Changhai Road, Shanghai, People’s Republic of …24 Parth U. Thakker, [email protected] 25 Mehran Abolbashari, [email protected] 26 Tae-Hyoung Kim, [email protected]

1

Reno-protection of Urine-derived Stem Cells in A Chronic Kidney Disease Rat Model Induced by 1

Renal Ischemia and Nephrotoxicity 2

3

4

Chao Zhang1,2, Sunil K. George1, Rongpei Wu1,3, Parth Udayan Thakker4, Mehran Abolbashari1, 5

Tae-Hyoung Kim1,5, In Kap Ko1, Yuanyuan Zhang1*, Yinghao Sun2, John Jackson1, Sang Jin Lee1, 6

James J. Yoo1, and Anthony Atala1 7

8

9

1 Wake Forest Institute for Regenerative Medicine, Wake Forest School of Medicine, Medical 10

Center Boulevard, Winston-Salem, NC, USA 11 2 Department of Urology, Changhai Hospital, the Second Military Medical University, 168 12

Changhai Road, Shanghai, People’s Republic of China 13 3 Department of Urology, First Affiliated Hospital of Sun Yat-Sen University, Guangzhou, Guang 14

Dong, People's Republic of China. 15 4 Department of Urology, Wake Forest Baptist Medical Center, Medical Center Boulevard, 16

Winston-Salem, NC, USA 17 5 Department of Urology, College of Medicine, Chung-Ang University, Seoul, South Korea 18

19

20

Chao Zhang, [email protected] 21

Sunil K. George, [email protected] 22

Rongpei Wu, [email protected] 23

Parth U. Thakker, [email protected] 24

Mehran Abolbashari, [email protected] 25

Tae-Hyoung Kim, [email protected] 26

In Kap Ko, [email protected] 27

Yuanyuan Zhang, [email protected] 28

Yinghao Sun, [email protected] 29

John Jackson [email protected] 30

Sang Jin Lee, [email protected] 31

James J. Yoo, [email protected] 32

Anthony Atala, [email protected] 33

*Corresponding author: Yuanyuan Zhang, [email protected] 34

2

Abstract 35

Purpose: Drug-induced nephrotoxicity can occur in patients with pre-existing renal dysfunction 36

or renal ischemia, potentially leading to chronic kidney disease (CKD) and end-stage renal 37

disease (ESRD). Prompt treatment of CKD and the related side effects is critical in preventing 38

progression to ESRD. The goal of this study was to demonstrate the therapeutic potential of 39

urine-derived stem cells (USC) to treat chronic kidney disease-induced by nephrotoxic drugs and 40

renal ischemia. 41

Materials and methods: Human USC were collected, expanded and characterized by flow 42

cytometry. A CKD model was induced by creating an ischemia-reperfusion injury and gentamicin 43

administration. Twenty-eight adult immunodeficient rats were divided into three groups: 44

PBS-treated group (n=9), USC-treated group (n=9), and sham group with age-matched control 45

animals (n=10). Cell suspension of USC (50 x 106 / 100µl / kidney) or PBS was injected bilaterally 46

into the renal parenchyma 9 weeks after CKD model creation. Renal function was evaluated by 47

collection blood and urine samples to measure serum creatinine and glomerulus filtration rate. 48

The kidneys were harvested 12 weeks after cell injection. Histologically, the extent of 49

glomerulosclerosis and tubular atrophy, the amount of collagen deposition, interstitial fibrosis, 50

inflammatory monocyte infiltration, and expression of transforming growth factor beta 1 (TGF-ß1), 51

and superoxide dismutase 1 (SOD-1) were examined. 52

Results: USC expressed renal parietal epithelial cells (CD24, CD29 and CD44). Renal function, 53

measured by GFR and serum Cr in USC-treated group were significantly improved compared to 54

PBS-treated animals (p<0.05). The degree of glomerular sclerosis and atrophic renal tubules, 55

the amount of fibrosis, and monocyte infiltration significantly decreased in USC-treated group 56

compared to the PBS group (p<0.05). The level of TGF-ß1 expression in renal tissues was also 57

significantly lower in the PBS group, while the level of SOD-1 expression was significantly 58

elevated in the USC group, compared to PBS group (p<0.05). 59

3

Conclusions: The present study demonstrates the nephron-protective effect of USC on renal 60

function via anti-inflammatory, anti-oxidative stress, and anti-fibrotic activity in a dual-injury CKD 61

rat model. This provides an alternative treatment for CKD in certain clinical situations, such as 62

instances where CKD is due to drug-induced nephrotoxicity and renal ischemia. 63

4

1. Introduction 64

Chronic kidney disease (CKD) is a major health problem characterized by a gradual loss of 65

kidney function over time. Over 30 million people or 15% of US adults are estimated to have 66

CKD [1]. CKD may progress to end-stage renal disease (ESRD), which is a public health 67

concern and big economic burden. ESRD costs the United States taxpayers approximately 32.8 68

billion dollars in annual Medicare expenditures [2]. Causes of CKD vary widely, however 69

drug-induced nephrotoxicity has been an increasingly recognized complication of many 70

therapeutic agents in the clinical setting. In patients with pre-existing renal dysfunction or renal 71

ischemia, the effects of drug-induced nephrotoxicity can be profound and may accelerate 72

progression to ESRD [2]. At present, hemodialysis and renal transplantation are both effective 73

treatment modalities for ESRD. However, hemodialysis is expensive with many potential 74

complications while renal transplant as a treatment modality is limited by the number organ 75

donors available [3]. 76

Cell therapy has emerged as a promising therapeutic approach for the treatment of CKD 77

with great potential [4]. Several stem cell types from a variety of sources have been used in 78

animal experiments [5-10]. Bone marrow-derived stem cells were found to significantly improve 79

renal function in mice models [5], and their safety was subsequently studied in a phase 1 clinical 80

trial [11]. In addition, stem cells derived from amniotic fluid [6, 7], endothelium [8], adipose 81

tissue [9] and primary renal cells [10] have demonstrated therapeutic effect in renal regeneration. 82

However, obtaining these stem cells requires invasive procedures, which may cause iatrogenic 83

injury to adjacent organs and increases infection risk among other complications. Thus, the 84

desire to obtain stem cells from a simple, non-invasive and inexpensive route is of utmost 85

necessity. 86

Recently, we demonstrated that stem cells exist in human urine and possess the potential for 87

clinical applications [12-20]. Urine derived stem cells (USC) are thought to originate from the 88

parietal cells in renal glomeruli [14] and have the potential for tissue regenerative effects, 89

5

including robust proliferative potential, multi-potential differentiation, and the ability to exert 90

regenerative effects via paracrine factors [14, 21, 22]. USC secrete multiple paracrine factors to 91

recruit native cells to participate in tissue regeneration and induce immune-modulatory changes 92

in vivo. Studies from our group and other laboratories have demonstrated that USC or the media 93

derived from USC significantly improved renal function by reducing inflammation and fibrin matrix 94

deposition within the kidney in different rodent CKD models, including age-related kidney 95

disease [23] and streptozotocin-induced diabetic nephropathy [24, 25]. Furthermore, USC 96

enhanced renal function in an acute renal ischemia model [26]. Nephrotoxic drug induced renal 97

damage superimposed on pre-existing renal damage is one of the more common causes of 98

iatrogenic CKD [2]. However, it is challenge to revise the renal function and prevent histological 99

structure damages from the chronic renal damages in clinical settings. Compared to stem cells 100

derived from other sources such as amniotic fluid, placenta, bone marrow, adipose tissues, USC 101

have unique advantages in the treatment of CKD. USC can be obtained from simple, safe, 102

non-invasive and low-cost approaches. A large of amount of stem cells can be generated 103

through a few weeks. In addition, as the kidney tissue specific stem cells, USC might be optimal 104

for the kidney tissue repair. As such, the goal of this study was to determine therapeutic impact of 105

USC on renal parenchyma architecture and renal function in a rat CKD model, induced by dual 106

nephrotoxic drug-renal ischemia injury. 107

6

2. Methods 108

109

2.1 USC isolation and identification 110

Collection of human urine samples for USC isolation in this study was approved by the 111

Wake Forest University Health Sciences Institutional Review Board. USC culture and 112

characterization have been described previously [12-20]. In brief, USC were collected from 113

healthy adult males (n=6) and expanded in culture media comprised of keratinocyte serum-free 114

medium (Gibco, Gaithersburg, MD) and Dulbecco's Modified Eagle's medium (DMEM, high 115

glucose, Gibco, Gaithersburg, MD) with 5% fetal bovine serum (FBS) (Gibco, Gaithersburg, MD) 116

containing supplements [12], see Table 1. Expanded USC were trypsinized and incubated with 117

anti-human antibody labeled with cell surface markers of renal parietal epithelial cells (CD24, 118

CD29 and CD44) (BD, Franklin Lakes, NJ) [27], mesenchymal stem cells (MSC) (CD73, CD90, 119

CD105, and CD146) (BD, Franklin Lakes, NJ), embryonic stem cells (SSEA4), hematopoietic 120

stem cell (CD31, CD34, CD45) (BD, Franklin Lakes, NJ) and STRO-1 (BioLegend, San Diego, 121

CA). Fluorescein isothiocyanate (FITC) or phycoerythrin (PE) conjugated isotype antibodies 122

were used to determine background fluorescence. All cells were analyzed using BD FACS 123

Caliber analytical fluorescence activated cell sorter (BD, Franklin Lakes, NJ). 124

125

2.2 Osteogenic and adipogenic differentiation of USC 126

Adipogenic induced differentiation 127

USC were seeded at a density of 21,000 cells/cm2 and cultured in serum containing DMEM 128

low-glucose medium with 1 µM dexamethasone, 500 µM 3-isobutyl-1-methylxanthine, 10 µg/ml 129

insulin, and 100µM indomethacin for 28 days with medium changes every third day. 130

Differentiated cells were then fixed with 4% paraformaldehyde for 30 min at room temperature 131

and stained with fresh 0.3% Oil Red O solution for 50 min. An inverted microscope was used to 132

identify red stained areas, indicating fat droplets. Adipose derived stem cells were used as 133

7

positive controls. 134

135

Osteogenic induced differentiation 136

USC were seeded at a density of 4,000 cells/cm2 and cultured in serum containing DMEM 137

low-glucose medium with 100 nM dexamethasone, 10 mM β-glycerophosphate and 50 µM 138

ascorbic acid-2-phosphate (Wako Chemicals, Richmond, VA) for 28 days with medium changes 139

every third day. Differentiated osteocytes were then fixed by ice-cold 95% ethanol for 5 minutes 140

at 4 ℃ and stained for calcium deposits with 2% Alizarin Red Solution (pH 4.0). An inverted 141

microscope was used to identify orange-red stained areas, indicating calcium deposits. 142

143

2.3 An athymic rat model of dual injury-induced CKD 144

Animal experiments were approved by Wake Forest Institutional Animal Care and Use 145

Committee (Protocol no A11-085). Animals were housed in a temperature-controlled 146

environment with free access to food and water. A 12-hour light and 12-hour dark cycle was 147

provided. 148

To avoid the potential risk of immunoreaction caused by the implanted xenogenic cells, 149

athymic rats (RNU316) were used for this study. These athymic nude rats were T-cell deficient 150

and demonstrated depleted cell populations in thymus-dependent areas of peripheral lymphoid 151

organs. In total, 58 male athymic rats (age 9 weeks, Charles River, NC) were used in this study. 152

To optimize the doses of gentamicin and renal ischemia time frame, 30 animals were used for 153

creation model of dual injury-induced CKD before USC or PBS injection. Among these animals, 154

22 animals died and 8 animals recovered renal function (failure of model creation) during the 155

model creation processes. 156

157

Once the model was successfully established, 28 animals were divided into three groups: a 158

PBS-treated group (n=9), an USC-treated group (n=9), and an age-matched control group 159

8

(n=10). A CKD model was induced by creating an ischemia-reperfusion injury and gentamicin 160

injection as demonstrated in Figure 1 [28]. Briefly, the rats were anesthetized and laid in the 161

supine position with continuous 4% isoflurane via nose cone. A midline incision was made and 162

intestines were retracted to expose both kidneys. Using vascular clamps, the renal pedicles were 163

clamped bilaterally, for 60 minutes and then released to induce ischemia-reperfusion injury. The 164

clamps were then removed and the incision was closed. Two weeks after surgically-created 165

renal ischemia injury, gentamicin (100 mg/kg/day, Phoenix Pharmaceutical Saint Joseph, MO) 166

was injected subcutaneously for 5 consecutive days. Serum creatinine levels were used to 167

evaluate renal function. Blood and urine samples were collected every 2 weeks and analyzed 168

with blood chemistry machine (Beckman Coulter Inc., Brea, CA). Glomerulus filtration rate (GFR) 169

was calculated as follows: GFR=urine creatinine x urine volume/serum creatinine/time. 170

171

2.4 Injections of USC into rat renal parenchyma 172

Eight weeks after the dual injury model was created, a USC cell suspension was injected 173

into the kidneys for the USC-treated group. The abdominal contents were exposed once again 174

and both kidneys were identified. A cell suspension of USC was administered into the 175

parenchyma of both kidneys. Each kidney received a total of 5x106/ml in 100 µl of PBS, 8 weeks 176

after CKD creation. To distribute injected cells evenly, 50% of the aforementioned cell 177

suspension was injected into the upper pole and lower pole of each kidney. PBS (100 µl) only 178

was injected in the same fashion as above for the PBS-control group. The model was not created 179

in aged-matched control animals and no injections were performed. Blood and urine samples 180

were collected to test renal function parameters i.e. serum creatinine and GFR, at 2 week 181

intervals. Kidneys were harvested from each animal for histological evaluation, 12 weeks after 182

injection. 183

184

2.5 Histological and Immuno-histochemical analysis 185

9

Harvested kidney tissues were fixed in 10% paraformaldehyde solution and processed for 186

paraffin embedding. Five µm sections were obtained. Slides were stained with hematoxylin and 187

eosin (H&E), and numbers of glomeruli were counted under high power fields (200X) microscope 188

(Leica, Germany). Normal glomeruli are uniform in size, visceral (inner) structures and Bowman's 189

space. Amounts of collagen deposited were assayed by computerized ImageJ (ImageJ software 190

package, NIH) based on Masson’s Trichrome staining. Immunohistochemistry was performed 191

with anti-human leukocyte antigen (HLA) and anti CD68 antibodies as a marker of macrophage 192

(BD Biosciences, CA). All slides were blocked in serum-free protein blocker (DAKO, Carpinteria, 193

CA) for 20 minutes. 194

195

Transforming growth factor beta 1 (TGF-ß1, DAKO) was tested as dysregulation of TGF-β 196

activation and signaling may result in apoptosis of resident functional cells but proliferation of 197

fibroblast. In addition, superoxide dismutase 1 (SOD-1, DAKO) as an antioxidant enzyme 198

protecting the cell from reactive oxygen species toxicity were tested in USC-treated group, 199

compared to PBS-treated control. Anti-superoxide dismutase 1 antibody (EPI722Y) was used at 200

a 1:400 dilution. After incubation with the primary antibodies, tissue sections were incubated with 201

biotin-conjugated secondary antibodies (1:300 dilution) (Vector Laboratories, Burlingame, CA, 202

USA) at room temperature for 60 min, followed by incubation with HRP-conjugated streptavidin 203

(Vector Laboratories) for 30 min at room temperature and further developed with a DAB 204

substrate kit (Vector Laboratories). 205

206

Positive cells were counted under lower and high power fields (100X, 400X). Total number 207

of glomeruli with abnormal structures, increased glomerular size, pathological changes of renal 208

tube, intestinal fibrosis, collagen deposits and inflammatory were quantified. The degree of 209

histological abnormalities on nephrons, such as dilation or collapse, epithelium and obliterated 210

lumen necrosis in renal tubule and glomerular was evaluated. Quantitative comparison of 211

10

histology and immunohistochemical staining was measured by digital image analysis. 212

213

2.6 Statistical analysis 214

All data were analyzed using software Statistical Package for the Social Sciences (SPSS) 215

version 16.0 for windows (SPSS, Inc., Chicago, IL). Normality of the analyzed data was 216

corroborated with Shapiro-Wilk test. When parametric analysis was possible, data were 217

expressed as mean ± SD and one-way analysis of variance (ANOVA) was used for comparison 218

among the three groups. Least Significant Difference test was applied to further compare 219

differences among three groups. P<0.05 was considered statistically significant. 220

11

Results 221

Individual USC appeared consistently as rice grain-like at primary culture (Supplementary 222

Figure 1). USC at passage 3-5 were strong positive for glomerular parietal epithelial cells (CD24, 223

CD29 and CD44), and also classical cell surface markers for MSC (CD73, CD90, CD105, CD146 224

and SSEA-4), but lack of the expression of hematopoietic markers (CD 31, CD34 and CD45). 225

These results were consistent with our previous reports (Supplementary Figure 1) [12-20]. In 226

addition, USC differentiated into osteogenic and adipogenic cell lineages, respectively (Figure 227

2). 228

For evaluation of renal function, serum creatinine (Cr) was evaluated and GFR was 229

calculated every 2 weeks. Creatinine rise and GFR depression were used as indicators of 230

effective CKD model creation. Creatinine levels increased significantly in all animals assigned to 231

USC and PBS groups after one week, post-model creation and remained elevated compared to 232

AMC (p<0.01). Likewise, GFR declined significantly for animals assigned to USC and PBS 233

groups after one week, post-model creation (p<0.01). The changes in creatinine and GFR were 234

similar in both USC and PBS groups until injection was performed at 13 weeks post-model 235

creation. In the USC-treated group, serum creatinine decreased and GFR increased significantly 236

improved compared to PBS-treated animals two weeks after cell implantation (p=0.027 and 237

p=0.037, respectively). Th difference in GFR and Cr remained significantly different between 238

these groups until the study endpoint 12 weeks after cell implantation. GFR in the USC-treated 239

group improved to 50% of that in AMC whereas PBS-treated animals only exhibited a 25% 240

recovery of GFR compared to AMC (Figure 3). 241

242

Grossly, the surface of the kidneys in PBS-treated group was coarse and pale, while 243

kidneys in USC-treated group appeared similar to AMC (Figure 4). In addition, the kidney weight 244

in PBS-treated group (1.48±0.31g) was significantly lower compared to those of the AMC 245

(1.98±0.19 g) and USC-treated groups (1.69±0.22 g) 12 weeks after injection therapy. The 246

12

kidney weight in USC-treated group significantly increased compared to that in PBS-treated 247

group (p<0.01) despite not returning to kidney weights exhibited in the AMC group. 248

249

Immunocytochemical analysis using HLA staining revealed that the implanted human USC 250

were present around Bowman’s capsule or scattered within renal tubules for the entire study 251

period, however numbers of the grafted cell decreased significantly, 12 weeks after injection 252

compared to immediately post-injection (Figure 5). To determine the reno-protective effects of 253

USC, we evaluated the changes in glomerular, renal tubular, and tubulo-interstitial structure 254

among the three groups. In the PBS-treated group, about 38% of glomeruli displayed normal 255

architecture (4.0±2.0 normal glomeruli/high power field [HPF]) within the renal cortex while 62% 256

(6.6±2.0) demonstrated glomerulosclerosis evidenced by wrinkling and collapse of the basement 257

membrane and constricted glomerular capillaries, compared to AMC. This pathological changes 258

in nephrons were significantly lower than the USC-treated group (6.3±2.1) and the AMC group 259

(10.6±2.2) (p<0.01). Conversely, the USC-treated group demonstrated 60% normal glomeruli 260

with 40% displaying glomerular sclerosis and dilated, atrophic tubules (Figure 6) compared to 261

the AMC group (p<0.01). 262

263

Collagen deposition in the renal parenchyma was investigated using Masson’s Trichrome 264

staining (Figure 7). Collagen deposition was elevated in PBS-treated animals (28.4±6.3). This 265

was significantly higher than in USC-treated group (8.3±3.0) and the AMC group (1.3±0.3) 266

(p<0.01), which represented over a 20-fold increase. There was also significantly higher collagen 267

deposition in the USC-treated group compared to the AMC group which was only a 6-fold 268

increase. To assess monocyte infiltration (Figure 8), numbers of cells expressing CD68+ was 269

significantly higher in the PBS group (8.7±2.54) compared to both USC-treated group (4.7±2.12) 270

and AMC group (2.0±1.05) (p<0.01). The USC-treated group exhibited an elevated numbers of 271

cells expressing CD68+compared to the AMC group as well, however albeit not as drastic. 272

13

To evaluate tissue fibrosis within the renal tissues, TGF-ß1 as an inflammatory marker 273

(Figure 9) was used. TGF-ß1 expression was significantly high in the PBS-treated groups in 274

both the cortex and medulla (1.2±0.1 and 1.3±0.08) which represented a 58% and 69% increase 275

compared to AMC (0.8±0.05 and 0.7±0.06). Expression of TGF- ß1 in the cortex and medulla of 276

the USC-treated group (0.95±0.09 and 0.99±0.06), respectively, represented as 23% and 29% 277

increase compared to AMC. The relative increase in TGF-ß1 for PBS-treated animals was 278

significantly higher than for the USC-treated group (p<0.01). 279

280

Superoxide dismutase (SOD-1) as important mediator in the oxidative stress response was 281

assessed (Figure 9). SOD-1 expression in the PBS-treated group in the cortex and medulla was 282

1.49±0.17 and 1.53±0.14, which represented 84% and 89% of that in the AMC group (1.8±0.16 283

and 1.7±0.11). Expression of SOD-1 in the USC-treated group was 2.35±0.1 and 2.43±0.11 in 284

the cortex and medulla, respectively. This represented a 33% and 42% increase in SOD-1 285

expression compared to AMC. The relative changes in SOD-1 expression were significantly 286

different among the three study groups (p<0.01). 287

288

Taken together, there is a tendency between the creatinine clearance and both the 289

inflammatory and antioxidant parameter amount three groups, i.e. renal function improved with 290

less inflammatory and increase level of antioxidate in USC treated group than those in 291

PBS-treated group, competed to AMC (Figure 10) . 292

14

Discussion 293

Chronic kidney disease is caused by progressive injury to nephrons including glomeruli and 294

renal tubules, interstitium and vasculature. Etiologies include genetic predisposition, metabolic 295

derangements, renal ischemia, environmental and infectious causes, autoimmune conditions, 296

and drug-related nephrotoxicity. Repeated or superimposed insults may ultimately lead to 297

chronic kidney disease, which ultimately may progress to end-stage renal disease without 298

appropriate therapy. Of particular interest to us was drug-induced nephrotoxicity. 299

300

Drug-induced nephrotoxicity occurs more frequently in patients with pre-existing renal 301

dysfunction or ischemia due to acute kidney injury, chronic kidney disorder, diabetic nephropathy, 302

congestive heart failure, among other causes. Gentamicin is notorious for its nephrotoxic 303

side-effects, which is characterized by proximal convoluted tubular necrosis and glomerular 304

congestion, resulting in decreased GFR and renal dysfunction [29]. Similarly, renal ischemia 305

often causes acute tubular necrosis [30]. Dual insults with renal ischemia and drug nephrotoxicity 306

may lead to severe CKD and ESRD without prompt treatment and thus it is necessary to develop 307

an effective approach to prevent progression of CKD to ESRD. 308

309

Several animal models of CKD have been reported in the literature [31], however, due to the 310

robust regenerative capacity of rodents, it is challenging to study long-term efficacy of cellular 311

therapy. The subtotal nephrectomy model [32-34] requires a complex procedure and does not 312

reflect the mechanism of renal failure most often observed in humans. The streptozotocin model 313

mimics tissue injury observed in diabetic nephropathy. However, as CKD caused by 314

drug-induced nephrotoxicity increases, the need for another animal model is required. Herein, 315

we have established a model using ischemia-reperfusion injury combined with 316

gentamicin-induced nephrotoxicity to provide an easy and reliable animal model to study CKD 317

[28]. 318

15

Given that definitive treatment i.e. renal transplant is not commonplace; it is of utmost 319

importance to prevent progression of CKD to ESRD. The nephron is the functional unit of the 320

kidney and thus we quantified the number of normal nephrons in each treatment group. In the 321

USC treated group, 70% of glomeruli displayed normal architecture while this number decreased 322

to 40% in the PBS-treated group. In contrast, numbers of abnormal nephrons significantly lower 323

in USC-treated group than those in PBS group. Likewise there was a significant decrease in 324

serum Cr and increase in GFR in the USC group compared to PBS treated animals. It is likely 325

that the preservation of glomerular architecture, evidenced by a lower level of glomerulosclerosis 326

and dilatated, atrophic tubules, in the USC treated group compared to the PBS group is 327

responsible for the chemical (serum Cr) and functional (GFR) findings. 328

329

Tissue fibrosis and collagen deposition are important contributors to progressive renal 330

failure. This is a complex process involving fibroblasts, immune reactions and free radical 331

damage to the renal parenchyma. We evaluated these processes through gross examination of 332

the kidneys and through the chemical investigation of mediators of fibrosis and free radical 333

damage. Monocyte infiltration as cells expressing CD68+ was elevated in the PBS group to a 334

greater extent than in USC treated group, compared to AMC. TGF- ß1 as a surrogate for tissue 335

fibrosis followed a similar trend. Superoxide dismutase is an important mediator in inhibiting free 336

radical induced tissue damage and its expression was lowest in the PBS treated group and 337

highest in the USC treated group. Given the lower levels of TGF- ß1 and CD68+ cells in the USC 338

group compared to the PBS group, it is unsurprising the see a lower level of collagen deposition 339

and improved renal function in the latter (Figure 9). Furthermore, collagen and fibrosis lead to 340

glomerular and tubular atrophy, explaining the gross appearance and the weights of the kidneys 341

in the PBS group compared to the USC and AMC groups. 342

343

Our study is not without limitations which may need to be addressed prior to beginning 344

16

safety trials. First, we only administered a single dose of a given concentration of USC. It is 345

feasible that multiple injections may further improve the renoprotection imbued by USC. These 346

injections may need to be distributed over several weeks to months for maximal effect. Second, it 347

is possible that increasing the number of injected cells will improve kidney function even further. 348

Third, very few studies have been conducted investigating routes of administration for urologic 349

stem cell therapy and thus it may be that intravenous therapy may provide improved benefit over 350

intrarenal injection. Finally, molecular analysis of pro-regenerative, anti-apoptotic or anti-fibrotic 351

pathways needs further investigation. 352

353

Herein, we have demonstrated that local implantation of human USC significantly improved 354

renal function, imbued a protective effect on nephrons, reduced renal scaring, and mollified the 355

inflammatory response in a renal ischemia-reperfusion, gentamicin, dual injury-induced CKD 356

rodent model (Figure 10). Through our chemical and functional analysis, it is clear that USC 357

have some impact on the aforementioned parameters insofar as USC treatment ameliorates 358

renal degradation seen in the group that received no treatment. 359

17

360

361

Table 1. Antibodies, mediums, and serums used in this study.

Items

Antibody

Type

Source

Catalog No.

A list of

Antibodies

CD24 FITC-conjugate BD 560992

CD29 PE-conjugated BD 555443

CD31 FITC-conjugated BD 560984

CD34 APC-conjugated BD 560940

CD44 PE-conjugated BD 550989

CD45 APC-conjugated BD 340943

CD73 PE-conjugated BD 550257

CD90 APC-conjugated BD 561971

CD105 PE-conjugated BD 560839

CD146 PE-conjugated BD 550315

SSEA4 PE-conjugated BD 560128

STRO-1 FITC-conjugated Biolegend 340105

HLA Rabbit monoclonal Abcam 52922

CD68 Mouse monoclonal AbD Serotec MCA341R

TGF-ßI Rabbit monoclonal Abcam 170874

SOD-1 Rabbit monoclonal Abcam EP1727Y

Medium and Serum

Source

Catalog No.

Culture

Mediums

KSFM Gibco 17005042

DMEM Gibco 41966-052

Serums

Fetal bovine serum

Gibco

26140079

18

362

363

Table 2. Pathohistological changes in kidney of USC-treated rats, compared to those in controls 12

weeks after cell therapy.

Pathological

changes

PBS-treated

Group*

USC-treated

Group*

Age-matched

Control*

No. of rats n=9 n=9 n=10

Ratio of normal nephrons

(%) (Fig. 6)

38% (4.0±2.0)

60% (6.33±2.1)

100% (10.6±2.2)

Ratio of abnormal

nephrons (%) (Fig. 6)

62% (6.6±2.0)

40% (4.27±2.1)

0% (0.0±0.0)

Ratio of collagen

deposition (%) ( Fig. 7)

322% (16.2±4.7)

152% (7.6±3.3)

100% (5.0±2.0)

Ratio of inflammatory

(%) (Fig. 8)

435% (8.7±2.5)

235% (4.7±2.1)

100% (2.0±1.1)

Increased interstitial

fibrosis (cortex vs.

medulla) (%) (Fig. 9)

158% (1.22±0.1)

vs.

182% (1.32±0.08)

123% (0.95±0.09)

vs.

128%(0.99±0.06)

100% (0.77±0.05)

vs.

100%(0.72±0.06)

Abbreviations:

PBS-Phosphate buffered saline; USC-Urine-derived stem cells.

Notes: Glomerular sclerosis, dilated or atrophic renal tubule structure; Inflammatory response -

measured by monocyte infiltration. One-way ANOVA statistical analysis was formed among three

groups, based on the average relative intensity of ten fields per sample, each taken at a 200x

magnification for immunostaining. Percentages were relative to AMC. All values are reported as mean

± SEM.

*All differences listed in this table were statistically significant between the PBS, USC-treated groups,

and AMC, p<0.05 or p<0.01.

19

Figures and legends: 364

365

366

367

368

369

370

371

372

373

374

375

376

377

378

379

380

381

382

383

Figure 1. A brief flow chart of the study design. The CKD model was created using dual-injury 384

of renal ischemia and nephrotoxicity using 5 consecutive days of gentamycin injection, 2 weeks 385

after renal ischemia-reperfusion injury. Eight weeks after confirmation of model creation using 386

serial serum Cr and GFR, animals received either USC or PBS via direct injection into renal 387

parenchyma. Renal function was performed at different time points. Finally, histological and 388

immunocytochemical analysis were performed 12 weeks after injection. Age matched animals 389

were applied as controls (AMC). 390

391

Gentamycin

injection daily

for 5 days

USC or PBS

Groups

AMC

Group

Renal ischemia

/reperfusion

USC or PBS

injection

Histology

analysis

2 wks 8 wks

12 wks

3rd

wk 11th

wk 23th

wk 0

20

392

A 393

100 µm

21

394

395 396

B 397

398

399

Supplement Figure 1. (A) Human USC all displayed ''rice grain''-like morphology in the early 400

passage (p2). (B) USC expressed markers of glomerular parietal epithelial cells (CD24, CD29 401

and CD44), and also classical cell surface markers of MSC (CD73, CD90, CD105, CD146 and 402

SSEA-4) and negative expression of hematopoietic markers (CD 31, CD34 and CD45). 403

22

404

405

406

407

408

409

410

411

412

413

414

415

Figure 2. In vitro differentiation potential of USC. (A) Adiopogenic differentiation of USC was 416

demonstrated with oil red O (lipid droplet). (B) Osteogenic differentiation of USC was 417

demonstrated with Alizarin Red. (C) Non-induced USC acted as a control group. 418

23

419

420

421

422

423

Figure 3. Dynamic changes of renal function at different time points. A CKD model was 424

successfully established in athymic rats. (A) Serum creatinine levels were significantly lower in 425

USC-treated rats, compared to PBS-treated group on weeks 4 and 6 after cell therapy, 426

respectively. (B) Glomerular filtration rate was significantly higher in USC-treated group at each 427

time period after cell therapy, compared to the PBS-treated group (* p<0.01). 428

A

B

S

eru

m C

reatin

ine (

mg

/dl)

G

FR

(m

l/m

in)

Final evaluation 12 weeks after cell therapy

USC Injection

Gentamycin

Clamp bilateral renal vessels

Final evaluation 12 weeks after cell therapy

USC Injection

Gentamycin

Clamp bilateral renal vessels

A

* *

* * * *

24

429

430

431

Figure 4. The gross appearance and weight changes of the kidneys. (A) The gross 432

appearance of kidneys in the three treatment groups was noted. (B) The weight of the kidneys in 433

USC-treated rats was significantly elevated compared to the PBS-treated animals but also 434

significantly lighter than that in AMC group, * Significance at the p<0.05 level. 435

436

B

We

igh

t o

f k

idn

ey *

*

25

437

438

439

440

441

442

443

444

445

446

447

448

449

450

451

Figure 5. Tracking implanted USC within renal tissues by immuncytochemistry analysis 452

with anti-HLA A antibody. More grafted USC (brown) were found in the tubule interstitial 453

regions one week (A) and two weeks (B) after implantation compared to 12 weeks. Number of 454

the implanted USC decrease with time during the 12 week follow-up. A few cells were still 455

identified in the renal tubules and the medulla (C) as well as around Bowman’s capsule in the 456

renal cortex (D). 457

50 µm

50 µm

50 µm

A B

C D

26

458

459

460

461

462

463

464

465

466

467

468

469

470

471

Figure 6. of Potho-histological changes in nephrons within renal cortex and medulla after 472

USC implantation. (A) About 60% of glomeruli increased in size with collapse of some 473

glomerular tufts (short arrow) within the renal cortex and 62% of renal tubules were dilated (long 474

arrow) within the medulla in PBC-treated rats. In contrast, a majority of glomeruli and renal 475

tubules displayed normal structure and only 40% of nephrons displayed abnormal configuration 476

in USC-treated group. (B) Expressed as the average numbers of relative abnormal nephrons in 477

six fields per sample at 200x magnification. Numbers of normal nephron within renal tissue under 478

high power field with H&E staining. * Significance at the p<0.05 level. 479

PBS USC AMC A

Me

du

lla

C

ort

ex

B B

PBS USC AMC

No

. o

f n

orm

al

ne

ph

ron

s/

Lo

w p

ow

er

fie

ld

B

27

480

481

482

483

484

485

486

487

488

489

490

491

492

493

494

495

496

Figure 7. Histological appearance of collagen deposition and fibrotic changes of the 497

kidneys. (A) Among of collagen deposition in Masson Trichrome staining was significantly 498

elevated in PBS-treated rats, and minimal changes in kidney of rats treated with USC-treated 499

group compared to that in AMC. (B) Percentage of collagen deposition within the renal 500

parenchyma. * Significance at the p<0.05 level 501

502

503

A

Me

du

lla

C

ort

ex

B

PBS USC AMC

PBS USC AMC

Pe

rce

nta

ge

of

co

lla

ge

n

de

po

sit

ion

28

504

505 506

Figure 8. CD68 expression as a surrogate for monocyte infiltration within the kidney. (A) 507

Number of infiltrated monocyte expressing positive for CD68 significantly increased in 508

PBS-treated group, compared to AMC. (B) Number of cells expressing CD68 significantly 509

decreased in USC group compared (C) compared to PBS group. (D) Analysis of CD68 510

expression in the three groups with immunocytochemical staining. * Significance at the p<0.05 511

level 512

513

D

Pe

rce

nta

ge

of

mo

no

cyte

infi

ltra

tio

n

29

514

515

516

517

518

519

520

521

522

523

524

525

526

527

528

529

530

531

532

533

534

535

536

537

538

539

540

541

542

543

544

545

546

547

548

549

550

551

552

Me

du

lla

PBS USC AMC

TGF-1 expression in renal tissues

50 m

50 m 50 m

PBS

50 m

USC

50 m

50 m

A

AMC

*

*

Co

rte

x

Rela

tive inte

nsity o

f T

GF

-ß1 e

xpre

ssio

n

30

553

554

555

PBS USC AMC

Co

rte

x

Me

du

lla

50 m 50 m

50 m 50 m 50 m

50 m

A

SOD-1 expression in renal tissues

PBS USC AMC

* *

Rela

tive inte

nsity o

f S

OD

1 e

xpre

ssio

n

31

Figure 9. Effect of USC implantation on anti-fibroblast and redox properties in CKD 556

rodent model. (A) Amount of TGF-β1 expression significantly increased in renal cortex and 557

medullar tissue of PBS treated animals compared to that of AMC group. However, USC 558

implantation significantly decreased levels of TGF-β1 expression, compared PBS treatment. (B) 559

Amount of SOD-1 expression levels significantly increased in renal tissue of USC-treated groups 560

compared to that of PBS group. Semi-quantification is expressed as the average relative 561

intensity of six fields around defect site per sample, each taken at a 200x magnification. 562

Representative images are shown at 200x magnification. *p < 0.01 compared to AMC as a unit 563

set. 564

32

565

566

Figure 10. Relationship between serum creatinine and both the oxidative stress parameter 567

and the inflammatory parameters with regression line analysis. Antioxidant and 568

antiinflammatory activity of USC is behind the renoprotective activity, compared to those of 569

PBS-treated and AMC groups. 570

SOD1

Serum Creatinine

TGF-ß1

No

rma

lize

d v

alu

es

33

Abbreviations 571

AMC, age-matched control 572

CKD, chronic kidney disease 573

ESRD, end-stage renal disease 574

H&E, hematoxylin and eosin 575

HLA, human leukocyte antigen 576

HPF, high power field 577

FITC, Fluorescein isothiocyanate 578

GFR, Glomerulus filtration rate 579

iPSC, induced pluripotent stem cells 580

PBS, phosphate buffered saline 581

PE, phycoerythrin 582

SSEA-4, stage-specific embryonic antigen-4 583

SPSS, Statistical Package for the Social Sciences 584

USC, Urine-derived stem cells 585

34

Declarations 586

Acknowledgement 587

The authors would like to acknowledge all parties that participated in this study. Authors also 588

would like to thank Dr. Marshall Z. Schwartz in Department of Urology at Wake Forest School of 589

Medicine for his editing and helpful discussions 590

35

Funding 591

Not applicable 592

36

Availability of data and materials 593

All data generated and/or analyzed during this study are included in this published article and its 594

supplementary files. 595

37

Competing interests 596

The authors declare that there are no competing interests associated with this manuscript. 597

38

Author contributions 598

CZ, SG, MA, TK, and PT carried out the cell culture, molecular studies, the animal experiments 599

and drafted the manuscript. CZ, SG, and PT carried out immunoassays and performed statistical 600

analyses. AA and YZ participated in the design of the study. YZ, IK, and PT conceived the study, 601

participated in its design, and completed the manuscript. JY, YS, JJ and AA participated in 602

administration and coordination. All authors read and approved the final manuscript. 603

39

Consent for publication for data from individual person 604

Not applicable 605

40

Ethics approval and consent to participate 606

Collection of human urine samples in this study was approved by the Wake Forest University 607

Health Sciences Institutional Review Board and was performed in accordance with the 608

Declaration of Helsinki (2004). Signed, informed consent was collected from patients. 609

41

Ethics approval for animal study 610

Rodent experiments were approved by Wake Forest Institutional Animal Care and Use 611

Committee (IACUC). 612

42

Reference 613

[1] Centers for Disease Control and Prevention. National Chronic Kidney Disease Fact Sheet, 614

2017. 615

[2] L. Awdishu, R.L. Mehta, The 6R's of drug induced nephrotoxicity, BMC Nephrol 18(1) 616

(2017) 124. 617

[3] 618

https://www.statista.com/statistics/398442/total-number-of-us-organ-donors-by-age-gro619

up/, (2017). 620

[4] L.J. Hickson, A. Eirin, L.O. Lerman, Challenges and opportunities for stem cell therapy in 621

patients with chronic kidney disease, Kidney Int 89(4) (2016) 767-78. 622

[5] E.I. Prodromidi, R. Poulsom, R. Jeffery, C.A. Roufosse, P.J. Pollard, C.D. Pusey, H.T. Cook, 623

Bone marrow-derived cells contribute to podocyte regeneration and amelioration of renal 624

disease in a mouse model of Alport syndrome, Stem Cells 24(11) (2006) 2448-55. 625

[6] I.L. Noronha, R.C. Cavaglieri, F.L. Janz, S.A. Duarte, M.A. Lopes, M. Zugaib, S.P. Bydlowski, 626

The potential use of stem cells derived from human amniotic fluid in renal diseases, Kidney 627

Int Suppl (2011) 1(3) (2011) 77-82. 628

[7] C. Rota, B. Imberti, M. Pozzobon, M. Piccoli, P. De Coppi, A. Atala, E. Gagliardini, C. Xinaris, 629

V. Benedetti, A.S. Fabricio, E. Squarcina, M. Abbate, A. Benigni, G. Remuzzi, M. Morigi, Human 630

amniotic fluid stem cell preconditioning improves their regenerative potential, Stem Cells 631

Dev 21(11) (2012) 1911-23. 632

[8] O. Sangidorj, S.H. Yang, H.R. Jang, J.P. Lee, R.H. Cha, S.M. Kim, C.S. Lim, Y.S. Kim, Bone 633

marrow-derived endothelial progenitor cells confer renal protection in a murine chronic 634

renal failure model, Am J Physiol Renal Physiol 299(2) (2010) F325-35. 635

[9] S. Yokote, Y. Katsuoka, A. Yamada, I. Ohkido, T. Yokoo, Effect of adipose-derived 636

mesenchymal stem cell transplantation on vascular calcification in rats with 637

adenine-induced kidney disease, Sci Rep 7(1) (2017) 14036. 638

[10] L.M. Yamaleyeva, N.K. Guimaraes-Souza, L.S. Krane, S. Agcaoili, K. Gyabaah, A. Atala, T. 639

Aboushwareb, J.J. Yoo, Cell therapy with human renal cell cultures containing 640

erythropoietin-positive cells improves chronic kidney injury, Stem Cells Transl Med 1(5) 641

(2012) 373-83. 642

[11] A. Makhlough, S. Shekarchian, R. Moghadasali, B. Einollahi, M. Dastgheib, G. Janbabaee, 643

S.E. Hosseini, N. Falah, F. Abbasi, H. Baharvand, N. Aghdami, Bone marrow-mesenchymal 644

stromal cell infusion in patients with chronic kidney disease: A safety study with 18 months 645

of follow-up, Cytotherapy 20(5) (2018) 660-669. 646

[12] Y. Zhang, E. McNeill, H. Tian, S. Soker, K.E. Andersson, J.J. Yoo, A. Atala, Urine derived 647

cells are a potential source for urological tissue reconstruction, J Urol 180(5) (2008) 648

2226-33. 649

[13] S. Bharadwaj, G. Liu, Y. Shi, C. Markert, K.E. Andersson, A. Atala, Y. Zhang, 650

Characterization of urine-derived stem cells obtained from upper urinary tract for use in 651

43

cell-based urological tissue engineering, Tissue Eng Part A 17(15-16) (2011) 2123-32. 652

[14] S. Bharadwaj, G. Liu, Y. Shi, R. Wu, B. Yang, T. He, Y. Fan, X. Lu, X. Zhou, H. Liu, A. Atala, J. 653

Rohozinski, Y. Zhang, Multipotential differentiation of human urine-derived stem cells: 654

potential for therapeutic applications in urology, Stem Cells 31(9) (2013) 1840-56. 655

[15] R. Lang, G. Liu, Y. Shi, S. Bharadwaj, X. Leng, X. Zhou, H. Liu, A. Atala, Y. Zhang, 656

Self-renewal and differentiation capacity of urine-derived stem cells after urine 657

preservation for 24 hours, PLoS One 8(1) (2013) e53980. 658

[16] G. Liu, R.A. Pareta, R. Wu, Y. Shi, X. Zhou, H. Liu, C. Deng, X. Sun, A. Atala, E.C. Opara, Y. 659

Zhang, Skeletal myogenic differentiation of urine-derived stem cells and angiogenesis using 660

microbeads loaded with growth factors, Biomaterials 34(4) (2013) 1311-26. 661

[17] P.R. Liu G, Wu R, Shi Y, Deng C, Zhou X, Atala A, Opara E, Zhang Y Skeletal Myogenic 662

Differentiation of Urine-Derived Stem Cells, Angiogenesis and Innervation Using Hydrogel 663

Loaded with Growth Factors for Potential in Treatment of Urinary Incontinence, J Urol 664

193(4) (2015) e74. 665

[18] Y. Liu, W. Ma, B. Liu, Y. Wang, J. Chu, G. Xiong, L. Shen, C. Long, T. Lin, D. He, D. Butnaru, L. 666

Alexey, Y. Zhang, D. Zhang, G. Wei, Urethral reconstruction with autologous urine-derived 667

stem cells seeded in three-dimensional porous small intestinal submucosa in a rabbit 668

model, Stem Cell Res Ther 8(1) (2017) 63. 669

[19] R.P. Wu, G. Liu, Y.A. Shi, S. Bharadwaj, A. Atala, Y. Zhang, Human urine-derived stem 670

cells originate from parietal stem cells, J Urol 191(4) (2014) e1-e958. 671

[20] S. Wu, Y. Liu, S. Bharadwaj, A. Atala, Y. Zhang, Human urine-derived stem cells seeded in 672

a modified 3D porous small intestinal submucosa scaffold for urethral tissue engineering, 673

Biomaterials 32(5) (2011) 1317-26. 674

[21] D. Zhang, G. Wei, P. Li, X. Zhou, Y. Zhang, Urine-derived stem cells: A novel and versatile 675

progenitor source for cell-based therapy and regenerative medicine, Genes Dis. 1(1) (2014) 676

8-17. 677

[22] D. Qin, T. Long, J. Deng, Y. Zhang, Urine-derived stem cells for potential use in bladder 678

repair, Stem Cell Res. Ther. 5(3) (2014) 69. 679

[23] T.J. Long, Y.D.; Shi, H.; Zhong, L.R.; Zhang, D.Y.; Li, W.; Zhang, Y.; Chen, D.; Jiao, Y.; Diz, 680

D.; Zhang, Y., Human urine-derived stem cells lessen inflammation and fibrosis within 681

kidney tissue in a rodent model of aging-related renal insufficiency, J Urol 197( 4) (2017) 682

e293. 683

[24] X. Dong, T. Zhang, Q. Liu, J. Zhu, J. Zhao, J. Li, B. Sun, G. Ding, X. Hu, Z. Yang, Y. Zhang, L. Li, 684

Beneficial effects of urine-derived stem cells on fibrosis and apoptosis of myocardial, 685

glomerular and bladder cells, Mol Cell Endocrinol 427 (2016) 21-32. 686

[25] Z.Z. Jiang, Y.M. Liu, X. Niu, J.Y. Yin, B. Hu, S.C. Guo, Y. Fan, Y. Wang, N.S. Wang, Exosomes 687

secreted by human urine-derived stem cells could prevent kidney complications from type I 688

diabetes in rats, Stem Cell Res Ther 7 (2016) 24. 689

[26] S.F. Tian, Z.Z. Jiang, Y.M. Liu, X. Niu, B. Hu, S.C. Guo, N.S. Wang, Y. Wang, Human 690

44

urine-derived stem cells contribute to the repair of ischemic acute kidney injury in rats, Mol 691

Med Rep 16(4) (2017) 5541-5548. 692

[27] C. Sagrinati, G.S. Netti, B. Mazzinghi, E. Lazzeri, F. Liotta, F. Frosali, E. Ronconi, C. Meini, 693

M. Gacci, R. Squecco, M. Carini, L. Gesualdo, F. Francini, E. Maggi, F. Annunziato, L. Lasagni, M. 694

Serio, S. Romagnani, P. Romagnani, Isolation and characterization of multipotent progenitor 695

cells from the Bowman's capsule of adult human kidneys, J Am Soc Nephrol 17(9) (2006) 696

2443-56. 697

[28] H.J. Wang, A. Varner, T. AbouShwareb, A. Atala, J.J. Yoo, Ischemia/reperfusion-induced 698

renal failure in rats as a model for evaluating cell therapies, Ren Fail 34(10) (2012) 699

1324-32. 700

[29] E. Ozbek, M. Cekmen, Y.O. Ilbey, A. Simsek, E.C. Polat, A. Somay, Atorvastatin prevents 701

gentamicin-induced renal damage in rats through the inhibition of p38-MAPK and 702

NF-kappaB pathways, Ren Fail 31(5) (2009) 382-92. 703

[30] D.P. Basile, M.D. Anderson, T.A. Sutton, Pathophysiology of acute kidney injury, Compr 704

Physiol 2(2) (2012) 1303-53. 705

[31] H.C. Yang, Y. Zuo, A.B. Fogo, Models of chronic kidney disease, Drug Discov Today Dis 706

Models 7(1-2) (2010) 13-19. 707

[32] R. Kelley, E.S. Werdin, A.T. Bruce, S. Choudhury, S.M. Wallace, R.M. Ilagan, B.R. Cox, P. 708

Tatsumi-Ficht, E.A. Rivera, T. Spencer, H.S. Rapoport, B.J. Wagner, K. Guthrie, M.J. Jayo, T.A. 709

Bertram, S.C. Presnell, Tubular cell-enriched subpopulation of primary renal cells improves 710

survival and augments kidney function in rodent model of chronic kidney disease, Am J 711

Physiol Renal Physiol 299(5) (2010) F1026-39. 712

[33] H.C. Caldas, I.M. Fernandes, R.S. Kawasaki-Oyama, M.A. Baptista, A.M. Plepis, V.A. 713

Martins, T.M. Coimbra, E.M. Goloni-Bertollo, D.M. Braile, M. Abbud-Filho, Effect of stem cells 714

seeded onto biomaterial on the progression of experimental chronic kidney disease, Exp 715

Biol Med (Maywood) 236(6) (2011) 746-54. 716

[34] G.J. Becker, T.D. Hewitson, Animal models of chronic kidney disease: useful but not 717

perfect, Nephrol Dial Transplant 28(10) (2013) 2432-8. 718