N-acetyl-β-D-glucosaminidase is a high–molecular-weight (150-kDa) lysosomal enzyme that may be detected in many mammalian tissues, serum, and urine.1,2 In the kidney, the highest activity of NAG is found in the epithelial cells of the proximal convoluted tubule, where glycoprotein reabsorption and degradation occur.3 There are 2 isoenzymes of NAG that differ in their tubular distribution, subunit composition, pI, and thermostability; NAG A is located in the soluble intralysosomal compartment, and NAG B is bound to the lysosomal membrane.4 In humans with normal renal function, low NAG activity can be detected in urine as a result of exocytosis. This is predominantly the NAG A isoenzyme,5 as has been similarly reported for cats.6 When direct damage to proximal tubular cells occurs, both NAG B activity and total NAG activity measured in the urine increase.7,8 However, a number of factors can influence total NAG activity, including urinary pH (particularly alkaline pH > 8.0),9 the conditions by which the samples are stored,10,11 and the method used for assessment of NAG activity.12 In human medicine, the NAG index, defined as the ratio of NAG concentration to urine creatinine concentration, is used as an indicator of tubular damage.13 There is also evidence from experimental studies14 that increased NAG activity may reflect increased lysosomal turnover and protein presentation to the proximal tubules rather than direct tubular damage.
Chronic kidney disease is a common condition affecting geriatric cats that is currently diagnosed on the basis of a consistent history and clinical signs, persistent azotemia, and an inability to appropriately concentrate urine.15 Survival rates for cats with CKD have been significantly associated with the severity of both azotemia and proteinuria.16 However, it has commonly been stated that up to 75% of nephrons must no longer function before azotemia will be detected. In an experimental study17 in which cats underwent subtotal nephrectomy with removal of 50% to 75% of the kidney, urine concentrating ability was maintained and azotemia was mild. Limited information is available on the pathological changes that occur early in the course of CKD of cats. However, findings from studies18,19 that have evaluated affected cats after subtotal nephrectomy and postmortem examination have revealed that tubulointerstitial nephritis is the predominant feature. Few studies6,20 have examined NAG activity in feline urine. Recently, the NAG index was investigated for use as a biomarker to assess the development of azotemia in cats with hyperthyroidism that were receiving treatment with methimazole, versus untreated cats.21 The NAG index did not differentiate the 2 groups in the study21; however, the small number of cats enrolled may have affected the results. We suggest that measurement of urine NAG activity may represent a useful marker for early tubular damage in cats with CKD and may indicate active tubular damage in cats at risk for progressive CKD. The objectives of the study reported here were to present preliminary validation data for a commercially available assay for the quantification of urinary NAG activity in cats and to perform a cross-sectional analysis comparing NAG index in geriatric (≥ 9 years old) cats with variable plasma creatinine concentration and with proteinuria.
Materials and Methods
Case selection—Cats were recruited from 2 clinics for geriatric (≥ 9 years old) cats located in central London (Beaumont Animals' Hospital and the People's Dispensary for Sick Animals). Cats were enrolled prospectively between January 2001 and December 2006 in an ongoing geriatric health-monitoring program. To be eligible for inclusion, cats were ≥ 9 years of age and either had a current diagnosis of azotemic CKD or were apparently healthy on the basis of history, physical examination, and plasma biochemical testing. Cats with systemic disease other than CKD, those receiving long-term medication other than parasiticides, and cats with systemic hypertension were excluded from the study. The Ethics and Welfare Committee of the Royal Veterinary College and the WALTHAM Ethics Committee approved the study protocol. Collection and storage of blood and urine samples were performed with the informed consent of the cats' owners.
Upon study enrollment, serum biochemical analysis and urinalysis were performed concurrently for all cats, and on the basis of these results, cats were classified either as healthy nonazotemic (plasma creatinine concentration " 2.0 mg/dL; laboratory reference range 0.22 to 2.0 mg/dL) or as potentially having a diagnosis of CKD. Cats suspected of having CKD were reexamined in 4 to 6 weeks at which time a further serum biochemical analysis and urinalysis were performed. The diagnosis of CKD was made on the basis of a persistently high plasma creatinine concentration (> 2.0 mg/dL), in association with inappropriate urine concentrating ability. Healthy nonazotemic cats were reexamined every 6 months, whereas cats with CKD were offered reexamination appointments every 8 weeks. Repeat serum biochemical testing and, where possible, urinalysis were performed at every visit for the nonazotemic cats and every other visit for cats with a diagnosis of CKD.
At each consultation, a history was obtained, a full physical examination was performed, and SBP was measured by use of Doppler ultrasonogrphy.a An indirect fundic examination was performed in cats with an SBP > 160 mm Hg.22 Systemic hypertension was diagnosed when the mean of 5 sequential measurements of SBP was > 170 mm Hg on at least 2 visits 1 to 2 weeks apart or on a single visit if hypertension occurred with hypertensive target organ damage, most commonly a hypertension-induced retinopathy, choroidopathy, or both. Cats with a diagnosis of systemic hypertension were routinely treated with amlodipine besylate.b
Samples for assessment of NAG activity were collected from healthy nonazotemic cats at the first examination and at a follow-up visit for cats with a diagnosis of CKD. The staging scheme for plasma creatinine concentration advocated by the International Renal Interest Society was applied to all cats (stage I, < 1.6 mg/dL; stage II, 1.6 to 2.8 mg/dL; stage III, 2.9 to 5.0 mg/dL; stage IV, > 5.0 mg/dL).23 Stage II was further subdivided into stage IIa (1.6 to 2.0 mg/dL) and IIb (2.1 to 2.8 mg/dL).
Cats with a previous diagnosis of CKD were offered a low-phosphorus, low-protein dietc that was provided free of charge. Compliance with the diet was variable. If indicated on the basis of the results of serum biochemical analysis, cats with CKD were also offered a potassium supplementd and additional phosphate binders.e Owners were asked to withhold food from cats for 8 hours prior to all hospital visits.
Sample collection—Blood samples were obtained by jugular venipuncture and were collected into lithium heparin-containing tubes. When the bladder was palpable, a urine sample was obtained by cystocentesis as a routine screen for bacterial urinary tract infections. Urine pHf was assessed immediately after cystocentesis. Plasma and urine samples were stored on ice for < 6 hours prior to analysis. Urinalysis included measurement of specific gravity, dipstick analysis,g and microscopic examination of the sediment. Samples were excluded if there was evidence of pyuria, gross hematuria, or bacteriuria. Remaining plasma and urine samples were centrifugedh at 1,000 X g for 10 minutes at 4°C, placed in aliquots, and stored at −80°C. An aliquot of the stored sample was transported to an external commercial laboratoryi where a colorimetric pyrogallol red method was used to determine urine protein concentration, and a total colorimetric picric acid method was used to determine urine creatinine concentration. The UP:C was then calculated by use of these 2 values. Heparinized plasma was used for biochemical analysis,i and total thyroxine concentration was evaluated in all nonazotemic cats and in any cat with a history (eg, polyphagia or weight loss), clinical examination findings (eg, palpable goiter, tachycardia, or low body condition score), or results of biochemical analysis (eg, high serum alanine aminotransferase activity or serum alkaline phosphatase activity) that could be consistent with a diagnosis of hyperthyroidism. Cats with a diagnosis of hyperthyroidism (total thyroxine > 55 nmol/L) were excluded from the study.
Measurement and validation of NAG—Urinary NAG activity was measured by use of a commercially available colorimetric assay.j This is a nonautomated technique that uses 3-cresolsulfonphthaleinyl-N-acetyl-β-glucosaminide as a substrate and is hydrolyzed by NAG to produce 3-cresol-sulfonphthalein (3-cresol purple) and N-acetyl-glucosamine. The concentration of 3-cresol purple produced was measured photometricallyk at a wavelength of 580 nm. Unless otherwise stated, the assay was run in accordance with the manufacturer's instructions for the microassay by use of half volumes (ie, 25-μL sample) and measured against a blank standard with an incubation period of 45 minutes at 37°C. The NAG activity was reported as U/L, and NAG activity was subsequently expressed by use of the NAG index, which is the ratio of urinary NAG activity to grams of urine creatinine.
Urine samples from cats with variable azotemia that had been stored at −80°C for < 1 year but that had not previously been subjected to a freeze-thaw cycle were used for assay validation. During validation, the length of incubation of sample and substrate were adjusted according to the activity of NAG in the test samples. To evaluate the linearity of NAG activity and to determine the optimum length of incubation, NAG activity was measured after incubation of both sample and substrate at 37°C, at 15-minute intervals from 0 to 60 minutes. The shortest incubation time, where NAG activity in the lowest feline standard was substantially greater than NAG activity in the blank (substrate), was chosen as the incubation time for future NAG activity assessment. The precision and repeatability of the assay was assessed by evaluating the inter- and intra-assay variability for urine samples with low, medium, and high NAG activity. Urine samples that were used for calculation of the inter- and intra-assay CV had been stored at −80°C for > 1 year.
The effect of urine storage on NAG activity was assessed. Fresh urine samples that had not previously been frozen and that had been stored at 4°C for a maximum of 6 hours were selected. The NAG activity was measured in a fresh sample and after 5 rapid freeze-thaw cycles consisting of a period of 10 minutes on dry ice followed by a further 10 minutes at room temperature (approx 21°C). The NAG activity was also evaluated in fresh urine samples that had been stored at 4°C for a maximum of 6 hours and after storage at −20°C for 5 weeks.
Cross-sectional analysis of NAG index—A cross-sectional analysis was performed to evaluate application of the NAG index to cats with variable plasma creatinine concentration and proteinuria. Cats were grouped according to their plasma creatinine concentration (stage I, < 1.6 mg/dL; stage IIa, 1.6 to 2.0 mg/dL; stage IIb, 2.1 to 2.8 mg/dL; stage III, 2.9 to 5.0 mg/dL; and stage IV, > 5.0 mg/dL) and also according to the magnitude of proteinuria by use of the UP:C (nonproteinuric, < 0.2; borderline proteinuria, 0.2 to 0.4; and proteinuria, > 0.4). These groupings are in accordance with the staging schemes advocated by the International Renal Interest Society but include cats that did not have a diagnosis of CKD.23 Stored urine samples were used for the retrospective assessment of NAG index. Urine samples were selected on the basis of whether they either had never been thawed or had been through only 1 freeze-thaw cycle.
Statistical analysis—All statistical analyses were performed by use of commercially available software packages.l,m For all analyses, values of P < 0.05 were considered significant. Unless otherwise stated, data are presented as median (25th and 75th percentile) values. Normality of the distribution of the data was assessed by use of the Kolmogarov-Smirnov test, and if indicated, UP:C, NAG activity, and NAG index data were logarithmically transformed for parametric statistical analysis. Precision and repeatability of the NAG assay were assessed by calculation of the inter- and intra-assay CVs. The effect of a series of repeat freeze-thaw cycles on NAG activity was assessed by use of a nonparametric Friedman test, and a Dunnett multiple comparison post hoc analysis was performed if indicated. The NAG activity in fresh urine and urine that had been stored at −20°C for 5 weeks was compared by use of a Wilcoxon signed rank test, and an independent Student t test was used to assess log NAG index for neutered male versus neutered female cats that were not azotemic. A 1-way ANOVA was applied to compare log NAG index and clinical variables for all cats grouped according to plasma creatinine concentration and proteinuria, and where necessary, a Bonferroni posttest was applied. The relationship between log NAG index, log UP:C, and plasma creatinine concentration, and also between urinary pH and log NAG activity, was assessed by use of the Pearson correlation.
Results
Measurement and validation of NAG assay—Linearity of NAG activity for 10 samples with incubation times of up to 60 minutes was demonstrated (Figure 1). Intra-assay CVs (n = 5) in urine samples with mean NAG activity of 0.84, 4.27, and 7.43 U/L were 12.4%, 20.4%, and 9.9%, respectively. Interassay CVs (n = 5) for samples with mean NAG activity of 0.64, 4.38, and 8.48 U/L were 25.9%, 14.4%, and 25.1%, respectively. The NAG activity was assessed in 7 fresh urine samples (median, 5.37 U/L; range, 0.37 to 16.43 U/L) and after a series of 5 freeze-thaw cycles. From the first to fifth freeze-thaw cycle, 94.0%, 99.7%, 82.2%, 83.0%, and 67.0%, respectively, of NAG activity that was originally detected in the fresh urine remained. A significant (P = 0.017) difference in NAG activity was detected. However, by use of the Dunnett post hoc analysis, this difference was significant (P < 0.05) between only the fresh sample and the sample after the fifth freeze-thaw cycle (median, 2.47 U/L; range, 0.07 to 8.83 U/L). Ten fresh urine samples with a median NAG activity of 2.62 U/L (range, 0.62 to 3.74) were reevaluated after 5 weeks of storage at −20°C and had a median NAG activity of 1.78 U/L (range, 0.46 to 3.53 [78.0%]); this decrease in NAG activity was not significant (P = 0.153).
Linearity of NAG activity in urine samples from 10 cats for a variable incubation period. Optical density was measured at a wavelength of 580 nm.
Citation: American Journal of Veterinary Research 71, 2; 10.2460/ajvr.71.2.241
Linearity of NAG activity in urine samples from 10 cats for a variable incubation period. Optical density was measured at a wavelength of 580 nm.
Citation: American Journal of Veterinary Research 71, 2; 10.2460/ajvr.71.2.241
Linearity of NAG activity in urine samples from 10 cats for a variable incubation period. Optical density was measured at a wavelength of 580 nm.
Citation: American Journal of Veterinary Research 71, 2; 10.2460/ajvr.71.2.241
Cross-sectional analysis of NAG index—The NAG index was assessed in 197 cats (78 neutered females and 119 neutered males) that had variable plasma creatinine concentrations with or without proteinuria. This group included 94 nonazotemic healthy cats (43 neutered females and 51 neutered males) with a median NAG index of 1.06 (range, 0.44 to 1.91). There was no significant (P = 0.425) difference when log NAG index in these nonazotemic cats was compared by sex. Median urinary pH for 148 cats was 6.24 (range, 5.83 to 6.67). A significant (P = 0.045) but weak negative correlation was found between urinary pH and log NAG activity (r2 = 0.027). Clinical data for cats were divided according to the plasma creatinine concentration (Table 1). No significant (P = 0.06) difference was found for NAG index when cats were grouped according to plasma creatinine concentration, and although significant (P = 0.028), the correlation between log NAG index and plasma creatinine concentration was poor (r2 = 0.025; Figure 2). However, a significant (P < 0.001) difference was found when cats were grouped according to the magnitude of proteinuria (Figure 3). A significant (P < 0.001) positive correlation was found between log UP:C and log NAG index (r2 = 0.249).
Clinical data for 197 cats in the cross-sectional analysis grouped according to plasma creatinine concentration.
| Variable | Plasma creatinine concentration (mg/dL) | ||||
|---|---|---|---|---|---|
| < 1.6 | 1.6–2.0 | 2.1–2.8 | 2.9–5.0 | > 5.0 | |
| No. of cats | 50 | 54 | 44 | 38 | 11 |
| Creatinine (mg/dL) | 1.37 (1.26, 1.50) | 1.78 (1.67, 1.87) | 2.44 (2.22, 2.68) | 3.23 (2.99, 3.88) | 5.50 (5.36, 6.11) |
| NAG activity (U/L) | 2.00 (1.05,3.57) | 1.93 (0.91, 3.53) | 1.90 (0.73, 4.06) | 1.58 (0.51, 3.56) | 1.50 (0.67, 2.60) |
| NAG index | 1.11 (0.51, 1.73) | 1.09 (0.38, 3.78) | 2.56 (0.76, 3.92) | 1.63 (0.43, 4.55) | 3.23 (1.00, 3.57) |
| SBP (mm Hg)* | 133.4 (120.6, 147.8) | 145.6 (134.2, 172.0) | 144.8 (128.6, 152.0) | 148.8 (135.2, 158.8) | 135.0 (125.2, 144.8) |
| USG | 1.051a (1.036, 1.053) | 1.037b (1.022, 1.053) | 1.020c (1.016, 1.028) | 1.018c (1.015, 1.020) | 1.012c (1.012, 1.014) |
| UP:C | 0.16c (0.13, 0.20) | 0.18c (0.12, 0.33) | 0.17c (0.10, 0.34) | 0.13c (0.09, 0.23) | 0.62b (0.47, 0.67) |
Data are reported as median values (25th, 75th percentile).
SBP measurements were not available for 1 cat with stage IIa and 1 cat with stage III azotemia.
USG = Urine specific gravity.
Values with different superscripts within a row were significantly different (P < 0.05) from each other.
Scatterplot of the correlation between log urinary NAG index and plasma creatinine concentration for 197 cats (Pearson correlation coefficient, r2 = 0.025; P = 0.028). Notice that y-axis data are logarithmically transformed.
Citation: American Journal of Veterinary Research 71, 2; 10.2460/ajvr.71.2.241
Scatterplot of the correlation between log urinary NAG index and plasma creatinine concentration for 197 cats (Pearson correlation coefficient, r2 = 0.025; P = 0.028). Notice that y-axis data are logarithmically transformed.
Citation: American Journal of Veterinary Research 71, 2; 10.2460/ajvr.71.2.241
Scatterplot of the correlation between log urinary NAG index and plasma creatinine concentration for 197 cats (Pearson correlation coefficient, r2 = 0.025; P = 0.028). Notice that y-axis data are logarithmically transformed.
Citation: American Journal of Veterinary Research 71, 2; 10.2460/ajvr.71.2.241
Scattergram of the urinary NAG index for 197 cats categorized according to the degree of proteinuria. Notice that y-axis data are logarithmically transformed. Horizontal bars represent the median NAG index for each group. *Denotes a significant (P < 0.05) difference in log NAG index between groups determined by use of a 1-way ANOVA and Bonferroni posttest procedure. Nonproteinuria = UP:C < 0.2 (n = 116); borderline proteinuria = UP:C 0.2 to 0.4 (45); proteinuria = UP:C > 0.4 (36).
Citation: American Journal of Veterinary Research 71, 2; 10.2460/ajvr.71.2.241
Scattergram of the urinary NAG index for 197 cats categorized according to the degree of proteinuria. Notice that y-axis data are logarithmically transformed. Horizontal bars represent the median NAG index for each group. *Denotes a significant (P < 0.05) difference in log NAG index between groups determined by use of a 1-way ANOVA and Bonferroni posttest procedure. Nonproteinuria = UP:C < 0.2 (n = 116); borderline proteinuria = UP:C 0.2 to 0.4 (45); proteinuria = UP:C > 0.4 (36).
Citation: American Journal of Veterinary Research 71, 2; 10.2460/ajvr.71.2.241
Scattergram of the urinary NAG index for 197 cats categorized according to the degree of proteinuria. Notice that y-axis data are logarithmically transformed. Horizontal bars represent the median NAG index for each group. *Denotes a significant (P < 0.05) difference in log NAG index between groups determined by use of a 1-way ANOVA and Bonferroni posttest procedure. Nonproteinuria = UP:C < 0.2 (n = 116); borderline proteinuria = UP:C 0.2 to 0.4 (45); proteinuria = UP:C > 0.4 (36).
Citation: American Journal of Veterinary Research 71, 2; 10.2460/ajvr.71.2.241
Discussion
Urine NAG activity was rapidly quantifiable by use of a nonautomated colorimetric technique that uses 3-cresolsulfonphthaleinyl-N-acetyl-β-glucosaminide as a substrate. This technique has previously been used to evaluate NAG activity in urine from young healthy cats and to demonstrate that there was no apparent circadian variation in urinary NAG excretion.20 However, no validation of the technique was reported. Analysis of NAG by use of 3-cresolsulfonphthaleinyl-N-acetyl-β-glucosaminide was first reported by Noto et al24 in 1983, with the proposed advantages that the substrate was highly water soluble, that spectral overlap of interfering substances such as hemoglobin or bilirubin was negligible at 580 nm, that no prior preparation or filtration of samples was required,25 and that results correlated well with fluorometric assessment of NAG activity by use of 4-methylumbelliferyl N-acetyl-β-D-glucosaminide.12,26 In the present study, the high interassay CVs (25.9%, 14.4%, and 25.1% for mean NAG activity of 0.64, 4.38, and 8.48 U/L respectively) were of concern. In previous a report24, interassay CVs ranged from 5% to 11% when the same nonautomated technique was applied, and other clinical veterinary studies20,21 have subsequently relied on this technique. However, it should also be mentioned that the amount of NAG activity in the samples used by Noto et al24 was considerably higher than the NAG activity found in cat urine in the study presented here. In addition, it was reported that between 8 and 17 repeats were performed to achieve the substantially lower CVs in the original report.24
Interassay samples in the present study were divided into aliquots prior to analysis, to prevent variation in the number of freeze-thaw cycles, and were evaluated over a 1-week period, such that the difference in length of storage time at −20°C was minimal. Other factors that have been suggested to influence NAG activity include alkaline pH and inhibitors of NAG activity present in urine, such as urea, an effect that may be species dependent.12,27 In fact, it has been shown that NAG A activity may be more affected by pH than NAG B activity.9,27 In human urine, the optimal pH for assessment of NAG activity by use of the current assay is a pH of 4.75 to 5.25.24 However, as found in the present study, pH of urine collected from geriatric cats is rarely alkaline, and the correlation between log NAG activity and urinary pH, although significant, was poor. Therefore, these factors seem unlikely to have contributed to the suboptimal interassay CVs.
Results of studies that have evaluated the effect of storage on NAG activity have been controversial. A study by Loeb et al28 suggested that approximately 15% of NAG activity is lost with an initial freeze-thaw cycle at −20°C but that, thereafter, the length of time and the temperature at which samples are frozen have minimal effect on NAG activity. Others have reported that NAG activity was well preserved at −20° and −70°C for 1 month.11 A study11 of longer duration revealed that NAG activity was preserved (84.1% of basal activity) after storage at −70°C for 1 year, but that there was a significant decrease in NAG activity after storage for 1 year at −20°C. The results of our study suggest that NAG activity is maintained at > 90% of the activity detected in fresh urine for up to 2 freeze-thaw cycles. However, although there was no significant reduction in NAG activity after storage for 5 weeks at −20°C, there was reduced activity. Future studies should therefore consider the duration of storage of samples prior to analysis, and we would suggest a maximum of 2 freeze-thaw cycles when analysis of urine samples for NAG activity is planned.
The NAG index in healthy geriatric cats in the study reported here was lower than has previously been reported by Uechi et al20 in healthy young cats, in a report that applied the same method for NAG activity quantification (n = 12; mean ± SD, 6.4 ± 2.0 U/g). Our results were comparable to those for 35 healthy young control cats in the study performed by Sato et al6 that applied the substrate p-nitrophenyl N-acetyl-β-D-glucosaminide; our results were also comparable to those of a more recent study21 that used 3-cresolsulfonphthaleinyl-N-acetyl-β-glucosaminide to evaluate the role of NAG index as a predictor of azotemia in cats with hyperthyroidism. It is also of note that even in a healthy population of nonazotemic geriatric cats, the presence of early kidney disease cannot be excluded and that, as such, a reduced number of functioning nephrons in the aging kidney could influence the level of NAG activity detected.
Urine NAG activity has been evaluated for use as a marker of renal tubular damage in a number of disease states in human patients and as an indicator of tubular nephrotoxicosis29 and heavy metal poisoning.30 Urine NAG activity has also been investigated as a predictor of early renal tubular damage in both insulin-dependent and non–insulin dependent diabetes mellitus31,32 and in human patients with systemic hypertension.33 In dogs, NAG activity has been used as a marker of tubular damage in patients with CKD resulting from various causes, in dogs with pyometra34 or leishmaniasis,35 and in experimental studies36 evaluating nephrotoxicosis associated with gentamicin administration. More recently, measurement of NAG activity has been suggested as a method of monitoring damage caused by repeated application of extracorporeal shock wave lithotripsy in dogs.37
In human patients with glomerulonephritis, the NAG index is not associated with GFR or with serum creatinine concentration.38 Similarly, in the study reported here, there was no significant association between high plasma creatinine concentration and an increase in the NAG index. In fact, when the statistical analyses were repeated including only cats with a diagnosis of CKD, there was no correlation between log NAG index and plasma creatinine concentration, suggesting that NAG index should not be considered for use as a marker of CKD.
The NAG index was more closely associated with proteinuria than plasma creatinine concentration. In human patients, proteinuria has been implicated as a factor involved in the progression of kidney disease.39 Study40 results revealed a high capacity for tubular protein reabsorption via megalin and cubulin receptor–mediated uptake, and it is suggested that such proteins are broken down and processed in the lysosomes of the proximal tubules. Findings of an experimental study41 indicate that increased presentation of protein to proximal tubular cells results in the basolateral release of various proinflammatory and profibrogenic cytokines that can potentiate tubulointerstitial nephritis and result in further loss of nephrons. In addition, results of experimental studies42,43 of rats reveal that after an initial renal insult, adaptive mechanisms result in the hypertrophy of the remaining functioning nephrons to maintain an effective GFR. It is therefore reasonable to speculate that, in such conditions, NAG activity within the lysosomes is likely to be upregulated and that, in conjunction with increased protein processing, this may result in increased urinary NAG activity.
In the present study, there was a significant difference in log NAG index when cats were grouped according to their magnitude of proteinuria. However, there was considerable overlap between groups, suggesting that on the basis of current evidence, interpretation of NAG index in an individual cat in a clinical situation is of questionable benefit.
In an experimental study, Bosomworth et al14 induced minimal change nephrotic syndrome and marked proteinuria in rats, demonstrating that peak urine NAG activity and a change in NAG isoenzyme pattern occurred at the same time that peak proteinuria was detected. Further, despite histologically normal tubules, urine protein concentration was significantly positively correlated with NAG activity. In human patients with multiple causes of kidney disease, a significant positive correlation was found between 24-hour NAG activity and 24-hour proteinuria.1 However, there was no significant difference in NAG activity between patients with renal disease and those with end-stage renal failure that were requiring hemodialysis.1 When NAG index was compared for healthy adults versus patients with insulin-dependent diabetes mellitus with and without diabetic nephropathy, a significant difference was found between all 3 groups. This was despite there being no significant difference in plasma creatinine concentration between groups.32 It can therefore be hypothesized that the correlation between proteinuria and NAG index in cats implies increased activity of lysosomes as a result of protein processing rather than direct proximal tubular cell damage.
There are a number of limitations to the present study, including the cross-sectional design and the use of spot urine samples for the assessment of NAG activity. Although 1 study44 indicates that spot urine samples can be used as a replacement for the assessment of 24-hour total NAG activity, other studies44,45 have shown that this may be inaccurate, particularly in human patients with preexisting kidney disease. In the present study, NAG was assessed in a relatively large number of cats, but approximately 50% were not azotemic. A more equal distribution across a range of plasma creatinine concentrations and degrees of proteinuria would have been optimal. In addition, NAG activity was assessed in the clinical setting of our study by use of urine samples that had been stored at −80°C for variable lengths of time. However, the samples that were chosen either had never been thawed or had been through 1 freeze-thaw cycle.
In the present study, thyroxine concentration was not evaluated in all cats with azotemia. It has previously been documented46 that renal disease may result in suppression of thyroxine concentration therefore substantially complicating the diagnosis of hyperthyroidism. A study47 of humans suggests that urinary NAG index is significantly higher in patients with hyperthyroidism prior to treatment, than in healthy control individuals. It is therefore possible that, in a small number of cats, the diagnosis of hyperthyroidism may have been missed and that this may have influenced NAG activity. A further limitation of our study was that there was no control of dietary protein intake. Cats with a diagnosis of CKD were routinely offered a low-protein, lowphosphorus renal diet, but compliance was variable. There is conflicting information from studies48,49 of cats that have undergone experimental nephrectomy, and there is a paucity of information from clinical studies50,51 regarding the influence of dietary protein intake on the magnitude of proteinuria. However, it is possible that variation in dietary protein intake in cats with CKD in the present study may have influenced the magnitude of proteinuria and subsequently NAG activity.
In future studies, the evaluation of NAG isoenzyme activities and calculation of the ratio of NAG A to NAG B activity could provide additional information that may differentiate increased activity from primary cellular damage. Future longitudinal assessment of NAG index may also give some indication of its role either as a predictor of early kidney damage prior to the development of azotemia or as a marker for active renal tubular damage and the risk for disease progression in cats with a previous diagnosis of CKD.
ABBREVIATIONS
| CKD | Chronic kidney disease |
| CV | Coefficient of variance |
| GFR | Glomerular filtration rate |
| NAG | N-acetyl-β-D-glucosaminidase |
| pI | Isoelectric point |
| SBP | Systolic blood pressure |
| UP:C | Urine protein-to-creatinine ratio |
Parks Electronic Doppler, model 811B, Perimed UK, Bury St Edmunds, Suffolk, England.
Amlodipine, Istin, Pfizer, Sandwich, Kent, England.
Feline low-phosphorus, low-protein diet, WALTHAM Pet Nutrition, Melton Mowbray, Leicestershire, England.
Tumil-K, Arnold's Veterinary Products, Shrewsbury, Shropshire, England.
Alu-cap aluminium hydroxide, 3M Health Care Ltd, Loughborough, Leicestershire, England.
HI 9224 pH meter, Hanna Instruments, Leighton Buzzard, Bedfordshire, England.
Multistix urine chemistry reagent strips, Bayer Diagnostics, Newbury, Berkshire, England.
Mistral 3000, Sanyo-Gallenkamp, Loughborough, Leicestershire, England.
Idexx Laboratories, Wetherby, West Yorkshire, England.
N-acetyl-β-glucosaminidase kit, catalog No. 875406, Roche, Basel, Switzerland.
Spectronic Genesys 2, Patterson Scientific, Luton, Bedfordshire, England.
SPSS, version 15.0 for Windows, SPSS Inc, San Diego, Calif.
GraphPad Prism, version 5.0 for Windows, GraphPad Software, San Diego, Calif.
References
- 1.↑
Costigan MG, Rustom R, Bone JM, et al. Origin and significance of urinary N-acetyl-β,-D-glucosaminidase (NAG) in renal patients with proteinuria. Clin Chim Acta 1996;255:133–144.
- 2.
Ikonne JU, Ellios RB. N-acetyl-β-D-hexosaminidase component A. Different forms in human tissues and fluids. Biochem J 1973;135:457–462.
- 3.↑
Bourbouze R, Baumann F-C, Bonvalet J-P, et al. Distribution of N-acetyl-β-D-glucosaminidase isoenzymes along the rabbit nephron. Kidney Int 1984;25:636–642.
- 4.↑
Robinson D, Stirling S. N-acetyl-β-D-glucosaminidase in human spleen. Biochem J 1968;107:321–327.
- 5.↑
Paigen K, Peterson J. Coordinancy of lysosomal enzyme excretion in human urine. J Clin Invest 1978;61:751–762.
- 6.↑
Sato R, Soeta S, Syuto B, et al. Urinary excretion of N-acetyl-B-D-glucosaminidase and its isoenzymes in cats with urinary disease. J Vet Med Sci 2002;64:367–371.
- 7.
Ellis BG, Tucker SM, Thompson AE. Presence of serum and tissue forms of N-acetyl-β-D-glucosaminidase in urine from patients with renal disease. Clin Chim Acta 1975;64:195–202.
- 8.
Tucker SM, Pierce RJ, Price RG. Characterization of human N-acetyl-β-D-glucosaminidase isoenzymes as an indicator of tissue damage in disease. Clin Chim Acta 1980;102:29–40.
- 9.↑
Mandic LM, Acimovic JM, Jovanovic VB. The possibility of determining N-acetyl-β-glucosaminidase isoenzymes under alkaline conditions. Clin Biochem 2005;38:384–389.
- 10.
Matteucci E, Pellegrini L, Uncini-Manganelli C, et al. More on effects of storage time and temperature on urinary enzymes: a 1-year study. Enzyme 1992;46:249–251.
- 11.↑
Matteucci E, Gregori G, Pellegrini L, et al. How can storage time and temperature affect enzymatic activities in urine? Enzyme 1991;45:116–120.
- 12.↑
Jung K, Priem F, Klotzek S, et al. Methods compared for determining activity of N-acetyl-β-D-glucosaminidase in urine without pretreatment of sample: different sensitivity and species effect. Enzyme 1991;45:215–221.
- 13.↑
Herget-Rosenthal S, Poppen D, Hüsing J, et al. Prognostic value of tubular proteinuria and enzymuria in nonoliguric acute tubular necrosis. Clin Chem 2004;50:552–558.
- 14.↑
Bosomworth MP, Aparicio S, Hay A, et al. Urine N-acetyl-β-D-glucosaminidase—a marker of tubular damage? Nephrol Dial Transplant 1999;14:620–626.
- 15.↑
Lulich JP, Osborne CA, O'Brien TD, et al. Feline renal failure: questions, answers, questions. Compend Contin Educ Pract Vet 1992;14:127–152.
- 16.↑
Syme HM, Markwell PJ, Pfeiffer D, et al. Survival of cats with naturally occurring chronic renal failure is related to severity of proteinuria. J Vet Intern Med 2006;20:528–535.
- 17.↑
Ross LA, Finco DR. Relationship of selected clinical renal function tests to glomerular filtration rate and renal blood flow in cats. Am J Vet Res 1981;42:1704–1711.
- 18.
DiBartola SP, Broome MR, Stein BS, et al. Effect of treatment of hyperthyroidism on renal function in cats. J Am Vet Med Assoc 1996;208:875–878.
- 19.
Mathur S, Brown CA, Dietrich UM, et al. Evaluation of a technique of inducing hypertensive renal insufficiency in cats. Am J Vet Res 2004;65:1006–1013.
- 20.↑
Uechi M, Uechi H, Nakayama T, et al. The circadian variation of urinary N-acetyl-β-D-glucosaminidase and γ-glutamyl-transpeptidase in clinically healthy cats. J Vet Med Sci 1998;60:1033–1034.
- 21.↑
Lapointe C, Belanger MC, Dunn M, et al. N-acetyl-β-D-glucosaminidase index as an early biomarker for chronic kidney disease in cats with hyperthyroidism. J Vet Intern Med 2008;22:1103–1110.
- 22.↑
Syme HM, Barber PJ, Markwell PJ, et al. Prevalence of systolic hypertension in cats with chronic renal failure at initial evaluation. J Am Vet Med Assoc 2002;220:1799–1804.
- 23.↑
Elliott J. Staging chronic kidney disease. In: Elliott J, Grauer GF, eds. BSAVA manual of canine and feline nephrology and urology. 2nd ed. Haryana, India: Replika Press Pvt Ltd, 2007;163.
- 24.↑
Noto A, Ogawa Y, Mori S, et al. Simple, rapid, spectrophotometry of urinary N-acetyl-β-D-glucosaminidase, with use of a new chromogenic substrate. Clin Chem 1983;29:1713–1716.
- 25.↑
Konings CH, Mulder C. Effect of gel filtration on urinary N-acetyl-β-D-glucosaminidase activity measured by the m-cresolsulfonphthaleinyl procedure (lett). Clin Chem 1990;36:402–403.
- 26.
Drake PL, Krieg E, Teass AW, et al. Two assays for urinary N-acetyl-β-D-glucosaminidase compared. Clin Chem 2002;48:1604–1605.
- 27.
Morita A, Numata Y, Kosugi Y, et al. Stabilities of N-acetyl-β-D-glucosaminidase (NAG) isoenzymes in urine: advantages of NAG isoenzyme B measurement in clinical applications. Clin Chim Acta 1998;278:35–43.
- 28.↑
Loeb WF, Das SR, Trout JR. The effect of erythritol on the stability of gamma-glutamyl transpeptidase and N-acetyl glucosaminidase in human urine. Toxicol Pathol 1997;25:264–267.
- 29.↑
Wiland P, Szechinski J. Proximal tubule damage in patients treated with gentamicin or amikacin. Pol J Pharmacol 2003;2003:631–637.
- 30.↑
Mandic L, Radmila M, Jelena A, et al. Change in the iso-enzyme profiles of urinary N-acetyl-beta-D-glucosaminidase in workers exposed to mercury. Toxicol Ind Health 2002;18:207–214.
- 31.
Hong CY, Chia KS, Ling SL. Urinary protein excretion in type 2 diabetes with complications. J Diabetes Complications 2000;14:259–265.
- 32.↑
Jung K, Pergande M, Schinrke E, et al. Urinary enzymes and low molecular mass proteins as indicators of diabetic nephropathy. Clin Chem 1988;34:544–547.
- 33.↑
Mueller PW, Hall WD, Caudill SP, et al. An in-depth examination of the excretion of albumin and other sensitive markers of renal damage in mild hypertension. Am J Hypertens 1995;8:1072–1082.
- 34.↑
Heiene R, Moe L, Molmen G. Calculation of urinary enzyme excretion, with renal structure and function in dogs with pyometra. Res Vet Sci 2001;70:129–137.
- 35.↑
Palacio J, Liste F, Gascon M. Enzymuria as an index of renal damage in canine leishmaniasis. Vet Rec 1997;140:477–480.
- 36.↑
Grauer GF, Greco DS, Behrend EN, et al. Estimation of quantitative enzymuria in dogs with gentamicin induced nephrotoxicosis using enzyme/creatinine ratios from spot urine samples. J Vet Intern Med 1995;9:324–327.
- 37.↑
Fortes MAQR, Andriolo A, Ortiz V, et al. Effect of shock wave reapplication on urinary N-acetyl-beta-glucosamindase in canine kidney. Int Braz J Urol 2004;30:148–154.
- 38.↑
Holdt-Lehmann B, Lehmann A, Korten G, et al. Diagnostic value of urinary alanine aminopeptidase and N-acetyl-β-D-glucosaminidase in comparison to α 1-microglobulin as a marker in evaluating tubular dysfunction in glomerulonephritis patients. Clin Chim Acta 2000;297:93–102.
- 39.↑
Locatelli F, Marcelli D, Comelli M, et al. Proteinuria and blood pressure as causal components of progression to end stage renal disease. Nephrol Dial Transplant 1996;11:461–467.
- 40.↑
Willnow TE, Nykjaer A, Herz J. Lipoprotein receptors: new roles for ancient proteins. Nat Cell Biol 1999;1:E157–E162.
- 41.↑
Remuzzi G, Bertani T. Pathophysiology of progressive nephropathies. N Engl J Med 1998;339:1448–1456.
- 42.
Brenner BM. Nephron adaptation to renal injury or ablation. Am J Physiol 1985;249:F324–F337.
- 43.
Hostetter TH, Olson JL, Rennke HG, et al. Hyperfiltration in remnant nephrons: a potentially adverse response to renal ablation. Am J Physiol 1981;241:F85–F92.
- 44.↑
Wellwood JM, Price RG, Ellis BG, et al. A note on the practical aspects of the assay of N-acetyl-β-glucosaminidase in human urine. Clin Chim Acta 1976;69:85–91.
- 45.
Tassi C, Mancuso F, Feligioni L, et al. Expression modes of urinary N-acetyl-β-D-glucosaminidase in patients with chronic renal insufficiency. Clin Chim Acta 2004;346:129–133.
- 46.↑
Peterson ME, Gamble DA. Effect of nonthyroidal illness on serum thyroxine concentrations in cats: 494 cases (1988). J Am Vet Med Assoc 1990;197:1203–1208.
- 47.↑
Tominaga M, Fujiyama K, Hoshino T, et al. Urinary N-acetyl-β-D-glucosaminidase in the patients with hyperthyroidism. Horm Metab Res 1989;21:438–440.
- 48.
Adams LG, Polzin DJ, Osborne CA, et al. Influence of dietary protein/calorie intake on renal morphology and function in cats with 5/6 nephrectomy. Lab Invest 1994;70:347–357.
- 49.
Finco DR, Brown SA, Brown CA, et al. Protein and calorie effects on progression of induced chronic renal failure in cats. Am J Vet Res 1998;59:575–582.
- 50.
Elliott J, Rawlings JM, Markwell PJ, et al. Survival of cats with naturally occurring chronic renal failure: effect of dietary management. J Small Anim Pract 2000;41:235–242.
- 51.
Ross SJ, Osborne CA, Kirk CA, et al. Clinical evaluation of dietary modification for treatment of spontaneous chronic kidney disease in cats. J Am Vet Med Assoc 2006;229:949–957.
