87
Journal of Immunological Methods, 138 (1991) 87-94 © 1991 Elsevier Science Publishers B.V. 0022-1759/91/$03.50 ADONIS 002217599100129T JIM 05878
Determination of urinary Tamm-Horsfall protein by ELISA using a maleimide method for enzyme-antibody conjugation Iwao Uto 1,2, Takashi Ishimatsu 1, Hideo H i r a y a m a 1, Shoichi U e d a 1, Junji Tsuruta 2 and Takeshi K a m b a r a 2 J Department of Urology, and 2 Department of Allergy, Institute for Medical Immunology, Kumamoto University Medical School, Kumamoto, Japan (Received 18 July 1990, revised 7 November 1990, accepted 17 December 1990)
A method for enzyme-antibody conjugation using a new maleimide derivative as coupling reagent has been developed. Since a monomeric conjugate of horseradish peroxidase and Fab' antibody could be readily prepared with high efficiency and reproducibility, the enzyme activity and antigen-binding activity were well preserved and nonspecific staining was greatly reduced. The conjugate is suitable for use in both ELISA procedures and immunohistochemistry. Using both methods we examined the pathophysiological significance of Tamm-Horsfall protein (THP) and the present study describes the ELISA method to quantify urinary THP using the new method with rabbit anti-THP antibody. A low concentration (0.04 M) of urea added to the urine samples increased the linearity of the standard curve and the sensitivity of the assay, permitting the detection of as little as 20 n g / m l THP. Freezing and thawing the urine resulted in variable or lower values of THP concentration. THP concentrations in urine as determined by ELISA were stable for at least one month after - 7 0 °C storage, but not after - 3 0 °C storage. There was no correlation between THP concentrations in 24 h urine samples and the morning urine of the same patient. These results suggest that it is essential to use fresh or - 7 0 °C stored 24 h urine samples with added urea (0.04 M) for the determination of THP concentrations in urine by the present enzyme-antibody conjugation method. The THP concentration in normal 24 h urine of young children was found to be less than 51.8 m g / g Cr. Key words: Tamm-Horsfall protein; ELISA; Urea; Maleimide method
Introduction Correspondence to: I. Uto, Department of Urology, Institute for Medical Immunology, Kumamoto University Medical School, Kumamoto, Japan. Abbreviations: THP, Tamm-Horsfall protein; ELISA, enzyme-linked immunosorbent assay; Cr, creatinine; SDS, sodium dodecyl sulfate; PAGE, polyacrylamide gradient slab gel; PBS, phosphate-buffered saline; VUR, vesicoureteral reflux; HRP, horseradish peroxidase; HS-CHM, N-hydroxysuccinimide ester of N-(4-carboxycyclohexylmethyl) maleimide.
THP is a high molecular weight glycoprotein found in normal urine (Tamm and Horsfall, 1950), and is produced in the ascending limb of Henle and the distal convoluted tubules of the kidney (Schenk et al., 1971). It has been demonstrated that THP is the major component of urinary casts
88 (McQueen, 1962) and triggers calcium phosphate crystal formation and clumping in the process of urinary stone formation (Rose and Sulaiman, 1984). Dulawa et al. (1988) have reported that T H P protects against bacterial infection in the urinary tract by aggregating bacteria. However, its physiologic role and possible significance under pathologic circumstances remain unclear. In order to measure the concentration of urinary T H P in various renal diseases, several methods for the quantitative determination of T H P have been described, including radial immunodiffusion (Mazzuchi et al., 1974; Hartmann et al., 1984; Nakagawa, 1987), electroimmunoassay (Wieslander et al., 1977; Samuell, 1978), radioimmunoassay (Grant and Neuberger, 1973; Dawnay et al., 1980; Lynn and Marshall, 1984; Hunt et al., 1985) and enzyme immunoassay (Hunt et al., 1986; Kjellsson et al., 1987). However, the results have been variable and the clinical application of these methods to indicate renal damage has been unsatisfactory. This is because the assay method, are complicated and the treatment and the sampling of the urine is different for each method. In the present paper, we propose a new ELISA method for the measurement of urinary T H P using a maleimide derivative as a coupling reagent. In addition we have studied the influence of various treatments, storage and sampling methods and have established the optimum assay conditions for the measurement of T H P concentrations in both normal and pathological urine.
Materials and methods
Purification of urinary THP Urinary T H P was prepared by precipitation with 0.58 M NaC1 from pooled samples of urine (10 liters) obtained from healthy adult volunteers, as previously described (Grant and Neuberger, 1973). This precipitate was dissolved in 14 mM Na2HPO4/16 mM NaH2PO 4, p H 6.8, containing 8 M urea, and then applied to Sephacryl S-200 (bed volume 390 ml) equilibrated with the same buffer containing 2 M urea using a flow rate of 15 m l / h at room temperature. The absorbance of fractions was monitored at 280 nm and three protein peaks were obtained. The fractions corre-
sponding to the second major peak were pooled and dialysed against distilled water before being lyophilized. The final product was stored at -70°C.
Preparation of anti-THP antibody Antiserum to human urinary T H P was raised in New Zealand White rabbits. Multiple subcutaneous injections were given of 1 ml emulsion containing equal volumes of T H P (1 m g / m l in water) and complete Freund's adjuvant. Immunization was performed four times in total at 1 week intervals. Blood was taken from the ear 1 week after the last injection, and serum collected. IgG was prepared by salting out of the sera with ( N H 4 ) 2 S O 4 followed by passage through a diethylaminoethyl cellulose column. A F(ab')2 fragment was obtained from the IgG fraction by the pepsin treatment at p H 3.9 for 8 h at 37°C, followed by salting out with 60% saturated ammonium sulfate and by Sephadex G-150 column chromatography as p r e v i o u s l y d e s c r i b e d (Yamamoto and Cochrane, 1982).
Conjugation of anti-THP Fab' fragments with HRP using the maleimide method Anti-THP Fab' was conjugated with H R P by the maleimide method according to the method of Imagawa et al. (1982) and Tsuruta et al. (1985). Briefly, 6 mg of H R P in 1.0 ml of 0.1 M sodium phosphate buffer, p H 7.0, were incubated with 2.4 mg of H S - C H M in 100 t~l of N,N'-dimethylformamide for 20 min at 30 ° C. After centrifugation, the supernatant was applied to a Sephadex G-25 column (1 x 45 cm; bed volume 35 ml) using 0.1 M sodium phosphate buffer at p H 6.0. The fractions corresponding to the first peak were pooled and concentrated to approximately 300 /~1 in a collodion bag at 4 °C. These procedures resulted in the coupling of H R P and CHM (HRP-CHM). The average amount of C H M introduced per HRP molecule was calculated to be from 1.0 to 1.5 with good reproducibility. In order to prepare the Fab' fragment, 500 /~1 of 1-4 m g / m l of F(ab')2 fragment were reduced with 10 mM 2-mercaptoethylamine in 0.1 M sodium phosphate buffer at pH 6.0 for 90 min at 37°C. To remove the remaining nonreactive 2-mercaptoethylamine, the reactant was applied to a Sephadex G-25 column,
89 and the void volume fractions were pooled and concentrated to approximately 300/~1 in a collodion bag at 4 o C. H R P - C H M and the Fab' fragment, approximately 30-50 /~M each, were mixed immediately and incubated for 60 min at 37 ° C or for 20 h at 4 ° C . Then, 2 /~1 of 100 mM 2-mercaptoethylamine were added to the mixture to block the free maleimide base, followed by application to an Ultrogel AcA44 colunm (1.5 x 45 cm; bed volume 80 ml) with 0.1 M sodium phosphate buffer at pH 6.5. Absorbances were measured at 280 and 403 nm for protein concentration and for H R P concentration, respectively. Three protein peaks were obtained and the fractions corresponding to the second major peak (containing monomeric conjugates composed of a single Fab' and a single HRP) were stored in the presence of 0.1% bovine serum albumin and 0.005% merthiolate at 4 ° C and used in subsequent experiments.
Electrophoresis and immunoblotting for determining the purity of THP and the specificity of anti-THP Ab SDS-PAGE (5-20%) was performed according to the method of Laemmli et al. (1970), and Western blotting was performed by a modified method using enzyme-labelled antibody (Towbin et al., 1979).
cubated for 30 min at room temperature. After three final washings, color was developed using o-phenylenediamine dihydrochloride (Wako, 100 /~l/well of 3.7 mM solution in 50 mM disodium phosphate-25 mM citric acid buffer pH 5.0 containing 0.012% H202) for 1 min at room temperature. This reaction time was determined by preliminary experiments and was found to be most reproducible in agreement with Silveria et al. (1986). The reaction was stopped by the addition of 1 M HzSO 4 (100 /~l/well) and absorbance at 492 nm was measured.
Construction of the standard curve A standard curve was established by a modification of the procedure of Hunt et al. (1985) with purified urinary T H P diluted over the range 0.0110 n g / m l in the following buffers. (1) assay buffer (PBS-Tween) alone; (2) assay buffer containing 0.1% SDS; (3) assay buffer containing 0.04 M urea or (4) assay buffer containing 2 M urea.
Urine samples The urine of seven healthy volunteers, aged 31.6 + 2.1 years, was collected into individual sterile bottles for 24 h and then divided into 5 ml samples. The urine samples were assayed immediately or was stored either at - 3 0 ° C or at 70 o C until assay. Urinary T H P concentration were also estimated by the ELISA procedure in 26 children without renal disease (15 males and 11 females, aged 6.3 + 2.4 years) and in patients with different grades of primary vesico ureteral reflux (VUR) (12 males and 12 females, aged 5.0 _+ 2.6 years). Reflux was unilateral in 14 cases. Of these, one had grade V reflux, seven grade IV, two grade III and four grade II. Grading was established according to the international classification (Report of the International Reflux Study Committee, 1981). 24 h urine samples were collected and aliquots stored at - 7 0 ° C for T H P estimation. The values obtained were corrected for the concentration of urinary creatinine in order to determine the urine volume. The values of 24 h urine samples were compared with those of the first urine sample obtained in the morning on the same day. -
THP ELISA procedure Polystyrene microtiter plates (Nunc) were incubated for 1 h at 3 7 ° C with rabbit anti-human T H P IgG (100 /xl/well of 2 /ag/ml in coating buffer: 50 mM carbonate buffer, pH 9.6). After three washings with PBS-Tween (10 mM phosphate buffer containing 150 mM NaC1 and 0.05% Tween 20), the plates were filled with 1% BSA (150/al/well of I%BSA in coating buffer) to block non-specific binding sites, and incubated for 1 h at room temperature. After three further washings, the wells were filled with T H P standard solution or diluted (1/50) urine samples (100/~l/well) and incubated for 1 h at room temperature. After three further washings, the plates were filled with HRP-conjugated anti-THP antibody Fab' (2 /~g/ml in PBS-Tween BSA, 100 ~tl/well) and in-
90
Results
Preparation of THP and specificity of anti THP antibodies Purified T H P showed one major band on SDS-PAGE run under reducing conditions and stained with Coomassie blue. The apparent molecular weight was approximately 100,000. Westernblotting showed that the anti-THP antibody reacted with one major band and two faint bands. The former corresponded to a molecular weight of 100,000, which suggested identity with the major band of purified THP. The other two bands had molecular weights of 300,000 and 400,000, respectively, and were probably trimers and tetramers respectively (Fig. 1).
Construction of the standard curve The addition of high concentrations of urea (2 M) partially inhibited the antigen-antibody reaction on ELISA compared with the standard curve of T H P in buffer alone. However, the addition of
low concentrations of urea (0.04 M) did not prevent the reaction and gave an adequate standard curve with low backgrounds, sufficient linearity and parallelism to that of T H P in buffer alone. In contrast, the addition of 0.1% SDS (Hunt et al., 1985) prevented a reaction in the ELISA and resulted in an inadequate standard curve (Fig. 2). In this assay, the lowest level of detection was 20 n g / m l . When specified amounts of T H P were added to the samples, the recovery rate reached 93.7%. The intra-assay and inter-assay coefficients of variation were 7.0% and 3.7%, respectively.
Effect of centrifugation of urine samples Urine samples were pretreated both with and without centrifugation at 13,000 x g for 10 min to remove cells and precipitants in the urine, and 2 M urea was then added. They were then diluted 1 / 5 0 (final concentration of urea: 0.04 M) and tested in the ELISA procedure. The results are shown in Table I. The urinary
b ii iill I
~¢~ra~ -~trimer
94,. •.~ m o n o m e r 4
67,.
67
4a
,. ,.
43"
..
30,-
i~i/~!~iiiii!ii~i
30"2 0 ,.
}
di!,i iii ~
URINE THP
URINE THP
Fig. 1. Electrophoretic (a) and Western-blotting (b) profiles of urine and purified THP. Urine, normal human urine; THP, standard human THP. Arrows represent positions of standards, which are (from the bottom); soybean trypsin inhibitor, carbonic arthydrase, ovalbumin, bovine serum albumin and phosphorylase b. Purified THP showed one major band and its molecular weight was approximately 100,000. Western-blotting showed that the anti THP antibody reacted with one major band and two faint bands. The former band corresponded to molecular weight of 100,000, and the minor bands to molecular weights of 300,000 and 400,000, respectively.
91 TABLE I E F F E C T OF T R E A T M E N T O F U R I N E SAMPLES W I T H U R E A A N D / O R A B S O R B A N C E D E T E R M I N E D BY ELISA Treatment
T H E E F F E C T S OF C E N T R I F U G A T I O N ON
OD492 nm Urea a + cfg b +
No. 1 2 3 4 5 6 7
0.27 0.21 0.24 0.44 0.21 0.10 0.20
*
+ _
(43.4%) (27.8%) (30.8%) (59.5%) (34.2%) (26.0%) (39.6%)
0.61 0.74 0.76 0.73 0.62 0.40 0.55
0.24 ± 0.10 (37.0 ± 11.5%)
_ + (100%) (100%) (100%) (100%) (100%) (100%) (100%)
0.19 0.15 0.11 0.39 0.10 0.09 0.10
0.63 ± 0.13 (100%)
(31.0%) (19.5%) (15.0%) (53.1%) (16.2%) (21.3%) (16.7%)
0.49 0.53 0.56 0.64 0.51 0.30 0.45
0.16 + 0.11 (24.7 _ 13.6%)
(79.5%) (70.9%) (73.8%) (87.1%) (83.5%) (74.6%) (81.6%)
0.50 ± 0.11 (78.7 + 5.8%)
Key." samples were measured with or without adding 2 M urea a n d / o r centrifugation (cfg). a 2-M urea, bcfg 13,000× g, 10 min, * m e a n + S D .
freezing at - 30 o C and - 70 o C for periods ranging from 2 weeks to 3 months. The T H P concentration of urine stored at - 7 0 ° C was similar to that of T H P in fresh urine when measured within the first month of storage (Table II). On the other hand the values varied widely in the initial period in the case of samples stored at 30 ° C (Table II).
T H P concentrations were considerably decreased by the centrifugation of urine samples independent of the addition of urea. Addition of urea without centrifugation gave the most suitable curve with the highest values and adequate linearity (Fig. 2).
Effect of storage on urine samples
-
The effects of storage temperatures and times were studied in seven urine samples from healthy volunteers. Samples were assayed before and after
Effect of freezing and thawing urine samples Six urine samples from healthy volunteers were
T A B L E II E F F E C T S O F D E L A Y E D S T O R A G E A N D F R E E Z E - T H A W I N G O F U R I N E SAMPLES O N T H E T H P C O N C E N T R A T I O N (#G/ME) a Urine
Fresh
no. 1 2 3 4 5 6 7 Mean ±SD
1.8 4.5 6.2 3.6 6.2 1.9 3.9 4.0 ±1.8
Storage at - 30 o C
Storage at - 70 o C
0.5
1
3 (month)
39.2 8.8 26.1 16.9 31.0 6.0 12.7 20.1 ±12.3
9.0 7.0 14.8 16.1 60.9 13.2 16.1 19.6 ±18.6
18.0 9.0 11.5 15.1 45.1 12.7 24.5 19.4 ±12.4
a All assay was performed in duplicate.
0.5 8.8 5.2 6.2 6.3 14.2 5.2 5.8 7.4 ±3.2
Freeze-thawing
1
3 (month)
1
2
5.8 6.3 5.1 10.9 10.8 7.1 5.1 7.3 ±2.5
19.4 11.0 13.1 12.8 26.2 14.9 15.0 16.1 ±16.1
6.9 13.8 4.0 4.3 5.2 7.6 3.6 4.3 13.1 9.5 5.0 4.8 Notdone 6.3 7.4 ±3.5 ±3.8
3
4
5 (times)
10
10.2 5.0 6.7 4.3 14.2 5.7
8.8 4.3 7.6 5.5 15.0 4.3
9.5 4.5 6.2 6.2 11.9 4.8
13.1 5.0 6.4 5.0 15.5 6.7
7.8 ±3.7
7.6 ±4.1
7.2 ±2.9
8.6 ±4.5
92
E c
1.0
£,J O3
¢-.
0.5
0
03
,< 0.0
........
, .1
.01
........
, 1
........ 10
THP concentration (gg/ml) Fig. 2. Standard curves of THP concentration. Standard curves for THP were prepared in assay buffer alone (o), 0.1% SDS (O), 0.04 M urea (A) or 2 M urea (A). 2 M urea slightly inhibited the antigen-antibody reaction in ELISA. 0.04 M urea did not prevent the reaction, and gave a curve which was similar to that obtained with assay buffer alone. 0.1% SDS interfered with the ELISA reaction. assayed after repeated freezing and thawing (1, 2, 3, 4, 5, and 10 times) by solid C O 2 / a c e t o n e . In four samples T H P levels were stable, but in two samples the levels were increased (Table II).
Difference in THP concentrations in 24 h urine collections and morning urine samples The T H P concentrations of matched 24 h and morning urine samples were c o m p a r e d and no correlation was f o u n d (r = 0.12, Fig. 3).
Clinical study In the normal control group (n = 26), the mean _+ SD for urinary T H P / C r was 19.2 _+ 10.5
loo-
E c
cD ..c
s'0 morning
,~0 unne
(mg/gCr)
Fig. 3. Comparison of THP concentration as measured in 24 h urine and morning urine, samples. The THP concentration in 24 h urine samples from 13 VUR patients was measured and compared with that of each morning urine. There was no correlation between the two measurements (r = 0.12).
m g / g C r . It was f o u n d that this value was 26.4 _+ 12.7 m g / g Cr in the y o u n g children, 2 - 5 years of age, and 13.6 _+ 3.2 m g / g Cr in the older children, 6 - 1 0 years of age. Accordingly, we conclude that the normal value of T H P in y o u n g children is below 51.8 m g / g Cr (mean + 2 SD). This study was extended to the patients with p r i m a r y V U R (n = 4 in grade II, 2 in grade III, 7 in grade IV and 1 in grade V). As a preliminary study, high values of T H P were observed from grade II to grade IV. The mean value for the patients was higher than that for normal children. However, wide variations were observed (mean _+ SD, 60.4 4-45.2 m g / g Cr) and since the numbers in each group were very few, the differences were not statistically significant.
Discussion
Since the first description of T H P in 1950 ( T a m m and Horsfall), m a n y studies have been reported on its localization ( K u m a r et al., 1985; Peach et al., 1988) and physiological function ( R o c h a and Kokko, 1973). It is accepted that T H P is p r o d u c e d and secreted by epithelial cells of distal tubules and the ascending limb of Henle. However, the physiological and pathological role of T H P remains unclear. N u m e r o u s quantitative assays of urinary T H P have been reported (Grant and Neuberger, 1973; Mazzuchi et al., 1974; Wieslander et al., 1977; Samuell, 1978; D a w n a y et al., 1980; H a r t m a n n et al., 1984; H u n t et al., 1985; H u n t et al., 1986; Kjellsson et al., 1987; Nakagawa, 1987) but the values obtained have been remarkably variable. Discrepancies between these previous reports m a y have been due to: (1) the use of m o r n i n g urine instead of 24 h urine samples, (2) differences in the storage conditions used, (3) the use of urea (8 M) as a dissociating agent in the purification of standard T H P , (4) differences in the assay system, i.e., r a d i o i m m u n o a s s a y or E L I S A , or (5) the treatment of the urine samples. H S - C H M was developed by Imagaiva et al. (1982) as a cross-linking reagent for enzyme labelling in enzyme immunoassay. In this method, a single a m i n o group of one H R P molecule is reactive with H S - C H M and allows the introduced maleimide group to react with a single thiol group
93 of the hinge portion of the Fab' fragment. Therefore, one can expect to obtain monomeric conjugates of Fab' and H R P with high efficiency and good reproducibility. In the present study, the conjugate was found to give very satisfactory resuits. Although the anti-THP antibody was a conventional preparation and not immunopurified, the use of a conjugate prepared by the maleimide method gave good reproducibility with negligible background. The standard curves produced on each plate, using prepared THP, showed good reproducibility and demonstrated an adequate linear range from 20 n g / m l to 10 t t g / m l for the measurement of unknown urinary T H P concentrations. In the present study the T H P concentration in 24-h urine samples did not correlate with the T H P levels in morning urine samples study and this suggests that urinary T H P concentrations exhibit circadian variations in vivo. We conclude that 24-h urine sampling is essential for the estimation of urinary THP. In previous studies the urine samples had been stored at between - 20 o C and - 30 ° C. In the present study T H P values of the urine samples stored at - 3 0 ° C showed wide variations. In contrast, the levels were relatively stable at - 7 0 ° C during the first month of storage and the reproducibility and recovery rate of the data were constant. Therefore, urine samples should be stored at - 7 0 o C, and measurements should be performed within the first month of storage. Urinary T H P concentrations were much lower after centrifugation than those values obtained without centrifugation. This phenomenon m a y be due to the removal of aggregated and precipitated T H P and adsorbed T H P on cells a n d / o r crystals in the urine. Therefore, centrifugation of urine samples to remove insoluble materials is not recommended in the measurement of urinary THP. As previously pointed out, it is necessary to add dissociating agents such as urea or SDS for measurement of urinary T H P by radial immunodiffusion and ELISA methods, because T H P easily forms macromolecular complexes. However, treatment with 2M urea or 0.1% SDS inhibited the binding between T H P and anti-THP Ab and the standard curve showed inadequate linearity and
sensitivity. In contrast, the presence of 0.04 M urea in the urine samples resulted in suitable linearity and sensitivity without any inhibition of the T H P and anti-THP Ab reaction. It is therefore suggested that 0.04 M urea should be added to the urine samples for measurements of urinary T H P excretion. In the present study, the range of urinary T H P in normal children was below 51.8 m g / g Cr and the value showed a tendency to decrease with age. This result is consistent with the report of Hartm a n n et al. (1984) in which the range was below 53 m g / g Cr as measured by the radial immunodiffusion method. These authors also found that urinary T H P tended to decrease with age. In a preliminary study in patients with VUR, urinary T H P excretion was examined and found to be higher in V U R grade I I - I V than in children without renal disease. It is suggested that production a n d / o r secretion from the epithelial cells or leakage of T H P from damaged epithelial cells is increasing in these patients. Further studies are now being undertaken to examine the relationships between the levels of urinary T H P and epithelial T H P using E L I S A and immunohistochemistry. The significance of other markers of renal function in V U R both before and after anti-reflux surgery will also be investigated. Our preliminary study suggests that the level of urinary T H P measured using this ELISA may be a useful indicator of distal tubular damage in patients with renal disease.
References Dawnay, A., McLean, C. and Cattell, W.R. (1980) The development of a radioimmunoassay for Tamm-Horsfall glycoprotein in serum. Biochem J. 185, 679. Dulawa, J., Jam, K., Thomsen, M., Rambausek, M. and Ritz, E. (1988) Tamm-Horsfall glycoprotein interferes with bacterial adherence to human kidney cells. Eur. J. Clin. Invest. 18, 367. Grant, A.M.S. and Neuberger, A. (1973) The development of radioimmunoassay for the measurement of urinary TammHorsfall glycoprotein in the presence of sodium dodecyl sulfate. Clin. Sci. 44, 163. Hartmann, M.P.O., Abadie, C.P., Bouillie, J. and Hartmann, L. (1984) Variation of urinary Tamm-Horsfall protein in humans during the first thirty years of life. Nephron 38, 163. Hunt, J.S., McGiven, A.R., Groufski, A., Lynn, K.L. and
94 Taylor, M.C. (1985) Affinity-purified antibodies of defined specificity for use in a solid-phase naicroplate radioimmunoassay of human Tamm-Horsfall giycoprotein in urine. Biochem. J. 227, 957. Hunt, J.S., Peach, R.J., Brunisholz, M.C., Lynn, K.L. and McGiven, A.R. (1986) A sensitive and specific ELISA using a monoclonal capture antibody for detection of TammHorsfall urinary glycoprotein in serum. J. Immunol. Methods 91, 35. Imagawa, M., Yoshitake, S., Hamaguchi, Y., Ishikawa, E., Niitsu, Y., Urushizawa, I., Kanazawa, R., Tachibana, S., Nakazawa, N., Ogawa, H. (1982) Characteristics and evaluation of antibody-horseradish peroxidase conjugates prepared by using a maleimide compound, glutaraldehyde, and periodate. J. Appl. Biochem. 115, 92. Kjellsson, B., Soderstrom, T. and Hanson, L.~.. (1987) An ELISA method for quantification of Tamm-Horsfall protein using monoclonal antibodies. J. Immunol. Methods 98, 105. Kumar, S., Jasani, B., Hunt, J.S., Moffat, D.B. and Asscher, A.W. (1985) A system for accurate immunolocalization of Tamm-Horsfall protein in renal biopsies. Histochem. J. 17, 1251. Laemmli, U.K. (1970) Cleavage of structural proteins during assembly of the head of bacteriophage T4. Nature 227, 680. Lynn, K.L. and Marshall, R.D. (1984) Excretion of TammHorsfall glycoprotein in renal disease. Clin. Nephrol. 22, 253. Mazzuchi, N., Pecarovitch, R., Ross, N., Rodriguez, I. and Sanguinetti, C.M. (1974) Tamm-Horsfall urinary glycoprotein quantitation by radial immunodiffusion. J. Lab. Clin. Med. 84, 771. McQueen, E.G. (1962) The nature of urinary casts. J. Clin. Pathol. 15, 367. Nakagawa, Y. (1987) Urinary Tamm-Horsfall glycoprotein in patients with various renal diseases. Jpn. J. Nephrol. 22, 529. Peach, R.J., Day, W.A., Ellingsen, P.J. and McGiven, A.R. (1988) Ultrastructural localization of Tamm-Horsfall protein in human kidney using immunogold electron microscopy. Histochem. J. 20, 156.
Report of the International Reflux Study Committee (1981) Medical versus surgical treatment of primary vesicoureteral reflux. Pediatrics 67, 392. Rocha, A.S. and Kokko, J.P. (1973) Sodium chloride and water transport in the medullary thick ascending limb of Henle. Evidence for active chloride transport. J. Clin. Invest. 52, 612. Rose, G.A. and Sulaiman, S. (1984) Tamm-Horsfall mucoprotein promotes calcium phosphate crystal formation in whole urine. Urol. Res. 12, 217. Samuell, C.T. (1978) A modified electroimmunoassay technique for uromucoid in urine. Clin. Chim. Acta 85, 285. Schenk, E.A., Schwartz, R.H. and Lewis, R.A. (1971) TammHorsfall mucoprotein. I. Localization in the kidney. Lab. Invest. 25, 92. Schwartz, R.H., Lewis, R.A. and Schenk, E.A. (1972) TammHorsfall mucoprotein. III. Potassium dichromate-induced renal tubular damage. Lab. Invest. 27, 24. Silveria, A.M.V., Yamamoto, T., Adamson, L., Hessel, B. and Blomb~ick, B. (1986) Application of an enzyme-linked immunosorbent assay (ELISA) to von Willebrand factor (vWF) and its derivatives. Thrombosis Res. 43, 91. Tamm, I. and Horsfall, F.L. (1950) Characterization and separation of an inhibitor of viral hemaggiutination found in urine. Proc. Soc. Exp. Biol. Med. 74, 108. Towbin, H., Staehelin, T. and Gordon, J. (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: Procedure and some applications. Proc. Natl. Acad. Sci. U.S.A. 76, 4350. Tsuruta, J., Yamamoto, T., Kozono, K. and Kambara, T. (1985) Application of a new method of antibody-enzyme conjugation with maleimide derivative for immunohistochemistry: Hepatocelular production, interstitial distribution, and renal cell reabsorption of plasma albumin in guinea pig. T. Histochem. Cytochem. 33, 767. Wieslander, J., Bygren, P. and Heineg~ird, D. (1977) Determination of the Tamm and Horsfall giycoprotein in human urine. Clin. Chim. Acta 78, 391. Yamamoto, T. and Cochrane, C.G. (1982) A protease-like permeability factor in guinea pig skin: immunological identity with plasma Hageman factor. Am. J. Pathol. 107, 127.
Journal of Immunological Methods, 138 (1991) 87-94 © 1991 Elsevier Science Publishers B.V. 0022-1759/91/$03.50 ADONIS 002217599100129T JIM 05878
Determination of urinary Tamm-Horsfall protein by ELISA using a maleimide method for enzyme-antibody conjugation Iwao Uto 1,2, Takashi Ishimatsu 1, Hideo H i r a y a m a 1, Shoichi U e d a 1, Junji Tsuruta 2 and Takeshi K a m b a r a 2 J Department of Urology, and 2 Department of Allergy, Institute for Medical Immunology, Kumamoto University Medical School, Kumamoto, Japan (Received 18 July 1990, revised 7 November 1990, accepted 17 December 1990)
A method for enzyme-antibody conjugation using a new maleimide derivative as coupling reagent has been developed. Since a monomeric conjugate of horseradish peroxidase and Fab' antibody could be readily prepared with high efficiency and reproducibility, the enzyme activity and antigen-binding activity were well preserved and nonspecific staining was greatly reduced. The conjugate is suitable for use in both ELISA procedures and immunohistochemistry. Using both methods we examined the pathophysiological significance of Tamm-Horsfall protein (THP) and the present study describes the ELISA method to quantify urinary THP using the new method with rabbit anti-THP antibody. A low concentration (0.04 M) of urea added to the urine samples increased the linearity of the standard curve and the sensitivity of the assay, permitting the detection of as little as 20 n g / m l THP. Freezing and thawing the urine resulted in variable or lower values of THP concentration. THP concentrations in urine as determined by ELISA were stable for at least one month after - 7 0 °C storage, but not after - 3 0 °C storage. There was no correlation between THP concentrations in 24 h urine samples and the morning urine of the same patient. These results suggest that it is essential to use fresh or - 7 0 °C stored 24 h urine samples with added urea (0.04 M) for the determination of THP concentrations in urine by the present enzyme-antibody conjugation method. The THP concentration in normal 24 h urine of young children was found to be less than 51.8 m g / g Cr. Key words: Tamm-Horsfall protein; ELISA; Urea; Maleimide method
Introduction Correspondence to: I. Uto, Department of Urology, Institute for Medical Immunology, Kumamoto University Medical School, Kumamoto, Japan. Abbreviations: THP, Tamm-Horsfall protein; ELISA, enzyme-linked immunosorbent assay; Cr, creatinine; SDS, sodium dodecyl sulfate; PAGE, polyacrylamide gradient slab gel; PBS, phosphate-buffered saline; VUR, vesicoureteral reflux; HRP, horseradish peroxidase; HS-CHM, N-hydroxysuccinimide ester of N-(4-carboxycyclohexylmethyl) maleimide.
THP is a high molecular weight glycoprotein found in normal urine (Tamm and Horsfall, 1950), and is produced in the ascending limb of Henle and the distal convoluted tubules of the kidney (Schenk et al., 1971). It has been demonstrated that THP is the major component of urinary casts
88 (McQueen, 1962) and triggers calcium phosphate crystal formation and clumping in the process of urinary stone formation (Rose and Sulaiman, 1984). Dulawa et al. (1988) have reported that T H P protects against bacterial infection in the urinary tract by aggregating bacteria. However, its physiologic role and possible significance under pathologic circumstances remain unclear. In order to measure the concentration of urinary T H P in various renal diseases, several methods for the quantitative determination of T H P have been described, including radial immunodiffusion (Mazzuchi et al., 1974; Hartmann et al., 1984; Nakagawa, 1987), electroimmunoassay (Wieslander et al., 1977; Samuell, 1978), radioimmunoassay (Grant and Neuberger, 1973; Dawnay et al., 1980; Lynn and Marshall, 1984; Hunt et al., 1985) and enzyme immunoassay (Hunt et al., 1986; Kjellsson et al., 1987). However, the results have been variable and the clinical application of these methods to indicate renal damage has been unsatisfactory. This is because the assay method, are complicated and the treatment and the sampling of the urine is different for each method. In the present paper, we propose a new ELISA method for the measurement of urinary T H P using a maleimide derivative as a coupling reagent. In addition we have studied the influence of various treatments, storage and sampling methods and have established the optimum assay conditions for the measurement of T H P concentrations in both normal and pathological urine.
Materials and methods
Purification of urinary THP Urinary T H P was prepared by precipitation with 0.58 M NaC1 from pooled samples of urine (10 liters) obtained from healthy adult volunteers, as previously described (Grant and Neuberger, 1973). This precipitate was dissolved in 14 mM Na2HPO4/16 mM NaH2PO 4, p H 6.8, containing 8 M urea, and then applied to Sephacryl S-200 (bed volume 390 ml) equilibrated with the same buffer containing 2 M urea using a flow rate of 15 m l / h at room temperature. The absorbance of fractions was monitored at 280 nm and three protein peaks were obtained. The fractions corre-
sponding to the second major peak were pooled and dialysed against distilled water before being lyophilized. The final product was stored at -70°C.
Preparation of anti-THP antibody Antiserum to human urinary T H P was raised in New Zealand White rabbits. Multiple subcutaneous injections were given of 1 ml emulsion containing equal volumes of T H P (1 m g / m l in water) and complete Freund's adjuvant. Immunization was performed four times in total at 1 week intervals. Blood was taken from the ear 1 week after the last injection, and serum collected. IgG was prepared by salting out of the sera with ( N H 4 ) 2 S O 4 followed by passage through a diethylaminoethyl cellulose column. A F(ab')2 fragment was obtained from the IgG fraction by the pepsin treatment at p H 3.9 for 8 h at 37°C, followed by salting out with 60% saturated ammonium sulfate and by Sephadex G-150 column chromatography as p r e v i o u s l y d e s c r i b e d (Yamamoto and Cochrane, 1982).
Conjugation of anti-THP Fab' fragments with HRP using the maleimide method Anti-THP Fab' was conjugated with H R P by the maleimide method according to the method of Imagawa et al. (1982) and Tsuruta et al. (1985). Briefly, 6 mg of H R P in 1.0 ml of 0.1 M sodium phosphate buffer, p H 7.0, were incubated with 2.4 mg of H S - C H M in 100 t~l of N,N'-dimethylformamide for 20 min at 30 ° C. After centrifugation, the supernatant was applied to a Sephadex G-25 column (1 x 45 cm; bed volume 35 ml) using 0.1 M sodium phosphate buffer at p H 6.0. The fractions corresponding to the first peak were pooled and concentrated to approximately 300 /~1 in a collodion bag at 4 °C. These procedures resulted in the coupling of H R P and CHM (HRP-CHM). The average amount of C H M introduced per HRP molecule was calculated to be from 1.0 to 1.5 with good reproducibility. In order to prepare the Fab' fragment, 500 /~1 of 1-4 m g / m l of F(ab')2 fragment were reduced with 10 mM 2-mercaptoethylamine in 0.1 M sodium phosphate buffer at pH 6.0 for 90 min at 37°C. To remove the remaining nonreactive 2-mercaptoethylamine, the reactant was applied to a Sephadex G-25 column,
89 and the void volume fractions were pooled and concentrated to approximately 300/~1 in a collodion bag at 4 o C. H R P - C H M and the Fab' fragment, approximately 30-50 /~M each, were mixed immediately and incubated for 60 min at 37 ° C or for 20 h at 4 ° C . Then, 2 /~1 of 100 mM 2-mercaptoethylamine were added to the mixture to block the free maleimide base, followed by application to an Ultrogel AcA44 colunm (1.5 x 45 cm; bed volume 80 ml) with 0.1 M sodium phosphate buffer at pH 6.5. Absorbances were measured at 280 and 403 nm for protein concentration and for H R P concentration, respectively. Three protein peaks were obtained and the fractions corresponding to the second major peak (containing monomeric conjugates composed of a single Fab' and a single HRP) were stored in the presence of 0.1% bovine serum albumin and 0.005% merthiolate at 4 ° C and used in subsequent experiments.
Electrophoresis and immunoblotting for determining the purity of THP and the specificity of anti-THP Ab SDS-PAGE (5-20%) was performed according to the method of Laemmli et al. (1970), and Western blotting was performed by a modified method using enzyme-labelled antibody (Towbin et al., 1979).
cubated for 30 min at room temperature. After three final washings, color was developed using o-phenylenediamine dihydrochloride (Wako, 100 /~l/well of 3.7 mM solution in 50 mM disodium phosphate-25 mM citric acid buffer pH 5.0 containing 0.012% H202) for 1 min at room temperature. This reaction time was determined by preliminary experiments and was found to be most reproducible in agreement with Silveria et al. (1986). The reaction was stopped by the addition of 1 M HzSO 4 (100 /~l/well) and absorbance at 492 nm was measured.
Construction of the standard curve A standard curve was established by a modification of the procedure of Hunt et al. (1985) with purified urinary T H P diluted over the range 0.0110 n g / m l in the following buffers. (1) assay buffer (PBS-Tween) alone; (2) assay buffer containing 0.1% SDS; (3) assay buffer containing 0.04 M urea or (4) assay buffer containing 2 M urea.
Urine samples The urine of seven healthy volunteers, aged 31.6 + 2.1 years, was collected into individual sterile bottles for 24 h and then divided into 5 ml samples. The urine samples were assayed immediately or was stored either at - 3 0 ° C or at 70 o C until assay. Urinary T H P concentration were also estimated by the ELISA procedure in 26 children without renal disease (15 males and 11 females, aged 6.3 + 2.4 years) and in patients with different grades of primary vesico ureteral reflux (VUR) (12 males and 12 females, aged 5.0 _+ 2.6 years). Reflux was unilateral in 14 cases. Of these, one had grade V reflux, seven grade IV, two grade III and four grade II. Grading was established according to the international classification (Report of the International Reflux Study Committee, 1981). 24 h urine samples were collected and aliquots stored at - 7 0 ° C for T H P estimation. The values obtained were corrected for the concentration of urinary creatinine in order to determine the urine volume. The values of 24 h urine samples were compared with those of the first urine sample obtained in the morning on the same day. -
THP ELISA procedure Polystyrene microtiter plates (Nunc) were incubated for 1 h at 3 7 ° C with rabbit anti-human T H P IgG (100 /xl/well of 2 /ag/ml in coating buffer: 50 mM carbonate buffer, pH 9.6). After three washings with PBS-Tween (10 mM phosphate buffer containing 150 mM NaC1 and 0.05% Tween 20), the plates were filled with 1% BSA (150/al/well of I%BSA in coating buffer) to block non-specific binding sites, and incubated for 1 h at room temperature. After three further washings, the wells were filled with T H P standard solution or diluted (1/50) urine samples (100/~l/well) and incubated for 1 h at room temperature. After three further washings, the plates were filled with HRP-conjugated anti-THP antibody Fab' (2 /~g/ml in PBS-Tween BSA, 100 ~tl/well) and in-
90
Results
Preparation of THP and specificity of anti THP antibodies Purified T H P showed one major band on SDS-PAGE run under reducing conditions and stained with Coomassie blue. The apparent molecular weight was approximately 100,000. Westernblotting showed that the anti-THP antibody reacted with one major band and two faint bands. The former corresponded to a molecular weight of 100,000, which suggested identity with the major band of purified THP. The other two bands had molecular weights of 300,000 and 400,000, respectively, and were probably trimers and tetramers respectively (Fig. 1).
Construction of the standard curve The addition of high concentrations of urea (2 M) partially inhibited the antigen-antibody reaction on ELISA compared with the standard curve of T H P in buffer alone. However, the addition of
low concentrations of urea (0.04 M) did not prevent the reaction and gave an adequate standard curve with low backgrounds, sufficient linearity and parallelism to that of T H P in buffer alone. In contrast, the addition of 0.1% SDS (Hunt et al., 1985) prevented a reaction in the ELISA and resulted in an inadequate standard curve (Fig. 2). In this assay, the lowest level of detection was 20 n g / m l . When specified amounts of T H P were added to the samples, the recovery rate reached 93.7%. The intra-assay and inter-assay coefficients of variation were 7.0% and 3.7%, respectively.
Effect of centrifugation of urine samples Urine samples were pretreated both with and without centrifugation at 13,000 x g for 10 min to remove cells and precipitants in the urine, and 2 M urea was then added. They were then diluted 1 / 5 0 (final concentration of urea: 0.04 M) and tested in the ELISA procedure. The results are shown in Table I. The urinary
b ii iill I
~¢~ra~ -~trimer
94,. •.~ m o n o m e r 4
67,.
67
4a
,. ,.
43"
..
30,-
i~i/~!~iiiii!ii~i
30"2 0 ,.
}
di!,i iii ~
URINE THP
URINE THP
Fig. 1. Electrophoretic (a) and Western-blotting (b) profiles of urine and purified THP. Urine, normal human urine; THP, standard human THP. Arrows represent positions of standards, which are (from the bottom); soybean trypsin inhibitor, carbonic arthydrase, ovalbumin, bovine serum albumin and phosphorylase b. Purified THP showed one major band and its molecular weight was approximately 100,000. Western-blotting showed that the anti THP antibody reacted with one major band and two faint bands. The former band corresponded to molecular weight of 100,000, and the minor bands to molecular weights of 300,000 and 400,000, respectively.
91 TABLE I E F F E C T OF T R E A T M E N T O F U R I N E SAMPLES W I T H U R E A A N D / O R A B S O R B A N C E D E T E R M I N E D BY ELISA Treatment
T H E E F F E C T S OF C E N T R I F U G A T I O N ON
OD492 nm Urea a + cfg b +
No. 1 2 3 4 5 6 7
0.27 0.21 0.24 0.44 0.21 0.10 0.20
*
+ _
(43.4%) (27.8%) (30.8%) (59.5%) (34.2%) (26.0%) (39.6%)
0.61 0.74 0.76 0.73 0.62 0.40 0.55
0.24 ± 0.10 (37.0 ± 11.5%)
_ + (100%) (100%) (100%) (100%) (100%) (100%) (100%)
0.19 0.15 0.11 0.39 0.10 0.09 0.10
0.63 ± 0.13 (100%)
(31.0%) (19.5%) (15.0%) (53.1%) (16.2%) (21.3%) (16.7%)
0.49 0.53 0.56 0.64 0.51 0.30 0.45
0.16 + 0.11 (24.7 _ 13.6%)
(79.5%) (70.9%) (73.8%) (87.1%) (83.5%) (74.6%) (81.6%)
0.50 ± 0.11 (78.7 + 5.8%)
Key." samples were measured with or without adding 2 M urea a n d / o r centrifugation (cfg). a 2-M urea, bcfg 13,000× g, 10 min, * m e a n + S D .
freezing at - 30 o C and - 70 o C for periods ranging from 2 weeks to 3 months. The T H P concentration of urine stored at - 7 0 ° C was similar to that of T H P in fresh urine when measured within the first month of storage (Table II). On the other hand the values varied widely in the initial period in the case of samples stored at 30 ° C (Table II).
T H P concentrations were considerably decreased by the centrifugation of urine samples independent of the addition of urea. Addition of urea without centrifugation gave the most suitable curve with the highest values and adequate linearity (Fig. 2).
Effect of storage on urine samples
-
The effects of storage temperatures and times were studied in seven urine samples from healthy volunteers. Samples were assayed before and after
Effect of freezing and thawing urine samples Six urine samples from healthy volunteers were
T A B L E II E F F E C T S O F D E L A Y E D S T O R A G E A N D F R E E Z E - T H A W I N G O F U R I N E SAMPLES O N T H E T H P C O N C E N T R A T I O N (#G/ME) a Urine
Fresh
no. 1 2 3 4 5 6 7 Mean ±SD
1.8 4.5 6.2 3.6 6.2 1.9 3.9 4.0 ±1.8
Storage at - 30 o C
Storage at - 70 o C
0.5
1
3 (month)
39.2 8.8 26.1 16.9 31.0 6.0 12.7 20.1 ±12.3
9.0 7.0 14.8 16.1 60.9 13.2 16.1 19.6 ±18.6
18.0 9.0 11.5 15.1 45.1 12.7 24.5 19.4 ±12.4
a All assay was performed in duplicate.
0.5 8.8 5.2 6.2 6.3 14.2 5.2 5.8 7.4 ±3.2
Freeze-thawing
1
3 (month)
1
2
5.8 6.3 5.1 10.9 10.8 7.1 5.1 7.3 ±2.5
19.4 11.0 13.1 12.8 26.2 14.9 15.0 16.1 ±16.1
6.9 13.8 4.0 4.3 5.2 7.6 3.6 4.3 13.1 9.5 5.0 4.8 Notdone 6.3 7.4 ±3.5 ±3.8
3
4
5 (times)
10
10.2 5.0 6.7 4.3 14.2 5.7
8.8 4.3 7.6 5.5 15.0 4.3
9.5 4.5 6.2 6.2 11.9 4.8
13.1 5.0 6.4 5.0 15.5 6.7
7.8 ±3.7
7.6 ±4.1
7.2 ±2.9
8.6 ±4.5
92
E c
1.0
£,J O3
¢-.
0.5
0
03
,< 0.0
........
, .1
.01
........
, 1
........ 10
THP concentration (gg/ml) Fig. 2. Standard curves of THP concentration. Standard curves for THP were prepared in assay buffer alone (o), 0.1% SDS (O), 0.04 M urea (A) or 2 M urea (A). 2 M urea slightly inhibited the antigen-antibody reaction in ELISA. 0.04 M urea did not prevent the reaction, and gave a curve which was similar to that obtained with assay buffer alone. 0.1% SDS interfered with the ELISA reaction. assayed after repeated freezing and thawing (1, 2, 3, 4, 5, and 10 times) by solid C O 2 / a c e t o n e . In four samples T H P levels were stable, but in two samples the levels were increased (Table II).
Difference in THP concentrations in 24 h urine collections and morning urine samples The T H P concentrations of matched 24 h and morning urine samples were c o m p a r e d and no correlation was f o u n d (r = 0.12, Fig. 3).
Clinical study In the normal control group (n = 26), the mean _+ SD for urinary T H P / C r was 19.2 _+ 10.5
loo-
E c
cD ..c
s'0 morning
,~0 unne
(mg/gCr)
Fig. 3. Comparison of THP concentration as measured in 24 h urine and morning urine, samples. The THP concentration in 24 h urine samples from 13 VUR patients was measured and compared with that of each morning urine. There was no correlation between the two measurements (r = 0.12).
m g / g C r . It was f o u n d that this value was 26.4 _+ 12.7 m g / g Cr in the y o u n g children, 2 - 5 years of age, and 13.6 _+ 3.2 m g / g Cr in the older children, 6 - 1 0 years of age. Accordingly, we conclude that the normal value of T H P in y o u n g children is below 51.8 m g / g Cr (mean + 2 SD). This study was extended to the patients with p r i m a r y V U R (n = 4 in grade II, 2 in grade III, 7 in grade IV and 1 in grade V). As a preliminary study, high values of T H P were observed from grade II to grade IV. The mean value for the patients was higher than that for normal children. However, wide variations were observed (mean _+ SD, 60.4 4-45.2 m g / g Cr) and since the numbers in each group were very few, the differences were not statistically significant.
Discussion
Since the first description of T H P in 1950 ( T a m m and Horsfall), m a n y studies have been reported on its localization ( K u m a r et al., 1985; Peach et al., 1988) and physiological function ( R o c h a and Kokko, 1973). It is accepted that T H P is p r o d u c e d and secreted by epithelial cells of distal tubules and the ascending limb of Henle. However, the physiological and pathological role of T H P remains unclear. N u m e r o u s quantitative assays of urinary T H P have been reported (Grant and Neuberger, 1973; Mazzuchi et al., 1974; Wieslander et al., 1977; Samuell, 1978; D a w n a y et al., 1980; H a r t m a n n et al., 1984; H u n t et al., 1985; H u n t et al., 1986; Kjellsson et al., 1987; Nakagawa, 1987) but the values obtained have been remarkably variable. Discrepancies between these previous reports m a y have been due to: (1) the use of m o r n i n g urine instead of 24 h urine samples, (2) differences in the storage conditions used, (3) the use of urea (8 M) as a dissociating agent in the purification of standard T H P , (4) differences in the assay system, i.e., r a d i o i m m u n o a s s a y or E L I S A , or (5) the treatment of the urine samples. H S - C H M was developed by Imagaiva et al. (1982) as a cross-linking reagent for enzyme labelling in enzyme immunoassay. In this method, a single a m i n o group of one H R P molecule is reactive with H S - C H M and allows the introduced maleimide group to react with a single thiol group
93 of the hinge portion of the Fab' fragment. Therefore, one can expect to obtain monomeric conjugates of Fab' and H R P with high efficiency and good reproducibility. In the present study, the conjugate was found to give very satisfactory resuits. Although the anti-THP antibody was a conventional preparation and not immunopurified, the use of a conjugate prepared by the maleimide method gave good reproducibility with negligible background. The standard curves produced on each plate, using prepared THP, showed good reproducibility and demonstrated an adequate linear range from 20 n g / m l to 10 t t g / m l for the measurement of unknown urinary T H P concentrations. In the present study the T H P concentration in 24-h urine samples did not correlate with the T H P levels in morning urine samples study and this suggests that urinary T H P concentrations exhibit circadian variations in vivo. We conclude that 24-h urine sampling is essential for the estimation of urinary THP. In previous studies the urine samples had been stored at between - 20 o C and - 30 ° C. In the present study T H P values of the urine samples stored at - 3 0 ° C showed wide variations. In contrast, the levels were relatively stable at - 7 0 ° C during the first month of storage and the reproducibility and recovery rate of the data were constant. Therefore, urine samples should be stored at - 7 0 o C, and measurements should be performed within the first month of storage. Urinary T H P concentrations were much lower after centrifugation than those values obtained without centrifugation. This phenomenon m a y be due to the removal of aggregated and precipitated T H P and adsorbed T H P on cells a n d / o r crystals in the urine. Therefore, centrifugation of urine samples to remove insoluble materials is not recommended in the measurement of urinary THP. As previously pointed out, it is necessary to add dissociating agents such as urea or SDS for measurement of urinary T H P by radial immunodiffusion and ELISA methods, because T H P easily forms macromolecular complexes. However, treatment with 2M urea or 0.1% SDS inhibited the binding between T H P and anti-THP Ab and the standard curve showed inadequate linearity and
sensitivity. In contrast, the presence of 0.04 M urea in the urine samples resulted in suitable linearity and sensitivity without any inhibition of the T H P and anti-THP Ab reaction. It is therefore suggested that 0.04 M urea should be added to the urine samples for measurements of urinary T H P excretion. In the present study, the range of urinary T H P in normal children was below 51.8 m g / g Cr and the value showed a tendency to decrease with age. This result is consistent with the report of Hartm a n n et al. (1984) in which the range was below 53 m g / g Cr as measured by the radial immunodiffusion method. These authors also found that urinary T H P tended to decrease with age. In a preliminary study in patients with VUR, urinary T H P excretion was examined and found to be higher in V U R grade I I - I V than in children without renal disease. It is suggested that production a n d / o r secretion from the epithelial cells or leakage of T H P from damaged epithelial cells is increasing in these patients. Further studies are now being undertaken to examine the relationships between the levels of urinary T H P and epithelial T H P using E L I S A and immunohistochemistry. The significance of other markers of renal function in V U R both before and after anti-reflux surgery will also be investigated. Our preliminary study suggests that the level of urinary T H P measured using this ELISA may be a useful indicator of distal tubular damage in patients with renal disease.
References Dawnay, A., McLean, C. and Cattell, W.R. (1980) The development of a radioimmunoassay for Tamm-Horsfall glycoprotein in serum. Biochem J. 185, 679. Dulawa, J., Jam, K., Thomsen, M., Rambausek, M. and Ritz, E. (1988) Tamm-Horsfall glycoprotein interferes with bacterial adherence to human kidney cells. Eur. J. Clin. Invest. 18, 367. Grant, A.M.S. and Neuberger, A. (1973) The development of radioimmunoassay for the measurement of urinary TammHorsfall glycoprotein in the presence of sodium dodecyl sulfate. Clin. Sci. 44, 163. Hartmann, M.P.O., Abadie, C.P., Bouillie, J. and Hartmann, L. (1984) Variation of urinary Tamm-Horsfall protein in humans during the first thirty years of life. Nephron 38, 163. Hunt, J.S., McGiven, A.R., Groufski, A., Lynn, K.L. and
94 Taylor, M.C. (1985) Affinity-purified antibodies of defined specificity for use in a solid-phase naicroplate radioimmunoassay of human Tamm-Horsfall giycoprotein in urine. Biochem. J. 227, 957. Hunt, J.S., Peach, R.J., Brunisholz, M.C., Lynn, K.L. and McGiven, A.R. (1986) A sensitive and specific ELISA using a monoclonal capture antibody for detection of TammHorsfall urinary glycoprotein in serum. J. Immunol. Methods 91, 35. Imagawa, M., Yoshitake, S., Hamaguchi, Y., Ishikawa, E., Niitsu, Y., Urushizawa, I., Kanazawa, R., Tachibana, S., Nakazawa, N., Ogawa, H. (1982) Characteristics and evaluation of antibody-horseradish peroxidase conjugates prepared by using a maleimide compound, glutaraldehyde, and periodate. J. Appl. Biochem. 115, 92. Kjellsson, B., Soderstrom, T. and Hanson, L.~.. (1987) An ELISA method for quantification of Tamm-Horsfall protein using monoclonal antibodies. J. Immunol. Methods 98, 105. Kumar, S., Jasani, B., Hunt, J.S., Moffat, D.B. and Asscher, A.W. (1985) A system for accurate immunolocalization of Tamm-Horsfall protein in renal biopsies. Histochem. J. 17, 1251. Laemmli, U.K. (1970) Cleavage of structural proteins during assembly of the head of bacteriophage T4. Nature 227, 680. Lynn, K.L. and Marshall, R.D. (1984) Excretion of TammHorsfall glycoprotein in renal disease. Clin. Nephrol. 22, 253. Mazzuchi, N., Pecarovitch, R., Ross, N., Rodriguez, I. and Sanguinetti, C.M. (1974) Tamm-Horsfall urinary glycoprotein quantitation by radial immunodiffusion. J. Lab. Clin. Med. 84, 771. McQueen, E.G. (1962) The nature of urinary casts. J. Clin. Pathol. 15, 367. Nakagawa, Y. (1987) Urinary Tamm-Horsfall glycoprotein in patients with various renal diseases. Jpn. J. Nephrol. 22, 529. Peach, R.J., Day, W.A., Ellingsen, P.J. and McGiven, A.R. (1988) Ultrastructural localization of Tamm-Horsfall protein in human kidney using immunogold electron microscopy. Histochem. J. 20, 156.
Report of the International Reflux Study Committee (1981) Medical versus surgical treatment of primary vesicoureteral reflux. Pediatrics 67, 392. Rocha, A.S. and Kokko, J.P. (1973) Sodium chloride and water transport in the medullary thick ascending limb of Henle. Evidence for active chloride transport. J. Clin. Invest. 52, 612. Rose, G.A. and Sulaiman, S. (1984) Tamm-Horsfall mucoprotein promotes calcium phosphate crystal formation in whole urine. Urol. Res. 12, 217. Samuell, C.T. (1978) A modified electroimmunoassay technique for uromucoid in urine. Clin. Chim. Acta 85, 285. Schenk, E.A., Schwartz, R.H. and Lewis, R.A. (1971) TammHorsfall mucoprotein. I. Localization in the kidney. Lab. Invest. 25, 92. Schwartz, R.H., Lewis, R.A. and Schenk, E.A. (1972) TammHorsfall mucoprotein. III. Potassium dichromate-induced renal tubular damage. Lab. Invest. 27, 24. Silveria, A.M.V., Yamamoto, T., Adamson, L., Hessel, B. and Blomb~ick, B. (1986) Application of an enzyme-linked immunosorbent assay (ELISA) to von Willebrand factor (vWF) and its derivatives. Thrombosis Res. 43, 91. Tamm, I. and Horsfall, F.L. (1950) Characterization and separation of an inhibitor of viral hemaggiutination found in urine. Proc. Soc. Exp. Biol. Med. 74, 108. Towbin, H., Staehelin, T. and Gordon, J. (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: Procedure and some applications. Proc. Natl. Acad. Sci. U.S.A. 76, 4350. Tsuruta, J., Yamamoto, T., Kozono, K. and Kambara, T. (1985) Application of a new method of antibody-enzyme conjugation with maleimide derivative for immunohistochemistry: Hepatocelular production, interstitial distribution, and renal cell reabsorption of plasma albumin in guinea pig. T. Histochem. Cytochem. 33, 767. Wieslander, J., Bygren, P. and Heineg~ird, D. (1977) Determination of the Tamm and Horsfall giycoprotein in human urine. Clin. Chim. Acta 78, 391. Yamamoto, T. and Cochrane, C.G. (1982) A protease-like permeability factor in guinea pig skin: immunological identity with plasma Hageman factor. Am. J. Pathol. 107, 127.
