Development of a Cross-Reactive Monoclonal Antibody for Detecting

Subscriber access provided by Linköping University Library

Article

Development of a Cross-Reactive Monoclonal Antibody for Detecting the Tumor Stroma Hallie M. Hintz, Aidan Cowan, Mariya Shapovalova, and Aaron LeBeau Bioconjugate Chem., Just Accepted Manuscript • DOI: 10.1021/acs.bioconjchem.9b00206 • Publication Date (Web): 10 Apr 2019 Downloaded from http://pubs.acs.org on April 11, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 37 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Bioconjugate Chemistry

Development of a Cross-Reactive Monoclonal Antibody for Detecting the Tumor Stroma Hallie M. Hintz†, Aidan Cowan†, Mariya Shapovalova†, and Aaron M. LeBeau*† †Department

of Pharmacology, University of Minnesota Medical School, Minneapolis, Minnesota, 55455

*Corresponding

author: Aaron M. LeBeau, Nils Hasselmo Hall Room 3-104, 312 Church Street SE, Minneapolis, MN, 55455. Phone: (612)-301-7231. Email: [email protected]

ACS Paragon Plus Environment

1

Bioconjugate Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 37

ABSTRACT Here we document the discovery of a monoclonal antibody that selectively binds to both human and murine fibroblast activation protein alpha (FAP), a serine protease that is over-expressed on cancerassociated fibroblasts (CAFs) making it an attractive therapeutic target for the aiding and abetting tumor microenvironment. The lead antibody, B12, was identified from a naïve murine single-chain variable fragment antibody phage display library screened against recombinant human FAP on magnetic beads. The heavy and light chains of B12 were cloned into full-length human immunoglobulin 1 (IgG) vectors and expressed as a chimeric monoclonal antibody (B12 IgG). We engineered a drug-resistant prostate cancer cell line, CWR-R1-EnzR, to express human FAP for antibody characterization and validation (R1EnzRFAP). B12 IgG selectively bound to the R1-EnzRFAP cells by flow cytometry and was internalized in vitro by confocal microscopy. B12 IgG was further evaluated as a near-infrared (NIR) optical imaging probe in R1-EnzRFAP and parental xenograft models. High tumor uptake and retention of the NIR probe was observed in the R1-EnzRFAP xenografts and endogenous expression of murine stromal origin FAP was detected in the parental xenografts. Ex vivo evaluation of these models by immunohistochemistry documented B12 IgG localization to both human and murine FAP-expressing cells.


2

Page 3 of 37 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


INTRODUCTION Solid tumors consist of malignant cells and a heterogeneous mixture of supporting stromal cells that are essential for tumor growth beyond a few millimeters1. Stromal transformation in the tumor microenvironment occurs during the early stages of carcinogenesis and has been likened to processes in normal wound healing2. Tumor growth is associated with adaptations in the supporting stroma, including neoangiogenesis; the recruitment of fibroblasts and immune cells; the release of peptide-signaling molecules and proteases; and the extensive remodeling of the extracellular matrix3-7. High stromal composition is a characteristic of many solid tumors, including prostate, breast, colon, and lung, with tumor-stroma ratios ranging from 20% to ≥ 50% of the tumor mass8. Even higher ratios, >90%, are seen in carcinomas with desmoplastic reactions. Cancer-associated fibroblasts (CAFs) are the major cell type in the stromal compartment and play a significant role in tumorigenesis and invasion. The exact origin of these transformed reactive fibroblasts is still unknown, but several cell types are suspected including fibroblasts, adipocytes, cancer stem cells, bone marrow mesenchymal stem cells, and endothelial cells that have undergone endothelial mesenchymal transition9, 10. Malignant cells directly contribute to the formation of the reactive stroma through mechanical stress and secretion of TGF-β, growth factors, and chemokines11, 12. This cross talk between cancer cells and CAFs creates a paracrine feedback loop that stimulates tumor growth, promotes inflammation and immunosuppression, and drives tumor invasion13-16. Furthermore, CAFs are involved in several mechanisms of therapeutic resistance and immune evasion which augments progression and metastasis of the disease17, 18. CAFs are a heterogeneous population and expression of cellular markers varies based on cell origin and tumor type. The membrane-bound serine protease fibroblast activation protein alpha (FAP) is highly expressed on CAFs in 90% of epithelial tumors and its expression is associated with aggressive


3


Page 4 of 37

phenotypes. FAP expression has also been detected in some soft tissue and bone sarcomas19. The protease was independently identified by two research groups in the 1990s and was found to have dipeptidyl peptidase, gelatinase, and collagenase activity20, 21. FAP activity contributes to tumorigenesis through extracellular matrix degradation and growth factor activation and FAP-expressing cells have been shown to mediate anti-tumor immunity in the tumor microenvironment22-24. In nonmalignant tissue, FAP is expressed by reactive fibroblasts in wound healing, and in diseased tissue such as rheumatoid arthritis, lung fibrosis, and liver cirrhosis19. Overall, the localization of FAP expression in the tumor microenvironment makes it an ideal candidate for therapeutic targeting and molecular imaging. In parallel to humans, murine FAP, which shares 89% sequence identity with the human enzyme, is expressed by murine reactive stroma within human cancer xenografts25. Preclinical studies show that targeting FAP in murine stroma leads to necrosis of the tumor in multiple cancer xenograft models23, 26-28. To date, no in vitro CAF cell line model exists. However, various in vivo models of the tumor stroma have been used for evaluation of diagnostic and therapeutic agents including cancer cell lines engineered to express human or murine FAP or inoculation of primary fibroblasts with xenografts. Limitations of these models include inaccurate distribution of the CAFs, artificial FAP expression, and modification of primary cells when cultured in vitro29, 30. Studies using murine models that mimic human tumor stroma formation and function are required to assess CAF-targeted therapies. The first generation anti-FAP monoclonal antibody, F19, was derived from a mouse immunized with lung fibroblasts. F19 was able to differentiate between malignant and benign tissues in vitro and was shown to bind to FAP-expressing stroma in humans31. Based on these results, a humanized version of F19, sibrotuzumab, was evaluated in phase I/II clinical trials for anti-tumor activity against metastatic colorectal cancer and non-small cell lung cancer. Although the humanized antibody demonstrated favorable tumor stroma targeting, it failed to demonstrate any clinical efficacy32. Furthermore, 8 of 26


4

Page 5 of 37 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


patients treated with sibrotuzumab developed human-anti-human antibodies, indicating an immune response to the humanized therapeutic33. As a result, sibrotuzumab was withdrawn from clinical trials. In this study, we describe the characterization and validation of a chimeric monoclonal antibody specific for FAP identified from a naïve murine single-chain variable fragment (scFv) antibody phage display library. The antibody, B12, was shown to detect FAP expression in cell lines by flow cytometry and was rapidly internalized by FAP-expressing cells in vitro. B12 demonstrated cross-reactivity with murine FAP, but not with the highly homologous protease human dipeptidyl peptidase IV. Further testing in vivo was performed to evaluate the ability of B12 conjugated to a near-infrared dye to non-invasively image preclinical models of prostate cancer. High tumor uptake and retention in FAP-expressing xenografts was observed, as well as, detection of endogenous FAP expressed by murine-origin cells. As additional proof of our model, immunohistochemistry (IHC) staining was used to document B12 penetration ex vivo in xenograft tissues.


5


Page 6 of 37

RESULTS Selection and characterization of anti-FAP scFv antibody fragments A naïve murine scFv antibody phage display library with a diversity of 1.9 x 109 was used to identify antibodies specific for human FAP (hFAP). Biopanning was performed on biotinylated recombinant hFAP attached to streptavidin-coated magnetic beads to conserve native protein structure. After four rounds of biopanning, 384 unpurified scFvs in culture supernatant were screened against biotinylated recombinant hFAP by enzyme-linked immunosorbent assay (ELISA) (Figure 1). Of these scFvs, 35 demonstrated hFAP binding as indicated by high ELISA signal (Figure 1A). The 35 unpurified scFvs in culture supernatant were serially diluted and analyzed for concentration-dependent binding. 21 clones were found to bind hFAP with a saturating ELISA signal which suggests high affinity for a single epitope (Figure 1B). Of the 21 clones, twelve scFvs had unique sequences. Finally, we identified scFv clones selective for hFAP by counter screening ELISA. The twelve unique scFvs in culture supernatant were screened against biotinylated human dipeptidyl peptidase IV (hDPP-IV), a serine protease which shares 52% homology with hFAP. The threshold for binding signal was set at 0.1 and four scFvs showed no cross-reactivity with hDPP-IV (Figure 1C). The FAP-selective clones were then expressed and purified for further testing. Since no in vitro cell line model for CAFs exists, we engineered a hFAP-expressing cell line to use for antibody characterization and validation (Supporting Figure S1). The human prostate cancer cell line CWR-R1-Enzalutamide Resistant/luciferase+ was lentivirally transduced to express hFAP (R1-EnzRFAP). Expression of hFAP in the engineered cell line was confirmed by quantitative RT-PCR (Supporting Figure S1A) and western blot (Supporting Figure S1B). Flow cytometry with a PE conjugated commercial antibody was used to detect membrane expression of hFAP (Supporting Figure S1C). No hFAP mRNA, protein, or membrane expression was observed in the parental R1-EnzR cells (Supporting Figure S1). Finally, the membrane-bound hFAP was found to be enzymatically active when tested by a whole cell-


6

Page 7 of 37 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 1. Identification of the B12 scFv from a murine antibody phage display library. (A) Identification of scFv clones with hFAP specificity by ELISA. 384 unpurified scFv clones in culture supernatant were added to hFAP protein coated plates and scFv binding was detected using a peroxidase conjugated anti-HA-tag monoclonal antibody and Turbo TMB reagent. 35 clones with a high ELISA signal were identified. (B) Identification of scFv clones with high affinity binding by ELISA. 35 unpurified scFv clones in culture supernatant were serially diluted from 1- 0.05 (relative concentration) and added to protein coated plates. 21 clones demonstrated saturating ELISA signals. (C) Identification of scFv clones with hFAP selectivity by ELISA. 12 scFv clones with unique sequences in culture supernatant were added to hDPP-IV protein coated plates. The threshold for binding signal was set at 0.1 (dashed line) and 4 scFv clones with a low ELISA signal were identified. (D) Identification of scFv clones with selectivity for membrane-bound hFAP by flow cytometry. The 4 scFv clones were expressed, purified, and labeled with Alexa Fluor 488. R1-EnzRFAP (orange line) and R1-EnzR (red line) cell lines were stained with 500 nM of scFv for one h and analyzed by flow cytometry. Unstained R1-EnzRFAP cells (blue line) were used as a negative control. B12 was the only scFv clone with selectivity for FAP-expressing cells but not the parental FAP negative cells.


7


Page 8 of 37

associated proteolysis assay using the prolyl peptidase-specific fluorogenic substrate Z-Gly-Pro-AMC (Supporting Figure S1D). The four expressed scFv clones were next investigated for the ability to selectively bind hFAPexpressing cells. The scFv clones were conjugated to Alexa Fluor 488 and tested in parallel by flow cytometry. R1-EnzRFAP and R1-EnzR cell lines were stained with each scFv at a concentration of 500 nM for one h and compared to unstained controls. It was determined that only scFv clone B12 bound to the transduced R1-EnzRFAP, but not to the parental R1-EnzR cells (Figure 1D, Supporting Figure S2). Based on these cumulative data, clone B12 was selected as the lead scFv for development as an anti-FAP IgG1 chimeric monoclonal antibody.

Generation and in vitro characterization of anti-FAP IgG1 chimeric monoclonal antibody The heavy and light chains of the B12 scFv were cloned into full-length human immunoglobulin 1 (B12 IgG) vectors for expression in HEK293T cells. After purification, the selectivity and affinity of B12 IgG were determined in vitro (Figure 2). By ELISA, B12 IgG was found to cross-react with both hFAP and murine FAP (mFAP) – a characteristic essential for validating B12 IgG in endogenous FAPexpressing murine models (Figure 2A). The concentration-dependent binding and saturation of the ELISA signal suggests that B12 IgG bound to a single unique and conserved epitope on both hFAP and mFAP. In addition, no B12 IgG binding was observed with the homologous prolyl protease hDPP-IV (Figure 2A). Surface plasmon resonance was next used to determine the bivalent affinity of B12 IgG for recombinant FAP protein. The KD of B12 IgG for hFAP and mFAP were calculated to be roughly similar with values of 3.39 nM and 1.86 nM respectively (Figure 2B). When active recombinant hFAP was pretreated with B12 IgG for 30 min before the addition of a Z-Gly-Pro-AMC substrate, no inhibition of hFAP enzymatic


8

Page 9 of 37 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 2. in vitro characterization of B12 IgG. (A) B12 IgG specificity for hFAP (circle), mFAP (square), and the homologous protease hDPP-IV (triangle) was determined by ELISA. B12 IgG was serially diluted from 500 nM to 0.1 nM and added to protein coated wells. B12 IgG binding was detected by a peroxidase conjugated human IgG monoclonal antibody and Turbo TMB reagent. (B) Surface plasmon resonance was used to calculate the binding affinity of B12 IgG for hFAP (KD=3.39 nM) and mFAP (KD=1.86 nM). (C) B12 IgG selectivity for membrane-bound hFAP was analyzed by flow cytometry. R1-EnzRFAP (orange line) and R1-EnzR (red line) cells were stained with Alexa Fluor 488 conjugated B12 IgG at a concentration of 100 nM for one h. Unstained R1-EnzRFAP cells (blue line) were used as a negative control. (D) mRNA levels of hDPP-IV were investigated by quantitative RT-PCR in prostate (R1-EnzR, R1-EnzRFAP, PC3) and colon (Caco2) cancer cells. mRNA fold change is normalized to R1-EnzR. **p

Development of a Cross-Reactive Monoclonal Antibody for Detecting

Recommend Documents