Abstract
Objectives
Feline autologous mesenchymal stem cells (MSCs) show promise for immunomodulatory activity, but the functional impact of chronic kidney disease (CKD), concurrent immunosuppressive drug administration or infection is unknown. The study objectives compare endogenous cytokine gene expression (interleukin [IL]-6, IL-10, IL-12p40, IL-18 and transforming growth factor beta [TGF-β]) in adipose-derived MSCs (aMSCs) from cats with and without CKD, following in vitro exposure to microbial ligands and treatment with common immunosuppressive drugs.
Methods
Previously obtained aMSCs, phenotype CD44+, CD90+, CD105+ and MHCII–, from cats with (n = 6) and without (n = 6) CKD were compared via real-time PCR (RT-PCR) for immunomodulatory gene expression. aMSCs were exposed in vitro to lipopolysaccharide (LPS), peptidoglycan or polyinosinic:polycytidylic acid (Poly I:C), simulating bacterial or viral exposure, respectively. aMSCs were also exposed to ciclosporin, dexamethasone or methotrexate. Gene expression was measured using RT-PCR, and Cq was utilized after each run to calculate the delta cycle threshold.
Results
aMSCs isolated from healthy and CKD cats showed no significant differences in gene expression in the five measured cytokines. No significant changes in measured gene expression after drug treatment or microbial ligand stimulation were observed between normal or CKD affected cats. Proinflammatory genes (IL-6, IL-12p40 and IL-18) showed altered expression in aMSCs from both groups when compared with the same cells in standard culture after exposure to methotrexate. Poly I:C altered IL-6 and TGF-β gene expression in aMSCs from both healthy and CKD cats when compared with the same cells in standard culture.
Conclusions and relevance
The five genes tested showed no statistical differences between aMSCs from healthy or CKD cats. There was altered cytokine gene expression between the control and treatment groups of both healthy and CKD cats suggesting feline aMSCs have altered function with immunosuppressive treatment or microbial ligand exposure. Although the current clinical relevance of this pilot study comparing brief exposure to select agents in vitro in aMSCs from a small number of cats is unknown, the study highlights a need for continued investigation into the effects of disease and concurrent therapies on use of cell-based therapies in feline patients.
Introduction
The use of mesenchymal stem cells (MSCs) as a treatment for immune-mediated disease has become increasingly attractive owing to the ability of MSCs to modulate the innate and adaptive immune response. 1 We, and others, have shown that feline MSCs inhibit T lymphocyte proliferation and reduce neutrophil production of reactive oxygen species.2,3 The adipose-derived MSC (aMSC)-induced alteration in immune response can have possible clinical benefits for the management of chronic immune-mediated diseases in cats. The successful use of MSCs has been described in cats with experimentally induced asthma. 4 The clinical use of MSCs for diseases such as inflammatory bowel disease and stomatitis has also been documented.5–7 The determinants of successful use in the aforementioned studies were decreased lung attenuation and bronchial wall thickening, complete resolution of clinical signs and reduction of clinical signs. However, the specific mechanisms of immunomodulation are not well established with regard to their clinical use. One study that evaluated the efficacy of aMSC autologous and allogenic administration for stomatitis therapy found that a decrease in CD8lo cells could serve as a biomarker for response to systemic aMSC therapy. 7
aMSCs may not only offer a novel mechanism for immunomodulation in naturally occurring immune-mediated disease, but their use may also have a place in the immunosuppressive regimens of cats following renal transplantation for chronic kidney disease (CKD). MSC administration peri-transplantation has been reported in a small number of cats. 8 The use of aMSCs in people after renal transplantation is an area of active investigation, and clinical trials are ongoing, with initial success.9–11 aMSC administration was safe and well tolerated by kidney transplant recipients,9,10 and administration of aMSCs after renal transplantation resulted in enhancement of both the regulatory T cell (Treg) population in the peripheral blood and control memory CD8+ T cell function. 9 When used in renal transplant recipients, autologous aMSCs are harvested from the recipient and cultured prior to transplantation. The cultured cells are administered to the recipient soon after allograft reperfusion with the intent of blunting the immune reaction12–15 and ischemia–reperfusion injury of the introduced allograft.15,16
Use of cultured autologous MSCs in transplant recipients has several potential advantages, such as dampening adaptive immune responses, 1 reducing host rejection 17 and decreasing infectious disease transmission, in addition to ensured availability. In a study involving rats with established CKD, human embryonic MSC-derived conditioned medium was administered. This is different from autologous administration of MSCs as MSC-derived conditioned medium is a cell-free treatment strategy containing MSC-derived secreted growth factors and cytokines. Administration of MSC-derived conditioned medium in this study resulted in a reno-protective effect causing an increased glomerular filtration rate and decreased glomerular damage. 18 However, humans and cats affected by CKD are known to have altered systemic homeostasis leading to dramatic effects on a variety of cellular functions.19,20 One study found that administration of autologous MSCs to rats with CKD exerted no benefit and the rats were found to have decreased regenerative potential, premature senescence, inability to induce angiogenesis and reduced proliferation in culture. 21 Human bone marrow-derived MSCs from uremic patients have also shown inhibition of metabolic activity and proliferation in vitro. 22 Although some studies suggest that there might be similar issues with feline MSCs,23,24 similar studies have not yet been completed in cats.
In order to better inform clinical trials in feline patients, it is necessary to evaluate the characteristics of autologous aMSCs harvested from cats with CKD undergoing renal transplantation. Additionally, renal transplantation necessitates immune suppression of the patient to prevent allograft rejection, and post-transplantation infection is a common complication effecting 28–37% of feline renal transplant recipients.25,26 Owing to the necessity for immunosuppressive medications in renal transplant recipients and the increased risk and common occurrence of post-transplantation infection, 25 it is critical to understand both the effect of immunosuppressive drugs and, conversely, microbial antigen stimulation on aMSCs prior to their clinical application in feline transplant recipients.
The aim of this study was to determine if CKD affected feline aMSC gene expression of five key immunomodulatory cytokines (interleukin [IL]-6, IL-10, IL-12p40, IL-18 and transforming growth factor beta [TGF-β]) vs gene expression of aMSCs from healthy cats. Gene expression was assessed without treatment and after in vitro exposure to microbial ligands and immunosuppressive therapies to evaluate for changes that could impact the autologous use of aMSCs from cats with CKD undergoing renal transplantation. Our null hypothesis was that there would be no difference in the expression of key immunomodulatory genes between aMSCs from cats with CKD and healthy cats in the naive state or when treated with microbial ligands or immunosuppressive drugs.
Materials and methods
aMSC harvest and culture
Adipose tissue was collected from client-owned cats with CKD. Written informed consent was obtained from all clients prior to harvest, culture and subsequent administration of aMSCs as part of the cat’s clinical care, which also allowed for samples to be used for research purposes. Four of the five client-owned cats (cats 2, 3, 4 and 6) were renal transplant recipients who met rigorous screening criteria for renal transplantation. Cat 5 was a cat with CKD whose owner desired aMSC harvest, culture and administration for the purpose of stem cell treatment of CKD. The falciform adipose tissue was collected under general anesthesia in sterile conditions at various times, but specifically either before, during or unrelated to a transplantation. Client-owned cats were anesthetized under the supervision of a board-certified anesthesiologist using a variety of anesthesia and analgesia protocols; postoperative analgesia was provided with intravenous or transmucosal opioid medication. Fat was collected prior to US Food and Drug Administration (FDA) guidance limiting collection and clinical use of MSCs in cats not enrolled in FDA-registered clinical trials. 27 aMSCs from healthy cats and a cat with ischemia-induced CKD described below were obtained from purpose-bred cats under approval of the Institutional Animal Care and Use Committee (IACUC). Blood used for leukocyte enrichment was also collected from purpose-bred cats with IACUC approval. After collection, tissue was immediately transported in sterile containers on ice and immediately processed in an on-site laboratory. Specific data regarding time from collection to processing are not available for any cat, but identical protocols were followed and no difference in aMSC collection is known between populations.
The affected sample group consisted of aMSC cell lines from five client-owned cats with CKD (Table 1) and a purpose-bred cat with ischemia-induced CKD (cat 1). The cat with ischemia-induced CKD underwent a 90 min unilateral renal ischemia followed by a delayed contralateral nephrectomy approximately 6 months prior to harvesting of adipose for aMSC culture and had stable International Renal Interest Society (IRIS) stage 3 CKD at the time of fat harvest. 28 The control group (Table 2) consisted of six aMSC cell lines, previously obtained from healthy, purpose-bred cats with normal renal function. These cats were being anesthetized for a purpose unrelated to this study, and falciform fat was obtained for the purpose of isolation and culture of aMSCs. Cats were premedicated with intramuscular acepromazine (0.01 mg/kg), buprenorphine (0.03 mg/kg) and ketamine (7 mg/kg), induced isoflurane delivered by mask, intubated and maintained on isoflurane in oxygen. Postoperative analgesia was provided by transmucosal buprenorphine (0.03 mg/kg q8h) for 24 h followed by transdermal fentanyl (3–5 µg/kg) for an additional 3 days. No drugs or treatments were used in the unrelated study that may influence MSCs. The adipose tissue was macerated and plated. An amount of macerated adipose tissue conducible to cell growth was placed in each Petri dish. Cell lines were cultured in MSC medium (DMEM with 1 g/l glucose [Corning cellgro] + 10% Hyclone fetal bovine serum [GE Healthcare] with 2 mM L-glutamine, 50 U penicillin and 50 µg/ml streptomycin [Corning]) within Petri dishes. Adipose tissue was removed after sign of cell attachment, and cells were transferred to T-75 flasks for expansion. Once plated, cells were observed daily for growth. At 80% confluency, cells were passaged with 2.5% trypsin (Corning) and frozen in MSC medium containing 10% dimethyl sulfoxide (Sigma), used for phenotype characterization or used to produce additional cultures. After expansion, cell lines were cryopreserved at 1 × 106 cells/ml freezing medium and stored in a slow-freezing container in a −80°C freezer for 24 h prior to transitioning to liquid nitrogen. Cell lines were utilized between the second to fifth passage, and passage number was determined after each trypsinization.
Summary of the cats from which adipose-derived mesenchymal stem cells were collected
All cats were assessed using the International Renal Interest Society (IRIS) chronic kidney disease (CKD) staging guidelines
MN = male neutered; DSH = domestic shorthair; SC = subcutaneous; DLH = domestic longhair; PKD = polycystic kidney disease; FS = female spayed; IBD = inflammatory bowel disease
Summary of normal cats from which adipose-derived mesenchymal stem cells were collected
All cats were purpose-bred research cats and were determined to be of normal health status based on physical examination and bloodwork at the time of adipose collection
Cats 7, 10, 11, 12 in the healthy cat group had elevated urinary protein to creatinine (UPC) levels (>0.4) based on International Renal Interest Society guidelines likely secondary to the presence of cauxin, a protein known to increase urine protein concentrations in intact male cats29,30 Bloodwork was performed prior to castration and fat collection, and this proteinuria resolved after castration. At the time of fat collection, all cats were neutered
MN = male neutered; DSH = domestic shorthair; BUN = blood urea nitrogen; USG = urine specific gravity
aMSC characterization and flow cytometry
Phenotyping of cryopreserved aMSC lines from the second to third passages was performed using a previously published protocol. 3 Cells were blocked with 1% bovine serum albumin (BSA; Sigma), washed in staining buffer (phosphate buffered saline [PBS], 0.5% BSA [Sigma], 0.1% sodium azide [Sigma]) and incubated with the following primary antibodies: rat anti-CD44 (clone OX-49 [Biolegend]) 1:1000; mouse anti-CD90 (clone 30-H12 [Biolegend]) 1:1000; mouse anti-human CD105 (clone SN6 [IgG1; Serotec/BioRad]) 1:500; and mouse anti-MHCII (monomorphic, clone TH14B [IgG2b; VMRD/Kingfisher Biotech]), 1:2. The samples were incubated for 60 mins at 4°C and then washed with staining buffer. The secondary antibodies were fluorescein isothiocyanate (FITC) goat anti-rat light chain (polyclonal [Biolegend]) 1:100 and FITC anti-mouse IgG (polyclonal, purified Fab2 fragments [Sigma]) 1:200. Secondary antibodies were added to cell samples that had not been exposed to primary antibody (to assess non-specific binding) and to samples with primary antibody from the appropriate species. The secondary antibody was incubated for 30 mins at 4°C and, after incubation, the samples were washed and run on a BD Accuri C6 gen 2 instrument (BD Biosciences).
Blood leukocyte isolation and stimulation
A total of 24 ml blood was collected in three heparinized (100 mM, 0.6 ml/syringe) syringes from the jugular vein of purpose-bred research cats. Blood was pooled and centrifuged after dilution with an equal volume of PBS (ThermoFisher) containing heparin (porcine origin [Sigma]) to generate a buffy coat. The buffy coat was collected, suspended and centrifuged. The pellet was diluted with tris-ammonium chloride (lysis) reagent and incubated, 31 and then washed twice with PBS containing heparin. Blood leukocytes were separated using Ficoll density gradient centrifugation. Cells were counted and assessed for viability using 0.04% Trypan blue (Sigma) and then suspended in MSC medium. Leukocytes werestimulated with either Concanavalin A (Sigma) 5 µg/ml, Eschericia coli 055:B5 lipopolysaccharide (LPS; List Biologicals), 1 μg/ml or Staphyloccus aureus killed whole cell antigen (UGA Veterinary Clinical Research Support Lab), 1:200, for 24 h. Cells were removed after incubation and stored at –80°C for extraction of RNA.
aMSC microbial ligand stimulation and exposure to immune suppressive drugs
To simulate bacterial or viral exposure in vivo, aMSCs from third to fourth passages were grown in T75 flasks to about 90% confluency. Cells were trypsinized (2.5% trypsin [Corning]) and counted. The cells were then adjusted to 105 viable cells/ml using the previously described MSC medium and evenly divided into 96 well plates with 2–3 replicates for each of the treatments described below. aMSCs were stimulated with LPS (E coli 055:B5, List Biologicals) at 100 ng/ml, peptidoglycan (PGN; Sigma) from S aureus at 100 ng/ml or polyinosinic:polycytidylic acid (Poly I:C; MP) at 100 μg/ml for 24 h at 37°C in 96-well round-bottom plates. To simulate immunosuppression, cells were exposed to ciclosporin (CsA; 10–7 M [Sigma]), dexamethasone (10 ng/ml; Sigma) or methotrexate (MTX; 100 nM [Sigma]) for 24 h at 37°C in 96-well round-bottom plates. Cells were removed with 2.5% trypsin (Corning) after incubation. MSC medium was removed after centrifugation, and cells were stored at −80°C as cell pellets for the extraction of RNA. A preliminary trial was conducted to determine the maximal dose non-toxic (<5% trypan blue positive and >95% of untreated Alamar Blue activity for two stem cell cultures) within the estimated in vivo dosage for each drug and the minimal active in vivo concentration of each microbial ligand. These doses were used in the experiments.
RNA extraction, cDNA synthesis and real-time quantitative PCR
RNA extraction was performed on cell lines varying from second to fifth passages using QIAshredder and RNeasy Mini Kits (Qiagen). RNA was purified as specified in the instructions for the RNeasy Mini Kit; RNAlater was not used. Quality and quantity were assessed with a Nanodrop spectrometer (ThermoFisher), and RNA with 260/280 greater than 1.8 was used. cDNA was synthesized using iScript (Bio-Rad) protocols and reagents. Real-time quantitative PCR (RT-qPCR) was performed using SybrGreen (Bio-Rad). The reference gene used for RT-qPCR for all samples was GAPDH (Table 3), and all samples were run within a thermocycler (Bio-Rad). Stimulated blood leukocytes collected from purpose-bred cats were used as positive control, Cq was generated after each run and Cq >40 was considered as no expression present. Cq used to calculate delta cycle threshold (∆Ct) (xCq(gene) – xCqGAPDH). ∆∆Ct was compared using the difference between ∆Ct (medium) and ∆Ct (treatment).
Real-time quantitative PCR (RT-PCR) primer sequences
All primers were diluted based on the manufacturer’s instructions (Sigma Aldrich). The primers and PCR method were designed and validated specifically for these studies based on previously published techniques 33
bp = base pairs; IL = interleukin; TGF-β = transforming growth factor beta; GAPDH = glyceraldehyde-3-phosphate dehydrogenase
Statistical analysis
Linear mixed regression models with a random cat effect were used to evaluate the fixed nominal effects of study group and incubation treatment on the relative concentrations of cytokines. Models were estimated using the method of restricted maximum likelihood, and denominator degrees of freedom for F-tests were estimated using the Kenward–Roger procedure. Briefly, ∆Ct values were treated as the dependent variable and regression coefficients from these models estimating the ∆∆Ct values for each factor level relative to their respective reference category. 32 The relative cytokine concentration of each factor level vs the reference category was calculated as 2–∆∆Ct. All tests assumed a two-sided alternative hypothesis, and values of P <0.05 were considered statistically significant. Outliers were defined as ∆Ct values that were <1 for IL-18 and <5 for IL-10 and IL-12p40. Statistical comparisons of the effect of treatment on cells were performed using the marginal means of both healthy and CKD cats. Analyses were performed by a statistician using commercially available statistical software (Stata version 16.1; StataCorp).
Results
Feline aMSC harvest population
The mean age of CKD cats from which aMSCs were collected was 6.7 years, while the mean age of healthy cats was 1.8 years. Only male neutered cats were present in the healthy cat population, while two female spayed cats were present in the CKD population. No medications were used at the time of collection in the healthy cats, while the CKD cats were receiving a variety of medications at the time of aMSC collection. Four of the cats in the healthy group (cats 7, 10, 11 and 12) had elevated urinary protein to creatinine levels (>0.4 as per IRIS guidelines). The proteinuria resolved after castration.
aMSC flow cytometry
All cell lines were aMSCs based on plastic adherence, morphology, and a CD44+ CD105+ CD90+ MHC II– phenotype. 3 Cell lines with a >80% of the population in the positive cell markers and <20% in the negative cell marker were considered to be aMSCs.
Immunomodulatory gene expression
There was no significant difference in key immunomodulatory gene expression in naive aMSCs from healthy cats and from cats with CKD for any of the cytokines examined (Figure 1). However, there were significant differences in expression levels for all the cytokines measured between naive aMSCs (media only) and aMSCs following each of the treatments, except for treatment with LPS and PGN (Table 4, Figure 1). The RT-qPCR primer sequences used for each gene and the reference gene are shown in Table 3. Compared with naive cells (media only), the expression levels of IL6 were higher for cells (both healthy and CKD) incubated with either MTX (P <0.001) or Poly I:C (P = 0.012) (Table 4, Figures 2 and 3). Similarly, the expression levels of IL10 were higher for cells that were incubated with CsA (P = 0.035) or MTX (P = 0.002) (Table 4, Figure 2). The expression levels of IL12p40 were higher for cells incubated with CsA (P = 0.007), dexamethasone (P = 0.004) or MTX (P = 0.001) (Table 4, Figure 2). The expression levels of IL18 were higher for cells that were incubated with MTX (P <0.001) (Table 4, Figure 2). Finally, the expression of TGFB was lower for all cells that were incubated with either dexamethasone (P = 0.012) or Poly I:C (P = 0.032) (Table 4, Figures 2 and 3). There was no significant difference between aMSC gene expression in healthy cats or aMSC gene expression in CKD cats for any treatment in culture for any of the cytokines evaluated (P >0.10). One aMSC cell line from a healthy cat (Table 2, patient 8) was consistent with an outlier under two treatments (PGN and medium), and we rejected these outliers (n = 5 [IL-10 medium, IL-12p40 medium, IL-18 medium, IL-10 PGN, IL-12p40 PGN]).
Dot plots of delta cycle threshold (ΔCt) values for five cytokines measured from the cells of six healthy cats (open circles) and six cats with chronic kidney disease (CKD; crosses) after incubation with seven different treatments. One healthy cell line was removed as an outlier for treatments with peptidoglycan (PGN) and medium. No statistical differences were observed for any values relative to cats with and without CKD.
Profile plots of mean (± 95% confidence interval) ΔCt values for five cytokines (interleukin [IL]-6, IL-10, IL-12p40, IL-18, transforming growth factor [TGF]-β) measured from the cells of six healthy cats (open circles) and six cats with chronic kidney disease (CKD) or insufficiency (crosses) after incubation with three immunosuppressive treatments (ciclosporin, dexamethasone [Dex], methotrexate) vs medium. Both healthy and CKD lines are represented to show the similarity between the response of both groups to treatments. One healthy cell line was removed for treatments with peptidoglycan (PGN) and medium as an outlier.
Profile plots of mean (± 95% confidence interval) ΔCt values for five cytokines (interleukin [IL]-6, IL-10, IL-12p40, IL-18, transforming growth factor [TGF]-β) measured from the cells of six healthy cats (open circles) and six cats with chronic kidney disease (CKD) or insufficiency (crosses) after incubation with three viral or bacterial ligands (lipopolysaccharide [LPS], peptidoglycan [PGN], polyinosinic:polycytidylic acid [Poly I:C]) vs medium. Both healthy and CKD lines are represented to show the similarity between the response of both groups to treatments. One healthy cell line was removed for treatments with PGN and medium as an outlier.
Summary of relative cytokine concentrations by study group and incubation treatment of adipose-derived mesenchymal stem cells collected from six healthy cats and six cats with chronic kidney disease (CKD)
(–) were used as the reference for comparison, and the marginal mean was used for comparison of treatments across both CKD and healthy cells
P <0.05 for each group was considered statistically significant
Ct = cycle threshold; CI = confidence interval; IL = interleukin; CsA = ciclosporin; LPS = lipopolysaccharide; MTX = methotrexate; PGN = peptidoglycan; Poly I:C = polyinosinic:polycytidylic acid
Discussion
This study had two major aims. The first was to compare the basal expression of immunomodulatory genes in aMSCs derived from cats with and without CKD. We found no statistical differences in gene expression in the five measured genes relative to CKD status. As a matter of clinical application, this is critically important as using autologous aMSCs is generally less complicated clinically, immunologically and, potentially, from a regulatory perspective. Further in vivo studies are necessary, but for use in transplantation, aMSCs could be harvested prior to the transplantation procedure from the kidney recipient with diminished concern that they may function differently than aMSCs from healthy cats relative to in vitro expression of the genes reported in this study. The five cytokines (IL-6, IL-12p40, IL-10, IL-18 and TGF-β) presented in this study are key immunomodulators released by MSCs to aid in inflammatory and immune responses within the body.
The second major aim of this study was to examine how infection and use of immunosuppressive drugs affects aMSC gene expression using in vitro models of infection (LPS, a product of Gram-negative bacterial infection; PGN from S aureus; and Poly I:C representing double-stranded RNA intermediates common to most viral replication in cells). In this study, we discovered that only Poly I:C, and not LPS or PGN, altered aMSC gene expression for the cytokines measured regardless of renal function. Simulation of immunosuppression was performed by exposure to commonly used immunosuppressive drugs (dexamethasone, MTX and ciclosporin) in culture. Our results showed that immunosuppressive drugs do alter aMSC gene expression for the cytokines measured, regardless of renal function.
CKD and healthy aMSCs
Our results showed there was no statistical difference in gene expression in the five genes evaluated between aMSCs from healthy and CKD cats. The results of one study comparing aMSCs between rats with CKD and healthy rats are consistent with our findings. 34 The study showed no significant difference in the gene expression of stem cell markers, morphology of cells or GAPDH activity between CKD and control groups, 34 but variabilities across patients and studies in the function and response of MSCs to various conditions were still evident.21,34,35 One study showed that aMSCs from rats with CKD exhibited premature senescence and decreased regenerative potential, inhibiting the function of aMSCs in vitro and in vivo. 21 Additionally, studies involving uremic rats have shown alterations in the regenerative potential of MSCs, proliferation of MSCs in culture and other key functions, 21 and the previously mentioned study found that uremic toxin in CKD rats had a small effect on the gene expression and differentiation of aMSCs. 34 Human-induced pluripotent stem cells from patients with diabetic nephropathy and glomerulonephritis were found to have a similar differentiation potential to stem cells from healthy individuals. 21
Regarding cats specifically, this is the first study to evaluate gene expression of aMSCs in cats with and without CKD. Other studies have documented and compared the immunomodulatory potential of aMSCs from young and geriatric cats. aMSCs from geriatric cats took longer to reach passage 2 than young cats, but there was no significant difference in the ability to suppress lymphocyte proliferation or expression of senescence-associated β-galactosidase. 24 Six of the geriatric cats had either CKD or neoplasia. 24 Geriatric cats with a serum creatinine >2 mg/dl had subjectively extended growth time in initial culture, but no conclusions could be drawn as to whether uremia had a negative effect on aMSC function. 24 Extended proliferation and decreased regenerative potential of MSCs in rats with CKD has been noted. 21 Of note, the mean age between the two groups of cats in our study varied greatly (CKD 6.7 years, healthy 1.8 years). Consistent with the previous study, which did not report changes in the immunomodulatory potential of aMSCs from young and geriatric cats, there was no difference between expression of the five genes studied here. 24 The potential effect of age and CKD on cell proliferation was not evaluated in this study but could affect autologous clinical application.Other important aspects of stem cell immunobiology may be influenced by systemic perturbations related to sex and sterilization status. All healthy aMSCs in this study were collected from males that were neutered at the time of fat collection, but the impact of sex on feline aMSC function remains unknown. 2 Steroid administration may also affect stem cell immunobiology. One CKD cell line in our study was collected from a cat receiving prednisolone. The interaction between MSCs and steroids during inflammation has been studied in mice. 36 Concurrent administration of dexamethasone systemically in combination with administration of MSCs intravenously in carbon tetrachloride-induced advanced liver fibrosis resulted in negation of the therapeutic effects of MSCs. 36 Further work on the effects of common pharmacologic therapy for CKD on MSC immunobiology, and more work on other aspects of stem cell function should be continued prior to implementation of MSCs into therapeutic protocols.
Immunosuppressive therapy
In veterinary transplantation, the typical post-renal transplant immunosuppressive protocol commonly includes CsA and steroid administration. 37 The immunosuppressive drugs in this study included CsA and dexamethasone, and concentrations for in vitro use were determined based on cell viability assessment at the estimated in vivo dosage.
MTX was also used in this study, and in vitro dosage was determined in the same manner. While not commonly used in veterinary transplantation protocols, MTX is commonly used in human medicine for graft vs host disease and hematopoetic stem cell transplantation, and is used in veterinary medicine for other applications. 38
In the aMSCs evaluated, exposure to CsA resulted in upregulation of IL12p40 and IL10 gene expression. Exposure of cells to MTX or dexamethasone also resulted in upregulation of IL12p40 gene expression. IL-12 plays a critical role in induction of interferon gamma and T helper-1 cells through the activation of the Janus kinase (JAK)-signal transducer and activator transcription (STAT) pathway, both of which may have clinical consequences in the transplant patient or any patient requiring immunomodulation. 39 Indeed, pre-transplantation elevations in IL-12 and IL-10 have been associated with acute rejection following cadaveric renal transplantation in people. 40 Changes in cytokine production in relation to feline renal transplantation are not as well understood. In a study including five feline renal transplant recipients, lymphocyte cytokine production in vitro was assessed after exposure to CsA, dexamethasone, CTLA4 immunoglobulin and a CsA–dexamethasone combination. The frequency of response of the lymphocytes and proliferative capacity of CD4+ and CD8+ T cells was decreased in immunosuppressed renal transplant recipients. 41 In the light of the findings here, and until further clarity on the clinical consequences of this in cats undergoing transplantation is obtained, clinicians should be aware that immunosuppressive medications may alter aMSC gene expression. How this clinically affects cats or alters treatment efficacy, or how those changes in gene expression relate to circulating cytokine levels, are currently unknown.
The upregulation of IL6 and IL18 gene expression by MTX may also be important in clinical patients. IL-18 is an inflammatory cytokine and important in the pathogenesis of renal injury and renal fibrosis, working through Toll-like receptors and activating protein-1. It is significantly upregulated following kidney injury; therefore, IL-18 may have a role in stimulating renal fibrosis in both mice and humans. 42 If the upregulation of IL18 gene expression by MTX has a clinical consequence of allograft fibrosis, its use with MSCs would be contraindicated, especially in renal transplant recipients. Further research of the role of IL-18 in cats is needed prior to clinical implementation of aMSCs into therapeutic regimens as the effect of IL-18 could be detrimental to feline transplant success.
IL6 expression has been shown to be upregulated as a result of oxidative stress, chronic inflammation and fluid overload with reduced clearance due to impaired renal function in humans and accelerating the progression of CKD. 43 Further upregulation of gene expression with the treatment of MTX could lead to acceleration of the progression or development of CKD in the contralateral kidney or graft. However, the variability of gene expression under various conditions should be further studied prior to altering clinical practice. It has also been previously suggested that MTX has a role in acute kidney injury in cats when used as a chemotherapeutic agent. 44 The administration of MTX to rats has been shown to decrease glutathione levels and increase myeloperoxidase and malondialdehyde levels, resulting in alteration in the oxidate stress response and induction of renal toxicity.45,46 Therefore, further research into the effects of MTX in cats with CKD is indicated.
Dexamethasone exposure to healthy and CKD aMSCs resulted in the downregulation of TGFβ gene expression. TGF-β controls the proliferation, differentiation and function of immune cells as an anti-inflammatory cytokine, 47 and in relation to articular and bone marrow-derived MSCs in humans and rats, respectively, TGF-β has been found to promote differentiation.48,49 One study in rats found that in the presence of estrogen, upregulation of TGF-β, along with other growth factors, resulted in increased osteogenic differentiation of bone-marrow derived MSCs. 49 Other studies not involving MSCs found that dexamethasone reduced TGF-β production in mouse bone, 50 or documented no change in TGF-β following steroid administration.51,52 Clinically, an increased amount of TGF-β in a transplant patient would be desirable to limit inflammation associated with the immune response. Increased renal tubular TGF-β expression in people was linked to improved acute allograft function, but overall effects were negative because of an association with interstitial fibrosis and chronic allograft vasculopathy. 53 Additionally, TGF-β has been described to be upregulated and activated in fibrotic diseases, and has a role in fibroblast phenotype and function in humans; 54 therefore, investigation into TGF-β-mediated pathways as a target for the treatment of feline patients with fibrotic conditions is warranted.
Other studies have shown that anti-inflammatory drugs influence proliferation and differentiation in MSCs. In horses, one study found that non-steroidal anti-inflammatory drugs (NSAIDs) had minimal influence on adipogenic and chondrogenic MSC differentiation, but osteogenic differentiation was significantly affected. 55 In contrast, a study involving human MSCs found that NSAIDs do not interfere with MSCs potential to proliferate and differentiate into osteogenic lineage but did inhibit the chondrogenic differentiation. 56 The study described here did not evaluate MSC proliferation and differentiation potential, and further studies to clarify the impact of common drugs on MSC function are needed.
Bacterial and viral ligands
Viral and bacterial infections, which commonly occur in immunosuppressed patients, were simulated with bacterial and viral ligands, with minimal effect on expression of immunomodulatory genes in feline aMSCs. Although our study did not show a significant increase in IL-6 with LPS or PGN stimulation, other studies in humans have found that LPS significantly increased IL-6 expression in aMSCs. 57 One study in humans found that treatment of MSCs with LPS resulted in upregulation of anti-inflammatory cytokine expression and promotion of macrophage activation. 58 Another study involving murine bone marrow-derived MSCs found that after 1 h of stimulation with LPS the cells produced a proinflammatory affect. 59 Additional studies have found that heat-inactivated E coli and LPS exposure reduced the adipogenic differentiation and induced osteogenic differentiation and proliferation in aMSCs from human adults. 60 It is possible that the length of exposure to LPS in our study was not long enough to elicit the previously observed effect. In a study involving murine bone marrow-derived MSCs, administration of both LPS and Poly I:C ligands resulted in a two-fold increase in the number of bone marrow-derived MSCs 24 h after injection but not before. 61 The variability between our study and others requires further investigation. Cell viability was evaluated and found to be unaffected with the LPS and PGN doses used in the study, but experiments should be replicated to ensure reproducibility.
Along with the increased incidence of viral and bacterial infections in the immunosuppressive state post-renal transplant, the effect of antibiotic administration on aMSCs should also be considered. Systemic administration of activated MSCs with antibiotic therapy has been beneficial for mice and canine patients with chronic, drug-resistant infections. The combination of both MSCs and antibiotics resulted in eradication of S aureus biofilm infections in difficult locations. 62 Further investigation into the role of aMSCs in response to infectious agents and antimicrobials in immunosuppressed veterinary patients along with the use of a larger population of samples is indicated.
Several limitations were present in the study. The small population utilized in the experiments reported here and the uncertainty of how these data relate to aMSC use in a clinical patient with a milieu of other cytokines limits the ability to accurately predict clinical treatment outcomes. Additionally, cells were exposed to treatment conditions for only 24 h. Other studies have found that cytokine production and function vary significantly throughout the exposure time.59,61 The outlier data here also limit the power of the data set. The inadequate expression of the reference gene by the cell line under the two treatment conditions excluded as outliers was not likely due to biologic variation. The reference gene was adequately expressed for the cell line rejected under other treatment conditions. Single-dose and single medication treatments were also used in this study. It is difficult to simulate all types of interactions that will occur in multi-component in vivo settings in a simple, controlled in vitro environment, specifically in relation to protein concentration. Studies have found that differential mRNA expression reflects a difference between conditions at the functional level of proteins, 63 and therefore the overall function of the cell. However, the response of aMSCs to the complex conditions in vivo may not be the same as the in vitro responses we observed in this study.
Conclusions
The results of this study demonstrate that commonly used immunosuppressive therapies and components of viral and bacterial organisms can have an impact on the gene expression of feline aMSCs. They provide an initial step toward understanding the critical interactions governing the role of aMSCs and may have important implications in the design of research to examine the in vivo effects of aMSCs in transplant models and for the future implementation of autologous stem cell therapy following renal transplantation. Further investigation into the function of MSCs in response to relevant drugs and the common types of infections is warranted.
Footnotes
Acknowledgements
The authors gratefully acknowledge Dr Roy D Berghaus for his statistical consultation.
Author note
A portion of the data from this study was previously presented at the 2019 ACVS Surgery Summit.
Conflict of interest
The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Funding
This work was partially supported by a grant from the University of Georgia Office of the Provost and the College of Veterinary Medicine Research Committee.
Ethical approval
The work described in this manuscript involved the use of experimental animals and the study therefore had prior ethical approval from an established (or ad hoc) committee as stated in the manuscript. The work also described in this manuscript involved the use of non-experimental (owned or unowned) animals and procedures that differed from established internationally recognized high standards (‘best practice’) of veterinary clinical care for the individual patient. The study therefore had prior ethical approval from an established (or ad hoc) committee as stated in the manuscript.
Informed consent
Informed consent (either verbal or written) was obtained from the owner or legal custodian of all animal(s) described in this work (either experimental or non-experimental animals) for the procedure(s) undertaken (either prospective or retrospective studies). No animals or humans are identifiable within this publication, and therefore additional informed consent for publication was not required.
