Quantitative determination of growth hormone by immunoblotting

ANALYTICAL

BIOCHEMISTRY

19 1,268-271

(1990)

Quantitative Determination lmmunoblotting Emilio

Fernbdez

and John

of Growth Hormone by

J. Kopchick’

Department of Zoologicaland Biomedical Sciences,Molecular and Cellular Biology Program and Edison Animal Biotechnology Center, Ohio University, Athens, Ohio 45701

Received

May

22,199O

Blotting techniques have been extensively used, not only analytically for protein identification, but also preparatively to isolate and purify specific proteins from a large variety of cellular extracts and biological fluids. The process involves the separation of proteins by sodium dodecyl sulfate-polyacrylamide gel electrophoresis, transfer of proteins to nitrocellulose membranes, and immunostaining to identify proteins which often are at very low concentrations. Because of the quantitative interactions of the proteins with specific antibodies, we have coupled the immunoblot procedure with photographic and densitometric methods for the quantitative determination of bovine growth hormone (bGH). In this way, the method is suitable for bGH detection and quantitation for a small number of samples by use of a single Western blot analysis. The sensitivity of this method permits determinations of bGH to 0.5 ng. The method uses a comparative procedure in which purified bovine growth hormone is used as a standard. 63 1900

Academic

Press,

Inc.

One of the most useful tools in gene regulation studies is the utilization of reporter gene systems for which expression can be easily detected and quantified. Many different proteins with structural, enzymatic, or regulatory functions are used as genetic markers to study gene expression. However, some of these proteins/peptides require prolonged and expensive methods for quantification in spite of the relatively simple analytical procedures used for their identification. Growth hormones as well as other peptide hormones are capable of regulating metabolic functions at very low concentrations. Also, growth hormone (GH)’ genes ’ To whom 268

correspondence

should

be addressed.

have been used by us and others as reporter molecules in gene expression studies (5-8). Although a number of GH detection methods exist (e.g., immunoprecipitation, Western blot, RIA, ELISA), RIA and ELISA are used to accurately determine GH levels (3,4). However, only human (h) GH has a commercially available kit for its quantitation. For other GHs, each laboratory must gen-

erate GH detection

and quantification

systems.

Bovine (b)GH is a single-chain polypeptide with a molecular mass of -22,000 Da (9-11). It contains 191 amino acids and contains four cysteine residues at positions 53, 164, 181, and 189 (12). Two disulfide bonds form between residues 53 and 164 and 181 and 189, which result in a large and small loop, respectively. Immunoblotting is routinely used by us to detect bGH secreted by either stable or transiently transfected cell lines and to detect bGH in the serum of transgenic mice. However, when bGH quantification is required one must use RIA or ELISA in addition to the immunoblotting. McGrane et al. (4) developed an ELISA to quantify bGH in serum of transgenic mice, but such an assay is not commercially available. The method described below, as well as RIA or ELISA, is based on the fact that antigen-antibody interactions are quantitative in the presence of an excess of antibody and, in this manner, linear correlations can be obtained between the amount of antigen present in the sample and the final measurement obtained in a particular assay. In the present report we describe a densitometric method coupled with

* Abbreviations used: bGH, bovine growth hormone; BSA, bovine serum albumin; DMEM, Dulbelco’s modified Eagle’s medium; ELISA, enzyme-linked immunosorbent assay; GH, growth hormone; OD, optical density; PA, peak area; PBS, phosphate-buffered saline; R, linear regression coefficient; RIA, radioimmunoassay; SDSPAGE, sodium dodecyl sulfate-polyacrylamide gel electrophoresis; TEMED, tetrametbyl ethylene diamine; TBS, Tris-buffered saline; vis, visible light. 0003-2697/90$3.00 Copyright 0 1990 by Academic Press, Inc. All rights of reproduction in’ any form resewed.

DETERMINATION

OF

GROWTH

HORMONE

the Western blotting technique which enables us to detect and quantitatively determine the bGH concentration in the same assay. MATERIALS

AND

METHODS

Materials. Purified bGH was obtained from Upjohn Co. (Kalamazoo, MI). Nitrocellulose membranes (0.2 pm) were purchased from Schleicher & Schuell (Keene, NY). A protein gel electrophoresis system and the transfer apparatus, as well as TEMED, acrylamide, bisacrylamide, gelatin, ammonium persulfate, and the color development reagent (4-chloro-1-naphthol) were obtained from Bio-Rad (Richmond, CA). Other chemicals were obtained from Sigma Chemical Co. (St. Louis, MO). Rabbit polyclonal anti-bGH serum was a gift of Dr. Fritz Rottman (Case Western Reserve University, Cleveland, OH). Goat anti-rabbit IgG-peroxidase conjugate was obtained from Boehringer-Mannheim (Indianapolis, IN). Black and white continuous tone transparency films (Polapan CT) were purchased from Polaroid (Cambridge, MA). Sample preparation. Purified bGH from a 0.2 mg/ml stock solution in (1%) BSA was diluted in freshly prepared culture media (Ham’s F-10 containing: 1.2 g/liter CO,HNa; 10% Nu-serum; 2.5% horse serum, and 10 mg/liter gentamicin) to a final concentration of 0.5 pg/ ml and used as a standard. This solution was stable for at least 3 months at -20°C. Different amounts of bGH, ranging from 0.5 to 20 ng, were subsequently diluted to 45 ~1 with culture media and mixed with 5 ~1 of sample buffer (1.0% glycerol, 1.5% fl-mercaptoethanol, 0.9% SDS, 0.04% bromphenol blue, 15 mM Tris, final concentrations; pH 6.7). Samples were boiled at 100°C for 5 min before loading onto SDS-polyacrylamide minigels. When biological samples of unknown bGH concentration were analyzed, different volumes were diluted to 45 ~1, mixed with 5 ~1 of sample buffer, and processed as described. SDS-polyacrylamide gel electrophoresis (SDSPAGE). Resolution of bGH using a SDS-PAGE minigel system (5 X 8 cm) was performed according to the procedure described by Laemmli (2) with minor modifications. Briefly, 12% (w/v) acrylamide: bisacrylamide (29:l) minigels were polymerized in gel running buffer (0.1% SDS in 0.4 M Tris, pH 8.9) by additions of 0.6% ammonium persulfate and 0.0005% TEMED. A 3% stacking gel, polymerized in 0.125 M Tris 1% SDS, pH 6.8, was positioned 5-7 mm above the running gel. The electrode buffer was 0.2 M glycine and 0.1% SDS in 0.025 M Tris, pH 8.3. Electrophoresis was carried out at 40 V for 2.5-3 h at room temperature. Electrophoretic transfer. Electroblotting of the proteins to nitrocellulose membranes was performed according to the procedure described by Burnette (1) with slight modifications. Following electrophoresis, the

BY

IMMUNOBLOTTING

269

stacking gel was removed and the running gel and a section of nitrocellulose membrane of the same size were soaked for 5-7 min in transfer buffer (96 mM glytine, 20% methanol, and 0.01% SDS in 12.5 mM Tris, pH 8.3). Gel and nitrocellulose were mounted in the transfer apparatus between several pieces of wet filter paper (gel blot paper GB002; Schleicher & Schuell) avoiding air bubbles. Transfer was carried out at 100 V for l-l.5 h. No differences were observed after overnight transfer (~15 h) at 25 V (data not shown). A system to refrigerate the buffer was necessary when transfers were performed at 100 V. Immunostaining of the nitrocellulose membranes. After transfer, nitrocellulose membranes were carefully removed from the gel and blocked in a 2% gelatin solution in TBS buffer (0.5 M NaCl, 50 mM Tris, pH 7.5) for l-2 h. After rinsing once with PBS-Tween 20 buffer (0.15 M NaCl, 0.05% Tween 20, 20 mM Na-phosphate, pH 7.5) to remove the excess of gelatin, membranes were incubated overnight (>15 h) with the primary antibody (rabbit anti-bGH serum in 1% gelatin-TBS). Membranes were washed 3 X 5 min with PBS-Tween 20 and incubated with a second goat anti-rabbit IgG peroxidase conjugate (l/1000 dil. in 1% gelatin-TBS) for at least 3 h. After treatment with the second antibody, membranes were washed 3 X 5 min with PBS-Tween 20 and developed for lo-20 min in a solution containing 0.015% hydrogen peroxide, 0.05% 4-chloro-1-naphthol, 16.6% methanol in TBS. Membranes were thoroughly rinsed with water to stop the reaction and to remove residual dye. Densitometric quantitation. Membranes were air dried at room temperature and photographed using 35mm black/white continuous tone transparency films (IS0 125; Polapan CT) and a f:50,1:3.5, FD-macro lens (Canon) at settings of l/60 s and fi5.6. Films were developed according to the manufacturer’s instructions. Developed films were then scanned at 690 nm using a gel scan program of a Response uv/vis scanning spectrophotometer (Corning) equipped with a 0.1 X 3-mm slit plate. Wavelength scanning (from 340 to 750 nm) was performed on overexposed vs underexposed film to select the optimal wavelength for subsequent studies.

RESULTS

AND

DISCUSSION

Although SDS-PAGE gels of variable sizes can be used depending on the sample requirements, we routinely use SDS-PAGE minigels (5 X 8 cm) in a standardized procedure for testing bGH gene expression levels in different biological fluids including culture fluids, cell extracts, animal sera, etc. This size gel permits loading up to 15 samples with an acceptable degree of resolution. Figure la. shows a standard Western blot of a minigel

270

FERNANDEZ

lal

0

0.5

1.0

2.0

ng

4.0

of

6.0

10.0

15.0

20.0

bGH/well

FIG. 1.

Western blot analysis of increasing levels of bGH. (a) After electrotransfer, a nitrocellulose membrane was immunostained by a double antibody peroxidase-conjugated method as described under Materials and Methods. Lanes A to I contain 0,0.5, 1, 2, 4, 6, 10, 15, and 20 ng of bGH, respectively. (b) Densitometry of the 35mm black and white transparency film from the nitrocellulose membrane in (a).

loaded with increasing amounts of purified bGH in cell culture media after immunostaining of the nitrocellulose filter. Following photography, bGH levels were determined by densitometry (Fig. lb). Optical densities, as well as the peak areas, were plotted as linear functions of the amounts of bGH loaded on the gel (R = 0.9791 for optical densities; R = 0.9823 for peak areas; P < 0.001 in both cases; Fig. 2). Linearity, however, was restricted to a mass of bGH ranging from 0.5 to 20 rig/well. Although linearity in this range remains constant between analyses, densitometric analysis of the film can vary depending on factors such as the performance of the antibodies, colored reagent, or films. Therefore, a standard curve in the range from 0.5 to 20 rig/well must be analyzed together with the samples. Exponential and polynomic (second through fifth orders) regression fits of the data were also performed in order to expand the dynamic range. Only linear and polynomic adjustments yielded acceptable correlation indexes. Second-order fitting of the data can be used when the bGH mass exceeds 20 ng; however, as the bGH mass approximates to the asymptotic limit of the curve, quantification becomes less precise. When polynomic fittings of both optical densities were used for bGH

AND

KOPCHICK

masses less than 20 ng, the numeric value of the coefficients in second and higher order terms were less than l/100 of the coefficient in the first-order term. Therefore we considered this influence insignificant relative to those terms over the linear fitting. Adjustments of third order and over do not accurately represent the averaged results for these types of experimental data. Sensitivity of bGH detection using this system was 0.1-0.2 rig/well. However, bGH levels of less than 0.5 rig/well could not be properly quantified since the standard curve was not linear at this level (results not shown). Nevertheless, this sensitivity is in the same range previously reported for chicken and bGH RIAs (3). From our results, concentrations as low as lo-15 rig/ml of bGH per sample can be measured by this method. For lower bGH concentrations either lyophilization or other methods for concentrating samples can be used. However, as the total protein concentration in the sample increases, the sensitivity of bGH quantitation can be affected, as in the case of culture media containing high levels of serum (e.g., 12.5% horse-serum and 2.5% fetal bovine serum in DMEM or Ham’s F-10; data not shown). Nevertheless, this method has been used to determine bGH levels in serum of transgenic mice (N. Y. Chen and J. J. Kopchick, unpublished resuits) . Both the efficiency of transfer to the nitrocellulose membrane and the degree of denaturation of the protein can affect the accuracy of the procedure. High molecular weight proteins require prolonged electrophoretic transfer times (1). On the other hand, the renaturation of the target protein during the transfer process could

0.004

. 0

.

.

.

I 5

.

.

.

.

I 10

.

.

.

.

I 15

.

.

.

I 20

’

’

’

.

I

25

bGW@a) FIG. 2.

Optical densities (OD, open boxes) and relative peak areas (Area, closed boxes) as a function of bGH (ng). The photographic and densitometric procedures as well as statistical correlations have been described in the text.

DETERMINATION

OF

GROWTH

also affect antibody binding. Therefore, a standard curve for each determination is essential. To our knowledge, this report represents the first case of the use of the Western blot as a detection and quantification method for determining the concentration of a specific protein in biological fluids. The approach consisting of photographic and densitometric methods coupled to the Western blot procedure is accurate, reproducible, and relatively inexpensive. Thus, a method for growth hormone determination alternative to RIA or ELISA is described.

ACKNOWLEDGMENTS We thank Drs. J. Cioffi, B. Kelder, T. Coleman, and W. Chen for helpful comments and suggestions in the preparation of this manuscript. E.F. is supported by a postdoctoral research fellowship from Servicio de Formacidn de Personal Investigador, Ministerio de Education y Ciencia (Spain). J.J.K. is supported in part by the State of Ohio’s Eminent Scholar Program, which includes a grant from Milton and Lawrence Goll.

HORMONE

BY

271

IMMUNOBLOTTING

REFERENCES 1. Burnette,

W. N. (1981)

And.

2. Laemmli,

U. K. (1970)

Nature

Biochem. (London)

112,195-203. 227,

680-685.

3. Leung, F. C., Taylor, J. E., Steelman, S. L., Bennett, C. D., Rodkey, J. A., Long, R. A., Serio, R., Weppelman, R. M., and Olson, G. (1984) Gen. Comp. Endocrinol. S&389-400. 4. McGrane, M. M., deVente, J., Jun, J., Park, E., Wynshow-Boris, A., Wagner, T., Rottman, F. M., and Hanson, R. W. (1988) J. Bid. Chem. 263(23), 11,443-11,451. 5. Pasleau, F., Leung, F., and Kopchick, J. J. (1987) Gene 57,47-52. 6. Martin-Gallardo, A., Montoya-Zavala, M., Kelder, B., Tayior, J., Chen, H., Leung, F. C., and Kopchick, J. J. (1988) Gene 70,51-56. 7. Kelder, B., Chen, H., and Kopchick, J. J. (1989) Gene 76, 75-80. 8. Selden, R. F., Howie, K. B., Rowe, M. E., Goodman, H. M., and Moore, D. D. (1986) Mol. Cell. Biol. 6,3173-3179. 9. Andrews, P. (1966) Nature fLondon) 209,155-157. 10. Dellacha, J. M., Santone, J. A., and Faiferman, L. (1966) Experientia 22, 16-17. 11. Miller, W. L., Martial, J. A., and Baxter, J. D. (1980) J. Biol. Chem. 255,7521-7524. 12. Hara, K., Chen, C. J., and Sonenberg, M. (1978) Biochemistry 17, 550-556.