Evaluation of urinary proteins by sodium dodecyl sulphate polyacrylamide disc gel electrophoresis and molecular mass analysis

Clinica Chimica Actu, 128 (1983) 151-167 Elsevier Biomedical Press

151

CCA 2413

Evaluation of urinary proteins by sodium dodecyl sulphate polyacrylamide disc gel electrophoresis and molecular mass analysis J. Lubega * Department

of Clinical Biochemistry,

(Received

University of Cambridge, Addenbrookes Cambridge CB2 2QR (UK) July Ist, revision October

Hospital, Hills Road,

22nd. 1982)

Summary Patterns of urinary proteins were examined on sodium dodecyl sulphate polyacrylamide gels and compared to those obtained using high voltage agarose electrophoresis. Both were found to be well adapted to routine laboratory analysis although the former had more sensitivity for early renal changes. Glomerular, mixed and tubular patterns were identifiable, and special precautions and pitfalls whilst interpreting the latter were pointed out. Restricted (Y, antitrypsin was shown in renal transplant proteinuria and its possible mechanisms were discussed. Molecular masses and frequencies of several discrete bands in heavy proteinurics were calculated, and attempts to identify them were made with reference to standard proteins.

Introduction From a qualitative demonstration of albumin as the coagulable material in urine by the classical heat test [l] and Bence Jones proteins [2] there has developed an overall analysis of urine protein composition by electrophoretic separation on various media. Paper electrophoresis with poor resolution coupled with protein trailing and adsorption [3] has given rise to starch gel [4], cellogel and finally agarose as separation media. A still better technique of Norman Anderson’s two-dimensional iso-Dalt system [5] with enormous resolution has revolutionalised protein separation in spite of its attendant difficulties in identifying a multiplicity of proteins spots. By contrast, quantitative assessment involves endogenous and exogenous protein clearances in order to construct regression lines the slopes of which are the selectivity indices of the glomerulus [6,7]. Endogenous studies utilise specific plasma proteins

* Present address for reprints: Dr. J. Lubega, Infirmary, Leicester LEI SWW, UK.

0009-898 l/83/0000-0000/$03.00

Department

0 1983 Elsevier Biomedical

of Chemical

Press

Pathology,

L&ester

Royal

152

[S], whereas polyvinyl pyrrolidone (PVP) [9], haemoglobin and labeiled albumin constitute some of the exogenous tests. Clearances are labour-intensive and unsuitable for routine analysis, and substances like PVP are potential inducers of renal failure. Sodium dodecyl sulphate polyacrylamide gel disc electrophoresis (SDSPAGE) is well suited to large batch analysis and has high resolution, sensitivity, and reproducibility [ 101. SDS-PAGE distinguishes between glomerular, tubular and mixed types of proteinuria in addition to molecular mass calculation of bands when suitable protein calibrators are used [ 111. When biopsy taking is difficult such as in children and the elderly, SDS-PAGE profiles could elucidate the predominant renal lesion. In this report such SDS-PAGE profiles have been studied and compared to those of the high voltage agarose electrophoresis (HVAE). Emphasis has been placed on inte~retation of these patterns and pointing out some pitfalls. The semi-quantitative parameters such as glomerular tubular protein ratio (GTPR) and ratio of glomerular protein (RGP) obtained by densitometric scanning of SDS-PAGE gels as proposed by Boesken et al [12j have also been re-evaluated. The GTPR classified proteinuria into glomerular and tubular types and was tested to determine if it could also segregate mixed types. The RGP ratio, which is supposed to depict the selectivity within the glomerular group, was compared to the conventional differential protein clearance (DPC) as measured by the IgG and transferrin clearances. The phenomenon of restricted cy, antitrypsin excretion in renal transplant proteinuria was pointed out together with its possible mechanisms. Furthermore, the importance and relative frequency of various molecular masses in heavy proteinuria were investigated. Materials and methods patients attending the renal outpatient and those on the renal ward were studied. The included nephrotics, transplanted cases, diabetics, and others with evidence of heavy proteinuria. Samples, from 24-h urine collections, were concentrated in Amicon chambers (Amicon Ltd., High Wycombe, Bucks, UK) in order to get 100-200 @g of protein per 100 ~1 of sample. Samples with l-2 g/24 h and above of protein were applied neat on SDS-PAGE, but concentrated for HVAE. Drops of sodium azide at a concentration of 0.2 g/l preserved the original urine samples. Serum samples were obtained at the end of the 24-h collection. All samples were stored at - 20°C until analysis. High voltage agarose e~ectr~phore~~s (HVAE). A 10 g/l agarose was dissolved in 100 ml of 0.1 mol/l barbital buffer (pH 8.6) and then boiled. This was poured on to pre-heated 11 X 20.5 x 0.1 cm plates (Boehring Diagnostics, Hounslow, Middlesex, UK). A slit former was overlaid until the gel set. 3 ~1 of urine concentrated 25-100 times were electrophoresed at 100 mA constant current at 250 mV for 55 min in a Shandon electrophoretic tank (Shandon Southern Products, Runcorn, Cheshire,

153

UK), filled with the same buffer. The gel was fixed in picric acid/acetic acid (acetic acid/water, 1 : 3, v/v, saturated with picric acid), pressed, dried and stained with 0.5% Coomassie Blue R250 and destained with methanol/water/acetic acid (5 : 5 : 1, v/v/v). The DPC was carried out by the immunochemical technique of Laurel1 [ 131 for measuring IgG and transferrin in serum and urine in a 10 g/l agarose gel, containing commercially obtained monospecific rabbit anti-human IgG and anti-transferrin sera, (Dakopatts, Mercia Broacades Ltd., Weybridge, Surrey, UK). Dividing the IgG clearance by that of transferrin gave the DPC. SDS-PAGE was a modification of the method of Weber et al [ 141. Briefly the tank buffer was made by dissolving 7.8 g of NaH,PO, .2 H,O, 38.6 g of Na,HPO, . 7 H,O and 2.0 g of sodium dodecyl sulphate (SDS) in a litre of distilled water to give 0.2 mol/l phosphate buffer (pH 7.2) and 2 g/l SDS. The gel buffer was made by diluting the tank buffer with distilled water (1 : 1, v/v) and the sample buffer was made by diluting tank buffer 1 : 20 with 1% SDS. A 100 g/l gel was prepared by adding 22.2 g of acrylamide to 0.6 g of N, N’methylene bisacrylamide in 100 ml distilled water. To 15 ml of gel buffer, 13.5 ml of 100 g/l acrylamide solution, previously de-aerated, were added, followed by 1.5 ml of freshly prepared 15 g/l ammonium per sulphate and the mixture boiled to 1OO’C for 5 min and cooled. 40 ~1 of N, N, N’, N’-tetramethyl ethyldiamine (TEMED) were added. Twelve gels were cast in 75 x 6.0 mm glass tubes with bottoms sealed with 3-4 layers of Nescofilm, which were removed after the gel polymerised. 20-50 yl of distilled water were layered on top of each gel. To 100 ~1 of concentrated or 10 ~1 of neat urines a drop of glycerol was added, followed by a drop of 0.5 g/l bromophenol blue marker. Electrophoresis at 100 mA

L%IMr -

1oJ.200

-

looBO60-

40-

?O-

Fig. 1. Standard curve for reference proteins used on SDS-PAGE.

154

per gel commenced till the dye front reached the bottom of the gel. Gels were fixed and stained in Amido black 10B in 70 ml/l acetic acid for 6 h and destained in acetic acid/methanol/water (7 : 10 : 100, v/v/v) overnight. Densitometric quantitation. Each gel was scanned using the CAMAG 80100 densitometer (CAMAG AB, Muttenz, Switzerland) with an integration unit and recorder. Transmission was set with a 595 nm filter, slit of 0.2 x 10 mm, scanning speed of 3 cm/min and chart speed of 12 cm/min. Molecular mass analysis was based on band mobilities relative to the dye front and then constructing a calibration curve by plotting log M, versus mobility of the standard proteins as shown in Fig. 1. Glomerular tubular protein ratio (GTPR)

the area above by that below the central

was calculated from the scans by dividing albumin band as shown in Fig. 6b.

Ratio of glomerular

the albumin band; between transferrin

proteins (RGP) was obtained by subdividing the area above namely the area above transferrin (M, 90000) divided by that and the central albumin (M, 67 000) as shown in Fig. 6b.

Results Fig. 2 shows typical normal, mixed glomerular-tubular, and tubular SDS-PAGE appearances, and more mixed patterns are shown in Fig. 3, Nos. 1, 2, 4, 5, and 7. Pure glomerular lesions with bands only in the upper half of the gels are shown in Nos. 3 and 6. No. 5 is tubular proteinuria with bands below the central albumin band. The concordance between HVAE and SDS-PAGE based on Fig. 3 is given in Table I. By comparison Fig. 4 shows typical HVAE of glomerular and tubular patterns. So often a drop in the glomerular filtration rate (GFR) can produce an overflow type of proteinuria due to an increased filtered protein load, which could saturate the reabsorptive mechanism to produce a tubular pattern on these gels, but with no real renal structural or functional anomaly. Measuring the GFR could indicate the source of this electrophoretic ambiguity. However, primary glomerular-tubular disease could in itself decrease the total number of effective nephrons with a net fall in GFR, which confuses the picture even further. Consequently, the real and apparent tubular proteinuria could become difficult to distinguish without resorting to a biopsy or other ancillary tests of renal tubular function such as urinary electrolyte status, amino acids and acidification tests. Fig. 5 shows an HVAE mixed pattern of proteinuria from renal transplants with marked absence of (Y, antitrypsin (M, 54 000) despite excretion of higher molecular masses such as albumin (M, 67 000), transferrin M, 90 000) and IgG (M, 150 000). This pattern has been observed in many transplant urines and possible mechanisms of this restricted (Y, antitrypsin excretion are being investigated in a separate study, but could include electrical repulsion by the negatively charged sialoglycoprotein

Fig. 2. Typical appearances on SDS-PAGE for: (left to right) normal, with only a central albumin band; mixed, with glomerular in the upper and tubular in the lower halves respectively; and tubular, with bands appearing only in the lower half of the gel.

glomerular basement membrane, immunosuppressive therapy or (Y, antitrypsin sequestration or localisation within the glomerulus so as to neutralise proteolysis due to the chronic inflammatory tendency of graft rejection. Densitometric sketches of scans are shown in Fig. 6 with glomerular in (a), mixed

156

Fig. 3. Various

SDS-PAGE

patterns

of proteinuria

in seven patients.

Gels are 1-7, from left to right.

glomerulo-tubular (b), and pure tubular (c) proteinuria. Table II shows data on 18 predominantly glomerular and five purely tubular proteinuria patients with clinical information, GTPR and RGP ratios, 24-h protein, SDS-PAGE appearances, and selectivity indices. The distinction between glomerular and tubular proteinuria by the GTPR amongst

qAT

Fig. 4. High voltage agarose electrophoretic appearances of glomerular of proteinuria. Alb, albumin; cu,AT, a, antitrypsin; Me, microglobulins;

(left) and tubular (right) patterns Tf, transferrin; yG, y-globulins.

23 patients in Table II is shown in Fig. 7. Values greater or less than unity were indicative of glomerular and tubular proteinuria respectively, which seem to confirm a previous report by Boesken et al ]12]. However, GTPR showed no particular values to distinguish the 11 mixed proteinurics from glomerular cases. Consequently this parameter cannot be useful in delineating mixed lesions. Out of 18 patients, ten bad unselective proteinuria with a DPC ratio > 0.2 and

158 TABLE

1

CORRELATION BETWEEN SHOWN IN FIG. 3

AGAROSE

AND

SDS-PAGE

ELECTROPHORESES

electrophoresis

SDS-PAGE

FOR

GELS

pattern

Gel

Clinical diagnosis

Agarose

1. 2. 3. 4.

glomerulo-tubular glomerulo-tubular glomerulo-tubular glomerulo-tubular

glomerulo-tubular glomerulo-tubular glomerulo-tubular glomerulo-tubular

5.

Renal transplant Renal transplant Chronic renal rejection Pyelonephritis with obstructive uropathy Acute monocytic leukaemia

tubular

only

6.

Myelomatosis

tubular

only

7.

Myelomatosis

tubular with lysozyme excretion tubular with intact IgG excretion glomerulo-tubular

glomerulo-tubular

Fig. 5. Agarose electrophoretic patterns in five renal transplant urines with absence of (I, antitrypsin band in a mixed type of proteinuria. az microglobulins are suggestive of tubular involvement. Alb, albumin; a, AT, a, antitrypsin; a2 MC, a2 microglobulins; Tf, transferrin; &M, & microglobulin; fg, fibrinogen; R, reference; y G, y-globulins.

159

a

bc

d

Fig. 6a. A sketch of scanning appearances and SDS-PAGE in a glomerular type of proteinuria. Peaks at a, b, c, d, and e represent IgG (M, 150000). polymeric albumin (M, 140000), polymeric albumin (M, 130000). transferrin (M, 90000) and central albumin (M, 67000), respectively.

the RGP range of 0.27-6.0, whereas the remaining eight selective proteinurics (DPC < 0.2) had a RGP range of 0.1 l-52. There was such a big overlap of RGP values between selective and unselective proteinuria, contrary to Boesken et al’s suggestion that unselective hyperacute glomerular nephritides were characterised by RGP values of < 2, whereas unselective chronic glomerular nephritides had RGPs between 2-20 and those with minimal change nephritides had values > 25. RGP showed poor correlation with DPC, thus rendering the latter a simpler and better determinant of selectivity. Molecular mass analysis of urine samples from patients with heavy proteinuria examined on SDS-PAGE is shown in Table III with the aim of determining the frequency of specific protein excretion. Albumin had loo%, transferrin 83%, IgG 73.3%, &’ microglobulin 73%, polymeric albumin 56%, cy, antitrypsin 50%, (Y, acid glycoprotein 46.4%, and a; microglobulin 44%, & microglobulin 46.6% and lysozyme 20%. The 14 patients with CQ microglobulin excretion included four renal transplants, two with myeloma and eight with chronic renal failure, all of whom showed tubular

160

Fig. 6b. A scanning and SDS-PAGE pattern in a mixed type of proteinuria. The ratios of A : B and C : D depict the GTPR for type and RGP for the selectivity of proteinuria, respectively. Notice that the central albumin band area is excluded in the calculation.

patterns of proteinuria. The two types of (Ye microglobulins corresponded to M, 19000 for CY;and M, 23 500 for CY;’subtypes, respectively. The former was approximately half as frequent as the latter. The two (Ye microglobulins were probably identical to those observed by Davis et al [ 151 in tubular proteinuria and eluting on Sephadex G,,; the early peak at M, 19 000 and the late one at M, 23 500. Albumin had the highest frequency of 100% and its glomerular passage seems to be slightly hindered. However, without an effective reabsorptive mechanism as much as 40 g per day would be lost [ 161. Molecular masses higher than albumin’s are excluded from the normal urine except in glomerular disease [ 17,181. The six cases (20%) with lysozyme (M, 15 000) had marked tubular involvement. Lysozyme present in trace amounts in normal urine increases substantially in tubular dysfunction [19] and Hayslett et al [20] correlated lysozymuria with histopathological diagnosis when its plasma threshold of 45 pg/l is exceeded. & Microglobulin at 46.6% was probably a better detector of tubular dysfunction [21]. Molecular masses of 10000, 30000 and 33 000 were not referable to any standard reference proteins. Bernier and Putnam [22] proved that molecular masses 22000

161

Fig. 6c. A sketch of scanning

appearances

and SDS-PAGE

in a tubular

type of proteinuria.

and 44000 were mono and dimeric forms of Bence Jones protein, respectively. Intermediate molecular masses were due to polymerisation and or dissociation. Davis et al [ 151 also found Bence Jones proteins anomalously eluting as though they had molecular masses of 53 000 and 3 1000 and or 15 000, due to dissociation of tetrameric and dimeric forms of Bence Jones protein. It is therefore probable that the above observed molecular masses were dissociated forms of Bence Jones proteins. IgG at 73.3% and transferrin at 83.3% were abundant and form the basis for selectivity of proteinuria determination [23]. Molecular masses of 130 000 at 44% and 140000 at 36% could represent polymeric albumin. These urine specimens were kept at -20°C which, according to Hardwicke et al [24], enhances dimerisation of albumin (Kistler regards this as a post glomerular event [25], and these polymers disappear completely from gel profiles upon treatment [24]. The combined total of 30 patients with these presumed molecular masses responded to treatment, which suggests polymeric albumin is a prognostic marker. The molecular mass of approximately 185 000 in 12 patients (40%) was assigned

II

with

Cryoglobulinaemia nephritis

9.

7. 8.

6.

3. 4. 5.

2.

Nephrotic syndrome Nephrotic syndrome proliferative nephritis Nephrotic syndrome Nephrotic syndrome Membranous proliferative glomerulo nephritis Nephrotic syndrome with chronic renal failure SLE nephrotic syndrome Heavy proteinuria

Diagnosis

1.

NO.

_

5.3 8.0

5.4

5.3 3.2 11.3

9.4 5.7

(g/24

h)

Proteinuria

THE DISCRIMINANT FUNCTION OF THE GTPR SELECTIVITIES BY THE RGP IN COMPARISON PROTEIN EXCRETION

TABLE BETWEEN TO THE

9.3

4.6 12.3

3.2

3.7 5.0 3.9

4.9 3.6

GTPR

0.34

0.20 0.20

2.00 2.90

0.27

mixed

0.73 0.44

glomerular mixed (predominantly glomerular) mixed (predominantly glomerular)

glomerular glomerular mixed

glomerular mixed

non-selective

non-selective non-selective

non-selective

non-selective non-selective non-selective

non-selective selective

Selectivity

AND THE DELINEATION OF CLINICAL DIAGNOSES AND

Disc electrophoretic pattern

0.20 0.78 0.20

2.31 1.90 0.88

0.26 0.06

IgG/Tf ratio

AND TUBULAR PROTEINURIA IN PATIENTS WITH VARIOUS

1.73 0.38

RGP

GLOMERULAR IgG/Tf RATIO

Membranous proliferative giomeruio nephritis Nephrotic syndrome Lupoid nephritis Renal transplant with chronic rejection Pyelonephritis with obstructive uropathy Chronic myeloid leukaemia

13. 14.

15.

21. 22. 23.

20.

19.

16. 17. 18.

Nephrotic Nephrotic

Il. 12.

Renal transplant Renal transplant Accelerated hypertension with chronic renal failure

syndrome syndrome

Nephrotic syndrome with haematuria Minimal change, nephritis Nephrotic syndrome

10.

0.4 0.8 0.4

0.3

0.2

0.4 0.2 3.0

1.0

4.6

5.3 0.2

21.2 1.4

4.0

_

_

0.39

0.13 0.05

1.25 0.11 6.00

8.4

11.0 0.18

0.04 0.17

9.90 3.10

11.6 12.0

5.4 9.3 3.80

0.08 0.04

52.00 0.90

3.8 36.0

3.0

0.52

4.60

12.3

10.0

tubular with muramidase excretion tubular tubular tubular

tubular

mixed

glomerular mixed (predominantly glomerular) glomerular predominantly glomerular predominantly glomerular mixed mixed

gtomerular

selective selective (borderline) selective (borderline) selective selective non-selective

non-selective non-selective

non-selective

164

.

20 -

lo8-

.

’

t

l

:

6-

.

. . . . .

GLOMERULAR t -

TUBULAR T SDS -PAGE -

Fig. 7. The distribution of the glomerular proteinurias as calculated from SDS-PAGE

tubular scans.

protein

ratio

(GTPR)

in glomerular

and

tubular

to apolipoprotein A (apo A) (M, 180 000), which Virella and Lopes-Virella [26] correlated with non-selective DPCs. These subjects had DPCs greater than 0.2, which suggests that apo-A, because of its high molecular mass, only leaks into urine if there is considerable damage.

165 TABLE

III

MOLECULAR Relative mobility related to the dye front

MASS ANALYSIS Frequencyofband (% of cases)

OF 30 PATIENTS Calculated molecular mass

0.95 0.92 0.88 0.81 0.74

6.6% (2) 6.6% (2) 46.6% (14) 20.0% (6) 44.0% (13)

10000

0.67

73.0% (22)

23 500

0.64 0.60 0.57 0.50 0.42 0.35 0.28 0.17 0.14 0.07 0.04

6.6% 23.0% 26.6% 46.6% 50.0% 100.0% 83.0% 44.0% 56.0% 73.3% 40.0%

(2) (7) (8) (14) (15) (30) (25) (13) (17) (22) (12)

URINES

EXAMINED

Molecular mass of standard reference protein

10000

11800 15000 19000

26000 30000 33000 43 000 54000 67000 90000 130000 140000 170000 185000

11800 15000 Sephadex G75 Peak IV late [ 15) (19000) Sephadex G75 Peak IV early [ 151 (23 500) 25 000

44 100 54000 67000 90 000 130000 140000 160000 180000

ON SDS-PAGE

GELS

Protein appearing in urine is likely to be

microglobulin P, Lysozyme e; microglobulin

a;’ microglobulin

chymotrypsinogen _ a, acid glycoprotein a, antitrypsin albumin transferrin polymeric albumin polymeric albumin IgG apolipoprotein

A

Conclusion

Interpreting patterns of urine HVAE and SDS-PAGE demands awareness of some pitfalls, especially with regard to tubular proteinuria. GFR, urine protein, pH, electrolytes, and other tests of tubular function, and finally a biopsy may be required. The phenomenon of restricted (Y, antitrypsin excretion in renal transplant urines was shown on HVAE gels but warrants more studies to elucidate its possible mechanisms. The GTPR could segregate glomerular from tubular defects but not mixed from glomerular patterns. Likewise the RGP was not as good as the DPC in indicating selectivities. Finally the importance of finding some molecular masses such as albumin, microglobulins, lysozyme, polymeric albumin, dissociated Bence Jones polymers and apo-A was briefly discussed.

166

Acknowledgements I wish to thank Dr. D.B. Evans of the Renal Unit of Addenbrookes Hospital, for allowing me access to his patients. Also I am grateful to Professor R.W. Carrel1 of the Clinical Biochemistry Department, University of Otago, New Zealand, for guidance. References 1 Antoine B. The history of albuminuria. Abbott Universal Ltd., Abbottempo Book 2, 1970, 10-13. 2 Lines JG. A chronicle of the development of clinical chemistry. IFCC Newsletter, October 1977. 3 Kunkel H, Slater RJ. Filter paper electrophoresis with special reference to urinary proteins. J Lab Clin Med 1953; 41: 619-631. 4 Butler EA, Flynn FV, Harris H, Robson EB. A study of urine proteins by two-dimensional electrophoresis with special reference to the proteinuria of renal tubular disorders. Clin Chim Acta 1962; 7: 34-41. 5 Anderson NG, Anderson NL, Tollarsen SL. Proteins of human urine, concentration and analysis by two-dimensional electrophoresis. Clin Chem 1979; 35: 1199-1210. 6 Joachim GR, Cameron JS, Schwartz M, Becker EL. Selectivity of protein excretion in patients with the nephrotic syndrome. J Clin Invest 1964; 43: 2332-2346. 7 MacLean PR, Petrie JJB. A comparison of gel filtration and immunodiffusion in the determination of selectivity of proteinuria. Clin Chim Acta 1966; 14: 367-376. 8 Bienenstock J, Poortmans J. Renal clearances of plasma proteins in renal disease. J Lab Clin Med 1970; 75: 297-306. 9 Hulme B, Hardwicke J. The measurement of renal permeability using labelled macromolecules, Proc Roy Sot Med 1966; 59: 509-512. 10 Pesce AJ, Boreisha I, Pollak VE. Rapid differentiation of glomerular and tubular proteinuria by sodium dodecyl sulphate polyacrylamide gel electrophoresis. Clin Chim Acta 1972; 40: 27-34. 11 Weber K, Osborn M. The reliability of molecular weight determination by sodium dodecyl polyacrylamide gel electrophoresis. J Biol Chem 1969; 244: 44064412. 12 Boesken WH, Kopf K, Schollmeyer P. Differentiation of proteinuric disease by disc electrophoretic molecular weight analysis of urinary proteins. Clin Nephrol 1973; 1: 31 l-317. I3 Laurel1 CB. Quantitative estimation of proteins by electrophoresis on agarose gel containing antibodies. Anal Biochem 1966; 15: 43-52. 14 Weber K, Pringle JR, Osborn M. Molecular weight determination on SDS gels. In: Colowick and Kaplan, eds. Methods in Enzymology, Vol. 26. New York: Academic Press, 1972: 3-27. 15 Davis JS, Flynn FV, Platt HS. The characterisation of urine protein by gel filtration, Clin Chim Acta 1968; 21: 357-376. 16 Gregoire F, Malmendier C, Lambert PP. The mechanism of proteinuria and a study of the effects of hormonal therapy in the nephrotic syndrome. Am J Med 1958; 29: 516-531. 17 Brenner BM, Hostetter TH, Humes DH. Molecular basis of proteinuria of glomerular origin. N Engl J Med 1978; 298: 826-833. 18 Pesce AJ, Gaizutis M, Pollak VE. Selectivity of proteinuria; an evaluation of the immunochemical and gel filtration techniques. J Lab Clin Med 1970; 75: 586-606. 19 Harrison JF, Lunt GS, Scott P, Blainey JD. Urinary lysozyme ribonuclease and low molecular weight protein. Lancet 1968; 1: 371-374. 20 Hayslett J, Perillie PE, Finch SC. Urinary muramidase and renal disease correlation with renal histology and implications for mechanism of enzymuria. N Engl J Med 1968; 279: 506-5 12. 21 Johansson BG, Ravnscov U. Serum levels and urinary excretion of a2 microglobulin, & microglobulin and lysozyme in renal disease. Stand J Urol Nephrol 1972; 6: 249-256. 22 Bernier GM, Putnam FW. Polymerism, polymorphism and impurities in Bence Jones proteins. Biochim Biophys Acta 1964; 86: 295-308.

167

23 Cameron JS, Blandford G. A simple assessment selectivity in heavy proteinuria. Lancet 1966; 2: 242-247. 24 Boesken WH, Schindera H, Billingham M, Hardwicke J, White RHR, Williams A. Polymeric albumin in the urine of patients with nephrotic syndrome. Clin Nephrol 1977; 8/3: 395-399. 25 Kistler P. International symposium on the purity of proteins. Basel: S. Karger, 1973. 26 Virella G, Lopes-Virella MFL. New approaches to the study of proteinuria. Clin Chem 1977; 23/9: 1793-1794.