J Papers - From BMJ and ACP · included were six patients with overt diabetic nephropathy, three with polyarteritis nodosa, and single cases ofWegener's granulomatosis, systemic lupus

J Clin Pathol 1992;45:561-567

Papers

Urinary excretion of f2-glycoprotein- 1

(apolipoprotein H) and other markers of tubularmalfunction in "non-tubular" renal disease

F V Flynn, M Lapsley, P A Sansom, S L Cohen

AbstractAim: To determine whether urinaryf12-glycoprotein-1 assays can provideimproved discrimination between chronicrenal diseases which are primarily oftubular or glomerular origin.Methods: Urinary /4-glycoprotein-l, reti-nol-binding protein, a,-microglobulin,/J,-microglobulin, N-acetyl-p-D-glucosa-minidase and albumin were measured in51 patients with primary glomerular dis-ease, 23 with obstructive nephropathy,and 15 with polycystic kidney disease, andexpressed per mmol ofcreatinine. Plasma/12-glycoprotein-I was assayed in 52patients and plasma creatinine in all 89.The findings were compared between thediagnostic groups and with previouslypublished data relating to primary tubu-lar disorders.Results: All 31 patients with plasma crea-tinine greater than 200 umolIl excretedincreased amounts of 12-glycoprotein-1,retinol-binding protein, and aL-microglo-bulin, and 29 had increased N-acetyl-/l-D-glucosaminidase; the quantititeswere generally similar to those found incomparable patients with primary tubularpathology. Among 58 with plasma crea-tinine concentrations under 200 pmolIl,increases in a12-glycoprotein-l, retinol-binding protein, and a1-microglobulinexcretion were less common and muchsmaller, especially in those with obs-tructive nephropathy and polycystic dis-ease. The ratios of the excretion of albu-min to the other proteins provided theclearest discrimination between thepatients with glomerular or tubular mal-function, but an area of overlap waspresent which embraced those withobstructive nephropathy and polycysticdisease.Conclusions: Increased excretion of/12-glycoprotein-l due to a raised plasmaconcentration or diminution of tubularreabsorption, or both, is common in allthe forms of renal disease investigated,and both plasma creatinine and urinaryalbumin must be taken into account wheninterpreting results. Ratios of urinaryalbumin:p2-glycoprotein-1 greater than1000 are highly suggestive of primaryglomerular disease and those less than 40of primary tubular disease. Used in thisway, 62-glycoprotein-l assays provide

superior discrimination between glomer-ular and tubular malfunction when com-pared with retinol binding protein but thebest discrimination is provided by albu-min: a1-microglobulin ratios.

In recent years increased urinary excretion ofretinol binding protein (RBP), a,-microglobu-lin (a,M), or fi2-microglobulin (f2M) havebeen used as markers of proximal renal tubularmalfunction. These low molecular weight plas-ma proteins are freely filtered at the glomerulusand normally virtually completely reabsorbedby the proximal renal tubule cells, where theyare catabolised.' Defective reabsorptionaccounts for their increased urinary excretionin renal tubular disorders. Unfortunately, find-ing them in excess is not specific to renaltubular disease because all three are excretedin increased quantity whenever the endoge-nous creatinine clearance is reduced-to25-30 ml/minute in the case ofRBP and /i2M,2and to less than 55 ml/minute in the case ofa1m.3

Increased excretion of N-acetyl-f,-D-gluco-saminidase (NAG), an enzyme found in thelysosomes of proximal renal tubule cells,should be more specific for tubular pathologybecause its molecular mass is large enough topreclude passage through the normal glo-merular basement membrane. However,increased excretion ofNAG only reflects activetubular damage and it has also been reportedin patients with glomerulonephritis.4 Increasedalbumin excretion usually reflects glomerulardisease but when the quantities are small itmay be due to impaired tubular reabsorption:thus some albuminuria is regularly found inpatients with renal tubular disorders.5

In the course of searching for an improvedprotein marker of renal tubular disease wediscovered that fl2-glycoprotein-1 (fl2Gl) isexcreted in increased quantity by patients withprimary renal tubular disorders, despite itsrelatively high molecular mass.6 To determinewhether increased excretion of ,2G1 might bea more specific marker for tubular malfunctionthan existing tests we have now investigated itsexcretion by patients with primary glomerularpathology, obstructive nephropathy, and poly-cystic kidney disease, comparing its excretionwith that of albumin, RBP, a,M, fl2M, andNAG. The study was approved by the hospitalethical committee.

Department ofChemical Pathology,University College andMiddlesex School ofMedicine, WindeyerBuilding, ClevelandStreet, LondonWlP6DBFV FlynnM LapsleyP A SansomDepartment ofMedicine, UniversityCollege and MiddlesexSchool of Medicine,Rayne Institute,LondonS L CohenCorrespondence to;Professor F V FlynnAccepted for publication28 November 1991

561

on January 21, 2021 by guest. Protected by copyright.

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.45.7.561 on 1 July 1992. D

ownloaded from

http://jcp.bmj.com/

Flynn, Lapsley, Sansom, Cohen

MethodsRandom midstream specimens of urine andsamples of blood anticoagulated were obtainedfrom 89 patients (54 men and 35 women) agedbetween 18 and 90 years. All were under thecare of a nephrologist and had had someincrease in urinary protein excretion. They

included 51 with a variety of diseases asso-

ciated primarily with glomerular pathology, 23with nephropathy following unequivocallydocumented renal tract obstruction, and 15with radiologically confirmed polycystic kid-ney disease. Those with primary glomerularpathology included eight with IgA and one

with IgM nephropathy eight with focal seg-

mental glomerulosclerosis, seven with mem-branous nephropathy, five with minimalchange nephropathy, five with renal amyloido-sis, four with mesangiocapillary glomerulone-phritis, and one with membranoproliferativeglomerulonephritis, in all of whom the diag-nosis had been confirmed by renal biopsy. Alsoincluded were six patients with overt diabeticnephropathy, three with polyarteritis nodosa,and single cases ofWegener's granulomatosis,systemic lupus erythematosus, and hyperten-sive nephropathy.The urine and blood specimens were cen-

trifuged at 1800 x g for 10 minutes and thesupernatant divided into portions for the dif-ferent assays. All the urine fractions werefrozen within four hours of collection and keptat - 20'C until shortly before analysis.

All the urine samples were assayed forcreatinine, albumin, fl2G, RBP, a,M, fl2Mand NAG, and all the plasma samples forcreatinine. Plasma f2G1 concentrations weremeasured in 52 patients. All the assays, exceptthose for creatinine, were performed in dupli-cate. To reduce the effects of the differentconcentrations of the urine specimens and tofacilitate comparison of the excretion of thedifferent proteins, all results were expressed inmg per mmol of creatinine.The concentration of creatinine was meas-

ured by the Jaffe reaction, using the kineticmethod employed in the American Monitor

Perspective analyser. Albumin was measuredby an immunoturbidimetric method with aCobas-Bio centrifugal analyser, using theAmes "Microalb" kit from Miles LaboratoriesLtd (Stoke Poges, Slough, England). f,2G1 wasmeasured by in-house methods described else-where,7 using a sandwich enzyme linkedimmunosorbent assay (ELISA) for urine assaysand a radial immunodiffusion procedure forplasma measurements. RBP was measured bythe ELISA method ofTopping et al,8 and aiMby the ELISA procedure ofTakagi et al,9 usingthe "Imzyne aiM" kit from Mast Diagnostics(Bootle, Merseyside, England). f2M wasmeasured by a radioimmunoassay methodusing the ,82-Micro RIA kit from PharmaciaLtd (Midsummer Boulevard, Milton Keynes,England). NAG activity was measured by acolorimetric procedure, based on the methodofYuen et al,"O using a Cobas-Bio centrifugalanalyser and the kit available from CortecsDiagnostics Ltd (Deeside Industrial Park,Clwyd, Wales).

ResultsThe findings in the 51 patients with chronicrenal disease primarily ofglomerular origin, 23with obstructive nephropathy, and 15 withpolycystic kidney disease are summarised intables 1-3, respectively. Table 1 also citescorresponding findings from 60 patients withchronic renal tubular disorders studied earlier'and in whom the diagnoses had been made byphysicians with special expertise in the diseasesconcerned.Among the 37 patients with primary glo-

merular pathology and a plasma creatinine ofless than 200 4umol/l, an increased excretion ofNAG, a M, f,2G1 and RBP was found, respec-tively, in 73%, 59%, 57%, and 49% of thesubjects. The excretion of #2M was within thereference range in 33 of the patients, but in thiscontext it should be noted that the pH of thespecimens was below 5-5 in 11 cases and hadnot been recorded in two others. When an

Table I Urinary excretion offive plasma proteins and ofNAG by 5I patients with chronic glomerular pathology, 37with plasma creatinine concentrations under 200 (group GI) and 14 with concentrations above 200 pmolll (group G2)

All patients Patients with increased excretion

Mean and range of excretion* % ofgroup Mean excretion

(GI) (G2) (GI) (G2) (GI) (G2)ALB 190 (<0.05-760) 322 (86-1100) 97% 100% 217 x URL 358 x URLfl,Gl 0-121 (0 008-1 54) 3 97 (0-059-18-7) 57% 100% 5-6 x URL 115 x URLRBP 0 079 (0-003-1-36) 2-26 (0-084-13-2) 49% 100% 9-5 x URL 141 x URLaiM 1-4 (<0 01-8-7) 9 3 (2 6-28 3) 59% 100% 2-9 x URL 13 x URLflM 0-06 (0-007-0-91) 3-17 (0-025-20-1) 11% 71% 3-2 x URL 44 x URLWAG 56.2 (8-144) 127 (26-6-576) 73% 100% 2-9 x URL 5-1 x URL

(Ti) (T2) (Ti) (T2) (Ti) (T2)ALB 15-6 (<0.05-118) 64-7 (2-2-197) 83% 100% 21 x URL 72 x URLfl GI 0 940 (0 011-30 8) 4-96 (0-025-21-9) 79% 92% 34 x URL 157 x URLiRP 5-42 (0-004-184) 7-31 (0 049-20 7) 75% 100% 452 x URL 457 x URLa,M 4-5 (0-1-62-5) 11-7 (2-6-283) 75% 100% 8-5 x URL 17 x URLfi M 2-7 (0-01-68-7) 7-38 (0-06-19-4) 40% 92% 68 x URL 80 x URLI4AG 36.7 (1-5-176) 52-5 (12-7-121) 50% 90% 2-4 x URL 2-3 x URL

Corresponding findings applying to 60 patients with chronic renal tubular disorders (n = 5) are included for comparison, 48with plasma creatinine concentrations under 200 (group T1) and 12 with concentrations above 200 umol/l (group T2). URL =upper reference limit.*Protein values are in mg and enzyme activity in jsumol of substrate hydrolysed per hour, and all results are expressed per mmolof urinary creatinine, and the reference ranges applying are:albumin (ALB) = < 0-9; a,-microglobulin (a,M) = < 0-7; ,a-glycoprotein-I (#,G1) = 0-0069-0-0345; fl-microglobulin ((iM) =< 0-1; retinol binding protein (RBP) = 0-001-0-016; N-acetyl-ft-D-glucosaminidase (NAG) = < 25.

562


http://jcp.bmj.com

/J C


ownloaded from

http://jcp.bmj.com/

Excretion of markers of tubular malfunction in non-tubular renal disease

Table 2 Urinary excretion offive plasma proteins and ofNAG by 23 patients with obstructive nephropathy, 13 withplasma creatinine concentrations under 200 (group 01), and 10 with concentrations above 200 umolll (group 02)


Mean and range of excretion* % ofgroup Mean excretion

(01) (02) (01) (02) (01) (02)ALB 9 7 (<0.05-24) 167 (6 2-389) 85% 100% 12-8 x URL 185 x URLfl,G1 0-056 (0-019-0-108) 1 38 (0 067-3-86) 62% 100% 2-4 x URL 40 x URLRBP 0036 (0003-0-153) 7-38 (3 15-15 4) 46% 100% 4-3 x URL 461 x URLaiM 05 (<001-14) 7-20 (2-5-13-5) 31% 100% 1-7 x URL 10 x URL,B M 0 075 (0 014-0 19) 6-21 (0-028-17-7) 31% 90% 1.5 x URL 62 x URLI4AG 29-5 (8 7-87) 63-3 (19 5-237) 46% 80% 1-8 x URL 2-5 x URL

Table 3 Urinary excretion offive plasma proteins and ofNAG by 15 patients with polycystic kidney disease, 8 withplasma creatinine concentrations under 200 (group PI), and 7 with concentrations above 200 ,umolll (group P2)


Mean and range of excretion * % ofgroup Mean excretion

(P1) (P2) (P1) (P2) (P1) (P2)ALB 5-5 (0.7-23) 115 (8-4-527) 88% 100% 7-1 x URL 128 x URLfl,G1 0 043 (0-02-0 075) 1-32 (0049-4 07) 63% 100% 1-6 x URL 38 x URLRBP 0-013 (0 007-0 036) 17-5 (0-201-57-3) 25% 100% 1-7 x URL 1090 x URLa,M 0 5 (0 1-1 3) 28-3 (3-0-118) 38% 100% 1-6 x URL 40 x URLfi,M 0-031 (0-021-0-038) 10-2 (0-103-32-5) 0% 100% 102 x URLNAG 22-0 (8-5-70) 78-5 (36-5-191) 25% 100% 2-0 x URL 3-1 x URL

increased excretion was present, the magni-tude of the increase decreased in the orderRBP > fl2G1 > /2M > a,M and NAG. Sixofthe subjects had a normal excretion offl2G,RBP, a,M, fl2M and NAG and three anincreased excretion of all these constituents.Eleven were excreting increased quantitites offl,2G1, RBP, and a,M, and 13 increasedamounts of two of these three proteins. Theincidence and magnitude of the increase inexcretion of f2Gl and low molecular weightproteins, but not NAG, was appreciably lessthan in patients with renal disease primarily oftubular origin, and albumin excretion was onaverage 12 times higher.

All 14 patients with primary glomerularpathology and a plasma creatinine above200 4umol/l were excreting increased quantitiesof albumin, f,2G1, RBP, a,M and NAG, thefirst three usually in substantially increasedamounts. Ten had raised amounts off2M, butamong the four patients in whom the excretionwas within the reference range the pH of theurine specimens was below 5-5 in three. Theexcretion of NAG, fl2GI, and the low molec-ular weight proteins was generally much great-er than in those with a plasma creatinine of lessthan 200 pmol/l: the average excretion of fJ2Mwas 53 times greater, and the correspondingfigures for fl2G, RBP, a,M, NAG and albu-min were, respectively, 33, 29, 7, 2-3 and 1-7times greater. When compared with patientswith primary tubular pathology and similarplasma creatinine concentrations, this groupexcreted five times as much albumin, twice asmuch NAG, and about two thirds the quantityof the other proteins.Among the 13 patients with obstructive

nephropathy and a plasma creatinine under200 ,umol/l the incidence of increased excre-tion ofthe constituents measured was generallyless than in the patients with primary glo-merular and primary tubular pathology and itsmagnitude consistently so. Three had normalexcretion of all the constituents measuredexcept albumin, and three had increased excre-

tion of f2G1, RBP, and a1M. Of the 10 withobstruction and a plasma creatinine exceeding200 umol/l, the excretion of all the con-stituents measured was raised in seven and theoutput of fi2G, RBP, and a,M was increasedin all. The amounts excreted were less than inpatients with primary glomerular and primarytubular pathology except in the case of RBP,and sometimes f2M, which reached excep-tionally high values in the six patients withplasma creatinine concentrations above600 ,umolIl.Among the eight patients with polycystic

kidney disease and a plasma creatinine concen-tration of less than 200 pumol/l, the incidenceand magnitude of increased excretion of theconstituents measured was less than that foundin patients with primary glomerular and pri-mary tubular pathology. Two had f2GI, RBP,a,M and NAG all within the reference range,one had all the measured constituents raised,and one had increased excretion off2Gl, RBP,and a,M. All seven who had a plasma crea-tinine greater than 200 umol/l had increases ofall the constituents measured, the rise in RBP,a,M, and f2M being particularly pronouncedin the four subjects with plasma creatinineconcentrations in excess of 600 pmolIl.

Figure 1 shows the excretion of fl2G 1, RBP,a1M and NAG by each of the 89 patientsstudied, plotted against their plasma creatinineconcentration. These plots show that in allthree diagnostic groups, when the plasmacreatinine concentration exceeds 200 umol/l,the excretion of the proteins is consistentlyraised and, with few exceptions, the same istrue of urinary NAG activity. That the excre-tion of f,2Gl may be reflecting an increase inthe plasma concentration can be deduced fromfig 2, where the plasma f2G1 concentrationsfound in 52 patients are plotted against theirplasma creatinine concentrations.

Figure 3 shows the excretion off2G1, RBP,alM and NAG by each of the 58 patientswhose plasma creatinine concentrations wereless than 200 pumol/l, plotted against their

563


http://jcp.bmj.com

/J C


ownloaded from

http://jcp.bmj.com/


100000

D 1000j

AL 100'

0oL 10I0)

C~ 1 I

0-1

10 000.

IC 1000

c* 100'.0

0o 10

EIL. 1I

0

II

a'o 1Iol1

's ° 'Ioi 0o

-A

a

a

a

a

0a

0

0 a.

0

a0

0

-i

0

Ca

.5

0

0 200 400 600 800 1000

I a

o roa.aA a

0 00:

£0a

W;IAv I

10 000-

1000.

100.

10

01,

-J

D 10 000

0

X 1000

._

E 1000

10.cm

I 1 i

0

X 0-1.

q a

I

a l

I

A

a 0200 A

a 0 0o

a

o a &

& ao £0

a

alp II

l-

200 400 600 800 1000

aA

A

o o,

, .I I I I ..I I

0 200 400 600 800 1000 0 200 400 00 800 1000

Plasma creatinine FmolAl Plasma creatinine p.molAlFigure 1 Plots showing correlation between plasma creatinine concentration and urinary excretion of /12 GI, RBP, a,M,and NAG among 5I patients with primary glomerular pathology ([), 23 with obstructive nephropathy (0), and 15with polycystic kidney disease (A). In each the horizontal line marks the upper reference limit for the urinary constituentand the vertical line the 200 pmolll creatinine concentration. The urinary values are plotted as multiples of the upperreference limit (MURL) on logarithmic scales.

albumin excretion. A tendency for the excre-tion off2G1, a,M, and NAG to rise in parallelwith increasing albumin loss can be seen butthe correlation is poor. Figure 4 shows theratios of the excretion of albumin to that of,62G1, RBP, and a,M among the patients withprimary glomerular pathology, plotted along-side the corresponding results obtained fromthose with obstructive nephropathy and poly-cystic kidney disease, and those derived frompatients with chronic renal tubular disordersstudied earlier.5 In a few cases albumin excre-tion values close to zero precluded calculationof the ratios. The figure shows the expectedtendency for the proportion of albumin to behigher among those with primary glomerular

_ 500-

E

,-. 400-._

0

NO 300-

200-

E 100-

0)

'a 0

o~

0

0 0 00

0

0 0

0 00

0

a 0 f

0B;00

0

. * . * .

200 400Plasma creatinine (gLmolIl)

Figure 2 Plot showing the correlation between plasmaconcentrations of creatinine and l2G I among 52 patients.The horizontal line marks the upper reference limit for theprotein measurement.

pathology and that the ratio of urinary albu-min:aIM gives the best discrimination betweenpatients with primary glomerular and tubularpathology. Figure 4 also shows that the ratiosof albumin:)2Gl and albumin:a,M among thepatients with obstructive nephropathy andpolycystic kidney disease occupy an inter-mediate position between that of patients withprimary glomerular and primary tubularpathology.

DiscussionThe results reported confirm that manypatients with chronic renal disease primarily ofglomerular origin excrete increased quantitiesof the low molecular weight proteins that areused to detect renal tubular disorders, andestablish that the same applies to f12G1. Theamounts excreted by the patients with plasmacreatinine concentrations above 200 jumol/l aregenerally comparable with those we found inpatients with chronic renal tubular disorders,5and fi2G1 assays do not themselves offersuperior discrimination between conditionsthat primarily affect the glomeruli or thetubules. The relatively low incidence ofincreased f2M excretion found in the patientswith primary glomerular disease almost cer-tainly does not indicate any greater specificty ofthis protein for detecting tubular malfunctionbut rather reflects its instability in acid urine"which can account for appreciable degradationtaking place during the time that the urine isheld in the bladder."2Thus three ofthe patients

U-1 =C I I

I~~i I I v

1 t t I

564

1

t

f


http://jcp.bmj.com

/J C


ownloaded from

http://jcp.bmj.com/


r = 0-36

I 0

£

0

I 0ma

__!

- Aa0,;boc%-Ia,IIIII

a

-Ja a:

ac

a a a -

Oa 0._

._CL

.0

ir

1 10 100 1000

' r = 0-S9

II

0 AO 0 ,p!.

~I:a

a

a

a

a aa

I O1 1

w-.> ..........11110-1 1 10 100 1W0

100I

10.

-i

D 101

00

0CE 1(Ea00

06AI0

1 r = 0.31

I

a

I 0

D

a

a&0O o

aI

Oa

0-10

0

aa

a

a0

oa

O a0 *

aa

a

ao aa

a

aa

aa al

a

a

1 10 100 1000

r = 0-55

a

00

1 a a a 0t F ad ; ..0-1

mao O

a 0 a

1 10 100 1000Albumin MURL

Figure 3 Plots showing the correlation between urinary excretion of albumin and f)2G 1, RBP1 a,-M, and NAG among37 patients with primary glomerular pathology (O), 13 with obstructive nephropathy (0), and eight with polycystickidney disease (A), all with plasma creatinine concentrations below 200 pmol/l. The lines mark the upper referencelimits. All results are plotted as multiples of the upper reference limit (MURL) on logarithmic scales.

with a plasma creatinine concentration exceed-ing 200 1umol/l and an apparently normal l2Mexcretion had provided urine specimens ofpH 5-6. The infrequency of finding a raised/J2M excretion supports the case for droppingthis measurement in the assessment of renalfunction'3 because of the impractability ofroutinely administering alkali before collectingurine specimens.Our findings suggest that more than one

mechanism may be responsible for theincreased excretion of the constituents meas-ured. It has previously been established that inpatients with renal failure the plasma concen-trations of RBP, aiM, and fl2M areincreased,'4-'8 and assay of serum low molec-ular weight proteins has been proposed todetect and monitor changes of the glomerularfiltration rate.3 '9 The basis of the rise inplasma concentrations is that their main routeof elimination is via the glomeruli, final dis-posal being attributable to catabolism withinthe renal tubule cells. A raised plasma concen-tration of the free form of these proteins can beexpected to result in their increased filtrationby glomeruli that remain functional and this inturn could lead to saturation of the tubularreabsorptive capacity of such intact nephronsand consequent increased urinary loss. Ourfindings relating the excretion of RBP, aiMand fl2M to plasma creatinine concentrationssupport earlier suggestions that such a mecha-nism might account for the increased lowmolecular weight protein output found whenthe plasma creatinine concentration exceeds200 Pumol/1.215

Patients with chronic glomerulonephritis,

chronic pyelonephritis, and systemic lupuserythematosus have been reported as havingincreased serum fi2G1 concentrations,20 butwe have now shown that when the plasmacreatinine is raised the plasma /32G1 concentra-tion almost always exceeds the upper limit ofthe reference range. The exceptions seen in fig2 might well be heterozygotes for a fl2G1 genewhich depresses the plasma concentration andis known to occur with a prevalence of about5% in the caucasian population.2' This finding,together with our observation that increasedurinary output of fJ2G1 consistently occurswhen the plasma creatinine exceeds 200 pmol/1, suggest that it is handled by the kidney in asimilar way to the low molecular weight pro-teins. It might not be expected to be filtered soreadily because it has a molecular mass of 50kilodaltons,22 but it has a range of isoelectricpoints which are relatively high and so it isprobably retarded less than more anionicproteins by the glomerular charge barrier.6The increase of urinary excretion of low

molecular weight proteins and JJ2G1 found insome of our subjects with plasma creatinineconcentrations within the reference range isunlikely to be explained by an overflow mecha-nism. One possibility is that there might becompetition from albumin and other proteinsfor reabsorption by the proximal tubule cells.In rats Bernard et al have clearly shown thatsmall and large proteins can compete for renaluptake and have postulated that there arecommon reabsorption sites for which proteinsexhibit different affinities depending on theircharge, size, and configuration.23 However,Mutti et al found no relation between albumi-

100l

10j

-J

c0

._

2

*00

CL4I

0.1,0.1

100

10

-J

C

.00cDm010L.

1

0.1

Albumin MURL

IWA.

* - -- *I I I - -- -l * *

565

0-1

1-


http://jcp.bmj.com

/J C


ownloaded from

http://jcp.bmj.com/


100 000'

WI'

cm

._D

.c

10 000

1000

100

10

1

0-1

1 000 0001

100 000

X 10 000-

*E 1000

D 100'

0

c 10'.r 1

1 1

E

.0

0

C_

v1

100000

10 000

1000

1001

101

1I

0.1 '

a

8

o 8 10

GI 01 P1 TI

0

0

0

I 0

8ao

Gl 01 P1 Ti

~i0

-3-.8*

a 0 8 0

l

Gl 01 P1 Ti

0

a

0

8

G2 02 P2 T2

I

0

0

t,0

1-

0

o

0

0

G2 02 P2 T2

0

41-8

8 fgo

0

-.- 0

0

G2 02 P2 T2

Figure 4 Ratios plotted on logarithmic scales of theexcretion of albumin to f2G1, to RBP and to a,M amongpatients with plasma creatinine concentrations under (Gl,01, Pl, Ti) and over 200 ,umolll (G2, 02, P2, T2).Groups Gl and G2 comprise 36 and 14 patients withprimary glomerular pathology; groups 01 and 02, 11and 10 patients with obstructive nephropathy; groups Pland P2, eight and seven with polycystic kidney disease;and groups Tl and 72, 42 and 12 with renal tubularmalfunction. Horizontal lines indicate mean values.

nuria and low molecular weight proteinuriaamong diabetic patients,24 and recently Ber-nard et al, studying similar patients, producedevidence that suggested that there are siteswhere albumin selectively competes with Pro-tein-i but not RBP.25An alternative explanation for the increased

loss of low molecular weight proteins and fl2GIamong the patients with primary glomerularpathology and a plasma creatinine concentra-tion below 200 ,umol/l is that it results fromdiminution in the ability ofthe proximal tubulecells to bind and incorporate these proteinsinto endocytic vesicles or to hydrolyse themwithin lysosomes. The increase in NAG excre-

tion found in so many of these patients wouldbe consistent with the disease process havingproduced injury to the proximal tubule cells.Increased glomerular filtration of protein byitself can seemingly cause tubular pathology,because patients with nephrotic syndrome due

to minimal change glomerulonephritis havelarge lipid and protein laden vacuoles withintheir proximal tubule cells. The correlation weobserved between albumin excretion and NAGloss would be consistent with proteinuriacausing tubular injury but could also resultfrom the primary disease process in otherways.Our findings regarding NAG excretion are

as expected: increased urinary NAG has pre-viously been reported among patients withvarious forms of renal disease,4 26 27 diabeteswith microalbuminuria,28 and essential hyper-tension.29The conventional view would be thatthe increase of NAG represents enzyme thathas escaped from the lysosomes of damagedproximal renal tubule cells. In those patientswith considerable albuminuria the increasecould conceivably represent escape of the 140kilodalton enzyme into the glomerular filtrate,particularly in those with diabetes or athero-sclerosis in whom increased plasma con-centrations may have been present.30 However,the demonstration that the isoenzyme patternin the urine in glomerulonephritis differs fromthat of the plasma,4 suggests that NAG excre-tion reflected the direct escape of enzyme fromtubule cells injured by the primary diseaseprocess.The ratios of the excretion of albumin to that

of the other proteins measured provided themost effective means of differentiating betweenpatients whose renal disease was of glomerularorigin and those where tubular disease wasprimary. As would be expected, high ratioswere found in the former and low ratios in thelatter. The ratios of albumin:f12G1 gave betterdiscrimination than the ratios of albumin:RBP; albumin:fl2Gl ratios of less than 40 wereonly found in patients with primary tubulardisease and ratios greater than 1000 were onlyfound in those with primary glomerularpathology. The best discrimination with leastoverlap was given by ratios of albumin:a,M;ratios less than 17 and greater than 160correlated exactly with the primary site ofpathology in 61 of the 78 (81%) patients witha plasma creatinine of less than 200,pmol/l,and in 19 of 26 (73%) with a plasma creatinineconcentration above 200 ,mol/l. Ratiosbetween the discriminating values may point tosome reduction in function in both areas of thenephron and in this context it is of interest thatnearly all the ratios in the patients withobstructive nephropathy and polycystic kidneydisease were included within the zone ofoverlap, because this corresponds with what isknown of the development of these conditions.Finding overlap in the ratios of those classifiedas having primary glomerular or tubularpathology fits with the concept that regardlessof where the pathological process starts otherparts of the nephron are liable to be affected.

This work was supported by a grant from the Locally OrganisedResearch Scheme of the North East Thames Regional HealthAuthority.We also thank DrMA Mansell, ProfessorG H Nield,and Dr F D Thompson for permission to study their patients.

{l-] I 0

0% a I

566

cc


http://jcp.bmj.com

/J C


ownloaded from

http://jcp.bmj.com/


1 Maack T, Johnson V, Kau ST, Figueiredo J, Sigulem D.Renal filtration, transport, and metabolism of low-molecular-weight proteins: a review. Kidney Int1979;16:251-70.

2 Bernard A, Vyskocyl A, Mahieu P, Lauwerys R. Effect ofrenal insufficiency on the concentration of free retinol-binding protein in urine and serum. Clin Chim Acta1988;171:85-94.

3 ItohY, KawaiT. Human a -microglobulin: its measurementand clinical significance. J Clin Lab Anal 1990;4:376-84.

4 Hultberg B, Ravnskov U. The excretion of N-acetyl-16-glucosaminidase in glomerulonephritis. Clin Nephrol1981;15:33-8.

5 Lapsley M, Sansom PA, Marlow CT, Flynn FV, NordenAGW. fl2-glycoprotein-l (apolipoprotein H) excretion inchronic renal tubular disorders: a comparison with otherprotein markers of tubular malfunction. J Clin Pathol199 1;44:812-6.

6 Norden AGW, Fulcher LM, Lapsley M, Flynn FV. Excre-tion of fl2-glycoprotein-I (apolipoprotein H) in renaltubular disease. Clin Chem 1991;37:74-7.

7 Sansom PA, Marlow CT, Lapsley M, Flynn FV. A sandwichenzyme-linked immunosorbent assay for 112-glycoprotein1. Ann Clin Biochem 1991;28:283-9.

8 Topping MD, Forster HW, Dolman C, Luczynska CM,Bernard AM. Measurement of urinary retinol-bindingprotein by enzyme-linked immunosorbent assay, and itsapplication to detection of tubular proteinuria. Clin Chem1986;32: 1863-6.

9 Takagi K, Koyamaishi Y, Itoh Y, Enomoto H, Kawai T.Enzyme-immunoassay of human a,-microglobulin. Jpn JClin Chem 1981;10:30-9.

10 Yuen CT, Kind PRN, Price RG, Praill PFG, RichardsonAC. Colorimetric assay for N-acetyl-fl-D-glucosamin-idase (NAG) in pathological urine using the o-nitrostyrylsubstrate: the development of a kit and the comparisonof manual procedure with the automated fluorimetricmethod. Ann Clin Biochem 1984;21:295-300.

11 Evrin P-E, Wibeli L. The serum levels and urinary excretionof f, -microglobulin in apparently healthy subjects. ScandJ Clin Lab Invest 1972;29:69-74.

12 Davey PG, Gosling P. fl2-microglobulin instability in path-ological urine. Clin Chem 1982;28:1330-3.

13 Flynn FV. Assessment of renal function: selected develop-ments. Clin Biochem 1990;23:49-54.

14 Scarpioni L, Dall'Aglio PP, Poisetti PG, Buzio C. Retinolbinding protein in serum and in urine of glomerular andtubular nephropathies. Clin Chim Acta 1976;68: 107-13.

15 Yu H, Yanagisawa Y, Forbes MA, Cooper EH, CrocksonRA, MacLennan ICM. Alpha-l-microglobulin: an indica-tor protein for renal tubular function. J Clin Pathol1983;36:253-9.

16 Cooper EH, Forbes MA, Crockson MA, MacLennan ICM.

Proximal renal tubular function in myelomatosis:observations from the fourth Medical Research Counciltrial. J Clin Pathol 1984;37:852-8.

17 Schardijn GHC, Statius van Eps LW. / -microglobulin: itssignificance in the evaluation of renal function. Kidney Int1987;32:635-41.

18 Wibell L, Evrin P-E, Berggard I. Serum fl2-microglobulin inrenal disease. Nephron 1973;10:320-31.

19 Jung K, Schulze B-D, Sydow K, Pergande M, Precht K,Schreiber G. Diagnostic value of low-molecular massproteins in serum for the detection of reduced glomerularfiltration rate. J Clin Chem Clin Biochem 1987;25:499-503.

20 Cohnen G. Immunochemical quantitation of /1 -glycopro-tein 1 in various diseases. J Lab Clin Mfed 1970;75:212-6.

21 Sepehrnia B, Kamboh MI, Adams-Campbell LL, et al.Genetic studies of human apolipoproteins VIII. Role ofthe apolipoprotein H polymorphism in relation to serumlipoprotein concentration. Hum Genet 1989;82:118-22.

22 Schousboe I. Purification, characterisation and identifica-tion of an agglutinin in human serum. Biochem BiophysActa 1979;579:396-408.

23 Bernard A, Viau C, Ouled A, Lauwerys R. Competitionbetween low- and high-molecular-weight proteins forrenal tubular uptake. Nephron 1987;45:115-8.

24 Mutti A, Alinovi R, Ghiggeri GM, et al. Urinary excretion ofbrush-border antigen and plasma proteins in early stagesof diabetic nephropathy. Clin Chim Acta 1990;188:93-100.

25 Bernard AM, Lauwerys RR, Noel A, Vandeleene B,Lambert A. Urine Protein 1: a sex-dependent marker oftubular or glomerular dysfunction. Clin Chem 1989;35:2141-2.

26 Sherman RL, Drayer DE, Leyland-Jones BR, ReidenbergMM. N-acetyl-f-glucosaminidase and 112-microglobulin.Their urinary excretion in patients with renal paren-chymal disease. Arch Intern Med 1983;143:1183-5.

27 Swedenborg P, Hultberg B, Thysell H. Urinary 1)-hex-osaminidase excretion in polycystic kidney disease. ActaMed Scand 1981;210:471-3.

28 Gibb DM, Tomlinson PA, Dalton NR, Turner C, Shah V,Barratt TM. Renal tubular proteinuria and microal-buminuria in diabetic patients. Arch Dis Child1989;64:129-34.

29 Alderman MH, Melcher L, Drayer DE, Reidenberg MM.Increased excretion of urinary N-acetyl-/i-glucosamin-idase in essential hypertension and its decline withantihypertensive therapy. N Engi J Med 1983;309:1213-7.

30 Belfiore F, Napoli E, Vecchio LL, Rabuazzo AM. Increasedserum activity of beta-n-acetyl-glucosaminidase in ath-erosclerosis. Am J Med Sci 1974;268:235-9.

567


http://jcp.bmj.com

/J C


ownloaded from

http://jcp.bmj.com/

Documents

J Papers - From BMJ and ACP · included were six patients with overt diabetic nephropathy, three with polyarteritis nodosa, and single cases ofWegener's granulomatosis, systemic lupus