- Email: [email protected]
Quantitation of androgen receptor messenger RNA from genital skin fibroblasts by reverse transcription — competitive polymerase chain reaction
PII:
J. Steroid Biochem. Molec. Biol. Vol. 66, No. 1-2, pp. 35±43, 1998 # 1998 Elsevier Science Ltd. All rights reserved Printed in Great Britain S0960-0760(98)00006-5 0960-0760/98 $19.00 + 0.00
Quantitation of Androgen Receptor Messenger RNA from Genital Skin Fibroblasts by Reverse Transcription Ð Competitive Polymerase Chain Reaction Philippe NirdeÂ,1 Virginie Georget,1 BeÂatrice TeÂrouanne,1 ReneÂ-Benoit Galifer,3 Charles Belon4 and Charles Sultan1,2* INSERM U439, Pathologie MoleÂculaire des ReÂcepteurs NucleÂaires, 70, route de Navacelles, F 34090 Montpellier, France; 2Unite B.E.D.R., Hopital Lapeyronie, F 34295 Montpellier, France; 3Chirurgie PeÂdiatrique, Hopital Arnaud de Villeneuve, 34000 Montpellier, France and 4Laboratoire de Biochimie, Faculte de Pharmacie, F 34080 Montpellier, France 1
To gain further information concerning the regulation by androgen of AR mRNA expression in cultured genital skin ®broblasts (GSF), we ®rst developed a quantitative reverse transcription-competitive polymerase chain reaction (RT-PCR). This method used an ethidium bromide stain analysis of the PCR products for the accurate quantitation of low levels of human androgen receptor (hAR) mRNA in GSF. To control for variations due to sample preparation, and to minimize the disparity of the reverse transcriptase ef®ciency between samples after the RT procedure, we produced an initial PCR for the glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene, and then adjusted the amount of cDNA to that of this housekeeping gene. Competitive PCR for hAR was then immediately performed on normalized cDNA with a competitor DNA that exhibited a 13 bp deletion as compared to the 163 bp for the target fragment, and the PCR products were easily separated by 3.5% agarose gel electrophoresis. This quantitation procedure involved no additional steps, such as enzymatic cleavage of the PCR products, nor the use of radioactivity. In GSF from individuals, we found that the normal amount of AR mRNA was 5.6 attomoles/m mg RNA, (21.0, s.e.m.) with an intra- and an inter-assay of 8.4 and 14.7%, respectively. We observed a biphasic pattern of AR mRNA expression in normal human GSF in the presence of physiological concentration of androgen. Quantitative RTPCR of AR mRNA may be useful for studying AR mRNA expression in experimental or clinical conditions. # 1998 Elsevier Science Ltd. All rights reserved. J. Steroid Biochem. Molec. Biol., Vol. 66, No. 1-2, pp. 35±43, 1998
INTRODUCTION
electrophoresis [15], Northern blotting [3± 8, 10, 16, 17], or RNase protection assay [14, 18]. These studies have shown that AR mRNA is a lowabundance target for biochemical analysis and its quantitation remains dif®cult to achieve by these conventional methods. New insights have been gained into the quantitation of scarce and low-abundant molecules with the emergence of enzymatic ampli®cation of genes by polymerase chain reaction (PCR) [19] after reverse transcribing the cellular RNA. Due to the exponential nature of the ampli®cation process [20], minute variations in the ef®ciency of the enzymatic reactions result in signi®cant changes in the product yields. The main dif®culty lies
The regulation of androgen receptor (AR) in target tissues has been investigated, and con¯icting data regarding regulation of hAR or hAR mRNA by androgen have been reported [1±14]. The advantage of studying AR mRNA lies in the possibility of establishing the kinetics of its expression for various periods of time. To evaluate AR expression, several analytical methods have been used, such as capillary *Correspondence to Pr. Ch. Sultan. Tel. 467 338 696; Fax: 467 338 327; e-mail: [email protected]. Received 18 Jul. 1997; accepted 27 Nov. 1997. 35
P. Nirde et al.
36
in the control of the ef®ciency of both the reverse transcription and the polymerase chain reaction [21]. During PCR, the most accurate way to compensate these variations is by the addition of an internal standard, also called a competitor, to the PCR mixture together with the target gene present in the extracts [22, 23]. Based on such control templates, several RT-PCR protocols have been reported for the study of AR mRNA in biological extracts for either relative quantitation [18, 24±27] or precise quantitation [13, 28, 29]. However, the previously proposed competitive PCR protocols involve at least three enzymatic reactions in sequence, in which the last one is an enzymatic digestion of the PCR products. This carries the disadvantage of introducing an additional enzyme reaction to differentiate the internal standard from the target products for gel analysis. In addition, the normalization of samples is another key parameter of the RT-PCR process that needs to be resolved [17]. Indeed, to verify that similar amounts of RNA are loaded in assays, it is common practice to determine the quantity of applied ribosomal RNA in samples, but a recent report has demonstrated by nuclear run-on assays that androgen stimulation may speci®cally increase ribosomal RNA [30]. Such an occurrence both prevents the use of a ribosomal RNA subpopulation as a reliable control against which the level of expression of the AR gene under study can be compared, and raises the question of a reliable control throughout the RTPCR process. Given these dif®culties, we propose in this study an RT- competitive PCR for the determination of AR mRNA that involves assessment of the housekeeping GAPDH gene by an initial PCR. The expression of this gene was demonstrated to be not in¯uenced by androgen stimulation [10, 31]. Equivalent cDNA solution was then loaded for the competitive hAR PCR. This sensitive method was used to circumvent the variations in RT ef®ciency from sample to sample. It was ®rst applied to study the regulation of hAR mRNA expression in normal human GSF from 30 min onwards and then during the ®rst 36 h following incubation with 10ÿ9 M dihydrotestosterone (DHT).
MATERIALS AND METHODS
Materials Dulbecco's modi®ed Earle's minimum medium, fetal calf serum (FCS), penicillin, streptomycin, and Murine Moloney leukemia virus reverse transcriptase were obtained from Gibco BRL, Life Technologies (Gercy-Pontoise, France); Taq polymerase and random primers were purchased from Promega (Lyon, France); Metaphore-agarose was purchased from FMC Bioproducts (Bioprobe system, Montreuil,
France); and RNAble was obtained from Eurobio (Strasbourg, France). DHT was purchased from Sigma (St. Quentin Fallavier, France). Standard 13S/ 23S ribosomal E. coli MRE600 RNA was obtained from Boehringer (Mannheim, Germany). Ethanol and isopropanol were purchased from Carlo Erba (Roma, Italy), and chloroform from Prolabo (Lyon, France). Cell culture Following routine neonatal circumcision, human preputial skins were obtained from the pediatric surgery department at Lapeyronie Hospital, Montpellier. GSFs were established as previously described [32] and cultured in DMEM supplemented with 5% fetal calf serum, penicillin (100 mg/ml), and streptomycin (100 mg/ml). Cells were grown at 378C in a humidi®ed atmosphere of 95% air and 5% CO2 in either 100 or 50 mm diameter dishes (Falcon, Becton Dickinson, Marseille, France). Hormonal incubation From the 100 mm dishes, cells grown near 80% con¯uency were trypsinized, diluted to 1/3 in 50 mm dishes with DMEM, and supplemented with 5% FCS and antibiotics. Two days before hormonal stimulation, cells were cultured in DMEM without FCS. During hormonal incubation, cells were cultivated with DMEM supplemented with 5% charcoal-treated FCS, antibiotics and 10ÿ9 M of DHT previously diluted to 1/1000 in ethanol. Several DHT concentrations ranging from 10ÿ7 to 10ÿ10 M were checked for mRNA stimulation at 12 h, and mRNA stimulation was maximum from 10ÿ9 M. Controls were run in the same conditions except that the hormonal solution was substituted by ethanol (1/1000; v/v). At the selected times (0, 0.5, 1, 3, 6, 12, 24 and 36 h), cells were washed twice at 48C in cold 0.1 M PBS, pH 7.4, prior to being used for RNA preparation. Total RNA preparations Total RNA isolation was carried out according to the manufacturer's instructions using an RNAble reagent kit based on guanidium isothiocyanate±phenol± chloroform extraction [33]. Cells were lysed with 1 ml of RNAble. The integrity of the RNA was visualized in agarose gel and the ribosomal RNA was quanti®ed by digital image processing using known amounts of E. coli ribosomal standards, representing 80% of total RNA determined by UV260 absorbance [34]. Product quantitation The ¯uorescence intensity of the standards was quanti®ed by digital image processing of agarose gel bands on a Argus 100 biological imaging workstation (Hamamatsu Photonics, Hamamatsu, Japan) equipped with a high performance CCD camera (Cohu, Japan). After background subtraction, the net
Quantitation of androgen receptor mRNA
¯uorescence was graphed as a function of the standard amount. The standard curve was then ®tted by linear regression of the standard points, and the amount of nucleic acid from the samples were determined by their net ¯uorescence. The same methodology was used for ribosomal RNA and PCR products analysis Reverse transcription Two micrograms of total cellular RNA were reverse-transcripted at 378C for 60 min in a 50 ml solution containing 400 units of Mu-MLV transcriptase reverse, 50 ng of random primers, 10 mM DTT, and 500 mM of each dNTP and 1 RT buffer. Reverse transcriptase was inhibited by heating at 998C for 5 min, and samples were either used for GAPDH ampli®cation, or stored at ÿ208C until further analysis. GAPDH ampli®cation The same volume of cDNA (3 ml) was subjected to 24 cycles of PCR, each cycle consisting of 1 min at 948C, 1 min at 558C, and 1 min at 728C (MJ Research PCT 200, Watertown, MA). Total reaction volume was 25 ml, containing 480 nM of the sense (5' TCGCCAGCCGAGCCACATCG) and the antisense primers (5' GAACATGTAAACCATGTAG), 200 mM dNTP, 1.5 mM MgCl2, 1 U Taq polymer-
37
ase, and 1 Taq polymerase buffer. The PCR product was loaded onto 1.8% agarose gel containing 0.5 mg/ml ethidium bromide and GAPDH fragments quanti®ed from their relative ¯uorescence intensity by computer imaging. hAR competitor The competitive hAR DNA corresponded to a previously described natural deletion of 13 bp from position 2119 to 2132 in exon 4 of hAR from a patient presenting a complete androgen insensitivity syndrome [35]. By PCR, it generates a 150 bp product instead of 163 bp for the wild-type hAR fragment (Fig. 1). Brie¯y, pCMV5-hAR vector (a generous gift from T. R. Brown, Johns Hopkins University, Baltimore, MD) was cleaved with KpnI/ BamH1 and this fragment was subcloned into the corresponding sites of pUC19 (pUC19-hAR). The PCR of the patient's exon 4 was digested by Tth111I/StuI and this fragment ligated to pUC19-hAR from which the Tth111-I/StuI fragment had been previously removed. Finally, the deleted exon 4 KpnI/ BamHI was inserted into the corresponding sites of pCMV5-hAR-del4. Ampli®cations were performed in DH5a cells, and exon 4 was veri®ed by sequencing. The 150 bp competitor was ampli®ed by PCR (same conditions as for competitive PCR), puri®ed, and calibrated.
Fig. 1. Schematic diagram of hAR cDNA organization and the region spanning exons 4 and 5 of the gene. Starting position of exon 4 and termination position of exon 5 are numbered in brackets. The nucleotide positions of ¯anking primers are indicated. Sense primer within exon 4 and antisense primer within exon 5 gave a 163 bp PCR product, whereas mutant gave a 150 bp product with the same primers.
P. Nirde et al.
38
Competitive PCR The AR ampli®cation was performed using primer pairs within the hormonal domain of the hAR, corresponding to exons 4 and 5 at positions 2032±2051 and 2176±2195, respectively (Fig. 1), which produces a 163 bp fragment with the wild-type gene. Equivalent amounts of cDNA, with regards to GAPDH production, were subjected to three hAR ampli®cations with known amounts of competitor hAR-DNA of 0.03, 0.4 and 5 attomoles, respectively. Final PCR volume was 25 ml containing dNTP, MgCl2, and Taq polymerase, as for GAPDH, and 320 nM sense primer (5' GAAGCCATTGAGCCAGGTGT), 320 nM antisense primer (5' TCGTCCACGTGTAAGTTGCG). PCR ampli®cation was carried out in a thermocycler for 30 cycles. Each cycle consisted of denaturation at 928C for 40 s, followed by annealing at 628C for 40 s and elongation at 748C for 90 s. After ampli®cation, the PCR products were separated on a 3.5% Metaphore-agarose slab gel stained with ethidium bromide and run in a 0.5
TBE at 48C. For each individual reaction, intensities were determined for the competitor (Ic) and sample (Is) after background subtraction. The logarithm Is/Ic was plotted as a function of the logarithm of the amount for the internal competitor, and the curve ®tting was computed by linear regression. The data generated from these curves were used to interpolate the initial amount of AR mRNA in each sample, calculated from the ratio of the sample to the standard at equivalence [36, 37]. Final amounts of AR mRNA were expressed as attomoles per mg RNA (2s.e.m). RESULTS
Control of the reverse transcription ef®ciency The same volume of total RNA extracted from human GSF at the pre-selected times during a 36 h DHT stimulation period was analyzed by gel electrophoresis (Fig. 2(A)). As expected, an increase in the RNA contents was observed during stimulation of the GSF by DHT. An up to threefold increase in the
Fig. 2. (A) Ethidium bromide staining of denaturing 1% agarose gel of total RNA extracted from GSF under hormonal stimulation. An equal volume from each sample (3 ml) was loaded onto the gel run at 100 mA (constant intensity). The positions of 18S and 28S ribosomal RNA are indicated on the right, and those of 13S and 23S ribosomal E. coli used as standards are indicated on the left. Standard points (S1 = 0.5 mg, S2 = 1.0 mg, S3 = 1.5 mg, S4 = 2 mg) and samples from lanes 1 to 8 (0, 0.5, 1, 3, 6, 12, 24 and 36 h, respectively) of a typical extraction are shown. (B) At the desired time of stimulation, RNA was reverse transcribed and the relative amounts of the housekeeping GAPDH gene were obtained by a foremost PCR. Samples (lanes 1±8 for 0 to 36 h incubation) were loaded onto 2% Metaphore-agarose gel run in a 0.5 TBE, and each band was analyzed by digital imaging.
Quantitation of androgen receptor mRNA
ribosomal RNA was recorded between the unstimulated and the 36 h stimulated GSF. From the standard points, a precise quanti®cation of the amount of ribosomal RNA was determined, and equal amounts of RNA were reverse transcripted into cDNA and then subjected to GAPDH ampli®cation. PCR products were separated by gel electrophoresis (Fig. 2(B)). The relative GAPDH amounts for the samples (Fig. 2(B), lanes 1±8) displayed a eightfold overall variation. The equation that describes the PCR process is N = N0(1 + e)n, where N and N0 are the number of molecules after and before ampli®cation, respectively, e the ef®ciency of the enzymatic
39
reaction, and n the number of cycles. This difference in the GAPDH content is consistent with a difference in the ef®ciency of ampli®cation between samples of 9% with regard to this equation. The cDNA from each sample was then adjusted for their GAPDH contents, and another PCR was performed to verify that the known independent-androgen-stimulated GAPDH gene was accurately determined in each sample (data not shown). Competitive PCR Figure 3(A) exhibits typical results obtained for the quantitation of AR mRNA after a competitive PCR.
Fig. 3. Typical quantitative RT-PCR pro®le obtained by using the 13 bp mutated AR exon as competitor template. (A) A constant amount of the same cDNA was co-ampli®ed with three increasing amounts of the AR competitor (a = 5 attomoles; b = 0.4 attomoles; c = 0.03 attomoles). After 30 cycles of ampli®cation, PCR products were loaded on a 3.5% Metaphore-agarose gel run in a 0.5 TBE at 48C (50 V constant voltage). (B) The ¯uorescence intensities of the bands corresponding to the target and the competitor, minus lane background, were quanti®ed by computer imaging of the gel. The logarithm of the ratio of ampli®ed sample to competitor products (log S/C) was plotted as the function of the logarithm of the known amount of competitor added (log C) to the reaction.
40
P. Nirde et al.
PCR mixtures containing a constant amount of GAPDH cDNA for the same sample were added to dilutions of the hAR competitor (Fig. 3(A)) and ampli®ed with the same primer pairs. The 13 bp difference in size that separated the mutant from the normal gene was properly separated by gel electrophoresis. For the same competitor concentration, the log of ratios between the ¯uorescence intensities was graphed as a function of the competitor concentration (Fig. 3(B)). The equivalence point was computed for similar signals for the competitor and the sample, i.e., log(Is/Ic) = 0. Intra-assays were performed on eight different extracts from the same origin at the same passage run at the same time, and the variations in the mRNA determination were 8.4%. In the inter-assays, the samples were run at two different periods (four times for GAPDH) and the variations in the mRNA determination were found to be 14.7%. Quantitation of hAR MRNA in human GSF In four newborn foreskin cultured ®broblasts, the mean level of AR mRNA was 5.6 21.0 attomoles/mg RNA (range 4.2±7.1) (Fig. 4). Physiological concentration of DHT (10ÿ9 M) induced a biphasic pattern of AR mRNA concentration: a rapid decrease in the mean mRNA levels to 1.7 2 1.6 attomoles/mg RNA (range 0.4±4.7) occurred 1 h after hormonal stimulation. Between 0.5 and 1 h, the decrease ranged
Fig. 4. Time course of AR mRNA levels in normal GSF stimulated with 10ÿ9 M DHT during a 36 h period. RT-competitive PCR was performed on equivalent cDNA amounts. Levels of AR mRNA were resolved as stated in Section 2 and expressed as attomoles of mRNA/m mg RNA. (w) sampling 1, () sampling 2, (R) sampling 3, (r) sampling 4.
from 60 to 70%. An increase in the mRNA was then observed, reaching a peak of 11.5 24.3 attomoles/mg RNA (range 6.8±18.6) 12 h after stimulation, which represented a 205% increase. The AR mRNA level then slowly decreased to reach the basal AR mRNA expression after 24 h of stimulation.
DISCUSSION
mRNA evaluation is generally considered to be an effective way to study the regulation of AR. Competitive PCR has been proposed for AR mRNA quantitation by constructing mutated [29] or deleted AR sequence as competitive template [24, 28, 38]. We previously described a short deletion in the ligandbinding domain of the human androgen receptor which is responsible for a complete androgen insensitivity syndrome [35]. In the present study, we selected the natural portion of DNA that encompasses this deletion to produce an internal competitor for the competitive PCR. It has the advantage of being directly produced by PCR without additional construction, and the ampli®ed product is well separated by agarose gel electrophoresis from the wild-type PCR fragment without additional enzymatic reaction. This suggests that a natural short deletion or a natural short insertion found within the AR gene from patients could be easily used as a competitor fragment in a competitive PCR whose primers encompass the modi®ed portion of that gene. Our competitor sequence generated a PCR fragment as short as 150 bp. It exhibits only a 13 bp difference in size from the wild template as compared with the 74 bp difference reported by Edelstein [29], the 165 bp difference reported by Prins [28], and the 38 bp difference reported by Marlucelli [24] for PCR products of 294, 485 and 172 bp, respectively. Because the ampli®cation ef®ciency is higher for both shorter and closely related sequences [39], this natural human AR mutant ful®lls all the conditions to be used as a competitor. Furthermore, the successive use of enzyme conversion and ampli®cation procedure introduces inevitable sample to sample variations. For precise mRNA quantitation in minute amounts of biological material these variations should be minimized to the greatest extent possible, particularly for reactions in sequence. To reach this goal in this study, we controlled the ef®ciency of the reverse transcriptase promptly after the reaction. Thus, the use of the halfway GAPDH PCR signal allowed a better determination of the cDNA concentration in each sample and minimized the impact of the difference in the RT ef®ciency between samples. The ®nal competitive PCR was then performed on homogeneous sample concentrations based on the GAPDH product used as reference, which is not in¯uenced by androgen
Quantitation of androgen receptor mRNA
stimulation [10, 31]. In contrast, such validation for the b-actin gene needs to be demonstrated [17]. A housekeeping gene determination in sample extracts has often been reported. It should be emphasized that in these studies, the housekeeping gene was then only used as a mathematical factor to compute the ®nal AR mRNA evaluation after the RT-PCR process, and not to both truly load equivalent amounts of cDNA and compensate formally the PCR ef®ciency between samples. It clearly appears that the housekeeping gene that is used must represent a key halfway parameter in the accurate determination of hAR mRNA by RT-competitive PCR. We therefore must ascertain that the expression of the housekeeping gene, used as a reference gene to normalize the samples, is not in¯uenced by experimental conditions, i.e., the androgen stimulation. We applied the method described herein to quantify AR mRNA expression in normal GSF during an androgen stimulation. The mRNA level in control GSF (5.6 attomoles/mg RNA) was consistent with the data reported by Marcelli [40] (10 attomoles/mg RNA) and was of the same magnitude as that reported by Choong [13] (3.8 attomoles/mg RNA) in control GSF. In addition to the study of the AR mRNA levels in normal GSF at steady-state, we analyzed the ®rst minutes, and up to 36 h after DHT stimulation, in greater detail because scant data are available on quantitative AR mRNA levels during an androgen stimulation, and because GSF cells are often used as control cells. DHT stimulation of normal GSF cells resulted in a rapid, but transient, decrease in AR mRNA expression. This down-regulation occurred, although possibly more than 83% of the initial DHT concentration remained present in the medium [41] and diffusion of the hormone within the GSF had taken place. It is unlikely that this decline in AR mRNA expression was the result of a marked decrease in AR mRNA stability since androgen primarily controls RNA stability [42, 43]. This down-regulation preceded the up-regulation of AR mRNA expression, which displayed a maximum mRNA level 12 h following stimulation. Up-regulation correlates with the androgen-dependent augmentation of androgenreceptor activity in normal GSF reported by Kaufman [41]. AR mRNA up-regulation by DHT has also been demonstrated in mouse and rat prostate by in situ hybridization [44]. The basal mRNA level was observed in GSF 36 h following stimulation with physiological amounts of DHT in three out of four GSF cell cultures. Other authors have observed that GSF stimulated with the synthetic androgen mibolerone exhibited a slight decrease in the AR mRNA level 49 h following hormonal administration [6], or that the mRNA level remained similar to that of the control [10] for prolonged hormonal treatment at 96 h. These authors have concluded that androgens
41
have no action on GSF AR mRNA. With regards to the kinetics studies, we can complete these conclusions, which remained valid for prolonged androgen stimulation. Indeed, in our study, autologous expression regulation of AR mRNA occurred for shorter periods (0.5±12 h) of hormonal induction. We demonstrated the presence of a biphasic regulation of AR mRNA expression in GSF as soon as 30 min following DHT stimulation, and a return to the basal mRNA level at the end of the experiments. Such biphasic regulation has also been observed in Sertoli cells for AR mRNA expression after stimulation by follicle-stimulating hormone or by dibutyryl cyclic AMP [7]. Similar results have been reported for estradiol and its estrogen receptor [45]. The mechanisms of such biphasic expression for AR mRNA in androgen-stimulated GSF remain to be explained. In conclusion, an accurate assessment of hAR mRNA by quantitative RT-PCR is made possible by the normalization of cDNA contents performed from GAPDH evaluation, preliminary to the competitive PCR for the AR gene. We demonstrated that the regulation of hAR mRNA expression by DHT in the GSF displayed a biphasic pro®le which was timedependent. The applicability of this quantitative RT-PCR of AR mRNA could be extended to cases of androgen hypersensitivity, such as hirsutism or androgen insensitivity [46], in which decreases in AR mRNA levels have recently been reported [47].
REFERENCES 1. Blondeau J. P., Baulieu E. E. and Robel P., Androgen-dependent regulation of androgen nuclear receptor in the rat ventral prostate. Endocrinology 110 (1982) 1926±1932. 2. Kaufman M., Pinsky L., Hollander R. and Bailey J. D., Regulation of the androgen receptor by androgen in normal and androgen-resistant genital skin ®broblasts. Journal of Steroid Biochemistry 18 (1983) 383±390. 3. Quarmby V. E., Yarbrough W. G., Lubahn D. B., French F. S. and Wilson E. M., Autologous down-regulation of androgen receptor messenger ribonucleic acid. Molecular Endocrinology 4 (1990) 22±28. 4. Shan L., Rodriguez M. C. and JaÈnne O. A., Regulation of androgen receptor protein and mRNA concentrations by androgens in rat ventral prostate and seminal vesicles and in human hepatoma cells. Molecular Endocrinology 4 (1990) 1636± 1646. 5. Mestayer C., Gabou L., Portois M. C., Boucekkine C., Chang C. and Mowszowicz I., Expression du geÁne du reÂcepteur des androgeÁnes chez des sujets normaux et des patients preÂsentant une insenbilite compleÁte aux androgeÁnes. Annales d'Endocrinologie (Paris) 52 (1991) 431±434. 6. Krongrad A., Wilson C. M., Wilson J. D., Allman D. R. and McPhaul M. J., Androgen increases androgen receptor protein while decreasing receptor mRNA in LNCap cells. Molecular and Cellular Endocrinology 76 (1991) 79±88. 7. Blok L. J., Themmen A. P., Peters A. H., Trapman J., Baarends W. M., Hoogerbrugge J. W. and Grootegoed J. A., Transcriptional regulation of androgen receptor gene expression in sertoli cells and other types. Molecular and Cellular Endocrinology 88 (1992) 153±164.
42
P. Nirde et al.
8. Lin M. C., Rajfer J., Swerdloff R. S. and Gonzalez-Cadavid N. F., Testosterone down-regulates the levels of androgen receptor mRNA in smooth muscle cells from the rat corpora cavernosa via aromatization to estrogens. Journal of Steroid Biochemistry and Molecular Biology 45 (1993) 333±343. 9. Shan L. X., Zhu L. J., Bardin W. and Hardy M. P., Quantitative analysis of androgen receptor messenger ribonucleic acid in developing leydig cells and sertoli cells by in situ hybridization. Endocrinology 136 (1995) 3856±3862. 10. Wolf D. A., Herzinger T., Hermeking H., Blaschke D. and HoÈrz W., Transcriptional and posttranscriptional regulation of human androgen receptor expression by androgen. Molecular Endocrinology 7 (1993) 924±936. 11. Wiren K. M., Zhang X., Chang C., Keenan E. and Orwoll E. S., Transcriptional up-regulation of the human androgen receptor by androgen in bone cells. Endocrinology 138 (1997) 2291±2300. 12. Bentvelsen F. M., Brinkman A. O., Van der schoot P., Van der linden J. E. T. M., Van der kwast T. H., Boersma W. J. A., SchroÈder F. H. and Nijman J. M., Developmental pattern and regulation by androgens of androgen receptor expression in the urogenital tract of the rat. Molecular and Cellular Endocrinology 113 (1995) 245±253. 13. Choong C. S., Sturm M. J., Strophair J. A., McCulloch R. K., Tilley W. D., Leedman P. J. and Hurley D. M., Partial androgen insensitivity caused by an androgen receptor mutation at amino acid 907 (Gly±Arg) that results in decreased ligand binding af®nity and reduced androgen receptor messenger ribonucleic acid levels. Journal of Clinical Endocrinology and Metabolism 81 (1996) 236±243. 14. Le Dai J., Maiorino C. A., Gkonos P. J. and Burnstein K. L., Androgenic up-regulation of androgen receptor cDNA expression in androgen-independent prostate cancer cells. Steroids 61 (1996) 531±539. 15. Nathakarnkitkool S., Oefner P. J., Bartch G., Chin M. A. and Bonn G. K., High-resolution capillary electrophoretic analysis of DNA in free solution. Electrophoresis 13 (1992) 18±31. 16. Blok L. J., Hoogerbrugge J. W., Themmen A. P., Baarends W. M., Post M. and Grootegoed J. A., Transcriptional regulation of androgen receptor gene expression in Sertoli cells and other types. Endocrinology 131 (1992) 1343±1349. 17. Mora G. R., Prins G. S. and Mahesh V. B., Autoregulation of androgen receptor protein and messenger RNA in rat ventral prostate is protein synthesis dependent. Journal of Steroid Biochemistry and Molecular Biology 58 (1996) 539±549. 18. Takane K. K., Husmann D. A., McPhaul M. J. and Wilson J. D., Androgen receptor levels in the rat penis are controlled differently in distinctive cell types. Endocrinology 128 (1991) 2234±2238. 19. Saiki R. K., Scharf S., Faloona F., Mullis K. B., Horn G. T., Erlich H. A. and Arnheim N., Enzymatic ampli®cation of beta-globin genomic sequences and restriction site analysis for diagnosis of sickle cell anemia. Science 230 (1985) 1350±1354. 20. Chelly J., Kaplan J. C., Marie P., Gautron S. and Kahn A., Transcription of the dystrophin gene in human muscle and non-muscle tissues. Nature 333 (1988) 856±860. 21. Barragan-Gonzales E., Lopez-Guerrero J. A., Bolufer-Gilabert P., Sanz-Alonso M., De la Rubia-Comos J. and SempereTalens A., The type of reverse transcriptase affects the sensitivity of some reverse transcription PCR methods. Clinica Chimica Acta 260 (1997) 73±83. 22. Becker-Andre M. and Hahlbrock K., Absolute mRNA quanti®cation using the polymerase chain reaction (PCR). A novel approach by a PCR aided transcript titration assay (PATTY). Nucleic Acids Research 17 (1989) 9437±9446. 23. Wang A. M., Doyle M. V. and Marrk D. F., Quantitation of mRNA by the polymerase reaction . Proceedings of the National Academy of Sciences U.S.A. 86 (1989) 9717±9721. 24. Malucelli A., Sauerwein H., Pfaf¯ M. W. and Meyer H. D., Quanti®cation of androgen receptor mRNA in tissues by competitive co-ampli®cation of a template in reverse transcriptionpolymerase chain reaction. Journal of Steroid Biochemistry and Molecular Biology 58 (1996) 563±568. 25. Rossi R., Fransceschetti P., Maestri I., Magri E., Cavazzini L., Degli-Uberti E. C. and Del Senno L., Evidence for androgen receptor gene expression in human thyroid cells and tumours. Journal of Endocrinology 148 (1996) 77±85.
26. Grant E. S., Batchelor K. W. and Habib F. K., Androgen independence of primary epithelial cultures of the prostate is associated with a down-regulation of androgen receptor gene expression. Prostate 29 (1996) 339±349. 27. Nastiuk K. L. and Clayton D., Seasonal and tissue-speci®c regulation of canary androgen receptor messenger ribonucleic acid. Endocrinology 134 (1994) 640±649. 28. Prins G. S. and Woodham C., Autologous regulation of androgen receptor messenger ribonucleic acid in the separate lobes of the rat prostate gland. Biology of Reproduction 53 (1995) 609±619. 29. Edelstein R. A., Carr M. C., Caesar R., Young M., Atala A. and Freeman M. R., Detection of human androgen receptor mRNA expression abnormalities by competitive PCR. DNA and Cell Biology 13 (1994) 265±273. 30. Kabler R. N., Srinivasan A., Taylor L. J., Mowad J., Rothblum L. I. and Cavanaugh A. H., Androgen regulation of ribosomal RNA synthesis in LNCaP cells and rat prostate. Journal of Steroid Biochemistry and Molecular Biology 59 (1996) 431±439. 31. Burnstein K. L., Maiorino C. A., Dai J. L. and Cameron D. J., Androgen and glucocorticoid regulation of androgen receptor cDNA expression. Molecular and Cellular Endocrinology 115 (1995) 177±186. 32. Sultan C., Picard J. Y., Josso N. and Migeon C. J., Incomplete androgen insensitivity syndrome. Clinical Endocrinology 28 (1983) 565±568. 33. Chomczynski P. and Sacchi N., Single step method of RNA isolation by acid guanidium thiocyanate±phenol±chloroform extraction. Analytical Biochemistry 162 (1987) 156±159. 34. Ausubel F. M. et al. In: Current protocols in molecular biology, Vol. 2. Greene Publishing Associates and Wiley, New York, 1993. 35. Lobaccaro J. M., Lumbroso S., Poujol N., Georget V., Brinkman A. O., Malpuech G. and Sultan C., Complete androgen insensitivity syndrome due to a new frameshift deletion in exon 4 of the androgen receptor gene: Functional analysis of the mutant receptor. Molecular and Cellular Endocrinology 111 (1995) 21±28. 36. Gilliland G., Perrin S., Blanchard K. and Boun H. F., Detection and quanti®cation by competitive polymerase chain reaction. Proceedings of the National Academy of Sciences U.S.A. 87 (1990) 2725±2729. 37. Siebert P. D. and Larrick J. W., Competitive PCR. Nature 359 (1992) 557±558. 38. Nastiuk K. L. and Clayton D., Seasonal and tissue-speci®c regulation of canary androgen receptor messenger ribonucleic acid. Endocrinology 134 (1994) 640±649. 39. Nedelman J., Heagerty P. and Lawrence C., Quantitative PCR with internal controls. Computer and Applied Bioscience 8 (1992) 65±70. 40. Marcelli M., Tilley W. D., Zoppi S., Grif®n J. E., Wilson J. D. and McPhaul M. J., Androgen resistance associated with a mutation of the androgen receptor at amino acid 772 (Arg±Cys) results from a combination of decreased messenger ribonucleic acid levels and impairment of receptor function. Journal of Clinical Endocrinology and Metabolism 73 (1991) 318±325. 41. Kaufman M., Pinsky L., Hollander R. and Bailey J. D., Regulation of the androgen receptor by androgen in normal and androgen-resistant genital skin ®broblasts. Journal of Steroid Biochemistry 18 (1983) 383±390. 42. Page M. J. and Parker M. G., Effect of androgen on the transcription of rat prostatic binding protein genes. Molecular and Cellular Endocrinology 27 (1982) 343±355. 43. Berger F. G., Loose D., Meisner H. and Watson G., Androgen induction of messenger RNA concentrations in mouse kidney is post-transcriptional. Biochemistry 25 (1986) 1170±1175. 44. Takeda H., Nakamoto T., Kokontis J., Chodak G. W. and Chang C., Autoregulation of androgen receptor expression in rodent prostate: immunohistochemical and in situ hybridization analysis. Biochemical and Biophysical Research Communications 177 (1991) 488±496. 45. Saceda M., Lippman M. E., Chambon P., Lindsey R. L., Ponglikitmongkol M., Puente M. and Martin M. B., Regulation of the estrogen receptor in MCF-7 cells by estradiol. Molecular Endocrinology 2 (1988) 1157±1162.
Quantitation of androgen receptor mRNA 46. Sultan C., Lumbroso S., Poujol N., Belon C., Boudon C. and Lobaccaro J. M., Mutations of the androgen receptor gene in androgen insensitivity syndrome. Journal of Steroid Biochemistry and Molecular Biology 46 (1993) 519±530.
43
47. Choong C. S., Sturm M. J., Strophair J. A., McCulloch R. K. and Hurley D. M., Reduced expression and normal nucleotide sequence of androgen receptor gene coding and promoter regions in a family with partial androgen insensitivity syndrome. Clinical Endocrinology 46 (1997) 281±288.
J. Steroid Biochem. Molec. Biol. Vol. 66, No. 1-2, pp. 35±43, 1998 # 1998 Elsevier Science Ltd. All rights reserved Printed in Great Britain S0960-0760(98)00006-5 0960-0760/98 $19.00 + 0.00
Quantitation of Androgen Receptor Messenger RNA from Genital Skin Fibroblasts by Reverse Transcription Ð Competitive Polymerase Chain Reaction Philippe NirdeÂ,1 Virginie Georget,1 BeÂatrice TeÂrouanne,1 ReneÂ-Benoit Galifer,3 Charles Belon4 and Charles Sultan1,2* INSERM U439, Pathologie MoleÂculaire des ReÂcepteurs NucleÂaires, 70, route de Navacelles, F 34090 Montpellier, France; 2Unite B.E.D.R., Hopital Lapeyronie, F 34295 Montpellier, France; 3Chirurgie PeÂdiatrique, Hopital Arnaud de Villeneuve, 34000 Montpellier, France and 4Laboratoire de Biochimie, Faculte de Pharmacie, F 34080 Montpellier, France 1
To gain further information concerning the regulation by androgen of AR mRNA expression in cultured genital skin ®broblasts (GSF), we ®rst developed a quantitative reverse transcription-competitive polymerase chain reaction (RT-PCR). This method used an ethidium bromide stain analysis of the PCR products for the accurate quantitation of low levels of human androgen receptor (hAR) mRNA in GSF. To control for variations due to sample preparation, and to minimize the disparity of the reverse transcriptase ef®ciency between samples after the RT procedure, we produced an initial PCR for the glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene, and then adjusted the amount of cDNA to that of this housekeeping gene. Competitive PCR for hAR was then immediately performed on normalized cDNA with a competitor DNA that exhibited a 13 bp deletion as compared to the 163 bp for the target fragment, and the PCR products were easily separated by 3.5% agarose gel electrophoresis. This quantitation procedure involved no additional steps, such as enzymatic cleavage of the PCR products, nor the use of radioactivity. In GSF from individuals, we found that the normal amount of AR mRNA was 5.6 attomoles/m mg RNA, (21.0, s.e.m.) with an intra- and an inter-assay of 8.4 and 14.7%, respectively. We observed a biphasic pattern of AR mRNA expression in normal human GSF in the presence of physiological concentration of androgen. Quantitative RTPCR of AR mRNA may be useful for studying AR mRNA expression in experimental or clinical conditions. # 1998 Elsevier Science Ltd. All rights reserved. J. Steroid Biochem. Molec. Biol., Vol. 66, No. 1-2, pp. 35±43, 1998
INTRODUCTION
electrophoresis [15], Northern blotting [3± 8, 10, 16, 17], or RNase protection assay [14, 18]. These studies have shown that AR mRNA is a lowabundance target for biochemical analysis and its quantitation remains dif®cult to achieve by these conventional methods. New insights have been gained into the quantitation of scarce and low-abundant molecules with the emergence of enzymatic ampli®cation of genes by polymerase chain reaction (PCR) [19] after reverse transcribing the cellular RNA. Due to the exponential nature of the ampli®cation process [20], minute variations in the ef®ciency of the enzymatic reactions result in signi®cant changes in the product yields. The main dif®culty lies
The regulation of androgen receptor (AR) in target tissues has been investigated, and con¯icting data regarding regulation of hAR or hAR mRNA by androgen have been reported [1±14]. The advantage of studying AR mRNA lies in the possibility of establishing the kinetics of its expression for various periods of time. To evaluate AR expression, several analytical methods have been used, such as capillary *Correspondence to Pr. Ch. Sultan. Tel. 467 338 696; Fax: 467 338 327; e-mail: [email protected]. Received 18 Jul. 1997; accepted 27 Nov. 1997. 35
P. Nirde et al.
36
in the control of the ef®ciency of both the reverse transcription and the polymerase chain reaction [21]. During PCR, the most accurate way to compensate these variations is by the addition of an internal standard, also called a competitor, to the PCR mixture together with the target gene present in the extracts [22, 23]. Based on such control templates, several RT-PCR protocols have been reported for the study of AR mRNA in biological extracts for either relative quantitation [18, 24±27] or precise quantitation [13, 28, 29]. However, the previously proposed competitive PCR protocols involve at least three enzymatic reactions in sequence, in which the last one is an enzymatic digestion of the PCR products. This carries the disadvantage of introducing an additional enzyme reaction to differentiate the internal standard from the target products for gel analysis. In addition, the normalization of samples is another key parameter of the RT-PCR process that needs to be resolved [17]. Indeed, to verify that similar amounts of RNA are loaded in assays, it is common practice to determine the quantity of applied ribosomal RNA in samples, but a recent report has demonstrated by nuclear run-on assays that androgen stimulation may speci®cally increase ribosomal RNA [30]. Such an occurrence both prevents the use of a ribosomal RNA subpopulation as a reliable control against which the level of expression of the AR gene under study can be compared, and raises the question of a reliable control throughout the RTPCR process. Given these dif®culties, we propose in this study an RT- competitive PCR for the determination of AR mRNA that involves assessment of the housekeeping GAPDH gene by an initial PCR. The expression of this gene was demonstrated to be not in¯uenced by androgen stimulation [10, 31]. Equivalent cDNA solution was then loaded for the competitive hAR PCR. This sensitive method was used to circumvent the variations in RT ef®ciency from sample to sample. It was ®rst applied to study the regulation of hAR mRNA expression in normal human GSF from 30 min onwards and then during the ®rst 36 h following incubation with 10ÿ9 M dihydrotestosterone (DHT).
MATERIALS AND METHODS
Materials Dulbecco's modi®ed Earle's minimum medium, fetal calf serum (FCS), penicillin, streptomycin, and Murine Moloney leukemia virus reverse transcriptase were obtained from Gibco BRL, Life Technologies (Gercy-Pontoise, France); Taq polymerase and random primers were purchased from Promega (Lyon, France); Metaphore-agarose was purchased from FMC Bioproducts (Bioprobe system, Montreuil,
France); and RNAble was obtained from Eurobio (Strasbourg, France). DHT was purchased from Sigma (St. Quentin Fallavier, France). Standard 13S/ 23S ribosomal E. coli MRE600 RNA was obtained from Boehringer (Mannheim, Germany). Ethanol and isopropanol were purchased from Carlo Erba (Roma, Italy), and chloroform from Prolabo (Lyon, France). Cell culture Following routine neonatal circumcision, human preputial skins were obtained from the pediatric surgery department at Lapeyronie Hospital, Montpellier. GSFs were established as previously described [32] and cultured in DMEM supplemented with 5% fetal calf serum, penicillin (100 mg/ml), and streptomycin (100 mg/ml). Cells were grown at 378C in a humidi®ed atmosphere of 95% air and 5% CO2 in either 100 or 50 mm diameter dishes (Falcon, Becton Dickinson, Marseille, France). Hormonal incubation From the 100 mm dishes, cells grown near 80% con¯uency were trypsinized, diluted to 1/3 in 50 mm dishes with DMEM, and supplemented with 5% FCS and antibiotics. Two days before hormonal stimulation, cells were cultured in DMEM without FCS. During hormonal incubation, cells were cultivated with DMEM supplemented with 5% charcoal-treated FCS, antibiotics and 10ÿ9 M of DHT previously diluted to 1/1000 in ethanol. Several DHT concentrations ranging from 10ÿ7 to 10ÿ10 M were checked for mRNA stimulation at 12 h, and mRNA stimulation was maximum from 10ÿ9 M. Controls were run in the same conditions except that the hormonal solution was substituted by ethanol (1/1000; v/v). At the selected times (0, 0.5, 1, 3, 6, 12, 24 and 36 h), cells were washed twice at 48C in cold 0.1 M PBS, pH 7.4, prior to being used for RNA preparation. Total RNA preparations Total RNA isolation was carried out according to the manufacturer's instructions using an RNAble reagent kit based on guanidium isothiocyanate±phenol± chloroform extraction [33]. Cells were lysed with 1 ml of RNAble. The integrity of the RNA was visualized in agarose gel and the ribosomal RNA was quanti®ed by digital image processing using known amounts of E. coli ribosomal standards, representing 80% of total RNA determined by UV260 absorbance [34]. Product quantitation The ¯uorescence intensity of the standards was quanti®ed by digital image processing of agarose gel bands on a Argus 100 biological imaging workstation (Hamamatsu Photonics, Hamamatsu, Japan) equipped with a high performance CCD camera (Cohu, Japan). After background subtraction, the net
Quantitation of androgen receptor mRNA
¯uorescence was graphed as a function of the standard amount. The standard curve was then ®tted by linear regression of the standard points, and the amount of nucleic acid from the samples were determined by their net ¯uorescence. The same methodology was used for ribosomal RNA and PCR products analysis Reverse transcription Two micrograms of total cellular RNA were reverse-transcripted at 378C for 60 min in a 50 ml solution containing 400 units of Mu-MLV transcriptase reverse, 50 ng of random primers, 10 mM DTT, and 500 mM of each dNTP and 1 RT buffer. Reverse transcriptase was inhibited by heating at 998C for 5 min, and samples were either used for GAPDH ampli®cation, or stored at ÿ208C until further analysis. GAPDH ampli®cation The same volume of cDNA (3 ml) was subjected to 24 cycles of PCR, each cycle consisting of 1 min at 948C, 1 min at 558C, and 1 min at 728C (MJ Research PCT 200, Watertown, MA). Total reaction volume was 25 ml, containing 480 nM of the sense (5' TCGCCAGCCGAGCCACATCG) and the antisense primers (5' GAACATGTAAACCATGTAG), 200 mM dNTP, 1.5 mM MgCl2, 1 U Taq polymer-
37
ase, and 1 Taq polymerase buffer. The PCR product was loaded onto 1.8% agarose gel containing 0.5 mg/ml ethidium bromide and GAPDH fragments quanti®ed from their relative ¯uorescence intensity by computer imaging. hAR competitor The competitive hAR DNA corresponded to a previously described natural deletion of 13 bp from position 2119 to 2132 in exon 4 of hAR from a patient presenting a complete androgen insensitivity syndrome [35]. By PCR, it generates a 150 bp product instead of 163 bp for the wild-type hAR fragment (Fig. 1). Brie¯y, pCMV5-hAR vector (a generous gift from T. R. Brown, Johns Hopkins University, Baltimore, MD) was cleaved with KpnI/ BamH1 and this fragment was subcloned into the corresponding sites of pUC19 (pUC19-hAR). The PCR of the patient's exon 4 was digested by Tth111I/StuI and this fragment ligated to pUC19-hAR from which the Tth111-I/StuI fragment had been previously removed. Finally, the deleted exon 4 KpnI/ BamHI was inserted into the corresponding sites of pCMV5-hAR-del4. Ampli®cations were performed in DH5a cells, and exon 4 was veri®ed by sequencing. The 150 bp competitor was ampli®ed by PCR (same conditions as for competitive PCR), puri®ed, and calibrated.
Fig. 1. Schematic diagram of hAR cDNA organization and the region spanning exons 4 and 5 of the gene. Starting position of exon 4 and termination position of exon 5 are numbered in brackets. The nucleotide positions of ¯anking primers are indicated. Sense primer within exon 4 and antisense primer within exon 5 gave a 163 bp PCR product, whereas mutant gave a 150 bp product with the same primers.
P. Nirde et al.
38
Competitive PCR The AR ampli®cation was performed using primer pairs within the hormonal domain of the hAR, corresponding to exons 4 and 5 at positions 2032±2051 and 2176±2195, respectively (Fig. 1), which produces a 163 bp fragment with the wild-type gene. Equivalent amounts of cDNA, with regards to GAPDH production, were subjected to three hAR ampli®cations with known amounts of competitor hAR-DNA of 0.03, 0.4 and 5 attomoles, respectively. Final PCR volume was 25 ml containing dNTP, MgCl2, and Taq polymerase, as for GAPDH, and 320 nM sense primer (5' GAAGCCATTGAGCCAGGTGT), 320 nM antisense primer (5' TCGTCCACGTGTAAGTTGCG). PCR ampli®cation was carried out in a thermocycler for 30 cycles. Each cycle consisted of denaturation at 928C for 40 s, followed by annealing at 628C for 40 s and elongation at 748C for 90 s. After ampli®cation, the PCR products were separated on a 3.5% Metaphore-agarose slab gel stained with ethidium bromide and run in a 0.5
TBE at 48C. For each individual reaction, intensities were determined for the competitor (Ic) and sample (Is) after background subtraction. The logarithm Is/Ic was plotted as a function of the logarithm of the amount for the internal competitor, and the curve ®tting was computed by linear regression. The data generated from these curves were used to interpolate the initial amount of AR mRNA in each sample, calculated from the ratio of the sample to the standard at equivalence [36, 37]. Final amounts of AR mRNA were expressed as attomoles per mg RNA (2s.e.m). RESULTS
Control of the reverse transcription ef®ciency The same volume of total RNA extracted from human GSF at the pre-selected times during a 36 h DHT stimulation period was analyzed by gel electrophoresis (Fig. 2(A)). As expected, an increase in the RNA contents was observed during stimulation of the GSF by DHT. An up to threefold increase in the
Fig. 2. (A) Ethidium bromide staining of denaturing 1% agarose gel of total RNA extracted from GSF under hormonal stimulation. An equal volume from each sample (3 ml) was loaded onto the gel run at 100 mA (constant intensity). The positions of 18S and 28S ribosomal RNA are indicated on the right, and those of 13S and 23S ribosomal E. coli used as standards are indicated on the left. Standard points (S1 = 0.5 mg, S2 = 1.0 mg, S3 = 1.5 mg, S4 = 2 mg) and samples from lanes 1 to 8 (0, 0.5, 1, 3, 6, 12, 24 and 36 h, respectively) of a typical extraction are shown. (B) At the desired time of stimulation, RNA was reverse transcribed and the relative amounts of the housekeeping GAPDH gene were obtained by a foremost PCR. Samples (lanes 1±8 for 0 to 36 h incubation) were loaded onto 2% Metaphore-agarose gel run in a 0.5 TBE, and each band was analyzed by digital imaging.
Quantitation of androgen receptor mRNA
ribosomal RNA was recorded between the unstimulated and the 36 h stimulated GSF. From the standard points, a precise quanti®cation of the amount of ribosomal RNA was determined, and equal amounts of RNA were reverse transcripted into cDNA and then subjected to GAPDH ampli®cation. PCR products were separated by gel electrophoresis (Fig. 2(B)). The relative GAPDH amounts for the samples (Fig. 2(B), lanes 1±8) displayed a eightfold overall variation. The equation that describes the PCR process is N = N0(1 + e)n, where N and N0 are the number of molecules after and before ampli®cation, respectively, e the ef®ciency of the enzymatic
39
reaction, and n the number of cycles. This difference in the GAPDH content is consistent with a difference in the ef®ciency of ampli®cation between samples of 9% with regard to this equation. The cDNA from each sample was then adjusted for their GAPDH contents, and another PCR was performed to verify that the known independent-androgen-stimulated GAPDH gene was accurately determined in each sample (data not shown). Competitive PCR Figure 3(A) exhibits typical results obtained for the quantitation of AR mRNA after a competitive PCR.
Fig. 3. Typical quantitative RT-PCR pro®le obtained by using the 13 bp mutated AR exon as competitor template. (A) A constant amount of the same cDNA was co-ampli®ed with three increasing amounts of the AR competitor (a = 5 attomoles; b = 0.4 attomoles; c = 0.03 attomoles). After 30 cycles of ampli®cation, PCR products were loaded on a 3.5% Metaphore-agarose gel run in a 0.5 TBE at 48C (50 V constant voltage). (B) The ¯uorescence intensities of the bands corresponding to the target and the competitor, minus lane background, were quanti®ed by computer imaging of the gel. The logarithm of the ratio of ampli®ed sample to competitor products (log S/C) was plotted as the function of the logarithm of the known amount of competitor added (log C) to the reaction.
40
P. Nirde et al.
PCR mixtures containing a constant amount of GAPDH cDNA for the same sample were added to dilutions of the hAR competitor (Fig. 3(A)) and ampli®ed with the same primer pairs. The 13 bp difference in size that separated the mutant from the normal gene was properly separated by gel electrophoresis. For the same competitor concentration, the log of ratios between the ¯uorescence intensities was graphed as a function of the competitor concentration (Fig. 3(B)). The equivalence point was computed for similar signals for the competitor and the sample, i.e., log(Is/Ic) = 0. Intra-assays were performed on eight different extracts from the same origin at the same passage run at the same time, and the variations in the mRNA determination were 8.4%. In the inter-assays, the samples were run at two different periods (four times for GAPDH) and the variations in the mRNA determination were found to be 14.7%. Quantitation of hAR MRNA in human GSF In four newborn foreskin cultured ®broblasts, the mean level of AR mRNA was 5.6 21.0 attomoles/mg RNA (range 4.2±7.1) (Fig. 4). Physiological concentration of DHT (10ÿ9 M) induced a biphasic pattern of AR mRNA concentration: a rapid decrease in the mean mRNA levels to 1.7 2 1.6 attomoles/mg RNA (range 0.4±4.7) occurred 1 h after hormonal stimulation. Between 0.5 and 1 h, the decrease ranged
Fig. 4. Time course of AR mRNA levels in normal GSF stimulated with 10ÿ9 M DHT during a 36 h period. RT-competitive PCR was performed on equivalent cDNA amounts. Levels of AR mRNA were resolved as stated in Section 2 and expressed as attomoles of mRNA/m mg RNA. (w) sampling 1, () sampling 2, (R) sampling 3, (r) sampling 4.
from 60 to 70%. An increase in the mRNA was then observed, reaching a peak of 11.5 24.3 attomoles/mg RNA (range 6.8±18.6) 12 h after stimulation, which represented a 205% increase. The AR mRNA level then slowly decreased to reach the basal AR mRNA expression after 24 h of stimulation.
DISCUSSION
mRNA evaluation is generally considered to be an effective way to study the regulation of AR. Competitive PCR has been proposed for AR mRNA quantitation by constructing mutated [29] or deleted AR sequence as competitive template [24, 28, 38]. We previously described a short deletion in the ligandbinding domain of the human androgen receptor which is responsible for a complete androgen insensitivity syndrome [35]. In the present study, we selected the natural portion of DNA that encompasses this deletion to produce an internal competitor for the competitive PCR. It has the advantage of being directly produced by PCR without additional construction, and the ampli®ed product is well separated by agarose gel electrophoresis from the wild-type PCR fragment without additional enzymatic reaction. This suggests that a natural short deletion or a natural short insertion found within the AR gene from patients could be easily used as a competitor fragment in a competitive PCR whose primers encompass the modi®ed portion of that gene. Our competitor sequence generated a PCR fragment as short as 150 bp. It exhibits only a 13 bp difference in size from the wild template as compared with the 74 bp difference reported by Edelstein [29], the 165 bp difference reported by Prins [28], and the 38 bp difference reported by Marlucelli [24] for PCR products of 294, 485 and 172 bp, respectively. Because the ampli®cation ef®ciency is higher for both shorter and closely related sequences [39], this natural human AR mutant ful®lls all the conditions to be used as a competitor. Furthermore, the successive use of enzyme conversion and ampli®cation procedure introduces inevitable sample to sample variations. For precise mRNA quantitation in minute amounts of biological material these variations should be minimized to the greatest extent possible, particularly for reactions in sequence. To reach this goal in this study, we controlled the ef®ciency of the reverse transcriptase promptly after the reaction. Thus, the use of the halfway GAPDH PCR signal allowed a better determination of the cDNA concentration in each sample and minimized the impact of the difference in the RT ef®ciency between samples. The ®nal competitive PCR was then performed on homogeneous sample concentrations based on the GAPDH product used as reference, which is not in¯uenced by androgen
Quantitation of androgen receptor mRNA
stimulation [10, 31]. In contrast, such validation for the b-actin gene needs to be demonstrated [17]. A housekeeping gene determination in sample extracts has often been reported. It should be emphasized that in these studies, the housekeeping gene was then only used as a mathematical factor to compute the ®nal AR mRNA evaluation after the RT-PCR process, and not to both truly load equivalent amounts of cDNA and compensate formally the PCR ef®ciency between samples. It clearly appears that the housekeeping gene that is used must represent a key halfway parameter in the accurate determination of hAR mRNA by RT-competitive PCR. We therefore must ascertain that the expression of the housekeeping gene, used as a reference gene to normalize the samples, is not in¯uenced by experimental conditions, i.e., the androgen stimulation. We applied the method described herein to quantify AR mRNA expression in normal GSF during an androgen stimulation. The mRNA level in control GSF (5.6 attomoles/mg RNA) was consistent with the data reported by Marcelli [40] (10 attomoles/mg RNA) and was of the same magnitude as that reported by Choong [13] (3.8 attomoles/mg RNA) in control GSF. In addition to the study of the AR mRNA levels in normal GSF at steady-state, we analyzed the ®rst minutes, and up to 36 h after DHT stimulation, in greater detail because scant data are available on quantitative AR mRNA levels during an androgen stimulation, and because GSF cells are often used as control cells. DHT stimulation of normal GSF cells resulted in a rapid, but transient, decrease in AR mRNA expression. This down-regulation occurred, although possibly more than 83% of the initial DHT concentration remained present in the medium [41] and diffusion of the hormone within the GSF had taken place. It is unlikely that this decline in AR mRNA expression was the result of a marked decrease in AR mRNA stability since androgen primarily controls RNA stability [42, 43]. This down-regulation preceded the up-regulation of AR mRNA expression, which displayed a maximum mRNA level 12 h following stimulation. Up-regulation correlates with the androgen-dependent augmentation of androgenreceptor activity in normal GSF reported by Kaufman [41]. AR mRNA up-regulation by DHT has also been demonstrated in mouse and rat prostate by in situ hybridization [44]. The basal mRNA level was observed in GSF 36 h following stimulation with physiological amounts of DHT in three out of four GSF cell cultures. Other authors have observed that GSF stimulated with the synthetic androgen mibolerone exhibited a slight decrease in the AR mRNA level 49 h following hormonal administration [6], or that the mRNA level remained similar to that of the control [10] for prolonged hormonal treatment at 96 h. These authors have concluded that androgens
41
have no action on GSF AR mRNA. With regards to the kinetics studies, we can complete these conclusions, which remained valid for prolonged androgen stimulation. Indeed, in our study, autologous expression regulation of AR mRNA occurred for shorter periods (0.5±12 h) of hormonal induction. We demonstrated the presence of a biphasic regulation of AR mRNA expression in GSF as soon as 30 min following DHT stimulation, and a return to the basal mRNA level at the end of the experiments. Such biphasic regulation has also been observed in Sertoli cells for AR mRNA expression after stimulation by follicle-stimulating hormone or by dibutyryl cyclic AMP [7]. Similar results have been reported for estradiol and its estrogen receptor [45]. The mechanisms of such biphasic expression for AR mRNA in androgen-stimulated GSF remain to be explained. In conclusion, an accurate assessment of hAR mRNA by quantitative RT-PCR is made possible by the normalization of cDNA contents performed from GAPDH evaluation, preliminary to the competitive PCR for the AR gene. We demonstrated that the regulation of hAR mRNA expression by DHT in the GSF displayed a biphasic pro®le which was timedependent. The applicability of this quantitative RT-PCR of AR mRNA could be extended to cases of androgen hypersensitivity, such as hirsutism or androgen insensitivity [46], in which decreases in AR mRNA levels have recently been reported [47].
REFERENCES 1. Blondeau J. P., Baulieu E. E. and Robel P., Androgen-dependent regulation of androgen nuclear receptor in the rat ventral prostate. Endocrinology 110 (1982) 1926±1932. 2. Kaufman M., Pinsky L., Hollander R. and Bailey J. D., Regulation of the androgen receptor by androgen in normal and androgen-resistant genital skin ®broblasts. Journal of Steroid Biochemistry 18 (1983) 383±390. 3. Quarmby V. E., Yarbrough W. G., Lubahn D. B., French F. S. and Wilson E. M., Autologous down-regulation of androgen receptor messenger ribonucleic acid. Molecular Endocrinology 4 (1990) 22±28. 4. Shan L., Rodriguez M. C. and JaÈnne O. A., Regulation of androgen receptor protein and mRNA concentrations by androgens in rat ventral prostate and seminal vesicles and in human hepatoma cells. Molecular Endocrinology 4 (1990) 1636± 1646. 5. Mestayer C., Gabou L., Portois M. C., Boucekkine C., Chang C. and Mowszowicz I., Expression du geÁne du reÂcepteur des androgeÁnes chez des sujets normaux et des patients preÂsentant une insenbilite compleÁte aux androgeÁnes. Annales d'Endocrinologie (Paris) 52 (1991) 431±434. 6. Krongrad A., Wilson C. M., Wilson J. D., Allman D. R. and McPhaul M. J., Androgen increases androgen receptor protein while decreasing receptor mRNA in LNCap cells. Molecular and Cellular Endocrinology 76 (1991) 79±88. 7. Blok L. J., Themmen A. P., Peters A. H., Trapman J., Baarends W. M., Hoogerbrugge J. W. and Grootegoed J. A., Transcriptional regulation of androgen receptor gene expression in sertoli cells and other types. Molecular and Cellular Endocrinology 88 (1992) 153±164.
42
P. Nirde et al.
8. Lin M. C., Rajfer J., Swerdloff R. S. and Gonzalez-Cadavid N. F., Testosterone down-regulates the levels of androgen receptor mRNA in smooth muscle cells from the rat corpora cavernosa via aromatization to estrogens. Journal of Steroid Biochemistry and Molecular Biology 45 (1993) 333±343. 9. Shan L. X., Zhu L. J., Bardin W. and Hardy M. P., Quantitative analysis of androgen receptor messenger ribonucleic acid in developing leydig cells and sertoli cells by in situ hybridization. Endocrinology 136 (1995) 3856±3862. 10. Wolf D. A., Herzinger T., Hermeking H., Blaschke D. and HoÈrz W., Transcriptional and posttranscriptional regulation of human androgen receptor expression by androgen. Molecular Endocrinology 7 (1993) 924±936. 11. Wiren K. M., Zhang X., Chang C., Keenan E. and Orwoll E. S., Transcriptional up-regulation of the human androgen receptor by androgen in bone cells. Endocrinology 138 (1997) 2291±2300. 12. Bentvelsen F. M., Brinkman A. O., Van der schoot P., Van der linden J. E. T. M., Van der kwast T. H., Boersma W. J. A., SchroÈder F. H. and Nijman J. M., Developmental pattern and regulation by androgens of androgen receptor expression in the urogenital tract of the rat. Molecular and Cellular Endocrinology 113 (1995) 245±253. 13. Choong C. S., Sturm M. J., Strophair J. A., McCulloch R. K., Tilley W. D., Leedman P. J. and Hurley D. M., Partial androgen insensitivity caused by an androgen receptor mutation at amino acid 907 (Gly±Arg) that results in decreased ligand binding af®nity and reduced androgen receptor messenger ribonucleic acid levels. Journal of Clinical Endocrinology and Metabolism 81 (1996) 236±243. 14. Le Dai J., Maiorino C. A., Gkonos P. J. and Burnstein K. L., Androgenic up-regulation of androgen receptor cDNA expression in androgen-independent prostate cancer cells. Steroids 61 (1996) 531±539. 15. Nathakarnkitkool S., Oefner P. J., Bartch G., Chin M. A. and Bonn G. K., High-resolution capillary electrophoretic analysis of DNA in free solution. Electrophoresis 13 (1992) 18±31. 16. Blok L. J., Hoogerbrugge J. W., Themmen A. P., Baarends W. M., Post M. and Grootegoed J. A., Transcriptional regulation of androgen receptor gene expression in Sertoli cells and other types. Endocrinology 131 (1992) 1343±1349. 17. Mora G. R., Prins G. S. and Mahesh V. B., Autoregulation of androgen receptor protein and messenger RNA in rat ventral prostate is protein synthesis dependent. Journal of Steroid Biochemistry and Molecular Biology 58 (1996) 539±549. 18. Takane K. K., Husmann D. A., McPhaul M. J. and Wilson J. D., Androgen receptor levels in the rat penis are controlled differently in distinctive cell types. Endocrinology 128 (1991) 2234±2238. 19. Saiki R. K., Scharf S., Faloona F., Mullis K. B., Horn G. T., Erlich H. A. and Arnheim N., Enzymatic ampli®cation of beta-globin genomic sequences and restriction site analysis for diagnosis of sickle cell anemia. Science 230 (1985) 1350±1354. 20. Chelly J., Kaplan J. C., Marie P., Gautron S. and Kahn A., Transcription of the dystrophin gene in human muscle and non-muscle tissues. Nature 333 (1988) 856±860. 21. Barragan-Gonzales E., Lopez-Guerrero J. A., Bolufer-Gilabert P., Sanz-Alonso M., De la Rubia-Comos J. and SempereTalens A., The type of reverse transcriptase affects the sensitivity of some reverse transcription PCR methods. Clinica Chimica Acta 260 (1997) 73±83. 22. Becker-Andre M. and Hahlbrock K., Absolute mRNA quanti®cation using the polymerase chain reaction (PCR). A novel approach by a PCR aided transcript titration assay (PATTY). Nucleic Acids Research 17 (1989) 9437±9446. 23. Wang A. M., Doyle M. V. and Marrk D. F., Quantitation of mRNA by the polymerase reaction . Proceedings of the National Academy of Sciences U.S.A. 86 (1989) 9717±9721. 24. Malucelli A., Sauerwein H., Pfaf¯ M. W. and Meyer H. D., Quanti®cation of androgen receptor mRNA in tissues by competitive co-ampli®cation of a template in reverse transcriptionpolymerase chain reaction. Journal of Steroid Biochemistry and Molecular Biology 58 (1996) 563±568. 25. Rossi R., Fransceschetti P., Maestri I., Magri E., Cavazzini L., Degli-Uberti E. C. and Del Senno L., Evidence for androgen receptor gene expression in human thyroid cells and tumours. Journal of Endocrinology 148 (1996) 77±85.
26. Grant E. S., Batchelor K. W. and Habib F. K., Androgen independence of primary epithelial cultures of the prostate is associated with a down-regulation of androgen receptor gene expression. Prostate 29 (1996) 339±349. 27. Nastiuk K. L. and Clayton D., Seasonal and tissue-speci®c regulation of canary androgen receptor messenger ribonucleic acid. Endocrinology 134 (1994) 640±649. 28. Prins G. S. and Woodham C., Autologous regulation of androgen receptor messenger ribonucleic acid in the separate lobes of the rat prostate gland. Biology of Reproduction 53 (1995) 609±619. 29. Edelstein R. A., Carr M. C., Caesar R., Young M., Atala A. and Freeman M. R., Detection of human androgen receptor mRNA expression abnormalities by competitive PCR. DNA and Cell Biology 13 (1994) 265±273. 30. Kabler R. N., Srinivasan A., Taylor L. J., Mowad J., Rothblum L. I. and Cavanaugh A. H., Androgen regulation of ribosomal RNA synthesis in LNCaP cells and rat prostate. Journal of Steroid Biochemistry and Molecular Biology 59 (1996) 431±439. 31. Burnstein K. L., Maiorino C. A., Dai J. L. and Cameron D. J., Androgen and glucocorticoid regulation of androgen receptor cDNA expression. Molecular and Cellular Endocrinology 115 (1995) 177±186. 32. Sultan C., Picard J. Y., Josso N. and Migeon C. J., Incomplete androgen insensitivity syndrome. Clinical Endocrinology 28 (1983) 565±568. 33. Chomczynski P. and Sacchi N., Single step method of RNA isolation by acid guanidium thiocyanate±phenol±chloroform extraction. Analytical Biochemistry 162 (1987) 156±159. 34. Ausubel F. M. et al. In: Current protocols in molecular biology, Vol. 2. Greene Publishing Associates and Wiley, New York, 1993. 35. Lobaccaro J. M., Lumbroso S., Poujol N., Georget V., Brinkman A. O., Malpuech G. and Sultan C., Complete androgen insensitivity syndrome due to a new frameshift deletion in exon 4 of the androgen receptor gene: Functional analysis of the mutant receptor. Molecular and Cellular Endocrinology 111 (1995) 21±28. 36. Gilliland G., Perrin S., Blanchard K. and Boun H. F., Detection and quanti®cation by competitive polymerase chain reaction. Proceedings of the National Academy of Sciences U.S.A. 87 (1990) 2725±2729. 37. Siebert P. D. and Larrick J. W., Competitive PCR. Nature 359 (1992) 557±558. 38. Nastiuk K. L. and Clayton D., Seasonal and tissue-speci®c regulation of canary androgen receptor messenger ribonucleic acid. Endocrinology 134 (1994) 640±649. 39. Nedelman J., Heagerty P. and Lawrence C., Quantitative PCR with internal controls. Computer and Applied Bioscience 8 (1992) 65±70. 40. Marcelli M., Tilley W. D., Zoppi S., Grif®n J. E., Wilson J. D. and McPhaul M. J., Androgen resistance associated with a mutation of the androgen receptor at amino acid 772 (Arg±Cys) results from a combination of decreased messenger ribonucleic acid levels and impairment of receptor function. Journal of Clinical Endocrinology and Metabolism 73 (1991) 318±325. 41. Kaufman M., Pinsky L., Hollander R. and Bailey J. D., Regulation of the androgen receptor by androgen in normal and androgen-resistant genital skin ®broblasts. Journal of Steroid Biochemistry 18 (1983) 383±390. 42. Page M. J. and Parker M. G., Effect of androgen on the transcription of rat prostatic binding protein genes. Molecular and Cellular Endocrinology 27 (1982) 343±355. 43. Berger F. G., Loose D., Meisner H. and Watson G., Androgen induction of messenger RNA concentrations in mouse kidney is post-transcriptional. Biochemistry 25 (1986) 1170±1175. 44. Takeda H., Nakamoto T., Kokontis J., Chodak G. W. and Chang C., Autoregulation of androgen receptor expression in rodent prostate: immunohistochemical and in situ hybridization analysis. Biochemical and Biophysical Research Communications 177 (1991) 488±496. 45. Saceda M., Lippman M. E., Chambon P., Lindsey R. L., Ponglikitmongkol M., Puente M. and Martin M. B., Regulation of the estrogen receptor in MCF-7 cells by estradiol. Molecular Endocrinology 2 (1988) 1157±1162.
Quantitation of androgen receptor mRNA 46. Sultan C., Lumbroso S., Poujol N., Belon C., Boudon C. and Lobaccaro J. M., Mutations of the androgen receptor gene in androgen insensitivity syndrome. Journal of Steroid Biochemistry and Molecular Biology 46 (1993) 519±530.
43
47. Choong C. S., Sturm M. J., Strophair J. A., McCulloch R. K. and Hurley D. M., Reduced expression and normal nucleotide sequence of androgen receptor gene coding and promoter regions in a family with partial androgen insensitivity syndrome. Clinical Endocrinology 46 (1997) 281±288.