Preliminary study of urinary excretion of liver-type fatty acid-binding protein in a cat model of chronic kidney disease

Abstract

Urinary liver-type fatty acid-binding protein (uL-FABP) is a clinically useful biomarker for monitoring chronic kidney disease (CKD) in humans. However, long-term monitoring of uL-FABP in CKD cats has not been reported. The objective of this preliminary study was to investigate whether the urinary excretion of L-FABP could predict the deterioration of renal function in 2 CKD model cats. Urinary liver-type fatty acid-binding protein (uL-FABP) increased before standard renal biomarkers, including serum creatinine, blood urea nitrogen, and symmetric dimethylarginine, in 1 cat with deteriorating renal function, but remained low and relatively stable in another cat with stable renal function. Our results suggest that uL-FABP is a potential clinical biomarker for predicting the progression of CKD in cats, as it is in humans.

Chronic kidney disease (CKD) is a leading cause of illness and death in aged cats, resulting in uremia at the end stages. The estimated prevalence of CKD in cats ranges from 28 to 50% (1,2). Although serum creatinine (sCre), blood urea nitrogen (BUN), and symmetric dimethylarginine (SDMA) concentrations are the markers of glomerular filtration rate (GFR) that are most often used to clinically diagnose CKD (3), there are few renal biomarkers for predicting and monitoring the progression of CKD in cats.

The urinary excretion of liver-type fatty acid-binding protein (L-FABP) is a useful biomarker for the early detection and monitoring of CKD in humans (4). We recently showed that L-FABP is expressed in renal proximal tubular cells and liver cells in cats, as in humans (5). Liver-type fatty acid-binding protein (L-FABP) is excreted into the urine in renal tubular ischemia and oxidative stress before the progression of renal damage in cats (6).

Kamijo et al (7) reported that urinary L-FABP (uL-FABP) reflected the clinical prognosis of CKD in humans. In another study, renal function deteriorated significantly in patients with “high” uL-FABP, compared to those with “low” uL-FABP. Therefore, it might be valuable to monitor cats with high uL-FABP indexes in order to predict the outcome of CKD, regardless of the increases in other renal biomarkers. Little is known, however, about the pathophysiological role of L-FABP and temporal changes in uL-FABP excretion during the progression of CKD in cats.

In this study, we used reversible unilateral ureteral obstruction and contralateral nephrectomy (RUUO-N) (8) to construct a cat model of CKD. We investigated the relationship between CKD progression and uL-FABP excretion, evaluating the potential of uL-FABP as a biomarker for predicting CKD progression in cats.

This study was carried out in accordance with local animal ethics guidelines and was approved by the Animal Research Committee of Iwate University (A201827). Two healthy female mixed-breed cats (Cat 1 — 4.5 kg, age 8 y; Cat 2 — 4.3 kg, age 7 y) were used to produce models of CKD. Before this study, both cats were healthy based on a physical examination, complete blood (cell) count, biochemical profiles, and urinalysis.

All surgeries were conducted in the following manner. Food and water were restricted for 12 h before all procedures. Subcutaneous cefovecin sodium (Convenia; Zoetis, Tokyo, Japan), 8 mg/kg body weight (BW), was administered preoperatively. The cats were premedicated with intravenous (IV) midazolam (Maruishi Pharmaceutical, Osaka, Japan), 0.3 mg/kg BW, and buprenorphine (Otsuka Pharmaceutical, Tokyo, Japan), 0.02 mg/kg BW. Anesthesia was induced with IV propofol (Mylan, Osaka, Japan), 7 mg/kg BW, until an effect was achieved and maintained with isoflurane in 100% oxygen. Temperature, pulse, respiration, blood pressure, and oxygen saturation were recorded at 10-minute intervals during anesthesia. A balanced electrolyte solution of lactated Ringer’s solution (5 mL/kg BW per hour) was administered intraoperatively. The cats were monitored continuously for 24 h; IV buprenorphine, 0.02 mg/kg BW was administered for postoperative pain relief. Both cats were given single doses of lactated Ringer’s solution (200 mL/d) subcutaneously at 24-hour intervals postoperatively until they began eating and drinking. Urine production, appetite, and mental status were monitored daily in each cat during the experiment.

Each cat underwent RUUO-N surgery. The left ureter was visualized, and the distal ureter was ligated with 4-0 monofilament absorbable sutures at 2 points (on day 0). On day 28 (4 wk later), the left ureter was released and anastomosed to the bladder mucosa with a modified ureteroneocystostomy technique using 8-0 nylon in a simple interrupted pattern. The right kidney was allowed to remain functional for 2 wk, whereas the left kidney function recovered. Then, a right nephrectomy was carried out on day 42. The duration of ureteral obstruction was based on the period during which no renal function was observed in a canine kidney obstructed for > 28 d (9).

The BUN, sCre, SDMA, serum phosphorus (sP), serum calcium (sCa), urine protein/creatinine (UPC) ratio, urine specific gravity (USG), and uL-FABP concentrations in each cat were measured at 2-day intervals for the first postoperative week, weekly until 4 mo after the right nephrectomy (day 161), and monthly thereafter. Symmetric dimethylarginine (SDMA) was among the renal biomarkers used to estimate the glomerular filtration rate (GFR); this measure is included in the clinical diagnostic criteria for CKD in cats [International Renal Interest Society (IRIS) CKD staging system].

The cats were euthanized with a pentobarbital overdose and a complete necropsy was conducted. The monitoring period was tentatively set at 560 d based on studies in which the median survival of cats with ureteral obstructions treated using a surgical device was > 463 d (10,11). Cat 2 was euthanized with a pentobarbital overdose on day 505 because its general health had worsened despite the fact that the sCre level was < 8.0 mg/dL. Cat 1 was euthanized on day 556 despite normal renal function.

The kidneys were assessed histopathologically, with percutaneous renal biopsies done monthly until 4 mo after the right nephrectomy (day 161) and at 3-month intervals thereafter. Renal tissue samples were fixed in 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4) at 4°C for 12 h, embedded in paraffin, sectioned at 2-μm thicknesses, and stained with hematoxylin-eosin (H&E) and Masson’s trichrome. Tubulointerstitial fibrosis and inflammation in the renal tissue samples were evaluated as follows: normal, no or rare small foci; mild, scattered multifocal areas affecting < 5% of the section; moderate, moderate fibrosis/inflammation affecting 25 to 50% of the section; and severe, diffuse or coalescing fibrosis/inflammation (12).

The uL-FABP concentration was measured using a 2-step sandwich enzyme-linked immunosorbent assay (ELISA) (Feline L-FABP ELISA Kit; CMIC Holdings, Tokyo, Japan). This ELISA kit is sensitive within an L-FABP concentration range of 1.9 to 375 ng/mL, according to the manufacturer’s documentation. Urine samples with L-FABP concentrations of > 375 ng/mL were measured after dilution with the standard diluent solution (0 ng/mL) included in the kit. The uL-FABP concentrations were normalized to the urinary creatinine (uCre) concentration to avoid variation caused by urine volume; the results were expressed as the uL-FABP index. The uCre concentrations were measured by Folin’s method (13).

Figure 1 shows the time courses of uL-FABP excretion and renal function parameters in both model cats. The uL-FABP concentrations of Cat 1 peaked after the right nephrectomy on day 43 (13 μg/g uCre in Figure 1A) and then decreased below the limit of detection by ELISA. In Cat 2, the uL-FABP increased after the right nephrectomy, peaked on day 64 (208 μg/g uCre in Figure 1B), and decreased gradually in the same manner as in Cat 1. Finally, the uL-FABP concentration in Cat 1 stabilized at < 6 μg/g uCre; it increased gradually in Cat 2 after day 323, peaked on day 456 (263 μg/g uCre), and then decreased to 10.9 μg/g uCre on day 505.

Figure 1.

Time course of urinary liver-type fatty acid-binding protein (uL-FABP) and common renal function markers, including serum creatinine (sCre), blood urea nitrogen (BUN), and symmetric dimethylarginine (SDMA) in feline chronic kidney disease (CKD) models. A, C, E, and G — Cat 1. B, D, F, and H — Cat 2. The left ureter was ligated on day 0. The left ureter was released and anastomosed to the bladder mucosa on day 28. Right nephrectomy was conducted on day 42.

In Cat 1, the conventional renal markers sCre (Figure 1C), BUN (Figure 1E), and SDMA (Figure 1G) peaked on day 46 (sCre, 2.8 mg/dL; BUN, 51.6 mg/dL; SDMA, 28 μg/dL) and then stabilized within their reference ranges (reference ranges: sCre, < 1.6 mg/dL; BUN, < 36 mg/dL; SDMA, < 14 μg/dL). Serum phosphorus (sP), sCa, and UPC remained within the reference ranges throughout the study (reference ranges: sP, 2.2 to 6.5 mg/dL; sCa, 8.0 to 11.1 mg/dL; UPC, < 0.4). Urine specific gravity (USG) remained in the range of 1.015 to 1.066 until the end of the study.

In Cat 2, sCre, BUN, SDMA, and sP increased 2 d after the right nephrectomy and then stabilized until day 456, when they increased again and peaked on day 505 (sCre, 5.7 mg/dL in Figure 1D; BUN, 111.2 mg/dL in Figure 1F; SDMA, 59 μg/dL in Figure 1H; and sP, 12.5 mg/dL). Serum calcium (sCa) remained within the reference range throughout the study, as in Cat 1. Urine protein/creatinine (UPC) remained within the reference range except on days 428 and 456 (0.5 and 0.55, respectively). Urine specific gravity (USG) indicated isosthenuria immediately after the right nephrectomy and then remained in the range of 1.014 to 1.034. After day 456, continuous isosthenuria was observed. Based on the IRIS CKD staging system, Cat 1 was evaluated as Stage I and Cat 2 as Stage IV at the end of the study.

Figure 2 shows representative histopathology with Masson’s trichrome staining in CKD model cats. In the normal right kidney harvested from Cat 2, there were no obvious histopathological abnormalities anywhere (Figure 2A). (In the left obstructed kidneys, mild tubulointerstitial inflammation and fibrosis were initially observed via biopsy on day 28 in both cats (data not shown). At necropsy, renal interstitial lesions and tubular degeneration were worse in Cat 2 than in Cat 1. The final histopathological evaluations were rated as “mild” in Cat 1 (Figure 2B) and “moderate” in Cat 2 (Figure 2C).

Figure 2.

Histopathological images of Masson’s trichrome staining in CKD model cats. A — Right non-obstructed kidney tissues of Cat 2. B — Left obstructed kidney of Cat 1. Mild interstitial fibrosis, interstitial inflammation, and tubular atrophy were observed. C — Left obstructed kidney of Cat 2. Moderate interstitial fibrosis, interstitial inflammation, tubular dilation, and tubular atrophy were observed. Bar = 50 μm.

In this study, Cat 2 had high uL-FABP concentrations and reduced renal function. Cat 1 continued to have low uL-FABP concentrations and maintained renal function. Previously, we reported weak correlations between uL-FABP and both sCre and SDMA in cats (5). This is presumably because the mechanisms underlying elevated uL-FABP excretion differ from those underlying changes in sCre or SDMA. Urinary liver-type fatty acid-binding protein (uL-FABP) was reported as a sensitive biomarker for detecting mild renal damage that would not affect sCre or the estimated GFR in humans (14). Excretion of uL-FABP reflects the tubular response to ischemia and oxidative stress; it increases during the early stages of pathological events, before structural damage to glomerular and tubular cells (6,15). Based on these findings, we propose that uL-FABP may serve as a useful biomarker for monitoring the kidney pathophysiology that results in CKD progression in cats, as in human patients.

In this study, both model cats showed mild renal fibrotic changes in kidney biopsy samples obtained 4 wk after ureteral obstruction (data not shown). Only Cat 2, however, with high uL-FABP levels after the right nephrectomy, had a poor outcome at the end of the study. An increased uL-FABP level in acute kidney injury (AKI) was significantly correlated with the degree of tubulointerstitial fibrosis in sequential CKD in an ischemia–reperfusion mouse model (16).

Fujita et al (14) also reported that sustained elevation of uL-FABP levels for 48 h after coronary angiography could be an indicator of poor renal outcome at 1 y in human patients with ischemic heart disease. Our study showed that the uL-FABP levels in Cat 2 remained higher for more than 48 h, but not in Cat 1. This suggests that the individual difference in renal outcome in our study was partially associated with uL-FABP levels after acute kidney injury (AKI), as in humans. In Cat 2, the uL-FABP values declined over the days before euthanasia, despite the severe elevation in the sCre, BUN, and SDMA levels. Cat 2 exhibited moderate renal interstitial fibrosis (Figure 2C), possibly attributable in part to a reduction in the number of proximal tubular epithelial cells that could secrete uL-FABP.

It is important to establish a model of CKD in cats in order to understand the initiation and progression of interstitial fibrosis, as well as to investigate potential therapeutics. Brown et al (17) reported that unilateral renal ischemia for 90 min resulted in chronic tubulointerstitial fibrosis after 6 mo. In their model, the function of the impaired kidney could not be monitored because no contralateral nephrectomy was carried out.

Unilateral ureteral obstruction (UUO) is a well-described model of CKD in rodents because it exhibits the fundamental pathophysiological mechanisms that typify all forms of CKD relatively quickly (18). Tapmeier et al (8) used that approach to construct the RUUO-N model by re-implanting the obstructed ureter and subsequently removing the contralateral kidney. The advantages of RUUO-N for studying CKD development and progression include the ability to: i) evaluate functional sequels of kidney injury using serum or urine markers; ii) investigate renal pathophysiological changes during/after injury; and iii) conduct long-term studies in living animals with CKD (19,20). Therefore, RUUO-N may be a beneficial model of CKD in cats for the long-term monitoring of renal structures and functions; notably, this model in cats seems to have relatively high variability in renal function loss and histopathologic lesions, unlike in rodents.

We previously showed that the feline kidney seemed to recover relatively quickly after 40 to 50 min of renal ischemia; renal function returned to normal within 14 d (6). In the current study, the period of ureteral obstruction may have been short and could be extended to over 4 wk to increase the stability of the CKD model. Further study should investigate the conditions needed to generate a more stable CKD model in cats.

The severity of CKD differed markedly between the 2 cats, despite the fact that the RUUO-N treatments were similar. Interindividual differences may have been a factor as cats have a more complicated genetic background than laboratory inbred mice. Considering that Cat 2 exhibited higher levels of uL-FABP than Cat 1 after nephrectomy and suffered more tubular damage than Cat 1 according to histopathological evaluations, the sustained elevation of uL-FABP after nephrectomy in Cat 2 may explain the poor renal outcome by ischemia and oxidative stress, similar to that in mice and humans (4,7).

We determined that RUUO-N in cats mainly induced tubulointerstitial injury rather than glomerular injury. A feline model of CKD induced by another condition, such as glomerular disease, may have a different time course of uL-FABP secretion.

In conclusion, uL-FABP excretion increased before the excretion of such common renal biomarkers as sCre, BUN, and SDMA in a feline model of CKD with deteriorating renal function. Our study showed that uL-FABP is a potential biomarker for the early detection of pathophysiological events that result in CKD progression in cats. A larger prospective study is recommended in order to explore the impact of uL-FABP on CKD progression in cats.