Human chimeric antigen receptor (CAR) T-cell therapy has been used to successfully treat B-cell leukemia and lymphoma as well as multiple myeloma, and there are currently 7 US FDA–approved CAR T-cell products.1 Chimeric antigen receptor T therapy is being explored in preclinical models and clinical trials for additional indications, including infectious diseases and autoimmunity.2–8 CAR T therapy involves infusion of engineered autologous T cells expressing a chimeric receptor specific for an antigen on target cells.9 The CAR is comprised of a single-chain variable fragment tethered by a hinge domain to a transmembrane domain and 1 or more cytoplasmic domains derived from the T-cell receptor signaling complex and costimulatory molecules.10,11
The success of CAR T therapy in treating human cancer patients has spurred investigation in veterinary medicine into the design and manufacturing of CAR T cells for companion animals. The first report of canine CAR T cells tested growth conditions, such as artificial antigen-presenting cells, anti-canine CD3/CD28 beads, or concanavalin A (ConA) to expand T cells, and generated the first-in-canine CAR T cells by electroporating mRNA encoding a canine CD20 CAR to treat a patient with B-cell lymphoma.12 Subsequent studies13,14 utilized viral transduction for CAR T generation and compared various media formulations with mitogenic stimuli like ConA, phytohemagglutinin, and phorbol myristate acetate (PMA) with ionomycin (PMA/I) for their ability to support efficient canine T-cell transduction.15 Additional studies16,17 have generated canine CAR T cells directed toward solid tumor antigens, such as B7-H3 and IL-13Rα2. Despite these developments in CAR T therapy for dogs, no such investigations have been conducted for cats. Humans, dogs, and cats share similar environmental exposures, and all develop spontaneous cancer; therefore, in addition to direct benefits for treating feline cancer, feline CAR T development may provide opportunities for translation to human oncology.
Here, we tested the optimal growth conditions for feline T cells using cytokine and mitogen combinations previously demonstrated to grow human and canine T cells. We also tested whether feline T cells could be polarized with different cytokine and mitogenic stimuli in vitro to more TH1-like or Treg-like phenotypes that could have future use in augmenting CAR T therapy for infectious and autoimmune diseases. We found that recombinant feline IL-2 (rfIL-2) with ConA or PMA/I yielded robust proliferation and activation, but PMAI/I resulted in higher response to polarizing cytokines and ex vivo expansion. Finally, we utilized an FIV-based lentiviral system to transduce primary feline T cells with a human CD19 CAR and observed specific cytotoxicity against human CD19+ target cells. Together, these data demonstrate the optimal conditions for feline T-cell growth and polarization as well as the novel generation of functional feline CAR T cells.
Methods
Primary feline T-cell isolation
Venipunctures on healthy donor cats were carried out by the Cornell Center for Animal Resources and Education following protocols approved by IACUC. Individual donor characteristics are listed in Supplementary Table S1. Heparinized blood was diluted up to 35 mL with 1X Dulbecco PBS (DPBS) layered over Ficoll Paque PREMIUM (Cytiva) and spun at 400 X g for 30 minutes at 20 °C with no brake. Peripheral blood mononuclear cells were washed twice with magnetic-activated cell sorting (MACS) buffer (1X DPBS + 0.5% bovine serum albumin [Roche] + 2 mM EDTA) at 200 X g for 10 minutes at 20 °C to remove platelets. T cells were incubated with anti-feline CD3 clone NZM118 hybridoma supernatant (provided by Y. Nishimura, National Institute of Infectious Diseases, Tokyo, Japan) for 30 minutes at 4 °C and washed with MACS buffer, followed by incubation with anti-mouse IgG Microbeads (Miltenyi Biotec) at 4 °C for 15 minutes. Cells were washed with MACS buffer and positively selected using a MS column (Miltenyi Biotec). Cells were stained for flow cytometry and/or put into culture in complete media (XVIVO-15 [Lonza] + 10% fetal bovine serum [Avantor Seradigm] + 2 mM L-glutamine [Corning]) as specified for each assay.
T-cell labeling and stimulation
Peripheral blood mononuclear cells were washed twice with MACS buffer and labeled in DPBS with 5 μM CellTrace Violet (Invitrogen) for 20 minutes at 37 °C prior to the addition of 5 volumes of complete XVIVO-15. An aliquot was removed to compare T-cell percentage pre- and post-enrichment before proceeding with T-cell enrichment as described above. 1 X 105 enriched T cells/well were seeded in 100 μL of complete XVIVO-15 in 96-well round-bottom tissue culture plates (Thermo) with 10 ng/mL rfIL-2 (R&D Systems) alone or in combination with 100 ng/mL recombinant human IL-21 (rhIL-21; Peprotech), 5 μg/mL ConA (Sigma), or 25 ng/mL PMA (Sigma) + 1 μg/mL ionomycin (Sigma) for 24 hours. Medium was exchanged to remove mitogenic stimuli, rfIL-2 (for rfIL-2 alone, ConA, and PMA/I) or rfIL-2 with rhIL-21 was added, and cells were cultured for an additional 3 days. Cells were either stained for viability and CD4/CD8/CD134 expression or CD3 purity. The gating strategy is detailed in Supplementary Figure S1.
Flow cytometry
Cell viability was determined by staining cells in LIVE/DEAD Near IR (Invitrogen) diluted in DPBS according to manufacturer's instructions. Antibodies used were anti-feline CD4 FITC (Bio-Rad; clone vpg34), anti-feline CD8 PE (Southern Biotech; clone fCD8), anti-feline CD134 AlexaFluor 647(Bio-Rad; clone 7D6), goat anti-mouse IgG3 AlexaFluor 488 (Invitrogen), anti-human CD19 FITC (BioLegend; clone HIB19), and anti-human CD271 (for truncated nerve growth factor receptor [LNGFR]) BV421 (BD; clone C40–1457). Primary and secondary antibodies were diluted in flow cytometry buffer (DPBS + 1% bovine serum albumin + 0.1% NaNH3 + 1 mM EDTA) and stained for 30 minutes at 4 °C in the dark. Cells were washed with flow cytometry buffer before fixing and staining in Cytofix/Cytoperm solution (BD) for 5 minutes at 4 °C. Proliferation was determined by flow cytometry through CellTrace violet dilution. All T-cell experiments were analyzed on an Attune NxT Acoustic Focusing Cytometer (Thermo). Analysis was carried out using FlowJo, version 10.6.2 (BD). Antibody dilutions are listed in Supplementary Table S2.
T-cell expansion
1 X 106 primary T cells/well were seeded onto a G-Rex 24-well plate (Wilson Wolf) in a total volume of 1 mL/well complete XVIVO-15 with 10 ng/mL rfIL-2 and either 5 μg/mL ConA or 25 ng/mL PMA + 1 μg/mL ionomycin. Twenty-four hours later, media volume was increased to 8 mL/well with complete XVIVO-15 + 10 ng/mL rfIL-2. On days 4, 8, and 12 post seeding, half of the media was replenished with XVIVO-15 + 10 ng/mL rfIL-2, and viable cells were counted on a hemocytometer using trypan blue.
Thelper cell polarization
1 X 105 enriched T cells/well were seeded in 100 μL complete XVIVO-15 in 96-well round-bottom tissue culture plates with 10 ng/mL rfIL-2 and either 5 μg/mL ConA or 25 ng/mL PMA + 1 μg/mL ionomycin. Polarizing conditions consisted of 10 ng/mL rfIL-2 without additional cytokines (TH0) or with 10 ng/mL recombinant feline IL-12 (TH1) or 2 ng/mL recombinant human TGF-β1 (Treg) (both from R&D Systems). The next day, wells were adjusted to 200 μL by removing 50 μL and adding 150 μL fresh polarization mixture without mitogens but adjusting to the polarizing conditions with cytokines. On day 4, each well was split into 2 wells, respective cytokines were added, and cells were cultured for an additional 2 days. Detailed RNA isolation, cDNA generation, and quantitative PCR protocols are listed in Supplementary Methods. All primers were purchased as oligonucleotides (Integrated DNA Technologies) and are listed in Supplementary Table S3.
Enzyme-linked immunosorbent assay
On day 6, supernatants from each polarization condition per donor were clarified at 400 X g, pooled, and stored at −80 °C. Interferon-γ and IL-10 were quantified using Feline Interferon-γ and Feline IL-10 DuoSet ELISA kits (R&D Systems) according to the manufacturer's instructions in triplicate technical replicates per condition from each donor. Absorbance was measured at 450 nm on a BioTek Synergy H1 plate reader (Agilent). Data were analyzed and graphed in Prism, version 10 (GraphPad).
1928Z-LNGFR CAR vector cloning
A Fragment Gene (Azenta Life Sciences) was designed encoding the second-generation anti-human CD1928Z-LNGFR CAR sequence19 (provided by M. Sadelain, Memorial Sloan Kettering Cancer Center) and a Kozak consensus sequence GCCACCATGG. 5’ PmeI and 3’ BamHI sites were added by PCR with Pfu Ultra II Fusion HS polymerase (Agilent). pTiger (Addgene #1728) and the 1928Z-LNGFR gene fragment were digested with PmeI and BamHI-HF (New England Biolabs [NEB]), run on a 1% agarose gel, and extracted with the GeneJET gel extraction kit (Thermo) before subsequent ligation and transformation into NEB Stable Escherichia coli (NEB). A cloned plasmid was selected from a single colony, expanded, and purified using EndoFree Plasmid Maxi Kit (Qiagen). The sequence was confirmed by whole-plasmid sequencing (Plasmidsaurus) and annotated with PLannotate, version 1.2.2.20
Cell line culture
HEK293T cells (provided by H. Aguilar-Carreno, Cornell) and Raji B cells (provided by K. Richards, Cornell) were cultured in Dulbecco Modified Eagle Medium (Corning) or RPMI 1640 (Corning), respectively, and supplemented with 10% fetal bovine serum, 2 mM L-glutamine, 1% penicillin-streptomycin (Corning), 2.5 μg/mL amphotericin B (Gibco), 1 mM sodium pyruvate (Corning), and 10 mM HEPES (Corning).
Primary T-cell lentiviral transduction
The day before transfection, 2.5 X 106 HEK293T cells per plate were seeded onto duplicate 10-cm tissue culture plates. The next day, medium was exchanged to 8.8 mL per plate complete Dulbecco Modified Eagle Medium without antibiotics 2 hours prior to transfection. Seven and a half μg pTiger-1928Z-LNGFR, 12.5 μg CF1ΔEnv FIV packaging vector21 (provided by E. Poeschla, U Colorado), and 5 μg HDM-VSV-G (provided by G. Mostoslavsky, Boston University) were added to 1.1 mL of Opti-MEM (Gibco) in duplicate Eppendorf tubes prior to adding 75 μL of FUGENE HD (Promega) to each tube and incubating for 15 minutes before adding to cells. Fresh media was exchanged 24 hours later. Forty-eight- and 72-hour supernatants were spun at 2,000 X g at 4 °C and clarified through a 0.45-μm polyethersulfone syringe filter (Nest Scientific). Clarified supernatants were pooled and concentrated at 100,000 X g for 90 minutes at 4 °C in an Optima XPN-80 ultracentrifuge (Beckman Coulter) using a SW32 Ti rotor and ultracentrifuge tubes (Beckman Coulter). The virus pellet was resuspended in 400 μL basal XVIVO-15 without additives, aliquoted, and stored at −80 °C. Lentivirus was titered using HEK293T and calculated based on LNGFR expression by flow cytometry.22 1 X 106 CD3-enriched primary feline T cells expanded in G-Rex plates were resuspended in a total volume of 500 μL of basal XVIVO-15 with 10 ng/mL rfIL-2 and 8 μg/mL polybrene (Sigma) with or without lentivirus (Multiplicity of infection ≈ 1) and aliquoted to 1 well of a 24-well tissue culture plate. Cells were spinoculated at 400 X g for 2 hours at 32 °C then placed in the 37 °C, 5% CO2 incubator for another 2 hours. Cells were then transferred to 1 well of a G-Rex 24-well plate in 8 mL total volume complete XVIVO-15 with 10 ng/mL rfIL-2.
Cytotoxicity assay
Raji wild-type (WT) or CD19-deficient target cells (generated using clustered regularly interspaced short palindromic repeats/clustered regularly interspaced short palindromic repeats–associated protein 9; Supplementary Methods) were stained with 5 μM CellTrace Violet. 2 X 105 mock- or lentivirus-transduced effector T cells (51% to 77% transduction) were seeded with 2 X 104 target cells in a total of 200 μL complete XVIVO-15 with 10 ng/mL rfIL-2 per well of a 96-well V-bottom plate (Nest Scientific). Cells were cocultured for 4 hours in triplicate for both target cell lines. Target cells cultured alone were used to determine spontaneous death. The cells were washed with DPBS, stained with LIVE/DEAD Near IR, washed with flow cytometry buffer, fixed with Cytofix/Cytoperm, washed 2 more times, and resuspended in 200 μL flow buffer. Viability was determined by flow cytometry using an Attune NxT CytKick Max Autosampler (Thermo) recording 1 X 104 single CellTrace Violet+ events per well. The percentage of cytotoxicity was calculated by subtracting the mean percentage of spontaneous dead targets (CellTrace Violet+LIVE/DEAD+) from the percentage dead in coculture for both respective target cell lines.
Statistical analyses
All statistical analyses were performed using Prism, version 10. A Shapiro-Wilk test was used to confirm normality. Comparisons of 2 data sets were analyzed by unpaired t tests with Welch correction for unequal SDs. Comparisons of 3 or more data sets were analyzed by 2-way ANOVA with a Tukey multiple comparisons test. Significance was defined as P < .05. Statistically significant outliers from quantitative PCR data were identified using GraphPad outlier calculator and excluded prior to normality testing. All data points were included for other experiments. P values are summarized as *P < .05; **P < .01; ***P < .001; and ****P < .0001.
Results
Concanavalin A and PMA/I are potent stimuli for feline T-cell proliferation and activation
We first isolated CD3+ T cells from peripheral blood mononuclear cells collected from healthy feline donors. Prior to MACS isolation, CD3+ cells were 24.2%, 53.2%, and 37.3% from 3 donors, in line with previously reported frequencies detected with this antibody.18 Following MACS, purity increased to 98.7%, 97.6%, and 97.5%, respectively (Figure 1). Human and canine primary T cells are conventionally activated and expanded with CD3/CD28-activating antibodies,12,17,23,24 and although we obtained a feline CD3 antibody, feline CD28 antibodies are not commercially available. Thus, we tested alternative conditions previously demonstrated to expand T cells from other species as well as feline T cells, which included IL-2, IL-21, ConA, or PMA/I with cells from 3 independent donors.12,14,25,26 Cells treated with either rfIL-2 or rfIL-2 with rhIL-21 maintained the highest viability, which was similar from 48 to 96 hours (rfIL-2, 61.5% to 86.4%; rfIL-2 with rhIL-21, 74.8% to 86%; Figure 1). Cells treated with rfIL-2 plus ConA had an average viability of 79.3% for the first 48 to 72 hours, but viability declined at 96 hours (41.6% to 58.4%; Figure 1). Conversely, the viability of T cells cultured in rfIL-2 plus PMA/I was low at 48 hours but increased by the 96-hour timepoint (59.1% to 66.2%; Figure 1). We next measured T-cell proliferation. Gating on CD4+ and CD8+ T cells, rfIL-2 did not induce proliferation of either subset, whereas the addition of rhIL-21 induced proliferation at 72 hours within a small subset of CD4+ and a slightly larger subset of CD8+ T cells (Figure 1). Recombinant feline IL-2 plus ConA or PMA/I resulted in at least 1 round of division by 72 hours, with multiple rounds of division detected at 96 hours in both CD4+ and CD8+ populations (Figure 1). To determine the level of activation conferred by each condition throughout the time course, we measured the expression of the T-cell activation marker CD134, also known as OX40.27,28 rfIL-2 alone or with rhIL-21 yielded minimal CD134 expression in both CD4+ and CD8+ populations, whereas significantly higher expression was induced following exposure to ConA or PMA/I at 48 hours in both populations (Figure 1). While CD134 expression remained relatively unchanged in rfIL-2 plus ConA throughout the 96-hour time course, rfIL-2 plus PMA/I treatment induced a time-dependent increase in CD134 expression in both subsets and resulted in the highest overall expression of CD134 in the different CD8+ cell treatment conditions (Figure 1). Together, these results demonstrate the utility of using rfIL-2 in combination with ConA or PMA/I for expanding and activating feline T cells.
Concanavalin A (ConA) and phorbol myristate acetate plus ionomycin (PMA/I) are potent stimuli for feline T-cell proliferation and activation. Peripheral blood mononuclear cells from whole-blood samples were stained with CellTrace Violet proliferation dye, and T cells were enriched by magnetic activated cell sorting for CD3 followed by flow cytometry analysis. A—CD3 expression on peripheral blood mononuclear cells (pre) and enriched T cells (post) on day 0; n = 3 donors. B–D—Enriched T cells were incubated with recombinant feline IL-2 (rfIL-2) alone or with recombinant human IL-21, ConA, or PMA/I for 24 hours. Medium was exchanged to remove mitogenic stimuli, rfIL-2 (for rfIL-2 alone, ConA, and PMA/I) or rfIL-2 with recombinant human IL-21 was added, and cells were cultured for an additional 3 days. B—Percentage of viable cells stained with LIVE/DEAD dye on day 0 (untreated) or at each timepoint. Representative of n = 3 donors. C—CellTrace Violet dilution in viable, single CD4+ or CD8+ cells. Representative of n = 3 donors. D—CD134 geometric mean fluorescence intensity of viable, single CD4+ or CD8+ cells. Mean ± SD; n = 3. Two-way ANOVA with Tukey multiple comparisons test was used to compare treatment conditions at each timepoint. *P < .05; ***P < .001; ****P < .0001. ns = Not significant.
Citation: American Journal of Veterinary Research 86, 1; 10.2460/ajvr.24.08.0247
Concanavalin A and PMA/I combined with polarizing cytokines generate feline THelper subsets
Next, we aimed to determine if we could replicate the conditions previously used with murine T cells to generate TH1 and Tregs,29 2 CD4+ T-cell subsets with major roles in controlling viral infections and autoimmunity, respectively, and that could be used to augment CAR T-cell function in disease-specific contexts.30 We cultured CD3-enriched T cells in ConA or PMA/I in combination with rfIL-2 alone (TH0), rfIL-2 with IL-12 (TH1), or rfIL-2 with rhTGF-β1 (Treg) for 6 days. In TH1-polarizing cytokines with ConA treatment, there was minimal upregulation of the TH1 driver transcription factor Tbet, which was similar with PMA/I treatment (Figure 2). Concanavalin A induced a slight increase in Foxp3, a driver transcription factor for Treg, in cells cultured with rhTGF-β1, but PMA/I induced an almost 7.5-fold higher increase in expression relative to ConA (Figure 2). These data contrast with a previous report where Foxp3 was upregulated in CD4+CD25− cells treated with ConA and rhTGF-β131 but are consistent with our observations that PMA/I induces more proliferation and activation of feline T cells. Importantly, TH1-polarizing conditions resulted in a marked increase in Ifng gene levels in cells cultured in either ConA or PMA/I. Again, the relative levels were around 4.5-fold higher in cells treated with PMA/I (Figure 2). Regulatory T-cell polarizing conditions induced little change in Il10 gene expression for ConA-treated cells, whereas PMA/I-treated cells had an average 18-fold induction in Il10 gene expression (Figure 2).
Primary feline T cells polarized with T helper cytokine milieu upregulate lineage cytokine expression and secretion. Enriched primary feline T cells were stimulated with rfIL-2 and either ConA or PMA/I alone (TH0) or in combination with either recombinant feline IL-12 (TH1) or recombinant human TGF-β1 (Treg) for 24 hours, then media was exchanged and cells cultured with rfIL-2 and either recombinant feline IL-12 or recombinant human TGF-β1 for an additional 5 days. Cells were harvested on day 6, lysed with TRIzol reagent, reverse transcribed into cDNA, and analyzed by quantitative PCR for (A) lineage transcription factors or (B) lineage cytokines. Minimum to maximum with mean of technical replicates indicated from n = 4 to 6 donors. Data were analyzed by unpaired t tests with Welch correction for unequal SDs comparing ConA to PMA/I. Culture supernatant from each condition for each donor was harvested on day 6, pooled, and analyzed in triplicate for (C) interferon-γ (IFN-γ) or (D) IL-10 secretion (pg/mL). Each data point represents the mean of triplicate technical replicates from n = 3 individual donors from 2 independent experiments. Data were analyzed by unpaired t tests with Welch correction for unequal SDs comparing TH0 to TH1 or Treg conditions, respectively. *P < .05.
Citation: American Journal of Veterinary Research 86, 1; 10.2460/ajvr.24.08.0247
We next tested whether these polarized subsets also secreted their respective lineage-associated cytokines. Both ConA and PMA/I resulted in marked secretion of interferon-γ in the TH1 condition compared to the TH0 and Treg conditions, though the relative secretion level with PMA/I was 48- to 236-fold of that induced by ConA (Figure 2). Interleukin-10 secretion from cells in the Treg condition was similar with ConA and PMA/I, and the PMA/I, but not the ConA, Treg condition induced much more IL-10 than the TH0 and TH1 conditions (Figure 2). Thus, the culture of primary feline T cells with recombinant feline IL-12 or rhTGF-β1 induced transcription factors and cytokines consistent with TH1 and Treg CD4+ subsets.
Phorbol myristate acetate with ionomycin yields greater ex vivo feline T-cell expansion than ConA
A critical aspect of cell-based immunotherapy is the ex vivo expansion of therapeutically relevant cell numbers for infusion into patients. Since PMA/I- and ConA-stimulated T cells had similar viability at 96 hours and similar proliferation profiles, we next asked which condition would result in greater overall T-cell expansion using the G-Rex system. Seeding 1 million T cells per well and culturing with rfIL-2 and ConA resulted in expansion of up to 18 million cells within 12 days, whereas rfIL-2 and PMA/I yielded nearly double the number of cells (31 million; (Figure 3). The results were consistent across 4 independent donors where PMA/I led to a greater number of T cells. These data demonstrate the potential to expand primary feline T cells ex vivo into clinically relevant numbers and the comparatively better yield achieved with PMA/I.
PMA/I stimulation yields greater ex vivo expansion of primary feline T cells than ConA. 1 X 106 T cells were seeded in 1 mL per well of a G-Rex 24-well plate and stimulated with either rfIL-2 and ConA or PMA/I. At 24 hours, 7 additional mL of media with rfIL-2 was added, and cells were counted on the indicated days. Two-way ANOVA with a Tukey multiple comparisons test, n = 4 donors, was used to compare ConA and PMA/I at each timepoint. Lines connect data from the same donor. *P < .05; ***P < .001.
Citation: American Journal of Veterinary Research 86, 1; 10.2460/ajvr.24.08.0247
Generation of functional primary feline CAR T cells
Protocols to generate human CAR T cells using lentivirus have been well documented (reviewed in Labbé32), and FIV-vectored lentiviral systems have been used to transduce feline cells originating from a variety of tissues, including bone marrow cells as well as a feline T-cell line.21,33,34 We next tested whether a 3-vector FIV lentiviral system could efficiently transduce primary feline T cells on day 4 of PMA/I expansion in the G-Rex plate to express a human CD19 CAR with CD28 and CD3ζ cytoplasmic domains (1928Z). Seventy-two hours post-transduction, there was a roughly 2:1 ratio of CD4 to CD8 T cells. We tracked transduction using the coexpressed LNGFR marker (Figure 4). Compared to mock transduction with polybrene alone (Mock), transduction with 1928Z lentivirus resulted in an average of 60.4% LNGFR expression in CD4+ T cells and 68.2% in CD8+ T cells, which was consist across 3 independent donors (52.5% to 65.5% CD4+ and 51.7% to 77.2% CD8+; Figure 4).
Primary feline CAR T cells generated by lentiviral transduction. A—Schematic of primary feline T-cell expansion, transduction, and cytotoxicity assay. Enriched primary feline T cells were stimulated in a G-Rex 24-well plate with rfIL-2 and PMA/I for 24 hours in 1 mL total volume before diluting media to 8 mL with rfIL-2 for an additional 3 days. 1 X 106 T cells were either left untransduced (UTD), or transduced with either polybrene alone (Mock) or in combination with hCD19–28Z-LNGFR lentivirus (1928Z). Image prepared in BioRender. Cells were expanded in a G-Rex well before harvesting for (B) flow cytometry analysis of transduced and expanded viable, single CD4+ or CD8+ cells for LNGFR expression (transduction efficiency marker). C—Wild-type (WT) and CD19 knockout (CD19KO) Raji cells were stained for CD19 and analyzed by flow cytometry. D—2 X 105 Mock-transduced T cells or 1928Z feline CAR T cells were incubated with 2 X 104 WT or CD19KO target cells, and cytotoxicity was measured after 4 hours by staining with LIVE/DEAD dye and analyzing by flow cytometry. Mean ± SD from n = 3 donors. Two-way ANOVA with a Tukey multiple comparisons test across conditions. *P < .05; ****P < .0001. CMV = Cytomegalovirus promoter. FITC = Fluorescein isothiocyanate. LNGFR = Truncated nerve growth factor receptor. PE = Phycoerythrin. RRE = Rev response element. WPRE = Woodchuck hepatitis virus post-transcriptional regulatory element.
Citation: American Journal of Veterinary Research 86, 1; 10.2460/ajvr.24.08.0247
We next tested whether the transduced feline CAR T cells were functional and antigen specific. We generated CD19-deficient Raji cells (CD19KO) (Figure 4), which were used in parallel with CD19+ wild-type (WT) Raji cells as targets in a cytotoxicity assay. Mock transduction with polybrene alone feline T cells demonstrated low feline anti-human xenogeneic activity with less than 10% cytotoxicity against WT or CD19KO targets (Figure 4). 1928Z CAR T cells demonstrated significantly higher cytotoxicity against specific targets compared to Mock-transduced T cells, confirming that the CAR was functional. Moreover, 1928Z CAR T cells also had significantly higher cytotoxicity against specific targets (WT) compared to nonspecific targets (CD19KO), demonstrating CAR specificity. 1928Z CAR T cells exhibited slightly higher activity toward nonspecific targets compared to Mock-transduced T cells, suggesting that there may be an elevated level of baseline activation. Together, these data demonstrate the generation of functional and specific feline CAR T cells.
Discussion
This study describes the conditions for feline T-cell expansion and engineering for use in CAR T-cell immunotherapy. We successfully enriched primary T cells to > 97% purity, activated them, and expanded them ex vivo. We consistently observed the greatest ex vivo expansion and activation using rfIL-2 plus PMA/I. We also describe a method to transduce expanded primary feline T cells with lentivirus to express a specific CAR. These feline CAR T cells demonstrated specific cytotoxicity against target cells.
Although CAR T-cell therapy has been most widely investigated to treat cancers, a growing interest has developed in repurposing it for other diseases (reviewed in Maldini et al35). Multiple groups have reported preclinical studies using CAR T cells to treat viral infections, such as HIV, hepatitis C, and SARS-CoV-2.5–8 CD19 CAR T cells have also been utilized to eliminate autoreactive B cells in human patients with systemic lupus erythematosus.2–4 Chimeric antigen receptor T therapy for both viral infection and autoimmunity could be enhanced by taking advantage of the natural function of TH1 or Treg phenotypes, respectively. We showed that after 6 days with ConA or PMA/I plus polarizing cytokines, feline T-cell cultures upregulated expression of the respective lineage transcription factors and cytokines and secreted those cytokines. Although culturing in TH1 cytokines, rfIL-2 plus rfIL-12, slightly increased Tbet expression, these cytokines induced a larger increase in Ifng expression, which was higher with PMA/I than ConA. The disconnect in lineage transcription factor and cytokine expression could be due to the harvest timepoint as it is possible that Tbet expression is upregulated early and then decreases back to basal levels before day 6. Importantly, the upregulation of Ifng correlated with cytokine secretion. The differences in polarization between ConA and PMA/I stimulation could be attributed to the fact that PMA/I confers both signal 1 (T-cell receptor signaling) and signal 2 (costimulation), whereas ConA only confers signal 1. Concanavalin A induced nearly equal to or higher IL-10 secretion in the TH1 condition compared to the Treg condition, which consisted of rfIL-2 plus rhTGF-β1. This could be indicative of the T cells receiving proinflammatory signal 1 and signal 3 (cytokine stimulation) without signal 2 as PMA/I stimulation resulted in greater upregulation of Il10 and Foxp3 and secretion of IL-10. Although the individual levels of cytokine produced were variable, the trends of secretion between polarizing conditions were consistent between donors. Of note, our polarizing cultures also contained CD8+ T cells, which might also respond to the polarizing cytokines and have an impact on CD4+ T cell differentiation. Future studies could replicate our experiments by first isolating CD4+ T cells. However, the polarized cells would then have to be characterized for lineage-specific gene signatures at the single cell level, lineage markers by flow cytometry, or in vitro functional validation to confirm polarization. Our difference in culture conditions might explain the discrepancy in our ConA Treg polarization condition results compared to a previous study31 that stimulated sorted CD4+CD25− cells with ConA and rhTGF-β1 and observed marked upregulation of Foxp3 as well as suppressor function in vitro. That study also utilized a higher rhTGF-β1 dose (10 ng/mL) compared to ours (2 ng/mL). Nonetheless, our results demonstrate that primary feline T cells can be polarized into proinflammatory or regulatory subsets to a higher degree with PMA/I than ConA.
To generate a cell-based immunotherapy, the cells must be expanded ex vivo into clinically relevant numbers. One expansion platform that has been widely adopted in academia and industry is the G-Rex system, which promotes maximum cell growth through enhanced gas exchange. We found that rfIL-2 plus PMA/I resulted in up to a 30-fold expansion by day 12 (Figure 3; 4 donors), whereas rfIL-2 plus ConA led to up to an 18-fold expansion. These data are consistent with a previous study26 showing that PMA/I-treated cells had a higher stimulation index compared to ConA-treated cells. Another study25 noted that PMA/I failed to induce feline T-cell stimulation compared to ConA, but this could be attributed to toxicity from the high dose of PMA used in that study (10 to 15 μg/mL compared to 25 ng/mL in our study). Our cells remained viable during expansion, could be genetically engineered to express a CAR, and retained function.
Though protocols for CAR T-cell generation have been established for human and, more recently, canine, none have been published to date for feline. Chimeric antigen receptor T cells are typically generated through transduction with γ-retrovirus or lentivirus encoding for the CAR and any transduction markers (reviewed in Levine9). Previous studies21,33,34 have reported success using FIV vectors to transduce feline cell lines and primary cells. Although cells from many different tissues, including a feline T-cell line, were successfully transduced, none of the studies reported transduction of enriched primary feline T cells. Here, we transduced primary T cells with an FIV-based lentivirus encoding a human CD19 CAR and truncated human nerve growth factor receptor transduction marker and observed efficient transduction of both CD4+ and CD8+ T-cell populations. Although these transgenes are under the control of a cytomegalovirus immediate-early enhancer/promoter that has been reported to have poor function in feline cells transduced with lentivirus,34 we observed surface expression of LNGFR in transduced cells and cytotoxicity by CAR-expressing cells against specific target cells. Reduced cytomegalovirus promoter activity in T cells may occur due to silencing over time36,37; thus, future development may include using a different promoter for CAR expression in primary feline T cells. The CF1ΔEnv packaging vector20 we used to generate lentivirus encodes FIV vif in addition to the necessary gag, pol, and rev elements. FIV vif has been shown to degrade feline APOBEC3, a host restriction factor to lentivirus infection.38,39 Future studies would need to compare transduction with lentiviruses packaged with and without vif to determine its importance in transducing feline T cells. In our hands, feline CAR T cells demonstrated significant cytotoxicity toward specific target cells compared to mock-transduced T cells. This demonstrates that human signaling domains are functional in feline T cells. We also observed low but significant CAR T-cell killing of nonspecific targets as compared to mock-transduced T cells. This nonspecific activity was significantly less than CAR T-cell killing of specific target cells. The nonspecific killing by CAR T cells may be due to the elevated activation state imparted by tonic CD28 signaling emanating from the CAR. Future studies should include developing a fully feline CAR with alternative costimulatory domains to reduce background signaling and potential rejection in vivo in immune competent cats.
Here, we provide a proof of concept for generating functional primary feline CAR T cells. To fully capitalize on feline CAR T therapy for specific diseases, additional work is needed to generate monoclonal antibodies toward target antigens and develop corresponding single-chain variable fragments. Human T-cell activation and expansion for clinical use is most often induced by CD3/CD28 activation beads, so to develop good manufacturing practice–grade feline CAR T cells, similar feline-specific activation beads will also need to be developed and validated. Despite these limitations, our study provides the first step in advancing feline immunotherapy to improve the health of our feline companions.
Supplementary Materials
Supplementary materials are posted online at the journal website: avmajournals.avma.org.
Acknowledgments
We would like to thank Dr. Carolyn McDaniel for providing access to her colony of teaching cats.
Disclosures
The authors have nothing to disclose. No AI-assisted technologies were used in the generation of this manuscript.
Funding
James R. Cockey was supported by the Liz Hanson Fellowship from the Cornell Feline Health Center. This material is based upon work supported by the Cornell Feline Health Center Research Grants Program, a grant made available to the College of Veterinary Medicine, Cornell University.
ORCID
J. R. Cockey https://orcid.org/0000-0001-6227-2077
C. A. Leifer https://orcid.org/0000-0003-0987-9747
References
- 1.↑
Approved cellular and gene therapy products. US FDA. Updated November 8, 2024. Accessed November 13, 2024. https://www.fda.gov/vaccines-blood-biologics/cellular-gene-therapy-products/approved-cellular-and-gene-therapy-products
- 2.↑
Krickau T, Naumann-Bartsch N, Aigner M, et al. CAR T-cell therapy rescues adolescent with rapidly progressive lupus nephritis from haemodialysis. The Lancet. 2024;403(10437):1627–1630. doi:10.1016/S0140-6736(24)00424-0
- 3.
Wang W, He S, Zhang W, et al. BCMA-CD19 compound CAR T cells for systemic lupus erythematosus: a phase 1 open-label clinical trial. Ann Rheum Dis. 2024;83(10):1304–1314. doi:10.1136/ard-2024-225785
- 4.↑
Mackensen A, Müller F, Mougiakakos D, et al. Anti-CD19 CAR T cell therapy for refractory systemic lupus erythematosus. Nat Med. 2022;28(10):2124–2132. doi:10.1038/s41591-022-02017-5
- 5.↑
Guo X, Kazanova A, Thurmond S, Saragovi HU, Rudd CE. Effective chimeric antigen receptor T cells against SARS-CoV-2. iScience. 2021;24(11):103295. doi:10.1016/j.isci.2021.103295
- 6.
Maldini CR, Claiborne DT, Okawa K, et al. Dual CD4-based CAR T cells with distinct costimulatory domains mitigate HIV pathogenesis in vivo. Nat Med. 2020;26(11):1776–1787. doi:10.1038/s41591-020-1039-5
- 7.
Anthony-Gonda K, Bardhi A, Ray A, et al. Multispecific anti-HIV duoCAR-T cells display broad in vitro antiviral activity and potent in vivo elimination of HIV-infected cells in a humanized mouse model. Sci Transl Med. 2019;11(504):eaav5685. doi:10.1126/scitranslmed.aav5685
- 8.↑
Sautto GA, Wisskirchen K, Clementi N, et al. Chimeric antigen receptor (CAR)-engineered T cells redirected against hepatitis C virus (HCV) E2 glycoprotein. Gut. 2016;65(3):512–523. doi:10.1136/gutjnl-2014-308316
- 9.↑
Levine BL. Performance-enhancing drugs: design and production of redirected chimeric antigen receptor (CAR) T cells. Cancer Gene Ther. 2015;22(2):79–84. doi:10.1038/cgt.2015.5
- 10.↑
Guedan S, Calderon H, Posey AD, Maus MV. Engineering and design of chimeric antigen receptors. Mol Ther Meth Clin Dev. 2019;12:145–156. doi:10.1016/j.omtm.2018.12.009
- 11.↑
Sadelain M, Brentjens R, Rivière I. The basic principles of chimeric antigen receptor design. Cancer Discov. 2013;3(4):388–398. doi:10.1158/2159-8290.CD-12-0548
- 12.↑
Panjwani MK, Smith JB, Schutsky K, et al. Feasibility and safety of RNA-transfected CD20-specific chimeric antigen receptor T cells in dogs with spontaneous B cell lymphoma. Mol Ther. 2016;24(9):1602–1614. doi:10.1038/mt.2016.146
- 13.↑
Sakai O, Igase M, Mizuno T. Optimization of canine CD20 chimeric antigen receptor T cell manufacturing and in vitro cytotoxic activity against B-cell lymphoma. Vet Comp Oncol. 2020;18(4):739–752. doi:10.1111/vco.12602
- 14.↑
Panjwani MK, Atherton MJ, MaloneyHuss MA, et al. Establishing a model system for evaluating CAR T cell therapy using dogs with spontaneous diffuse large B cell lymphoma. Oncoimmunology. 2020;9(1):1676615. doi:10.1080/2162402X.2019.1676615
- 15.↑
Sakai O, Yamamoto H, Igase M, Mizuno T. Optimization of culture conditions for the generation of canine CD20-CAR-T cells for adoptive immunotherapy. In Vivo. 2022;36(2):764–772. doi:10.21873/invivo.12763
- 16.↑
Zhang S, Black RG, Kohli K, et al. B7-H3 specific CAR T cells for the naturally occurring, spontaneous canine sarcoma model. Mol Cancer Ther. 2022;21(6):999–1009. doi:10.1158/1535-7163.MCT-21-0726
- 17.↑
Yin Y, Boesteanu AC, Binder ZA, et al. Checkpoint blockade reverses anergy in IL-13Rα2 humanized scFv-based CAR T cells to treat murine and canine gliomas. Mol Ther Oncolytics. 2018;11:20–38. doi:10.1016/j.omto.2018.08.002
- 18.↑
Nishimura Y, Shimojima M, Sato E, et al. Downmodulation of CD3epsilon expression in CD8alpha+beta- T cells of feline immunodeficiency virus-infected cats. J Gen Virol. 2004;85(Pt 9):2585–2589.
- 19.↑
Zhao Z, Condomines M, van der Stegen SJC, et al. Structural design of engineered costimulation determines tumor rejection kinetics and persistence of CAR T Cells. Cancer Cell. 2015;28(4):415–428. doi:10.1016/j.ccell.2015.09.004
- 20.↑
McGuffie MJ, Barrick JE. pLannotate: engineered plasmid annotation. Nucleic Acids Res. 2021;49(W1):W516–W522.
- 21.↑
Poeschla EM, Wong-Staal F, Looney DJ. Efficient transduction of nondividing human cells by feline immunodeficiency virus lentiviral vectors. Nat Med. 1998;4(3):354–357. doi:10.1038/nm0398-354
- 22.↑
Saenz DT, Barraza R, Loewen N, Teo W, Poeschla EM. Titration of feline immunodeficiency virus-based lentiviral vector preparations. Cold Spring Harb Protoc. 2012;2012(1):126–128. doi:10.1101/pdb.prot067561
- 23.↑
Rotolo A, Atherton MJ, Kasper BT, Haran KP, Mason NJ. Genetic re-direction of canine primary T cells for clinical trial use in pet dogs with spontaneous cancer. STAR Protoc. 2021;2(4):100905. doi:10.1016/j.xpro.2021.100905
- 24.↑
Gagliardi C, Khalil M, Foster AE. Streamlined production of genetically modified T cells with activation, transduction and expansion in closed-system G-Rex bioreactors. Cytotherapy. 2019;21(12):1246–1257. doi:10.1016/j.jcyt.2019.10.006
- 25.↑
Cridge H, Kordon A, Pinchuk LM, et al. Effects of cyclosporine on feline lymphocytes activated in vitro. Vet Immunol Immunopathol 2020;219:109962. doi:10.1016/j.vetimm.2019.109962
- 26.↑
Aronson LR, Drobatz KJ, Hunter CA, Mason N. Effects of CD28 blockade on subsets of naïve T cells in cats. Am J Vet Res. 2005;66(3):483–492. doi:10.2460/ajvr.2005.66.483
- 27.↑
Shimojima M, Miyazawa T, Ikeda Y, et al. Use of CD134 as a primary receptor by the feline immunodeficiency virus. Science. 2004;303(5661):1192–1195. doi:10.1126/science.1092124
- 28.↑
Mallett S, Fossum S, Barclay AN. Characterization of the MRC OX40 antigen of activated CD4 positive T lymphocytes–a molecule related to nerve growth factor receptor. EMBO J. 1990;9(4):1063–1068. doi:10.1002/j.1460-2075.1990.tb08211.x
- 29.↑
Sekiya T, Yoshimura A. In vitro Th differentiation protocol. In: Feng XH, Xu P, Lin X, eds. TGF-β Signaling. Vol 1344. Springer New York; 2016:183–191. doi:10.1007/978-1-4939-2966-5_10
- 30.↑
Fung VCW, Rosado-Sánchez I, Levings MK. Transduction of human T cell subsets with lentivirus. In: Annunziato F, Maggi L, Mazzoni A, eds. T-Helper Cells. Vol 2285. Springer US; 2021:227–254. doi:10.1007/978-1-0716-1311-5_19
- 31.↑
Petty CS, Tompkins MB, Tompkins WA. Transforming growth factor-β/transforming growth factor-βRII signaling may regulate CD4+CD25+ T-regulatory cell homeostasis and suppressor function in feline AIDS lentivirus infection. J Acquir Immune Defic Syndr. 2008;47(2):148–160. doi:10.1097/QAI.0b013e318160df70
- 32.↑
Labbé RP, Vessillier S, Rafiq QA. Lentiviral vectors for T cell engineering: clinical applications, bioprocessing and future perspectives. Viruses. 2021;13(8):1528. doi:10.3390/v13081528
- 33.↑
Curran MA, Kaiser SM, Achacoso PL, Nolan GP. Efficient transduction of nondividing cells by optimized feline immunodeficiency virus vectors. Mol Ther. 2000;1(1):31–38. doi:10.1006/mthe.1999.0007
- 34.↑
Price MA, Case SS, Carbonaro DA, et al. Expression from second-generation feline immunodeficiency virus vectors is impaired in human hematopoietic cells. Mol Ther. 2002;6(5):645–652. doi:10.1006/mthe.2002.0725
- 35.↑
Maldini CR, Ellis GI, Riley JL. CAR T cells for infection, autoimmunity and allotransplantation. Nat Rev Immunol. 2018;18(10):605–616. doi:10.1038/s41577-018-0042-2
- 36.↑
Milone MC, Fish JD, Carpenito C, et al. Chimeric receptors containing CD137 signal transduction domains mediate enhanced survival of T cells and increased antileukemic efficacy in vivo. Mol Ther. 2009;17(8):1453–1464. doi:10.1038/mt.2009.83
- 37.↑
Jones S, Peng PD, Yang S, et al. Lentiviral vector design for optimal T cell receptor gene expression in the transduction of peripheral blood lymphocytes and tumor-infiltrating lymphocytes. Hum Gene Ther. 2009;20(6):630–640. doi:10.1089/hum.2008.048
- 38.↑
Yoshikawa R, Takeuchi JS, Yamada E, et al. Feline immunodeficiency virus evolutionarily acquires two proteins, vif and protease, capable of antagonizing feline APOBEC3. J Virol. 2017;91(11):e00250–17. doi:10.1128/JVI.00250-17
- 39.↑
Troyer RM, Thompson J, Elder JH, VandeWoude S. Accessory genes confer a high replication rate to virulent feline immunodeficiency virus. J Virol. 2013;87(14):7940–7951. doi:10.1128/JVI.00752-13
