Year : 2015 | Volume
: 2 | Issue : 3 | Page : 75--80
Renal Tubulointerstitial Fibrosis: A Review in Animal Models
Jie Zhao1, Li Wang2, Aili Cao2, Minqian Jiang1, Xia Chen1, Wen Peng3,
1 Department of Nephrology, Putuo Hospital, Shanghai University of Traditional Chinese Medicine, Shanghai 200062, China
2 Laboratory of Renal Disease, Putuo Hospital, Shanghai University of Traditional Chinese Medicine, Shanghai 200062, China
3 Department of Nephrology; Laboratory of Renal Disease, Putuo Hospital, Shanghai University of Traditional Chinese Medicine, Shanghai 200062, China
Department of Nephrology, Laboratory of Renal Disease, Putuo Hospital, Shanghai University of Traditional Chinese Medicine, 164 LanXi Road, Shanghai 200062
Deterioration of renal function is closely related to renal interstitial fibrosis (RIF). With animal models, we can simulate the pathological changes and progression of disease. Various animal models of RIF are pivotal for clinical treatment and investigation of new drugs. Reviewed here are methods of establishment and mechanism of commonly used animal models of RIF to help researchers select the optimal animal model for different research purposes.
|How to cite this article:|
Zhao J, Wang L, Cao A, Jiang M, Chen X, Peng W. Renal Tubulointerstitial Fibrosis: A Review in Animal Models.J Integr Nephrol Androl 2015;2:75-80
|How to cite this URL:|
Zhao J, Wang L, Cao A, Jiang M, Chen X, Peng W. Renal Tubulointerstitial Fibrosis: A Review in Animal Models. J Integr Nephrol Androl [serial online] 2015 [cited 2020 Jul 8 ];2:75-80
Available from: http://www.journal-ina.com/text.asp?2015/2/3/75/161428
Renal interstitial fibrosis (RIF), the common pathway, and major pathological change for different kidney diseases to progress to chronic kidney failure, is also a crucial factor in determining the progressive change of renal function. Recent studies have indicated that tubulointerstitial fibrosis (TIF) may not only develop from tubulointerstitial diseases, but also from certain glomerular diseases, with reciprocal effects. Further, impairment in renal function is also closely correlated with tubulointerstitial injury, the severity of which predicts the degree of damaged renal function, and even surpasses that caused by glomerular diseases.  Therefore, it is of great significance to treat chronic kidney diseases by investigating therapeutic measures and search for new medications to ameliorate the progression of TIF with the aid of animal models.
Animal models with RIF is generally established by medications such as cyclosporine A (CsA), aminoglycoside antibiotics or nephrotoxic drugs, or by surgery such as unilateral ureteral obstruction (UUO), 5/6 nephrectomy, kidney ischemia-reperfusion, and so forth.
Unilateral ureteral obstruction model
Unilateral ureteral obstruction is the most commonly used animal model to investigate the pathogenesis of RIF and evaluate antifibrotic treatment. In 1972, Tanagho et al. first introduced this method by ligating the ureter of lambs aged 70-75 days. Later, rats, possums, rabbits, pigs, dogs, monkeys, and other animals were also proposed for establishment of animal model, but rats are still the most common ones. Ureteral ligation is the most commonly adopted method, but new tubular organs may form in the surrounding area of the ligation to allow urine outflow via collateral pathways. Also, adhesion of surrounding tissue increases the difficulty to remove ligation in the future, especially in adult rats. One solution is to place small vascular clamps surrounding the ureter, but ureteral damage or misplacement following urine outflow are likely to occur. A silicone tube may be applied by ligating the ureter perpendicularly or by creating a partial obstruction with a parallel sleeve. It is a rather more convenient way to establish a UUO model with adult rats, but when mice are used, a thermostat operation console after anesthesia with isoflurane and a microscope with a clear visual field are detrimental in maintaining a low morbidity and mortality rate; based on these factors, the dexterity of the operator, postoperative adhesion, and tissue injury are also closely relevant.  Establishment of model: Inhalational anesthesia with isoflurane/oxygen under sterile condition, followed by median incision of lower abdomen. The left ureter is then divided, and two ties are knotted around the mid portion with 5/0 Mersilk suture, and the ureter was sectioned between the ligatures. The surgical incision was then closed with continuous peritoneal sutures and interrupted skin sutures.  Establishment of model was completed 7 days thereafter with mice and 2-3 weeks with rats.
Recent studies have suggested the mechanism of RIF after UUO as follows:
Infiltration of macrophages, causing release of cytokines to induce tubular cell apoptosis, activate and increase proliferation of fibroblasts;Glomerulopathy and tubular atrophy, induced by tubular cell apoptosis and necrosis; andTransformation of kidney cell phenotype.Chronic UUO produces active oxygen and nuclear factor kappa B by activation of renin-angiotensin system to further regulate preapoptotic signal transduction and subsequent tubular cell apoptosis. Contrarily, macrophages replaced produce anti-inflammatory factors to stimulate cell survival and proliferation. The two factors combined causes metabolic disorders and hemodynamic disturbances followed by tubular injury and cell apoptosis or necrosis and interstitial macrophage infiltration. Proliferation of interstitial fibroblasts and phenotype transformation of myofibroblasts causes extracellular matrix (ECM) accumulation that eventually induces fibrosis.
Interstitial fibrosis is the common pathological change of advanced nephropathy with various causes. Rodent UUO model causes progressive interstitial fibrosis. It is easily controllable in the timing, severity, and duration for ligation induced models. It is also easily duplicated and reversible and can induce fibrosis with only a short amount of time; therefore, it is considered a fast and reliable model with high success rate and low animal injury when considering pathogenic patterns and mechanisms of interstitial fibrosis. However, UUO is achieved with acute mechanical injury, and may not apply for all conditions of tubular and interstitial injury, and fibrosis.
Platt established RIF model with 5/6 nephrectomy in 1952, in which 2/3 of the left kidney was first excised, followed by right nephrectomy 1 week thereafter. Compensative renal hypertrophy, glomerular necrosis, capillary wall collapse, progressive mesangial distention, tubular cell atrophy, tubular distention, interstitial inflammatory cell infiltration, fibrosis, and other postoperative complications, were observed. The 5/6 nephrectomy model refers to the remnant kidney model established by right 2/3 nephrectomy and left total nephrectomy. The following two-phase method is applied:  In the first phase, rats were anesthetized by enflurane inhalation and placed on thermostat operation console with a rectal temperature maintained at 37-38°C. A right flank incision was made to expose the right kidney, and the adrenal gland along with renal capsule was carefully dissected; then, approximately 2/3 of the right kidney (including the upper and lower poles) was excised. The second phase of surgery was performed 1 week later when the rats were anesthetized with the same method followed by an incision made on the left flank and total excision of the left kidney after dissection of the left adrenal gland. Sham operation was performed for rats in control group, in which the peritoneum was opened, but the kidneys were not removed. Both groups were monitored postoperatively for 8-13 weeks. It was also proposed by some researchers to make a 2-cm abdominal incision to expose fat tissue and adrenal gland and excise 1/3 of the left kidney (including the upper and lower poles) with electrotome; the remnant kidney was placed back to original position and sutured with 6-0 silk. The second phase was performed about a week later to ligate the pedicle with 5-0 silk and excise the right kidney. For sham operation group, surgery was performed simultaneously but with only anesthesia of the lower abdomen and surface incisions. 
An improved method was proposed by Ma and Xie  as follows: Wistar rats were anesthetized with 2% pentobarbital sodium (30 mg/kg) and an oblique incision was made on the left flank. The adipose capsule was dissected and renal artery exposed. One to two branches of the renal artery near the pedicle were clamped for 30 s to interrupt blood flow, and the corresponding ischemic areas appeared in a darkened purple-black color. Then the clamps were removed to selectively excise a total of 2/3 left kidney including the upper and lower pole, and peripheral areas. The scope of ischemia and relevant margins were used as reference sites for surgical excision. Gelfoam was used for hemostatic, and the remnant kidney was restored and incisions sutured. The right kidney was excised 1 week later. A total of 5/6 of kidney is excise. Typical pathological changes include glomerular hypertrophy, glomerular cell hyperplasia, glomerular wall adhesion, thrombus formation, glomerular fibrosis, interstitial distention, increased fibrous tissue, inflammatory cell infiltration, as well as coexistence of focal tubular atrophy and hypertrophy. Establishment of such model is easy and highly repeatable and is considered an ideal method to investigate the pathogenesis and mechanism of kidney fibrosis and ameliorate deterioration of renal function.
The unique structure, vasculature and blood supply in kidney makes the tubule highly sensitive to ischemia and hypoxia. Rats, rabbits, and dogs are commonly used animals to investigate acute ischemic injury. The time of ischemia is typically 30-60 min, and heparin is often administered preoperatively to prevent intrarenal coagulation of the ischemic kidney. The kidney injury of such model often presents as detachment of brush border of tubular epithelial cells, epithelial cell degeneration, necrosis, detachment and apoptosis, interstitial edema, and inflammatory cell infiltration. The degree of pathological change is closely relevant with ischemia-reperfusion time, therefore, it is a classic model for acute tubular injury.
Jain et al. used male Wistar rats for left nephrectomy and clamped the right pedicle for 45 min that rapidly induced nocturia, interstitial fibrosis, and loss of renal function. Male large white pigs (Le Magneraud, Surgθres, France) were also used by other scholars as animal model. Standard chow and water were provided. Midazolam nasal drip (0.2 mg/kg) was administered for rapid sedation followed by inhalational anesthesia by flurane and 100% oxygen. The operation was performed under sterile condition. A 20-caliber catheter was inserted into the auricular vein, and atropine sulfate (10 μg/kg) was administered intravenously to reduce pharyngeal and tracheal secretion. Left pedicle and ureter were gently dissected, and heparin sodium (100 U/kg) was rapidly infused 10 min before resection of the kidney.  To verify the correlation of warm ischemia time and status of different nephrons, researchers further adopted the three following groups: Bilateral nontraumatic clamping of the pedicles for 90 min to induce long-time warm ischemia (WI90, n_7), unilateral warm ischemia combined with opposite nephrectomy (50% of nephron mass reduction 1/2N_90, n_7), and unilateral warm ischemia with intact opposite kidney (WIuni90, n_7). Then the clamp was removed to carefully monitor adequate reperfusion. No thrombus in the renal artery was observed. Our results showed that three methods all successfully created interstitial fibrosis model, in which the unilateral heat ischemia combined with opposite nephrectomy model showed the most severe pathological change.  Acute ischemic kidney failure induces permanent necrosis of peritubular capillaries that is likely to develop progressive chronic kidney failure. Therefore, the ischemia-reperfusion model is mainly used for mechanistic investigations.
Unilateral renal vein ligation
In 1988, Gonzαlez-Avila et al. reported the unilateral renal vein ligation model of in rats. Thirty male Wistar rats weighed 200-250 g were anesthetized by intraperitoneal injection of chloral hydrate (0.1 mL/kg body weight). Left renal vein was exposed and ligated with 4-0 suture. Incision was then sutured, with a total operation time of 20 min. The advantages of this model are:
Easy operation technique that simply requires renal vein dissection and ligation;Low cost;The healthy opposite kidney compensates after unilateral operation, posing less impact on the general status to the animal;Shorter experiment time, in which the model is completed 10 days postoperatively;The pathological changes ideally simulate that occur in human.
Interstitial congestion and edema occurs at early stage with mononuclear cell infiltration, collagen hyperplasia, tubular epithelial degeneration and atrophic tissue changes; in mid to late stages, increase of interstitial fibroblasts, collagen disposition, tubular atrophy, and necrosis, as well as thickened arteriolar wall, are typically observed. 
Cyclosporine A nephropathy
Cyclosporine A activates intrarenal renin-angiotensin system that stimulates OPN expression and macrophage infiltration and enhances cell apoptosis, promotes growth factor (e.g., transforming growth factor-β1) release, and ECM aggregation to induce interstitial fibrosis. Male Sprague-Dawley (SD) rats weighed 180-240 g were given free access to water and standard moist salt-free chow (AltrominW standard 1320 with 0.2% sodium, large, Germany) with constant room temperature and a 12 h light/dark cycle. Rats were gavaged with 15 mg/kg/day CsA (dosages were slightly adjusted according to increase of body weight) for 21 days in total.  It was also proposed that after low-salt gavage for 1 week, CsA was diluted to 15 mg/mL with olive oil and injected subcutaneously at 1 mL/kg for the next 4 weeks,  or alternatively, dissolve CsA with olive oil for subcutaneous injection at 15 mg/kg/day. 
The CsA nephropathy model is typically applied to investigate the mechanism of CsA-induced RIF and search for new medications to prevent CsA renal toxicity. Despite of easy operation and high success rate, this model is of limited application value due to high cost, liver toxicity, long experiment duration, and mild interstitial lesions; in addition, the proportion of CsA-induced RIF is low among all etiological factors, and its course of fibrosis development is different from typical congestion-induced RIFs.
Adriamycin (ADR) or doxorubicin, a member of the anthracycline antibiotics, is a cell-cycle nonspecific agent that inhibits RNA and DNA synthesis, in which the inhibitory effect on RNA is the strongest available. ADR is majorly used clinically as chemotherapy agent, which kills tumor cells in different stages of the cell cycle. It generates active oxygen through metabolism that induces hyperoxygenation of lipid in glomerular epithelial cells guided by various lipid mediators and destroys the structure and function of glomerular filtration barrier to ultimately induce proteinuria. ADR-induced rat nephropathy model first causes vacuolar degeneration, massive proteinuria, and a gradual congestion in the proximal tubule and broken tubular basement membrane, thereby causing an inflammatory response and fibrosis in the interstitium.  When using mice as model, Balb/c mice aged 8-10 week were given free access to food and water and kept under constant temperature and a 12 h light/dark cycle. The mice were administered intravenously via tail vein with 10 mg/kg ADR without anesthesia. Pathology indicated that ADR caused glomerular and tubulointerstitial lesions, and lesions in the cortex aggravated gradually after 7-14 days and leveled off by day 21. 
Adriamycin induces protein casts in the distal tubule and congestive lesions, basement membrane damage, and interstitial inflammation and fibrosis in the proximal tubule. The pathological changes are similar with that found in human, therefore it is considered a classic model for kidney fibrosis. One disadvantage is the side effect of ADR. When leaked outside the blood vessel, local tissue ulcer and necrosis may occur. ADR has varied toxicity in different systems. For instance, it inhibits hematopoietic functions in the bone marrow and may lead to congestive heart failure in severe cardiac toxicity. In addition, a comprehensive evaluation standard for ADR-induced animal nephropathy model is still not available, and this has limited the evaluation and repeat applications of such model.
In 1975, Roth et al. first reported the effect of adenine on kidney function in dogs. In 1986, Yokozawa et al. gavaged rats with adenine (150 mg/kg/day) for 17 days consecutively to construct chronic renal failure model. It was also proposed that C57BL/6J mice were kept in standard cages with sawdust and paper roll with a room temperature of 21-22°C and a humidity of 40-50% in a 12 h light/dark cycle. All animals were given free access to water and chow and were adaptively fed for 1 week before the experimentation. To ensure intake of Adenine, the chow was mixed with tyrosine to cover the odor and taste of Adenine. Other ingredients included corn starch (39.3%), casein (20.0%), maltodextrin (14.0%), sucrose (9.2%), corn oil (5%), cellulose (5%), compound vitamins (1.0%), methionine (0.3%), and choline bitartrate salts (0.2%). Total phosphate: 0.9%, total calcium: 0.6%. After 8 weeks of feeding, fibroblast hyperplasia and interstitial fibrosis in tubular interstitium were observed.  Other scholars have proposed the use of male SD rats with 14-day consecutive feeding with chow mixed with 0.7% ADR. Pathology revealed partially damaged and atrophic, and partially distended tubules with interstitial lymphocyte and mononuclear cell infiltration, in which fibroblast hyperplasia and interstitial fibrosis were observed. 
Long-term administration in adenine-induced nephropathy model induces a metabolic disorder that is similar to human chronic renal failure. With adjustments in dosage and duration of drug delivery, chronic renal failure with different degrees can be designed. This model is easily quantified and reproduced, but due to long preparation time and high individual differences in drug reactions, the application of such model is still limited.
Aristolochic acid-induced nephropathy
In 1993, Belgium scientists Vanherweghem et al. first reported that the Chinese herb "Miao Tiao Wan" (slimming pills), containing Aristolochia fangchi and Magnolia officinalis, may induce progressive renal failure with a pathology of rapid progressive interstitial fibrosis, and therefore named such disease as "Chinese herbs nephropathy," or, "aristolochic acid nephropathy."
Huang et al. administered various dosages of Aristolochic acid I (AAI, Sigma I Aldrich, China) via intraperitoneal injection in C57BL/6 mice aged 8 weeks. All mice were fed at 20°C room temperature and 45% humidity with 12 h light/dark cycle and free access to water and food. The results indicated that 110 mg/kg AAI induces death caused by acute renal failure in 14 days. Tubular cells were detached from the basement membrane causing partial tubular congestion. Single or two-time (d1 + d14) administration at 2.5 mg/kg would not cause fibrosis in a month. Weekly administration of AAI at 2.5 mg/kg four times consecutively plus 1 week of disease advancement causes partial tubular damage in peripheral cortex and medulla. Administration at 3 mg/kg every 3 days for 6 weeks plus an additional 6 weeks led to a more severe interstitial tubular fibrosis. The last method is more suitable in etiological analysis in that a markedly reduced renal function is accompanied with a clear tubular interstitial fibrosis of the peripheral cortex and medulla, and the lesion presented in C57BL/6 mice was in consistent with that in human. This model has a similar pathological characteristic compared with that in human chronic TIF and is widely applied in investigation of renal toxicity in Aristolochia and Asarum genus herbs. However, it takes about 6-21 months for model preparation, a long cycle, complicated method and long feeding time, which all contribute to a high total cost that limits its application.
Gilbert et al. administered rats with 40 mg/kg gentamycin for 14 days and observed renal failure and proximal tubular necrosis; interstitial fibrosis occurred after 30 days of withdrawal. Wang Fei et al. administered rats peritoneally with 80 mg/kg/day gentamycin sulfate, twice a day for 9 days, and blood was harvested via abdominal aorta 5 and 30 days before sacrifice; a same amount of saline was injected peritoneally in control group. Five days after administration, kidney coefficient, blood and urine creatinine, and blood urea nitrogen were markedly higher than control group, with tubular epithelial cell degeneration and necrosis, massive protein casts and mild cell, and granular casts observed under microscope. Tubular atrophy and distention, epithelial cell necrosis, inflammatory cell infiltration were present, along with a marked hyperplasia of interstitial fibrous tissue and distended interstitium, but the structure of glomerulus remained intact. This model is primarily used to investigate the renal damage of aminoglycoside antibiotics to further elaborate the metabolic pathway of drugs via kidney and their nephrotoxicity effects.
In all, there are many ways to construct a kidney fibrosis animal model, each with its own advantages and disadvantages. Although surgery-induced UUO model is prone to intraoperative bleeding and with certain damage to the animal itself, it is still a convenient way to establish homogeneous lesions with a short preparation time and great repeatability. Therefore, it is considered an officially acknowledged model with the widest application in etiological and mechanistic researches. Drug-induced and other models have common shortages of long preparation period, nonhomogeneous lesions, expensive costs, extrarenal damage, late or insignificant appearance of fibrosis, or difficulties in model preparation, specimen harvest, as well as varied evaluation criteria and poor repeatability. These shortcomings have limited the wide application in RIF researches in varying degrees.
This work was supported by General Medicine of Key Discipline Construction Project, State Administration of Traditional Chinese Medicine of the People's Republic of China; Leading Academic Discipline Project of State Administration of Traditional Chinese Medicine of China, Talent Project of Integrative Medicine of Shanghai Municipal Health Bureau (ZYSNXD012-RC-ZXY); Key Medical Discipline Project of Shanghai Municipal Health Bureau (ZK2012A34); Independent Innovation Research Fund of Putuo District Science and Technology Committee (2012PTKW002).
|1||Hodgkins KS, Schnaper HW. Tubulointerstitial injury and the progression of chronic kidney disease. Pediatr Nephrol 2012;27:901-9.|
|2||Chevalier RL, Forbes MS, Thornhill BA. Ureteral obstruction as a model of renal interstitial fibrosis and obstructive nephropathy. Kidney Int 2009;75:1145-52.|
|3||Kim MJ, Turner CM, Hewitt R, Smith J, Bhangal G, Pusey CD, et al. Exaggerated renal fibrosis in P2X4 receptor-deficient mice following unilateral ureteric obstruction. Nephrol Dial Transplant 2014;29:1350-61.|
|4||Kim JA, Choi HJ, Kwon YK, Ryu do H, Kwon TH, Hwang GS. 1H NMR-based metabolite profiling of plasma in a rat model of chronic kidney disease. PLoS One 2014;9:e85445.|
|5||Zou H, Zhao X, Sun N, Zhang S, Sato T, Yu H, et al. Effect of chronic kidney disease on the healing of titanium implants. Bone 2013;56:410-5.|
|6||Ma L, Xie R. A modified chronic renal failure model induced by 5/6 nephrectomy. Chin J Trauma Disabil Med 2010;18:5-7.|
|7||Jain S, Bicknell GR, Nicholson ML. Tacrolimus has less fibrogenic potential than cyclosporin A in a model of renal ischaemia-reperfusion injury. Br J Surg 2000;87:1563-8.|
|8||Hauet T, Goujon JM, Baumert H, Petit I, Carretier M, Eugene M, et al. Polyethylene glycol reduces the inflammatory injury due to cold ischemia/reperfusion in autotransplanted pig kidneys. Kidney Int 2002;62:654-67.|
|9||Cau J, Favreau F, Zhang K, Febrer G, de la Motte GR, Ricco JB, et al. FR167653 improves renal recovery and decreases inflammation and fibrosis after renal ischemia reperfusion injury. J Vasc Surg 2009;49:728-40.|
|10||González-Avila G, Vadillo-Ortega F, Pérez-Tamayo R. Experimental diffuse interstitial renal fibrosis. A biochemical approach. Lab Invest 1988;59:245-52.|
|11||Guo H, Zou W. Establishment of animal models of renal tubulointerstitial fibrosis. J Clin Exp Pathol 2003;19:211-2.|
|12||Nielsen FT, Jensen BL, Hansen PB, Marcussen N, Bie P. The mineralocorticoid receptor antagonist eplerenone reduces renal interstitial fibrosis after long-term cyclosporine treatment in rat: Antagonizing cyclosporine nephrotoxicity. BMC Nephrol 2013;14:42.|
|13||Kim JY, Ghee JY, Lim SW, Piao SG, Chung BH, Yoon HE, et al. Comparison of early and late conversion of sirolimus in experimental model of chronic cyclosporine nephropathy. J Korean Med Sci 2012;27:160-9.|
|14||Shi SH, Zheng SS, Jia CK, Zhu YF, Xie HY. Inhibitory effect of tea polyphenols on transforming growth factor-beta1 expression in rat with cyclosporine A-induced chronic nephrotoxicity. Acta Pharmacol Sin 2004;25:98-103.|
|15||Wang Y, Wang YP, Tay YC, Harris DC. Progressive adriamycin nephropathy in mice: Sequence of histologic and immunohistochemical events. Kidney Int 2000;58:1797-804.|
|16||Silveira KD, Barroso LC, Vieira AT, Cisalpino D, Lima CX, Bader M, et al. Beneficial effects of the activation of the angiotensin-(1-7) MAS receptor in a murine model of adriamycin-induced nephropathy. PLoS One 2013;8:e66082.|
|17||Roth GJ, Moore GL, Kline WE, Poskitt TR. The renal effect of intravenous adenine in humans. Transfusion 1975;15:116-23.|
|18||Yokozawa T, Zheng PD, Oura H, Koizumi F. Animal model of adenine-induced chronic renal failure in rats. Nephron 1986; 44:230-4.|
|19||Jia T, Olauson H, Lindberg K, Amin R, Edvardsson K, Lindholm B, et al. A novel model of adenine-induced tubulointerstitial nephropathy in mice. BMC Nephrol 2013;14:116.|
|20||Nicholas SB, Yuan J, Aminzadeh A, Norris KC, Crum A, Vaziri ND. Salutary effects of a novel oxidative stress modulator on adenine-induced chronic progressive tubulointerstitial nephropathy. Am J Transl Res 2012;4:257-68.|
|21||Vanherweghem JL, Depierreux M, Tielemans C, Abramowicz D, Dratwa M, Jadoul M, et al. Rapidly progressive interstitial renal fibrosis in young women: Association with slimming regimen including Chinese herbs. Lancet 1993;341:387-91.|
|22||Huang L, Scarpellini A, Funck M, Verderio EA, Johnson TS. Development of a chronic kidney disease model in C57BL/6 mice with relevance to human pathology. Nephron Extra 2013; 3:12-29.|
|23||Gilbert DN, Plamp C, Starr P, Bennet WM, Houghton DC, Porter G. Comparative nephrotoxicity of gentamicin and tobramycin in rats. Antimicrob Agents Chemother 1978;13:34-40.|
|24||Wang Fei, Lu Haiyan, Zhang Yue, Hao Yan-Peng, Zhou Juan, Liu Yu-Min, et al. Tubulointerstitial fibrosis model in rat induced by gentamycin. Chin J Comp Med 2008;8:60-4.|