Developing better mouse models to study cisplatin-induced kidney injury

Abstract

Cisplatin is a potent chemotherapeutic used for the treatment of many types of cancer. However, its dose-limiting side effect is nephrotoxicity leading to acute kidney injury (AKI). Patients who develop AKI have an increased risk of mortality and are more likely to develop chronic kidney disease (CKD). Unfortunately, there are no therapeutic interventions for the treatment of AKI. It has been suggested that the lack of therapies is due in part to the fact that the established mouse model used to study cisplatin-induced AKI does not recapitulate the cisplatin dosing regimen patients receive. In recent years, work has been done to develop more clinically relevant models of cisplatin-induced kidney injury, with much work focusing on incorporation of multiple low doses of cisplatin administered over a period of weeks. These models can be used to recapitulate the development of CKD after AKI and, by doing so, increase the likelihood of identifying novel therapeutic targets for the treatment of cisplatin-induced kidney injury.

Cisplatin-Induced AKI in the Clinical Setting

cisplatin is a chemotherapeutic used in the treatment of many solid cancers, including head/neck cancer, lung cancer, and germline cancers (3, 11, 36). Although it is a potent choice for first-line treatment, cisplatin has several adverse side effects of which nephrotoxicity is its dose-limiting toxicity (3, 36). Thirty percent of patients that receive cisplatin treatment (either alone or in combination with other chemotherapeutics) will develop kidney damage in the form of acute kidney injury (AKI) (3, 5, 25, 34, 36).

Although there are several guidelines for defining AKI, it is generally regarded as a rapid decline in kidney function over only a period of days (42, 59). The pathology of AKI, particularly cisplatin-induced AKI (CDDP-AKI), is complex and consists of inflammation, vascular injury, oxidative stress, and proximal tubular injury. Mechanisms of proximal tubule injury are widely studied and include autophagy, DNA damage, mitochondrial dysfunction, and cell cycle dysregulation, pathways that converge on apoptotic and necrotic cell death. This ultimately leads to tissue damage followed by a decline in renal function as measured by serum creatinine (SCr) (28, 30, 34, 36, 38).

Although these processes have been well documented in animal models of cisplatin-induced AKI, what is known about these processes in humans is very limited. For one, AKI is often diagnosed using a criteria such as RIFLE that focuses more on clinical markers [namely glomerular filtration rate (GFR) and SCr] rather than pathology (43, 59). This is due to the fact that obtaining a kidney biopsy is an invasive procedure. Acute tubular necrosis (ATN) is often used as an indicator of AKI when biopsies are taken (46). In one study of kidney biopsies taken from patients who had AKI that was further exacerbated by nephrotoxicity, very little ATN was present in these biopsies (46). With animal models of cisplatin-induced AKI, there is a large amount of ATN. However, the lack of ATN in this biopsy study could be attributed to several confounding factors. For example, only a very small amount of kidney tissue was taken during biopsy and may not have accurately sampled the area where damage had occurred (46). This study sheds light on the fact that if patient biopsies were used to diagnose AKI, the severity of the AKI could be greatly underestimated. This study also indicates that with animal models of AKI, we can model the development of ATN that is seen in patients, although it may be a more severe form in animals (46).

Generally speaking, patients diagnosed with AKI have poor prognoses. Ten to 15 percent of patients with AKI will require dialysis, which is costly and does not ameliorate the damage to the kidney (34, 58, 61, 62). In addition, patients with AKI have a 50% mortality rate (8, 61, 62). Large-scale longitudinal studies have indicated that patients who sustain an AKI episode are also more likely to develop long-term adverse outcomes, namely chronic kidney disease (CKD) (6, 7, 15, 37).

The likelihood of developing CKD after AKI increases with the severity of AKI sustained, as well as multiple occurrences of AKI (7, 10, 15, 20). It is estimated that there are 600,000 new cases of AKI each year, and ∼20% of these patients will develop CKD in as little as 18–24 mo (13). CKD is a progressive kidney disease that is marked by a gradual decline in kidney function, proteinuria, and development of both interstitial and glomerular fibrosis (55). The prognoses for CKD, much like AKI, are adverse. Patients with CKD may require dialysis, and the death rate from CKD alone is 400 people per million per year in the US (55). Other studies have indicated that not only does AKI increase a patient’s likelihood of developing CKD but that AKI can progress to CKD (6, 7). Animal models of AKI have shown that there is irreversible damage to peritubular capillaries and that AKI can trigger prolonged inflammation even after SCr returns to baseline levels (15, 67). This long-term inflammation is one of the factors contributing to the development of renal fibrosis.

Of the deaths related to AKI, ∼19% can be attributed to CDDP-AKI, making it a major health problem (34, 58). With regard to CKD, there are few studies that look specifically at the development of CKD after CDDP-AKI. However, a case report from Sasaki et al. (47) documented two cases of permanent renal failure in patients who had undergone cisplatin treatment for esophageal cancer. In this case study, two males in their mid-60s received 80 mg/m² cisplatin as part of neoadjuvant therapy. Treatment with cisplatin was followed by a severe peak in SCr that required immediate hemodialysis (47). A kidney biopsy revealed mild glomerulosclerosis, mild interstitial infiltration of lymphocytes, and the presence of interstitial fibrosis (47). These pathologies are indicative of CKD. However, data on the development of CKD are clouded by the fact that adult patients may have risk factors that predispose them to the development of CKD. These risk factors include diabetes, hypertension, other cardiovascular diseases, and smoking (18). In this case study, one of the patients was also suffering from congestive heart failure at the time of biopsy (47).

To better understand the development of CKD after cisplatin-induced AKI, a few studies have been done in the pediatric cancer population. This population is ideal for long-term studies since pediatric patients tend to have fewer CKD risk factors compared with the adult population. Therefore, better conclusions can be made about the rate of development of long-term renal effects after cisplatin treatment (4, 18, 50). Pediatric cancer patients have a ninefold greater risk of developing long-term kidney damage as a result of the disease itself or treatment of the disease (4). In a study by Arga et al. (4), ∼50% of pediatric cancer patients showed signs of nephrotoxicity 4 yr posttreatment with cisplatin. Although these children did not present with severe nephrotoxicity clinically, 36.4% maintained a lower GFR, a common measure used to diagnose CKD (4). In another study by Skinner et al. (50), a 10-yr followup study indicated that GFR was still decreased in 11% of pediatric cancer patients treated with cisplatin, carboplatin, or combination of the two drugs. In addition, 15% of the patients had severe nephrotoxicity even 10 yr posttreatment (50). These studies indicate that the nephrotoxic side effects of cisplatin treatment are long-lasting, and GFR remains decreased in a large percentage of pediatric cancer patients, which is indicative of a permanent loss of function associated with CKD.

Lack of Therapies for Cisplatin-Induced Kidney Injury

Ideally, physicians want to protect from CDDP-AKI or ameliorate kidney damage that has already been sustained from cisplatin treatment, especially once AKI has progressed to CKD. Unfortunately, no therapies are available (21, 32, 36). Intravenous saline infusions are often administered to patients before and during cisplatin treatment. Although the mechanism of action is unknown, saline infusions do seem to have some renoprotective effects against cisplatin treatment (54). However, even with this precaution in place, 30% of patients will still develop CDDP-AKI (34, 36). An alternative option is to utilize a less nephrotoxic platinum chemotherapeutic (carboplatin or oxaliplatin) to treat patients who may have an increased risk of developing CDDP-AKI (52, 65). Although these drugs are less nephrotoxic, they are not as effective as cisplatin is for treatment of certain cancers (65).

The only FDA-approved renoprotective agent is amifostine, which can deplete the nephrotoxic metabolite of cisplatin (12, 23, 45, 48). However, clinical trials leading to amifostine’s FDA approval were performed mainly in ovarian cancer patients. The dosing regimen for cisplatin treatment of ovarian cancer consists of one high dose of cisplatin (12, 45). Currently, most cancers treated with cisplatin utilize a repeated, low-dosing regimen to curtail the risk of developing AKI while maintaining therapeutic efficacy. When amifostine has been used with this type of cisplatin-dosing regimen, severe nephrotoxicity has occurred, suggesting that either amifostine is not renoprotective or amifostine itself may be nephrotoxic (12, 48). For this reason, amifostine is no longer used to prevent CDDP-AKI for most types of cancer.

Limitations of Preclinical Mouse Models of Cisplatin-Induced AKI

The lack of therapies for CDDP-AKI and subsequent progression to CKD may stem from the lack of clinically relevant animal models used to study AKI. The most commonly used animal model for studying CDDP-AKI is a mouse model in which mice are administered one very high dose of cisplatin (10–30 mg/kg). With this model, many potential renoprotective/injury ameliorating therapies have been tested and found to provide some therapeutic effect, but have then failed in early-stage clinical trials (21, 32). The reason for this outcome is multifaceted (Table 1). First, this high-dose cisplatin mouse model of AKI does not recapitulate the cisplatin dosing regimen humans receive for most forms of cancer. As previously described, patients usually receive low doses of cisplatin over an extended period of time to lower risk of nephrotoxicity while the drug’s therapeutic efficacy against cancer is maintained. Furthermore, mice treated with high-dose cisplatin become moribund 3–4 days after treatment. In humans, the goal of cisplatin treatment is to kill cancer, thereby improving patient longevity. In addition, the high-dose mouse model does not allow for long-term sequelae studies that may be associated with CKD. Finally, mice used for CDDP-AKI studies are usually young (8–12 wk of age) and do not have cancer. This does not take into consideration that only patients with cancer receive cisplatin treatment or that the median age of all cancer diagnoses is 65 yr of age, suggesting that cancer is a disease associated with aging (31).

Table 1.

The current mouse model used to study CDDP-AKI does not recapitulate the dosing regimen humans receive

Mouse dosing regimen of CDDP used to study AKI	Human dosing regimen of CDDP used to treat cancer
Mice receive 1 high dose of CDDP (10–30 mg/kg)	Humans receive low doses of cisplatin over a span of weeks/months
Mice die 3–4 days after CDDP treatment; long-term studies of kidney function cannot be performed	Goal of cisplatin treatment is to cure cancer without developing AKI, improve patient longevity
Mice used do not have cancer	Only patients with cancer receive cisplatin treatment
Mice used are usually young (8–10 wk of age)	The median age for all cancer diagnoses is 65

CDDP-AKI, cisplatin-induced acute kidney injury. The mouse model of CDDP-AKI consists of 1 high dose of cisplatin that causes mice to become moribund 3–4 days after treatment. In addition, mice used with this model are usually young and do not have cancer. To improve clinical relevancy, these limitations must be addressed.

The development of a more clinically relevant mouse model of cisplatin-induced kidney injury (CDDP-KI) would likely increase the chances of identifying novel therapeutic targets for treatment of this disease. To develop a more translational mouse model, limitations of the single high-dose cisplatin mouse model used to study AKI need to be addressed. Specifically, a clinically relevant mouse model of CDDP-KI would have to include a repeated low-dose regimen of cisplatin, be able to be used to study long-term outcomes of AKI leading to CKD, and include comorbidities of advanced age and cancer.

Addressing the Limitations of the Currently Used Mouse Model of Cisplatin-Induced AKI

Several groups have begun to develop more translational models of CDDP-KI that can be used to bridge the gap between AKI and CKD to specifically look at the development of fibrosis. Torres et al. (56) developed both a repeated dosing model of cisplatin injury and an innovative way to track changes to nephron structure as injury occurs using multiphoton microscopy. This model consisted of administering 15 mg/kg ip cisplatin to 10-wk-old C57BL6 mice, followed by another dose 2 wk later. These mice were able to survive for 25 wk. At 9 wk posttreatment, mice treated with cisplatin showed an increase in plasma creatinine levels that was indicative of kidney function loss. In addition, there was a >10-fold increase in apoptotic cells and a greater than threefold increase in Ki-67-positive cells, indicating that cell turnover was occurring (56). Using multiphoton microscopy Torres et al. (56) found that the glomerular capsules were abnormally altered in cisplatin-treated mice, although the glomeruli themselves remained largely unaltered. In addition, these abnormal glomerular capsules were highly associated with a decrease in GFR. However, these cisplatin-treated mice did not show overt collagen accumulation, as indicated by Sirius red staining, and there was not an increase in macrophage activity or myofibroblast proliferation (56). Because these are markers of CKD, this may suggest that the CKD phenotype was not overt and that fibrosis had not fully developed.

Ravichandran et al. (41) developed a model of repeated cisplatin injury that consisted of administering 10 mg/kg ip cisplatin once/wk for 4 weeks in 8- to 10-wk-old male C57BL6 mice. At the end point of this study, mice treated with this dosing regimen of cisplatin had a significant increase in blood urea nitrogen (BUN) and serum creatinine (SCr) that was indicative of severe loss of kidney function (41). Pathology revealed moderate levels of apoptosis and necrosis. There was also a significant increase in Tnfα and Cxcl1, which is indicative of an inflammatory response (41). In addition, mice treated with this repeated dosing regimen of cisplatin had increased collagen levels, as evidenced by Sirius red staining, although fibrosis was not overt (41). Although this model utilized a dosing regimen that better recapitulates the human dosing regimen, the high levels of cell death indicate more of an AKI phenotype than progression to CKD.

Similarly, Katagiri et al. (22) also administered 10 mg/kg cisplatin in 7- to 8-wk-old male C57BL6 mice for only three weekly doses. With this model, Katagiri et al. (22) showed that there was an increase in BUN and SCr at the week 4 end point. In addition, pathological analysis revealed loss of brush borders and ATN (22). There was also a significant increase in fibrosis, as evidenced by Masson trichrome staining (22). mRNA levels of Tgfβ and αSma also increased significantly, further supporting the presence of fibrosis (22). However, all mice in these studies died by week 4, indicating that, although this model can be used to study kidney fibrosis, studies lasting longer than 4 wk are limited. This is somewhat contradictory to the study performed by Ravichandran et al. (41), in which the mice survived four doses of 10 mg/kg cisplatin and decreased survival of the mice was not mentioned.

The models discussed above utilized C57BL6 mice. It is accepted that this strain of mouse is resistant to developing glomerular changes associated with kidney fibrosis and CKD, and some studies suggest that this strain is also resistant to the development of tubulointerstitial fibrosis (53, 63). Using streptozotocin to induce diabetic nephropathy, a common cause of end-stage renal disease, Sugimoto et al. (53) demonstrated that C57BL6 mice developed glomerular lesions and albuminuria associated with the pathophysiology of diabetic nephropathy but did not develop tubulointerstitial fibrosis, whereas other strains of mice used did. Both Ravichandran et al. (41) and Sugimoto et al. (53) indicated only slight fibrosis development with their models, and while Katagiri et al. (22) showed robust fibrosis, these mice do not survive long-term. To address this issue, Sharp et al. (49) developed a repeated, low dose cisplatin model utilizing FVB background mice. In this model, 8-wk-old FVB mice were administered 7 mg/kg cisplatin once/wk for 4 wk and were euthanized on day 24 (49). With this dosing regimen, the survival rate was 100% (49). Blood urea nitrogen (BUN) increased in mice treated with cisplatin, indicating a loss of kidney function (49). Analysis by a pathologist indicated that there was very little ATN, and Western blot analysis for cleaved caspase-3 (a marker of apoptosis) indicated little apoptosis in cisplatin-treated kidneys (49). However, pathology revealed a significant increase in tubulointerstitial fibrosis, and this was validated with Sirius red staining for collagen deposition and α-smooth muscle actin (α-SMA) immunohistochemistry for myofibroblasts (49). FVB mice treated with this repeated dosing regimen of cisplatin had increased transforming growth factor-β (TGFβ) and fibronectin, as shown with Western blot analysis (49). Taken together, these data suggested a robust fibrotic phenotype. In addition, the mice treated with repeated dosing regimens of cisplatin can be aged out for ≥6 mo and have increased SCr indicative of persistent loss of kidney function (data not published), providing a useful tool for studying long-term kidney outcomes. Sharp et al. (49) also indicated that it is the repeated dosing of cisplatin that induces fibrosis, as only a single dose of 7 mg/kg cisplatin does not lead to overt changes in collagen deposition or activation of the TGFβ signaling pathway involved in fibrosis.

A question generally posed for all of these models is how the dose used for repeated administration compares with the doses patients receive. Although this question is one of high importance, it is also one that is difficult to answer. Simple conversions of doses between the two species are not accurate even when body surface area is taken into account. In addition, mice and humans differ in their rate of cisplatin clearance. The peak plasma concentration of cisplatin is 2.4–20 times greater in mice than in humans (60). However, the half-life of the drug is also six times shorter in mice than in humans. This corresponds to a more rapid distribution of cisplatin to tissues, suggesting that mice may be at higher risk for cisplatin nephrotoxicity (60). Additionally, the dosing of cisplatin used in humans varies based on the patient’s individual ability to tolerate the drug as well as the type of cancer. For cancers where treatment with cisplatin alone is meant to cure the cancer (as is the case with testicular cancer), the dose of cisplatin used is generally high (17, 65). However, in cancers where cisplatin is used in conjunction with other treatment (radiation, surgery, etc.), the doses are usually lower (17, 65). Generally speaking, patients will receive 60–100 mg/m² every 3–4 wk for several months (51); 100 mg/m² is considered to be a highly toxic dose in human and is always accompanied with intense hydration therapy (60). Although the dosing used in these mouse studies described may not fully mimic the dose patients receive, these models do indicate that it is the repeated dosing of cisplatin that leads that to poor long-term renal outcomes and that changes indicative of CKD do occur.

Future Directions for Improving Clinical Relevancy of Newly Developed Models of Cisplatin-Induced Kidney Injury

Taken together, the models discussed represent a foray into the development of more clinically relevant mouse models of CDDP-KI. Although the clinical data on kidney pathology associated with CKD that develops after CDDP-AKI is limited, the models here do have similar pathologies. This highlights the fact that not only are these models are clinically relevant but, that these models provide researchers the best opportunity to identify novel therapeutic targets as well as test therapeutics with the intent of ameliorating kidney injury sustained from cisplatin. Although these models incorporate a dosing regimen similar to what patients receive, the comorbidities of advanced age and cancer have not been fully addressed.

Aging has been indicated as a risk factor for development of worsened kidney injury, regardless of the type of insult to the kidney (44, 64, 66). Elderly patients are more likely to have comorbidities such as congestive heart failure and renovascular disease that can cause renal damage (1, 9, 64). In addition, comorbidities such as these increase elderly patients’ likelihood of being exposed to surgeries or drugs that may further stress the kidney. Even without these comorbidities, the aging kidney undergoes a multitude of structural and functional changes (1, 64). During normal aging, there is a decrease in renal growth factors that may make cells in the kidney more susceptible to apoptosis, which can result in reduction of total renal mass (1). Furthermore, there is marked glomerulosclerosis and thickening of renal vessel walls. This results not only in increased susceptibility to kidney injury but also a decline in kidney function (1, 64). It has also been suggested that some people will develop fibrosis indicative of CKD without any apparent insult throughout their life (26, 27, 66). Furthermore, patients who are older and have altered kidney function are often disqualified from receiving cisplatin treatment. Since many of the processes that change during normal kidney aging are also involved in the pathophysiology of CDDP-KI, this may mean that elderly patients are at a higher risk of developing worsened injury and fibrosis after repeated cisplatin treatment (15). Thus, determining how age may exacerbate kidney injury or fibrosis associated with CKD may provide us with a therapeutic target that would enable elderly patients to receive cisplatin treatment and also improve long-term overall survival in this population.

Incorporation of cancer into a clinically relevant model of cisplatin treatment is perhaps the most important factor to include, as only patients with cancer receive cisplatin treatment. In a seminal study by Oliver et al. (33), an LSL-Kras^G12D/+ model of lung adenocarcinoma was used in combination with a more clinically relevant dosing regimen of cisplatin. Seven milligrams per kilogram of cisplatin was administered once/week for 2 wk, followed by a 2-wk rest period to allow for recovery from cisplatin-associated toxicities, including nephrotoxicity. This cycle was repeated for a total of four doses of cisplatin. At the end of treatment, Oliver et al. (33) reported that tumors present in LSL-Kras^G12d/+ mice treated with cisplatin had increased genomic instability indicative of a more aggressive tumor. It was concluded that either cisplatin treatment was selected for already existing tumor cells with these aberrant phenotypes or that repeated administration of cisplatin caused irreparable DNA damage in select tumors, ultimately resulting in chromosomal instability (33). However, reports on the toxicities sustained from this dosing regimen were not included in this study.

The nephrotoxicity of cisplatin in addition to the comorbidity of cancer was first addressed by Pabla et al. (35). In this model, 7- to 8-wk-old athymic female mice were inoculated with ovarian cancer cells subcutaneously. After tumors had formed, mice were injected intraperitoneally with 10 mg/kg cisplatin or saline vehicle weekly. Only 30% of mice survived until week 4, and none were able to receive a fifth dose of cisplatin (35). SCr and BUN levels increased significantly in cisplatin-treated mice. When mice were treated with repeated doses of cisplatin and rottlerin (an inhibitor of PKCδ), survival significantly improved, and 25% of mice were able to receive a sixth dose of cisplatin (35). In addition, treatment with rottlerin significantly decreased BUN and SCr. However, there were no data reported on whether or not these mice developed fibrosis with this dosing regimen of cisplatin. Furthermore, weekly cisplatin treatment resulted in a high level of apoptosis in the kidney, as evidenced by TUNEL staining (35). With rottlerin treatment, the levels of apoptosis were decreased (35). These data suggest that rottlerin was protective due to alleviation of high levels of cell death. Most therapeutics for CDDP-AKI that have been tested in clinical trials also target cell death, suggesting that although rottlerin may be protective in animal models, its success in clinical trials would be unlikely (21, 32).

A limitation of this model is that Pabla et al. (35) utilized a xenograft model of ovarian cancer. Mouse xenograft models of cancer are limited in scope, as the tumors that develop do not form in the location of origin, and it is widely recognized that the tumor microenvironment plays a large role in the response of the tumor to chemotherapy (2, 57). Furthermore, xenograft models utilizing human cancer cell lines employ immunocompromised mice for engraftment of the cancer cells. Although this model is good for determining how human cancers respond to treatment, using this model to study cisplatin nephrotoxicity is limited. Several groups have indicated that immune cells play an important role in cisplatin nephrotoxicity (14, 15, 30, 39–41, 54). Without an intact immune system, the kidney’s response to cisplatin may be completely altered. These limitations also apply to Ravichandran et al.’s (41) repeated dosing of cisplatin work in a xenograft model of lung adenocarcinoma.

A better approach would be to use a genetically engineered mouse model of cancer that is based on known driver mutations that are relevant to human cancers, as they better mimic the gradual progression of cancer in patients, and the mice have intact immune systems. An example of such a model is the Tet-O-Kras4bG12D transgenic mouse model of lung adenocarcinoma. In this model Tet-O-Kras4bG12D transgenic mice are crossed with mice that have a reverse tetracycline transactivator (rTtA) linked to a CCSP promoter that is specific for type II lung epithelial cells. The development of tumors occurs gradually in mice containing the transgenes for both rTtA linked to the CCSP promoter and Kras4bG12D after administration of doxycycline for 3–6 mo (16). Therefore, this mouse model incorporates both aspects of age and cancer that would be important for increasing clinical relevancy of the repeated dosing regimen models of cisplatin. In addition, a transgenic model of cancer provides the opportunity to explore cancer as the systemic disease it is and may indicate that there is interaction between the cancer itself and distal organs, including the kidney. This becomes apparent as recent studies have shown that lung adenocarcinoma acts as an endogenous circadian organizer and changes the metabolic capability of the liver (29). The kidney is also known to be highly circadian regulated, so in the presence of lung adenocarcinoma it may also have altered metabolic capabilities. Ultimately, this could lead to altered kidney injury outcomes with cisplatin treatment. Although there are still limitations that must be addressed, the repeated dosing models of cisplatin described here acknowledge the need for more clinically relevant/highly translational animal models (Table 2). In addition, these models are providing an innovative way to study onconephrology, which has become an emerging research topic in recent years. Clinically, patients with cancer will see an oncologist who will track their SCr levels during the course of their treatment. Only when patients show signs of AKI are they referred to nephrologists. If a patient recovers from AKI, few receive follow-up consultations with nephrologists. However, recent studies have indicated that nonsignificant increases in SCr (≤25%) are associated with the development of CKD, so patients without AKI that meet clinical criteria may require followup visits to nephrologists, as they run an increased risk of developing CKD (6, 7, 24, 37). Bridging the gap between oncology and nephrology, both through improving mouse models to study CDDP-KI as well as altering clinical care, will be crucial to improving longevity and quality of life for patients that develop CDDP-KI.

Table 2.

New clinically relevant models have been developed to study CDDP-AKI

Reference	Dose of Cisplatin Used, mg/kg	Frequency of Dose	Mice Used	Results
Torres et al. (56)	15	Two doses with a 2-wk period in between each dose	C57BL6; 10 wk old, sex not identified	Mild fibrosis but abnormal glomerular capsules corresponding to a decrease in GFR; mice survived ≥25 wk
Ravichandran et al. (34)	10	One dose/wk for 4 wk	C57BL6; 8–10 wk old, male	Mild fibrosis but high levels of cell death indicative of AKI; no indication of overall survival
Katagiri et al. (17)	10	One dose/wk for 3 wk	C57BL6; 7–8 wk old, male	Development of overt fibrosis; mice died within 4 wk
Sharp et al. (39)	7	One dose/wk for 4 wk	FVB; 8 wk old; male	Mice develop overt fibrosis with repeated dosing; mice can be aged out ≥6 mo

These models all utilize a repeated dosing regimen of cisplatin, and can be used to study the fibrosis that develops after AKI that is indicative of a CKD phenotype even though the severity of fibrosis varies between models.

GRANTS

Support for this work was provided by National Institute of Diabetes and Digestive and Kidney Diseases Grant R01-DK-093462 (to L. J. Siskind).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

C.N.S. drafted manuscript; C.N.S. and L.J.S. edited and revised manuscript; C.N.S. and L.J.S. approved final version of manuscript.