Engineered Immune Cells Targeting Fibroblast Activation Protein Cause Cell Death in Mouse Heart Scar Cells

The mouse cardiac fibroblasts (MCFs; BNCC 340098) and human T lymphocyte cells (Jurkat; Clone E6-1, ATCC TIB-152) cell lines were grown in RPMI-1640 (GIBCO) containing 10% fetal bovine serum (FBS; GIBCO) and 1% penicillin–streptomycin (Biosharp). HEK-293T cells (ATCC CRL-3216) were cultured in high-glucose DMEM (Servicebio) supplemented with 10% FBS and 1% penicillin–streptomycin (Biosharp). All cells were cultured at 37°C with 5% CO₂.

FAP-expressing HEK-293T cells (FAP-293T) were generated by transducing cells with a lentiviral vector (GeneCopoeia) coexpressing murine FAP and eGFP, linked by an internal ribosome entry site (IRES) element, followed by puromycin (3 μg/mL) selection to establish stable expression. FAP overexpression was confirmed by fluorescence microscopy, flow cytometry, and Western blotting. Nontransduced HEK-293T cells (NC) were used as the control group.

Ang II (A800620) and TGF-β (HY-P70648) were purchased from MedChemExpress (Shanghai, China). Cholesterol (C6213) and DSPC (D875630) were obtained from Macklin (Shanghai, China), while DOTAP (132172-61-3) and DSPE-PEG2000-NH2 (474922-26-4) were sourced from AVT (Shanghai, China). Lenti-X Concentrator and RetroNectin were purchased from TaKaRa (Tokyo, Japan). Alexa Fluor 647F(ab⁣′)2 IgG (115-606-003) was from Jackson ImmunoResearch (West Grove, PA). CD247 rabbit polyclonal antibody (12837-2-AP), Collagen Type I monoclonal antibody (66761-1), and Collagen Type III polyclonal antibody (22734-1) were obtained from Proteintech (Wuhan, China). Mouse FAPα antibody was sourced from R&D Systems (Minneapolis, MN, United States). Antibodies including rabbit FAP1 (AF5344), mouse vimentin (BF8006), mouse alpha-SMA (BF9212), rabbit GAPDH (AF7021), rabbit Bax (AF0120), rabbit Bcl-2 (AF6139), rabbit Caspase 3 (AF6311), goat anti-rabbit IgG HRP (S0001), and goat anti-mouse IgG HRP (S0002) were from Affinity Biosciences (Jiangsu, China). Rabbit β-actin (AC038) was acquired from ABclonal Biotechnology. Rabbit Alexa Fluor 647 goat IgG isotype (403006) was acquired from BioLegend (San Diego, CA).

Proliferating MCFs were passaged into culture dishes. When cultures reached 50%–60% confluency, cells were treated with Ang II or TGF-β to induce fibrotic activation. Ang II treatment: The medium was replaced with complete medium (RPMI‐1640 + 10% FBS) containing Ang II after PBS rinse. TGF-β treatment: The medium was replaced with complete medium (RPMI‐1640 + 10% FBS) containing TGF-β after PBS rinse. Control group: The medium was replaced with complete medium (RPMI‐1640 + 10% FBS) without additives. Cells were incubated for 48 h to generate myoFbs.

Two vectors—lentiviral and LNPs—were developed for CAR-Jurkat cells generation. Full-length CAR DNA was synthesized by GenScript and cloned into the pCDH-EF1-MCS vector to create the FAP-CAR-pCDH-EF1-MCS plasmid. This plasmid, along with psPAX2 and pMD2.G packaging plasmids, was cotransfected into HEK-293T cells for lentiviral production. Viral supernatants were collected at 48 and 72 h, concentrated with Lenti-X Concentrator. For LNP vector synthesis, CAR mRNA was synthesized, purified, and encapsulated in LNPs composed of DOTAP, DSPC, cholesterol, and DSPE-PEG2000-NH2 and characterized by transmission electron microscopy (TEM) and dynamic light scattering (DLS).

For lentiviral transduction, plates were first coated with RetroNectin solution and incubated overnight at 4°C. After removing the solution, wells were blocked with sterile 2% BSA for 30 min at room temperature (RT), followed by a PBS wash. Lentiviral particle suspension was added, and plates were centrifuged at 1200 × g for 2 h at 32°C to enhance viral binding. The supernatant was discarded, and target cells were added for incubation. LNPs, confirmed to be nontoxic, were used to transduce Jurkat cells by incubation with a nontoxic LNP concentration for 48 h, generating FAP-CAR-Jurkat cells.

After 48 h of transduction with lentivirus or LNPs, 2 × 10⁵ cells from both the control and transduced groups were collected, washed twice with PBS, and centrifuged. The pellet was resuspended in PBS, and Alexa Fluor 647 AffiniPure F(ab⁣′)₂ IgG antibody was added per the manufacturer's instructions. Cells were incubated on ice for 20–30 min in the dark, then washed twice with PBS, and resuspended in 500 μL of PBS. The suspension was filtered through a 200-mesh nylon filter and analyzed by flow cytometry, with blank and isotype controls for gating.

FAP-293T cells were generated by lentiviral transduction of FAP as described (Section 2.1). Target cells included FAP-293T cells, Ang II or TGF-β induced myoFbs (Section 2.3), NC, and untreated MCFs (nonactivated controls). CAR-Jurkat or nontransduced control Jurkat (NC-Jurkat) cells (non-CAR-transduced) were cocultured with target cells at effector-to-target (E:T) ratios of 1:1, 5:1, and 10:1 in triplicate 12 or 96-well plates. After 24 h, supernatants were collected for IFN-γ and IL-6 quantification via ELISA. Cytotoxicity was assessed by LDH release.

Cytokine levels in supernatants were quantified using commercial sandwich ELISA kits (IFN-γ, MEIMIAN, MM-0033H2; IL-6, MEIMIAN, MM-0049H2) according to the manufacturer's instructions. Briefly, 96-well plates precoated with capture antibodies were incubated with supernatants or standards (diluted in assay buffer) for 2 h at RT. After washing, biotinylated detection antibodies were added for 1 h, followed by streptavidin-HRP (30 min, RT). TMB substrate was used for colorimetric detection (OD450 nm). Standard curves were generated using recombinant cytokines, and sample concentrations were interpolated from linear regression curves (triplicate measurements).

CAR-Jurkat cell cytotoxicity was assessed using a LDH release assay (Beyotime, C0017). CAR-Jurkat or NC-Jurkat and target cells were cocultured at specified E:T ratios for 24 h. Supernatants were mixed with LDH substrate (30 min, dark), and absorbance (490 nm) was measured. Controls included a spontaneous condition (targets alone) and a maximum condition (targets +1% Triton X-100). Cytotoxicity (%) was calculated as (experimental − spontaneous)/(maximum − spontaneous) × 100.

Total protein was extracted and quantified using the BCA Protein Assay Kit (Beyotime). Proteins were separated by SDS-PAGE, transferred to PVDF membranes, and blocked with 5% nonfat milk in TBST for 2 h at RT. The membranes were then incubated with primary antibodies diluted in TBST under gentle agitation at 4°C overnight, followed by incubation with horseradish peroxidase–conjugated secondary antibodies—anti-rabbit (1:10,000, S0001, Affinity Biosciences) or anti-mouse (1:10,000, S0002, Affinity Biosciences)—for 1 h at RT. The following primary antibodies were used: CD247 (rabbit, 1:1500, 12837-2-AP, Proteintech), FAP1 (rabbit, 1:1000, AF5344, Affinity Biosciences), vimentin (mouse, 1:2000, BF8006, Affinity Biosciences), α-smooth muscle actin (α-SMA) (mouse, 1:2000, BF9212, Affinity Biosciences), Collagen Type I (mouse, 1:3000, 66761-1-Ig, Proteintech), Collagen Type III (rabbit, 1:1000, 22734-1-AP, Proteintech), Bax (rabbit, 1:2000, AF0120, Affinity Biosciences), Caspase 3 (rabbit, 1:1500, AF6311, Affinity Biosciences), Bcl-2 (rabbit, 1:1500, AF6139, Affinity Biosciences), GAPDH (mouse, 1:5000, AF7021, Affinity Biosciences), and β-actin (rabbit, 1:30000, AC038, ABclonal Biotechnology). Protein bands were visualized using an ECL reagent (Proteintech) and detected with a chemiluminescence imaging system (GE Healthcare). Band intensities were quantified using ImageJ software, and relative protein expression levels were normalized to GAPDH or β-actin.

Cells (control, Ang II, or TGF-β-treated) were fixed with 4% paraformaldehyde (15 min), permeabilized with 0.1% Triton X-100 (10 min), and blocked with 5% BSA (1 h). Primary antibodies α-SMA (mouse, 1:200, BF9212, Affinity Biosciences), FAP (rabbit, 1:200, A23789, ABclonal), and vimentin (mouse, 1:200, BF8006, Affinity Biosciences) were incubated overnight (4°C), followed by Alexa Fluor–conjugated secondary antibodies (1:500, 1 h, RT). Nuclei were counterstained with DAPI. Images were captured using a fluorescence microscope.

To assess the proliferative capacity of engineered CAR-T cells, CAR-Jurkat cells and negative control-Jurkat cells were seeded in 96-well flat-bottom plates at a density of 1 × 10⁴ cells per well in 100 μL of complete RPMI-1640 medium. On Days 1–5, 10 μL of CCK8 solution (Beyotime, C0038) was added to each well and incubated for 2 h at 37°C. Absorbance was measured at 450 nm using a microplate reader.

To evaluate the cytotoxicity of LNPs, Jurkat T cells were seeded in 96-well plates at 1 × 10⁴ cells per well and treated with increasing concentrations of LNPs (0, 1.25, 2.5, 5, 10, and 20 μg/mL). After 24 h of incubation at 37°C, 10 μL of CCK8 reagent was added to each well, followed by a 2-h incubation. Absorbance at 450 nm was measured to determine cell viability.

Synthesized LNPs were diluted in ultrapure water and deposited onto carbon-coated copper grids (10 μL, 1 min adsorption). Excess liquid was blotted, and grids were negatively stained with 2% phosphotungstic acid (PTA, pH 7.0) for 30 s. After air-drying, samples were imaged using a JEOL JEM-1400Plus TEM operated at 120 kV. Size distribution was analyzed from ≥ 100 particles (ImageJ v1.53).

The hydrodynamic diameter, polydispersity index (PDI, a dimensionless parameter reflecting sample homogeneity; PDI < 0.2 indicates monodisperse particles), and zeta potential of LNPs were measured using a Zetasizer Nano ZS90 (Malvern Panalytical, United Kingdom). Samples were dispersed in deionized water (0.1 mg/mL), sonicated for 3 min, and loaded into disposable polystyrene cuvettes. Measurements were performed at 25°C with a 633 nm laser, 90° scattering angle, and automatic attenuator optimization. Data were acquired as 10 runs (6 s each) per sample, averaged over three independent replicates. Correlation curves were inspected to exclude aggregation.

Data were confirmed to follow a normal distribution. Differences between two groups were assessed using Student's t-test, while comparisons among multiple groups were performed using one-way ANOVA followed by Tukey's post hoc test. All experiments were independently repeated at least three times (biological replicates), and results are presented as mean ± SD. Statistical significance was defined as p < 0.05.

To evaluate the ability of CAR-T cells to target and eliminate FAP-expressing myoFbs, we engineered a second-generation FAP-targeting CAR. This construct comprises an scFv derived from a murine anti-FAP monoclonal antibody (clone 73.3), linked to a human CD8α chain forming the hinge and transmembrane regions, and an intracellular domain containing the human 4-1BB and CD3ζ signaling components (Figure 1a). Additionally, a Regulatory Subunit I Anchor Disturber (RIAD) peptide was incorporated to enhance resistance to adenosine and prostaglandin E2-mediated immunosuppression, thereby improving CAR-T cell efficacy in suppressive environments [24].

We cloned the FAP-CAR nucleic acid sequence into the lentiviral vector pCDH-EF1-MCS and validated the construct via double enzyme digestion (EcoRI and KpnI) (Figure 1b). The resulting pCDH-EF1-MCS-FAP-CAR plasmid was confirmed (OD260/280 = 1.93). Additionally, we synthesized and purified the FAP-CAR mRNA, which featured 101 poly A tails and a cap structure, achieving a high purity of 92.6% (OD260/280 > 1.8 and OD260/OD230 > 2.0) (Figure 1c).

The recombinant expression plasmid pCDH-EF1-MCS-FAP-CAR was cotransfected into HEK-293T cells along with lentiviral packaging plasmids psPAX2 and pMD2.G to obtain lentiviral particles (Figure 2a). The viral suspension was concentrated and then evaluated for lentivirus concentration by measuring HIV-1 P24 antigen content using an ELISA assay (Figure S1 and Table 1).

We synthesized empty LNPs and LNPs loaded with FAP-CAR mRNA using the thin-film hydration method (Table 2). DLS analysis (Figure 2b) showed that both the empty and mRNA-loaded LNPs had particle sizes under 200 nm, with the latter being slightly larger due to mRNA loading. Both LNPs exhibited a PDI (a dimensionless parameter reflecting particle size distribution; PDI < 0.2 indicates monodispersity) around 0.2, indicating good homogeneity, and Zeta potentials above +40 mV. TEM images confirmed that the LNPs were spherical (Figure 2c). LNPs were stored at 4°C, and their particle sizes and PDIs were measured on Days 1, 3, and 7 to assess stability. The results indicated minimal variation in particle size over the first 3 days, suggesting that the LNPs remain stable at 4°C for short-term storage and should be used soon after synthesis (Figure 2d).

To validate the functionality of the synthesized FAP-CAR lentivirus, we first transduced HEK-293T cells—a model system with high lentiviral susceptibility—to confirm CAR expression and viral efficacy before applying the lentivirus to Jurkat cells. Western blot analysis of transduced HEK-293T cells revealed robust CAR protein expression compared to nontransduced controls (Figure). SnapGene software was used to translate the CAR nucleic acid sequence into protein fragments, showing a full-length protein of 621 amino acids (~68 kDa) (Figure). Flow cytometry further demonstrated CAR surface expression in > 40% of transduced 293T cells (Figure), confirming lentiviral functionality. S2A S2B S2C

We then transduced Jurkat cells with the validated FAP-CAR lentivirus at escalating concentrations (9711.37, 19422.74, and 38845.48 pg/mL). Western blot analysis showed significant CAR protein expression in all transduced groups compared to controls, with expression levels positively correlating with lentiviral concentration (Figure 3). Flow cytometry confirmed surface CAR expression across all transduced groups, with the highest positive rate (28.5%) in the highest concentration group (Figure 3). These results demonstrate successful transduction of Jurkat cells. The group with the highest transduction efficiency was selected for expansion. CCK8 assays indicated enhanced proliferation in CAR-Jurkat cells compared to controls, likely due to the inclusion of costimulatory domains (Figure 3). After over 10 passages, the CAR-positive rate remained stable in the highest transduction group (Figure 3b, right), suggesting stable long-term CAR expression in these cells.

A key component of the LNPs synthesized in this study is the cationic lipid DOTAP, widely used in delivery systems [25]. Notably, DOTAP forms the basis of EndoTAG-1, a paclitaxel-encapsulated cationic liposome currently in Phase III clinical trials [26]. However, DOTAP's positive charge can be toxic to normal tissues if too high, making it essential to assess the LNPs' safety. The cytotoxicity of the synthesized LNPs was evaluated by measuring Jurkat cell survival after incubation with LNPs using the CCK8 assay. Results showed that LNPs (1.25–10 μg/mL) maintained Jurkat cell viability > 80%, while 20 μg/mL reduced viability to < 80% (Figure 4a), suggesting mild cytotoxicity at the highest dose. Jurkat cells (5 × 10⁵) were incubated with 2.5, 5, 10, and 20 μg/mL of LNPs/FAP-CAR-mRNA for 48 h. Flow cytometry of CAR-Jurkat cells showed dose-dependent CAR expression from 2.5 to 10 μg/mL LNPs. However, at 20 μg/mL, CAR expression decreased (Figure 4b), potentially due to higher LNPs cytotoxicity at this concentration. LNPs allowed FAP-CAR mRNA to escape into the cytoplasm for translation, resulting in transient expression without genomic integration. After more than 7 days of expansion, flow cytometry revealed that CAR expression was nearly undetectable in all groups, indicating loss of CAR molecules during cell division and expansion (Figure 4c).

HEK-293T cells were stably transduced with a lentiviral vector encoding murine FAP and eGFP, followed by selection with 3 μg/mL puromycin. Fluorescence microscopy revealed > 90% eGFP-positive cells (Figure 5a), and Western blot confirmed markedly elevated murine FAP protein expression compared to controls (Figure 5b). Flow cytometry analysis showed 81.6% of cells expressing eGFP, with 71.8% coexpressing FAP on the surface (Figure 5c), confirming the successful generation of a murine FAP-overexpressing HEK-293T cell line (termed FAP-293T) for downstream CAR-T targeting studies.

To induce fibrotic activation of cardiac fibroblasts, MCFs were treated with either Ang II (0, 0.1, 1, and 10 μM) or TGF-β (0, 5, 10, and 20 ng/mL) for 48 h. Western blot analysis showed a dose-dependent upregulation of fibrosis-associated proteins, including α-SMA, vimentin, FAP, Collagen I, and Collagen III in response to Ang II stimulation, with peak expression observed at 1 μM (Figure 5d). Consistently, immunofluorescence staining confirmed increased vimentin, FAP, and α-SMA expression in the 1 μM Ang II group (Figure 5e). Similarly, treatment with TGF-β led to increased fibrosis-associated protein expression, with the most robust changes seen at 10 ng/mL (Figure S4A). Immunofluorescence staining confirmed increased α-SMA and FAP expression in the 10 ng/mL TGF-β group (Figure S4B). These findings demonstrate that both Ang II and TGF-β treatments can effectively induce fibroblast-to-myoFb transition. Based on the protein expression levels, 1 μM Ang II and 10 ng/mL TGF-β were selected for subsequent experiments to induce fibrotic activation of MCFs and upregulate FAP expression, allowing evaluation of CAR-Jurkat cell targeting efficiency.

CAR-Jurkat cells, stably expressing a FAP-specific CAR via lentiviral transduction, or NC-Jurkat cells, were cocultured with target cells at varying E:T ratios for 24 h. Target cells included FAP-293T cells, Ang II- or TGF-β-induced myoFbs, untreated MCFs, and NC. CAR-T cells eliminate target cells primarily via cytokine secretion, such as IFN-γ, while excessive IL-6 secretion is associated with cytokine release syndrome (CRS), a systemic inflammatory response associated with immune cell overactivation [27]. As shown in Figure 6a, ELISA results demonstrated a significant, E:T ratio-dependent increase in IFN-γ secretion by CAR-Jurkat cells when cocultured with myoFbs. At a 10:1 ratio, IFN-γ levels were significantly higher than those in cocultures of CAR-Jurkat with MCFs or NC-Jurkat with myoFbs, indicating specific activation. In contrast, IL-6 levels remained below 25 ng/L in all groups, suggesting no risk of CRS. Consistently, LDH release assays revealed enhanced cytotoxicity by CAR-Jurkat cells in a ratio-dependent manner, peaking at 10:1 and significantly exceeding levels in control groups (Figure 6b). Confocal imaging (Figure 6c) showed clustering of CAR-Jurkat cells around FAP-positive myoFbs, but not around FAP-negative MCFs, indicating specific recognition. Similar results were observed in cocultures with FAP-293T versus 293T cells (Figure S5). These findings confirm that CAR-Jurkat cells mediate target-specific lysis without detectable nonspecific killing, demonstrating precise antigen-dependent cytotoxicity while maintaining a favorable safety profile. Apoptosis staining by fluorescence microscopy and flow cytometry confirmed significant apoptosis in FAP-positive myoFbs, but not in MCFs (Figure 6d,e). Western blot analysis further revealed that CAR-Jurkat coculture reduced Bcl-2 levels and increased Bax and Caspase 3 expression (Figure 6f), suggesting apoptosis as a mechanism of cytotoxicity.