Absence of increased urinary excretion of adenosine-deaminase-binding protein by patients with chronic renal tubular malfunction

Clinica Chimica Acta, 200 (1991) 183-190 0 1991 Elsevier Science Publishers B.V. All rights reserved 0009-8981/91/$03.50 ADONIS 000989819100166F

183

CCA 05033

Absence of increased urinary excretion of adenosine-deaminase-binding protein by patients with chronic renal tubular malfunction Frederick V. Flynn ‘, Marta Lapsley ‘, Paul A. Sansom ’ and Anthony G.W. Norden * ’ Department of Chemical Pathology, University College and Middlesex School of Medicine, London and ’ Department of Chemical Pathology, Enfield, Disrricr Hospital, Middlesex (UK) (Received 6 November 1990; revision received 6 March 1991; accepted 7 May 1991) Key words: Proteinuria; Adenosine-deaminase-binding

protein; Renal tubular function; Renal tubular

disorders

Summary The urinary excretion of adenosine-deaminase-binding protein, a constituent of the brush border of proximal renal tubule cells, has been investigated in 39 patients with disorders associated with malfunction of the renal tubules, and its excretion has been compared with that of two low molecular mass plasma proteins and an enzyme derived from renal tubular cells. None of the 36 patients~ with disorders associated with chronic renal tubular malfunction were found to be excreting significantly increased quantities of adenosine-deaminase-binding protein but 30 had increased excretion of retinol-binding protein, cY,-microglobulin, or N-acetyl-P-D-glucosaminidase. Measurement of urinary adenosine-deaminase-binding protein may be useful in the assessment of acute renal tubular injuries but it is not of value in the detection of chronic renal tubular disorders.

Introduction An increased urinary excretion of low molcular mass plasma proteins characterises many pathological conditions affecting the proximal renal tubule [l]. Pro-

Correspondence to: F.V. Flynn, Department of Chemical Pathology, University College and Middlesex School of Medicine, Windeyer Building, Cleveland Street, London WlP 6DB, UK.

184

teins such as retinol-binding protein (RBP) and aI-microglobulin (cr,M) are normally freely filtered at the glomerulus and virtually completely reabsorbed and catabolised by the proximal renal tubule cells; decreased reabsorption accounts for their increased urinary excretion in renal tubular disorders. Unfortunately, increased excretion of these proteins does not always reflect tubular malfunction because it may also result from saturation of a normal tubular resorptive capacity due to an increase in the filtered load of such proteins secondary to nephron loss. Likewise, while measurement of urinary N-acetyl-p-D-glucosaminidase (EC 3.2.1.30) (NAG) activity permits the early detection of renal tubular damage, the specificity and sensitivity of this test leaves something to be desired [2]. Consequently, when Thompson et al. [3] reported that the excretion of adenosinedeaminase-binding protein &BP), a constituent of the brush border of proximal renal tubular cells, was markedly increased in patients with acute tubular injury, and that it was only barely detectable in the urine of healthy persons and patients with glomerular disease, we were interested to explore its potential diagnostic usefulness in patients with chronic renal tubular disorders. Patients and samples Random specimens of urine were obtained from 36 patients unequivocally diagnosed as having conditions associated with chronic renal tubular malfunction and also from three patients with acute renal tubular failure and six healthy adults. The diagnoses applying to the patients are given in Table I; the two with renal transplants had received an allograft some months earlier, the two with nephrocalcinosis had previously suffered from primary hyperparathyroidism, and the patient with oxalate nephropathy had previously had jejunal-ileal bypass surgery for gross obesity. The urine specimens were centrifuged at 1,800 X g for 10 min and the supernatant divide into portions for the different assays. The portions for the RBP, (YAM and NAG assays were frozen at -20 ’ C without preservative. The portions for the ABP assay were stored at 4 o C after addition of a preservative to inhibit proteinase activity: 25 ~1 of the buffered preservative solution, pH 7.6, was added per ml of urine to give the following final concentrations; Tris 20 mmol/l, benzamidine 1 mmol/l, and EDTA 10 mmol/l. Methods All samples were assayed for creatinine, ABP, RBP, and a,M and all but three for NAG; those omitted from the latter assay had not been satisfactorily stored. All the assays except those of creatinine were performed in duplicate and the mean values recorded. To reduce the effects of the different concentrations of the urine specimens, excretion of the proteins was expressed per mm01 creatinine; this was not practical in the case of the ABP assays because the concentration in most specimens could only be recorded as < 0.1 unit/l.

185 TABLE I Urinary excretion of adenosine-deaminase-binding protein (ABP), iV-acetyl-/3-D-glucosaminidase (NAG), retinol-binding protein (RBP), and a,-microglobulin (aiM1, by 39 patients with conditions associated with renal tubular malfunction Adult reference range

Sex

Age

ABP (arbitrary U/l) < 0.17

NAG (~mol/h/mmol creatinine) < 25

RBP (pg/mmol creatinine) < 17

o1M (mg/mmol creatinine) < 0.7

Acute tubular necrosis Acute tubular necrosis Acute tubular necrosis Renal transplant Renal transplant Adult Fanconi syndrome Lowe’s syndrome Interstitial nephritis Interstitial nephritis Dent’s disease [4] Dent’s disease Cadmium poisoning Nephrocalcinosis Nephrocalcinosis Wilson’s disease Wilson’s disease Oxalate nephropathy Analgesic nephropathy Familial RTA a Familial RTA Familial RTA Familial RTA Familial RTA Familial RTA Familial RTA Familial RTA Familial RTA Immune mediated RTA Immune mediated RTA Immune mediated RTA Immune mediated RTA Medullary sponge kidney Medullary sponge kidney Polycystic kidney Polycystic kidney Polycystic kidney Polycystic kidney Polycystic kidney Polycystic kidney

M M M F M M M M F M M M F F M F F F M M M M F F M M M F F F F F F F F M M F F

59 58 69 52 50 33 16 37 79 36 35 73 69 60 31 25 56 57 41 38 37 37 35 64 47 20 57 52 27 56 43 37 42 60 78 61 43 51 33

2.90 2.60 1.60 0.18 < 0.10 < 0.10 < 0.10 < 0.10 < 0.10 < 0.10 < 0.10 < 0.10 < 0.10 < 0.10 0.18 < 0.10 < 0.10 < 0.10 0.10 < 0.10 < 0.10 < 0.10 < 0.10 <,O.lO < 0.10 < 0.10 < 0.10 0.14 < 0.10 < 0.10 < 0.10 0.19 < 0.10 < 0.10 < 0.10 < 0.10 < 0.10 < 0.10 < 0.10

10,190 839 1,553 24 138 77 61 55 36 36 27 26 52 27 12 6 13 89 142 121 21 20 14 7

400 219 164 129 34,102 16,860 183,870 1,533 38 209 20,731 10,436 8,019 33 4 9 50 169 303 49 82 1,481 169 11 50 26 18 6 14 19 8 298 5 57,290 34,255 18 12,448 211 7

1.0 7.0 16.5 2.9 7.1 13.8 41.9 6.7 3.0 23.3 17.5 12.3 16.1 1.1
a RTA = Renal tubular acidosis

42 97 29 13 18 17 191 112 70 54 36 13

186

The concentration of creatinine was measured by the Jaffe reaction using the kinetic method employed in the American Monitor Perspective analyser. ABP was assayed by the sandwich enzyme-linked immunosorbent procedure of Thompson et al. [3] using the ‘Nephroscreen’ kit from Signet Laboratories Inc. Massachusetts; this uses two monoclonal antibodies which recognise different epitopes of the protein. RBP was measured by the sandwich enzyme-linked immunosorbent method of Topping et al. [5], and cu,M by the similar procedure of Takagi et al. [6] using the ‘Imzyne cu,-M’ kit from Mast Diagnostics, Bootle, Merseyside. NAG activity was measured by a calorimetric procedure based on the method of Yuen et al. [7], using a Cobas-Bio centifugal analyser and a kit from Cortecs Diagnostics Ltd., Deeside, Clwyd. Results The results obtained on the urine specimens from the healthy controls all fell within the reference intervals quoted in Table I. The ABP concentration in all six healthy controls was < 0.1 arbitrary U/l, and the mean values for the NAG, RBP, and cu,M results, expressed per mmol of creatinine, were 15 pmol/h, 5 pg, and 0.3 mg, respectively. The results obtained on the patients are shown in Table I. This shows that 31 of the 36 patients with chronic renal tubular malfunction had urinary ABP levels less than 0.1 unit/l, and that 33 had concentrations below 0.16 U/l, which was the highest level found among 30 healthy adults by Thompson et al. [3]. The excretion of ABP by all of our patients with chronic tubular malfunction was below the level of 0.2 U/l, which is accepted as the lowest level compatible with tubular injury. On the other hand the three patients with acute tubular failure were found to be excreting ABP at concentrations between 1.2 and 2.6 U/l of urine. Of the 36 patients with conditions associated with chronic renal tubular malfunction, 30 were excreting increased quantities of either or both of the low molecular mass proteins measured or of the enzyme NAG; the 6 with normal values for all these measurements included the two patients with Wilson’s disease who had received chelation therapy for several years. Twenty-eight were excreting increased quantities of RBP, i.e. more than 16 pgg/mmol creatinine, and 27 increased amounts of cz,M, i.e. more than 0.6 mg/mmol creatinine. Twenty-one of the 33 patients whose specimens were suitable for the assay of NAG were found to have increased urinary enzyme activity, i.e. the urine was capable of hydrolysing > 24 pmol of the substrate/h/mmol creatinine. Discussion Zager and Carpenter [S] were the first to produce evidence that suggested that measurement of the urinary excretion of kidney tissue proteins might be useful for the detection of injury to the renal tubules. Further evidence in favour of this concept was produced when Mutti et al. [9] reported that the urinary excretion of BB-50, a 50 kDa brush border antigen, provided a very sensitive and specific test

187

for the presence of toxic nephropathies, and Thompson et al. f3] reported that the excretion of ABP, also derived from the brush border of proximal renal tubular cells, reflected the severity of the injury in patients with acute tubular necrosis and acute renal allograft rejection. Subsequently a detailed study of the use of serial ABP measurements in renal transplant patients by Tolkoff-Rubin et al. [lo] demonstrated that urine ABP was elevated l-7 days before rejection was recognised clinically, and that ABP measurements were valuable in monitoring the effectiveness of anti-rejection therapy and in distinguishing cellular rejection from other allograft injuries. The findings reported by Tolkoff-Rubin [ll] and Falkenberg et al. [12] on the clinical application of a battery of assays designed to measure kidney-derived urinary antigens provided further evidence of the diagnostic value of measuring the excretion of proteins derived from renal cells. Falkenberg et al. included a few patients with chronic tubulo-interstitial disease, and they reported substantially increased excretion of two proximal tubular antigens by individual patients with analgesic nephropathy and chronic urinary tract infection, and lesser increases in individuals with medullary cystic disease and acute pyelonephritis. Our findings regarding ABP excretion by healthy subjects, and the increases we found among three patients with acute tubular necrosis, correspond closely with what has been reported by others. Surprisingly, none of the 36 patients with conditions associated with chronic renal tubular malfunction had excretion levels above 0.2 U/l, which is regarded as clearly indicative of tubular injury. However not all of this group of patients can be assumed to have damage to the proximal renal tubular cells: for instance those with distal renal tubular acidosis might not be expected to have their proximal tubules affected. We included such patients in our study because we have come to appreciate that patients with renal tubular disorders rarely have pure distal or proximal tubular malfunction; in this context it is notable that of the ten patients with proven renal tubular acidosis who had their excretion of NAG measured, five were excreting increased amounts of this proximal tubular enzyme. Twenty-seven of these 36 patients had an increased excretion of both RBP and LY~M,and one an increase of RBP alone, but it is difficult to be certain that such findings necessarily reflect impairment of proximal tubular cell function. However, Bernard, Vyskocil, Mahieu and Lauwerys [13] have suggested that an increased urinary excretion of RBP could be regarded as diagnostic of proximal tubular impai~ent if the plasma ~reatinine is less than 182 ~mol/I. As 16 of the 28 patients with increased ‘excretion of RBP had plasma creatinine levels below 170 ,umol/l, there should be no doubt that at the time that their ABP excretion was investigated many of our patients had satisfactory evidence of impaired function of their proximal renal tubules. Urinary albumin was also measured in all our patients and the fact that we could find no consistent relation between the excretion of albumin and RBP, or albumin and (YAM,also suggests that an increased filtered load of other proteins, due to increased glomeruiar permeability, is unlikely to account for the ,increased excretion of low molecular weight proteins by many of the patients. The difference in the excretion of ABP between those with acute and those with

188

chronic damage to the renal tubules probably reflects increased amounts of ABP being shed from the brush border of proximal renal tubular epithelial cells only at the time of active cellular injury. Tubular proteinuria on the other hand, while also a sensitive index of proximal renal tubular damage, reflects the long term functional consequence of damage to the proximal tubular cell, and may persist therefore long after the causative factor of any acute injury has been removed. Our finding of normal or low excretion of ABP by patients with chronic renal tubular disorders was paralleled by Dr A. Mutti not finding an increase of BB-50 antigen in the urine from six of the patients reported here, nor in seven other patients attending University College Hospital who have similar disorders [14]. The 50 kDa antigen that Mutti’s ELISA measures is different from ABP, because the latter is a glycoprotein of 120 kDa which is excreted as a dimer. Nevertheless Mutti’s results on our patients give added confidence to our own findings. The contrast between ABP and NAG excretion in the patients with chronic renal tubular disorders is intriguing. Both are constituents of proximal tubule cells and both are excreted in increased quantity in patients with acute tubular damage, yet only NAG is excreted in excess by patients with chronic tubular disorders. ABP is a protein found at the surface of the brush border while NAG is a lysosomal enzyme, and so it is conceivable that in response to acute cellular injury ABP might be shed from a given cell in totality and either not regenerated or only re-formed at a rate less than at which it is normally lost, while NAG being within lysosomes might only be partially released and possibly later regenerated; the consequence would be that as damage continued only NAG would remain to be released. Another possible explanation might be that the agents which are responsible for chronic tubular damage, such as the gradual intracellular accumulation of cadmium or copper, cause only increased permeability of the cell membrane rather than its disruption, whereas those agents that cause acute tubular injury, such as hypoxia or nephrotoxic drugs, have more devastating effects and therefore cause the release of both ABP and NAG. A further alternative explanation for ABP excretion not mirroring that of NAG activity could be that the toxic agents that slowly accumulate to cause chronic malfunction might be inducing increased lysosomal formation with consequential increased production of NAG. We conclude from our findings that while assays of urinary ABP and NAG are of value for the detection of acute renal tubular damage, measurement of the urinary excretion of low molecular weight plasma proteins should be used for the detection of chronic renal tubular malfunction because their increased excretion will persist after any acute injury to the proximal renal tubular cells has ceased to cause the shedding of cellular proteins. NAG assays might be indicated for the detection of continuing proximal tubular damage. Acknowledgements

This work was supported by a grant from the Locally Organised Research Scheme of the North East Thames Regional Health Authority which we gratefully acknowledge. We also thank Prof. 0. Wrong, Dr. J.M. Walshe, Dr. D.P. Brenton

189

and Dr. S.L. Cohen for pe~iss~on to study their patients, and Dr. A. Mutti for kindly permitting us to refer to his findings regarding the excretion of BB-50 antigen. References 1 Wailer KV, Ward KM, Mahan JD, Wismatt DK. Current concepts in proteinuria. Clin Chem 1989;35:755-765. 2 Jung K, Diego J, Strobelt V, Scholz D, Schreiber G. Diagnostic significance of some urinary enzymes for detecting acute rejection crises in renal-transplant recipients: alanine aminopeptidase, alkaline phosphatase, y-glutamyltransferase, N-ace@-P-D-glucosaminidase, and lysozyme. Clin Chem 1986;32:1807-1811. 3 Thompson RE, Piper DJ, Galberg C, Chan ‘IX, Tolkoff-Rubin NE, Rubin RH. Adenosine deaminase binding protein, a new diagnostic marker for kidney disease. Clin Chem 1985;31:679-683. 4 Wrong OM, Norden AGW, Feest TG. Dent’s disease: a familial renal tubular syndrome with hypercalciuria, tubular proteinuria, rickets, nephrocalcinosis and eventual renal failure. Q J Med 1990;77:1086,1087. 5 Topping MD, Forster HW, Dolman C, Luczynska CM, Bernard AM. Measurement of urinary retinal-binding protein by enzyme-linked immunosorbent assay, and its application to detection of tubular proteinuria. Clin Chem 19~,3~1863-18~. 6 Takagi K, Koyamaishi Y, Itoh Y, Enomoto H, Kawai T. En~e-immunoassay of human t~i-microglobulin. Jpn 3 Clin Chem 1981;10:30-39. 7 Yuen CT, Kind PRN, Price RG, Praill PFG, Richardson AC. Calorimetric assay for N-acetyl-fi-Dglucosaminidase (NAG) in pathological urine using the w-nitrostytyl substrate: the development of a kit and the comparison of manual procedure with the automated fluorimetric method. Ann Clin Biochem 1984;21:295-300. 8 Zager RA, Carpenter CB. Rad~o~mmunoassay for urinary renal tubular antigen: a potential marker of tubular injury. Kidney fnt 1978;13:505-512. 9 Mutti A, Lucertini S, Valcavi P, et al. Urinary excretion of brush-border antigen revealed by monoclonal antibody: early indicator of toxic nephropathy. Lancet 1985;ii:914-917. 10 Tolkoff-Rubin NE, Cosimi AB, Delmonico FL, et al. Diagnosis of tubular injury in renal transplant patients by a urinary assay for a proximal tubular antigen, the adenosine-deaminase-binding protein. Transplantation 1986;41:593-597. 11 Tolkoff-Rubin NE. Mon~lonal antibodies in the diagnosis of renal disease: a prelimina~ report. Kidney Int 1986;29:142-152. 12 Falkenberg FW, Mai U, Puppe C, et al. Kidney-derived urinary antigens assayed with monoclonal antibodies for the detection of renal damage. Clin Chim Acta 1986;160:171-182. 13 Bernard AM, Vyskocil AA, Mahieu P, Lauwerys RR. Assessment of urinary retinal-binding protein as an index of proximal tubular injury. Clin Chem 1987;33:775-779. 14 Mutti A. Detection of renal diseases in humans: developing markers and methods. Toxicology Letters 1989;46:177-191.