help button home button Endocrine Society Endocrinology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Virlon, B.
Right arrow Articles by Elalouf, J.-M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Virlon, B.
Right arrow Articles by Elalouf, J.-M.
Endocrinology Vol. 139, No. 6 2784-2795
Copyright © 1998 by The Endocrine Society

ARTICLES

Rat G Protein-Coupled Receptor Kinase GRK4: Identification, Functional Expression, and Differential Tissue Distribution of Two Splice Variants1

Bérangère Virlon2, Dmitri Firsov2,3, Lydie Cheval, Eric Reiter, Carine Troispoux4, Florian Guillou and Jean-Marc Elalouf

Département de Biologie Cellulaire et Moléculaire (B.V., D.F., L.C., J.-M.E.), Service de Biologie Cellulaire, CEA Saclay, 91191 Gif-sur-Yvette Cedex, France; Institut National de la Recherche Agronomique (INRA) (E.R., C.T., F.G.), Station de Physiologie de la Reproduction des Mammifères Domestiques, URA CNRS 1291, 37380 Nouzilly, France

Address all correspondence and requests for reprints to: J. M. Elalouf, Département de Biologie Cellulaire et Moléculaire, Service de Biologie Cellulaire, CEA SACLAY, 91191 Gif-sur-Yvette Cedex, France. E-mail: elalouf{at}dsvidf.cea.fr


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
G protein-coupled receptor kinases (GRKs) specifically phosphorylate the agonist-occupied form of G protein-coupled receptors, leading to the homologous mode of desensitization. We report here on the cloning of complementary DNAs that encode two rat GRK4 variants. Rat GRK4A (575 amino acids) displays 76% identity with the long human GRK4 splice variant. Rat GRK4B (545 amino acids) delineates a new variant that is identical to GRK4A except for a 31-amino acid deletion in the N-terminal domain, corresponding to exon VI in the human GRK4 gene. GRKs4A and B are likely produced by alternative splicing from a single gene, the partial characterization of which revealed a structural organization similar to that of the human GRK4 gene. GRK4A messenger RNA (mRNA) is abundant only in testis. A combination of in situ hybridization and quantitative RT-PCR studies demonstrated that GRK4A mRNA level increases during testicular development and predominates in leptotene to late pachytene primary spermatocytes and round spermatids. GRK4B mRNA is poorly expressed in testis and most rat tissues but is heterogeneously distributed in the kidney, with 20-fold enrichment in the outer medulla. GRKs4A and B are both functional protein kinases, as demonstrated in a rhodopsin phosphorylation assay. The differential tissue distribution of GRKA4 and GRK4B suggests that individual GRK4 variants may serve distinct physiological functions.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
G PROTEIN-COUPLED receptors can undergo two types of desensitization that have been shown to depend on different types of protein kinases (1). The nonspecific, or heterologous, desensitization is initiated by second messenger-dependent protein kinases, such as protein kinases A or C. On the other hand, the specific, or homologous mode of desensitization is triggered by G protein-coupled receptor kinases (GRKs). These serine/threonine kinases have the unique property of recognizing only agonist-occupied receptors, leaving unaltered the function of nonactivated receptors (2). According to the current model of rhodopsin or ß2-adrenergic receptor desensitization, GRKs-phosphorylated receptors interact with arrestin or arrestin-like proteins that prevent their further coupling to G proteins, leading to a desensitized state (1, 2).

In mammals, complementary DNAs (cDNAs) corresponding to six different GRK genes have been characterized (3, 4, 5, 6, 7, 8). Structural and functional similarities among GRKs enabled their classification into three subfamilies (7). The first subfamily only includes GRK1 (rhodopsin kinase). The second subfamily is composed of the two ß-adrenergic receptor kinase subtypes (ßARK1, or GRK2 and ßARK2, or GRK3), and the third subfamily includes GRK4, GRK5, and GRK6. The number of mammalian GRKs has been expanded to ten members through identification of alternative splice variants. Four GRK4 variants have been described in human (9, 10), and two GRK6 variants were recently described in the rat (11).

Expression of GRK4 has been investigated in human and baboon tissues and was found abundant only in the testis (6, 10, 12). The long GRK4 messenger RNA (mRNA) variant is held to be predominant in this tissue (10), whereas the tissue distribution of the other variants is unknown. The only available information indicates that a sequence portion common to all four variants is expressed at faint levels in several human tissues (6), and that a variant lacking a fragment encoding part of the N-terminal domain is expressed to much higher levels than the long mRNA form in human brain (9). Turning to the other GRKs, either tissue-specific (e.g. retina for GRK1 (5)) or widespread distribution (GRKs2, -3, 5, and -6) has been reported (3, 4, 7, 8, 11, 13, 14, 15). However, it is striking to note that none of the GRKs identified up to now is abundantly expressed in the kidney. As part of an effort to unravel the mechanisms of desensitization to hormone action in the kidney, we have looked for GRKs expressed in this tissue. RT-PCRs carried out on the rat kidney with degenerate primers enabled us to obtain cDNA fragments homologous to human GRK4 and GRK6. Rat GRK6 has been described previously (11). We report here on the cloning and functional expression of two rat GRK4 variants and demonstrate the differential tissue distribution of their mRNAs. We also show that the GRK4 gene transcript, which undergoes extensive alternative splicing, gives rise to species-specific variants.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Radioactive nucleotides and deoxynucleotides were from Amersham (Arlington Heights, IL). T7 DNA polymerase, Moloney murine leukemia virus reverse transcriptase (MMLV-RT), Superscript MMLV-RT, and reagents for cell culture and transfection were purchased from Life Technologies (Gaithersburg, MD). Taq Polymerase was from Eurobio (Les Ulis, France). Other modification enzymes as well as ß-agarase and restriction enzymes were from New England Biolabs (Beverly, MA). Oligonucleotide primers were purchased from Bioprobe Systems (Montreuil, France).

RNA preparation
Pachytene spermatocytes and round spermatids were isolated from 90-day-old rats. Both cell populations were prepared and characterized as described by Weiss et al. (16). Total RNAs were extracted from tissues and enriched testis cell populations by the acid guanidinium thiocyanate-phenol-chloroform method (17). When extraction was carried out on isolated nephron segments, a microadaptation of this method was used (18). Poly (A) RNAs were selected on oligo(deoxythymidine) [oligo(dt)] columns (Clontech, Palo Alto, CA).

RT-PCR amplification of GRK-related cDNAs
Degenerate primers were chosen in the cDNA portion encoding the catalytic domain of GRKs. The sequences of the sense and antisense primers were 5'-GGIGGITTYGGIGARGT-3' and 5'-ASYTCIGGIGCCATGWAIC-3', respectively (I, inosine; R, A or G; S, G or C; W, A or T; Y, C or T). RT (45 min at 41 C) was carried out using total RNAs extracted from 3 mm of microdissected medullary thick ascending limbs of Henle’s loop, corresponding to about 1,000 cells (19). The reaction also contained (50 µl final volume) 200 U of MMLV-RT, 10 nM of the antisense primer, 8 mM DTT, 400 µM of each dNTP, 3.5 mM MgCl2, and 5 µl of 10x Taq PCR buffer (500 mM KCl, 200 mM Tris-HCl (pH 8.3), 1 mg/ml gelatin). After completion of RT, the temperature was raised to 95 C for 30 sec, then equilibrated to 80 C. PCR was initiated by adding 50 µl of a mix containing 1.25 U of Taq Polymerase, 200 nM of sense and antisense primers, 1.5 mM MgCl2 and 5 µl of 10x Taq PCR buffer. The samples were processed for 35 cycles (95 C, 30 sec; 47 C, 30 sec; 72 C, 1 min), the last of which had an elongation time of 10 min. The content of six similar reactions were pooled, electrophoresed through a 2% low melting point (LMP) agarose gel, and the approximately 500-bp DNA fragment recovered by agarose digestion. An aliquot (5%) of the gel-purified DNA was reamplified with the phosphorylated degenerate primers using 16 PCR cycles, as indicated above. The PCR sample was again purified by agarose gel electrophoresis, blunt-ended with T4 DNA polymerase and cloned into the EcoR V site of pBluescript (BSSK+).

Cloning of the 3' and 5' ends of rat GRK4
The 3' and 5' ends of rat GRK4 were isolated using the rapid amplification of cDNA ends (RACE) method. For 3' RACE, first strand cDNA was synthesized from 1 µg of rat kidney poly (A) RNAs primed with 1 pmol of a lock-docking oligo(dT) primer (11). RT was initiated at 37 C (30 min) in the presence of 200U of Superscript MMLV-RT, then extended for 45 min at 42 C after adding fresh enzyme. The cDNA was purified using GlassMAX columns (Life Technologies). A series of PCR was then carried out using the lock-docking oligo(dT) primer (step 1) or the adapter 5'-GGCCGCAGATCTAGATATCG-3' (step 2 and 3) as antisense primers, and three nested sense primers selected from the partial cDNA sequence of rat GRK4. Their positions (608–633, 882–907 and 1032–1055), as well as those of all subsequent primers used in this study are numbered from the ATG translation initiation codon of the sequence depicted in Fig. 1Go.



View larger version (72K):
[in this window]
[in a new window]
  		Figure 1. Nucleotide and deduced amino acid sequence of rat clones coding for type IV G protein-coupled receptor kinase (GRK4). Nucleotides are numbered from the A residue (+1) of the translation initiation codon, and numbering of the predicted amino acid sequence is indicated just below. Nucleotides -188 to 1809 correspond to the sequence of clones 1T12 and 2T12 (GRK4A, long form). Bracketed residues 441 to 533 were absent in clone 3T12 (GRK4B, short form), which was otherwise identical to the long cDNA variant. Additional variants were evidenced by RT-PCR analysis of different parts of the coding region (see Figs. 9Go and 10Go): in GRK4C, the deletion observed in GRK4B extended to nucleotide 597 (white arrow); in GRK4D, bracketed residues 1405–1542 were deleted. The end of the 3' untranslated region (nucleotides 1810 to 1955) was deduced from the sequence of the cDNA clones obtained by 3' RACE. Underlined sequences correspond to consensus polyadenylation signals, and black arrows indicate beginning of the poly (A) tail in each 3' RACE clone.

 
For 5' RACE, first strand cDNA was synthesized from 2 µg of total testis RNAs primed with an antisense oligonucleotide (position 1972–1997) specific for rat GRK4. PCR was then performed on the tailed cDNA using the 5'-AmpliFinder RACE kit (Clontech, Palo Alto, CA). Three antisense primers (location: 1972–1997, 986-1010, and 630–656) were successively used in conjunction with the anchor primer provided with the kit. The final DNA fragment generated by this procedure corresponded to nucleotides 60–656 of the coding region. To obtain the missing 5' coding sequence, the RACE product generated with the antisense primer corresponding to nucleotides 986-1010 was reamplified. This was performed using a new antisense oligonucleotide (location: 69–94) and the anchor primer.

Isolation of the full-length coding region
The full-length coding region of rat GRK4 cDNA was obtained through a single-step RT-PCR procedure using sense and antisense primers corresponding to nucleotides (-)188–(-)167 and (+)1972–(+)1997, respectively. RT (50 µl reaction volume) was carried out as described above using 2 µg of rat testis total RNAs and 125 nM of the antisense primer. Aliquots (5 µl) of the cDNA were then processed for 40 PCR cycles (95 C, 30 sec; 58 C, 30 sec; 72 C, 2 min 10 sec) carried out using Vent DNA polymerase. The reaction (100 µl final volume) contained 1U of enzyme, 100 nM of phosphorylated sense and antisense primers, 400 µM dNTP, 0.1 mg/ml acetylated BSA, and 10 µl of 10x Vent PCR buffer. The predominant product (2.0 kb) obtained from 10 separate reactions was recovered from a preparative agarose gel and cloned into the EcoR V site of BSSK. Three clones (1T12, 2T12 and 3T12) were rescued and fully sequenced on both strands.

Southern blot analysis and cloning of rat GRK4 gene fragments
For Southern hybridization analysis, single restriction digests of rat genomic DNA (20 µg) were performed. Duplicate aliquots (10 µg) of the samples were electrophoresed through a 0.8% agarose gel, transferred by capillary blotting onto a nylon membrane (Hybond N+, Amersham), and hybridized either to a rat or a human GRK4 cDNA probe. The rat cDNA probe corresponded to nucleotides 875-1067 of rat GRK4 coding region. The human cDNA probe corresponded to the homologous portion of human GRK4{alpha} (nucleotides 878-1064 of the coding sequence). The probes were 32P-labeled to a specific activity of 109cpm/µg with the random hexamers priming method. Prehybridization (8 h at 40 C) was performed in a buffer containing 8 mM Tris-HCl (pH 7.5), 40% formamide, 4 x SSC (1 x SSC = 150 mM NaCl, 15 mM sodium citrate, pH 7.0), 5 x Denhardt’s, 0.2% SDS, 50 mM sodium phosphate, and 100 µg/ml heat-denatured salmon sperm DNA. Hybridization (16 h at 40 C) was performed in the same solution supplemented with 10% dextran sulfate and the DNA probe (2 x 106cpm/ml). The blot was first washed under low stringency conditions (two washes in 0.5 x SSC/0.1% SDS at room temperature, followed by three 20 min washes in the same solution at 45 C), and exposed 24 h at -70 C for autoradiography. Thereafter it was washed under moderate (0.1 x SSC/0,05% SDS at 45 C) and then under high stringency conditions (0.1 x SSC/0.05% SDS at 65 C), and each time reexposed for autoradiography.

For construction of genomic DNA sublibraries, 300 µg of rat genomic DNA were digested with HindIII and size fractionated on a 1% LMP preparative agarose gel. Two strips of 0.4 cm, corresponding to DNA fragments of approximately 3 or 6 kb, were sliced, then treated with ß-agarase. The purified DNA was ligated to HindIII-, calf intestinal alkaline phosphatase-treated BSSK. The ligation product was used to transform Escherichia coli (E. coli) XL2-blue ultracompetent cells (Stratagene, La Jolla, CA). Bacterial cells were plated at a density of 6,000 colonies per dish onto 140-mm Petri dishes containing X-Gal and IPTG. 105 and 4 x 104 recombinant clones (as judged from blue/white color selection) were obtained for the 3- and 6-kb genomic DNA insert, respectively. Screening was carried out by colony hybridization using the rat GRK4 cDNA probe, prehybridization, and hybridization conditions described above. Final washes were performed in 0.2 x SSC/0.05% SDS at 45 C for the 3-kb DNA fragment, and in 0.1 x SSC/0.05% SDS at 50 C for the 6-kb fragment. After two rounds of screening, one clone was isolated for each construct. The presence of the desired genomic DNA fragment was checked by restriction map analysis and Southern blot analysis, and the insert was fully sequenced.

Rhodopsin phosphorylation assay
The full-length rat GRK4 cDNAs were recovered from pBluescript constructs through XbaI-KpnI digestion. They were inserted under the control of the cytomegalovirus promoter into the XbaI-KpnI sites of the pCB6 expression vector containing the hygromycin gene as a selection marker (11). CHO cells grown in Ham’s F12 medium supplemented with 10% FCS and 2 mM L-glutamine were transfected using lipofectamine. Recombinant clones were selected using 500 µg/ml hygromycin B (Boehringer Mannheim, France) and checked for GRK4 expression by RT-PCR.

For phosphorylation assays, the cells were grown in 225-cm2 flasks. The culture medium was supplemented with 10 mM sodium butyrate for 16 h before the experiment. The cells were scrapped into 600 µl of ice-cold buffer (25 mM Tris-HCl (pH 7.5), 5 mM EDTA, 10 mM Mg(OAc)2, 5 mM DTT, 0.1 mM phenylmethylsulfonyl fluoride, 10 µg/ml pepstatine, 10 µg/ml leupeptine, and 10 µg/ml benzamidine), homogenized with a Dounce tissue disrupter. The samples were then centrifuged at 300,000 x g for 30 min at 4 C, and the supernatant recovered.

Rhodopsin phosphorylation was carried out using urea-treated bovine rod outer segments obtained as described (20). The assay (50 µl final volume) contained 5 µM rhodopsin, 20 mM Tris-HCl (pH7.5), 2 mM EDTA, 6 mM MgCl2, 100 µM ({gamma}-32P) ATP (1000–1500 cpm/pmol) and 1 µg of CHO cells cytosolic proteins. The samples were incubated in the dark or in room light for 20 min at 25 C, quenched by adding 12.5 µl of SDS sample buffer, and electrophoresed through a 10% homogeneous SDS polyacrylamide slab gel. The gel was fixed, stained with Coomassie blue, vacuum-dried and exposed 16–48 h at -70 C for autoradiography.

Northern blot analysis
The GRK4 probe was identical to that used for Southern experiments and library screening. The GRK6 probe corresponded to bases 583–968 of the rat GRK6 coding region (11). Poly (A) RNAs were denatured in glyoxal-dimethyl sulfoxide solution, electrophoresed through a 1% agarose gel, and transferred onto a nylon membrane (HybondN+, Amersham). Labeling of DNA probes, prehybridization, and hybridization conditions were performed as described above. Final washes were performed at 45 C in 0.1 x SSC and 0.05% SDS.

In situ hybridization
Tissues and slides preparation. Ten-day-old, 25-day-old, and 8-month-old rats were anesthetized and testis were fixed in vivo by intracardiac perfusion with 4% paraformaldehyde in 0.1 M PBS (pH 7.4). Testis were removed and postfixed by immersion in 4% paraformaldehyde in PBS for 24 h. Subsequently, they were dehydrated, paraffin embedded, and sectioned at a thickness of 8 µm. The tissue samples were then transferred to slides that had undergo the followings: cleaning in ethanol/ether (10:1), immersion for 10 sec in a 1% 3-aminopropyltriethoxylane (Tespa) (Aldrich Chemical Company, Saint Quentin, Fallavier, France) solution in acetone, then in acetone, two rinsings in distilled water, and overnight drying at 37 C.

Probes. We used [35S]-labeled sense and antisense cRNA probes. The full-length rat GRK4A cDNA cloned in BSSK was linearized with either SmaI (sense) or HindIII (antisense). Briefly, 1 µg of linearized plasmid was incubated in appropriate transcription buffer containing 500 µM rATP, rGTP, and rCTP, 50 µCi of [35S]-UTP, and 5 µM UTP in the presence of T3 (antisense) or T7 (sense) RNA polymerase (Promega) for 1 h at 37 C. Then, samples were incubated with DNAse RQ1 (Promega) and purified on a Sephadex G50 column (Pharmacia, Piscataway, NJ). The amount of radioactivity incorporated in the probe was always >80%.

In situ hybridization procedures. Tissue sections were processed through deparaffinization and rehydration following a routine histologic procedure. Then, sections were heated in 0.01 M sodium citrate buffer (pH 6.0) in a microwave oven (21). After two heating cycles of 5 min each, sections were successively let to cool in the same solution for 20 min, postfixed in 4% paraformaldehyde, digested with proteinase K, dehydrated, and finally air-dried.

For prehybridization, sections were incubated for 2 h at 50 C in 50% formamide, 10% dextran sulfate, 1 mg/ml herring sperm DNA, 2 x SSC, and 70 mM DTT. For hybridization, 35S-labeled antisense and sense cRNA probes were diluted in the prehybridization buffer and applied on sections (400,000 cpm/80 µl) that were then covered with coverslips. Hybridization was performed overnight at 50 C in a sealed humidified container and was followed by stringent washings: 1) 5 x SSC, 1 mM DTT for 30 min at room temperature; 2) 5 x SSC, 1 mM DTT for 30 min at 50 C; 3) 50% formamide, 2 x SSC, 1 mM DTT for 20 min at 50 C; 4) 0.5 M NaCl, 5 mM EDTA, 10 mM Tris-HCl pH 7.5 (NaCl-TE) for 10 min at 37 C twice; 5) 15 µg/ml RNAse A in NaCl-TE for 1 h at 37 C; 6) NaCl-TE for 20 min at 37 C; 7) 0.1 x SSC for 15 min at room temperature. Finally, the sections were dehydrated through increasing concentrations of ethanol diluted in 0.3 M ammonium acetate and were air-dried. After immersion in liquid photographic emulsion NTB2 (Kodak, Rochester, NY), the slides were exposed in the dark for 1 month, then photographically developed and fixed, and finally counterstained with hematoxyline. The microphotographs were taken from slides processed in the same conditions within the same experiment after the same exposure time to ensure an accurate comparison of the labeling signal.

RT-PCR analysis of GRK4 mRNAs expression in rat tissues
Quantitative RT-PCR. These experiments were performed using primers (location: sense, 404–429; antisense, 665–690) flanking the 93-bp sequence specific for the long cDNA form (see above). For standardization purpose, RT-PCR was carried out in the presence of a mutant cRNA, used as internal standard.

A GRK4 mutant with a 46-bp deletion was generated by inverse PCR (22) using antisense and sense primers corresponding to nucleotides 511–535 and 582–604 of the coding region, respectively. Inverse PCR was carried out in a 100-µl reaction volume containing 0.5 ng of the long BSSK-GRK4 cDNA construct, 100 nM of phosphorylated primers, 4 mM DTT, 200 µM dNTP, 5 U of Taq polymerase, 5 U of Taq Extender, and 10 µl of 10 x Taq Extender mix (Stratagene). Following PCR (95 C, 30 sec; 57 C, 30 sec; 74 C, 5 min; 20 cycles), the DNA fragment was purified by gel electrophoresis, blunt-ended with T4 DNA polymerase, circularized with T4 DNA ligase, and used to transform E. coli XL1-blue competent cells. Recombinant clones were screened by PCR with the primers used for quantitative RT-PCR, and the deletion was checked by DNA sequencing. The cRNA was synthesized by in vitro transcription as previously described (18), using 1 µg of XbaI-linearized template.

The experimental procedures for quantitative RT-PCR were similar to those described previously for other such assays (18, 23). The PCR cycles (95 C, 30 sec; 60 C, 30 sec; 72 C, 45 sec) were performed in the presence of [{alpha}-32P]deoxy-CTP (dCTP) (specific activity in the assay: 0.5 µCi/nmol). The reaction included 28 PCR cycles, the last of which had an elongation time of 10 min. Under these conditions, the amount of DNA formed was below or just above the detection threshold of the ethidium bromide staining method (end product concentration <1 nM) but could be detected by autoradiography. These experimental conditions allow exponential accumulation of DNA to occur throughout the PCR reaction, and hence the direct comparison between wild-type and mutant signals (18, 23). Quantitation of mutant and wild-type signals was performed by storage phosphor technology (24) using a Storm 840 and ImageQuant software (Molecular Dynamics, Sunnyvale, CA).

Hybridization analysis of GRK4 variants. The 5' end coding region was analyzed for alternative splicing using sense and antisense primers corresponding to nucleotides 1–21 and 382–403, respectively. In the 3’ end coding region, alternative splicing processes were studied using sense and antisense primers corresponding to nucleotides 1381–1401 and 1776–1796, respectively.

In a first series of experiments, RT-PCR was performed in the presence of [{alpha}-32P]dCTP (0.5 µCi/nmol), and the different splice variants were detected by autoradiography. RT (primed with 125 nM of antisense primer) and PCR (95 C, 30 sec; 55 C, 30 sec, and 72 C, 45 sec; 28 cycles, with a 10-min elongation step for the final one) steps were carried out using standard buffers (see amplification of GRK-related cDNAs). In a second series of experiments, RT-PCR was performed under similar conditions, except that [{alpha}-32P]dCTP was omitted from the reaction. The splice variants were then characterized by Southern hybridization. Briefly, duplicate aliquots (10 µl) of the PCR products were fractionated on a 2% agarose gel and transferred onto a HybondN+ membrane. The blot was divided in two. Each part was hybridized with an oligonucleotide probe corresponding either to residues 1497–1516 or 1552–1570 of rat GRK4A cDNA. The oligonucleotides were end-labeled with [{gamma}-32P] ATP to a specific activity of 109 cpm/µg, using T4 polynucleotide kinase. Prehybridization and hybridization were performed at 50 C in 6 x NET (1 x NET = 150 mM NaCl; 15 mM Tris HCl (pH 7.5), 1 mM EDTA), 0.05% Nonidet P40, 100 µg/ml yeast transfer RNA and 5 x Denhardt’s solution. The blots were then washed twice for 20 min at room temperature in 2 x SSC/0.1% SDS and exposed for autoradiography. The relative amount of each splice variant was quantitated as described above by storage phosphor technology (24).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
The GRKs so far identified display a similar overall organization characterized by an approximately 250 amino acids long catalytic domain flanked by an amino-terminal domain of approximately 190 amino acids and a carboxy-terminal domain of variable length (105–223 amino acids) (3, 4, 5, 6, 7, 8, 10, 11). Because the centrally located catalytic domain is the most conserved portion of GRKs (45–95% sequence identity), we designed degenerate oligonucleotide primers to amplify part of the cognate cDNA sequence. RT-PCR carried out with such primers on RNAs obtained from isolated renal tubular cells yielded a product of the expected size (approximately 500 bp), which was subsequently cloned. Of the 48 clones sequenced, 5 were identical and had a 485-bp insert with strong homology to human GRK4 (88% identity). The other sequences were either homologous to human GRK6 (17 clones), identical with the published rat GRK2 sequence (1 clone), or unrelated to GRKs (25 clones).

Full-length GRK2 (3, 13, 25, 26), GRK3 (3, 13, 27), GRK5 (7, 14, 15) and GRK6 (8, 11) coding sequences from several species have already been published. By contrast, GRK4 is only available in human (6, 10). We therefore sought to obtain additional sequence information by using the RACE method. The sequence of the 5' and 3' RACE products then enabled us to design rat-specific primers that were used to isolate the full-length coding region through a single step RT-PCR procedure. Because GRK4 is mainly expressed in the testis (6), RT was carried out on rat testis RNAs, and amplification was performed using a high fidelity thermophilic (Vent) DNA polymerase. The DNA fragment of the predicted size (2 kb) was cloned and three clones (1T12, 2T12 and 3T12) were fully sequenced on both strands.

Clones 1T12 and 2T12 were identical and displayed an insert of 1997 bp (Fig. 1Go). The cloned cDNA fragment contains a 1725-bp open reading frame starting with a Kozak consensus sequence (28) preceded by a 188-bp G-C rich (65%) region. These features, and the strong homology of the N-terminal region with that of several GRKs helped to define the translation initiation codon. The ORF encodes a protein of 575 amino acids with a predicted molecular mass of 66.8 kDa and an isoelectric point (pI) of 8.41. With regards to the third clone analyzed, its sequence was identical to that described above except for a 93-bp deletion in the coding region. The deletion does not change the reading frame, so that the protein predicted from the short cDNA variant is only reduced in length (544 instead of 575 amino acids). The two protein sequences were compared with mammalian GRKs. The highest homology was found with GRKs 4–6, and the alignment is shown on Fig. 2Go. The protein encoded by the long cDNA form is much more related to human GRK4{alpha} (76% amino acid identity) than to human GRKs5–6 (63% amino acid identity with both kinases). This sequence alignment suggests that we have isolated cDNAs encoding two rat GRK4 subtypes, which may be either produced from two highly related genes or from a single gene by a process of alternative splicing. The sequence deleted in the short variant matches that encoded by exon VI in the human GRK4 gene (10).



View larger version (65K):
[in this window]
[in a new window]
  		Figure 2. Alignment of the amino acid sequences of rat GRK4A and GRK4B with human (Hu) GRK4, GRK5, and GRK6. Hu GRK4, GRK5, and GRK6 sequences are from Premont et al. (10), Kunapuli and Benovic (7), and Benovic and Gomez (8), respectively. For Hu GRK4, the sequence of the long variant (GRK4{alpha}) is shown; residues absent in the amino-terminal (GRK4ß), carboxy-terminal (GRK4{gamma}) or in both domains (GRK4{delta}) of the short human variants are boxed. Only the amino acids different from those present in rat GRK4A are shown. Identical residues are indicated as a dash (-), whereas gaps introduced to maximize homology are shown as points (.). Numbers in parentheses indicate the size of the different polypeptides.

 
To analyze the genetic diversity of GRK4 coding sequences in the rat, Southern hybridization experiments were performed using a cDNA probe corresponding to part of the catalytic domain. As shown in Fig. 3AGo, under low stringency conditions, a 193-bp probe hybridized to 2–3 restriction fragments for each tested enzyme. Under high stringency conditions (membrane washing in 0.1 x SSC at 65 C), a single hybridizing fragment was detected in each case (Msc I, 5.0 kb; BglII, 0.6 kb; EcoRI, 5.6 kb; HindIII, 6.3 kb; PstI, 9.0 kb; data not shown). To unravel this hybridization pattern, we decided to perform the sequence analysis of fragments specific to moderate stringency conditions, or detected under both hybridization conditions. The two HindIII-hybridizing fragments were well resolved from each other and judged suitable for purification. Therefore, HindIII-digested rat genomic DNA was preparatively size-fractionated on an agarose gel; the 3- and 6-kb regions were isolated and used to construct two genomic sublibraries that were screened with the cDNA probe under low or high stringency hybridization conditions, as appropriate. The insert of one positive clone from each sublibrary was analyzed by Southern hybridization and then fully sequenced.



View larger version (71K):
[in this window]
[in a new window]
  		Figure 3. Southern blot analysis of the rat genome with rat (A) and human (B) GRK4 cDNA probes. The probes corresponded to nucleotides 875-1067 and 878-1064 of the coding region of rat GRK4A and human GRK4{alpha} cDNA, respectively. The blot was washed under low stringency conditions (0.5 x SSC/0.1% SDS, 45 C) and exposed 24 h at -80 C for autoradiography. Note that both probes detected the same DNA fragments, except for the 3.3 kb-BglII restriction fragment that was specific for the rat cDNA probe. This fragment was shown by DNA sequencing to contain a 38-bp sequence that shares 100% identity with nucleotides 930–967 of the rat GRK4A cDNA coding region and displays six mismatches with the homologous human GRK4 (exon X) sequence. Position of the molecular weight markers ({lambda}BstEII digests, in kb) is shown on the left.

 
The clone isolated using low stringency hybridization conditions displayed an insert of 3056 bp that contains a 191-bp fragment sharing 100% identity with the rat cDNA sequence and 93% identity with exon 9 of the human GRK4 gene (data not shown). In the rat genomic sequence, the nucleotides found immediately before (tag) and after (gt) this 191-bp fragment fit with consensus sequences for the end and the beginning of an intron, respectively (29). The 3' end of the sequence shared with the cDNA corresponds to the initial 55 nucleotides of the probe and confers to the insert its hybridization properties. Indeed, there were no additional regions of homology between the probe and the insert.

The clone isolated using high stringency conditions contained a 6440-bp insert. Sequencing of the insert revealed the presence of a 38- and a 90-bp sequence each displaying 100% identity with the rat cDNA (data not shown). Their length and position are consistent with the hybridization pattern obtained on genomic DNA (e.g. the 38-bp sequence locates on a 3.3-kb-BglII restriction fragment, and 74 nucleotides of the 90-bp sequence lies on a 0.66-kb-BglII restriction fragment). The 38- and 90-bp fragments match with two exons (exons 10 and 11, respectively) of the human GRK4 gene, and are flanked by consensus sequences for the end ((c/t)ag) or the beginning (gt) of introns, as appropriate. They are colinear in the rat cDNA and, together, represents the bulk of the probe. No additional regions of homology were found between the probe and the 6440-bp insert.

The experiments reported above demonstrate that the different genomic fragments hybridizing with the rat cDNA probe belong to a single gene, the partial characterization of which revealed an intron/exon organization identical to that of the human GRK4 gene. Furthermore, the same hybridization pattern was obtained on the rat genome using either a rat (Fig. 3AGo) or a human (Fig. 3BGo) GRK4 cDNA probe. There is therefore no rat sequence much more homologous to human GRK4 than the one described here. All these data are consistent with the notion that different rat GRK4 variants are produced from a single gene by alternative mRNA splicing.

To check that the two rat cDNAs encode functional protein kinases, a phosphorylation assay was carried out using rhodopsin as a substrate. The full-length coding regions of the two isoforms were separately inserted into an expression vector (pCB6) and stably transfected into CHO cells. Phosphorylation of rhodopsin was performed using cytosolic proteins extracted from cells transfected with pCB6 or pCB6-GRK4 constructs (Fig. 4Go). Phosphorylation of light-bleached rhodopsin was approximately 4-fold higher with cytosolic proteins of GRK4A or GRK4B-transfected cells than with those obtained from control cells. Phosphorylation was fully agonist-dependent, as shown by the uniformly negative results obtained for reactions incubated in the dark.



View larger version (61K):
[in this window]
[in a new window]
  		Figure 4. Expression of rat GRK4A and GRK4B in CHO cells. CHO cells were transfected with the parental expression vector pCB6 (vector), the pCB6-GRK4A (GRK4A), or the pCB6-GRK4B (GRK4B) expression construct. Stable transfectants were selected and cytosol cell extracts (1 µg of protein) used for rhodopsin phosphorylation. Reactions were incubated for 20 min at 25 C in the dark or under room light, as indicated. The samples were fractionated on a SDS/10% homogeneous polyacrylamide gel and exposed for autoradiography.

 
The tissue distribution of GRK4 mRNAs was first investigated by Northern hybridization (Fig. 5Go). Expression of GRK6 mRNAs was also studied in these experiments and acted as control. Using a rat GRK6 specific probe, a major hybridizing fragment of approximately 3 kb was detected. GRK6 mRNA levels varied from one tissue to another but always reached measurable amounts, confirming the broad distribution of this kinase in rat tissues (11). Analysis of GRK4 mRNAs was carried out using larger poly (A) RNA amounts but, in this case, positive results were only obtained in testis. A single hybridizing fragment of approximately 2.5 kb was detected in this tissue. Additional rat tissues (heart, liver, spleen, stomach, colon, caecum, and seminal vesicle) were studied by Northern hybridization. In all of them, only GRK6 mRNAs were detected (data not shown).



View larger version (46K):
[in this window]
[in a new window]
  		Figure 5. Differential expression of GRK4 and GRK6 mRNAs in rat tissues. Different amounts (either 10 or 3 µg, as indicated) of the same poly (A) RNAs samples were electrophoresed through a 1% agarose gel, transferred to a nylon membrane and hybridized to a GRK4 (left part) or a GRK6 (right part) probe. Final washes were performed in 0.1 x SSC and 0.05% SDS at 45 C, and the membrane was exposed 20 h at -70 C with one intensifying screen for autoradiography. Positions of 28S [4700 nucleotides (nt)], 18S (1900 nt) and E. coli23S (2900 nt) and 16S (1500 nt) ribosomal RNAs, used as molecular weight markers, are shown on the right.

 
To investigate cellular localization of GRK4 mRNAs within testis, we first used in situ hybridization (Fig. 6Go). Testis of immature (10 and 25 days old) and adult rats (8 months old) were hybridized with an antisense [35S]-labeled GRK4 cRNA probe. The corresponding sense cRNA probe was used as a negative control and systematically displayed a faint background.



View larger version (171K):
[in this window]
[in a new window]
  		Figure 6. Localization of GRK4 mRNA in the testis of immature (25 days old, A–D) and adult (8 months old, E–H) rats. In situ hybridization was carried out as described in methods. Testis sections were hybridized with antisense (A, B, E, F) or sense (C, D, G, H) [35S]-labeled rat GRK4 cRNA probes. Darkfield (A, C, E, G) (x160) and brightfield (B, D, F, H) (x250) photomicrographs are shown. AS, antisense probe hybridized sections; S, sense probe hybridized sections. I, area containing spermatogonia, primary spermatocytes (from preleptotene to early pachytene) and Sertoli cell nuclei; II, area containing late pachytene and round spermatid germ cells; es, elongated spermatids; i, interstitium; m, myoid cells.

 
In testis of 10-day-old rats, a weak and homogeneous labeling was observed over the tubules (data not shown). At this stage of development, the seminiferous tubules are made of myoïd cells, Sertoli cells, spermatogonia, and a few primary spermatocytes. No signal was observed over the interstitial cells.

At day 25 after birth (see Immature in Fig. 6Go, A–D), a strong specific hybridization was seen over the inner part of the tubules. Labeling intensity clearly varied with the stage of spermatogenesis: the peaks of mRNA levels were present in the tubules containing many leptotene to late pachytene primary spermatocytes. A weak specific signal was observed over the outer part of the tubules, whereas no signal was seen over the interstitial cells. The same expression pattern was observed over the adult testis sections (Fig. 6Go, E–H). Microscopic observation of the slides revealed that stage IX-XI tubules, containing many leptotene/late pachytene primary spermatocytes, displayed the stronger labeling. The lowest signal was systematically observed over stage XIII tubule sections, where there is no leptotene/late pachytene primary spermatocytes. As shown on Fig. 6FGo, no specific signal was seen over the elongated spermatids and spermatozoa; basal myoïd cells and interstitial cells were also unlabeled.

To confirm the cellular localization of GRK4 transcripts and to quantitatively assess the expression of GRK4A and GRK4B in the testis, RT-PCR experiments were carried out on RNAs extracted from the testis and enriched cell populations. The assay was performed using known amounts of a GRK4 cRNA deletion mutant, which acted as an internal standard. Figure 7Go compares the GRK4 mRNA levels in testes from 12-day-old to 6-month-old rats vs. pachytene primary spermatocytes and round spermatids enriched cell populations harvested from adult animals. These studies demonstrate that GRK4 transcripts reach detectable amounts between day 17 and 24 after birth, and thereafter continue to increase. It is also clear that, in all testicular samples, GRK4A mRNA was the predominant variant. Quantitative analysis of GRK4A mRNA revealed a significant enrichment in pachytene cells and round spermatids: the absolute mRNA levels in these cells (molecules/ng of total RNAs) were approximately 13-, 3-, and 2-fold higher than in the testis of 24-day-, 45-day-, and 6-month-old rats, respectively.



View larger version (36K):
[in this window]
[in a new window]
  		Figure 7. Quantitative RT-PCR analysis of GRK4 mRNA levels in the testis and in enriched germ cell populations. Total RNAs (10 ng) extracted from testis of 12-day-old to 6-month-old rats, or from germ cells of 3-month-old rats, were submitted to RT-PCR using primers flanking the 93-bp sequence specific for GRK4A. Each reaction included 500 molecules of a mutant GRK4 cRNA (Mut), used as an internal standard. PCR (28 cycles) was performed in the presence of [{alpha}-32P]-dCTP. Amplified products were resolved on a 2% agarose gel and detected by autoradiography. The size of the different fragments (A, GRK4A; Mut, GRK4 deletion mutant; B, GRK4B) is indicated on the right. All the reactions were carried out in the presence (+) or the absence (-) of reverse transcriptase. Without reverse transcriptase, negative results (only shown for pachytene cells) were systematically obtained. Blank, Reaction performed without RNAs.

 
To further study the distribution of GRK4 mRNAs, we performed quantitative RT-PCR on RNAs extracted from various rat tissues (Fig. 8Go). The primers flanking the GRK4A-specific insert were used again in these experiments. Based on the preceding Northern analysis and on preliminary RT-PCR experiments, different amounts of RNAs were introduced in the assay for testis (5 ng) and for the other tissues (100 ng). This new series of experiments confirmed that GRK4 mRNAs are mainly expressed in the testis, with 10- to 100-fold higher levels than in the other tissues. It is also striking to note that the relative abundance of GRK4A and GRK4B was different in testis and in the other samples: the GRK4A fragment was predominant in testis only (90% of the total PCR signal, and 20% in the other tissues). Faint amounts of a third fragment lacking nucleotides 441–597 of the coding region and referred to as GRK4C were present in some tissues. Quantitative comparison of the GRK4B signal in the various tissues indicated the following order of expression: kidney OM>testis>adrenal>brain = lung>kidney IM~heart>kidney = ovary>liver = colon>>stomach = spleen = seminal vesicle. Thus, GRK4B mRNA was clearly heterogeneously distributed in the kidney, being enriched 20-fold in the outer medulla. On the other hand, expression of GRK4B mRNA was roughly similar in the ovarian tissue of pregnant and postparturient females.



View larger version (39K):
[in this window]
[in a new window]
  		Figure 8. Quantitative RT-PCR analysis of GRK4 mRNAs in rat tissues. Total RNAs (5 or 100 ng, as indicated) were submitted to RT-PCR using primers flanking the 93-bp sequence specific for GRK4A. Each reaction also included 500 molecules of the GRK4 cRNA deletion mutant (Mut). PCR (28 cycles) was performed in the presence of ({alpha}-32P)dCTP. The fragments were fractionated on a 3% agarose gel and detected by autoradiography. The size of the different bands, identified as GRK4A, -B, or -C by DNA cloning and sequencing is indicated on the right. Reactions were carried out in the presence (+) or the absence (-) of reverse transcriptase. For testis (last right lane) as well as for all other tissues (not shown), reactions performed without reverse transcriptase always gave negative results. Blank, Reaction performed without RNAs; Seminal V., seminal vesicle; Ov., ovary; IM, inner medulla; OM, outer medulla. Ovary samples were harvested from female rats on day 21 (21D), day 22 (22D) of gestation, or 24 h post partum (PP). All other samples were obtained from adult male rats.

 
Because the sequence fragments absent in GRKs4B, -C have not been reported to undergo alternative splicing in human, it was of interest to further compare the splicing of human and rat GRK4. In human, the four GRK4 variants are generated through alternative splicing of exons II and XV, which are located near the 5' and 3' end coding region, respectively (10). RT-PCR were carried out on several tissues using primers flanking the corresponding portions of the rat sequence.

In the 5' end coding region, a single DNA fragment was obtained in rat tissues. Its size matched that predicted for a full-length 5' end coding region, indicating that this part of the molecule does not undergo alternative splicing in rat tissues (data not shown).

In the 3' end coding region, two DNA fragments were generated by RT-PCR. As shown in Fig. 9AGo, the predominant fragment had the expected size (417 bp) for a cDNA with a full-length 3' end coding region, whereas the minor fragment was shorter. The abundance of both fragments dramatically increased with testicular development, but the ratio of the long vs. short variant increased from 0.8 in immature to 6.3 in adult rat testis. Considering that our primers encompass the sequence corresponding to human GRK4 exons XIV–XVI, and that exons XIV–XV have an identical size (138 bp), we realized that splicing of either of these exons could account for our data. Therefore, Southern blot analysis of the RT-PCR products were carried out with exon-specific probes. The hybridization pattern (Fig. 9Go, B and C) clearly demonstrated that, in the rat, an exon corresponding to human exon XIV is spliced, instead of exon XV in human GRK4. Splicing of a fragment that exactly matches human GRK4 exon XIV was confirmed by cloning and sequencing of the short variant. Similar results were obtained in rat brain, kidney, and adrenal (data not shown). The differential splicing patterns of human and rat GRK4 are summarized in Fig. 10Go.



View larger version (59K):
[in this window]
[in a new window]
  		Figure 9. Analysis of GRK4 mRNA splicing in the 3' end coding region. A, 100 ng of testis total RNAs extracted from 12-day-old (12 d) to 6-month-old (6 m) rats were submitted to RT-PCR using rat GRK4 primers corresponding to nucleotides 1381–1401 and 1776–1796. PCR (28 cycles) was performed in the presence of [{alpha}-32P]-dCTP. Amplified products were resolved on a 2% agarose gel and detected by autoradiography. B and C, The same RT-PCR was carried out in duplicate without labeled nucleotide. Amplified products were resolved on a 2% agarose gel, blotted onto HybondN+ membranes, and hybridized with 32P end-labeled oligonucleotides corresponding to either exon XIV (B) or exon XV (C) of human GRK4. Blots were washed and exposed for autoradiography. The size of the different bands is indicated on the right. Reactions performed in the absence of reverse transcriptase (not shown in the figure) gave negative results.

 


View larger version (12K):
[in this window]
[in a new window]
  		Figure 10. Differential splicing pattern of human (Hu) and rat GRK4. Exon numbering (from I to XVI) refers to the structure of the human GRK4 gene (10). In Hu GRK4{alpha}, the coding sequence (1734 bp) extends from the ATG initiation codon to the stop codon depicted in the figure and includes all 16 exons. Hu GRK4ß and -{gamma} lack exon II and XV, respectively, and GRK4{delta} lacks both of these exons. Shaded boxes indicate rat GRK4 exons for which boundaries could be deduced from the analysis of the gene structure (exons IX-XI) or from the sequence of the splice variants (exons VI, VI, and XIV). Sequences matching Hu GRK4 exons VI, VI-VII, or XIV are spliced in rat variants referred to as GRK4B, -C, and -D, respectively. Positions of the cDNA probes used for Southern blot analyses are indicated by interrupted lines.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
This work provides the molecular characterization of two rat GRK variants, which only differ for the presence or absence of a 31-amino acid sequence in the N-terminal domain. Their cognate mRNAs display different tissue distribution. The long form is mainly expressed in testis germinal cells and is much more abundant in adult than in immature animals. On the other hand, the short form is present at low levels in testis and in several other tissues, but is enriched in the kidney outer medulla.

Previous studies enabled the characterization of rat GRKs1–3, GRK5, and GRK6 (11, 13, 15, 30). The long GRK isoform described here displays striking similarities with the long human GRK4 isoform (GRK4{alpha}) but also exhibits significant differences (Fig. 2Go). The overall amino-acid identity between the two proteins is 76%, whereas identity between the rat and human GRKs formerly identified is in the range 86–98%. We therefore considered the possibility that another rat gene, much more related to the human GRK4 gene, might exist. Thus, Southern blot analyses of the rat genome were performed using a rat cDNA probe and the corresponding region of human GRK4 cDNA. These experiments demonstrated that the human and rat probes hybridize to the same genomic DNA fragments (Fig. 3Go). The coding sequence described here is therefore the rat counterpart of human GRK4. Furthermore, cloning and sequencing of the different hybridizing fragments indicated that they contain sequences 100% identical to the rat probe, consisting of three successive exons that match exons IX, X, and XI of the human GRK4 gene. This partial characterization of the rat gene reveals an organization identical to that of the human GRK4 gene. It is thus reasonable to conclude that, in both human (10) and rat, different GRK4 variants are produced from a single gene by alternative splicing.

The conserved organization of the human and rat GRK4 genes likely extends beyond the characterized rat genomic portion. In the 5' coding region, one short rat GRK4 variant lacks a 93-bp sequence portion matching human GRK4 exon VI, and the other is further deprived of residues corresponding to exon VII (Fig. 10Go). In the 3' end coding region, the additional variant found in rat tissues lacks the sequence homologous to human GRK4 exon XIV (Figs. 9Go and 10Go).

Premont et al. (10) referred to as GRK4{alpha}, ß, {gamma}, and {delta} the four different human splice variants. The long rat GRK4 variant is three amino acids shorter than human GRK4{alpha} but is nevertheless clearly related to it and could be considered as its rat counterpart. By contrast, there are no GRK4 variants homologous to human GRK4{gamma}-{delta} in rat tissues. The different rat GRK4 variants may not be therefore appropriately named GRK4{alpha}-{delta}. We have called GRK4A the long rat isoform, GRK4B the variant lacking only 93-bp in the 5' end coding region, and GRK4C the next shorter variant that displays a further 64-bp deletion. The variant lacking the 138-bp sequence that constitutes part of the 3' end coding region is referred to as GRK4D.

Rat GRK4sA and B were fully characterized through isolation of their complete coding sequence. The motif G-X-G-X-X-G (residues 193–198 and 162–167 in GRK4A and GRK4B, respectively) and the R residue at position 431 (GRK4A) or 400 (GRK4B) are conserved at the same relative positions in serine/threonine kinases (31) and delineates a centrally located domain of 239 amino acids, flanked by an N-terminal domain of variable length (161–192 residues), and a C-terminal domain of 143 amino acids. The catalytic domain contains the tripeptide sequence DLG, which is unique to the GRK family (other protein kinases have DFG). In this domain, amino-acid identity with human GRK4 reaches 88%. Beside these structural features, an essential hallmark of GRKs is their ability to preferentially phosphorylate agonist-occupied receptors (2). Our expression studies, demonstrating a light-dependent phosphorylation of rhodopsin by cytosolic extracts of GRK4A- and GRK4B-transfected cells establish that both proteins are active GRKs. The possibility that other rat GRK4 variants (i.e. GRK4C, D) also exhibit functional kinase activity was not evaluated because their complete sequence was not isolated. With regard to GRK4C, it should be stressed that in the absence of further insertion or deletion event, a stop codon at position 652–654 predicts a truncated protein (164 amino-acids) lacking the catalytic domain.

Premont et al. (10) demonstrated that human GRKs4{alpha}-{delta} are able to attenuate signaling through the LH/CG receptor in transfected HEK293 cells, indicating that all four human GRK4 variants may be involved in desensitization processes. However, expression studies carried out by Sallese et al. (12) using the same cell line documented a light-dependent phosphorylation of rhodopsin by human GRK4{alpha}, whereas the three other variants were devoid of activity. This suggests that exons II and XV of human GRK4 confer substrate specificity and/or are critical for some components of the functional assay. Recent studies demonstrated that members of the GRK4 subfamily (i.e. GRKs4–6) are stimulated by phosphatidylinositol 4,5-biphosphate (PIP2) (32) and inhibited by calmodulin (12, 33). The effects of PIP2 and calmodulin depends on their binding to N-terminal protein motifs (a gelsolin-like PIP2 binding site, and a sequence homologous to the calmodulin-binding domain of MARCKS) that are highly conserved in GRKs4–6 and are present in exon II of human GRK4. In addition, Sallese et al. (12) demonstrated that human GRK4 variants lacking either exon II or XV fail to binds calmodulin, suggesting that both of them may bear a calmodulin binding-site. Taken together, all these data indicate that the various human and rat GRK4 variants can be of valuable help to delineate structure-function relationships among GRKs. The 31-amino acid sequence absent in rat GRK4B does not seem a prerequisite for rhodopsin phosphorylation, but rhodopsin is unlikely to be regulated by GRK4 in vivo. Further studies carried out on other G protein-coupled receptors are needed to explore the functional role, if any, of this protein domain.

The N-terminal domain of GRKs was initially thought to be mainly involved in substrate recognition. Indeed, an antibody directed against residues 17–34 of rhodopsin kinase blocked phosphorylation of rhodopsin but not of a synthetic peptide substrate (34). Additional roles of the N-terminal domain may include GRK targeting to intracellular organelles. Thus, its central part (residues 88–145) was demonstrated to mediate GRK2 association to microsomal membranes through a high affinity binding site (35). Human GRK4 has also been shown to associate to intracellular structures, such as acrosomal and outer mitochondrial membranes of germinal cells (12). Interestingly, the central part of the N-terminal domain is poorly conserved in human and rat GRK4. This raises the possibility that they associate with different intracellular structures. The possible role of GRK4 on such structures is completely unknown and can only be inferred from ultrastructural data (12).

Data previously published by Premont et al. (10) and by Sallese et al. (12), together with the results presented here (Figs. 5Go and 8Go), clearly demonstrate that both human and rat GRK4 genes are mainly expressed in the testis. Immunocytochemistry and Western blot analysis carried out on human and bovine testes by Sallese et al. (12) indicated that the {gamma} isoform of human GRK4 is selectively expressed in spermatozoa and germinal cells. The lack of available data concerning GRK4 mRNA distribution and ontogenesis within the testis, and the possibility of stage specific expression for this gene, prompted us to perform in situ hybridization and RT-PCR experiments. We observed that GRK4 mRNA abundance in whole testis increases with age (Fig. 7Go). This probably reflects the growing proportion of germ cells vs. somatic cells in the testis from immature animal to adulthood. It clearly appears from the in situ hybridization studies that GRK4 transcripts are predominantly expressed in the developing germ cells (i.e. from leptotene primary spermatocytes to round spermatids), but not in differentiated spermatids and spermatozoa. We conclude from these data that the GRK4 gene is already turned on in leptotene primary spermatocytes. During spermiation, however, a dramatic chromatin condensation occurs that turns off all the genes in elongated spermatids and spermatozoa. Sallese et al. (12) observed the stronger signals for protein expression in spermatozoa. This could be explained by: 1) a delayed translation of previously accumulated GRK4 transcripts; 2) or a concentration of the GRK4 protein in spermatozoa consecutive to their maturation [i.e. the so-called residual bodies exclusion process (36)].

Our results also establish that the predominant mRNA variant is GRK4A in spermatogenic cells (Fig. 7Go), whereas GRK4B is clearly more abundant in all other tissues analyzed. This suggests different physiological functions for the various GRK4 variants, but protein expression data are clearly needed to substantiate such an hypothesis. Surprisingly, although GRK4{alpha} mRNA is presumably the predominant variant in human testis (10), polyclonal antibodies only allowed to detect GRK4{gamma} in human sperm membranes (12).

At present, nothing is known about the possible receptor substrates of GRK4. The physiological relevance of impaired signaling properties of the LH/CG receptor by human GRKs4{alpha}-{delta}, demonstrated in transfected cells (10), is questionable because this receptor and GRK4 are present in different cell types (e.g. Leydig and germinal cells, respectively). Testis germ cells express the bombesin BRS-3 receptor (37), which act through the phospholipase C pathway, and several olfactory-like receptors (38). According to Pronin et al. (33), phospholipase C-coupled receptors may not be physiologically regulated by GRKs4–6. Indeed, such receptors increase free Ca2+ levels, presumably leading to a Ca2/calmodulin-dependent inhibition of GRKs4–6 activity. Thus, these kinases would be inhibited, rather than stimulated, when Ca2+-mobilizing receptors are in the agonist-occupied form. Turning to GRK4B, the highest mRNA levels were observed in adrenal, brain, lung, heart, ovary, and kidney. Because some of these tissues display marked cellular heterogeneity, it is conceivable that high expression levels could be present in a few specialized cell types. RT-PCR carried out on different kidney fractions indicated that GRK4B mRNAs are enriched in the outer medulla. This kidney area is mainly composed of medullary thick ascending limbs (80% of the total epithelial cells), the nephron segment from which we initially isolated a GRK4 cDNA fragment. Medullary thick ascending limb cells express several adenylyl cyclase stimulating [vasopressin V2, glucagon, and calcitonin CT1A receptors (23, 39)] or inhibiting receptors [Prostaglandin EP3, and arachidonic acid receptors (40, 41)]. Examination of the effects of the different GRK4 variants on these receptors and on testis-specific receptors is now clearly warranted to disclose the natural substrates of these enzymes and ultimately correlate the molecular diversity of GRK4 with precise physiological functions.


   Acknowledgments
 
We are greatly indebted to Dr. Nelly Bennett (CEA, Grenoble) for providing rod outer segments, to Dr. Antonio De Blasi for the human GRK4{alpha} cDNA, and to Mrs. Anne-Christine Bellanger for technical assistance. We also wish to thank Ms. Lola Pellanda for help in sampling rat tissues.


   Footnotes
 
1 The nucleotide sequences reported in this paper have been submitted to GenBank (accession number: X97568). Back

2 These two authors contributed equally to the present work. Back

3 Recipient of a grant from the Fondation pour la Recherche Médicale. Back

4 Supported by a doctoral fellowship from the Ministère de la Recherche et de l’Education. Back

Received March 5, 1998.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Dohlman HG, Thorner J, Caron MG, Lefkowitz RJ 1991 Model systems for the study of seven-transmembrane-segment receptors. Annu Rev Biochem 60:653–688[CrossRef][Medline]
  2. Premont RT, Inglese J, Lefkowitz RJ 1995 Protein kinases that phosphorylate activated G protein-coupled receptors. FASEB J 9:175–182[Abstract]
  3. Benovic JL, de Blasi A, Stone WC, Caron MG, Lefkowitz RJ 1989 ß-adrenergic receptor kinase: primary structure delineates a multigene family. Science 246:235–240[Abstract/Free Full Text]
  4. Benovic JL, Onorato JJ, Arriza JL, Stone WC, Lohse M, Jenkins NA, Gilbert DJ, Copeland NG, Caron MG, Lefkowitz RJ 1991 Cloning, expression, and chromosomal localization of ß-adrenergic receptor kinase 2. A new member of the receptor kinase family. J Biol Chem 266:14939–14946[Abstract/Free Full Text]
  5. Lorenz W, Inglese J, Palczewski K, Onorato JJ, Caron MG, Lefkowitz RJ 1991 The receptor kinase family: primary structure of rhodopsin kinase reveals similarities to the ß-adrenergic receptor kinase. Proc Natl Acad Sci USA 88:8715–8719[Abstract/Free Full Text]
  6. Ambrose C, James M, Barnes G, Lin C, Bates G, Altherr M, Duyao M, Groot N, Church D, Wasmuth JJ, Lehrach H, Housman D, Buckler A, Gusella JF, MacDonald ME 1992 A novel G protein-coupled receptor kinase gene cloned from 4p16.3. Hum Mol Genet 1:697–703[Abstract/Free Full Text]
  7. Kunapuli P, Benovic JL 1993 Cloning and expression of GRK5: a member of the G protein-coupled receptor kinase family. Proc Natl Acad Sci USA 90:5588–5592[Abstract/Free Full Text]
  8. Benovic JL, Gomez J 1993 Molecular cloning and expression of GRK6. A new member of the G protein-coupled receptor kinase family. J Biol Chem 268:19521–19527[Abstract/Free Full Text]
  9. Sallese M, Lombardi MS, de Blasi A 1994 Two isoforms of G protein-coupled receptor kinase 4 identified by molecular cloning. Biochem Biophys Res Commun 199:848–854[CrossRef][Medline]
  10. Premont RT, Macrae AD, Stoffel RH, Chung N, Pitcher JA, Ambrose C, Inglese J, MacDonald ME, Lefkowitz RJ 1996 Characterization of the G protein-coupled receptor kinase GRK4. Identification of four splice variants. J Biol Chem 271:6403–6410[Abstract/Free Full Text]
  11. Firsov D, Elalouf JM 1997 Molecular cloning of two rat GRK6 splice variants. Am J Physiol 273:C953–C961
  12. Sallese M, Mariggio S, Collodel G, Moretti E, Piomboni P, Baccetti B, De Blasi A 1997 G protein-coupled receptor kinase GRK4: molecular analysis of the four isoforms and ultrastructural localization in spermatozoa and germinal cells. J Biol Chem 272:10188–10195[Abstract/Free Full Text]
  13. Arriza JL, Dawson TM, Simerly RB, Martin LJ, Caron MG, Snyder SH, Lefkowitz RJ 1992 The G-protein-coupled receptor kinases ßARK1 and ßARK2 are widely distributed at synapses in rat brain. J Neurosci 12:4045–4055[Abstract]
  14. Premont RT, Koch WJ, Inglese J, Lefkowitz RJ 1994 Identification, purification, and characterization of GRK5, a member of the family of G protein-coupled receptor kinases. J Biol Chem 269:6832–6841[Abstract/Free Full Text]
  15. Nagayama Y, Tanaka K, Hara T, Namba H, Yamashita S, Taniyama K, Niwa M 1996 Involvement of G protein-coupled receptor kinase 5 in homologous desensitization of the thyrotropin receptor. J Biol Chem 271:10143–10148[Abstract/Free Full Text]
  16. Weiss M, Vigier D, Hue D, Perrard-Sapori, Marret C, Avallet C, Durand P 1997 Pre- and postmeiotic expression of male germ cell specific genes throughout 2-week cocultures of rat germinal and Sertoli cells. Biol Reprod 57:68–76[Abstract]
  17. Chomczynski P, Sacchi N 1987 Single-step method of RNA isolation by acid guanidinium thiocyanate-phenol-chloroform extraction. Anal Biochem 162:156–159[Medline]
  18. Elalouf JM, Buhler JM, Tessiot C, Bellanger AC, Dublineau I, de Rouffignac C 1993 Predominant expression of ß1-adrenergic receptor in the thick ascending limb of rat kidney. Absolute mRNA quantitation by reverse transcription and polymerase chain reaction. J Clin Invest 91:264–272
  19. Vandewalle A 1986 Heterogeneity of uridine incorporation along the rabbit nephron. II. Effect of DOCA. Am J Physiol 246:F427–F436
  20. Bennett N, Sitaramayya A 1988 Inactivation of photoexcited rhodopsin in retinal rods: the roles of rhodopsin kinase and 48-kDa protein (arrestin). Biochemistry 27:1710–1715[CrossRef][Medline]
  21. Sibony M, Commo F, Callard P, Gasc JM 1995 Enhancement of mRNA in situ hybridization signal by microwave heating. Lab Invest 73:586–591[Medline]
  22. Ochman H, Ayala FJ, Hartl DL 1993 Use of polymerase chain reaction to amplify segments outside boundaries of known sequences. Methods Enzymol 218:309–321[Medline]
  23. Firsov D, Bellanger AC, Marsy S, Elalouf JM 1995 Quantitative RT-PCR analysis of calcitonin receptor mRNAs in the rat nephron. Am J Physiol 269:F702–F709
  24. Johnston RF, Pickett SC, Barker DL 1990 Autoradiography using storage phosphor technology. Electrophoresis 11:355–360[CrossRef][Medline]
  25. Benovic JL, Stone WC, Huebner K, Croce C, Caron MG, Lefkowitz RJ 1991 cDNA cloning and chromosomal localization of the human ß-adrenergic receptor kinase. FEBS Lett 283:122–126[CrossRef][Medline]
  26. Hughes RJ, Anderson KL, Kiel D, Insel PA 1996 Cloning of GRK2 cDNA from S49 murine lymphoma cells. Am J Physiol 270:C885–C891
  27. Parruti G, Ambrosini G, Sallese M, De Blasi A 1993 Molecular cloning, functional expression and mRNA analysis of human ß-adrenergic receptor kinase 2. Biochem Biophys Res Commun 190:475–481[CrossRef][Medline]
  28. Kozak M 1987 An analysis of 5'-non coding sequences from 699 vertebrate messenger RNAs. Nucleic Acids Res 15:8125–8148[Abstract/Free Full Text]
  29. Ohshima Y, Gotoh Y 1987 Signals for the selection of a splice site in pre-mRNA. Computer analysis of splice junction sequences and like sequences. J Mol Biol 195:247–259[CrossRef][Medline]
  30. Zhao X, Haeseleer F, Fariss RN, Huang J, Baehr W, Milam AH, Palczewski K 1997 Molecular cloning and localization of rhodopsin kinase in the mammalian pineal. Vis Neurosci 14:225–232[Medline]
  31. Hanks SK, Quinn AM 1991 Protein kinase catalytic domain sequence database: identification of conserved features of primary stucture and classification of family members. Methods Enzymol 200:38–62[Medline]
  32. Pitcher JA, Fredericks ZL, Stone WC, Premont RT, Stoffel RH, Koch WJ, Lefkowitz RJ 1996 Phosphatidylinositol 4,5-biphosphate (PIP2)-enhanced G protein-coupled receptor kinase (GRK) activity. Location, structure, and regulation of the PIP2 binding site distinguishes the GRK subfamilies. J Biol Chem 271:24907–24913[Abstract/Free Full Text]
  33. Pronin AN, Satpaev DK, Slepak VZ, Benovic JL 1997 Regulation of G protein-coupled receptor kinases by calmodulin and localization of the calmodulin binding domain. J Biol Chem 272:18273–18280[Abstract/Free Full Text]
  34. Palczewski K, Buczylko J, Lebioda L, Crabb JW, Polans AS 1993 Identification of the N-terminal region in rhodopsin kinase involved in its interaction with rhodopsin. J Biol Chem 268:6004–6013[Abstract/Free Full Text]
  35. Murga C, Ruiz-Gomez A, Garcia-Higuera I, Kim CM, Benovic JL, Mayor F 1996 High affinity binding of ß-adrenergic receptor kinase to microsomal membranes. Modulation of the activity of bound kinase by heterotrimeric G protein activation. J Biol Chem 271:985–994[Abstract/Free Full Text]
  36. Clermont Y, Perey B 1957 Quantitative study of the cell population of the seminiferous tubules in immature rats. Am J Anat 100:241–267[CrossRef][Medline]
  37. Fathi Z, Corjay MH, Shapira H, Wada E, Benya R, Jensen R, Viallet J, Sausville EA, Battey JF 1993 BRS-3: a novel bombesin receptor subtype selectively expressed in testis and lung carcinoma cells. J Biol Chem 268:5979–5984[Abstract/Free&nb