press as: Stokol, T., et al., Generation and characterization of a monoclonal antibody against canine tissue factor. Vet. Immunol. Immunopathol. (2015), http://dx.doi.org/10.1016/j.vetimm.2015.07.001
G Model VETIMM-9366; No. of Pages 7
ARTICLE IN PRESS T. Stokol et al. / Veterinary Immunology and Immunopathology xxx (2015) xxx–xxx
Morrissey, J.H., Fair, D.S., Edgington, T.S., 1988. Monoclonal antibody analysis of purified and cell-associated tissue factor. Thromb. Res. 52, 247–261. Ogasawara, S., Daddona, J., Trimpert, J., Stokol, T., 2012. The effect of recombinant canine interleukin-6 and interleukin-8 on tissue factor procoagulant activity on canine peripheral blood mononuclear cells and purified canine monocytes. Vet. Clin. Pathol. 41, 325–335. Pawlinski, R., Mackman, N., 2010. Cellular sources of tissue factor in endotoxemia and sepsis. Thromb. Res. 125 (Suppl. 1), S70–S73. Rehemtulla, A., Ruf, W., Edgington, T.S., 1991. The integrity of the cysteine 186-cysteine 209 bond of the second disulfide loop of tissue factor is required for binding of factor VII. J. Biol. Chem. 266, 10294–10299. Stokol, T., Daddona, J.L., Mubayed, L.S., Trimpert, J., Kang, S., 2011. Evaluation of tissue factor expression in canine tumor cells. Am. J. Vet. Res. 72, 1097–1106. Tatsumi, K., Antoniak, S., Monroe 3rd, D.M., Khorana, A.A., Mackman, N., 2014. Subcommittee on H. malignancy evaluation of a new commercial assay to measure microparticle tissue factor activity in plasma: communication from the SSC of the ISTH. J. Thromb. Haemost. 12, 1932–1934.
7
Wagner, B., Hillegas, J.M., Babasyan, S., 2012. Monoclonal antibodies to equine CD23 identify the low-affinity receptor for IgE on subpopulations of IgM+ and IgG1+ B-cells in horses. Vet. Immunol. Immunopathol. 146, 125–134. Wagner, B., Radbruch, A., Rohwer, J., Leibold, W., 2003. Monoclonal anti-equine IgE antibodies with specificity for different epitopes on the immunoglobulin heavy chain of native IgE. Vet. Immunol. Immunopathol. 92, 45–60. Wagner, B., Robeson, J., McCracken, M., Wattrang, E., Antczak, D.F., 2005. Horse cytokine/IgG fusion proteins—mammalian expression of biologically active cytokines and a system to verify antibody specificity to equine cytokines. Vet. Immunol. Immunopathol. 105, 1–14. Wang, J.G., Geddings, J.E., Aleman, M.M., Cardenas, J.C., Chantrathammachart, P., Williams, J.C., Kirchhofer, D., Bogdanov, V.Y., Bach, R.R., Rak, J., Church, F.C., Wolberg, A.S., Pawlinski, R., Key, N.S., Yeh, J.J., Mackman, N., 2012. Tumor-derived tissue factor activates coagulation and enhances thrombosis in a mouse xenograft model of human pancreatic cancer. Blood 119, 5543–5552. Wang, J.G., Manly, D., Kirchhofer, D., Pawlinski, R., Mackman, N., 2009. Levels of microparticle tissue factor activity correlate with coagulation activation in endotoxemic mice. J. Thromb. Haemost. 7, 1092–1098.
Please cite this article in press as: Stokol, T., et al., Generation and characterization of a monoclonal antibody against canine tissue factor. Vet. Immunol. Immunopathol. (2015), http://dx.doi.org/10.1016/j.vetimm.2015.07.001