Cardiac regenerative potential of cardiosphere-derived cells from adult dog hearts

The regenerative potential of cardiosphere-derived cells (CDCs) for ischaemic heart disease has been demonstrated in mice, rats, pigs and a recently completed clinical trial. The regenerative potential of CDCs from dog hearts has yet to be tested. Here, we show that canine CDCs can be produced from adult dog hearts. These cells display similar phenotypes in comparison to previously studied CDCs derived from rodents and human beings. Canine CDCs can differentiate into cardiomyocytes, smooth muscle cells and endothelial cells in vitro. In addition, conditioned media from canine CDCs promote angiogenesis but inhibit cardiomyocyte death. In a doxorubicin-induced mouse model of dilated cardiomyopathy (DCM), intravenous infusion of canine CDCs improves cardiac function and decreases cardiac fibrosis. Histology revealed that injected canine CDCs engraft in the mouse heart and increase capillary density. Out study demonstrates the regenerative potential of canine CDCs in a mouse model of DCM.


Introduction
Numerous animal studies [1][2][3][4][5][6][7][8][9][10][11][12] and the first-in-human CADUCEUS trial [13,14] have demonstrated the regenerative potential of cardiospherederived cells (CDCs) in ischaemic cardiomyopathy. Like human beings, veterinary patients such as domestic dogs also suffer from heart diseases. For instance, the overall prevalence of dilated cardiomyopathy (DCM) in Doberman Pinschers is greater than 50% [15]. Once heart failure occurs, only palliative and sympathetic treatments exist. It is unknown whether canine CDCs could be derived from adult dogs and whether they are as potent as their human and rodent counterparts in cardiac regenera-tion. On the other hand, there is a growing need to improve (large) animal models for cell-based regenerative medicine [16]. Dogs with naturally occurring heart diseases may serves as animal models of human diseases. Here, we derived CDCs from the adult dog heart and tested their regenerative potency in various in vitro and in vivo assays. mately, 6 mm 9 6 mm piece of myocardial tissue was separated and washed with PBS (Life Technologies, Carlsbad, CA, USA). The tissue sample was then cut into smaller biopsy-sized pieces, and washed three times with PBS, followed by enzymatic digestion at 37°C in 5 mg/ml collagenase IV solution (Sigma-Aldrich, St. Louis, MO, USA) for 5 min. Iscove's Modified Dulbecco's Media (IMDM; Life Technologies, Carlsbad, CA, USA) containing 20% foetal bovine serum (FBS; Corning) is then added to the sample to inactivate the collagenase. After that, the tissue samples were further minced into smaller tissue explants (~0.5 9 0.5 mm) before plating. Approximately, 50 pieces of tissue explants were then placed onto a fibronectin-coated plate with approximately 1.5 cm between each explant and covered with 2 ml of IMDM with 20% FBS overnight to aid the attachment of tissue explants. The cultures were maintained in IMDM with 20% FBS and media change was performed every other day. In about 1 week, cells started to outgrow from the tissue explants. Once these outgrowth cells are about 70-80% confluent, they were harvested by 5-10 min. of incubation with TryPLE Select TM (Life Technologies). The cells were then seeded into an Ultra-Low-attachment flask (Corning) at a density of 100,000 cells/cm 2 and cultured in IMDM with 10% FBS. Phasebright canine cardiospheres (CSps) started to form in 3-7 days. Canine CSps were then collected from the low-attachment flasks and re-plated onto fibronectin-coated surface to produce adherent canine CDCs. Canine CDCs were cultured in IMDM with 20% FBS media and passaged every 3-5 days. We used Passage 2-3 CDCs for all in vitro and in vivo testing. Canine mesenchymal stem cells (MSCs) were derived from the femur bone marrow of the dog donor and used as the Control cells. Canine MSCs were maintained in the same aforementioned CDC culture media [11,12].

Clonal growth assay
Canine CDCs were seeded into a 96 well plate at a density of 1 cells per 100 ll per well. Only wells containing one cell were used for the experiment. Clonal growth of canine CDCs were tracked with a phasebright microscope during 1-week period of time.

Paracrine assay
Canine CDCs or MSCs were plated into a six-well plate and incubated in plain (serum-free) IMDM at a density of 1 million cell per 1 ml per well for 3 days. After that, the conditioned media (CM) from CDCs was harvested. Neonatal rat cardiomyocytes (NRCMs) were derived as described [19,20] and plated onto fibronectin-coated 6 well plate (Corning) with 1 9 10 6 cells with Medium 199 (Gibco). Neonatal rat cardiomyocytes were incubated with CDC-CM, MSC-CM, or plain IMDM as the Control. After 3 days, cells were fixed with 4% PFA and apoptotic cells were detected by terminal deoxynucleotidyl transferase dUTP nick end labelling (TUNEL) assay using the In Situ Cell Death Detection Kit (Roche Diagnostics, Mannheim, Germany), according to the manufacturer's instructions and DAPI for nuclei. Myocyte size quantification was performed by anti-a-SA antibody staining followed by cell area measurement using the NIH Image J software (Bethesda, MD, USA).
The pro-angiogenic effects of CDC-CM were studied by endothelial cell tube formation assay. Human umbilical vein endothelial cells (HU-VECs; from ATCC) were seeded onto growth factor-reduced Matrigel TM (BD Biosciences) in 96-well plates at a density of 2 9 10 4 cells per well. 100 ll of CDC-CM or MSC-CM were added into the wells. After 4 hrs, the wells were imaged with a Nikon (Chiyoda, Tokyo, Tokyo, Japan) white light microscope. The average tube length was then measured with NIH Image J Software.

Animal procedures
All animal work is compliant with Institutional Animal Care and Usage Committee at North Carolina State University. We employed an acute doxorubicin-induced DCM model as previously described [21]. Briefly, 6-to 8-week old female severe combined immunodeficiency mice (Charles River Laboratories, Wilmington, MA, USA) were given 10 mg/kg body weight intraperitoneally injected (bw) Doxorubicin to create acute DCM on Day 0. On Day 1, the mice were then randomized into the following two treatment groups (n = 9-11 mice per group): (i) Saline Control: Mice receiving 100 ll PBS injected through the tail vein; (ii) Cell Treatment: Mice receiving 100 ll of 1 9 10 6 canine CDCs in 100 ll PBS injected through the tail vein. anaesthetized with a 1.5% isofluorane-oxygen mixture and two-dimensional long axis images were record from the left caudal (apical) view, using a Philips CX30 ultrasound system couple with a L15 high-frequency probe. Twodimensional guided M-mode images at chordae tendineae level were evaluated. M-mode measurements of left ventricle end-diastolic and end-systolic dimensions (LVEDD and LVESD, respectively) were performed by using the leading-edge method of the American Society of Echocardiograph [22]. For estimation of each parameter, the average of three measurements from three different cycles in an image was obtained. Left ventricular end-diastolic and systolic volumes (LVEDV and LVESV, respectively) were calculated by the biplane method of disks (modified Simpson's rule). Ejection fraction (EF) was determined by using (LVEDV -LVESV/LVEDV) 9 100%, and fractional shortening (FS) was calculated from the M-mode echocardiography images as (LVEDD -LVESD/LVEDD) 9 100%.

Histology
All animals were killed 8 days after treatment. Mouse hearts were harvested and frozen in Optimal Cutting Temperature (OCT) compound (Tissue-Tek, Torrance, CA, USA). Cryo-sections (5 lm thick) were prepared. Masson's trichrome staining was performed as per manufacturer's instructions [HT15 Trichrome Staining (Masson) Kit; Sigma-Aldrich]. Fibrosis area was measured by NIH Image J as previously described [23]. For immunofluorescence staining, mouse heart cryosections were fixed with 4% PFA, blocked/permeabilized with Protein Block Solution (DAKO) containing 1% saponin (Sigma-Aldrich), and then stained with anti-a-SA (Sigma-Aldrich) and anti-von Willebrand factor (Abcam) antibodies. FITC or Texas-Red secondary antibodies were obtained from Abcam and used in conjunction with these primary antibodies. Images were taken by a Zeiss (Jena, Germany) confocal microscopy system.

Statistical analysis
Results are presented as mean AE SD unless specified otherwise. Comparisons between any two groups were performed with two-tailed unpaired Student's t-test. Comparisons among more than two groups were performed with one-way ANOVA followed by post hoc Bonferroni correction. Differences were considered statistically significant when P < 0.05.

Generation of canine CDCs
Using a three-stage 'adhesion-suspension-adhesion' culture process (Fig. 1A), we derived CDCs from dog myocardial tissues. Both phasebright and stromal-like cells started to outgrow from the canine heart tissue explants in a week after plating onto fibronectin-coated surfaces. Those outgrowth cells become confluent in~2-3 weeks (Fig. 1B). When seeded on low-attachment surfaces (to discourage cell attachment), the outgrowth cells spontaneously aggregate into three-dimensional canine cardiospheres (Fig. 1C). To streamline the cardiosphere forming process, we used Ultra-Low attachment surface instead of the previously reported poly-D-lysine coating method. We demonstrated that cardiosphere formation and the potency of the resulted CDCs of the Ultra-Low method are indistinguishable from those of the poly-Dlysine method (Fig. S1). When replated onto a fibronectin-coated surface, the cardiospheres dissociated into single cells which we termed CDCs (Fig. 1D and E). One biopsy-sized canine heart tissue can generate 50-200 millions of Passage 0 CDCs. When maintained and passaged in IMDM with 20% FBS, canine CDCs can further undergo 25-30 doublings in 35 days (Fig. 1F). Clonal growth was observed in canine CDCs (Fig. 1G). Flow Cytometry analysis and immunocytochemistry (Fig. 1H) revealed that canine CDCs were highly positive for CD105 while being negative for haematopoietic marker CD45. Only a negligible fraction of canine CDCs express CD90 and ckit. As a positive control, canine MSCs uniformly express CD90 (Fig. S2).

Differentiation potential of canine CDCs
After 12 days into differentiation, around 20% of canine C DCs started to express cardiomyocyte marker a-SA ( Fig. 2A and B) or cardiac troponin I (Fig. S3) while only~5% of canine MSCs expressed a-SA ( Fig. 2A and B). Instead, canine MSCs were potent in differentiation into smooth muscle cells:~90% of canine MSCs express alpha smooth muscle actin. The endothelial differentiation potentials of canine CDCs and MSCs were similar. These assays confirmed the cardiovascular differentiation potential of canine CDCs.

Paracrine assays on canine CDCs
Incubation in CDC-CM increased the size/spreading of NRCMs (CDC-CM versus IMDM: 1586.2 AE 87.7 versus 1030.6 AE 52.5 lm 2 ) (Fig. 3A), while MSC-CM did not have any positive effects on cardiomyocyte size (997.1 AE 56.7 lm 2 ). In addition, CDC-CM promoted the contractility of NRCMs as compared IMDM and MSC-CM (Videos S1-S3). In addition, CDC-CM protects cardiomyocytes from apoptosis: NRCMs incubated with IMDM and MSC-CM had a cell apoptosis rates (gauged by TUNEL staining) of 12.1 AE 2.1% and 12.3 AE 2.0%, respectively, while NRCMs incubated with CDC-CM had a markedly lower apoptosis rate of 5.3 AE 1.3% (Fig. 3B). These compound data suggest canine CDCs are capable of promoting cardiomyocyte contraction and survival through paracrine mechanisms [24]. Moreover, CDC-CM promotes tube formation of human umbilical vein endothelial cells on Matrigel TM (Fig. 3C): average tube length in MSC-CM versus CDC-CM: 19.5 AE 1.2 versus 33.3 AE 2.6 lm, suggesting its pro-angiogenic role in cardiac regeneration.

Canine CDC treatment improves cardiac function and reduces fibrosis
The animal study design is outline in (Fig. 4A). Heart sections were stained with Masson's Trichrome staining kit and analysed for fibrosis. Canine CDC injection significantly reduced fibrosis when compared to Control-treated hearts (Fig. 4B, Fig. S3) (fibrosis area % of hearts in  Control versus CDC therapy: 13.2 AE 1.4 versus 4.0 AE 0.8%). The bona fide therapeutic effects from stem cell therapy would be the improvement of heart pump functions. Echocardiography was performed at baseline and 7 days after cell or saline injection. Treatment effects (DEF% and DFS%) were calculated as the change in cardiac functions from the baseline. Saline injection had a negative treatment effects as EFs and FSs deteriorated over the 1 week time ( Fig. 4C and D, black bars). Cardiosphere-derived cell treatment ( Fig. 4C and D, red bars) protected cardiac functions from deteriorating because of the doxorubicin. These data suggest that canine CDC therapy mitigates fibrosis and pump function deterioration in DCM.

Canine CDCs engraft in the mouse heart, promotion of angiogenesis and apoptosis reduction
DiI-positive canine CDCs were detected in the mouse heart (Fig. 5A). However, very few engrafted cells acquired mature cardiomyocyte or endothelial phenotypes. Mounting lines of evidence suggest that stem cell transplantation (including CDCs) exerts benefit through paracrine mechanisms, i.e. transplanted cells secrete factors to promote endogenous repair [24]. Cardiosphere-derived cell therapy significantly increased vascular density in the DCM heart: vWF-positive vasculatures % of total nuclei in Control-treated hearts versus CDC-treated hearts: 11 AE 1.6 versus 30.4 AE 2.9% ( Fig. 5B and C). Cardiospherederived cell treatment reduced apoptosis in the post-MI hearts (TUN-EL-positive cells in Control versus CDC therapy: 1.44 AE 0.23 versus 0.53 AE 0.09%) (Fig. 5D and E).

Discussion
Only a few families in the world are not affected by cardiovascular diseases. As a result of the magnitude-high prevalence-high incidence of cardiovascular disease in the world, stem cell therapy offers a promising option for therapeutic cardiac regeneration. The last decade witnessed a burst of cell therapy trials for ischaemic cardiomyopathy. For the last 6 years, our lab has been studying CDCs. Results from a recent clinical trial indicate that infusion of autologous CDCs in mildto-moderate heart attack patients reduces scar and increases viable tissue. A Phase II clinical trial is ongoing to test the regenerative potential of allogeneic CDCs in patients with recent MI [25].
Veterinary patients such as domestic dogs suffer from heart diseases as well. For instance, it has recently been found that the overall prevalence of DCM in Doberman Pinschers is greater than 50% [18]. The expected survival time after diagnosis is strikingly less than 2 months. The autosomal dominant inheritance pattern in Doberman Pinschers is similar to most forms of DCM in human's heart [26,27]. Once heart failure has occurred, treatment is only symptomatic and palliative. It is intriguing to wonder whether CDCs can be derived from donor dog hearts and used for therapeutic regeneration in dog patients with DCM. In addition, there is a growing interest from grant agencies such as National Institute of Health to use domestic dogs with naturally occurring heart diseases as model systems for cellbased cardiac regenerative therapies.
In this study, we demonstrate that canine CDCs can be derived from adult dog hearts and they are very phenotypically similar to human and rodent CDCs (Fig. 1H). However, canine CDCs have low expressions in CD90 and ckit. Recent report from us and others has shown that ckit expression is irrelevant to the overall therapeutic benefit of CDCs and CD90 expression indeed undermines the regener-ative potential of CDCs [18,28]. Therefore, the low expressions in CD90 and ckit should not affect the therapeutic benefits of canine CDCs. Canine CDCs are cardiac stem cells and can express markers of cardiomyocytes, smooth muscle cells and endothelial cells in vitro (Fig. 2). Consistent with our previous findings, the cardiac differentiation potential of canine CDCs is greater than canine MSCs [8].
Mounting lines of evidence indicate that CDCs exert their therapeutic benefit through paracrine mechanisms. We collected CM from canine CDCs and MSCs and study its effects on cardiomyocytes and endothelial cells. Compared to MSC-CM, CDC-CM is more potent in promoting myocyte contraction, inhibiting myocyte cell apoptosis and promoting endothelial cell tube formation on Matrigel (Fig. 3).  Additional staining using cardiac troponin I is also included in the supplemental data (Fig. S4). We induced acute DCM in immunodeficiency mouse with injection of 10 mg/kg doxorubicin. This model has been reported to successfully induce DCM in less than 7 days [19]. Injection of canine CDCs ameliorates ventricular dysfunction (Fig. 4C) and reduces cardiac fibrosis (Fig. 4B). Such observed therapeutic benefits are not accompanied by sizable cell engraftment: only a few of DiI-positive canine CDCs (Fig. 5A) were detected in the mouse heart and even fewer differentiated into mature cardiac cells. The mechanisms underlying the therapeutic benefits of CDCs are not fully elucidated. It has been reported that CDCs are highly proangiogenic. Following this lead, we found that canine CDC treatment enhanced angiogenesis in the DCM heart (Fig. 5B). Cardiosphere-derived cell treatment also reduced apoptosis when compared to no treatment (Fig. 5D). The reduction in fibrosis may be a result of matrix metaloproteneases secreted by CDCs [29]. It has also been reported that CDC therapy stimulates cardiomyocyte proliferation and recruitment of endogenous cardiac stem cells [9]. Taking together, these mechanisms could explain the therapeutic benefits of CDCs in DCM.
In summary, we derived canine CDCs from adult dog hearts and showed their regenerative potential in a mouse model of induced DCM. Future studies will focus on testing canine CDCs in dogs with naturally occurring heart diseases such as DCM.    Video S1 NRCMs cultured with CDC-CM.