Footprint-free induced pluripotent stem cells can be successfully differentiated into mesenchymal stromal cells in the feline model

Abstract

Background

Induced pluripotent stem cells (iPSCs) can propagate indefinitely and give rise to every other cell type, rendering them invaluable for disease modelling, drug development research, and usage in regenerative medicine. While feline iPSCs have been described, there are currently no reports on generating genome integration (footprint)-free iPSCs from domestic cats. Therefore, the objective of this study was to generate feline iPSCs from fetal fibroblasts using non-integrative Sendai virus (SeV) vectors carrying human transcription factors. Moreover, these iPSCs were differentiated into mesenchymal stromal cells (MSCs), which can be used as an alternative to tissue-derived MSCs.

Methods

Feline fetal fibroblasts were transduced with CytoTune-iPS 2.0 Sendai Reprogramming vectors at recommended multiplicity of infections (MOI) and cultured for about 6 days. At 7 days post transduction cells were dissociated, replated on inactivated feeder cells and maintained in iPSC medium for 28 days with daily medium change. Emerging iPSC colonies were mechanically passaged and transferred to fresh feeder cells and further passaged every 6–8 days. Four feline iPSC lines were generated, with two selected for further in-depth characterization. Feline iPSCs were then differentiated into MSCs using a serial plating strategy and an inhibitor of the transforming growth factor-β (TGF-β) type I receptor.

Results

Feline iPSCs exhibited characteristic colony morphology, high nuclear-to-cytoplasmic ratio, positive alkaline phosphatase activity, and expressed feline OCT4, SOX2, and Nanog homeobox (NANOG) stem cell markers. Expression of SeV-derived transgenes decreased during passaging to be eventually lost from the host cells and feline iPSCs could be stably maintained for over 35 passages. Feline iPSCs differentiated into embryoid bodies in vitro and did not form fully differentiated teratomas; instead, they generated in vivo masses containing mesodermal tissue derivatives when injected into immunodeficient mice. Feline iPSC-derived MSCs were plastic adherent, displayed MSC-like morphology, expressed MSC-specific surface markers, and differentiated into cells from the mesodermal lineage in vitro. RNA deep sequencing identified 1,189 differentially expressed genes in feline iPSC-derived MSCs compared to feline iPSCs.

Conclusion

We demonstrated the generation of footprint-free iPSCs from domestic cats and their directed differentiation potential towards MSCs. These SeV-derived feline iPSCs and iPSC-derived MSCs will provide valuable models to study feline diseases and explore novel therapeutic strategies and can serve as translational models for human health, leading to increased knowledge on disease pathogenesis and improved therapeutic interventions.

Graphical Abstract

Introduction

Induced pluripotent stem cells (iPSCs) are genetically reprogrammed adult cells that exhibit a pluripotent stem cell-like state and are comparable to embryonic stem cells (ESCs) [1]. iPSCs are regarded as a valuable resource for disease modelling, drug development research, and usage in regenerative medicine, and for human iPSCs in particular, a useful source to avoid the ethical concerns associated with collecting embryos [2]. The first iPSCs were derived from mouse fibroblasts in 2006 [3] and soon after, were created from human fibroblasts [4, 5]. Since then, iPSCs or iPSC-like cells have been generated from diverse mammals, including but not limited to dogs [6, 7], farm animals [8], wild felines [9], primates [10, 11], marsupials [12, 13], and monotremes [14].

Research into domestic cats has both intrinsic and translational benefits. Cats are highly valued as companion animals, with the average US cat owner spending over $1,500 on primary cat care annually [15]. Cats also serve as a powerful animal model for several human diseases and organ-specific research. For instance, domestic cats are susceptible to a wide range of viral, bacterial, and protozoan infections with human homologues and are used to study basic pathogenesis and therapeutic trials [16]. Moreover, with nearly 250 naturally occurring genetic disorders, several of which mirror human conditions [17], domestic cats offer an ideal animal model for human hereditary disease pathology and diagnostic and therapeutic research. Further, cats have been used as models for human obesity and diabetic research for many years [18]. Domestic cats have also contributed substantially to neuroscience research [19] and serve as models for human reproduction [20]. Given the diversity of cat-based research models, feline iPSCs are a valuable tool for studying the biology of cats and modelling disease-specific processes to benefit both human and veterinary medicine.

Several drawbacks must be overcome to fully exploit the therapeutic potential of iPSCs and iPSC-derived cells. When generating iPSCs using traditional viral and non-viral vectors, reprogramming efficiency may be low, reprogramming may be incomplete, and iPSC lines may not be reproducible [21]. In addition, relying on genome-integrating retroviral or lentiviral-based vectors introduces the risk of insertional mutagenesis and malignant cell transformation, the main barriers to employing iPSCs for clinical applications [22]. In contrast, Sendai virus (SeV)-based reprogramming is an efficient and robust approach and viral sequences can be cleared entirely over cell passages, resulting in genome integration (footprint)-free iPSCs, which have been described for human [23], mouse [24], dog [25], and chimpanzee [10], but not for domestic cats. Applying this method to generate feline iPSCs may provide a cell source suitable for advanced research and clinical applications.

Mesenchymal stromal cells (MSCs) represent a heterogeneous population of fibroblast-like, plastic-adherent stem cells capable of self-renewal and differentiation along the mesodermal lineage. They are proposed to be useful for cell-based therapies against numerous inflammatory, degenerative, and immune-mediated diseases in vivo [26, 27]. However, several limitations have been reported. For example, MSCs are rare in adult tissues, requiring in vitro expansion after isolation [28], and based on their limited proliferative potential, it may not be possible to obtain enough cells for clinical use [29]. Moreover, when harvested from older donors, MSCs exhibit an age-related decrease in proliferation rate, capacity to differentiate, and potential to contribute to effective regeneration [30]. Further, variations across MSC cultures due to donor-related heterogeneity and changes accrued during long-term culture have led to variable results in both in vitro and in vivo studies [31], slowing the progression of MSC therapy from research labs to large-scale clinical trials. These limitations have led to an increased interest in obtaining MSCs from alternative sources, including iPSCs, to generate large numbers of reasonably homogenous, well-characterized, and low-passage MSCs. iPSC-derived MSCs have been reported for mouse [32], human [33, 34], dog [6] and horse [35], but not for domestic cats, and were found to closely resemble tissue-derived MSCs in both phenotype and function.

This study reports the first generation of footprint-free iPSCs from domestic cats by SeV reprogramming of fetal fibroblasts using human transcription factors, followed by their directed differentiation into MSCs. These SeV-derived feline iPSCs and iPSC-derived MSCs provide valuable tools for increasing our knowledge on disease pathogenesis, clinical research, and therapeutic applications that can benefit both human and feline health.

Materials and methods

Cell culture

With client consent, domestic cat fetuses (2.5 cm long) were obtained through routine spays of pregnant queens at a local shelter by Dr. Lena DeTar, DVM, DACVPM, DABVP-SMP, from Cornell University. We were not involved in any procedures with live animals, including the use of anesthetics, at any stage. Instead, we received euthanized fetuses from the shelter, with client consent, and animals were further managed by the local shelter following their routine spays. Feline fetal fibroblasts (FFFs) were isolated from four individual fetuses (culture numbers 1–4), as described for mouse embryonic fibroblasts [36]. FFFs at passage (P) 3 were used for reprogramming experiments.

FFFs and transformed mouse fetal fibroblast feeder cells (SNL76/7; Cell Bio Labs, San Diego, CA), were cultured in low-glucose Dulbecco’s Modified Eagle’s medium (DMEM) (Corning Inc., Tewksbury, Massachusetts) supplemented with 10% (v/v) fetal bovine serum (FBS) (Atlanta Biologicals, Flowery Branch, Georgia), 0.1 mM MEM Non-Essential Amino Acid (NEAA) Solution (Cytiva, Marlborough, MA), 6 mM L-glutamine, 1 mM sodium pyruvate, and 1% penicillin/streptomycin (P/S) (all from Corning Inc.). SNL feeders were mitotically inactivated by incubation with 10 µg/mL Mitomycin C (Sigma-Aldrich, St. Louis, MO) for 4 h and subsequently cryopreserved in medium consisting of 70% DMEM, 20% FBS and 10% dimethylsulfoxide (DMSO). Cryopreserved SNL feeder cells were revived at least one day before seeding iPSCs.

iPSC cultures were maintained on inactivated SNL feeder cells in 6-well plates coated with 0.1% gelatin (Sigma-Aldrich) in iPSC medium containing KnockOut DMEM (Thermo Fisher Scientific, Waltham, MA) supplemented with 15% (v/v) Embryonic Stem (ES) qualified FBS (Sigma-Aldrich), 0.1 mM NEAA, 1 mM L-glutamine, 0.1 mM β-mercaptoethanol (MP Biomedicals, Santa Ana, California), 1,000 IU/mL feline leukemia inhibitory factor (fLIF; Kingfisher Biotech, St. Paul, Minnesota) and 6.0 ng/mL human basic fibroblast growth factor (bFGF; Sigma-Aldrich).

iPSC-derived MSCs and feline bone marrow (BM)-derived MSCs, which were isolated from female domestic short-haired cats after euthanasia for reasons unrelated to this study, were maintained in MSC medium that consisted of low glucose DMEM supplemented with 30% FBS, 1% penicillin/streptomycin, and 2 mM L-glutamine.

All cultures were maintained at 37 °C and 5% CO₂.

Generation of feline iPSCs using Sendai virus (SeV) vectors

First, FFFs were transduced with CytoTune EmGFP Sendai Fluorescence Reporter (Thermo Fisher Scientific) at a multiplicity of infection (MOI) of 1, 3 and 5 to validate the cells’ amenability to transduction with SeV vectors. GFP expression was monitored at 24, 48 and 72 h post-transduction. The transduction efficiency percentage was defined as the number of GFP-positive cells compared to the total number of cells.

For reprogramming experiments, CytoTune-iPS 2.0 Sendai Reprogramming Kit (Thermo Fisher Scientific) was used, as per manufacturer’s instructions. Briefly, FFFs (P3) were seeded on 6-well plates at a density of 1.5 × 10⁵ cells/well, 2 days before transduction to achieve approximately 50–60% confluency. Next, FFFs were transduced with SeV vectors at recommended MOIs [polycistronic human (h) KOS, MOI = 5; hc-MYC: MOI = 5, and hKLF4: MOI = 3] in fibroblast medium. One day post transduction (dpt), medium was replaced with fresh fibroblast medium and cells were cultured for an additional 6 days. At 7 dpt, FFFs were dissociated with 0.05% trypsin-EDTA (Corning Inc.) and plated on inactivated SNL feeders grown on six-well plates. Cells were maintained in iPSC medium for 28 dpt, with daily medium change. Emerging iPSC colonies were mechanically removed at approximately 28 dpt and transferred to fresh SNL feeder cells. Colonies were passaged every 6–8 days. A total of four iPSC lines were generated and based on morphology, two were selected for further in-depth characterization.

Generation of feline iPSCs using retroviral (RV) vectors

Retroviral plasmids based on the Moloney murine leukemia virus (MMLV) were purchased from Addgene and contained the coding sequences for the human transcription factors: OCT4 (Addgene #17217), SOX2 (Addgene #17218), NANOG (Addgene #18115), c-MYC (Addgene #17220), and KLF4 (Addgene #17219). For production of MMLV virus, platinum-A packaging cells (Catalog Number: RV-102, Cell Biolabs, San Diego, California) were plated 24 h before transfection in T25 cell culture flasks dishes at a density of 1 × 10⁵ cells/cm², one dish per retroviral vector plasmid. Then, 6 µg of each retroviral vector was transfected to platinum-A cells separately using FuGENE^® HD transfection reagent (Fugene, Middleton, Wisconsin), as per the manufacturers’ instructions. FFFs were plated 24 h before transfection at a density of 1 × 10³ cells/cm² in 10 cm cell culture dishes in fibroblast media containing bFGF (10 ng mL^− 1). Supernatants containing retrovirus particles were collected from the platinum-A packaging cells after 48 h, filtered through a 0.45 μm filter, and supplemented with polybrene (8 µg/mL, Sigma-Aldrich) and bFGF (10 ng/mL). Virus supernatants were combined in equal volumes and used to inoculate FFFs and this process was repeated 24 h later. Twenty-four h after the second retroviral infection, inactivated SNL were added to cell culture dishes at a density of 2.5 × 10⁴/cm² in fibroblast media. This medium was replaced the following day by iPSC medium. Emerging iPSC colonies were mechanically removed around 8–10 dpt and transferred to fresh SNL feeder cells. Colonies were passaged every 5–6 days. A total of four iPSC lines were generated, and based on morphology, two were selected for further in-depth characterization.

Alkaline phosphatase staining

Alkaline phosphatase staining was carried out using the alkaline phosphatase detection kit (Sigma-Aldrich), according to manufacturer’s instructions.

RNA isolation, cDNA synthesis, and analysis of gene expression

Total RNA was extracted using the RNeasy mini kit (Qiagen, Germantown, MD) and reverse transcription (RT) was performed with the iScript cDNA Synthesis Kit (Bio-Rad Laboratories, Hercules, CA), both per manufacturers’ instructions.

Expression of viral transgenes, endogenous pluripotency genes, EB differentiation markers and MSC markers was determined using conventional reverse transcription–polymerase chain reaction (RT-PCR). PCR products were analyzed on 2% agarose gels with GelRed (Biotium, Fremont, CA) and visualized with a Bio-Rad Gel Doc XR + Imaging System. Feline glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as a loading control.

Quantitative PCR (qPCR) analysis was carried out on a QuantStudio3 Real-Time PCR System (Thermo Fisher Scientific) using SYBR Green qPCR Master Mix (Thermo Fisher Scientific). Expression data were normalized to the expression levels of GAPDH. The relative expression levels of selected genes were calculated by the Delta Delta Ct method. Primer sequences and product sizes are listed in Table 1.

Table 1 Primer sequences used in this study

For gene expression in tissue samples, RNA was extracted from formalin-fixed, paraffin-embedded (FFPE) tissue samples, as previously described [37]. Briefly, tissue scrolls were made, and paraffin was removed from FFPE tissue scrolls using xylene (cat#: 9460-11) (VWR, Radnor, Pennsylvania) and EtOH (cat #: E7023-500ML) (Sigma Aldrich). RNA was extracted from paraffin scrolls using the FFPE RNA extraction kit (Qiagen) per the manufacturer’s instructions. cDNA was synthesized and qPCR analysis was carried out as described above.

Karyotyping

Feline iPSCs were incubated overnight at 37 °C in iPSC medium containing 0.2 µg/mL colcemid (Thermo Fisher Scientific) to arrest cells in metaphase. Then, cells were harvested using 0.05% trypsin-EDTA, pelleted by centrifugation at 200 xg for 2 min, resuspended by adding 5 mL of 0.56% KCl dropwise, and incubated for 6 min at room temperature (RT). Cells were pelleted again at 200 xg for 5 min, fixed with 3:1 methanol/acetic acid, and incubated for 5 min at RT. This fixation step was repeated twice. Following the final centrifugation, cells were resuspended in 1 mL of fixative and were dropped from a height of one meter onto glass slides. Chromosomes were stained with DAPI and visualized with an FV3000 confocal microscope (Olympus, Japan). Metaphase spreads from at least 15 cells were counted, with a haploid number of 38 considered normal.

Embryoid body (EB) formation assay

Feline iPSC colonies were dissociated to generate small aggregates of cells and were cultured in ultra-low attachment plates (Corning Inc.) for 14 days in iPSC medium without LIF or bFGF. Culture medium was replaced every other day, floating EBs were harvested, and gene expression was analyzed.

In vivo differentiation

The work has been reported in line with the ARRIVE guidelines 2.0.

Study design: Two experiments were conducted with growth factor reduced (GFR) Matrigel solution (Corning Inc.) and with Laminin 111 (LN111) (BioLamina, Sweden). Each experiment had three groups, and each group had a single animal as the experimental unit. For the first experiment with Matrigel, the three groups were (i) Control group (ii) iPSCs-1 injected with Matrigel and (iii) iPSCs-2 injected with Matrigel. For the second experiment with Laminin, the three groups were (i) Control group (ii) iPSCs-1 injected with LN111 and (iii) iPSCs-2 injected with LN111. Sample size: A total of six mice were used, three mice per experiment and one animal per group as the experimental unit. We used a minimal number of mice to reduce animal use for these experiments. Since we only needed to see the growth of masses, replicates were not used. Inclusion and exclusion criteria: All mice were determined in advance and included in the experiments. One animal from each experimental group was used for the analysis. Randomization: Not applicable. Blinding: The collected masses were analyzed blindly by a pathologist. Outcome measures: The presence and absence of the masses were monitored, and the harvested masses were used for gene expression and histological analysis. Statistical methods: None. Experimental animals: All mice were female immunodeficient NOD-scid gamma (NSG) (NOD.Cg-Prkdcscid Il2rgtm1Wjl/SzJ) mice and were 8-weeks old at the time of injection. Female mice were used as we observed better results with this sex, based on a preliminary small experiment using both male and female mice (data not shown). Mice were housed under standard conditions at the animal facility at the Baker Institute for Animal Health, Cornell University. A veterinarian monitored the mice daily, and proper care was taken to reduce their pain, suffering, and distress, such as providing localized feeding to individual mice to reduce their locomotion. Experimental procedure: All injection procedures were performed in a biological safety cabinet under sterile conditions using sterile materials. Cell suspensions were kept on ice before the injection to avoid solidifying the Matrigel and Laminin. The skin at the injection site was wiped with a disinfectant solution and then with a paper towel soaked with 70% ethanol. For intramuscular injections, mice were restrained using the double-hand method. A sterile 1 ml syringe with a 25G needle with 50µL of injection volume was guided into the hind leg quadriceps along its long axis and toward the muscle center without anesthesia for all experiments. Mice were held for about 5 min after injection to ensure that the cell mixture remained localized to the injection site and to solidify and form an injection plug to enhance the cell engraftment. Eight weeks post-injection, all mice were humanely euthanized, and masses were harvested carefully, excluding any natural bone during mass prosecution. For euthanasia, carbon dioxide asphyxiation using a gradual fill method was used, and mice were observed continuously for the duration of the procedure. Then, masses were fixed in 10% neutral buffered formalin, embedded in paraffin, and stained with hematoxylin and eosin (H&E) for histological analysis.

Mesogenic differentiation of feline iPSCs

Three established protocols were tested to generate feline iPSC-derived MSCs. For the first protocol, EBs were generated and transferred to 6-well plates in MSC medium [38]. Floating EBs adhered to the flask and 5–6 days later, the undifferentiated center of each EB was manually removed, and remaining cells were cultured until they became confluent. After a single cell suspension was generated, cells were seeded directly into T75 cell culture flasks with MSC medium and expanded.

For the second protocol, feline iPSCs were enzymatically passaged with 0.05% trypsin-EDTA and iPSC bulk cultures were maintained in iPSC medium, supplemented with 10 µM of the transforming growth factor-β (TGF-β) type I receptor inhibitor, SB431542 (Stemgent, Cambridge, MA), in T25 cell culture flasks (Corning Inc.) for 12–14 days with daily medium change [33]. After a single cell suspension was generated, cells were seeded directly into T75 cell culture flasks with MSC medium and expanded.

For the third protocol, a serial plating strategy was used [35]. Briefly, iPSCs were enzymatically passaged with 0.05% trypsin/EDTA and seeded onto gelatin-coated 6-well plates at 1 × 10⁴ cells/cm² in MSC induction media, consisting of DMEM-high glucose supplemented with 10% FBS, 0.1 mM NEAA, 50 U/mL penicillin/streptomycin, 10 µM SB431542, and 5 ng/mL bFGF. Cells were grown until they reached 70–80% confluency and passaged onto fresh gelatin-coated wells. Hereafter, cells were passaged onto uncoated 6-well plates. SB431542 was removed from the MSC induction media 10 days after the initial passaging. Cells were routinely passaged every 3–4 days until Passage 12, when a single-cell suspension was generated, and cells were seeded directly into T75 cell culture flasks with MSC medium and expanded.

Immunofluorescence assays

Feline iPSC-derived MSCs were seeded at 1 × 10⁴ cells per cm² onto sterile coverslips, placed into 24-well plates, and cultured until 75–80% confluent. Cells were washed with PBS three times and fixed with 4% PFA for 15 min at RT. Cells were again washed with PBS three times and blocked with PBS + 2% bovine serum albumin (BSA) for 30 min at RT. Next, cells were incubated with primary antibodies in PBS + 3% (v/v) goat serum overnight at 4°C. Cells were incubated with PBS + 3% (v/v) goat serum only as controls. The next day, cells were washed three times with PBS and incubated with secondary antibodies for 1 h at RT. Cells were washed with PBS twice. DAPI was added during the third wash and incubated for 2 min at RT. After another wash with PBS, coverslips were extracted and placed on glass slides with a mounting medium (Dako, CA). Photomicrographs were captured with an Olympus FV3000 confocal microscope (Olympus, Japan). Primary and secondary antibodies and their dilutions are shown in Table 2.

Table 2 Antibodies used in this study

Flow cytometry

Feline iPSC-derived MSCs were dissociated with Accutase (Innovative Cell Technologies, San Diego, CA), and 2 × 10⁵ cells were distributed into 4 ml tubes. Cells were washed with PBS three times and fixed with 4% PFA for 15 min at RT. Cells were washed again with PBS three times and blocked with PBS + 3% BSA for 15 min at RT. Then, cells were incubated with primary antibodies, diluted in PBS, for 30 min at RT and after washing twice with PBS, incubated with secondary antibodies for 30 min at RT (Table 2). Cells were incubated with labelled secondary antibodies only as controls. After washing cells twice with PBS, 10,000 events per sample were analyzed using a Fortessa X-20 flow cytometer (Becton Dickinson, Franklin Lake, NY). Data were analyzed using FACSDiva (Becton Dickinson) and FlowJo (FlowJo LLC, Ashland, OR) software.

In vitro tri-lineage MSC differentiation assays

The tri-lineage differentiation potential of feline iPSC-derived MSCs was assessed using StemPro Adipogenesis, Osteogenesis, and Chondrogenesis differentiation kits (Thermo Fisher). Briefly, cells were cultured in adipogenic or osteogenic induction medium at 1 × 10⁴ cells/cm² density in 12-well tissue culture plates for 14 and 21 days, respectively. For the chondrogenic differentiation, 1 × 10⁶ cells were collected by centrifugation at 200 xg for 3 min in a 15 mL conical tube, the supernatant removed, and the cell pellet resuspended in a chondrogenic induction medium with the medium being replaced every 2–3 days for 14 days. Adipogenic, chondrogenic, and osteogenic differentiation was confirmed by staining with Oil Red O, Alcian blue, and Alizarin red S (all from Sigma-Aldrich), respectively. Brightfield images of stained cells were captured using an Olympus CKX41 inverted microscope and Infinity 2 digital camera.

RNA sequencing

Feline SV-derived iPSCs and iPSC-MSCs were rinsed with PBS, enzymatically passaged using 0.05% trypsin/EDTA, and collected by centrifugation at 200 × g for 3 min. Resulting cell pellets were snap-frozen in liquid nitrogen and stored at -80 °C until RNA extraction. Total RNA was extracted using the QIAGEN RNeasy Plus Mini Kit. Total RNA concentration was confirmed with a Qubit RNA HS Kit (Thermo Fisher) and RNA integrity was assessed using a Fragment Analyzer (Agilent, Santa Clara, CA). For library preparation, 1.3 µg of total RNA was used. RNA-seq libraries were prepared with the SMARTer mRNA-Seq Library Prep Kit, following the manufacturer’s instructions, and sequenced on an Illumina platform. Data processing was conducted using CLC Genomics Workbench v11.0.1. Sequence reads were trimmed to remove adaptor sequences and low-quality bases (quality limit: 0.05). At least 20 M reads were generated per library. To remove contaminating data from the mouse feeder cells, trimmed reads were first mapped to the GrCM38/mm10 reference genome (parameters: mismatch cost: 2; insertion cost: 3; deletion cost: 3; length fraction: 0.8; similarity fraction: 0.8). Read count extraction and normalization were performed using CLC genomic benchwork. These mouse-derived reads were subtracted from the dataset and the remaining reads were mapped to the Felis catus reference genome (Felis_catus_9.0 (GCA_000181335.4). [Parameters: – outSAMstrandField intronMotif,–outFilterIntronMotifs RemoveNoncanonical,–outSAMtype BAM SortedByCoordinate,–quantMode GeneCounts] SARTools and DESeq2 v1.26.0 were used to generate normalized counts and statistical analysis of differential gene expression (53–55). [Parameters: fitType parametric, cooksCutoff TRUE, independentFiltering TRUE, alpha 0.05, pAdjustMethod BH, typeTrans VST, locfunc median].

Cell growth kinetics

Feline iPSC-derived MSCs at P2, 5, and 8, were seeded in triplicate in 6-well plates at densities of approximately 2 × 10⁴ cells per well. Cells were collected from each well 1–9 days after seeding, mixed with Trypan blue (Thermo Fisher), and live cells were counted.

Statistical analysis

Data represent at least three independent experiments and are presented as the mean ± standard error of the mean. Differences between individual groups were analyzed by analysis of variance (ANOVA). Means were compared by Student’s t-test using GraphPad Prism 9 software (Graph Pad, San Diego, CA). Significance was indicated as either not significant (ns) or *: P ≤ 0.05, **: P ≤ 0.01, ***: P ≤ 0.001 or ****: P ≤ 0.0001.

Results

Successful generation of induced pluripotent stem cells (iPSCs) from feline fetal fibroblasts (FFFs) by Sendai virus (SeV) reprogramming vectors

The responsiveness of FFFs to transduction by non-integrating Sendai reprogramming vectors was first confirmed by using the CytoTune emerald, green fluorescent protein (EmGFP) Sendai fluorescence reporter assay. EmGFP expression became visible at 24 hpt and strong expression was observed at 48 and 72 hpt (Fig. 1A). The percentage of GFP-positive cells at MOI 1,3 and 5 was 7.4%, 37.8% and 54.2%, respectively (Fig. 1B).

Fig. 1

Generation of feline induced pluripotent stem cells (iPSCs) using Sendai virus (SeV) reprogramming vectors. (A). Transduction of feline fetal fibroblasts (FFFs) with an EmGFP SeV fluorescence reporter and GFP expression at different hours post transduction. (B). Transduction efficiency of SeV reporter at different multiplicities of infection (MOI). (C). A schematic timeline for the generation of feline iPSCs. FFFs were transduced with SeV reprogramming vectors that carried three vector preparations: human (h) KOS, hc-MYC, and hKLF4. (D). Phase contrast images of FFFs at different days post transductions

FFFs were then transduced with SeV vectors carrying the human transcription factors OCT-4, SOX-2, c-MYC, and KLF-4. A timeline for the reprogramming is shown in Fig. 1C. During reprogramming, mesenchymal to epithelial-like transition (MET) was initially observed around 18–20 dpt and colonies were first observed around 22–24 dpt (Fig. 1D). After passaging onto fresh feeder cells at around 28 dpt, colonies were recognizable and developed into tightly packed dome-shaped colonies with well-defined borders and bright phase surfaces containing cells with a high nuclear-to-cytoplasmic ratio (Fig. 2A). The morphology of the colonies closely resembled those of mouse embryonic stem cells (ESCs) and iPSCs [39, 40], and was comparable to iPSCs generated from domestic cat cells using integrative vectors [41, 42]. A total of four iPSC lines were generated (Fig. 1D and Additional File 1: Fig. S1A) and based on morphology, two were selected for further in-depth characterization. The two selected iPSC lines grew at similar rates, requiring subculture every 8–10 days, and were dependent on feline leukemia inhibitory factor (LIF) to maintain an undifferentiated state. Indeed, withdrawal of LIF from the iPSC medium resulted in the differentiation of feline iPSCs within a few days, as assessed by the loss of colony morphology without well-defined borders and the occurrence of fibroblast-like cells (Additional File 1: Fig. S1B). Both feline iPSC lines have been stable for more than 35 passages, providing evidence of their sustained proliferative capacity.

Fig. 2

Feline iPSCs express pluripotent characteristics, pluripotency markers, and silencing of exogenous transgenes. (A). Characteristic morphology of established iPSC colonies after being passaged onto feeder cells. Arrows indicate cells with a high nuclear-to-cytoplasmic ratios. (B & C). Conventional RT-PCR analysis of endogenously expressed feline pluripotency markers SOX2, NANOG and OCT4, and the loading control GAPDH, in feline iPSCs at passage (P) 0, 3, 6, 9 and 12 (B) and P15 and 25 (C). Full-length gels are presented in Additional File 9: Fig. 9). (D). Conventional RT-PCR analysis of human exogenous transcription factors c-Myc, KOS, KLF4, and Sendai virus (SV), in feline iPSCs at P 0, 3, 6, 9 and 12. Feline GAPDH was included as loading control. (E). Expression of SeV using qPCR analysis at different passages

Feline iPSCs express pluripotency markers and display properties of self-renewal

Initially, conventional RT-PCR was used to assess the expression of the feline-specific key pluripotency markers SOX2, NANOG, and OCT4 in both iPSC lines across passages. Expression of OCT4 and NANOG was initially noticed around passage (P)3 and expression of SOX2 was initially noticed around P6 (Fig. 2B). The relative mRNA expression levels of these endogenous factors were further confirmed using quantitative (q)RT-PCR and both iPSC lines showed significantly upregulated expression across passages (Additional File 1: Fig. S1C). Consistent and robust expression of these feline (f) endogenous pluripotency markers was further observed at P15, P25 and P35 (Fig. 2C and Additional File 2: Fig. S2). In contrast, non-transduced FFFs did not express any of these genes (Fig. 2B). Likewise, conventional RT-PCR was used to determine the expression of the human-specific exogenous transgenes KOS, c-MYC, and KLF-4, as well as presence of SeV, in reprogrammed feline cells. In both cell lines, expression of all three exogenous human transcription factors was not detectable beyond P3. The expression of SeV transgene was highest at P0 and gradually diminished until P12 (Fig. 2D). Expression of SeV was no longer detectable at P15 as shown by qPCR analysis (Fig. 2E). These results indicate that feline iPSCs can maintain their self-renewal profile for an extended period independent of exogenous transcription factors.

Feline iPSC colonies were positive for alkaline phosphatase staining when compared to unstained controls (Fig. 3A). Karyotype analysis of both feline iPSC lines at P20 confirmed a normal diploid chromosome number of 38 (Fig. 3B), indicating genomic stability at this stage. This chromosomal integrity was comparable to that of the parental fibroblasts, which also exhibited a normal karyotype (Additional File 3: Fig. S3A). Maintaining a stable karyotype is an important characteristic of high-quality iPSCs, as chromosomal abnormalities can impact pluripotency, self-renewal properties, differentiation potential, and downstream applications such as organoid generation. These findings thus further support the suitability of our feline iPSC lines for applications in veterinary medicine and translational studies.

Fig. 3

Alkaline phosphatase staining, karyotype analysis, and embryoid body (EB) formation assay. (A). Feline iPSC colonies were stained for alkaline phosphatase or left unstained (control). (B). Karyotype analysis of feline iPSCs-1 and iPSCs − 2 at P20 showing a normal diploid chromosome number of 38. (C). Feline iPSCs formed embryoid bodies (EBs) in suspension culture in differentiation medium. (D). Conventional RT-PCR analysis of feline markers of all three embryonic germ layers, including alpha-fetoprotein (AFP), GATA binding protein 6 (GATA6), and C-X-C chemokine receptor type 4 (CXCR4) for endoderm; smooth muscle actin (SMA) and GATA2 for mesoderm; and ENOLASE and NESTIN for ectoderm in feline EBs. Feline GAPDH was included as loading control. Full-length gels are presented in Additional File 10: Fig. S10. (E). Conventional RT-PCR analysis of endogenously expressed feline pluripotency markers OCT4, SOX2, and NANOG, and the loading control GAPDH, in feline iPSCs and EBs

Feline iPSCs differentiate into the three embryonic germ layers in vitro

Both iPSC lines readily formed embryoid bodies (EBs) when cultured in suspension with iPSC medium without LIF and basic fibroblast growth factor (bFGF) (Fig. 3C). Expression analysis of feline genes revealed significant upregulation of markers of all three embryonic germ layers: alpha-fetoprotein (AFP), GATA binding protein 6 (GATA6), and C-X-C chemokine receptor type 4 (CXCR4) for endoderm; smooth muscle actin (SMA) and GATA2 for mesoderm; and ENOLASE and NESTIN for ectoderm (Fig. 3D). In contrast, the parental iPSC lines displayed no or negligible expression of these markers (Fig. 3D). Moreover, during the germ layer specification, the feline pluripotency genes OCT4, SOX2, and NANOG, which showed expression in the parental iPSC lines, were no longer expressed in the EBs (Fig. 3E).

In vivo differentiation

An in vivo differentiation assay was performed to assess the pluripotency of feline iPSCs in vivo by injecting iPSCs in immunodeficient mice. Visible masses were first observed around five weeks post-injection for both iPSC lines. Mice were euthanized at 8 weeks post-injection and masses were removed from mice injected with iPSCs-2 in Matrigel (Fig. 4A) and injected with iPSCs-1 in Matrigel, iPSCs-1 in Laminin and iPSCs-2 in Laminin (Additional File 3: Fig. S3B). Histologic analysis of the mass from iPSCs-2 in Matrigel showed differentiation mainly towards the mesodermal lineage with cartilage and bone (Fig. 4B). Evaluation of gene expression patterns also showed a higher expression of the mesodermal marker SMA, in addition to the ectodermal marker NESTIN, compared to the endoderm markers AFP, GATA6 and CXCR4 (Fig. 4C). As expected, there was no expression of the human (h) pluripotency markers KOS, KLF4 and c-MYC, not SV (Fig. 4C).

Fig. 4

In vivo differentiation assay. (A). Macroscopic images of immunodeficient NOD-scid gamma (NSG) mice at week 8 post injection of feline iPSCs (i) and dissected mass (ii). (B). Higher magnification of square region in i highlighting a region of differentiation into cartilage (arrowhead) and bone (arrow) surrounded by germ cell tumor-like cells (asterisk). Scale bar in i = 2 mm and ii = 200 mm. H&E stain. (C). Quantitative polymerase chain reaction (qPCR) analysis of endogenously expressed feline differentiation markers, human transcription factors hKOS, hKLF4, hc-MYC, and SeV in dissected mass

Interestingly, masses generated in mice injected with iPSCs-1 in either Matrigel or Laminin and iPSCs-2 in Laminin contained a monomorphic population of large polygonal cells forming dense sheets and vague nests supported by a sparse fibrovascular stroma on histology (Additional File32: Fig. S3C). Cells displayed large and round nuclei with clumped chromatin and formed a poorly demarcated, infiltrative, unencapsulated mass with multiple areas of necrosis (Additional File 3: Fig. S3C). The overall features of these proliferative tissues resembled germ cell tumors rather than forming differentiated tissues representing the 3 germ layers. Evaluation at the mRNA level demonstrated elevated expression levels of feline mesodermal and ectodermal markers (data not shown), similar to what was observed for the mass generated by iPSCs-2 in Matrigel (Fig. 4C). Also here, the human (h) pluripotency markers KOS, KLF4, c-MYC and SV, were not expressed, however, these germ cell tumor-like masses did express the feline (f) pluripotency markers OCT4, SOX2 and NANOG, as well as the proto-oncogenes c-MYC and KLF4 (Additional File 3: Fig. S3D).

Generation and characterization of feline induced pluripotent stem cells (iPSCs) by retroviral vector (RV) reprogramming

Parallel to Sendai virus (SeV) reprogramming, feline iPSCs were also generated using the more traditional technique of integrative retroviral vectors (RV). A timeline for the retroviral reprogramming is shown in Additional File 4: Fig. S4A. Emerging iPSC colonies were first observed around 4–5 dpt and were relatively two-dimensional with a well-demarcated periphery with cells exhibiting a high nuclear to cytoplasm ratio (Additional File 4: Fig. S4B). Following passaging onto fresh SNL feeder cells after 8–10 dpt, feline iPSC colonies were tightly packed and dome-shaped (Additional File 4: Fig. S4B) and this morphology was comparable to previously reported retrovirally-derived feline iPSCs [41].

As per conventional RT-PCR analysis, these iPSCs expressed the feline-specific key pluripotency markers SOX2, NANOG, and OCT4 in both iPSC lines, starting around passage (P)1 and gradually increasing over passages (Additional File 4: Fig. S4C). So far, we have maintained these iPSCs over P12 without losing proliferative potential. Based on the quantitative (q)RT-PCR analysis for relative mRNA expression levels of the exogenous human pluripotency factors, these RV-derived feline iPSCs continued to show expression of viral transgenes both at P6 and P12 (Additional File 4: Fig. S4D).

Furthermore, RV-derived iPSCs readily formed embryoid bodies (EBs) in suspension culture, using iPSC medium without LIF and bFGF, and analysis for expression of markers of the three embryonic germ layers revealed significant upregulation of AFP, GATA6, CXCR4 for endoderm, SMA and GATA2 for mesoderm, and ENOLASE and NESTIN for ectoderm when compared to the parental iPSC lines (Additional File 4: Fig. S4E).

Feline iPSCs can be successfully differentiated into mesenchymal stromal cell (MSC)-like cells

We first attempted to generate feline iPSC-MSCs through EB-based differentiation (Fig. 5A) and by direct inhibition of transforming growth factor-β type I receptor (Fig. 5B). However, these efforts were not successful and resulted in an intermediate cell type without typical MSCs morphology {(Fig. 5A(ii) & Fig. 5B(ii)).

Fig. 5

Mesogenic induction of feline induced pluripotent stem cells (iPSCs). Unsuccessful generation of feline iPSC-derived mesenchymal stomal cells (MSCs) using embryoid body (EB)-based differentiation (A) and via direct inhibition of TGF-β type I receptor (B). Representative images during early differentiation (i) and at passage 3 (ii)

In contrast, a serial plating strategy in the presence of the transforming growth factor-β type I receptor inhibitor SB431542 was successful in differentiating feline iPSCs into MSCs (Fig. 6A). After approximately 8 days in MSC induction medium, iPSCs lost their typical phenotype, displayed epithelial to mesenchymal transition (EMT), and acquired a mixed cuboidal and fusiform-like morphology. Following subsequent passage onto uncoated plates after P12, a spindle-shaped-like morphology gradually began to dominate, and eventually, most of the differentiated cells displayed a typical MSC-like morphology (Fig. 6A). From this point onward, these cells were named feline iPSC-derived MSCs and were considered as a P1 MSC culture.

Fig. 6

Successful generation of feline iPSC-derived mesenchymal stromal cells (MSCs). (A). Morphology of iPSC-derived MSCs-1 and − 2, with evenly distributed plastic adherent cells. (B). Conventional RT-PCR analysis of feline MSC-specific surface markers CD29, CD44, CD73, CD90, and CD105, the hematopoietic stem cell marker CD34, and the loading control GAPDH, in feline iPSC-MSCs and iPSCs. Full-length gels are presented in Additional File 10: Fig. S10. (C & D). Immunofluorescence (C) and flow cytometric analysis (D) of MSC-specific feline surface proteins CD44, CD73, CD90, and CD105 (green). Nuclei were visualized by staining with DAPI (blue)

Both iPSC-derived MSC lines (iPSC-MSCs-1 and iPSC-MSCs-2) displayed a similar growth pattern when assessing their growth kinetics starting at P2 until P8 (Additional File 4: Fig. S5A). As expected, early passage MSCs (P2) displayed higher proliferative potential compared to late passage MSCs (P5 and P8) for both cell lines (Additional File 5: Fig. S5A). However, the rate of proliferation of iPSC-derived MSCs at each passage decreased gradually from day 1 to 9 for both cell lines (Additional File 5: Fig. S5A).

One of the classifying criteria for MSCs is their specific immunophenotypic profile, consisting of expression of various surface markers related to MSCs and absence of expression of markers associated with other types of stem and differentiated cells [43]. Conventional PCR analysis revealed the expression of CD29, CD44, CD73, CD90, and CD105, combined with a lack of expression of the hematopoietic stem cell (HSC) marker CD34 [44] (Fig. 6B). Importantly, the expression pattern noted in feline iPSC-MSCs was comparable to the one observed in bone marrow (BM)-derived feline MSCs. In contrast, the parental iPSC lines only showed positive expressions for CD29 and CD90, but not CD34, CD44, CD73, and CD105 (Fig. 6B). To confirm these results on a protein level, both immunofluorescence (IF) and flow cytometry was performed. As expected, feline iPSC-MSCs stained positive for MSC surface proteins CD44, CD73, CD90, and CD105 by IF (Fig. 6C) and flow cytometry (Fig. 6D). Negative controls to assess non-specific immunostaining are shown in Additional File 5: Fig. S5B.

Another important criterion to characterize MSCs is based on their trilineage differentiation potential [43]. To this end, we subjected feline iPSC-MSCs to standard chondrogenic, adipogenic, and osteogenic differentiation protocols. As expected, iPSC-MSCs grown in these differentiation media displayed robust chondrogenic differentiation, as demonstrated by positive Alcian blue staining, adipogenic differentiation, as demonstrated by Oil Red O positive intracellular lipid droplets, and osteogenic differentiation, as demonstrated by positive Alizarin Red staining (Fig. 7). In contrast, iPSC-MSCs grown in regular MSC medium and subjected to these staining did not show any positive signal (Fig. 7).

Fig. 7

In vitro differentiation of feline induced pluripotent stem cells (iPSC)-derived mesenchymal stromal cells (MSCs). Chondrogenic, adipogenic, and osteogenic differentiation of iPSC-MSCs-1 and iPSC-MSCs-2, as confirmed by Alcian Blue, Oil Red O, and Alizarin Red stainings, respectively

Global gene expression changes of feline iPSCs during MSC differentiation

Differentiation of feline iPSCs into an MSC-like phenotype involves dynamic and widespread changes in gene expression. Principal component analysis (PCA) showed distinct clustering of samples corresponding to differentiation state (Additional File 6: Fig. S6A). The differential gene expression was visualized using a MA plot, showing the significant transcriptional changes that occurred during this directed differentiation (Additional File 6: Fig. S6B). A total of 1,189 differentially expressed genes (DEGs), including 606 upregulated and 583 downregulated, were identified in feline IPSC-MSCs as a result of the differentiation process (Additional File 7).

Pluripotency-associated genes SOX2 and cMYC were significantly downregulated in IPSC-MSCs, suggesting the loss of stem cell characteristics (Additional File 6: Fig. S6C). In contrast, key genes associated with mesenchymal phenotype; integrin subunit beta 1 (ITGB1), pro-migratory and EMT-associated genes such as Zinc finger E-box binding homeobox 2 (ZEB2) and Cadherin 6 (CDH6), and the structural gene Desmin (DES), were upregulated in IPSC-MSCs (Additional File 6: Fig. S6C), highlighting the establishment of mesenchymal properties. Albeit not significant, other key genes related to a mesenchymal phenotype, such as vimentin (VIM), fibronectin 1 (FN1), a marker of mesenchymal differentiation; SRY-Box transcription factor 9 (SOX9) and cytoskeletal and structural genes such as α-Smooth Muscle Actin (ACTA2) and Secreted Protein Acidic and Rich in Cysteine (SPARC), were expressed at higher levels in iPSC-MSCs when compared to iPSCs (Additional File 7). Differentiation of feline iPSCs into MSCs was marked by downregulation of several oncogenes, including MYC proto-oncogene (MYC), Fos proto-oncogene (FOS), FosB proto-oncogene (FOSB), JunD proto-oncogene (JunD), RELB proto-oncogene (RELB) and Ret proto-oncogene (Ret) (Additional File 6: Fig. S6C). These genes are typically associated with proliferation, survival, and undifferentiated states. Conversely, differentiation of feline iPSCs into MSCs was also accompanied by an upregulation of several oncogenes, including KRAS proto-oncogene (KRAS), MN1 proto-oncogene (MN1), ABL proto-oncogene 2 (ABL2) ETS proto-oncogene 1 (ETS1), Pim-1 proto-oncogene (Pim-1), AKT serine/threonine kinase 3 (AKT3) and Yes associated protein 1 (YAP1) (Additional File 6: Fig. S6C). These genes, known for their roles in cellular growth, survival, and differentiation, may be indicative of mesenchymal lineage commitment rather than potential for tumorigenesis.

Discussion

This study demonstrates that fetal fibroblasts from domestic cats can be reprogrammed into induced pluripotent stem cells (iPSCs) using integration-free Sendai virus (SeV) vectors. Further, we demonstrated that these iPSCs can be successfully differentiated into mesenchymal stromal cells (MSCs) that are plastic adherent, have a fibroblast-like morphology, express MSC-specific surface markers, and possess trilineage differentiation potential.

The SeV reprogramming system utilizes non-integrative vectors that remain in the cytoplasm and, therefore, should be unable to alter the genetic information of the host cell [45]. This is in contrast with integrative vectors that require genomic integration to express reprogramming genes, a process which poses significant safety concerns and hampers the advancement of iPSC-based therapies due to the risk of insertional mutagenesis. So far, SeV-based reprogramming has been performed with human, murine, non-human primate and canine cells. For human cells, iPSC colonies typically began to emerge around 11 dpt with a multiplicity of infection (MOI) of 3 [23] and murine iPSC colonies were first noted around the same time, but with an MOI of 5 [46]. Canine embryonic fibroblasts (CEF) and peripheral blood mononuclear cells (PBMCs) have been successfully reprogrammed into canine iPSCs using an auto-erasable SeV vector at a MOI of 3–10 for CEF [25] and 2 for PBMCs [47]. Initial iPSC colonies appeared around 7 dpt using CEF and 10 dpt using PBMCs, with PBMC-derived iPSCs requiring approximately 22 days to reach a size suitable for picking and expansion. In a separate study, canine dermal fibroblasts and canine urine-derived cells have been reprogrammed using SeV containing 6 canine transcription factors: OCT3/4, SOX2, NANOG, KLF4, c-MYC and LIN28 at a MOI of 1 [48], with primary colonies first noted at 15 dpt and expansion between 15 and 30 dpt. In the case of non-human primate cells, peripheral T lymphocytes from chimpanzees have been reprogrammed with temperature-sensitive SeV vectors carrying human OCT3/4, SOX2, KLF4 and c-MYC and with colonies first noticed anywhere between 18 and 25 days, depending on the MOI used [10]. In the current study, the emergence of the first colonies was noted around 22–24 dpt and established colonies were observed as late as 28 dpt. It should be noted, however, that the lowest MOIs recommended for reprogramming were used, and using higher MOIs may have resulted in earlier colony formation. Moreover, SeV, also known as murine parainfluenza virus type 1, is a highly contagious respiratory tract infection that typically affects rodents, pigs, hamsters and guinea pigs [49]. However, SeV infection in feline species has not been well established, and thus, successful infection of feline fetal fibroblasts by SeV vectors and cautious long-term culture until the emergence of colonies during the reprogramming process remains a critical point for the induction of feline pluripotency. Furthermore, the use of SeV-based reprogramming in comparative and veterinary animal research necessitates broader adaptation and careful optimization of protocols to improve efficacy and reliability across various species.

Ideally, one would use feline-specific reprogramming factors to generate feline iPSC, as using human or mouse reprogramming factors could potentially raise safety concerns. For instance, reprogramming across species could introduce genetic abnormalities or unintended genetic modifications due to differences in regulatory mechanisms, potentially affecting the behavior and function of the reprogrammed cells and having unintended side effects on self-renewal, pluripotency and differentiation potential. Moreover, if the reprogramming process involves viral vectors to deliver the reprogramming genes of other species, there is a potential for cross-species viral transmission [50]. Further, reprogramming animal cells with human proteins could lead to developing cells or tissues with a mixture of human and animal characteristics which is not ideal for cellular therapies, for example by triggering an immune response due to the presence of “foreign” proteins [51]. Although the commercially available SeV reprogramming kit used in the present study comes with human reprogramming factors, we did find that the generated feline iPSCs displayed complete silencing of all human reprogramming transgenes by passage 6. iPSCs with incompletely silenced viral transgenes are considered to be only partially reprogrammed and continue to express both viral and endogenous pluripotency genes [52]. For example, when human or mouse somatic cells were reprogrammed by other viral vectors, viral transgenes were not silenced [45, 52] and retrovirally-derived feline iPSCs continued to express human transgene as late as P22 [41]. Further, RV-derived feline iPSCs generated in this study continued to express exogenous human transgenes at least up to P12. Despite the potential for rapidly and robustly generating iPSCs, these results suggest that maintaining self-renewal and pluripotency in RV-derived feline iPSCs necessitates both endogenous and exogenous pluripotency factors over an extended period. This outcome is common with retroviral reprogramming, as it often leads to integration-dependent expression of exogenous genes. Such dependency could affect the stability and differentiation potential of these iPSCs for more downstream applications, especially compared to non-integrative methods such as SeV-based reprogramming. Therefore, we propose the footprint-free iPSCs generated in the current study as a valuable tool for differentiation into organoids of interest for disease modelling, organ replacement, and precision medicine studies.

One of the most stringent tests for pluripotency of iPSCs is their capability to generate teratomas in vivo when injected into immunodeficient mice. Teratomas are solid, defined tumors composed of highly organized differentiated cells and tissues containing representatives of the three embryonic germ layers [53]. Generation of teratomas, however, can be challenging for non-traditional animal models, and variable outcomes have been reported for different species. Moreover, a common observation with in vivo pluripotent stem cell studies in non-traditional model species is the poor survival of transplanted pluripotent cells as well as limited differentiation potential in immunodeficient mice. In fact, in vivo teratoma formation has not been monitored or reported in many iPSCs from companion and other animals of veterinary importance [22]. For example, while there are numerous reports on the generation of equine iPSCs, some studies do not provide evidence of in vivo teratoma formation [54, 55]. Likewise, most of the reported canine iPSCs could not generate teratomas in immunodeficient mice [56]. In addition, most of the iPSCs reported from ruminant animals did not generate in vivo teratomas despite naming them as iPSCs [8]. These observations can be attributed to various factors such as incomplete reprogramming, differences in reprogramming vectors, lack of species-specific reprogramming genes, as well as a potential influence of the culture media used for the maintenance of undifferentiated iPSCs. For example, spontaneous differentiation inevitably occurs while iPSCs are maintained in culture conditions optimized for undifferentiated cells [57], which reduces the population size of self-renewing cells in the culture, and consequently, can reduce the differentiation potential when injected into immunodeficient mice.

In our study, injection of feline iPSCs in immunodeficient NSG mice resulted in visible masses within five weeks in 4 out of the 4 injected mice (success rate of 100%). Interestingly, one iPSC line displayed differentiation into mesodermal tissues whereas the other iPSC line generated tumors which appeared to resemble germ cell tumors rather than developing into clearly differentiated tissues, as typically seen with teratomas. Of note is that our attempt to promote differentiation by culturing feline iPSCs in recombinant laminin instead of the more traditionally used Matrigel did not alter the results. Moreover, human PBMC-derived iPSCs have also been reported to show a preferential differentiation potential into cells and tissues of mesodermal lineage [58]. In addition, the occurrence of tumors when performing in vivo teratoma assays has also previously reported for (i) canine iPSCs, which developed into germ-cell like tumors [6], similar to what we observed in our study with feline iPSCs, and (ii) human iPSCs, where one study found that 43% of human iPSCs generated yolk sac tumors and embryonal carcinomas [59]. The same group correlated their findings with the (re-) expression of one or several transgenes used for the reprogramming combined with a down-regulation of P21 and up-regulation of anti-apoptotic B-cell lymphoma 2 transcription (BCL-2) [6, 59]. Some reprogramming genes are well-recognized oncogenes, such as c-MYC and KLF4, and are associated with tumorigenesis because they are expressed in various human cancers and linked to poorly differentiated aggressive human malignancies [60,61,62]. Interestingly, upon analyzing the expression pattern of human transgenes in the feline iPSC-generated tumors, we did not detect human KOS, KLF4, or c-MYC, although we did see expression of feline KLF4 and c-MYC, indicating that re-expression of human transgenes upon injection was not the reason that some feline iPSC lines differentiated into germ cell-like tumors. Furthermore, feline iPSCs generated thus far using integrative retroviral vectors also do not provide evidence of in vivo teratoma formation [41, 42] and little is known about the potential risk of malignancy of reprogrammed feline iPSCs during in vivo differentiation. Therefore, assessing the safety of stem cells is critical for their potential applications in regenerative medicine and cellular therapy. Reprogramming somatic cells into iPSCs can sometimes induce genetic mutations, chromosomal abnormalities, or epigenetic changes that could pose risks and may promote cancer or other adverse effects during cellular therapies [63]. Moreover, iPSCs have a high proliferation rate, which also poses a risk of forming teratomas [63]. Even iPSC-derived cells, such as iPSC-MSCs can sometimes show uncontrolled cell division if they retain residual pluripotency or undergo genetic changes [64]. Ensuring the directed differentiation of iPSCs toward specific cell lineages or towards fully differentiated cells before therapeutic use minimizes this risk.

iPSC-derived MSCs have been proposed as an important alternative and standardized source of primary MSCs for research and cell-based therapies with less heterogeneity and higher ex vivo expansion potential [65]. They have been generated using different techniques, including culturing adherent outgrowths from embryonic bodies (EBs) [66], culturing on various surfaces such as fibrillar collagen, using cell selection [66], or by inhibiting TGF-β type I receptor activity [33], and have been described for human [33, 34], murine [32], canine [67], equine [35] and Tasmanian devil [68], but not for feline iPSCs. During this study, we successfully generated feline iPSC-derived MSCs for the first time using a serial plating strategy in the presence of TGF-β type I receptor inhibitor, as previously described for equine iPSCs [35].

The feline iPSC-MSCs generated in this study displayed low CD105 surface marker expression. According to the International Society for Cell and Gene Therapy (ISCT), human MSCs should demonstrate positive expression (at least 95%) of surface markers CD105, CD90, and CD73, while being negative (no more than 2% expression) for CD45, CD34, CD14/CD11b, CD79α/CD19, and HLA-DR [69]. Although this definition is still widely used today, it is generally accepted that MSCs constitute a heterogeneous population with distinct gene/protein expression profiles that are influenced by various factors such as species-specific differences, tissue sources, donor age, isolation procedures, culture conditions, and the passage number of cells at the time of characterization [70,71,72]. Indeed, various studies on human and mouse MSCs have reported variable CD105 expression levels [73,74,75] and the lack of CD105 expression has been documented for feline MSCs as well [76, 77]. Moreover, variable CD105 expression has also been noted in iPSC-derived MSCs, where one study found that CD105 expression was present in only a small subset of murine iPSC-derived MSCs [32]. Additionally, CD105 expression in human iPSC-MSCs generated using an inhibitory differentiation method was predominantly negative [33] and studies with equine iPSC-derived MSCs reported extremely low CD105 expression levels, with one study documenting no CD105 expression at all [35, 78].

The transition of feline iPSCs into an MSC-like phenotype was marked by extensive and coordinated transcriptional changes. Downregulation of key pluripotency-associated genes and upregulation of genes linked to the mesenchymal phenotype, suggest a loss of pluripotent characteristics of feline iPSCs and the acquisition of a more differentiated, MSC-like phenotype, during differentiation. Interestingly, the identified 1,189 differentially expressed genes (DEGs) in feline iPSC-derived MSCs are consistent with a previous study in which canine iPSC-MSCs showed 819 DEGs when compared to the parental iPSCs derived through Sendai viral reprogramming [79]. These findings not only offer valuable insights into the molecular mechanisms involved in differentiation of feline iPSCs into MSCs, but also highlight conserved pathways that govern cellular differentiation across species. The observed downregulation of several oncogenes associated with self-renewal and pluripotency in iPSC-MSCs indicates a decrease in proliferative potential and a shift towards a more differentiated cell type, whereas the upregulation of several other proto-oncogenes could merely indicate differentiation. Indeed, expression of oncogenes does not necessarily indicate that iPSC-derived MSCs are prone to become cancerous as oncogenes typically need to be overexpressed, activated, mutated, or simultaneously triggered in specific combinations and conditions for cancer to develop [80]. For example, the KRAS proto-oncogene, a key regulator in the Ras-MAPK signaling pathway [81] was found to be upregulated in feline IPSC-MSCs. KRAS plays a dual role in both promoting cell proliferation and supporting differentiation, depending on the cellular environment [82]. Similarly, the upregulation of MN1 and ETS1 may have contributed to the transcriptional regulation of genes associated with achieving a mesenchymal phenotype [83, 84]. Likewise, ABL2 is known for its role in controlling cytoskeletal changes and cell adhesion [85], both of which are crucial for MSC development. Pim-1 and AKT3, both serine/threonine kinases, play crucial roles in pathways that promote cell survival and metabolism during differentiation [86], and YAP1, a key regulator of the Hippo signaling pathway, has a role in supporting mesenchymal differentiation and maintaining cell viability under mechanical stress [87]. Overall, the sequencing results highlight a shift in gene expression that supports the differentiation of feline iPSCs into MSCs.

MSCs are considered an ideal source for therapeutics in tissue regeneration and autoimmune and hyperinflammatory diseases [88]. As the feline iPSC-derived MSCs generated in this study are also genome integration or (footprint)-free, they provide an unlimited source of fairly homogenous cells that can be used in both clinical trials and translational research. In fact, therapeutic trials with tissue-derived feline MSCs for immune-mediated disorders, such as chronic gingivostomatitis [89], feline chronic enteropathy [90], and feline asthma [91, 92], as well as chronic kidney disease [93] have been performed in feline patients with encouraging results, which is not only important for feline health but human health as well, as many human patients suffer from similar diseases.

Conclusion

We report the generation of integration- and footprint-free induced pluripotent stem cells (iPSCs) from domestic cat fetal fibroblasts via Sendai virus transduction of human transcription factors, further adding to the growing literature of the creation of feline iPSCs (Additional File 8: Table S1). These iPSCs were then successfully differentiated into functional MSCs using a serial plating strategy in the presence of an TGF-β type I receptor inhibitor. Integration-free feline iPSCs and iPSC-derived MSCs will provide a unique resource for generating cellular models for in vitro studies and, more importantly, will offer a platform for clinical and translational research that will benefit both human and feline health.