A monoclonal antibody against neem leaf glycoprotein recognizes carcinoembryonic antigen (CEA) and restricts CEA expressing tumor growth.

BASIC STUDY

A Monoclonal Antibody Against Neem Leaf Glycoprotein Recognizes Carcinoembryonic Antigen (CEA) and Restricts CEA Expressing Tumor Growth Arnab Das,*w Subhasis Barik,w Saptak Banerjee,w Anamika Bose,z Koustav Sarkar,y Jaydip Biswas,8 Rathindranath Baral,w and Smarajit Pal*

Summary: Carcinoembryonic antigen (CEA) is one of the promising tumor antigens mainly associated with carcinoma of the colon, lung, breast, etc. and received wide attention for cancer immunotherapy. Neem leaf glycoprotein (NLGP), an effective immunomodulator, is able to generate humoral and cellular immune responses in murine tumor models. We have generated a monoclonal antibody (mAb) against NLGP by fusing NLGP-immunized mice splenocytes with nonsecretory myeloma cells. A highly antiNLGP mAb secreting clone (1C8; IgG2a in nature) has been identified and propagated in culture. 1C8 recognizes human CEA as good as NLGP by enzyme linked immunosorbent assay, Western blotting, and immunoprecipitation. 1C8 detects CEA on colon cancer tissues by immunochistochemistry. By flow cytometry, 1C8 specifically reacts with CEA + human (Colo-205, HCT-116, and HT-29) and mouse (CT-26) colon cancer cells, but it showed minimum reactivity with CEA human (MCF7, SiHa, and SCC084) and mouse (B16MelF10) cancer cells. This anti-NLGP 1C8 mAb revealed significant antitumor activity and better survivability in vivo in animals bearing mouse (CT-26 in BALB/c) and human (Colo-205 in athymic nude) CEA + cancer cells. 1C8 has no direct influence on proliferation and migration of CEA + cells, however, NK cell–dependent strong antibody-dependent cellular cytotoxicity reaction toward CEA + cells and normalization of angiogenesis are chiefly associated with tumor growth restriction. Obtained results provided a new immunotherapeutic approach for the effective management of CEA + tumors. Key Words: colon cancer, carcinoembryonic antigen, monoclonal antibody, neem leaf glycoprotein, antibody-dependent cellular cytotoxicity

(J Immunother 2014;37:394–406)

C

arcinoembryonic antigen (CEA) was one of the first tumor antigens, isolated from colon cancer patients.1 It is a 180 kDa glycoprotein consisting of B60% carbohydrate and variation on it exhibits considerable heterogeneity.2 It also acts as an adhesion molecule and may play a role in metastasis.3,4 The generation of monoclonal antibodies (mAbs) against CEA allowed the detection of the Received for publication March 28, 2014; accepted May 29, 2014. From the *Clinical Biochemistry Unit; Departments of wImmunoregulation and Immunodiagnostics; 8Surgical Oncology and Medical Oncology, Chittaranjan National Cancer Institute; zDepartment of Molecular Medicine, Bose Institute, Kolkata, India; and yPediatric Hematology Oncology, University of Iowa Children’s Hospital, IA. Reprints: Smarajit Pal, Clinical Biochemistry Unit, Chittaranjan National Cancer Institute (CNCI), 37, S. P. Mukherjee Road, Kolkata 700026, India (e-mail: [email protected]). Supplemental Digital Content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Website, www.immunotherapy-journal.com). Copyright r 2014 by Lippincott Williams & Wilkins

394 | www.immunotherapy-journal.com

overexpressed CEA protein in a variety of adenocarcinomas, including gastric, colonic, rectal, ovarian, breast, etc.5–7 As a self oncofetal protein, CEA is weakly immunogenic in cancer patients, although, it generates both antibody8 and T-cell9 responses. Exploiting such knowledge several vaccine strategies have been formulated and tested, including generation of anti-CEA antibodies,10,11 anti-idiotypic antibodies mimicking CEA,12–14 use of CEA peptides and recombinant protein,15–18,19 recombinant microorganisms expressing CEA,20 CEA-pulsed dendritic cells,21–23 etc.24–26 to increase vaccine effectiveness. These strategies use CEA as a target molecule in various forms of active specific and passive immunotherapy. In case of CEA expressing tumors, CEA appeared as a self-antigen by the immune system, thus, patients with CEA + tumors are typically immunologically “tolerant” to CEA.12 Passive approach that involves infusion of huge amount of anti-CEA antibody is often unsuccessful as it generates neutralizing antibodies.27 Thus, alternate approach to target CEA is still in search. Despite the existing lacunae in the understanding, relationship of the CEA and plant materials is explored. Stoger et al28 successfully expressed a single-chain Fv antibody (ScFvT84.66) against CEA in the staple cereal crops—rice and wheat. Transient expression of a tumorspecific single-chain fragment and a chimeric antibody against human CEA was reported in tobacco leaves.29 Production of antibodies in plants against human CEA was also reviewed.30 In our effort to elucidate the adjuvant function of a preparation from neem (Azadirachta indica) leaves [neem leaf preparation (NLP)] in enhancement of antibody response against breast tumor–associated antigen,31 melanoma surface antigen,32 and CEA,33,34 we surprisingly found that the serum polyclonal antibody against this NLP reacts specifically with human CEA with minimum reactivity with other antigens.35 As, clinical significance of the CEA recognition by anti-NLP antibody was not explored, here, we attempted to generate and characterize a mAb against neem leaf glycoprotein (NLGP). Successful generation of mAb eventually showed its recognition to human CEA that have an enormous significance in the induction of anti-CEA antitumor immunity. In association with the in vivo testing of antiNLGP mAb for restriction of CEA + tumor growth, mechanism of antitumor action is also partially explored.

MATERIALS AND METHODS Reagents and Media Human CEA was procured from Aspen Bio Inc., CO. Anti-human CEA antibody, anti-mouse peroxidise-labeled IgG, CEA peptide (101-115), fluorescent stain DAPI, MTT, J Immunother

Volume 37, Number 8, October 2014

J Immunother


and pristane were procured from Sigma, St Louis and goat anti-mouse antibody isotypes/subclasses labeled with alkaline phosphatase were purchased from Southern Biotechnology, Birmingham, AL. Different tumor antigens (B16MelSAg, ECAg, and S180Ag) were prepared by the methods described.31,36 Flow cytometry antibodies (DX5FITC, CD11b-PE and CD16/32-FITC, anti-mouse IgGFITC) were procured from e-Biosciences, San Diego. Purified anti-mouse CD31 and VEGF/VEGFR2 were procured from Biolegend (San Diego, CA) and Santacruz Biotech (California), respectively. Trireagent and carboxyfluorescein succinimidyl ester (CFSE) were obtained from Invitrogen (CA). RT-PCR primers were procured from MWG-Biotech AG (Bangalore, India). TMB substrate solution (OptEIA) was purchased from BD-Pharmingen, San Diego. LDH release kit was procured from Roche Diagnostics (Mannham, Germany). AEC chromogen solution was purchased from VECTOR laboratories Inc. (Burlingame, CA). Different culture media, including HAT/HT media (50) and FBS, penicillin-streptomycin, L-glutamine (2 mM), HEPES, and sodium pyruvate were procured from GIBCO-Life Technologies, NY. Chemiluminescence detection kit was purchased from Pierce (Rockford, IL). BSA, EDTA, Protein G, DMSO, poly-L-lysine, Tween-20, TritonX-100, KLH, Freund’s adjuvant, p-nitrophenylphosphate (p-NPP), bmercaptoethanol, neuraminidase, etc. were procured from Sigma. Polyethylene glycol (MW-4000) was procured from Merck, Germany.

NLGP Extract from neem (A. indica) leaves was prepared by the method described.31,37 Briefly, pulverized leaf powder was soaked overnight in PBS, pH 7.4; supernatant was collected by centrifugation at 1500 rpm. NLP was then extensively dialyzed against PBS and concentrated by Centricon Membrane Filter (Millipore Corporation, Bedford, MA) with 10 kDa molecular weight cut off. Biochemical assessment suggests the presence 33% carbohydrate moiety and rest is protein (Fig. 1A1), hence, termed, NLGP.31,38 Nondenaturating polyacrylamide gel electrophoresis (PAGE) (Fig. 1A2) and high-performance liquid chromatography analysis (Fig. 1A4) revealed the presence of a single protein. However, in denaturating SDSPAGE, 3 subunits were appeared (Fig. 1A3).

Animals, Cell Lines, and Tumor Female BALB/c mice (4 wk age; 25 g average body weight) were obtained from Indian Institute of Chemical Biology, Kolkata and maintained under standard laboratory conditions. Autoclaved dry pellet diet (Epic Laboratory Animal Feed, Kalyani, India) and water were given ad libitum. Athymic nude (4–6 wk old) mice were purchased from National Centre for Laboratory Animal Sciences (NCLAS), NIN, Hyderabad and maintained in a specific pathogen-free facility. A nonsecretory P3 63Ag8.653 myeloma cell was obtained from National Center for Cell Sciences (NCCS), Pune and was maintained in RPMI1640 + 10% FBS with penicillin (100 U/mL), streptomycin (100 mg/mL), L-glutamine (2 mM), and HEPES (100 mM) for cell fusion and hybridoma. Three CEA expressing human colon cancer cell lines, like Colo-205 (RPMI-1640), HCT-116, and HT-29 (McCoy’s 5a) were procured from NCCS and maintained in their respective media as indicated with FBS, penicillin (100 U/mL), streptomycin (100 mg/mL), and L-glutamine (2 mM). In these experiments, three CEA /low cell lines, MCF7 (breast cancer), r

2014 Lippincott Williams & Wilkins

mAb Against NLGP Recognizes CEA

SiHa (cervical cancer), SCC084 (squamous cell cancer) were maintained in complete MEM and DMEM and used as control. All cell lines were maintained at 371C with the supply of 5% CO2 in humidified incubator. A CEA + murine colon carcinoma (CT-26) and a murine melanoma (B16MelF10) cell lines were maintained in complete RPMI1640 and DMEM, respectively. To develop in vivo solid tumors, BALB/c and athymic nude mice were inoculated subcutaneously (SC) in right hind leg quarters with CT-26 (2 106) and Colo-205 (2 106) cells, respectively. The Institutional Animal Ethics Committee approved the experimental design. Paraffinised sections of human colon tumor samples were collected from pathology archives of CNCI, Kolkata.

Generation and Purification of mAb Syngenic female BALB/c mice (n = 6 for each group) were immunized with NLGP (25 mg) coupled with KLH (1:1), along with incomplete Freund’s adjuvant (1:1) SC once a week for 4 weeks. Two days before fusion, mice were immunized with purified NLGP (25 mg) alone IV at mice tail vein. Mice were bled after each weekly immunization and sera were separated to check for anti-NLGP antibody titer by enzyme linked immunosorbent assay (ELISA).35 Before fusion, a murine myeloma P3 63Ag8.653 cells were maintained till they reach to log phase and spleen cells from immunized mice were fused with myeloma cells at a ratio of 3:1 in the presence of PEG 4000.39 Fused cells were cultured in complete RPMI 1640, supplemented with HAT (1 ), hybridoma cloning factor (10%), and b-ME (50 mM) and plated in 96-well microtiter plates as 8 104 original myeloma cells in each well. Plates were maintained at 371C in humidified incubator having 5% CO2. After 8–10 days postfusion, the colonies (1–2 mm in diameter) were macroscopically visible. Culture supernatants were screened for antibodies against NLGP by ELISA as described previously. The positive clones of mAb secreted against NLGP were recloned twice by limiting dilution to get purified antiNLGP mAb secreting hybridoma cells. Viable hybridoma cells (5106) were then grown in peritoneum of pristine primed mice as ascites for bulk antibody production. Two weeks postinoculation, mice ascites were collected and purified using Protein G as per manufacturer’s protocol. The protein concentration was estimated using Folin-phenol reagent.40

SDS-PAGE and Western Blotting Purified NLGP was electrophoresed on 12.5% SDSPAGE, transferred for Western blotting as described41 and incubated with anti-NLGP mAb. After washing, blots were incubated with HRP-conjugated secondary antibody for 2 hours at room temperature. Bands were detected using Western lighting chemiluminescence detection kit according to the manufacturer’s manual. In the first set of experiments, human and mouse CEA + and CEA cells were lysed by RIPA buffer. Cell lysates were blotted by using both anti-NLGP and anti-CEA mAbs. Band intensity was measured by densitometric scanning.

ELISA for CEA Specificity Generated anti-NLGP mAb was checked for its specificity towards different concentrations of NLGP (0.1–5 mg) by ELISA,35 as shown in Fig. S1 (Supplemental Digital Content 1, http://links.lww.com/JIT/A348). To check the CEA specificity, CEA whole molecule, CEA peptide www.immunotherapy-journal.com |

395

Das et al

J Immunother


FIGURE 1. Neem leaf glycoprotein (NLGP) and characterization of anti-NLGP monoclonal antibody (mAb). A, Carbohydrate and protein composition of NLGP is presented (A1). NLGP (25 mg) was analyzed by nondenatured (A2) and denatured (A3) PAGE after silver nitrate staining and by high-performance liquid chromatography (A4). B, Three groups of female syngenic BALB/c mice (n = 6) were immunized SC weekly for 4 weeks in total with NLGP, NLGP + KLH + Adj, and PBS as control to check the anti-NLGP antibody titer by ELISA. Bar diagram represents mean OD values at 450 nm ± SD of 5 individual observations. ***P < 0.001. C, NLGP reactive clones (n = 12) were screened, those were generated by fusing NLGP-immunized mice splenocytes with nonsecretory myeloma cells using hybridoma technique and 1C8 was identified as highly secreting clone. D, Purified NLGP was run on SDS-PAGE and Western blotting was performed using generated mAb (1C8) taking BSA as a negative control. E, Isotype profile of 1C8 (IgG, IgM, and IgA) was assessed. **P < 0.01. F, Subclass profile (IgG1, IgG2a, IgG2b, and IgG3) of 1C8 (IgG) was determined. **P < 0.01. In both (E and F), bar diagram represents mean ± SD of 3 individual observations.

fragment (Thr-Gly-Gln-Phe-Arg-Val-Tyr-Pro-Glu-Leu-ProLys-Pro-Ser-Ile) were used and detected by anti-NLGP mAb by ELISA, keeping anti-CEA antibody as positive control. CEA was also treated with neuraminidase before its reaction with both anti-NLGP and anti-CEA mAb. Briefly, 100 mL of the CEA solution (5 mg/mL) was incubated with neuraminidase and incubated at 371C for 3 hours. The specificity was also checked by reacting CEA + , CEA tumor cell lysates with anti-NLGP mAb. Isotypes (IgG, IgM, and IgA) and its subclasses (if it is IgG) were determined by ELISA. A panel of alkaline phosphatase-conjugated secondary antibodies, for example, anti-mouse IgG, IgM, IgA, IgG1, IgG2a, IgG2b, and IgG3 were used for this purpose. Generated anti-NLGP mAb was also confirmed for its CEA specificity by its reaction with a panel of antigens,


consisting of: NLGP, CEA, GD2, ECAg, S180Ag, B16MelSAg, and some plant derived components, like, quercetin dehydrate, etc. by ELISA. To determine the reactivity of anti-NLGP mAb with cell unbound CEA, some in vivo and in vitro experiments were designed. Some human cell lines (CEA + Colo-205, HCT-116 and HT-29 and CEA /low MCF7, SiHa and SCC084) and murine cell lines (CT-26 and B16MelF10) were maintained under respective culture conditions for 7 days to collect culture supernatants, and specific reaction of anti-NLGP mAb with secretory CEA was determined and compared with antiCEA mAb by ELISA. In in-vivo experiments, 1 group (n = 6) of CT-26 tumor bearing mice were treated with antiNLGP mAb once in a week for 4 weeks in total. Blood was collected weekly up to 4 weeks and sera were separated to determine the circulating CEA by using both anti-NLGP and r


J Immunother


anti-CEA mAbs. Normal BALB/c mice serum was taken as negative control. Blood from colon cancer patients and healthy individuals (n = 8, in each case) were also tested to determine the circulating CEA level by ELISA.

Immunoprecipitation Generated anti-NLGP and anti-CEA mAbs were incubated overnight with or without human CEA (0.5 mg/ 100 mL). Incubation mixture were assorted with Protein ASepharose 4B and further incubated for 60 minutes at 41C. Immunoprecipitated complex was then collected by centrifugation, washed thoroughly, denatured with sample buffer, centrifuged at 12000g, and electrophoresed on SDSPAGE. Following transfer of both immunoprecipitated complexes to nitrocellulose membrane, these were detected by anti-CEA mAb. Band intensity was measured by densitometric scanning.

Immunohistochemistry Sections (5 mm) from formalin-fixed paraffin-embedded colon cancer tissues (n = 10) and adjacent normal specimens were prepared as described41 using anti-NLGP and anti-CEA mAbs. Negative controls were included in


each staining step, where primary antibodies are replaced by mouse IgG. Murine CT-26 tumors and human colon cancer (Colo205) xenografted tumors from nude mice were harvested and snap-frozen in OCT compound to prepare sections (5 mm) using cryostat, air-dried, and fixed in ice-cold methanol for 20 minutes. After blocking with 5% BSA, cryosections were stained with both anti-NLGP and anti-CEA mAbs. Unrelated rat IgG was used as negative control. Here, murine B16MelF10 tumor tissue and human cervical cancer (SiHa) xenografted tumors from nude mice were taken as negative controls. These sections were also stained with antimouse VEGF and VEGFR2 antibodies using immunohistochemical method described earlier.41 For detection of the presence of CD31 + VECs, sections were incubated with rat anti-mouse CD31 followed by FITC conjugated anti-rat antibody. All sections were counterstained with DAPI and images were acquired using Leica DM 1000, Fluorescent Microscope (Leica, BM 4000B, Germany).

Flow Cytometry Various cancer cell lines including CEA + colon cancer cells of human and mouse origin; CEA cervical,

FIGURE 2. Recognition of carcinoembryonic antigen (CEA) by anti-neem leaf glycoprotein (NLGP) mAb. A, Generated anti-NLGP mAb was checked for its specificity to CEA by its reaction with a panel of antigens (NLGP, CEA, GD2, ECAg, S180Ag, B16MelSAg, and quercetin dehydrate) by coating them in 2 mg concentration by ELISA. Isotype-matched antibody was used as control. Bar diagram represents mean OD values ± SD of 5 individual observations. **P < 0.01. B, CEA was treated with neuraminidase and reacted with antiNLGP anti-CEA mAbs by ELISA, keeping CEA whole molecule as positive control. Bar diagram represents mean OD values ± SD of 3 individual observations. C, Purified CEA was electrophoresed in SDS-PAGE and detected by both anti-NLGP and anti-CEA mAbs in Western blotting (C1). Densitometric analysis (C2) of 3 individual observations was done and mean ± SD is presented. D, Different human CEA + human (Colo-205, HCT-116, and HT-29) and murine (CT-26) and CEA human (MCF7, SiHa, and SCC084) and murine (B16MelF10) cells were cultured to confluency and supernatants (1:50 dilution) were used to coat plate for adherence of secreted CEA and detected by ELISA using both anti-NLGP and anti-CEA mAbs. Bar diagram represents mean OD values ± SD of 5 individual observations. *P < 0.05; **P < 0.01. E, BALB/c mice (n = 6) were inoculated with CT-26 cells (2 106 cells/mice). In an interval of 7 days, serum from each immunized mice were used to detect secreting CEA by ELISA using both anti-NLGP and anti-CEA mAbs. Mean ± SD values of samples obtained from 6 individual mice in each time point is presented in bar diagram. An unrelated rat antibody was used as isotype control here. **P < 0.01. F, Circulating CEA in the serum of colon cancer patients was detected by ELISA using both of these mAbs. The bar diagram represents the mean OD values ± SD, obtained from 8 individual patients. **P < 0.01. G, A human CEA + (Colo205) and CEA (SiHa) and a murine CEA + (CT-26) and CEA (B16MelF10) cell lines were maintained in culture and cell lysates were prepared. CEA was then detected by both anti-CEA and anti-NLGP mAbs in ELISA. Bar diagram represents the mean OD values ± SD of 5 individual observations. **P < 0.01. H, CEA in CEA + tumor cell lysates (both human and murine) were detected by Western blotting using both anti-NLGP and anti-CEA mAbs. Representative band pattern (H1) and densitometric analysis of 3 individual observations are presented as mean ± SD (H2). r


www.immunotherapy-journal.com |

397

Das et al

J Immunother


FIGURE 3. Detection of carcinoembryonic antigen (CEA) by anti-neem leaf glycoprotein (NLGP) monoclonal antibody (mAb) using immunohistochemistry. A1, Different human colon adenocarcinoma samples (n = 10) were stained with anti-NLGP (1C8) and anti-CEA mAbs. Staining from 2 representative samples is presented. An unrelated rat antibody was used as isotype control here. Each section was stained by hematoxylin-eosin (H&E) method before immunostaining. A2, Number of CEA + cells as detected by both mAbs was presented in bar diagram with mean ± SD. A3, Adjacent normal colon tissues were reacted with same mAbs. B1, Solid subcutaneous murine CT-26 tumors were harvested, processed, and CEA expression was detected by both anti-NLGP and anti-CEA mAbs, with control as mentioned in (A1). B2, Human CEA + Colo-205 solid tumors developed in athymic nude mice (n = 6) were also stained with both the mAbs. An unrelated rat antibody was used as isotype control here. C, Number of murine and human CEA + cells as detected by both mAbs was presented in bar diagram (C1 and C2) with mean ± SD. D, Solid CEA mouse B16MelF10 tumors (D1) and human SiHa tumors (D2) developed in BALB/c and athymic nude mice, respectively, were also stained with these mAbs. Black arrows represent CEA expressing areas detected by both mAbs.


r


J Immunother


breast, and squamous cell carcinoma of human origin; and mouse melanoma were stained with anti-NLGP and antiCEA mAbs by the method described.36

In Vivo Growth Restriction of CEA + Murine and Human Colon Tumors Viable CEA + mouse (CT-26) and human (Colo-205) colon tumor cells (2 106 cells/100 mL PBS) were inoculated SC on the right hind leg quarter of 6-week-old female BALB/c and athymic nude mice, respectively. In both the tumor systems, after reaching the tumor volume of 24 mm3, a group of mice (n = 6) in each set was administered SC with anti-NLGP mAb (25 mg/mice/injection) on the left hind leg quarter once per week for 4 weeks in total. Other group of mice received unrelated rat IgG antibody as isotypic control. Tumor growth was monitored twice a week by caliper measurement and tumor volume was determined using the formula: tumor volume = (width)2 length/2. Survivability of 2 groups of mice was also monitored.

Antibody-dependent Cellular Cytotoxicity (ADCC)


Tumor Cell Migration Assay Cells were suspended in serum-free medium containing 1% BSA and then added to the upper chamber (10,000/ well) of transwells (8-mm pore size; Corning Costar, NY) in the presence of anti-NLGP, anti-CEA mAbs (5 mg/mL), or media alone. Lower chambers contained 20% FBS as chemoattractant. After incubation at 371C overnight, cells remained in the upper surface of membrane were removed to count, whereas the cells that migrated to the lower membrane surface were fixed with ethanol and stained with Giemsa solution. Cells migrating through the filter were counted and the number of migrating cells per optic field ( 20) was plotted.

Isolation of RNA and RT-PCR Analysis CT-26 tumors were collected from anti-NLGP mAb and isotype-matched control antibody-treated mice. Singlecell suspensions of solid tumors were obtained by enzymatic digestion, total RNA was isolated using the Trireagent. The cDNA synthesis was carried out using RevertAid First Strand cDNA Synthesis Kit (Fermentas, K1622) following the manufacturer’s protocol and PCR were carried out with gene-specific primers for VEGF, VEGFR2, CD31, and bactin genes used as described earlier.41 PCR products were identified by image analysis software for gel documentation (Gel Doc XR + system; BioRad) following electrophoresis on 2% agarose gels stained with ethidium bromide. Band intensity was measured by densitometric scanning.

For ADCC, CEA + , and CEA human cancer cells (1 104 cells) were cultured overnight in 96-well plates to use as target and those were incubated with normal human PBMC, as an effector in the presence of anti-NLGP and anti-CEA mAbs (5 mg/mL) for 4 hours. The LDH release assay was performed as per manufacturer’s guideline using different (10:1, 50:1, and 100:1) effector: target (E:T) cell ratios. The amount of spontaneously released LDH was measured in wells that contained only target cells. Total released LDH was assessed by lysing the target cells with 2% TritonX-100. The percentage of lysis was calculated according to the formula: % specific lysis = [(experimental lysis spontaneous lysis)/(maximum lysis spontaneous lysis)] 100. Similar ADCC assay was performed with murine CEA + CT-26 and CEA B16MelF10 cell lines as target cells and mouse splenic mononuclear cells, isolated from BALB/c mice as an effector, in presence of 2 mAbs.

To study the role of anti-NLGP mAb on tumor angiogenesis, 2 groups (n = 6, in each group) of BALB/c mice were inoculated with 2 106 cells/mice and after reaching tumor volume of 24 mm3 1 group was treated with anti-NLGP mAb keeping other group as isotype control. Both groups of mice were killed and skins were removed carefully from peritoneal region without disturbing the angiogenic blood vessels adjacent to tumors. Densitometric analysis was done using pixel values of the blood vessels.

In Vivo Effector Cell Status

Statistical Analysis

Blood was collected weekly from CT-26 tumor bearing mice (n = 6) 3 days after each anti-NLGP mAb treatment for 4 weeks. RBC-free mononuclear cells were stained with fluorescence-tagged anti-DX5 (for NK cells), anti-CD11b (for macrophages), and anti-CD16/32 (for Fc receptors) antibodies. Cells were fixed in 1% paraformaldehyde and analyzed flow cytometrically.

All results represent the average of 5 in vitro and 3 in vivo experiments. The number of experiments is mentioned in the Result section and Figure legends. In each experiment, a value represents the mean of 3 individual observations and is presented as mean ± SD. We compared all pairs of columns using t test with InStat3 software (GraphPad Software Inc.) and differences between groups attaining a P < 0.05 are considered significant.

Angiogenesis Study With Blood Vessels

Tumor Cell Proliferation Assay Different CEA + and CEA human and murine cells (2 105/well) were plated in each well of a 96-well microtiter plate and stimulated with anti-NLGP mAb (5 mg/mL) and nonspecific stimulator-Concanavalin A (ConA) (5 mg/ mL) for 72 hours. Cell proliferation was checked by MTT assay as described.32 To confirm the proliferation, CEA + and CEA cells were maintained in culture and resuspended in 0.1% BSA in PBS at a concentration of 1 106 cells/mL and labeled with 10 mM carboxyfluorescein succinimidyl ester at 371C for 10 minutes. Cells were then seeded in 6-well plates (1.5 106 cells/well) and 6 hours later anti-NLGP and anti-CEA mAbs (5 mg/mL) were added. Following incubation for 18 hours, cells were harvested and analyzed using a flow cytometer with 488 nm excitation and emission filters appropriate for fluorescein. r


RESULTS A mAb is Generated Against NLGP Anti-NLGP polyclonal antibody was generated in BALB/c mice by immunization with NLGP-KLH-Adj (Fig. 1B). Twelve clones for anti-NLGP antibody were generated and of 12 clones obtained, clone 1C8 was found to be highly antibody secreting one (Fig. 1C), thus, further monocloned by limiting dilution. Following selection in HAT and HT media, supernatants were checked by ELISA for antiNLGP mAb (1C8) titer (Fig. S1, Supplemental Digital Content 1, http://links.lww.com/JIT/A348). Hybrid cells were expanded in mouse peritoneum and ascitic fluid was checked for 1C8 (Fig. S1, Supplemental Digital Content 1, http:// links.lww.com/JIT/A348). Protein G-purified mAb showed www.immunotherapy-journal.com |

399

Das et al

J Immunother


FIGURE 4. Detection of carcinoembryonic antigen (CEA) by anti-neem leaf glycoprotein (NLGP) monoclonal antibody (mAb) using flow cytometry and in vivo restriction of CEA + tumors by same mAb. A, Different human CEA + (Colo-205, HCT-116, and HT-29) and CEA (MCF7, SiHa, and SCC084) tumor cells were maintained in their respective media and after 8–10 days in culture, cells were harvested and stained with both anti-NLGP and anti-CEA mAbs and detected flow cytometrically. Percent of positive cells in bar diagrams (A1) and representative figures in histogram as analyzed by flowJo software (A2) are presented. Mean values ± SD of 3 individual observations are presented in each case. B1, BALB/c mice (n = 6) were inoculated with murine colon carcinoma CT-26 cells (2 106 cells/mice); after the formation of palpable tumor, mice were treated with anti-NLGP mAb and tumor growth was monitored twice a week by caliper measurement and tumor volume was determined using the formula: tumor volume = (width)2 length/2. Other group of mice received unrelated rat IgG antibody as isotypic control. Tumor growth restriction (B2) and survivability (B3) curve are presented. C1, Athymic nude mice (n = 6) were inoculated with human colon cancer Colo-205 cells (2106 cells/mice); after the formation of palpable tumor, mice were treated with anti-NLGP mAb (25 mg) once in a week, and 4 injections in total. Other group of mice received unrelated rat IgG antibody as isotypic control. Tumor growth was monitored by the method as mentioned in (B1). Tumor growth restriction (C2) and survivability (C3) curve are presented. After killing, tumor weight of both isotype control and mAb-treated group were documented. Circles represent tumor areas. In both CT-26 and nude mouse models, growth restriction curves are presented in bar diagram with mean ± SD. ***P < 0.001. In both tumor growth restriction curve (B2 and C2), day of anti-NLGP mAb immunization was shown with arrows.

strong reactivity with NLGP, which was better than ascitic fluid and culture supernatants (Fig. S1, Supplemental Digital Content 1, http://links.lww.com/JIT/A348). In SDS-PAGE, heavy and light chains of molecular weight 50 and 25 kDa were detected (Fig. S1, Supplemental Digital Content 1, http:// links.lww.com/JIT/A348). Anti-NLGP mAb recognized NLGP in Western blotting (Fig. 1D). To reconfirm the specificity of mAb with NLGP, different NLGP concentrations (0.1–5 mg) were used and it can recognize a minimum concentration of NLGP by ELISA (Fig. S1, Supplemental Digital Content 1, http://links.lww.com/JIT/A348). Immunoglobulin class and subtype of 1C8 is IgG2a (Figs. 1E, F).


Anti-NLGP mAb is Reactive With Purified and Secretory CEA On the basis of our earlier preliminary observation on recognition of CEA by anti-NLP sera,36 we validated the reactivity of generated anti-NLGP mAb, 1C8, with CEA, in vivo and in vitro. 1C8 showed specific reaction with NLGP and CEA in comparison with other antigens tested (Fig. 2A). In addition to CEA whole molecule, CEA peptide fragment was checked whether 1C8 could recognize specific CEA epitope in comparison with anti-CEA mAb. The result showed (Fig. S1, Supplemental Digital Content 1, http://links.lww.com/JIT/A348) that 1C8 successfully r


J Immunother



FIGURE 5. Induction of antibody-dependent cellular cytotoxicity (ADCC) of CEA + tumors using anti-neem leaf glycoprotein (NLGP) mAb. A, ADCC of human (A1) CEA + (Colo-205, HCT-116, and HT-29) and CEA (MCF7, SiHa, and SCC084) and murine (A2) CT-26 (CEA + ) and B16MelF10 (CEA ) cells was assessed using anti-NLGP and anti-carcinoembryonic antigen (CEA) antibodies (5 mg/mL, in each case) with different effector: target (E:T) cell ratios (10:1, 50:1, and 100:1). RBC-depleted murine splenocytes were used as effectors. Mean values ± SD of 3 individual observations are presented in each case. *P < 0.05; **P < 0.01. B, Blood was collected weekly from CT-26 tumor bearing mice (n = 6) following 3 days of each anti-NLGP mAb treatment for 4 weeks. To check the effector cells (NK cells and macrophages) and Fc receptor status, RBC-free mononuclear cells were stained with fluorescence-tagged anti-DX5 (for NK cells) (B1), anti-CD11b (for macrophages) (B2), and anti-CD16/32 (for Fc receptors) (B3) antibodies. Positive cells were represented in bar diagram with mean values ± SD of 3 individual observations. **P < 0.01.

recognizes the particular CEA peptide. Again, by Western blotting purified CEA was reacted with both anti-NLGP (1C8) and anti-CEA mAbs and mAb 1C8 along with antiCEA mAb recognized CEA almost in equal extent (Figs. 2C1, C2). Neuraminidase treatment of CEA resulted significant reduction of reactivity of anti-NLGP mAb as well as anti-CEA mAb with neuraminidase-treated CEA (Fig. 2B). CEA + and CEA colon cancer cell lines (both human and murine), as mentioned in Materials and methods section, were cultured (initiating with equal number of cells in each case) for 72 hours after obtaining 60% confluency and cell-free supernatants were collected to check secretary soluble CEA. These supernatants were used to coat plates for ELISA, keeping corresponding culture media as controls. Using both anti-NLGP and anti-CEA mAbs, it was found that adsorbed compound of culture supernatants from CEA + cell lines showed significantly higher reactivity than those adsorbed from CEA cells. Reactivity was identical with both 1C8 and anti-CEA mAb (Fig. 2D). Serum obtained from CT-26 inoculated BALB/c mice was checked by ELISA using 1C8 and anti-CEA mAb. Anti-NLGP 1C8 mAb can detect soluble CEA in sera, might be secreted from inoculated tumors, in similar extent r


to anti-CEA mAb (Fig. 2E). The study demonstrated the increased level of circulating CEA, as detected by both antibodies, with progression of the tumorigenesis. This mAb is also reactive with sera from human colon cancer patients and showed no reactivity with the same from healthy individuals (Fig. 2F). In another experiment, cultured Colo-205, SiHa, CT26, and B16MelF10 cell lines were thoroughly lysed and same concentration of protein (that may contain CEA) was used to measure the reactivity of anti-NLGP and anti-CEA mAbs by ELISA and Western blotting. By ELISA, it was found that proteins isolated from Colo-205 and CT-26 showed higher reactivity with 1C8 than same from SiHa and B16MelF10 cells (Fig. 2G). Western blotting also confirmed this observation (Figs. 2H1, H2). Again, CEA was immunoprecipitated by using both mAbs and then detected using anti-CEA mAb (Fig. S1, Supplemental Digital Content 1, http://links.lww.com/JIT/A348).

Anti-NLGP mAb is Reactive With CEA + Colon Cancer Tissues and Xenografts As CEA is overexpressed in colon cancer,42 histopathologically confirmed well to moderately differentiated human colon cancer specimens (n = 10) were stained by immunohistochemical (Fig. 3A1) method. All 10 malignant www.immunotherapy-journal.com |

401

Das et al

J Immunother


FIGURE 6. Partial normalization of tumor angiogenesis in anti-neem leaf glycoprotein (NLGP) mAb-treated CT-26 tumor bearing mice. A, CT-26 tumor bearing mice were treated with anti-NLGP mAb and 1 group was taken as isotype control (n = 6, in each case). Following 30 days after initiation of the treatment, mice were killed, tumor harvested, skins removed, and studied for angiogenesis. Chaotic (thick arrows) and normalized (thin arrows) blood vessels (A1) with densitometric analysis (A2) of 3 individual observations are presented as mean ± SD. **P < 0.01. B, Total RNA was isolated from tumors of control and anti-NLGP mAb-treated mice (n = 3 in each case) to analyze genes, like, VEGF, VEGFR2, and CD31 at transcriptional level by RT-PCR (B1) and densitometric analysis (B2) was performed in each case and presented with mean values ± SD of 3 individual observations. *P < 0.05; **P < 0.01. C, Frozen sections of tumors from either isotype control or anti-NLGP mAb-treated mice were stained immunohistochemically with mAbs, specific for VEGF, VEGFR2, and with fluorescence-tagged anti-CD31 antibody.


r


J Immunother


specimens were immunologically reactive for anti-NLGP and anti-CEA mAbs. Overall quantitation suggests, in average, 310 cells were reacted with anti-CEA mAb, whereas, 1C8 detected 245 cells in immunohistochemistry (Fig. 3A2). Unrelated antibody, rat IgG showed almost negative staining. Cancer associated human normal colon tissues were also unreactive with these mAbs (Fig. 3A3). Other colon cancer tissues samples have shown similar kind of recognition with both antibodies (Fig. S2, Supplemental Digital Content 2, http://links.lww.com/JIT/A349). Presence of CEA was also detected in mouse CT-26 and human Colo-205 colon tumor xenografts by 1C8 and the expression was comparable with anti-CEA mAb (Figs. 3B1, B2, C1, C2, S2, Supplemental Digital Content 2, http://links.lww.com/JIT/A349). Murine melanoma and human cervical tumor xenograft (CEA ) appeared nonreactive with these mAbs (Figs. 3D1, D2).

Anti-NLGP mAb is Reactive to Cancer Cells Expressing CEA CEA is expressed on the outer membrane of most of the colon cancer cells.42 Human CEA + 3 colon cancer cell lines, along with CEA cells (Figs. 4A1, A2 and S3, Supplemental Digital Content 3, http://links.lww.com/JIT/ A350) and murine CT-26 and B16MelF10 (Fig. S3, Supplemental Digital Content 3, http://links.lww.com/JIT/ A350) cells were stained with anti-NLGP and anti-CEA mAbs for flow cytometry. Results revealed the detection of cell surface expression of CEA on CEA + cells in varying degrees. However, negligible reaction was detected in CEA cells. Among CEA expressing cells, Colo-205 and CT-26, showed greater surface CEA expression than others.

Anti-NLGP mAb Restricts CEA + Murine and Human Colon Tumor Growth in Vivo Given the importance of CEA in adhesion and metastasis of cancer cells,43 next, we examined the therapeutic potential of anti-NLGP mAb that can efficiently recognize the CEA on different colon tumor systems. In this study, we have selected 1 murine (CT-26 cells in BALB/c) and other human (Colo-205 cells in athymic nude mice) CEA + tumor xenograft models. After 1C8 mAb therapy (25 mg mAb/mice for 4 wk), CEA + solid CT-26 tumor growth in BALB/c mice (Figs. 4B1, B2) and Colo-205 in athymic nude mice (Figs. 4C1, C2) were significantly restricted in comparison with isotype control group and longer survivability in 1C8-treated groups was also noticed (Figs. 4B3, C3). There was no restriction observed when BALB/c or nude mice were inoculated with CEA tumors (data not shown).

Anti-NLGP mAb Induces ADCC Reaction to CEA + Colon Tumor Cells in Vitro As we observed anti-NLGP mAb, 1C8, can successfully restrict the CEA + murine and human colon tumor growth; mechanisms underlying the antitumor activity by this 1C8 mAb was evaluated. ADCC reaction using human and mouse effector cells and different CEA + and CEA cell lines as target in presence of 1C8/anti-CEA mAb revealed a strong cytotoxic reaction with CEA + cells, whereas, weak reaction in case of CEA cells. ADCC reaction was also tested selecting CEA + mouse CT-26 and CEA B16MelF10 as targets (Figs. 5A1, A2) and CEA specific reaction was noticed. r



Observed strong ADCC reaction prompted us to examine the NK cell and macrophage status, as they generally mediate ADCC. Flow cytometric analysis revealed that increased number of NK cells in mice is correlated well with extent of tumor growth restriction (Fig. 5B1). However, no significant change was noted in macrophages (Fig. 5B2). In addition, Fc receptor status was studied revealing significant reduction in CD16/32 detectable Fc receptor in 1C8-treated mice (Fig. 5B3) with respect to isotype control group.

Anti-NLGP mAb has Little Role in the Proliferation and Migration of CEA + Cells As we observed anti-NLGP mAb-induced cellular cytotoxicity is antibody dependent, next we checked if the generated mAb influences the tumor cell proliferation and migration. For proliferation, MTT assay and carboxyfluorescein succinimidyl ester staining assay showed that this generated mAb has no inhibitory effect on tumor cell proliferation (Fig. S4, Supplemental Digital Content 4, http:// links.lww.com/JIT/A351). In addition to proliferation, we detected no migration of human Colo-205 and murine CT-26 cells through transwell membranes under influence of mAbs (Fig. S4, Supplemental Digital Content 4, http://links.lww. com/JIT/A351).

Anti-NLGP mAb Normalizes Tumor Angiogenesis A recent report44 suggested that CEA may induce angiogenesis by regulating proangiogenic endothelial cells and altering VEGF signaling molecules. Accordingly, we assessed the role of CEA recognizing anti-NLGP mAb on angiogenic profile with special emphasis on VEGF, its receptor, VEGFR2, and CD31 in BALB/c mice with established CT-26 tumor. Visual (Fig. 6A1) and densitometric (Fig. 6A2) observations suggested that such mAb partially normalizes angiogenesis in 1C8-treated CT-26 tumor bearers than isotype controls. This mAb normalizes the chaotic nature of the tumor-associated angiogenic blood vessels in skin surrounding tumors. As evidenced by RT-PCR (Figs. 6B1, B2) or immunohistochemistry (Fig. 6C) analysis, VEGF, VEGFR2, CD31 were downregulated within tumors from anti-NLGP (1C8) treated CT-26 mice.

DISCUSSION CEA is a promising molecule as an immunotherapeutic target45 and several fundamental studies on cancer immunotherapy were conducted using CEA.11,46 Associated complications of targeting CEA using active specific immunotherapy have been reported time-to-time45 and selftolerance to CEA is considered as a major one.12 From this perspective, we reported here a novel anti-NLGP mAb 1C8, which was generated against a plant protein NLGP. It has surprising capability of targeting CEA, accumulating in CEA + tumor tissues and has striking functions of killing tumor cells to inhibit growth of both murine and human colon tumors in vivo. These useful properties of 1C8 make it a truly promising diagnostic as well as therapeutic tool for CEA + tumors, especially tumors of large intestinal origin. It needs to be mentioned here that a similar effect is achieved by using anti-CEA mAb, as discussed earlier by several groups.47,48 Use of both anti-CEA and anti-NLGP mAbs may avoid the complications related to tolerance during active specific immunotherapy using CEA molecule.45,49 Parkhurst et al50 discussed another novel www.immunotherapy-journal.com |

403

Das et al

approach of CEA targeting using autologous T lymphocytes genetically engineered to express a murine T-cell receptor against human CEA. Administration of such immune regimen in 3 patients with metastatic colorectal cancer refractory to standard treatments resulted profound decreases in serum CEA levels (74%–99%), and 1 patient had an objective regression of cancer metastatic to the lung and liver. In association to anti-CEA/antitumor immune response, they also experienced severe transient colitis. Thus, use of 1C8 for CEA targeting might offer comparable clinical benefit. There is scope to compare the clinical efficacy in further studies. On the basis of wide experience to work with NLGP in respect to immunomodulation,32,33,41 we have obtained a basic clue on recognition of CEA by anti-NLP polyclonal antibody.35 To further confirm this observation, we have attempted to generate a monoclone against NLGP, the purified fraction of NLP. The best monoclone, 1C8, was characterized by its reactivity to NLGP, CEA, and some other tumor antigens. The 1C8 (IgG2a) showed maximum reaction to NLGP and uniquely, it has strong reactivity to purified human CEA and a CEA peptide. Being a glycoprotein nature of CEA, exposure to neuraminidase significantly reduces its reactivity with anti-NLGP mAb (1C8). Again, 1C8 recognizes a known sequence of the CEA peptide, thus, there is a possibility that a homologous peptide is present in NLGP. However, confirmation of this hypothesis is pending due to unavailability of the complete sequence of NLGP. To understand the extent of CEA recognition by 1C8, here, we have used 3 different CEA + human colon carcinoma cell lines (Colo-205, HCT-116 and HT-29) in cultures, where, CEA content was found between 1 and 4 ng/ 106 cells and effectively recognized by 1C8. Firstly, supernatants from such cultures were used to coat ELISA plates, where strong binding of 1C8 was detected. Both anti-CEA and anti-NLGP mAbs reacted with CEA in identical extent, suggesting utility of anti-NLGP mAb in targeting this antigen. However, neuraminidase treatment significantly diminishes such reaction with anti-NLGP mAb. We have conducted in vivo experiments with mouse CEA + colon carcinoma cells (CT-26). Serum was collected from mice bearing CEA expressing CT-26 tumor and such serum showed reactivity with both of these mAbs. An interesting report suggested that the murine system have CEA-like genes and a murine CEA-like protein with a molecular weight of ˜ 120 kDa has been observed using antihuman CEA antibodies.51 This observation suggested that some antigenic determinants or epitopes may have been conserved between the human and murine CEA proteins. This reaction was increased with increment of the tumor load in mice, having more release of CEA in the circulation, thus, 1C8 might have utility in the diagnosis of CEA + tumors. As we detected more reaction with increment of the tumor load, this mAb may also reflect the tumor progression of host. In colon cancer patients, serum concentrations of CEA are to be higher in well-differentiated tumor-bearing patients compared with patients with poorly differentiated tumors.52,53 As 1C8 significantly detect serum CEA from patients, it can be used to monitor the progression of the disease. This mAb also can recognize CEA on (i) human (n = 10, along with corresponding adjacent normals) and mouse (CT-26 and Colo-205 xenograft) colon tumor sections; (ii) different CEA + colon cancer cell surfaces; and (iii) lysates from same cells. In every case, no reaction was


J Immunother


noticed in CEA cells. Actual reason of strong immunodetection of CEA by anti-NLGP mAb is unknown, but, this mAb may recognize some epitopes present in CEA molecule. Although complete structure of NLGP is not elucidated till date, it may be possible that some portions are structurally similar between 2 glycoproteins, NLGP and CEA. However, the puzzle needs to be solved by thorough biophysical analysis. Downstream signaling after binding of these mAbs with CEA on cancer cell surface may help in understanding of the process. Next, we asked the question whether observed recognition of CEA by anti-NLGP mAb has any antitumor effect for CEA + tumors. Therapy of BALB/c mice with murine CEA + CT-26 and nude mice with human CEA + Colo205 tumors with anti-NLGP mAb resulted significant restriction of tumor growth that might be due to the recognition of CEA by 1C8. MAb 1C8 not only effectively restricts colon tumor growth but also showed greater survivability of treated animals with no adverse effects (unpublished observation). Sections from these human tumor xenograft have intense expression of CEA, whereas, tumor sections from mice with anti-NLGP mAb therapy showed less expression of CEA. To elucidate the mechanism of antitumor action of anti-NLGP mAb, we have checked the ADCC reaction using CEA + either CT-26 or Colo-205 cells as a target. Strong ADCC reaction toward these CEA + tumor cells was noticed in presence of anti-NLGP mAb. Specificity of this reaction towards CEA + cells was confirmed by minimum reaction in case of CEA cells. In such ADCC reaction, FcR binding by mAbs played a crucial role.54 Apart from the target and antibody, possible effector cells were studied in CEA + tumor bearing mice under 1C8 therapy. Our results suggest that NK cells might serve as an effector in such ADCC, as number of DX5 + NK cells are increased with mAb therapy. Steady decrease in CD16/32 (FcR) was noted after mAb therapy indicating utilization of such receptors due to antibody binding. Thus, anti-NLGP mAb primarily works by induction of ADCC reaction recognizing cell surface CEA. It has undetectable effect on tumor cell proliferation and migration irrespective of their CEA status. In our study, another important observation was found that in mAb 1C8-treated CT-26 tumor bearing mice, angiogenesis was normalized with respect to chaotic vasculature of isotype control group (Fig. 6A1). We also observed a downregulation of main proangiogenic molecule VEGF and its receptor VEGFR2 by RT-PCR (Fig. 6B1) and IHC (Fig. 6C) of treated mice group compared with untreated ones. Similar downregulation was noticed in CD31 marker, indicating decrease in vascular endothelial cells. High tumor volume leads to higher release of soluble CEA into blood and these are responsible for colorectal cancer metastasis,44 because soluble CEA successfully promote angiogenesis by loosing adhesion and proliferation of proangiogenic endothelial cells both in vitro and in vivo.45 Thus, 1C8 mediated downregulation of CEA indirectly normalizes angiogenesis by downregulating VEGF and its related molecules. Accordingly, CEA + tumor growth is significantly restricted. Overall results suggest the generation and characterization of a novel mAb against a plant product NLGP that can recognizes human CEA to induce ADCC reaction and may normalize angiogenesis to restrict tumor growth. This anti-NLGP mAb is completely safe in relation to all r


J Immunother


physiological functions, as evaluated in tumor-free and tumor-bearing mouse models (unpublished observation). In addition to colon cancer, other CEA expressing tumors, like, lung and breast may be treated with developed 1C8 anti-NLGP mAb. However, humanization of such mAb would be most effective for clinical use.

ACKNOWLEDGMENTS The authors acknowledge Dr Jaydip Biswas, Director, CNCI, India for providing necessary facilities. They also acknowledge ICMR (61/5/2008-BMS) for funding this project. Their sincere thanks to Dr Abhijit Rakshit for providing experimental animals.

CONFLICTS OF INTEREST/ FINANCIAL DISCLOSURES The authors have declared there are no financial conflicts of interest in regards to this work.


14.

15.

16.

17.

18.

19.

REFERENCES 1. Gold P, Freedman SO. Demonstration of tumor-specific antigens in human colonic carcinoma by immunological tolerance and absorption techniques. J Exp Med. 1965;121: 439–462. 2. Thomas P, Toth CA, Saini KS, et al. The structure, metabolism and function of the carcinoembryonic antigen gene family. Biochim Biophys Acta. 1990;1032:177–189. 3. Benchimol S, Fuks A, Jothy S, et al. Carcinoembryonic antigen, a human tumor marker, functions as an intercellular adhesion molecule. Cell. 1989;57:327–334. 4. Oikawa S, Nakazato H, Kosaki G. Primary structure of human carcinoembryonic antigen (CEA) deduced from cDNA sequence. Biochem Biophys Res Commun. 1987;142:511–518. 5. Goldenberg DM, Sharkey RM, Primus FJ. Carcinoembryonic antigen in histopathology: immunoperoxidase staining of conventional tissue sections. J Natl Cancer Inst. 1976;57:11–22. 6. Grunert F, Luckenbach GA, Haderlie B, et al. Comparison of colon-, lung-, and breast-derived carcinoembryonic antigen and cross-reacting antigens by monoclonal antibodies and fingerprint analysis. Ann N Y Acad Sci. 1983;417:75–85. 7. Thompson JA, Pande H, Paxton RJ, et al. Molecular cloning of a gene belonging to the carcinoembryonic antigen gene family and discussion of a domain model. Proc Natl Acad Sci USA. 1987;84:2965–2969. 8. Freńoy N, Schauffler J, Cahour A, et al. Natural antibodies against the carcinoembryonic antigen (CEA) and a related antigen, NCA, in human sera. Anticancer Res. 1987;7: 1229–1233. 9. Tsang KY, Zaremba S, Nieroda CA, et al. Generation of human cytotoxic T cells specific for human carcinoembryonic antigen epitopes from patients immunized with recombinant vaccinia CEA vaccine. J Natl Cancer Inst. 1995;87:982–990. 10. Jothy S, Brazinsky SA, Chin-A-Loy M, et al. Characterization of monoclonal antibodies to carcinoembryonic antigen with increased tumor specificity. Lab Invest. 1986;54:108–117. 11. Hardman N, Murray B, Zwickl M, et al. Application of genetically-engineered anti-CEA antibodies for potential immunotherapy of colorectal cancer. Int J Biol Markers. 1992; 7:203–209. 12. Foon KA, John WJ, Chakraborty M, et al. Clinical and immune responses in resected colon cancer patients treated with anti-idiotype monoclonal antibody vaccine that mimics the carcinoembryonic antigen. J Clin Oncol. 1999;17: 2889–2895. 13. Bhattacharya-Chatterjee M, Mukherjee S, Biddle W, et al. Murine monoclonal anti-idiotype antibody as a potential r


20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

network antigen for human carcinoembryonic antigen. J Immunol. 1990;145:2758–2765. Pal S, Saha A, Mohanty K, et al. Generation of Her-2/neu vaccine utilizing idiotypic network cascade. Cancer Biol Ther. 2007;6:1916–1925. Chatterjee SK, Tripathi PK, Chakraborty M, et al. Molecular mimicry of carcinoembryonic antigen by peptides derived from the structure of an anti-idiotype antibody. Cancer Res. 1998;58: 1217–1224. Kawashima I, Tsai V, Southwood S, et al. Identification of HLA-A3-restricted cytotoxic T lymphocyte epitopes from carcinoembryonic antigen and HER-2/neu by primary in vitro immunization with peptide-pulsed dendritic cells. Cancer Res. 1999;59:431–435. Nukaya I, Yasumoto M, Iwasaki T, et al. Identification of HLA-A24 epitope peptides of carcinoembryonic antigen which induce tumor-reactive cytotoxic T lymphocyte. Int J Cancer. 1999;80:92–97. Zaremba S, Barzaga E, Zhu M, et al. Identification of an enhancer agonist cytotoxic T lymphocyte peptide from human carcinoembryonic antigen. Cancer Res. 1997;57:4570–4577. Bei R, Kantor J, Kashmiri SV, et al. Enhanced immune responses and anti-tumor activity by baculovirus recombinant carcinoembryonic antigen (CEA) in mice primed with the recombinant vaccinia CEA. J Immunother Emphasis Tumor Immunol. 1994;16:275–282. Kaufman H, Schlom J, Kantor J. A recombinant vaccinia virus expressing human carcinoembryonic antigen (CEA). Int J Cancer. 1991;48:900–907. Alters SE, Gadea JR, Philip R. Immunotherapy of cancer. Generation of CEA specific CTL using CEA peptide pulsed dendritic cells. Adv Exp Med Biol. 1997;417:519–524. Alters SE, Gadea JR, Sorich M, et al. Dendritic cells pulsed with CEA peptide induce CEA-specific CTL with restricted TCR repertoire. J Immunother. 1998;21:17–26. Nair SK, Hull S, Coleman D, et al. Induction of carcinoembryonic antigen (CEA)-specific cytotoxic T- lymphocyte responses in vitro using autologous dendritic cells loaded with CEA peptide or CEA RNA in patients with metastatic malignancies expressing CEA. Int J Cancer. 1999;82:121–124. McLaughlin JP, Schlom J, Kantor JA, et al. Improved immunotherapy of a recombinant carcinoembryonic antigen vaccinia vaccine when given in combination with interleukin-2. Cancer Res. 1996;56:2361–2367. Rao JB, Chamberlain RS, Bronte V, et al. IL-12 is an effective adjuvant to recombinant vaccinia virus-based tumor vaccines: enhancement by simultaneous B7-1 expression. J Immunol. 1996;156:3357–3365. Baral RN, Saha A, Chatterjee S, et al. Immunostimulatory CpG oligonucleotides enhance the immune response of antiidiotype vaccine that mimics carcinoembryonic antigen. Cancer Immunol Immunother. 2003;52:317–327. Foon KA, Yannelli J, Bhattacharya-Chatterjee M. Colorectal cancer as a model for immunotherapy. Clin Cancer Res. 1999; 5:225–236. Sto¨ger E, Vaquero C, Torres E, et al. Cereal crops as viable production and storage systems for pharmaceutical scFv antibodies. Plant Mol Biol. 2000;42:583–590. Vaquero C, Sack M, Chandler J, et al. Transient expression of a tumor-specific single-chain fragment and a chimeric antibody in tobacco leaves. Proc Natl Acad Sci USA. 1999;96: 11128–11133. Daniell H, Streatfield SJ, Wycoff K. Medical molecular farming: production of antibodies, biopharmaceuticals and edible vaccines in plants. Trends Plant Sci. 2001;6:219–226. Mandal-Ghosh I, Chattopadhyay U, Baral R. Neem leaf preparation enhances Th1 type immune response and antitumor immunity against breast tumor associated antigen. Cancer Immunity. 2007;7:8. Baral RN, Mandal I, Chattopadhyay U. Immunostimulatory neem leaf preparation acts as an adjuvant to enhance the

www.immunotherapy-journal.com |

405

Das et al

33.

34.

35. 36.

37.

38.

39. 40. 41.

42.

43.

efficacy of poorly immunogenic B16 melanoma surface antigen vaccine. Int Immunopharmacol. 2005;5:1343–1352. Sarkar K, Bose A, Chakraborty K, et al. Neem leaf glycoprotein helps to generate carcinoembryonic antigen specific anti-tumor immune responses utilizing macrophagemediated antigen presentation. Vaccine. 2008;26:4352–4362. Sarkar K, Goswami S, Roy S, et al. Neem leaf glycoprotein enhances carcinoembryonic antigen presentation of dendritic cells to T and B cells for induction of antitumor immunity by allowing generation of immune effector/memory response. Int Immunopharmacol. 2010;10:865–887. Sarkar K, Bose A, Laskar S, et al. Antibody response against neem leaf preparation recognizes carcinoembryonic antigen. Int Immunopharmacol. 2007;7:306–312. Mallick A, Barik S, Goswami KK, et al. Neem leaf glycoprotein activates CD8(+) T cells to promote therapeutic anti-tumor immunity inhibiting the growth of mouse sarcoma. PLoS One. 2013;8:e47434. Baral R, Chattopadhyay U. Neem (Azadirachta indica) leaf mediated immune activation causes prophylactic growth inhibition of murine Ehrlich carcinoma and B16 melanoma. Int Immunopharmacol. 2004;4:355–366. Chakraborty K, Bose A, Pal S, et al. Neem leaf glycoprotein restores the impaired chemotactic activity of peripheral blood mononuclear cells from head and neck squamous cell carcinoma patients by maintaining CXCR3/CXCL10 balance. Int Immunopharmacol. 2008;8:330–340. Kohler G, Milstein C. Continuous culture of fused cells secreting antibody of predefined specificity. Nature. 1975;256: 495–497. Bailey JL. Miscellaneous analytical methods. In: Bailey JL, ed. Techniques in Protein Chemistry. New York: Elsevier Science Publishing; 1967:340–346. Barik S, Banerjee S, Mallick A, et al. Normalization of tumor microenvironment by neem leaf glycoprotein potentiates effector T cell functions and therapeutically intervenes in the growth of mouse sarcoma. PLoS One. 2013;8:e66501. Conaghan PJ, Ashraf SQ, Tytherleigh MG, et al. Targeted killing of colorectal cancer cell lines by a humanised IgG1 monoclonal antibody that binds to membrane-bound carcinoembryonic antigen. Br J Cancer. 2008;98:1217–1225. Pakdel A, Naghibalhossaini F, Mokarram P, et al. Regulation of carcinoembryonic antigen release from colorectal cancer cells. Mol Biol Rep. 2012;39:3695–3704.


J Immunother


44. Bramswig KH, Poettler M, Unseld M, et al. Soluble carcinoembryonic antigen activates endothelial cells and tumor angiogenesis. Cancer Res. 2013;73:6584–6596. 45. Ho¨rig H, Medina FA, Conkright WA, et al. Strategies for cancer therapy using carcinoembryonic antigen vaccines. Expert Rev Mol Med. 2000;2:1–24. 46. Morse MA, Deng Y, Coleman D, et al. A phase I study of active immunotherapy with carcinoembryonic antigen peptide (CAP-1)-pulsed, autologous human cultured dendritic cells in patients with metastatic malignancies expressing carcinoembryonic antigen. Clin Cancer Res. 1999;5:1331–1338. 47. Zheng C, Feng J, Lu D, et al. A novel anti-CEACAM5 monoclonal antibody, CC4, suppresses colorectal tumor growth and enhances NK cells-mediated tumor immunity. PLoS One. 2011;6:e21146. 48. Stein R, Goldenberg DM. A humanized monoclonal antibody to carcinoembryonic antigen, labetuzumed, inhibits tumor growth and sensitizes human medullary thyroid cancer xenografts to decarbazine chemotherapy. Mol Cancer Ther. 2004;3:1559–1564. 49. Cloosen S, Arnold J, Thio M, et al. Expression of tumorassociated differentiation antigens, MUC1 glycoforms and CEA, in human thymic epithelial cells: implications for selftolerance and tumor therapy. Cancer Res. 2007;67:3919–3926. 50. Parkhurst MR, Yang JC, Langan RC, et al. T cells targeting carcinoembryonic antigen can mediate regression of metastatic colorectal cancer but induce severe transient colitis. Mol Ther. 2011;19:620–626. 51. Beauchemin N, Turbide C, Afar D, et al. A mouse analogue of the human carcinoembryonic antigen. Cancer Res. 1989;49: 2017–2021. 52. Thomas P, Forse AR, Bajenova O. Carcinoembryonic antigen (CEA) and its receptor hnRNP M are mediators of metastasis and the inflammatory response in the liver. Clin Exp Metastasis. 2011;28:923–932. 53. Dallas RM, Liu G, Chen WC, et al. Divergent roles of CD44 and carcinoembryonic antigen in colon cancer metastasis. The FASEB J. 2012;26:2648–2656. 54. Strome SE, Sausville EA, Mann D. A mechanistic perspective of monoclonal antibodies in cancer therapy beyond targetrelated effects. Oncologist. 2007;12:1084–1095.

r


Murine carcinoma expressing carcinoembryonic antigen-like protein is restricted by antibody against neem leaf glycoprotein.

A recombinant vaccinia virus expressing human carcinoembryonic antigen (CEA).

[Carcinoembryonic antigen (CEA) (author's transl)].

Carcinoembryonic antigen (CEA) levels and tumor histology in colon cancer.

Mouse-human chimeric monoclonal antibody to carcinoembryonic antigen (CEA): in vitro and in vivo activities.

First British standard for carcinoembryonic antigen (CEA).

Carcinoembryonic antigen (CEA): ten years' perspective.

A monkey antigen crossreacting with carcinoembryonic antigen, CEA.

Differential reactivities of carcinoembryonic antigen (CEA) and CEA-related monoclonal and polyclonal antibodies in common epithelial malignancies.

[Carcinoembryonic antigen (CEA) and CEA-like activities in ascites and pleura-effusions (author's transl)].

Multiple Smith-degradations of carcinoembryonic antigen (CEA) and of asialo CEA.

A predicted three-dimensional structure for the carcinoembryonic antigen (CEA).

Carcinoembryonic antigen (CEA): molecular biology and clinical significance.

[Carcinoembryonic antigen (CEA) in head and neck tumors (author's transl)].

Keratin and carcinoembryonic antigen (CEA) in human melanoma cells.

Carcinoembryonic antigen (CEA) in the gastric mucosa after partial gastrectomy.

Detection of carcinoembryonic antigen in colonic effluent by specific anti-CEA monoclonal antibodies.

Tissue demonstration of carcinoembryonic antigen (CEA) in ulcerative colitis.

Carcinoembryonic antigen (CEA) as prognostic marker in colonic cancer.

Immunohistochemical localization of carcinoembryonic antigen (CEA) in salivary gland tumors.

Carcinoembryonic antigen (CEA) levels in patients with brain tumours.

Carcinoembryonic antigen (CEA) assays in obstructive colorectal cancer.

Anti-CEA antibody fragments labeled with [(18)F]AlF for PET imaging of CEA-expressing tumors.

Biochemical characterization of 25 distinct carcinoembryonic antigen (CEA) epitopes recognized by 57 monoclonal antibodies and categorized into seven groups in terms of domain structure of the CEA molecule.