INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

DOPAMINE RECEPTORS HAVE been subdivided into two major classes both within the central nervous system (D₁ and D₂) and in the periphery (DA₁ and DA₂) (11, 16). This division developed from the pharmacological profiles observed in eliciting a physiological or biochemical response along with the elucidation of the signal transduction mechanisms activated by the receptor subtype. Thus D₁-like receptor activation resulted in an enhanced adenylyl cyclase activity, whereas D₂-like receptors inhibited or were not linked to this enzyme cascade. This pharmacological classification has been superseded by molecular characterization through the use of recombinant DNA cloning techniques that have revealed an extensive heterogeneity of the dopamine receptor system (2, 4, 5, 21, 22, 28-30, 32-34, 38). The demonstration that these newly cloned receptors have the pharmacological and biochemical properties of the native receptor has been achieved through their expression in heterologous cell lines (18, 26, 31).

Several D₁-like receptor subtypes have been cloned [D_1A, D_1B (or human homologue D₅), D_1C, and D_1D], and each is coupled to the stimulation of adenylyl cyclase through an interaction with the G_S guanine nucleotide binding protein. The tissue distribution for the D_1A receptor has been determined by Northern blot analysis and in situ hybridization histochemistry. At first, many of these studies were conducted on the brain, although, in more recent years, such studies have been extended to peripheral tissue sites (5, 9, 10, 12, 19, 20, 29, 36).

We have recently demonstrated that the D_1A receptor, which is present in the central nervous system, can be identified in the rat kidney at those sites previously labeled on the basis of pharmacological binding studies as D₁-like receptor sites (25). The dopamine D_1A receptor subtype was identified using both light microscopic immunohistochemistry and electron microscopic immunocytochemistry. Anti-peptide polyclonal antisera were directed to both extracellular and intracellular domains of the native receptor. The use of such receptor subtype-selective antibodies allowed for the identification of specific dopamine receptor subtype clones that are not distinguished by current pharmacological or ligand binding technology. These results suggested that renal dopamine DA₁ receptor corresponds structurally to the central dopamine D_1A receptor.

Peripheral tissues, such as the kidney, contain dopamine receptor subtype-specific mRNA in substantially lower quantities than that demonstrable in the brain, as evidenced by the failure of Northern blot analysis or standard in situ hybridization to confirm the presence of peripheral dopamine D_1A receptor mRNA in these organs (4, 21, 32, 38). Previously, the expression of the dopamine D_1A receptor gene in microdissected proximal tubules of rat kidneys has been demonstrated by reverse transcription-polymerase chain reaction (RT-PCR) with subsequent confirmation by solution hybridization and ribonuclease protection assay (36). The possibility of a broader renal expression of this dopamine receptor subtype was not explored. Therefore, we designed experiments that make use of a novel self-sustained sequence replication system to amplify in situ, in rat kidney sections, a unique portion of the rat D_1A cDNA, which was subsequently detected by nonradioactive in situ hybridization histochemistry. Other groups have similarly adapted this methodology for the amplification and localization of viral mRNA in cultured cells (14, 37). In addition, nonamplified brain sections known to express D_1A mRNA were probed to confirm the fidelity of the D_1A receptor mRNA signal in the kidney. We also demonstrated site-specific colocalization of D_1A receptor mRNA and protein in the renal proximal and distal tubule and confirmed the expression of the D_1A receptor gene by RT-PCR.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Male Wistar-Kyoto rats were anesthetized intraperitoneally with pentobarbital sodium. The animals were perfused transcardially with a regime of 0.9% saline, followed sequentially by 1% ribonuclease (RNase)-free sucrose in 0.12 M sodium phosphate and 4% paraformaldehyde in 0.12 M phosphate. After perfusion, the kidneys were removed, decapsulated, and postfixed in the 4% paraformaldehyde solution for 1 h at 4°C and thereafter cryoprotected overnight at 4°C in 30% RNase-free sucrose, 0.12 M phosphate. Coronal sections of rat brain including the corpus striatum were similarly treated. Frozen sections (7-12 µm thick) were cut in a cryostat at 26°C, mounted on poly-L-lysine-coated slides, and stored at 70°C until processed for in situ amplification and hybridization.

In situ amplification. The in situ amplification protocol involves the generation of duplex cDNA that contains a promoter for the bacteriophage T7 RNA polymerase, which in turn allows for repetitive RNA transcriptional synthesis (Fig. 1). The promoter sequence used was 5' AATTTAATACGACTCACTATAGGGA 3'. The amplified sequence can have the antisense sequence of the target, the sense sequence, or both, depending on which primer incorporates the T7 promoter binding sequence. The reaction depends on the continuous cycling of reverse transcription and RNA transcription reactions to replicate an RNA target by means of cDNA intermediates. Pairs of oligonucleotides are used to prime cDNA synthesis. The synthesis of a double-stranded cDNA then serves as transcription template for T7 RNA polymerase. Complete cDNA synthesis is dependent on the digestion of the RNA in the intermediate RNA-DNA heteroduplex by RNase H. Transcription-competent cDNAs yield antisense or sense RNA copies of the original target. These transcripts are converted to cDNAs containing double-stranded promoter sequences that can serve as a template for further RNA synthesis. In the present series of experiments, kidney sections were processed for D_1A receptor mRNA amplification, whereas brain sections were processed for standard in situ hybridization alone.

View larger version (22K): [in this window] [in a new window]   Fig. 1.   In situ amplification protocol involves the generation of duplex cDNA that contains a promoter for the bacteriophage T7 RNA polymerase which, in turn, allows for repetitive RNA transcriptional synthesis. Reaction depends on the continuous isothermal cycling of reverse transcription and RNA transcription reactions to replicate an RNA target by means of cDNA intermediates. Initial steps depict the synthesis of a double-stranded cDNA, which serves as transcription template for T7 RNA polymerase. Complete cDNA synthesis is dependent on the digestion of the RNA in the intermediate RNA-DNA heteroduplex by RNase H. RNA is shown by thick solid lines; DNA is shown dashed lines; AMV-RT, avian myeloblastosis virus-reverse transcriptase.

Because of the convention with in situ hybridization histochemistry of detecting mRNA by using antisense oligonucleotides, we amplified a sense product 308 bases in length of the 5'-flanking region of the mRNA encoding the NH₂ terminus of the D_1A receptor (30) (Fig. 2). The forward primer, containing the T7 promoter, was 5' GCAACTGGGGCTGAACAAGA 3', spanning nucleotides 112 to 93. The reverse primer was 5' CTAAAGAGATGACAAAGA 3', spanning nucleotides 196 to 179. The antisense probe directed to the amplified 308-bp sense product was 5' AAAGGAGAAATCCCTCTCCGCTGG 3', spanning nucleotides 40 to 63. 

View larger version (38K): [in this window] [in a new window]   Fig. 2.   Schematic of D1A receptor mRNA and receptor protein indicating location of mRNA sequence (112 to +196) in relation to the translation start site (+1) that was amplified in the self-sustained sequence replication reaction.

Tissue sections had placed on them a mixture of 4× standard sodium citrate (SSC) and 50% formamide, to which was added the antisense (reverse) primer at 1 ng/25 µl and which was allowed to hybridize overnight at room temperature. Unhybridized primer was washed off with two changes of 2× SSC at 15 min each at room temperature, followed by two changes of 0.5× SSC at 30 min each at 40°C in an OmniSlide In Situ system (Hybaid, Middlesex, UK). Sections then had added a solution containing dNTP mix at 5 mM, rNTP mix at 1.5 mM, reaction buffer [40 mM tris(hydroxymethyl)aminomethane hydrochloride (Tris · HCl), pH 8.1, 30 mM MgCl₂, 20 mM KCl, 10 mM dithiothreitol, 4 mM spermidine], forward and reverse primers at 1 ng/25 µl, diethyl pyrocarbonate-treated water, and an enzyme mix of avian myeloblastosis virus-reverse transcriptase (30 units), T7 RNA polymerase (100 units), and RNAse H (3 units), and transcriptional amplification was allowed to proceed for 2 h at 42°C. After amplification, the tissue sections were washed with two changes in 2× SSC for 15 min each at room temperature. Thereafter, the sections entered a standard in situ hybridization histochemistry protocol.

Hybridization histochemistry. In situ hybridization was conducted on kidney and brain sections using the same 24-mer high-performance liquid chromatography purified antisense and sense oligonucleotide probes that were complementary to a D_1A receptor mRNA sequence (nucleotides 40-63) lying within the amplified product and which codes for an NH₂-terminal amino acid residue sequence (Fig. 2). The oligonucleotides were labeled by tailing the 3' end with digoxigenin-11-dUTP, using terminal transferase from a commercially available kit (Genius 5 Oligonucleotide 3'-End Labeling Kit, Boehringer Mannheim). Prehybridization of both amplified and nonamplified tissue sections was achieved by placing an aliquot (300 µl) of an appropriate solution on each section (50% deionized formamide, 4× SSC, 1× Denhardt's bovine serum albumin, 0.5 mg/ml denatured sheared herring sperm DNA, 0.25 mg/ml yeast transfer RNA, 10% dextran sulfate) in a humid chamber. The slides were incubated for 1 h at room temperature. The hybridization solution was prepared by diluting the digoxigenin-labeled oligonucleotide probes in the prehybridization solution. The slides were removed from the incubation chamber and briefly submerged in 2× SSC. Thereafter, the hybridization solution (containing either sense or antisense probes) was applied onto the tissue sections and incubated overnight at 37°C.

After hybridization, the tissue sections were washed in decreasing concentrations of SSC as follows: 2× SSC for 1 h at room temperature, 1× SSC for 1 h at room temperature, 0.5× SSC for 0.5 h at 37°C, and 0.5× SSC for 0.5 h at room temperature. Immunoreactive hybridized digoxigenin-labeled probe was detected with an alkaline phosphatase-nitro blue tetrazolium-5-bromo-4-chloro-3-indolylphosphate p-toluidine salt (NBT-BCIP) technique using the Genius Nonradioactive Nucleic Acid Detection Kit (Boehringer Mannheim). After the posthybridization washes, the sections were immersed first for 5 min in buffer A (0.1 M Tris, pH 7.5, 1 M NaCl, 2 mM MgCl₂) containing 2% normal goat serum and 0.1% Triton X-100, then in the buffer A containing alkaline phosphatase-labeled anti-digoxigenin F(ab) fragment (1:5,000), and incubated overnight at 4°C. The sections then were washed in buffer A (three times for 10 min), in buffer B (0.1 M Tris, pH 9.5, 1 M NaCl, 5 mM MgCl₂) (5 min), and in buffer C (0.1 M Tris, pH 9.5, 0.1 M NaCl, 5 mM MgCl₂) (5 min). Tissue-bound alkaline phosphatase activity was detected by incubating the tissue sections with NBT [75 mg/ml NBT salt in 70% (vol/vol) dimethylformamide] and BCIP (BCIP and toluidinium salt in 100% dimethylformamide) diluted in buffer C in a humid chamber in the absence of light for 2-24 h. The enzymatic reaction was stopped by rinsing the sections in phosphate-buffered saline. The sections were counterstained with neutral red, delipidated with xylene, and coverslipped in DPX mounting media.

For the in situ hybridization procedure, controls included a sense oligonucleotide hybridization control, a no antisense or sense oligonucleotide probe-added control, or a no anti-digoxigenin secondary antibody-added control, each of which served to control for nonspecific hybridization and background. All of these treatments resulted in total loss of the hybridization signal. Controls for the amplification procedure included omission of the reverse transcriptase enzyme and omission of the amplification primers. Again, these resulted in total loss of the hybridization signal.

RT-PCR. Total RNA was isolated from kidney and brain tissue as well as from a murine fibroblast LTK cell line that had been stably transfected with a full-length rat D_1A receptor cDNA, using a modified calcium phosphate method as previously described (25). Nontransfected LTK cells were used as a control source of RNA. RNA extraction was carried out using a standard guanidium-thiocyanate protocol (36). First-strand cDNA synthesis (reaction volume, 30 µl), using 10-100 ng of RNA, was performed in the presence of Moloney murine leukemia virus-reverse transcriptase (MMLV-RT) buffer (New England Biolabs, Beverly, MA), with 0.5 µg oligo(dT)₁₅ primer (Promega, Southampton, UK), 0.20 mM dNTPS (Promega), 20 units RNase (Promega), and 100 units MMLV-RT. The reaction was carried out at 37°C for 90 min. A control in which all the components of the reaction were added except the reverse transcriptase was tested in parallel with each sample.

PCR amplification of the cDNA was performed using primers designed specifically for the rat D_1A receptor mRNA. The sense primer 5' AGATCTCTTGGTGGCTGTC 3', corresponding to nucleotides 263-281, and the antisense primer 5' ATAATGGCTACGGGGATGT 3', corresponding to nucleotides 688-706, resulted in the amplification of a 425-bp product. PCR was carried out (total volume, 50 µl) in the presence of 1 µl of the cDNA reaction, 25 pmol of each primer, 0.25 mM dNTPS, 5 µl of 10× Taq buffer (Promega), and 2 µl of 50 mM MgCl₂. The reaction contents were heated at 94°C for 1 min before the addition of 0.5 units of Taq polymerase (Promega). The 40 reaction cycles were set at 96°C for 15 s, 60°C for 30 s, and 72°C for 3 min. The integrity of the RNA was verified by amplification of the housekeeping gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH) under the same experimental conditions. Analysis of the amplified product was conducted by size fractionation of an aliquot of the RT-PCR sample, in conjunction with molecular weight markers of known size, in a 2% agarose gel and visualization with ethidium bromide staining. All of the RT-PCR products thus analyzed gave a band of the expected size.

Immunohistochemistry. Light microscopic immunohistochemistry was performed on rat kidney sections that had been prepared for in situ amplification of D_1A receptor mRNA as stated above. Antisera were directed toward a peptide sequence (GSEETQPFC) on the third extracellular domain of the predicted rat D_1A receptor amino acid sequence, and immunohistochemistry was performed according to our published methods (25) with the following modification. At the very end of the immunostaining procedure, the sections were counterstained with 0.2% neutral red. The antibody employed in these studies is specific for the D_1A receptor, as documented by previously published studies in D_1A receptor transfected LTK mouse fibroblast cells (25).

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

The antisense and sense oligonucleotide probes that were used in the in situ hybridization were chosen specifically because of their complementarity to a mRNA sequence lying within both the native and amplified mRNA product and their uniqueness in terms of sequence homology to known receptor or other mRNAs in the rat. Background labeling was minimal or nonexistent in all sections examined.

Sections processed for nonradioactive in situ hybridization for D_1A receptor mRNA revealed positive signal almost exclusively in the cortex of the kidney. Those sections hybridized with antisense probe demonstrated a signal in both vascular and tubular structures with a high signal-to-noise ratio. Those structures previously labeled with D_1A receptor antibodies by immunocytochemistry were also shown to contain D_1A receptor mRNA. Thus proximal tubules and cortical collecting ducts as well as various orders of renal arterioles demonstrated a positive signal with less signal in distal tubular structures and complete absence of reaction in the glomeruli (Figs. 3-6). In addition, the hybridization signal was very sparse or absent in the medulla. Consecutive kidney sections probed with sense oligonucleotide, absence of anti-digoxigenin or nonamplified sections probed with antisense oligonucleotide failed to reveal any hybridization signal.

View larger version (121K): [in this window] [in a new window]   Fig. 3.   Light micrograph of rat renal cortical and medullary tissue demonstrating positive hybridization signal for D1A receptor mRNA. A: positive staining was noted in tubules predominantly in the cortex of kidney sections, with virtually no signal in the medulla. PT, proximal tubule; DT, distal tubule. B: sense oligonucleotide control for A. Magnifications, ×85. All kidney sections in Figs. 3-7 underwent prior mRNA amplification unless otherwise stated.

View larger version (126K): [in this window] [in a new window]   Fig. 4.   A: in situ amplification for D1A receptor mRNA with a positive hybridization signal in the tubules; magnification, ×145. B: higher-power view (magnification, ×320) showing D1A receptor mRNA in tubules. C: high-power magnification (×620) showing D1A receptor mRNA in tubules. D: in situ amplification for D1A receptor mRNA in which amplification enzyme AMV-RT was omitted. No signal is present. Magnification, ×320. E: in situ amplification for DA receptor mRNA in which the secondary antibody was omitted. No signal is present. Magnification, ×320.

View larger version (119K): [in this window] [in a new window]   Fig. 5.   A: positive hybridization signal for D1A receptor mRNA in a proximal tubule. Note that signal appears to be localized toward apical portion of tubular cells. Magnification, ×830. B: sense oligonucleotide control; magnification, ×330. G, glomerulus.

View larger version (103K): [in this window] [in a new window]   Fig. 6.   A: in situ hybridization for D1A receptor mRNA. Hybridization signal is localized to the afferent arteriole of the juxtaglomerular apparatus and adjacent tubules. Interstitium and glomeruli are devoid of signal. Magnification, ×600. B: positive hybridization signal in a middle order renal arterial vessel in smooth muscle layer; magnification, ×600. C: similar order renal vessel probed with antisense oligonucleotide probe but without prior D1A receptor mRNA amplification; magnification, ×600. D: high-power magnification (×800) of a renal vessel showing a signal for D1A receptor mRNA in medial vascular smooth muscle layer.

Figure 3A is a typical example of the renal cortex exposed to the D_1A receptor digoxigenin-labeled antisense probe. It can be seen that there is a marked difference in the signal intensity in the cortex and medulla. Figure 3B depicts a consecutive section probed with sense oligonucleotide. No hybridization signal was detected. Figure 4 depicts several views of tubules in the renal cortex. In Fig. 4, A-C, D_1A receptor mRNA signal can be observed in proximal tubules at ascending magnification. In Fig. 4D, no signal is present when the amplification enzymes are omitted. In Fig. 4E, no signal is apparent in the absence of secondary antibody. Hybridization signals were detected in proximal tubular cells (Fig. 5A). The signal appeared more intense toward the apical pole of such tubules. Control sections exposed to sense hybridization did not produce signal (Fig. 5B).

Figure 6A shows a high power view of the juxtaglomerular apparatus. The hybridization signal was present in the afferent arteriole and adjacent tubules. No signal was detected in the glomerulus or the interstitium. Figure 6B depicts a positive signal in a large order renal arterial vessel. The signal is localized to the smooth muscle and adventitial layers of the blood vessel. The requirement for amplification of the D_1A receptor mRNA to detect signal is demonstrated in Fig. 6C, which depicts a similar renal arterial vessel probed with an antisense oligonucleotide, but which has not undergone prior amplification. Figure 6D shows a high-power magnification of a renal vessel showing signal for the D_1A receptor mRNA in the medial smooth muscle layer.

To confirm the fidelity of the D_1A receptor mRNA signal in the kidney, nonamplified brain sections containing regions known to express D_1A receptor mRNA also were probed with the antisense and sense oligonucleotides. Strong hybridization signals were detected throughout the caudate putamen and nucleus accumbens, as well as in the deep layers of the cerebral cortex, as would be expected from the known sites of D_1A receptor expression within the brain (data not shown).

Along with the above control experiments, we also demonstrated the in vitro amplification with RT-PCR of a predicted 425-bp product from RNA isolated from rat kidney and brain tissue, as well as from a murine fibroblast LTK cell line that had been stably transfected with a full-length rat D_1A receptor cDNA. Figure 7 shows the size analysis of the PCR products in an ethidium bromide stained 2% agarose gel. PCR reactions performed in the absence of reverse transcriptase or on nontransfected LTK cell RNA were negative. In addition to the above, we also colocalized D_1A receptor protein and mRNA using an antibody to the third extracellular loop of the rat D_1A receptor and the 3SR in situ mRNA amplification technique (data not shown). Receptor protein and mRNA were coexpressed at the various renal sites shown previously to contain D_1A receptor mRNA.

View larger version (93K): [in this window] [in a new window]   Fig. 7.   Polymerase chain reaction product analysis in 2% ethidium bromide-stained agarose gel. An amplification product of predicted 425 bp size was demonstrated for reverse transcription-polymerase chain reactions using extracted RNA from rat kidney (lane 3), rat brain (lane 4), and from a murine fibroblast cell line (LTK) that had been stably transfected with a full-length rat D1A receptor cDNA (lane 5). No amplified products were detected from RNA extracted from the nontransfected parent LTK cell line (lane 6) or from RNA extracted from rat kidney when PCR was performed in the absence of reverse transcriptase (lane 2).

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

In situ hybridization is a basic molecular biological technique, whose key feature is that the sample nucleic acid sequence is detected directly in the intact cell or tissue. An important limitation to standard nonamplified in situ hybridization is its detection threshold, which is much higher than that for other solution-based hybridization methods. This is probably due in part to difficulties of the nucleic acid probe finding its target sequence because it has to transverse the labyrinth of nuclear proteins and nucleic acids in the in situ analysis (35).

The detection of dopamine receptor subtype-specific mRNA in peripheral tissues using such techniques as Northern blot analysis and in situ hybridization histochemistry has been singularly unsuccessful, this failure being attributed to the low abundance of mRNA in such tissues (4, 21, 32, 38). PCR in situ hybridization could be used to overcome some of these difficulties. However PCR in situ hybridization suffers from important weaknesses (24). Nonspecific DNA synthesis secondary to primer-dependent and -independent pathways may predominate. These can result from mispriming, primer oligomerization (primer dependent), and DNA repair (primer independent). In addition, if primers span an intron of the genomic DNA, then the possibility of pseudo-gene amplification arises. Primer-independent pathways of nonspecific DNA synthesis probably predominate in in situ PCR and are the result of the repair of single-stranded DNA nicks in double-stranded DNA molecules generated by the heating steps of PCR (24).

The difficulties of nonspecific DNA signal in in situ PCR have been circumvented, in part, by the recent development of the reverse transcriptase in situ PCR (RT-in situ PCR) technique (17). This technique detects PCR-amplified RNAs (cDNAs). Because RT in situ PCR incorporates DNase digestion of all DNA, the possibility of a nonspecific DNA signal is removed, as indeed is the native DNA. With the addition of reverse transcriptase, mRNA can then be amplified. The use of RT-PCR in isolated renal structures and solution hybridization-RNase protection studies of poly(A)⁺ RNA extracted from such isolates has revealed dopamine receptor subtype-specific mRNAs (36). Despite these advances, however, the continued use of PCR generates significant problems. Among these problems is the number of steps involved in the thermal cycling of the reaction. For 30 cycles of PCR, 90 temperature transfers must be performed, and the incubation times must be monitored. Therefore, the use of expensive automated equipment is mandatory when PCR is to be routinely employed in assays. Moreover, the requirement for the multiple heat-denaturing steps obligates the use of tissue protection measures to ensure retention of adequate histology. Another problem is that, at best, the number of product molecules only doubles in each PCR cycle. Even this low per-cycle efficiency is rarely achieved in practice, requiring the number of cycles to be increased to reach a specific amplification level. In addition, PCR cannot distinguish RNA targets from DNA targets in a mixed nucleic acid sample when the RNA and DNA sequences are colinear. Direct detection of RNA targets in such samples requires the removal of all DNA sequences as in RT-PCR. By their nature, such techniques are best suited to isolated tissue components or extracted RNA and are unable to link the amplification signals to specific cell types. Indeed, RT-PCR was used in the present studies to confirm the presence of renal dopamine D_1A gene expression.

The present study outlines a novel method of amplification of specific mRNA sequences in whole tissue sections. RNA transcription is an inherent process of cellular and viral systems that allows the generation of multiple single-stranded RNA molecules from discrete nucleotide sequences. Guatelli et al. (13) have described a process of isothermal replication of a targeted nucleic acid sequence using a concerted three-enzyme in vitro reaction, which they have termed a self-sustained sequence replication reaction and which is modeled after the general scheme employed during retroviral replication. In the present study, we have modified the reaction that usually is conducted on isolated RNA to allow for its implementation on isolated tissue sections. Similar modifications of this methodology have been devised to amplify and localize viral mRNA in cultured cells (14, 37). The use of the self-sustaining transcription-based sequence replication, being isothermal and lacking the requirement of thermal cycling, is more appropriate to whole tissue sections than PCR and can delineate the cell type expressing specific mRNA sequences. In comparison to in situ PCR, the system does not require thermal cycling equipment, has rapid reaction kinetics (13) and has the inherent ability to distinguish between RNA and DNA targets. The lower thermal stringency employed raises the possibility of lower amplification specificity. However, the use of in situ hybridization with probes specific to the amplified product protects against this possible error. In addition, the amplification of RNA target molecules permits independent amplification of mRNA and not genomic copies that might be present in the same sample.

Using this unique method of in situ amplification, we have successfully demonstrated the presence of D_1A receptor-specific mRNA in the kidney both in situ in tissue sections, as well as from isolated kidney and D_1A receptor-transfected LTK cell mRNA. In situ mRNA expression can occur in a site-specific manner and correlates closely with the distribution of receptor-specific protein, as we have previously demonstrated using receptor antibodies (25). In conjunction with these studies, the present study provides evidence, at the level of gene expression, that the brain type D_1A dopamine receptor is also present in the kidney. In addition, the sites of mRNA expression identified in this study, together with the areas of receptor protein as delineated in our previous immunohistochemical studies, correspond closely to those areas previously identified with receptor-ligand binding and pharmacological studies as exhibiting D₁-like receptor activity (15).

In peripheral tissues, DA receptors are distributed in the sympathetic nervous system, the cardiovascular system, kidney and adrenal cortex (7). These have been classified into the D₁-like (DA₁) and D₂-like (DA₂) receptor subtypes on the basis of synaptic localization (11). The pharmacological properties of the D₁-like receptors roughly approximate those of D₁ and D₅ receptors, whereas those of the D₂-like receptors approximate those of the D₂, D₃ and D₄ receptors (1, 15). Within the kidney D₁-like receptors have been localized, by means of receptor ligand binding and pharmacological studies, to the medial layer of renal arterial vessels, the juxtaglomerular apparatus, the proximal convoluted tubule and cortical collecting duct. Such studies do not differentiate between the different subtypes of receptors given the similarities in affinities across receptor subtypes. However, differences in receptor-ligand binding capacity have suggested that these peripheral D₁-like receptors might differ from those described in the brain (8).

The D₁-like receptor in the kidney is associated with renal vasodilation and an increase in sodium excretion, mediated by cAMP which inhibits both Na⁺-H⁺ antiport and Na⁺-K⁺-adenosinetriphosphatase activities (1). DA is an intrarenal natriuretic hormone acting in a paracrine fashion on receptors adjacent to the site of production (27).

The molecular nature of dopamine receptors in the kidney has, until recently, remained undetermined. However, with the cloning of the D₃ receptor from rat brain, specific mRNA was coincidentally identified in the kidney (28). More recent studies have amplified D_1A receptor cDNA from microdissected proximal convoluted tubules of the rat kidney by RT-PCR and confirmed this signal presence by solution hybridization (36). With similar techniques, D₂ long and D₃ receptor mRNA has been detected in the rat kidney (10). The present study provides further evidence that the molecular structure of the D_1A receptor in the kidney is similar to that of the D₁ receptor newly cloned from the brain and is in agreement with the studies published by Healy and colleagues (12) and Caron et al. (23) suggesting that brain dopamine receptor subtypes are homologous to those expressed within peripheral tissue sites.

In conclusion, the present study demonstrates several points: 1) the rat kidney expresses the D_1A dopamine receptor gene, 2) D_1A receptor gene expression occurs in a site-specific manner, 3) sites of mRNA expression concur with those previously described as containing receptor protein, 4) D_1A receptor mRNA and protein are colocalized in the same renal cells, 5) the use of a novel transcription-based amplification system allows for detection of mRNA not feasible with Northern blot analysis or standard in situ hybridization histochemistry, and 6) these experiments provide further evidence that dopamine receptor subtypes cloned from the brain are homologous to those expressed in peripheral tissues.