Measurement of Human Growth Hormone Receptor Messenger ...

0021-972X/97/$03.00/0 Journal of Clinical Endocrinology and Metabolism Copyright © 1997 by The Endocrine Society

Vol. 82, No. 2 Printed in U.S.A.

Measurement of Human Growth Hormone Receptor Messenger Ribonucleic Acid by a Quantitative Polymerase Chain Reaction-Based Assay: Demonstration of Reduced Expression after Elective Surgery* MAJLIS HERMANSSON†, RUTH B. WICKELGREN†, FOLKE HAMMARQVIST, ¨ M, JAN WERNERMAN, RAGNAR BJARNASON, INGMAR WENNSTRO ¨ BJORN CARLSSON, AND LENA M. S. CARLSSON Research Center for Endocrinology and Metabolism, Department of Internal Medicine, Sahlgrenska Hospital (M.H., R.B.W., R.B., B.C., L.M.S.C.), and the Department of Pediatrics, East Hospital (R.B.), Goteborg University, Goteborg; and the Department of Surgery, St. Goran Hospital (F.H., I.W.), and the Anesthesiological Metabolism Unit, Clinical Research Center, Department of Anesthesiology and Intensive Care, Huddinge University Hospital (J.W.), Karolinska Institute, Stockholm, Sweden ABSTRACT Studies of GH receptor (GHR) gene expression in human tissues have been hampered by the limited amount of tissue available for analysis and the low sensitivity of conventional methods. We have developed a quantitative reverse transcriptase-PCR assay for measurement of GHR messenger ribonucleic acid levels in small human tissue biopsies. To compensate for sample to sample variation, an internal RNA standard, which differs from the wild-type GHR transcript by only a few nucleotides, was reverse transcribed and amplified together with the GHR transcripts. PCR was carried out using one biotinylated primer to permit the purification of single stranded PCR products on streptavidin-coated microtiter plates. The ratio between the wild-type and mutated transcripts was determined by two separate minisequence reactions in which a primer, annealed immediately 39 of a variable nucleotide, was extended by a single 3H-labeled nucleotide, complementary to either the wild-type or mutated se-

quence. The assay range was 0.125– 8 3 105 transcripts/sample, the mean intraassay coefficient of variation was 8.7%, and the lower limit of detection was 0.125 3 105 transcripts/sample. GHR messenger ribonucleic acid levels were detectable in small amounts (10 –100 ng) of total RNA extracted from adipose tissue, skeletal muscle, and liver. The GHR gene expression in liver was approximately 10-fold higher than that in skeletal muscle, whereas intermediate levels were found in adipose tissue. In nine patients undergoing elective abdominal surgery, GHR gene expression in skeletal muscle was reduced on day 3 after surgery compared to the baseline level. The decrease in GHR gene expression was accompanied by a decrease in skeletal muscle glutamine. This suggests that the postoperative protein catabolism may be caused at least partly by acquired GH insensitivity due to reduced expression of the GHR gene. (J Clin Endocrinol Metab 82: 421– 428, 1997)

T

jects. For example, endogenous GH levels in normally growing children vary over a wide range (5), and the response to exogenous GH in short children is highly variable (6). The molecular mechanisms behind differences in GH responsiveness are unknown, but possible explanations include alterations at the level of the GHR. There are many reports on structural defects in the GHR gene in patients with GH insensitivity (reviewed in Ref. 3), but much less is known about variations in GHR gene expression in patients with reduced GH sensitivity. Although GHR messenger ribonucleic acid (mRNA) has been demonstrated in several human tissues and cell types (1, 7–12), detailed studies of GHR gene expression in human disease have in many cases been hampered by the low sensitivity of conventional techniques, such as Northern blot and ribonuclease protection assays. Reverse transcription (RT) followed by amplification by the PCR allows the analysis of low abundance messages, such as the GHR mRNA, in a small amount of total RNA. We have developed a quantitative RT-PCR (Q-RT-PCR) assay for the measurement of GHR mRNA levels in human biopsies based on a competitive PCR strategy described by Ikonen et al (13). Little is known about the regulation of the GHR gene under physiological and pathophysiological conditions. One example is the acquired GH insensitivity described in

HE GROWTH-PROMOTING and metabolic actions of GH are mediated by specific GH receptors (GHR) on the surface of the target cells (1; reviewed in Ref. 2). In Laron’s syndrome, the absence of functional GHRs results in GH insensitivity and severe growth failure (reviewed in Ref. 3). However, in most cases, GH resistance is not inherited, but is instead acquired and transient. Acquired GH insensitivity has been implicated in several patient groups with systemic illness (reviewed in Ref. 4). Furthermore, several clinical observations indicate that there is considerable interindividual variation in GH responsiveness even in healthy subReceived July 12, 1996. Revision received October 14, 1996. Accepted October 21, 1996. Address all correspondence and requests for reprints to: Dr. L. M. S. Carlsson, Research Center for Endocrinology and Metabolism, Department of Internal Medicine, Sahlgrenska Hospital, Goteborg University, S-413 45 Goteborg, Sweden. E-mail: [email protected]. * This work was supported by the Swedish Medical Research Council (Grants 11285, 11502, 11331, 11576, and 04210), Emil och Wera Cornells Stiftelse, the Stockholm County Council, the Department of Public Health and Medical Science, and the Department of Research Development and Teaching, Stockholm, Sweden. † M.H. and R.W. contributed equally to this manuscript and are considered to be equal first authors.

421

422

JCE & M • 1997 Vol 82 • No 2

HERMANSSON ET AL.

patients with hypercatabolism, for example after surgery (14). The mechanisms behind this resistance are incompletely understood. We have used our Q-RT-PCR assay to measure the GHR mRNA levels in skeletal muscle from patients undergoing abdominal surgery to determine whether the acquired GH resistance in this group of patients may be due to reduced expression of the GHR gene. In addition, muscle amino acids and whole body nitrogen balance during the study period were determined. The decrease in muscle glutamine in particular is a well described parameter, reflecting the degree of muscle protein catabolism (15, 16). Subjects and Methods Subjects and samples The study was approved by the ethical committee of Goteborg University and Karolinska Institute. Cultured IM-9 cells and tissues obtained during surgery were rapidly frozen in liquid nitrogen and stored at 280 C. Total RNA was isolated from the biopsies using guanidium thiocyanatephenol-chloroform extraction, essentially as described by Chomczynski and Sacchi (17). Metabolically healthy patients (two women and seven men; mean age 6 sem, 63 6 4 yr), under 80 yr of age without weight loss, undergoing elective abdominal surgery were included in the study. The patients underwent colonic resection (n 5 6), anterior resection of the rectum (n 5 1), rectal amputation (n 5 1), or hemihepatectomy (n 5 1). Preoperatively, after induction of anesthesia a muscle biopsy was taken from the lateral portion of the quadriceps femoris muscle using the percutaneous needle biopsy technique. Muscle tissue of about 30–40 mg wet weight (ww) was used for determination of GHR mRNA and amino acid determination. The ww was determined immediately, and the muscle sample was then frozen in liquid nitrogen and stored at 280 C until analysis. A second muscle biopsy was taken on the third postoperative day under local anesthesia confined to the skin and fascia only. At the time of biopsy, blood samples were drawn from the antecubital vein for the determination of insulin-like growth factor I (IGF-I), IGF-binding protein-3 (IGFBP-3), and GH-binding protein (GHBP). The patients were given a standardized glucose infusion (3.0 g glucose/kg BWzday) postoperatively.

Primers Primers were purchased from Scandinavian Gene Synthesis (Koping, Sweden). Localization of the GHR primers is shown in Fig. 1, and the sequences of all primers are listed in Table 1.

Generation of synthetic RNA standards A 377-bp fragment of the human GHR complementary DNA (cDNA), extending from nucleotides 201–577 (1), was amplified by RT-PCR, using primers GHR3 and GHR4, and subcloned into pGEM-T (Promega, Madison, WI), generating pGHR34. The identity of the subcloned fragment was verified by DNA sequencing using the PRISM Sequenase Terminator Double Stranded DNA Sequencing Kit (Applied Biosystems, Foster City, CA) and an Applied Biosystems 373A automatic sequencer. Two point mutations were introduced at nucleotides 335 (G to C) and 333 (C to A) using PCR-based mutagenesis (18), introducing a BglII site and destroying a PvuII site. Two overlapping fragments were produced with pGHR34 as template in two separate PCR reactions using primers GHR3 and GHR4M, and GHR3M and GHR4, respectively. A mixture of the two amplified fragments was used as the template for amplification with primers GHR3 and GHR4, regenerating a 377-bp GHR cDNA fragment with the desired mutations. The mutated GHR fragment was subcloned into pGEM-T, generating pGHR34M, and the mutations were verified by DNA sequencing. RNA corresponding to the wild-type GHR standard (GHR-wt) and the mutated internal standard (GHR-mut) was obtained by in vitro transcription with SP6 RNA polymerase (Promega) from NcoI (Promega)-linearized pGHR34 and pGHR34M, respectively. In each case, a single band of complementary RNA of the expected size was observed on an ethidium bromide-stained 1.5% agarose gel. In later experiments, a strategy including purification of polyadenylated [poly(A)] transcripts was applied to ensure that full-length transcripts were produced. To obtain in vitro transcription products with poly(A) tails, restriction fragments from pGHR34 and pGHR34M, respectively, were subcloned into the pSP64 poly(A) vector (Promega), generating pGHR34A1 and pGHR34MA1. Synthetic RNA standards produced from EcoRI (Promega)-linearized pGHR34A1 and pGHR34MA1 were purified with PolyATract mRNA Isolation Systems (Promega).

cDNA synthesis and PCR First strand cDNA was generated from RNA in 1 3 reverse transcriptase buffer [50 mmol/L Tris-HCl (pH 8.3; 42 C), 5 mmol/L KCl, 1 mmol/L MgCl2, 10 mmol/L dithiothreitol, and 0.05 mmol/L spermidine; Promega] with 7 U AMV reverse transcriptase (Promega), 20 U RNAsin (Promega), 1.5 mmol/L deoxy (d)-NTP (Promega or Boehringer Mannheim, Mannheim, Germany), and 0.5 mg random hexamers (Promega) in a final volume of 20 mL. After annealing at 22 C for 5 min, the reaction was carried out at 42 C for 50 min and terminated at 70 C for 5 min. cDNA was used as the template for PCR in 1 3 PCR buffer [10 mmol/L Tris-HCl (pH 8.3; 20 C), 50 mmol/L KCl, and 1.5 mmol/L MgCl2; Boehringer] with 0.3 mmol/L dNTP (Promega), 3 U Taq Polymerase (Boehringer), primer B-GHR3 (15 pmol), and primer GHR5 (50 pmol) in a final volume of 100 mL. PCR was performed using GeneAmpPCR system 9600 (Perkin-Elmer/Cetus, Norwalk, CT) for 30 cycles (15-s denaturation at 94 C, 15-s annealing at 57 C, and 30-s elongation at 72 C). Negative controls were always included during both cDNA synthesis and PCR to confirm that there was no contamination.

Assay procedure FIG. 1. Schematic representation of GHR cDNA and localization of the primers used for the Q-RT-PCR assay.

To avoid contamination, the Q-RT-PCR assay was carried out in four distinct laboratory areas; sample preparation area, pre-PCR, PCR, and

TABLE 1. Primers used in the Q-RT-PCR

a

Primer

Sequence (59 to 39)

Location according to Ref. 1 and 19

B-GHR3 GHR3 GHR4 GHR3M GHR4M GHR5 GHR-SEQ B-CP1 CP2 CP-SEQ

Biotin-CTG GAC AGA TGA GGT TCA CTG GAC AGA TGA GGT TCA CAT ACT CCA GAA CCA TCC GAA AAC AGAa TCaT TAC TTT AAA GTA AGa A Ta CT GTT TTC GTG CTT CCC ATC TCA CTT AGG TAA ACG ATG AAT TAA AGT AA Biotin-GC AGA CAA GGT CCC AAA GA GCA GCG AGA GCA CAA AGA GGA CTT GCC ACC AGT GCC ATT A

201–218 201–218 577–560 325–342 342–325 532–515 358–336 76–94 231–209 608–591

Mutated nucleotide as described in Subjects and Methods.

QUANTITATIVE PCR FOR GHR mRNA

423

FIG. 2. Assay procedure for the Q-RT-PCR assay used for measurement of GHR mRNA. A sample of total RNA, containing the target mRNA, was mixed with a fixed amount of the internal standard (GHR-mut), reverse transcribed, and amplified by PCR. The use of one biotinylated primer allowed purification of single stranded PCR products on streptavidin-coated microtiter wells. The ratio between GHR-mu and GHR-wt transcripts was determined by two separate minisequencing reactions in which a primer complementary to the sequence immediately 39 of the variable nucleotide was extended with a single labeled nucleotide complementary to either the wild-type or mutated nucleotide. A standard curve was generated using serial dilutions of synthetic GHR-wt RNA (GHR-wt). post-PCR. The assay procedure is outlined in Fig. 2. Samples, consisting of total RNA extracted from human tissues or GHR-wt RNA (for construction of the standard curve), were mixed with 1 3 105 molecules of the internal standard, GHR-mut RNA. The RNA mixture was reverse transcribed into cDNA, and PCR was carried out as described above, using primers GHR5 and B-GHR3. To immobilize and purify single stranded PCR products, aliquots (10 mL) of the PCR reactions were mixed with 50 mL PBS [0.14 mol/L NaCl, 0.01 mol/L sodium phosphate buffer (pH 7.4), and 0.1% Tween-20] and dispensed into streptavidincoated microtiter plates (Streptavidin Covalent Strips, Wallac Oy, Turku, Finland). Plates were sealed, incubated at 37 C for 1.5 h with gentle agitation, and washed four times with TENT buffer [40 mmol/L TrisHCl (pH 8.8), 1 mmol/L ethylenediamine tetraacetate, 50 mmol/L NaCl, and 0.1% Tween-20] at room temperature. The immobilized PCR products were denatured by treatment with 100 mL 50 mmol/L NaOH-150 mmol/L NaCl for 5 min at room temperature. The denaturation step was repeated, and the plates were washed four times with TENT buffer. The ratio between GHR-wt and GHR-mut sequences was determined by two separate minisequence reactions, in which the primer GHR-SEQ, complementary to the sequence immediately 39 of the variable nucleotide, was extended with a radiolabeled nucleotide complementary to either the GHR-wt or the GHR-mut sequence. This reaction was carried out at 55 C for 10 min in PCR buffer [10 mmol/L Tris-HCl (pH 9.0; 25 C), 50 mmol/L KCl, 1.5 mmol/L MgCl2, and 1% Triton X-100; Promega] con-

taining 2 U Taq-polymerase (Promega), the primer GHR-SEQ (0.2 mmol/ L), either [3H]dGTP (64 Ci/mmol; TRK 625, Amersham International, Little Chalfont, UK) or [3H]dCTP (31 Ci/mmol; TRK 627, Amersham; 0.2 mmol/L), and dideoxy-ATP (Boehringer Mannheim; 0.8 mmol/L) in a total volume of 50 mL. The microtiter plates were washed four times with TENT buffer and counted in a microliquid scintillation counter (Wallac Oy).

Application of the Q-RT-PCR assay Total RNA was isolated from skeletal muscle from nine patients undergoing abdominal surgery as described in Subjects and samples above. GHR mRNA levels were measured as described above with minor modifications. The abundance of GHR transcripts was related to the expression of a housekeeping gene, cyclophilin (19), and GHR mRNA levels were expressed as GHR transcripts per cpm cyclophilin. RT-PCR of cyclophilin was carried out using primers B-CP1 and CP2 (Table 1) for 26 cycles. [3H]dTTP (116 Ci/mmol; TRK 933, Amersham), dideoxy-GTP (Boehringer Mannheim), and the primer CP-SEQ (Table 1) were used in the cyclophilin minisequencing reaction.

Immunoassays GHBP. Plasma levels of GHBP were measured by ligand mediated immunofunctional assay (LIFA) as previously described (20). All

424

HERMANSSON ET AL.

FIG. 3. Standard curves from three separate experiments. The standard curves were generated by the Q-RT-PCR assay as described in Subjects and Methods, using 0.125– 8 3 105 GHR-wt transcripts and 1 3 105 GHR-mut transcripts.

JCE & M • 1997 Vol 82 • No 2

FIG. 4. Assay linearity. The abundance of GHR mRNA was analyzed by the Q-RT-PCR assay as described in Subjects and Methods, using three different amounts of total RNA extracted from human liver. The results are shown as the number of transcripts per mg total RNA.

TABLE 2. Precision of the Q-RT-PCR assay

Intraassay precision Replicates Mean (transcripts/mg total RNA) SD (transcripts/mg total RNA) CV (%)

25 ng total RNA

50 ng total RNA

100 ng total RNA

10 2.73 0.19 7.1

9 2.98 0.20 6.9

10 3.61 0.58 15.9

samples were measured in the same assay, and the intraassay coefficient of variation was 7%. The reagents were kindly provided by Genentech (South San Francisco, CA). IGF-I. Plasma levels of IGF-I were measured by RIA (Nichols Institute Diagnostics, San Juan Capistrano, CA). IGF-I was separated from the binding protein using an acid-ethanol and alkaline precipitation step. The intraassay coefficient of variation was 5%. IGFBP-3. Plasma levels of IGFBP-3 were measured by RIA (Nichols Institute Diagnostics). The intraassay coefficient of variation was 4%.

Amino acid determination The free amino acids in skeletal muscle were determined with special emphasis on glutamine and branched chain amino acids (21). To denature the proteins, the biopsy specimens were homogenized in 4% sulfosalicylic acid containing nor-leucine as an internal standard. The precipitated proteins were sedimented by centrifugation, and the supernatant was used for determination of free amino acids. The free amino acids in the supernatant were separated by ion exchange chromatography (Alpha Plus, LKB, Bromma, Sweden), derivatized with o-phthaldialdehyde on DC-6 ion exchange resin (Durrum, CA) and lithium citrate buffers, and quantified using a fluorescence detector (21).

Nitrogen balance determination Urine was collected over 24-h periods. By determination of the nitrogen content of the urine, nitrogen losses could be calculated. The urinary nitrogen content was determined by chemiluminescent nitrogen analyzer (771 C pyroreactor, 720 C nitrogen detector, Antek Instruments, Houston, TX). The external nitrogen losses per 24 h were approximately 1.5 g and were included in the calculation. The cumulative nitrogen balance during the study period was calculated.

Statistical analysis For evaluation of GHR mRNA, GHBP, IGF-I, IGFBP-3, and amino acids before and after surgery, a paired t test was used. Data are represented as the mean 6 sem.

Results Assay range and sensitivity

Figure 3 shows three Q-RT-PCR standard curves obtained by serial dilutions of GHR-wt RNA mixed with a fixed amount (1 3 105 molecules) of GHR-mut RNA. The RNA samples were reverse transcribed, amplified, and detected as described in Subjects and Methods. The assay range for the Q-RT-PCR assay was 0.125– 8 3 105 GHR-wt transcripts/ sample. The lower limit of detection was 0.125 3 105, as defined by Rodbard et al. (22) to be equivalent to the concentration corresponding to the mean absorbance of zero plus twice the sd. Assay precision

Replicates of samples containing 25, 50, or 100 ng total RNA extracted from cultured IM-9 cells were analyzed for the assessment of intraassay precision (Table 2). The intraassay coefficients of variation ranged from 6.9 –15.9%, and the average coefficient of variation was 8.7%. Linearity of the assay

The linearity of the assay was determined by measurements of GHR mRNA by the Q-RT-PCR assay in serial dilutions of total RNA (40, 20, and 10 ng) extracted from human liver. After correction for the dilution factor, approximately 6.8 3 106 GHR transcripts/mg total RNA were found in all three samples (Fig. 4), indicating that the assay is linear. GHR mRNA abundance in skeletal muscle, adipose tissue, and liver

To compare GHR gene expression in different human tissues, total RNA was extracted from liver (n 5 1), skeletal muscle (n 5 3), and adipose tissue (n 5 5), and GHR mRNA concentrations were measured by Q-RT-PCR. The GHR gene expression was highest in liver (13 3 106 transcripts/mg total RNA), whereas the expression in skeletal muscle and adipose tissue was approximately 10 and 3 times lower, respectively


(Fig. 5). The expression levels also varied between subjects. GHR gene expression ranged between 2.4 –5.2 3 106 transcripts/mg total RNA in adipose tissue and between 0.4 and 1.9 3 106 transcripts/mg total RNA in skeletal muscle. GHR mRNA abundance in skeletal muscle biopsies from patients undergoing abdominal surgery

To determine whether surgical trauma is related to changes in GHR gene expression, the Q-RT-PCR assay was used to measure GHR mRNA in skeletal muscle from nine patients undergoing major abdominal surgery (Fig. 6A). GHR mRNA levels were significantly reduced from 450 6 85 transcripts/cpm cyclophilin at baseline to 291 6 48 transcripts/cpm cyclophilin on day 3 after surgery (P , 0.05). In addition, plasma levels of the GHR-related GHBP, IGF-I, and

425

IGFBP-3 were measured. GHBP levels were reduced from 232 6 40 pmol/L at baseline to 155 6 31 (P , 0.01), 138 6 30 (P , 0.01), and 160 6 33 (P , 0.05) on days 1, 2, and 3 after surgery, respectively (Figs. 6B and 7C). IGF-I decreased from 163 6 18 mg/L at baseline to 153 6 24 (P 5 NS), 135 6 21 (P , 0.01), and 123 6 24 (P , 0.01) mg/L on days 1, 2, and 3 after surgery, respectively (Fig. 7A). IGFBP-3 decreased from 2.36 6 0.15 mg/L at baseline to 1.99 6 27 (P 5 NS), 1.89 6 0.23 (P , 0.05), and 1.73 6 0.21 (P , 0.01) mg/L respectively on days 1, 2, and 3 after surgery (Fig. 7B). These changes were accompanied by a decrease in muscle free glutamine from 11.23 6 0.66 mmol/kg ww muscle at baseline to 5.57 6 0.81 mmol/kg ww muscle tissue on day 3 after surgery (P , 0.001; Table 3) and with an increase in muscle free branched chain amino acids from 0.35 6 0.03 mmol/kg ww muscle at baseline to 0.50 6 0.03 mmol/kg ww muscle tissue on day 3 after surgery (P , 0.05; Table 3). The total amino acid concentration decreased from 20.84 6 0.97 mmol/kg ww muscle at baseline to 13.28 6 1.23 mmol/kg ww muscle on day 3 after surgery (P , 0.001; Table 3). The cumulated nitrogen balance during the study period was 227.7 6 3.4 g. Discussion

FIG. 5. GHR gene expression in human tissues. GHR mRNA was measured by the Q-RT-PCR assay in total RNA extracted from liver (n 5 1), skeletal muscle (n 5 3), and adipose tissue (n 5 5) as described in Subjects and Methods.

FIG. 6. GHR gene expression in skeletal muscle biopsies from nine patients undergoing abdominal surgery. GHR mRNA abundance was measured by the Q-RT-PCR assay at surgery and on day 3 after surgery as described in Subjects and Methods. For comparison, plasma GHBP levels at the time of the biopsy are shown. *, P , 0.05, day 0 vs. day 3.

Quantitation of gene expression by means of Q-RT-PCR has become a standard method in several laboratories, and there are many alternative Q-RT-PCR protocols. The major methodological problems in using PCR as a quantitative method are caused by the nonquantitative conversion of RNA into cDNA and the fact that the amount of amplified fragment increases in a linear fashion for only a limited

426

JCE & M • 1997 Vol 82 • No 2

HERMANSSON ET AL.

TABLE 3. Amino acid concentration in skeletal muscle (millimoles per kg ww muscle)

Glutamine BCAA Total amino acids

FIG. 7. Plasma concentrations of GHBP, IGF-I, and IGFBP-3 in nine patients undergoing abdominal surgery. *, P , 0.05; **, P , 0.01 (day 0 vs. days 1, 2, and 3).

number of cycles (i.e. the plateau phenomenon). Furthermore, the extremely high sensitivity of the PCR makes DNA contamination an important issue. In this study we have tried to minimize these major sources of methodological errors.

Day 0

Day 3

P value

11.23 6 0.66 0.35 6 0.03 20.84 6 0.97

5.57 6 0.81 0.50 6 0.03 13.28 6 1.23

,0.001 ,0.05 ,0.001

The use of an appropriate internal standard has been shown to greatly improve the reproducibility of Q-RT-PCR assays (23, 24). To minimize differences in amplification efficiency, an ideal internal standard should be as similar as possible to its corresponding wild-type template and compete for the same set of primers. To accomplish this, we based our Q-RTPCR assay on a method described by Ikonen et al (13), in which the internal standard is identical to the wild-type sequence, with the exception of a few mutations that are introduced to allow separate detection of the two transcripts. The internal RNA standard is added to the samples before cDNA synthesis to control for variations in the efficiency of the RT reaction, and the corresponding cDNA then serves as a competitive template during the PCR. To avoid contamination, we perform the Q-RT-PCR assay in four distinct laboratory areas: sample preparation area, pre-PCR, PCR, and post-PCR (25). Negative controls were always included, during both cDNA synthesis and PCR, to confirm the lack of contamination. It is difficult to compare the quality and performance of different Q-RT-PCR assays because the validation is usually not published. In contrast to bioassays and RIAs, for which strict guidelines for assay validation are provided in the instructions to authors, there are no guidelines for how PCRbased assays should be validated. Because of the acknowledged pitfalls common to PCR, we suggest that validation of Q-RT-PCR assays should include assessment of assay precision and linearity. Measurement of GHR mRNA in samples containing different amounts of total RNA extracted from a human liver demonstrated parallelism of standard and unknown sample in our assay. Although the intraassay variability was in the same range as that for many immunoassays, in our experience the interassay variability was higher (;17–32% depending on transcript abundance in the sample; data not shown), suggesting that comparisons should preferably be made within the same assay. Estimation of the lower limit of detection directly provides important information for an immunoassay. A certain level of detection for the Q-RT-PCR assay, expressed as transcripts per sample, may imply that the assay can be used to measure mRNA levels in some tissues but not in others, as expression levels often vary between tissues. Our Q-RT-PCR assay was developed to allow studies of GHR gene expression in human tissues. The assay is very sensitive and can be used to measure GHR mRNA levels in total RNA isolated from needle biopsies. Biopsies of different human tissues are often taken to aid the diagnosis of various diseases, but unless PCR-based assays are used, the amount of tissue that can be obtained is too small to allow studies of low abundance mRNAs, such as GHR mRNA. Therefore, very little is known about the physiological and pathophysiological regulation of GHR gene expression in human tis-


sues. We used the Q-RT-PCR assay to measure the GHR mRNA levels in three of the major target tissues for GH (liver, skeletal muscle, and adipose tissue) and showed that GHR mRNA was clearly measurable in all samples tested. The GHR mRNA levels that we report are slightly higher than the levels measured by another Q-RT-PCR assay developed by Martini et al. (11). However, the relative difference between GHR gene expression in liver and muscle was similar and also fits well with reports of GHR gene expression in these tissues, measured by other techniques, in the rat (26). Indirect evidence suggests that many diseases and syndromes are accompanied by reduced sensitivity to GH. Hormone insensitivity may have several causes, including structural defects in the receptor, receptor down-regulation, or postreceptor defects. With the exception of patients with Laron’s syndrome and some children with idiopathic short stature, in whom GH resistance is genetic and caused by mutations in the GHR gene, (27; reviewed in Ref. 3), the mechanisms behind the GH resistance are unknown. Structural defects due to deletions or mutations in receptor genes are interesting and important because they provide information about the structure-function relationship of the receptor. Furthermore, structural defects can explain the hormone insensitivity in some patients. However, in most patients, hormone insensitivity is acquired, indicating that a decrease in the abundance or activity of the receptor or postreceptor molecules may be the cause. GH resistance is found in conditions associated with an increased catabolic rate, such as trauma, sepsis, and surgery (reviewed in Ref. 4). Typical changes in the GH-IGF-I axis in such conditions of GH resistance include up-regulated GH secretion, low levels of IGF-I and IGFBP-3, high levels of IGFBP-1, and reduced concentrations of GHBP (4, 14). A reduction in skeletal muscle mass is also a major consequence during catabolic illness. The loss of skeletal muscle mass is accompanied by well described metabolic changes in skeletal muscle, such as a decrease in protein synthesis (28), a reduction of muscle free glutamine, and increases in branched chain amino acids (29). The degree of catabolism can be assessed by the amount of nitrogen lost during the period studied. To determine whether these changes may be explained by changes in GHR gene expression, we used the Q-RT-PCR assay to measure GHR mRNA levels in skeletal muscle from nine subjects undergoing abdominal surgery. In addition, we measured GHBP, which corresponds to the extracellular domain of the GHR and is thought to be derived proteolytically from the cell membrane-bound receptor (1, 2, 30). The patients in the present study showed parallel decreases in GHR gene expression, plasma levels of GHBP, and muscle free glutamine as well as pronounced nitrogen losses. The changes in free glutamine and nitrogen losses indicate a pronounced state of muscle protein catabolism, with the loss of about 1 kg skeletal muscle during 3 days. It has been demonstrated that postoperative muscle protein catabolism can be reduced by treatment with recombinant human GH as an adjuvant to nutrition (31–33) as well as by the addition of glutamine to the parenteral nutrition (34). The results from the present study indicate that expression of the GHR gene in skeletal muscle is reduced postoperatively. Because skeletal muscle represents the bulk of peripheral tissue, this may

427

explain part of the metabolic events in catabolic illnesses characterized by an increased flow of substrate from peripheral protein stores to more central tissues, such as the splanchnic organs, and immunocompetent tissues (35–37). The factor mediating the reduction of GHR gene expression in skeletal muscle after major surgery is unknown. An interesting group of candidate molecules is the cytokines. Administration of several cytokines has been shown to induce catabolism (38, 39), and several cytokines are produced in response to trauma and infections (40). One of the cytokines, tumor necrosis factor-a, shows increased expression in adipose tissue in obesity and has been suggested to cause insulin resistance in skeletal muscle (41). A similar mechanism may be involved in the decrease in GHR gene expression in skeletal muscle in patients with inflammatory processes, for example after major surgery. Interestingly, cytokines, including interleukin-1b and tumor necrosis factor-a, inhibit GHR gene expression in rat hepatocytes in vitro (42). We conclude that the Q-RT-PCR assay described in this study is sensitive and reliable. Using this assay it was possible to demonstrate reduced GHR gene expression in skeletal muscle in patients after major surgery. Acknowledgments We thank Per-Arne Lundberg and Lisbeth Jonsson for the IGF-I and IGFBP-3 measurements. The skilled technical assistance of Mrs. Liselott Thunblad and Mrs. Marianne Zander is gratefully acknowledged.

References 1. Leung D, Spencer S, Cachianes G, et al. 1987 Growth hormone receptor and serum binding protein: purification, cloning and expression. Nature. 330:537–543. 2. Kelly PA, Djiane J, Postel Vinay MC, Edery M. 1991 The prolactin/growth hormone receptor family. Endocr Rev. 12:235–251. 3. Rosenfeld RG, Rosenbloom AL, Guevara Aguirre J. 1994 Growth hormone (GH) insensitivity due to primary GH receptor deficiency. Endocr Rev. 15:369 –390. 4. Ross RJ, Chew SL. 1995 Acquired growth hormone resistance. Eur J Endocrinol. 132:655– 660. 5. Martha P, Rogol A, Blizzard R, Shaw M, Baumann G. 1991 Growth hormonebinding protein activity is inversely related to 24-hour growth hormone release in normal boys. J Clin Endocrinol Metab. 73:175–181. 6. Attie K, Carlsson L, Rundle A, Sherman B. 1995 Evidence for partial growth hormone insensitivity among patients with idiopathic short stature. J Pediatr. 127:244 –250. 7. Carlsson B, Bergh C, Bentham J, et al. 1992 Expression of functional growth hormone receptors in human granulosa cells. Hum Reprod. 7:1205–1209. 8. Delehaye Zervas MC, Mertani H, Martini JF, Nihoul Fekete C, Morel G, Postel Vinay MC. 1994 Expression of the growth hormone receptor gene in human digestive tissue. J Clin Endocrinol Metab. 78:1473–1480. 9. Oakes SR, Haynes KM, Waters MJ, Herington AC, Werther GA. 1992 Demonstration and localization of growth hormone receptor in human skin and skin fibroblasts. J Clin Endocrinol Metab. 75:1368 –1373. 10. Tavakkol A, Elder JT, Griffiths CE, et al. 1992 Expression of growth hormone receptor, insulin-like growth factor 1 (IGF-I) and IGF-I receptor mRNA and proteins in human skin. J Invest Dermatol. 99:343–349. 11. Martini JF, Villares SM, Nagano M, et al. 1995 Quantitative analysis by polymerase chain reaction of growth hormone receptor gene expression in human liver and muscle. Endocrinology. 136:1355–1360. 12. Werther GA, Haynes K, Waters MJ. 1993 Growth hormone (GH) receptors are expressed on human fetal mesenchymal tissues: identification of messenger ribonucleic acid and GH-binding protein. J Clin Endocrinol Metab. 76:1638 –1646. 13. Ikonen E, Manninen T, Peltonen L, Syva¨nen A-C. 1992 Quantitative determination of rare mRNA species by PCR and solid-phase minisequencing. PCR Methods Appl. 1:234 –240. 14. Bentham J, Rodriguez Arnao J, Ross RJ. 1993 Acquired growth hormone resistance in patients with hypercatabolism. Horm Res. 40:87–91. 15. Hammarqvist F. 1989 Intravenous amino acid supply after trauma: the effects of glutamine on muscle protein metabolism. Stockholm: Karolinska Institute.

428

HERMANSSON ET AL.

16. Vinnars E, Hammarqvist F, von der Decken A, Wernerman J. 1990 The role of glutamine and its analogues in posttraumatic muscle protein and amino acid metabolism. J Parenter Enteral Nutr. 14:125S–129S. 17. Chomczynski P, Sacchi N. 1987 Single step method of RNA isolation by acid guanidium thiocyanate-phenol-chloroform extraction. Anal Biochem. 162: 156 –159. 18. Higuchi R, Krummel B, Saiki R. 1988 A general method of in vitro preparation and specific mutagenesis of DNA fragments: study of protein and DNA interactions. Nucleic Acids Res. 16:7351–7367. 19. Haendler B, Hofer-Warbinek R, Hofer E. 1987 Complementary DNA for human T-cell cyclophilin. EMBO J. 6:947–950. 20. Carlsson L, Rowland A, Clark R, Gesundheit N, Wong W. 1991 Ligandmediated immunofunctional assay (LIFA) for quantitation of growth hormone binding protein in human blood. J Clin Endocrinol Metab. 73:1216 –1223. 21. Hammarqvist F, Wernerman J, Ali MR, Vinnars E. 1990 Effects of an amino acid solution enriched with either branched chain amino acids or ornithinealpha-ketoglutarate on the postoperative intracellular amino acid concentration of skeletal muscle. Br J Surg. 77:214 –218. 22. Rodbard D, Munson J, De Jaen A. 1978 Radioimmunoassay and related procedures in medicine. Vienna: International Atomic Energy Agency; 1:469. 23. Vanden Heuvel J, Tyson F, Bell D. 1993 Construction of recombinant RNA templates for use as internal standards on quantitative RT-PCR. BioTechniques. 14:395–398. 24. Gilliland G, Perrin S, Blanchard K, Bunn H. 1990 Analysis of cytokine mRNA and DNA: detection and quantitation by competitive polymerase chain reaction. Proc Natl Acad Sci USA. 87:2725–2729. 25. Dieffenbach C, Dveksler G. 1993 Setting up a PCR laboratory. PCR Methods Appl. 3:s2–s7. 26. Carlsson B, Billig H, Rymo L, Isaksson OG. 1990 Expression of the growth hormone-binding protein messenger RNA in the liver and extrahepatic tissues in the rat: co-expression with the growth hormone receptor. Mol Cell Endocrinol. 73:R1–R6. 27. Goddard AD, Covello R, Luoh SM, et al. 1995 Mutations of the growth hormone receptor in children with idiopathic short stature. N Engl J Med. 333:1093–1098. 28. Esse´n P, McNurlan MA, Wernerman J, Vinnars E, Garlick PJ. 1992 Uncomplicated surgery, but not general anesthesia, decreases muscle protein synthesis. Am J Physiol. 262:E253–E260. 29. Hammarqvist F, Wernerman J, Ali MR, von der Decken A, Vinnars E. 1987

30. 31.

32. 33. 34.

35. 36. 37. 38. 39. 40. 41. 42.

JCE & M • 1997 Vol 82 • No 2

The effects of branched amino acids upon postoperative muscle protein synthesis and nitrogen balance. Clin Nutr. 7:171–175. Harrison SM, Barnard R, Ho KY, Rajkovic I, Waters MJ. 1995 Control of growth hormone (GH) binding protein release from human hepatoma cells expressing full-length GH receptor. Endocrinology. 136:651– 659. Hammarqvist F, Stro¨mberg C, von der Decken A, Vinnars E, Wernerman J. 1992 Biosynthetic human growth hormone preserves both muscle protein synthesis, the decrease in muscle free glutamine and improves whole body nitrogen economy postoperatively. Ann Surg. 216:184 –191. Manson JM, Wilmore DW. 1986 Positive nitrogen balance with human growth hormone and hypocaloric intravenous feeding. Surgery. 100:188 –197. Ward HC, Halliday D, Sim AJW. 1987 Protein and energy metabolism with biosynthetic human growth hormone after gastrointestinal surgery. Ann Surg. 206:56 – 61. Hammarqvist F, Wernerman J, Ali MR, von der Decken A, Vinnars E. 1989 Addition of glutamine to total parenteral nutrition after elective abdominal surgery spares free glutamine in muscle, counteracts the fall in muscle protein synthesis and improves nitrogen balance. Ann Surg. 209:455– 461. Aulick LH, Wilmore DW. 1979 Increased peripheral amino acid release following burn injury. Surgery. 85:560 –565. Clowes GHA, Randell HT, Cha C-J. 1980 Amino acid and energy metabolism in septic and traumatized patients. J Parenter Enteral Nutr. 4:195–203. Wilomore DW, Smith RJ, O’Dwyer ST, Jacob DO, Ziegler TR, Wang X-D. 1988 The gut: a central organ after surgical stress. Surgery. 104: 917–923. Yang ZJ, Koseki M, Meguid MM, Gleason JR, Debonis D. 1994 Synergistic effect of rhTNF-alpha and rhIL-1 alpha in inducing anorexia in rats. Am J Physiol. 267:R1056 –R1064. Tracey KJ, Wei H, Manogue KR. 1988 Cachectin/tumor necrosis factor induces cachexia, anemia and inflammation. J Exp Med. 167:1211–1227. Hopkins SJ, Rothwell NJ. 1995 Cytokines and the nervous system I: expression and recognition. Trends Neurosci. 18:83– 88. Hotamisligil GS, Shargill NS, Spiegelman BM. 1993 Adipose expression of tumor necrosis factor-a: direct role in obesity-linked insulin resistance. Science. 259:87–91. Thissen JP, Verniers J. 1996 The inhibition by interleukin-1b of the GHinduced IGF-I gene expression is associated with decreased GH receptor gene expression in rat primary cultured hepatocytes [Abstract OR 12–5]. Proc of the 10th Int Congr of Endocrinol. 1996; 66.