Quan~i~a~i~n of Urinary dtu-Cfubufin and Albumin by Reverse-Phase High Performance Liquid Cbr~ma~ogra~hy

LOIS D. ~~HMAN-~~K~~MA~ AND DOUGLAS CAUDILL

A rapid, reproducjble, and sensitive high-performance liquid chromatography fHPLC) method for the quant~tatioo of mu-gtobulin, the major urinary protein excreted by adult male rats* and albumin has been developed. Totaf urinal proteins, isolated by a simple Sephadex G-25 gel f~itrati~n step, are separated and quantitated by reverse-phase HPLC on a C, Macrosphere 300 column. The proteins are separated and eluted with a two-step gradient af acetonitriie in aqueous triffuoroacetic acid. Detection limits of 9 and 25 pg/mL of urine were established for albumin and ~Zu-globulin, respectively. Quantitatio~ of urinary excretion of the two proteins in young adult male and female rats and aging male rats showed that values obtained with this method compared favorably with values from previously developed immunological techniques. To quantitate total urinary protein excretion, we modified the Bradford protein assay to use rat urinary protein as standard. Given the established importance of a2u-giobulin in the development of maie rat-specific nephrotoxicity and nephrocarcjnogenici~~ these methods should be useful for studying the renal handling of this protein under normal and nephrotox~c conditions, Key Words: ~2u~~~obu~i~~ Quantjtat~on; specific nephrotoxicity.

Urinary protein excretion;

Mafe rat-

INTRODUCTION 1%is generally recognized that, under phys~ofogical conditions, rodents are sig nificantfy more proteinuric than humans (Anderson et al., 1979; Goldstein et al., 1988; Olson et al., 19901. In rats, a sex-dependent proteinuria is noted, as adult male rats excrete much more protein than do female rats (Selfers et al., 1950; Neuhaus and Flory, 1978). This sex-dependent difference arises from the significant quantities of one protein, ~~u-gf~b~fin~ which is the major urinary protein laccounting for ~pproximatefy 30% of urine protein) excreted by adult mafe rats (Roy et at,, 1966; Neuhaus and Flory, 1978). a&-Globulin has been detected in female rat urine, but at levels that are significantly fess than those observed in male rat urine (Vandoren et al., 1983). Whereas a family of proteins that are structurally similar to &&globulin and that are synthesized across species has been identified

From the Human and Environmental Safety Division, Miami Valley Laboratories, Procter & Gamble Campany, Cincinnati, Ohio. Address reprint requests to: Dr. Lois D. Lehman-~&~eema~, Miami Vaffey Laboratories, Procter & Gamble Company, P.O. Box 398707, Cincinnati, OH 45239-8707, U.S.A. Received February 13,1991; revised and accepted May 15,1991. ZB

240

1. Lehman-McKeeman

and D. Caudill

(Brooks, 1987; Pevsner et al., 1988), only adult male rats are known to synthesize and excrete a2uglobulin. A diverse group of chemicals, mainly hydrocarbons, has been shown to cause a male rat-specific nephrotoxicity manifested acutely as the accumulation of protein droplets within renal proximal tubule ceils. It has been determined that cY2uglobulin is the only protein accumulating within these droplets (Kanerva et al., 1987). Moreover, it has been shown that the binding of the chemical (or its metabolites) to IX~Uglobulin is prerequisite to the ensuing toxicity (Lock et al., 1987; Strasser et al., 1988; Lehman-McKeeman et al., 1989, 1991; Charbonneau et al., 1989). Because chronic exposure to these chemicals gives rise to renal tubular tumors, again exclusively in male rats, this unusual nephrotoxicity and nephrocarcinogenicity has received considerable attention in the toxicology and regulatory communities (Swenberg et al., 1989; Borghoff et al., 1990; Flamm and Lehman-McKeeman, 1991). To date, the procedures used to evaluate the role of a2u-globulin in this toxicity have been limited to qualitative analyses. Two-dimensional gel electrophoresis was used initially to establish the specificity of this toxicity to a2u-globulin (Kanerva et al., 1987; Ridder et al., 1990). Since that time, immunohistochemical techniques have been used to localize a2u-globulin to the protein droplets (Olson et al., 1988; Burnett et al., 1989). Quantitatively, radioimmunoassays have been developed for a2u-globulin (Neuhaus and Flory, 1975; Roy, 1977; Charbonneau et al., 1987). In this paper, the development of a reverse-phase HPLC method to quantify both a2uglobulin and albumin is described. This method offers a rapid, sensitive procedure for simultaneous analysis of these proteins, and because total urine protein is separated, the method also allows for qualitative comparisons to be made. METHODS Apparatus

The HPLC instrument consisted of three Rainin Rabbit HP pumps equipped with a variable wavelength ultraviolet-visible detector (Knauer, Model 87) and a pneumatically controlled auto-injector (ICI, Model AS2000) equipped with a lOo-t.~L sample loop (Rainin Instruments, Woburn, MA). The chromatography system was controlled by a Macintosh Plus computer using Dynamax HPLC Method Manager (Rainin Instruments). Chromatography was performed on a Macrosphere 300 Cq column (15 cm x 4.6 mm; Alltech Associates, Deerfield, IL) equipped with a directconnect guard column. The solvent system consisted of (A) aqueous 0.12% trifluoroacetic acid [TFA (Aldrich Chemical Co., Inc., Milwaukee, WI)]; (B) 0.12% TFA in 70% HPLC grade acetonitrile (Fischer Scientific, Cincinnati, OH); and (C) 100% acetonitrile. Initial mobile-phase conditions were 85% A mixing with 15% B. Urinary proteins were eluted with a two-step linear gradient from initial conditions to 50% A/50% B in 10 min followed immediately by a linear gradient from 50% A/50% B to 40% A/60% B over 10 min. The system was maintained at 60% B for 2 min and then stepped to 100% C for 2 min before returning to initial conditions. All mobile-phase solvents were filtered and degassed by vacuum filtration through a 0.45~(*m pore

HPLC Quantitation of ol2u-Globulin nylon membrane filter (Micron Separators, Inc., Westboro, 1 mL/min, and proteins were detected at 280 nm. Isolation and Purification

of a2u-globulin

MA). The flow rate was

as Standard for the HPLC Assay

The a2uglobulin used as standard for this assay was purified by a two-step chromatographic process described in detail by Lehman-McKeeman et al., 1990. Briefly, total urine protein was isolated by Sephadex G-25 gel filtration chromatography (Pharmacia, Piscataway, NJ). This total protein fraction was lyophilized and stored desiccated at -8O”C, where it was stable for at least 6 months. The lyophilized protein was used as standard for the Bradford protein assay (see below) or reconstituted in water at 50 mg/mL to purify a2u-globulin. a2u-Globulin was isolated from the total protein fraction by anion-exchange HPLC utilizing the HPLC system described above equipped with a Synchropak AX300 semi-preparative column (25 cm x IO mm i.d.). Proteins were eluted with a linear salt gradient (IO mM Tris-HCI, pH 7.4 at 25”C, containing 0.5 M NaCl; 3 mL/min) and detected at 280 nm. The peak representing cw2u-globulin was collected, desalted, lyophilized, and subsequently stored desiccated at -80°C. The purity of the cu2u-globulin obtained in this manner was verified by amino acid sequence analysis of the amino terminus (LehmanMcKeeman et al., 1990). The rat albumin used in this assay (Sigma Chemical Co., St. Louis, MO) was globulin and fatty acid free. Animals and Urine Collection Adult (approximately 80 days of age) male and female Fischer 344 rats (Charles River Laboratories, Portage, MI) were used. Rats were housed under a 12-hr light/ dark cycle in a humidity- and temperature-controlled room and allowed free access to food (Purina Laboratory Rodent Chow, Ralston-Purina Co., St. Louis, MO) and water. To collect urine, housed rats individually in stainless steel metabolism cages and then fitted with a fecal cup (Smyth, 1979) to prevent fecal contamination of urine. Urine was collected over dry ice in 24-hr intervals. All urine was kept frozen at -80°C prior to analysis. Preparation

of Urine Samples for HPLC Analysis

Urine samples were thawed at room temperature and placed on ice. A 2.5-mL aliquot of ice-cold urine was applied to a Sephadex G-25 gel filtration column (PDIO, Pharmacia) previously equilibrated with distilled, deionized water and kept at 4°C. Proteins were eluted from the column by applying 3.2 mL of water. This 3.2mL fraction, representing proteins greater than 5000 Da, was collected for HPLC analysis, and a 50-PL aliquot was injected onto the Macrosphere column. The G-25 protein fraction was stable for at least 6 months when stored frozen at -80°C. Quantitation

of Total Urinary Protein

The Bradford protein assay (Bradford, 1976) with Coomassie Brilliant Blue G-250 (Bio-Rad, Richmond, CA) was used to quantify daily urinary protein excretion. Total

241

242

1. Lehman-McKeeman

and D. Caudill

protein was isolated from urine (described above) prior to analysis. In preliminary experiments, standard curves for rat albumin and lyophilized total rat urine protein were

compared.

dard curves made

A stock solution

from

at 2 mg/mL

of albumin

0.1 to 0.5 mg/mL, to establish

595 nm was determined

whereas

standard

(1 mg/mL)

was used to establish

the total

urine

protein

curves from 0.1 to 2 mg/mL.

spectrophometrically

(Beckman,

Model

stan-

standard

was

Absorbance

DU50,

at

Fullerton,

CA). RESULTS Representative and albumin a2u-globulin slightly is quite

chromatograms,

as the Macrosphere good

column

on a day-to-day

both proteins. main

showing

the baseline

resolution

of cY2u-globulin

are presented in Figure 1. Albumin (peak 1) elutes at 15.9 min, whereas (peak 2) elutes at 16.4 min. Whereas the retention time may change

completely

Moreover,

regardless

resolved.

a coefficient

of column

differences

column

between

of the retention

of variation

age, a2u-globulin

Both the life of the column

are greatly enhanced by the 2-min of the solvent gradient. The qualitative

ages, the reproducibility

basis with

and albumin

and column

wash with 100% acetonitrile

old and young

1 1 Yr-Old Male Rat

time

less than 1 for re-

performance at the end

male rats and between

male

I

&--AJh2

N

8 i B

3 Month-Old Male Rat

a 8 3 Month-Old Female Rat

TIME (min)

The top profile FIGURE 1. Chromatographic separation of rat albumin and a2uglobulin. shows the separation of 25 pg (in a 50-pl injection) of each protein standard where 1 is albumin and 2 is a2u-globulin. Fifty microliters of the urinary protein fraction, prepared as described in the Methods section was injected for quantitation of the proteins. In male rats, both albumin and a2u-globulin are detected, with albumin excretion increasing as rats age.

HPLC Quantitation 25

of cuZu-Globulin

I lobulin

20

I 50

25

75

PROTEIN (ggis)

FIGURE 2. Represen~tive standard curves for quantitation of albumin and mu-globulin with the HPLC method. Results represent the mean f SE of five individua1 standard curves. Regression coefficients for the best-fit lines were 0.999 for both curves.

and female rats are also depicted in Figure 1. As rats age, the urinary excretion of albumin increases, whereas the output of ol2uglobulin gradually declines (Neuhaus and Flory, 1978). In young male rats, cr2uglobulin accounts for 35%-40% of the total chromatogram peak area. In the older male rats, cw2uglobulin accounts for approximately IS%, and albumin represents 10% of the total urinary protein. As shown in Figure 1, the major difference between male and female rat urinary protein profiles is the absence of detectable levels of c&-globulin in female rats. Typical standard curves for ~2u-globulin and albumin are shown in Figure 2. In a SO-PL injection, standard curves from 3 to 75 t~..gfor both proteins give an acceptable range for the analysis of undiluted urine protein fractions. For ol2u-globulin, the slope of the best-fit line was 15,385 t 375, whereas for albumin, the slope was 22,750 2 300 (n = 5 for both proteins). For individual data points, the coefficient of variation across multiple standard curves was less than 3. The detection limits for a2u-globulin and albumin were established as approximately 1 and 0.35 pg, respectively. Given a 50-t~,L injection of the urinary protein

l

e

PROTEIN

Albumin Urinary Protein

(mg)

FIGURE 3. Standard curves obtained for rat albumin versus total urinary protein with the Bradford protein assay. The slopes of the two curves were 0.664 and 0.206 for albumin and rat urinary protein, respectively.

243

244

L. Lehman-McKeeman TABLE 1

and 0. Caudilt

Urinary Protein Excretion in Fischer 344 Rats mg PROTEIN/DAF _______.ACE

Untreated Femaled Maled Male” d-Limonenef Female Male

cW2U-CLOBULINb

ALBUMINS

TOTALS

1 yr

ND 12.24 rt 0.82 9.87 rt 2.33

ND 0.86 c 0.07 24.68 c 8.01

4.33 ?I 0.29 38.55 * 1.90 69.85 2 10.42

3mo 3 mo

ND 30.47 rt 2.43*

ND 1.03 r 0.11

4.92 + 0.41 79.2 2 4.06*

3 mo 3 mo

Abbreviations: ND, not detected. ’ All values are expressed as mean r SE. b Determined by the reverse-phase HPLC method. ’ Determined with the modified Bradford protein assay. dn = 24. en = IO. f Rats (n = 6) were treated orally with d-limonene (150 mg/kg) for 5 days/week for 3 weeks. * Statistically different from 3-month-old untreated male rat as determined by Student’s fp < 0.05).

t test

fraction (diluted from 2.5 to 3.2 mt on the G-25 column), these detection limits correspond to protein concentrations of 25 &mL urine for ol2uglobulin and 9 TV@ mL urine for albumin. To determine the appropriate standard to use to quantitate total urine protein excretion with the Bradford assay, we compared standard curves for albumin and total urine protein (from male rats). As shown in Figure 3, an equal amount of albumin absorbs about four times more than does the total urine protein fraction. Furthermore, with albumin, the standard curve is linear only up to 0.5 mg, whereas the urinary protein standard curve is linear up to at least 2 mg of protein. Thus, in this assay, it is appropriate to use the rat urine protein as standard because, with the significant difference in the slope of the standard curves, albumin will yield incorrect, low values for urine protein excretion in rats. To demonstrate the utility of the reverse-phase HPLC and modified Bradford protein assays, the daily excretion of ol2uglobulin, albumin, and total urinary protein in male and female Fischer 344 rats is presented in Table 1. There is approximately a tenfold difference in total protein excreted between male and female rats. As male rats age, albumin excretion increases, but there is significant variation among animals. Repeated dosing of d-limonene, a known hyaline droplet inducer, increases total protein excretion, an effect that appears to result primarily from enhanced elimination of a2u-globulin. DISCUSSION This paper reports the first HPLC method developed to quantitate a2u-globulin, the major urinary protein excreted by adult male rats, A variety of chromatographic

HPLC Quantitation of a2wGfobulin

techniques have been developed to profile urinary proteins from many species (Ratge and Wisser, 1982; Lindblom et al., 1983; Weber, 1988). In addition, Olson et al. (1990) have made a direct qualitative comparison of urinary protein excretion in male rats and humans to assess the relevance of the male rat-specific nephrotoxicity for human risk assessment. The method presented herein describes the simultaneous quantitation of ol2u-globulin and albumin in rats. The results obtained with the HPLC method are in good agreement with previously reported values obtained primarily with immunological techniques (Roy et al., 1966; Lane and Newhaus, 1972; Newhaus and Flory, 1978; Newhaus et al., 1981; Vandoren et al., 1983). This is the first time that excretion of ~2u-globulin in male rats with hyaline droplet nephropathy has been quantitated, and as noted in Table 1, d-limonene treatment markedly increased a2uglobulin elimination. Therefore, this HPLC method will be useful in studying how the renal handling of a2u-globulin is altered by the development of this toxicity. Given the abundance of a2u-globulin in urine, method sensitivity is not an issue for quantitation of this protein. The double antibody radioimmunoassay (RIA) developed by Roy (1977), which is capable of detecting picogram levels of the protein, offers the greatest sensitivity of all assays developed for cY2u-globulin. With the exception of this RIA, the HPLC method described herein compares favorably with other immunological procedures used to quantitate ~2u-globulin. For example, radial immunodiffusion techniques have detection limits in the microgram per milliliter range (Neuhaus and Flory, 1975; Kurtz and Feigelson, 1978; Neuhaus and Lerseth, 19791, similar to the HPLC method. Furthermore, the HPLC method is more sensitive than published enzyme-linked immunosorbent assays (ELISA) for 012uglobulin, as Charbonneau et al. (1987) reported a detection limit of 2 mg/g of kidney, which according to the procedures described for tissue preparation is equivalent to 670 Fg/mL. This ELISA is more than 20 times less sensitive than the HPLC method is where the detection limit is approximately 25 pg/mL. However, relative to immunological procedures, the HPLC method offers the advantage of simultaneous quantitation of albumin and cu2u-globulin, along with qualitative comparisons protein profiles. To evaluate total urinary protein excretion, we selected the widely used Bradford protein assay (Bradford, 1976). In quantitating urinary protein, bovine serum albumin, as described by Kluwe (1981) has typically been used as standard. However, as demonstrated in Figure 3, there is a significant difference between the binding of Coomassie Blue to albumin and rat total urinary protein. Therefore, for urinary protein quantitation, it is more appropriate to use lyophilized urinary protein as standard. In summary, we have developed a reverse-phase HPLC method for the quantitation of cY2u-globulin, the major urinary protein excreted by adult male rats. With this method, urinary albumin can also be quantified. In addition to the quantitation of both proteins, the method allows for the qualitative assessment of urinary protein profiles. The results presented also demonstrate the need to use the appropriate standard, namely the total urinary protein fraction, to quantitate accurately urinary protein excretion with the Bradford protein assay. These methods will allow for the

245

246

1. Lehman-McKeeman

and D. Caudill

specific analysis of the renal handling of a2u-globulin, elimination under normal and nephrotoxic conditions.

together

with

total protein

REFERENCES Anderson NC, Anderson NL, Tollakson SL (1979) Proteins of human urine. I. Concentration and analysis by two-dimensional electrophoresis. C/in Chem 25:1199-1210. Borghoff SJ, Short BG, Swenberg JA (1990) Biochemical mechanisms and pathobiology of a2uglobulin nephropathy. Annu Rev fharmacol Toxice/30:349-367.

Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal Biochem 721248-254. Brooks DE (1987) The major androgen-regulated secretory proteins of the rat epididymus bear sequence homology with members of the a2uglobulin superfamily. Biochem Int 14:235-240. Burnett VL, Short BC, Swenberg JA (1989) Localization of cu2u-globulin within protein droplets of male rat kidney: immunohistochemistry using perfusion-fixed, GMA-embedded tissue sections. / Histochem Cytochem 37:813-818. Charbonneau M, Lock EA, Strasser I, Cox MC, Turner MJ, Bus JS (1987) 2,2,4_Trimethylpentaneinduced nephrotoxicity. I. Metabolic disposition of TMP in male and female Fischer 344 rats. Toxicol Appl Pharmacol91: 171-I 81. Charbonneau M, Strasser J, Lock EA, Turner MI, Swenberg JA (1989) Involvement of reversible binding to cw2u-globulin in 1,4-dichlorobenzeneinduced nephrotoxicity. Toxicol Appl Pharmacol 99:122-132.

Flamm WG, Lehman-McKeeman LD (1991) The human relevance of the renal tumor-inducing potential of d-limonene in male rats. Implications for risk assessment. Reg Toxicol Pharmacol 13:70-86. Goldstein RS, Tarloff JB, Hook JB (1988) Age-related nephropathy in laboratory rats. FASEB / 2:22412251.

Kanerva RL, Ridder GM, Stone LC, Alden CL (1987) Characterization of spontaneous and decalin-induced hyaline droplets in kidneys of adult male rats. Fd Chem Toxic 25:63-82. Kluwe WM (1981) Rapid, automated measurements

of urinary protein and glucose concentrations. Pharmacol

Meth

/

5:235-240.

Kurtz DT, Feigelson P (1978) Multihormonal control of the messenger RNA for the hepatic protein a2u-globulin. In Biochemical Actions of Hormones, Volume V. Ed., C Litwack. New York: Academic Press. Lane SE, Neuhaus OW (1972) Further studies on the isolation and characterization of a sex-dependent protein from the urine of male rats. Biochim Biophys

Acta 257:461-470.

Lehman-McKeeman LD, Rivera-Torres Ml, Caudill D (1990) Lysosomal degradation of a2u-globulin and a2u-globulin-xenobiotic conjugates. Toxicol Appl

Pharmacol103:539-548.

Lehman-McKeeman LD, Rodriguez PA, Caudill D, Fey ML, Eddy CL, Asquith TN (1991) Hyaline droplet nephropapthy resulting from exposure to 3,5,5,-trimethylhexanoyloxybenzene sulfonate. Toxicol

Appl

Pharmacol107:429-438.

Lehman-McKeeman LD, Rodriguez PA, Takigiku R, Caudill D, Fey ML (1989) d-Limonene-induced male-rat-specific nephrotoxicity: evaluation of the association between d-limonene and a2uglobulin. Toxicol Appl Pharmacol99:250-259. Lindblom H, Soderberg L, Cooper E, Turner R (1983) Urinary protein isolation by high performance ion-exchange chromatography./ Chromatogr

266:187-196.

Lock EA, Charbonneau M, Strasser J, Swenberg JA, Bus JS (1987) 2,2,4-Trimethylpentane-induced nephrotoxicity. II. The reversible binding of a TMP metabolite to a renal protein fraction containing cx2u-globulin. Toxicol Appl Pharmacol 91:182192.

Neuhaus OW, Flory W (1975) The effect of dietary protein on the excretion of (u2u, the sex-dependent protein of the adult male rat. Biochim Biophys Acta 41: 74-86. Neuhaus OW, Flory W (1978) Age-dependent changes in the excretion of urinary proteins by the rat. Nephron 22:570-576. Neuhaus OW, Flory W, Biswas N, Hollerman CE (1981) Urinary excretion of u2uglobulin by adult

HPLC Quantitation male rats following treatment agents. Nephron 28:133-140.

with nephrotoxic

Neuhaus OW. Lerseth DS (1979) Dietary control of the renal reabsorption and excretion of u2uglobulin. Kidney Int 16:409-415. Olson MI, Johnson JT, Reidy CA (1990) A comparison of male rat and human urinary proteins: implications for human resistance to hyaline droplet nephropathy. Toxicol Appl Pharmacot 102:524-536. Olson MJ, Mancini MA, Garg BD, Roy AK (1988) Leupeptin-mediated alteration of renal phagolysosomes: similarity to hyaline droplet nephropathy of male rats exposed to unleaded gasoline. Toxicol Lett 41~245-254. Pevsner J, Reed RR, Feinstein PC, Snyder SW (1988) Molecular cloning of odorant-binding protein: member of a ligand carrier family. Science 241:336-339. Ratge 0, Wisser H (1982) Urinary protein profiling by high-performance gel permeation chromatography. 1 Chromatogr 230:47-56. Ridder GM, Von5argen EC, Alden CL, Parker RD (1990) Increased hyaline droplet formation in male rats exposed to decalin is dependent on the presence of a2uglobulin. Fund Appl Toxicol 15:732-743.

of du-Globulin

Roy AK (1977) Early events in the steroidal regulation of cY2u-globulin in rat liver. Evidence for both androgenic and estrogenic induction. Eur J Biochem 73: 537-543. Roy AK, Neuhaus OW, Harmison HR (1966) Preparation and characterization of a sex-dependent rat urinary protein. Biochim Biophys Acta 127:72-81. Sellers AL, Goodman HC, Marmorston (1950) Sex difference in proteinuria Amer J Phys~ol163:662-667.

J, Smith M in the rat.

Smyth RE (1979) Fecal cup for coilection of feces in male rats. Laboratory Animal Science 29:677-678. Strasser J, Charbonneau M, Borghoff SJ, Turner MJ, Swenberg JA (1988) Renal protein droplet formation in male Fischer 344 rats after isophorone (IPH) treatment. Toxicologist 8:136 (abstract). Swenberg IA, Short 5, Borghoff S, Strasser J, Charbonneau M (1989) The comparative pathobiology of u2u-globulin nephropathy. Toxicol Appl Pharmacol97: 35-46. Vandoren G, Mertens B, Heyns W, Van Baelen H, Rombauts W, Verhoeven G (1983) Different forms of ~2u-globulin in male and female rat urine. Eur J Biochem 134:175-181. Weber MH (1988) Urinary protein analysis. / Chromatogr 429:315-344.

247