Safety and effects of intravenous frozen donor stem cell treatment for chronic kidney disease in cats from three pilot studies

Cats with stable CKD, serum creatinine 1.6 to 5.0 mg/dl, were recruited from the patient population at the Veterinary Teaching Hospital at Colorado State University. Cats were determined to have stable CKD based on two repeated biochemical evaluations performed at least 2 weeks apart and ultrasonographic evidence of CKD. Pretreatment evaluation included complete blood count (CBC), biochemistry profile, urinalysis, urine culture, blood pressure, total T4, urine protein creatinine ratio (UPC), feline leukemia/feline immunodeficiency virus serology, and a renal ultrasound. Cats were excluded from the study if they had evidence of ureteroliths, pyelonephritis, anatomic abnormalities such as polycystic kidney disease or masses, uncontrolled hypertension, or concurrent systemic disease. Administration of concurrent supportive therapies was allowed provided there were no changes in therapy during the study period. The study was approved by the Institutional Animal Care and Use Committee at Colorado State University (#10-1603A and 11-2915A), and all owners reviewed and signed consent forms prior to participation in the study.

MSCs were isolated from adipose tissue of donor SPF cats obtained from the same subcutaneous site on the ventral abdomen just caudal to the umbilicus, as previously described [,]. For isolation of the stromal vascular fraction, the tissue was minced and digested with 1 mg/ml collagenase (Sigma Aldrich, St Louis, MO, USA) for 30 minutes at 37°C. The sample was centrifuged, and the stromal vascular fraction was plated in MSC medium (low glucose Dulbecco’s modified Eagle medium (DMEM), 100 U/ml penicillin, 100 μg/ml streptomycin, 2 μM-glutamine, 1% essential amino acids without-glutamine, 1% non-essential amino acids, 0.075% sodium bicarbonate (Invitrogen/Gibco, Carlsbad, CA, USA) plus 15% fetal bovine serum (Cell Generation, Fort Collins, CO, USA)). The aMSCs were incubated until approximately 70% confluent with media changes every 2 to 3 days. After reaching passage 3, cells were harvested, divided into treatment aliquots in freezing medium (11% dimethyl sulfoxide, 14% MSC medium, 75% fetal bovine serum), and stored in liquid nitrogen for no greater than 1 year prior to use. For injection in pilot studies 1 and 2, cryopreserved cells were removed from liquid nitrogen, incubated in MSC medium for 10 minutes, washed three times in Dulbecco’s phosphate buffered saline (DPBS; Sigma) and then counted and checked for viability using Trypan Blue (cells with less than 95% viability were not used). For cats in pilot study 1, 2 × 10cells were resuspended in 5 ml Hank’s balanced salt solution (HBSS) with 200 IU heparin sulfate and administered slowly by hand over 15 minutes; aMSCs were administered at weeks 0, 2 and 4. For cats in pilot study 2, 4 × 10cells were initially resuspended in 10 ml HBSS with 200 IU heparin sulfate and administered as a slow intravenous push over 30 minutes. However, the first cat to receive MSCs in this manner experienced an increased respiratory rate. Subsequently, cells for the high-dose group were resuspended in 20 ml HBSS/200 IU heparin and administered over 45 to 60 minutes with a syringe pump with frequent agitation of the syringe to prevent settling of the cells. Cats in pilot study 2 received aMSCs at weeks 2, 4, and 6. All three injections received by cats treated in pilot studies 1 and 2 were from the same donor cat. A total of four donor cats were used to provide aMSCs for those studies. ^{21 22} l l 6 6

MSCs were isolated from adipose tissue of donor SPF cats as described above. For preparation of the adipose tissue for cryopreservation, the tissue was minced and divided into 1-g aliquots in 1 ml of freezing medium (11% dimethyl sulfoxide, 14% MSC medium, 75% fetal bovine serum), and stored in liquid nitrogen for no greater than 1 year prior to use. Cryopreserved adipose was then later thawed, immediately washed twice with DPBS, and then prepared for culture as described above. For injection in pilot study 3 cats, cells were harvested by trypsinization at passage 3 and resuspended in 10 ml DPBS with 200 IU heparin sulfate and administered as a slow intravenous push over 20 minutes. Cats in pilot study 3 received aMSCs at weeks 2, 4, and 6. All three injections received by cats treated in pilot study 3 were from different donor cats given in a prerandomized order. A total of three donor cats were used to provide aMSCs for the study.

aMSCs cultured from cryopreserved adipose tissue were characterized by surface marker expression using flow cytometry and a panel of crossreactive antibodies specific for surface determinants expressed by MSCs from other species [-]. Specifically, feline aMSCs were analyzed for surface expression of CD44 (anti-mouse/human, antibody clone:IM7, eBioscience, San Diego, CA, USA) and CD90 (anti-human, antibody clone:eBio5E10, eBioscience). MSCs were also assessed for expression of CD4 (anti-feline antibody clone: 3-4F4, Southern Biotech, Birmingham, AL, USA) and major histocompatibility complex (MHC) class II (anti-feline antibody clone: TU39, BD Bioscience, San Jose, CA, USA). Samples were analyzed using a Cyan ADP flow cytometer (Beckman Coulter, Brea, CA, USA). Approximately 25,000 events were collected for analysis per sample. ^{23 26}

differentiation assays were conducted to confirm the multipotency of feline aMSCs, as assessed by their ability to differentiate into three cell lineages (osteoblasts, chondrocytes, and adipocytes) that are characteristic of MSCs []. For differentiation into adipocytes, aMSCs at confluency were incubated with MSC medium supplemented with 0.5 μM dexamethasone (Sigma Aldrich), 50 μM indomethacin (Sigma Aldrich) and 0.5 μM 3-isobutyl-1-methylxantine (Sigma Aldrich) for 3 weeks with media changes every 3 to 4 days. Chondrogenic differentiation medium consisted of DMEM 1 × (Cellgro, Manassas, VA, USA) supplemented with 15% fetal bovine serum (FBS; Cell Generation), 10 nM dexamethasone (Sigma Aldrich), 10 ng/ml transforming growth factor (TGF)-β (R&D Systems, Minneapolis, MN, USA), 50 μg/ml ascorbic acid (Sigma Aldrich), and 40 μg/ml proline (Sigma Aldrich). Osteogenic differentiation medium consisted of MSC medium supplemented with 10 nM dexamethasone, 50 μM ascorbic acid, and 20 mM β-glycerophosphate (Sigma Aldrich) At the end of the differentiation period, cells were fixed with 10% neutral buffered formalin and stained with Oil Red O (Sigma Aldrich) for presence of lipid or with Alizarin Red (Sigma Aldrich) for the presence of calcium []. Cell pellets from cartilage differentiation were harvested and placed in OCT Compound (Sakura Finetek USA, Inc., Torrance, CA, USA) and flash frozen prior to staining with toluidine blue (Richard-Allan Scientific, Kalamazoo, MI, USA) for cartilage matrix. aMSCs cultured in MSC medium alone under identical conditions were used as differentiation controls. In vitro ^{27 27}

Urine cytokine concentrations were determined using commercial ELISA kits, as we have described recently []. Briefly, urine concentrations of interleukin (IL)-6, IL-8, and IL-10 were measured using feline specific ELISA kits (R&D Systems), while monocyte chemoattractant protein 1 (MCP-1) and vascular endothelial growth factor (VEGF) were measured using commercial canine MCP-1 and VEGF kits respectively (R&D Systems). Urine transforming growth factor β1 (TGF-β1) concentrations were measured using an ELISA kit from Invitrogen (Camarillo, CA, USA). ³¹

Urine samples were collected by cystocentesis, and an aliquot of urine was immediately frozen at -20°C and then stored at -80°C for cytokine measurements. Samples were thawed immediately prior to analysis and kept on ice during sample preparation. All samples were centrifuged at 2000 rpm (350) for 5 minutes at 4°C, and the supernatant was collected for analysis. With one exception, ELISA kits were run according to manufacturer instructions for serum samples. As one exception, urine samples were diluted 5-fold rather than 40-fold for detection of TGF-β1. g

Urine and serum creatinine concentrations were measured with a Roche Cobas Integra Chemistry Analyzer (Roche Diagnostics Limited, West Sussex, UK) at the Colorado State University Diagnostic Laboratory. All reported cytokine concentrations were normalized for differences in urine concentration as a urine cytokine-to-urine creatinine ratio. In order to maximize cytokine yield in the urine samples, additional sample processing steps were performed as previously described prior to running the IL-8, MCP-1, and VEGF ELISA kits []. ³¹

Changes in GFR, serum creatinine, blood urea nitrogen (BUN), electrolytes, urine protein-to-creatinine (UPC) ratio, urine specific gravity, and packed cell volume (PCV) data over time in the aMSC-injected cats were compared statistically by means of repeated measures analysis of variance (ANOVA) followed by Bonferroni’s correction. Values were considered statistically different for<0.05. Statistical analyses were performed using Prism5 software (GraphPad, La Jolla, CA, USA). P

A total of 165 cats with CKD were screened for entry into this series of pilot studies. In all, 52 cats were excluded for geographical reasons (extensive travel required), 40 cats were excluded due to concurrent illness, 18 cats were excluded due to end-stage disease, 13 cats were excluded as the owners thought the study too complex, 11 cats were excluded due to unstable disease, 8 cats were excluded for anatomic abnormalities, and 2 cats were excluded due to fractious attitude in hospital. Descriptive data for the 21 cats with CKD that were enrolled in the aMSC pilot studies are presented in Table. Two cats that were enrolled in pilot study 3 were unable to finish the trial due to development of unrelated illnesses. A summary of cat weights throughout the study and aMSC dose by weight are presented in Table. No statistically significant difference in weight was seen in any pilot study as determined by a Wilcoxon sign rank test. 1 2

Duringculture feline aMSCs were observed to develop into a relatively homogeneous population of plastic-adherent cells with fibroblast-like morphology. Adipose-derived MSCs expressed high levels of CD44 and CD90 and were negative for expression of CD4 and MHC class II (Figure). Both cryopreserved aMSCs and aMSCs cultured from cryopreserved fat were capable of trilineage differentiation (Figure). in vitro 1 2

Cats in pilot study 1 treated with three separate intravenous infusions of 2 × 10cryopreserved aMSCs and cats in pilot study 3 treated with three separate intravenous infusions of 4 × 10aMSCs cultured from cryopreserved fat tolerated their injections without apparent adverse effects. During the infusions and immediately after, the cats appeared clinically normal and unperturbed. However, cats in pilot study 2, each having received an intravenous infusion of 4 × 10cryopreserved aMSCs, experienced a number of treatment-related adverse effects. The adverse effects included vomiting (two cats) and increased respiratory rate (four cats). In two cats, vomiting and nausea occurred within 2 minutes of initiating the first aMSC infusion and were not associated with respiratory distress. Administration of diphenhydramine (1 mg/kg subcutaneously) and maropitant (1 mg/kg subcutaneously) resulted in cessation of clinical signs of vomiting and nausea. Affected cats were premedicated with diphenhydramine and maropitant for subsequent treatments, and additional adverse gastrointestinal reactions were not noted. One cat that experienced vomiting at the time of the first injection subsequently developed increased respiratory rate at the time of the third aMSC injection. 6 6 6

Increased respiratory rate and effort following aMSC infusion was not pronounced in three of the four pilot study 2 cats that experienced respiratory signs but were quite pronounced in one cat. Average baseline respiratory rate was 44 breaths/minute (range 36 to 60) and average increased respiratory rate was 80 breaths/minute (range 60 to 112 breaths/minute). Increased respiratory rate and effort was generally noted after approximately two-thirds of the aMSCs had been administered (that is, after 30 minutes of infusion) in the three cats with mild respiratory signs. One cat, however, had mildly increased respiratory rate on the first infusion, a normal second infusion, and developed marked overt respiratory distress on the third infusion after approximately one-third of the aMSCs had been administered. The respiratory distress was characterized by a sudden onset of increased respiratory rate that quickly led to nasal flare, open mouth breathing, and an orthopnic stance with head extended and elbows out. The injection was stopped, the cat was moved to a critical care ward and placed in an oxygen chamber, and diphenhydramine (1 mg/kg subcutaneously) was administered. After 20 minutes no improvement was noted, and the cat became notably nauseated (foaming at the mouth and subsequently vomiting). Maropitant (1 mg/kg subcutaneously) and dexamethasone (0.05 mg/kg intravenously) were administered. Clinical signs began to abate within 20 minutes of administration of dexamethasone. Open mouth breathing and orthopnea improved within 1 h. Increased respiratory effort continued for the next 8 h, but by 24 h post injection the cat was clinically normal and was discharged without further incident.

Serum creatinine concentrations decreased significantly (= 0.01) in pilot study 1 cats over the study period but were unchanged in cats in pilot studies 2 and 3 (FigureA-C), however it is our opinion that the degree of observed decrease (<0.5 mg/dl) in serum creatinine concentrations is of questionable clinical significance. It should be noted that one cat from pilot study 1 was removed from analysis of serum creatinine concentration because the owners failed to administer fluids during the week prior to assessment, in violation of the study protocol. Values for BUN, serum phosphorus, serum potassium, packed cell volume, and urine specific gravity or UPC did not change significantly following administration of aMSCs as compared to pretreatment values in any of the three pilot studies (Table). P 3 3

The GFR was not measured in pilot study 1 cats. In pilot study 2 cats, there was an overall trend towards improvement in estimated GFR by iohexol clearance values (= 0.056), with individual increases of (75%, 10%, 61%, 7%, 6%) compared to untreated CKD control cats (7%, -18%, -22%) (FigureD). In pilot study 3 cats there was no statistically significant difference in total or individual kidney GFR values as determined by nuclear scintigraphy however variable individual increases were seen (63%, 7%, -34%, 13%, 91%) (FigureE). P 3 3

In cats in pilot study 1, urine MCP-1 and IL-8 concentration data following MSC administration as measured by ELISA are depicted in FigureA,B. Using repeated measures ANOVA, a statistically significant decrease in MCP-1 (= 0.0001) as well as in IL-8 (= 0.01) was detected. However due to the variability in cytokine concentration changes (some cats decreased significantly while others increased), these results may not have clinical relevance. 4 P P

In cats in pilot studies 2 and 3, urine concentrations of IL-8, MCP-1, TGF-β1 and VEGF were assessed (FigureC-I). In the case of all four cytokines measured, repeated measures ANOVA failed to detect any significant changes in cytokine concentrations when pretreatment urine cytokine concentrations were compared to samples taken throughout the treatment period. In pilot study 3, MCP-1 levels were inconsistently detected in study samples. 2