Alternative splicing of the G protein-coupled receptor superfamily in human airway smooth muscle diversifies the complement of receptors

Richard Einstein, Heather Jordan, Weiyin Zhou, Michael Brenner, Esther G. Moses, and Stephen B. Liggett
PNAS April 1, 2008 105 (13) 5230-5235; published ahead of print March 24, 2008 https://doi.org/10.1073/pnas.0801319105

  1. Communicated by Rita R. Colwell, University of Maryland, College Park, MD, February 8, 2008 (received for review October 27, 2007)

Abstract

G protein-coupled receptors (GPCRs) are the largest signaling family in the genome, serve an expansive array of functions, and are targets for ≈50% of current therapeutics. In many tissues, such as airway smooth muscle (ASM), complex, unexpected, or paradoxical responses to agonists/antagonists occur without known mechanisms. We hypothesized that ASM express many more GPCRs than predicted, and that these undergo substantial alternative splicing, creating a highly diversified receptor milieu. Transcript arrays were designed detecting 434 GPCRs and their predicted splice variants. In this cell type, 353 GPCRs were detected (including 111 orphans), with expression levels varying by ≈900-fold. Receptors used for treating airway disease were expressed lower than others with similar signaling properties, indicating potentially more effective targets. A disproportionate number of Class-A peptide-group receptors, and those coupling to Gq/11 or Gs (vs. Gi), was found. Importantly, 192 GPCRs had, on average, five different expressed receptor isoforms because of splicing events, including alternative splice donors and acceptors, novel introns, intron retentions, exon(s) skips, and novel exons, with the latter two events being most prevalent. The consequences of splicing were further investigated with the leukotriene B4 receptor, known for its aberrant responsiveness in lung. We found transcript expression of three variants because of alternative donor and acceptor splice sites, representing in-frame deletions of 38 and 100 aa, with protein expression of all three isoforms. Thus, alternative splicing, subject to conditional, temporal, and cell-type regulation, is a major mechanism that diversifies the GPCR superfamily, creating local recepteromes with specialized environments.

G protein-coupled receptors (GPCRs) represent the largest superfamily of receptors in the human genome. They are expressed on every cell and carry out an extensive array of biologic signaling. These include receptors for vision and smell, neurotransmission, and thousands of endocrine, autocrine and paracrine functions throughout the body. Critical GPCR circuits have been identified in central nervous system, cardiovascular, gastrointestinal, pulmonary, endocrine, and renal disease (reviewed in ref. 1), resulting in enhanced understanding of the pathologic basis of disease, new diagnostics, and new drug targets. Indeed, it has been estimated that ≈50% of pharmaceuticals are directed at GPCRs or their immediate downstream signaling events (2, 3). There are several well recognized phenomena in GPCR signaling events, at both the in vitro and clinical levels, which have revealed a complexity beyond that initially expected. One issue has been the recognition that one receptor can regulate another through any number of heterologous mechanisms (4, 5). This “cross-talk” appears to be necessary for the cell to integrate the large number of GPCR signaling events underway at any given time (6, 7). Thus, the complement of receptors expressed on a cell, even those that do not appear to be directly involved with a pathway, can be a determinant of the cellular response to a given GPCR. Extensive heterogeneity in GPCR signaling has also been observed between individuals that is not readily resolved by physiologic setting or disease state. This interindividual variability in GPCR function, which may be responsible for disease risk or modification and altered response to therapy, has led to the notion of heterogeneity of either structure or expression of GPCRs within the human population. And indeed, quantitative measurements of GPCR expression and/or function in otherwise matched individuals typically reveal as much as 10-fold variability (8). Similarly, common polymorphisms of GPCRs with functional implications have been identified, which appear to associate with some, but not all, of this variability (9). Another mechanism that introduces diversity of expression or function is alternative mRNA splicing, giving rise to several “isoforms” of a receptor. The identification and characterization of alternatively spliced genes within the group of known and orphan GPCRs have not been delineated, so the extent of diversity imposed by this mechanism within this important superfamily of signaling modules is unknown.

In the airways, GPCRs expressed on smooth muscle regulate airway contraction and relaxation and have been extensively studied in regard to asthma pathogenesis and treatment (10). Local increases in endogenous agonists acting at histamine, leukotriene, prostaglandin, muscarinic, thromboxane, and other GPCRs result in markedly enhanced bronchial hyperresponsiveness to contraction, leading to airway constriction. Other receptors, such as the β2-adrenergic receptor (β2AR), appear to oppose constriction or evoke relaxation (10), and thus airway smooth muscle (ASM) GPCRs have represented important targets for therapeutic agents, both agonists and antagonists, in asthma treatment. Nevertheless, a large body of pharmacologic data reveals complex, contradictory, or paradoxical responses of ASM GPCRs, bringing into question the role of certain receptors or pathways in smooth muscle pathophysiology and the potential utility of novel drugs targeting members of this superfamily. We hypothesized that ASM expresses many more GPCRs than predicted, and that these undergo frequent alternative splicing, resulting in unexpected receptor isoforms and thus creating a highly diversified receptor milieu.

Results

Complement of Expressed GPCRs from Human ASM.

To identify expression of the GPCR superfamily in ASM cells, RNA was isolated and analyzed by using expression microarrays that provide for resolution of transcripts produced from the same locus by alternative splicing. The overall expression level per GPCR can be determined by using probes monitoring the classic “nonspliced” form. The arrays used here contain probes that monitor a total of 434 GPCR genes. The expression results from cultured ASM cells derived from five individuals are provided in supporting information (SI) Dataset S1. A total of 353 GPCR genes had statistically significant expression levels, defined as a 2-fold signal over background, and all such signals had Bonferroni-corrected P values of <10−4. Of interest, because they are targets for current pharmacologic therapy in obstructive lung disease, are the M3-muscarinic receptor (CHRM3) and β2-adrenergic receptor (β2AR, ADRB2). Knowledge of expression of these “benchmark” therapeutic targets places into perspective expression results from other GPCRs, because they may relate to function or suitability as new drug targets. Antagonists to the former receptor are potent blockers of bronchoconstriction, whereas agonists acting at the β2AR are the most efficacious bronchodilators known. Yet, both of these were among the lower-expressing GPCRs in ASM (≈100-fold lower than the highest expressors), suggesting an unrecognized complement of receptors in this cell type that may be involved in pathogenesis of airway obstruction or may be new therapeutic targets. Among all significantly expressed GPCRs, expression varied by as much as ≈900-fold. Of the 353 receptors that were expressed, 111 were orphan GPCRs. Receptors were stratified by class (11) and by the major G proteins to which they couple based on published studies or by modeling techniques. As expected, the expressed GPCRs were dominated by members of Class A, which accounted for 72% of the nonorphan receptors. Within this class, the group of peptide receptors was most prevalent (60%), followed by the amine receptors (22%). The remainder of the Class A receptors were primarily of the hormone, prostanoid, and nucleotide families. The majority of non-Class A receptors were Class B secretin-like receptors. Examples of the heterogeneity of GPCRs expressed in human ASM and the relationship between G protein-coupling and expression level are shown in Fig. 1. Gq/11- and Gs-coupled receptors were the most prevalent in human ASM (Fig. 1). This appears to be independent of the level of receptor expression, because the number of detected receptors which couple to Gq/11 and Gs is similar in all expression quartiles, and greater than the number of detected Gi/o-coupled receptors.

Fig. 1.
Fig. 1.

Distribution of Gs-, Gq/11-, and Gi/Go-coupled GPCRs stratified by relative expression in human ASM cells. Data represent mean expression of the unspliced (“wild-type”) transcripts from cells derived from five individuals stratified by quartiles. Orphan receptor coupling was predicted by an algorithm as described in Methods. The diversity of GPCR expression is further illustrated by representative genes expressed in each quartile assembled by GPCR Class (A, rhodopsin-like; B, secretin receptor-like) and group (amine, peptide, or other). Gene names in black indicate disputed or dual G protein coupling. See Dataset S1 for expression results for specific GPCR genes.

Diversity of GPCR Expression Due to Alternative Splicing in Human ASM.

The number of GPCRs with potential splice events that were identified in silico amounted to 252 of the 434 GPCR genes, representing a total of 1,500 events. Probe sets on the array were designed to detect each specific splicing event using the strategy depicted in Fig. S1. Alternative splicing events of GPCR genes detected in human ASM are shown in Fig. 2 classified by event type. The data from individual GPCR splicing events from these cells are provided by sample in Dataset S2. A specific variant can be further assessed as to event-type, location within the gene of the splice event(s), and splice-site scores from the frequency matrix (12) in Dataset S3. Of the 252 potentially alternatively spliced GPCR genes, 192 had one (or more) alternative isoforms, with a total of 967 events detected. Exons and exon(s) skip events were the most common in ASM (Fig. 2). Of note, these two categories should be considered similar, because the difference lies in which sequence has been selected as the reference (Fig. S1). Of the ≈600 potential novel exons, ≈50% were expressed in ASM, which represented the smallest percentage of detected vs. probed events. A somewhat higher percentage was found for exon skip, alternative splice donor, and alternative splice acceptor events, and particularly so for the less-common novel intron and intron retention events. There was no apparent relationship between “wild-type” expression levels and the occurrence of a particular splicing event. That is, these events occurred regardless of the degree of overall expression of a given GPCR gene.

Fig. 2.
Fig. 2.

Distribution of GPCR splice variants detected in human ASM cells. The number of splice events detected in human ASM cells is shown by event type. The six types of alternative splicing events considered are defined in Methods and summarized in Fig. S1. See Dataset S2 for results from specific genes and Dataset S3 for detailed splice-event annotation.

Three Expressed Forms of the Leukotriene B4 Receptor Due to Alternative Splicing.

A select group of 30 GPCRs was chosen for further study based on high, intermediate, and low levels of expression, and their known relevance to ASM signal transduction (Table 1). Overall expression levels within this group of genes (taking into account all variants) differed by up to ≈400-fold. Of the 30 genes selected for further analysis, 13 had no evidence of alternative splicing, whereas the remaining 17 genes contained 70 splicing events. Of these, 39 events were identified where novel exons or exon extension events were present and led to additional sequence in the alternatively spliced transcripts. Thirty-eight of these events were found to be significantly expressed above background, whereas one event was not statistically significant. In contrast, 31 events were found to consist of exon skip events or exon truncations, where splicing activity led to a loss of sequence. Twenty-four of these events were found to be significantly above background, whereas seven were not found to be expressed. From this set of expressed events, 17 were selected for validation by RT-PCR. The final list represented a variety of splice events (alternative donor and acceptor sites, novel introns, and novel and skipped exons), with multiple categories of supporting evidence for the event (sequences from mRNAs, ESTs, and RefSeqs). Primers were specifically designed for capturing spliced isoforms or were specifically targeted to amplify only the longer isoform (Fig. S2 and Table S1). Data from the high-affinity leukotriene B4 receptor (13) (LTB4R, also denoted LTB4R1 and BLT1, GenBank accession no. NM_181657) showed several potential splicing events (Table 2, Fig. 3) and was further explored to elucidate the ramifications of this diversification mechanism with a relevant signaling pathway in ASM. LTB4R has four splice events that were monitored on the array, and two were unequivocal, as determined from RT-PCR followed by sequencing. The two validated events, the alternative donor and acceptor sites, are illustrated in Fig. S2. Both events were identified as part of the same transcript, so primers were designed to amplify the variant and the reference form of the transcript (primers are indicated in Fig. 3; F19 and R19 amplify both the reference and the variant, whereas F20 and R20 amplify only the reference). Amplicons were pooled, and bands were purified, cloned, and sequenced to verify the sequence structure for each variant. Slightly different variants were found in the 5′UTR but generated sequences that had exactly the same coding region as the reference sequence. Two additional variants contained a deletion of part of the 5′UTR and 5′-coding sequence (Fig. 3); the junction sequences for both of these variants were identical (Fig. 4A). This resulted in a translated protein lacking the first 100 aa (LTB4R-AS2 in Fig. 4B) compared with the reference protein. Another variant was found (Fig. 3, clones A03 and A06), with alterations in the coding region distinct from the previous clones. The distinct junction for this clone was identified and revealed the loss of a portion of the second exon. This loss from alternative splicing is in-frame and results in the exclusion of amino acids 43–81 of the protein (LTB4R-AS1, Fig. 4B).

View this table:
Table 1.

Expression levels of selected GPCR transcripts from airway smooth muscle

Fig. 3.
Fig. 3.

Genomic organization and identification of potential splice variants in the LTB4R gene. A schematic diagram for LTB4R is illustrated approximately to scale to indicate the exons and introns for the reference LTB4R gene and three splice variants that were detected. The location of RT-PCR primers is indicated and was used to isolate amplicons that verified the sequences for each variant. The sequences were aligned to the genome and UTR and coding regions are indicated. LTB4R-AS1 has lost 39 aa internal of the coding region, whereas LTB4R-AS2 lost the initial 100 N-terminal aa.

Fig. 4.
Fig. 4.

Nucleotide and translated sequence of LTB4R and the two splice variants. In A, sequence electropherograms from the RT-PCR products representing the indicated receptors are shown, with the point of deviation from wild-type LTB4R indicated by the arrows. In B, translated sequences are shown. The full-length sequence represents wild-type LTB4R, the bracketed red sequence represents amino acids absent in the LTB4R-AS1, and the entire red sequences represent the amino acids absent in LTB4R-AS2. *, +, and # indicate the specific splice sites in the DNA (A) or amino acid (B) sequence.

Transmembrane modeling (Fig. 5 A–C) reveals that LTB4R-AS1 lacks the second transmembrane domain and part of the 5′ portion of the first extracellular loop. LTB4R-AS2 lacks the N terminus, first transmembrane domain, first intracellular loop, second transmembrane domain, first extracellular loop, and ≈8 aa from the 5′ portion of the third transmembrane domain. To ascertain whether the LTB4R-AS1 and -AS2 receptor proteins, which lack 39 and 100 aa, are expressed on human ASM cells, Western blots were carried out with a polyclonal antibody directed to a peptide representing the third intracellular loop of the human LTB4R. As shown in Fig. 5D, the full-length receptor was identified at its predicted molecular mass of ≈38 kDa. Two other forms were also identified at the predicted molecular mass of the two shorter forms at ≈33 kDa (LTB4R-AS1) and ≈27 kDa (LTB4R-AS2). The wild-type and -AS2 forms were expressed at approximately equivalent levels, whereas the -AS1 form was ≈10-fold lower in expression compared with either of the other two.

Fig. 5.
Fig. 5.

Predicted topology of LTB4R and the two splice variants. Transmembrane-spanning domains and the extracellular and intracellular loops are depicted, with the x axis representing the amino acid number (A, LTB4R; B, LTB4R-AS1; C, LTB4R-AS2). The probabilities (P) were calculated by using the program as described in Methods. Representative Western blot (D) shows the three isoforms expressed in ASM at the predicted molecular mass of the receptors at 37.5 (full length), 33.2 (LTB4R-AS1), and 26.9 kDa (LTB4R-AS2). Results are from cells derived from two individuals.

Discussion

In the human genome, typical pre-mRNAs consist (on average) of 8 introns and 9 exons, spanning ≈27,000 nt (14). RNA splicing at intron–exon boundaries provides for deletion of introns, resulting in mature exon sequence. Such splicing, performed by ribonucleoprotein complexes (spliceosomes), is subject to conditional, temporal, and cell-type-specific regulation, resulting in alternatively spliced transcripts and thus various isoforms of the expressed protein (reviewed in refs. 15⇓–17). Alternative splicing is one mechanism by which the number of expressed proteins is thought to exceed the number of genes in the human genome. Thus, considerable structural diversity of expressed proteins can be realized from alternative splicing of a single gene. In the current work, we identified and characterized the splice variants of the GPCR superfamily because (i) it represents the largest family of signaling proteins in the genome, (ii) these receptors play a significant role in normal homeostasis and pathophysiology, and (iii) extensive efforts are underway to design agonists and antagonists for pharmacologic therapy. Genes encoding the majority of known and orphan GPCRs (not including those for smell or sight) were considered, and the cell-type interrogated was human ASM cells. This cell type was chosen because of its central role in the pathogenesis of airway obstruction in asthma and the known actions of GPCRs in ASM contraction, relaxation, and proliferation.

Of the 434 GPCRs monitored on the splice variant-specific array, 353 were detected over a wide range of expression levels in this cell type. This number of GPCRs is much greater than the ≈50 that were estimated to be expressed on human ASM based on evidence from pharmacologic and limited expression studies (10). The several hundred GPCRs with relatively low expression levels should not be considered biologically insignificant, because this group includes the M3-muscarinic receptor, the β2AR, and receptors for leukotrienes and prostanoids, all of which are important in the pathogenesis of obstructive lung disease or receptor-targeted therapy. Surveying the wild-type isoform expression levels of the Gq-coupled receptors, which mediate bronchoconstriction, we noted expression of ≈150 receptors, extensive variation in levels of expression, and diversification via splice variants. These findings indicate potential new targets for antagonists and potential complexities because of multiple isoforms. For example, the CysLT1 receptor is currently a target for antagonists such as montelukast in the treatment of asthma. However, the orphan receptor GPCR 18, also predicted to be Gq-coupled, was shown here to be expressed ≈4-fold higher than CysLT1 in ASM (Dataset S1) and may be an efficacious target for therapy, assuming that local levels of its endogenous ligand [N-arachidonylglycine (18)] are increased in asthma. Similarly, the β2AR, a Gs-coupled receptor which is the target for β-agonists used to dilate the airways during bronchospasm, is expressed ≈25-fold lower than the orphan receptor GPCR 101, which is also predicted to be Gs coupled. Thus, agonists to GPCR 101 could be more effective bronchodilators than β-agonists, or these agents could be coadministered to achieve maximal bronchodilation. Interestingly, GPCR 101 does not have the two consensus sequences for protein kinase A phosphorylation found in the β2AR, which contributes to agonist-promoted desensitization of this receptor (19). Thus, agonists for GPCR 101 may show less tachyphylaxis than β-agonists. Of the 353 GPCRs expressed on ASM, nearly two-thirds had in silico evidence for the potential to be alternatively spliced. Of these, 192 were found to express more than one isoform in this cell type, with a total of 967 splicing events detected in the superfamily. Of the spliced GPCRs, then, an average of approximately five different isoforms per receptor were expressed in ASM. This represents substantial diversity in a cell type where new pharmacologic agents are needed to control bronchospasm in asthma and chronic obstructive lung disease.

We chose the LTB4R for additional investigation because of its role in asthma and to emphasize the potential impact of splice variants on pathophysiology and treatment strategies. LTB4, like a number of other arachidonic acid metabolites, is increased in inflammatory states such as asthma, with elevated local levels in the airway (20). The high-affinity receptor for LTB4, LTB4R, is well recognized as being expressed on neutrophils and other leukocytes (eosinophils, T-lymphocytes) and, when activated by its endogenous agonist, promotes leukocyte migration to the lung and participates in activation and degranulation (21). To our knowledge, LTB4R expression on human ASM cells has not been previously identified. In recombinant expression systems, the receptor has been shown to couple to both Gαi and Gαq (22). Given that Gαq-coupled receptors act to constrict ASM, this suggests that LTB4 could evoke bronchoconstriction directly, rather than via leukocyte recruitment. In guinea pig trachea, LTB4 alone has been reported to have no effect on isometric contraction, but significantly enhances acetylcholine promoted contraction mediated by the Gq-coupled M3-muscarinic receptor (23). Other reports, though, have shown significant contraction of lung parenchymal strips by nanomolar concentrations of LTB4 (24). Some studies have concluded that LTB4 antagonists have minor effects on leukocyte accumulation but decrease bronchial hyperreactivity (25, 26), another showed decreased leukocyte recruitment but no effect on pulmonary function (27), and another found that antigen-induced bronchoconstriction was decreased but eosinophil accumulation and acute bronchial hyperreactivity were unaffected (28).

The discrepant bronchoconstriction effects from LTB4R antagonists may be due to the presence of splice variants of the receptor in ASM. We show here three forms of the transcribed LTB4R and two additional variations in the 5′UTR, which may alter the rate of transcription. The two splice variants involving the coding region lack domains that are likely necessary for LTB4 binding to the receptor. However, both have intact G protein coupling domains (second and third intracellular loops) and thus may act as dominant-negative proteins during wild-type LTB4R signaling. Of note, traditional transcript-detection methods would have failed to identify the two alternatively spliced LTB4R transcripts, giving a false impression of the level of expression of the full-length receptor. Given that LTB4R-AS2 protein is expressed at approximately the same level as wild-type LTB4R, functional LTB4R expression is at least ≈1/2 that expected. And, if the splice variants act as dominant negatives, functional expression could be substantially lower or perhaps trivial. Development of LTB4R antagonists directed at the smooth muscle receptor would thus appear to have a low probability of therapeutic efficacy.

Taken together, the current results reveal an unexpectedly large number of receptors in the GPCR superfamily expressed in human ASM and extensive variability in structure because of alternative splicing leading to many receptor isoforms. The utility of these findings was explored in the LTB4R gene, where splicing events resulted in two previously unrecognized isoforms, which may explain discrepant results with antagonists for treating asthma. In regard to the superfamily, alternative splice donors and acceptors, skipped exons, novel exons, and novel and retained introns were all observed, leading to marked expansion of this signaling modality. Given that GPCRs are expressed on every cell type in the body and play prominent roles in signaling events in many diseases, consideration of this previously unrecognized diversity is necessary for understanding GPCR-related pathophysiology and therapeutic targets.

Methods

Bioinformatic Identification of Potential GPCR Splice Variants.

An initial list of 434 GPCR gene representatives was determined by processing a set of nonviral and nonolfactory GPCR sequences obtained from public databases to a nonredundant set through genomic position (by alignment) clustering. These 434 representatives were compared with the exon structure of overlapping alignments at the same locus of the human genome in silico.

Sequences were aligned to the human genome (build 33) using BLAT (29). RefSeq sequences met the criteria of 97% alignment identity and a minimum BLAT score of 200; in addition, EST alignments also had to include at least one splice junction, thereby having more than one exon, and >70% of the total sequence had to be represented. Orientation of the EST alignment and sequence was determined by comparison of consensus splice site scores of all splice junctions for the sequence performed in both sequence orientations. Overlapping sequences were compared by using genomic mapping coordinates, and differences in sequence content were determined through mathematical differences in the derived coordinates. Each event was given a unique identification number. For example, 106.008.001 indicates an event found in the 106th locus as found in the original gene list; the alignment was the eighth variant sequence checked in the locus, and this event was the first (001) event found in that variant sequence. Probe names consist of the event number with an underscore followed by a letter to indicate whether the probe is a common probe (A5 or A3), a long-form exon probe (B), a long-form junction probe (C or D), or a short-form junction probe (E) (Fig. S1). Splice-site scores were calculated by using a frequency matrix from the AltExtron Database (www.ebi.ac.uk/asd/altextron/data/splice_site_dists.html) (12).

Probe Design and Microarrays for Splice Variant Identification.

A set of probes was synthesized for each event identified bioinformatically and was designed to obtain a common thermodynamic profile as described (30). Probes directed against exon–exon junctions were limited to 24 bases to limit cross-reactivity from hybridization of half the junction probe to a single exon (30). The microarrays were manufactured by Agilent Technologies with each probe printed in duplicate and randomly distributed on the microarray slide. Primary human ASM (bronchial) cells derived from five individuals were obtained from Clonetics and were maintained in monolayers in SMBM media (Clonetics). Cells at passages 3 or 4 were detached and total RNA prepared. (We are unaware of data suggesting that short-term culture alters the fundamental processes of splicing, but overall expression levels of genes, including GPCRs, could be affected by culture conditions.) Five micrograms of total RNA from each sample was labeled by using a random primed labeling protocol described by Xiang (31) with the following modifications: the final concentrations of dTTP and aadUTP were 250 μM, instead of 300 and 200 μM, respectively; the first-strand synthesis was not stopped with EDTA but directly hydrolyzed for 10 min at 65°C; after neutralization, the labeled material was purified by using the Rapid PCR purification system (Marligen) and eluted twice with 50 μl of each of RNase-free water preheated to 65°C. For hybridization, sets of dye-swap experiments were performed with each sample using 2.5 μg each of Cy3- and Cy5-labeled cDNA along with 40 μg of herring sperm DNA (Invitrogen). The mixture was heated at 98°C for 3 min and then brought to room temperature for 5 min in the presence of 25 μl of 10× Control Targets from an in situ hybridization kit (Agilent Technologies). After addition of the 2× hybridization buffer, the mixture was hybridized to the GPCR arrays at 65°C for 17 h. The arrays were washed by using a 60-mer oligo microarray processing protocol (Agilent), with washes of 5 min each. The arrays were scanned by using an Agilent Microarray Scanner. The images were analyzed with the Feature Extraction software, version 8.1 (Agilent Technologies), using the default settings. Dataset S2 contains probe level information from all 10 hybridizations (five samples processed in duplicate). Intensity dye-normalized values and processed background values are given for each probe (see Data Analysis and Modeling for details).

Data Analysis and Modeling.

The mean signal and background values from the output files were used to produce datasets that were compared to determine probes significantly expressed over background. The dye normalization factor was calculated from values in the file for each probe/sample combination and multiplied by the background and mean signals to produce dye normalized values. To calculate gene level expression values, the median value from all “A” probes for each gene or locus was calculated (A1 and A2 were used for genes without splice events, A3 and A5 probes for genes with splice events, Dataset S1). The median data values were log2 transformed, and a two-way ANOVA model was performed by using signal type (signal vs. background) and dye as a factor to generate Bonferroni-corrected P values for all probes to determine probes that were expressed significantly above background. Receptors were classified by their signaling to the three major classes of G protein: Gi/o, Gs, and Gq/11, using the International Union of Pharmacology database (32) and literature searches. Predicted G protein-coupling specificity for unclassified orphan receptors was determined by a support vector machine and hidden Markov model using the program GRIFFIN (33). Transmembrane spanning predictions were made by the TMHMM Server v2.0 (www.cbs.dtu.dk).

Western Blots.

To detect expression of the LTB4R isoforms, a polyclonal antibody (Sigma) directed against a peptide representing the third intracellular loop of LTB4R was used in Western blots from whole-cell ASM lysates. Studies were performed with 40 μg of protein and a primary antibody titer of 1:1,000 using methods described (4).

Acknowledgments

This work was supported by National Institutes of Health Grants HL071609 and HL065899 (to S.B.L.).

Footnotes

  • ↵‡To whom correspondence should be addressed. E-mail: sligg001{at}umaryland.edu

  • Author contributions: R.E. and S.B.L. designed research; H.J., M.B., E.G.M., and S.B.L. performed research; R.E. and S.B.L. contributed new reagents/analytic tools; R.E., W.Z., M.B., E.G.M., and S.B.L. analyzed data; and R.E., H.J., and S.B.L. wrote the paper.

  • The authors declare no conflict of interest.

  • This article contains supporting information online at www.pnas.org/cgi/content/full/0801319105/DCSupplemental.

  • Received October 27, 2007.

Freely available online through the PNAS open access option.

References

  1. ↵
    1. Lundstrom KH,
    2. Chiu ML
    , eds (2006) G Protein-Coupled Receptors in Drug Discovery (CRC Taylor & Francis Group, Boca Raton, FL).
    .
  2. ↵
    1. Drews J
    (2000) Drug discovery: a historical perspective. Science 287:1960–1964.
    .
    OpenUrlAbstract/FREE Full Text
  3. ↵
    1. Hopkins AL,
    2. Groom CR
    (2002) The druggable genome. Nat Rev Drug Discov 1:727–730.
    .
    OpenUrlCrossRefPubMed
  4. ↵
    1. McGraw DW,
    2. Almoosa KF,
    3. Paul RJ,
    4. Kobilka BK,
    5. Liggett SB
    (2003) Antithetic regulation by β-adrenergic receptors of Gq-receptor signaling via phospholipase-C underlies the airway β-agonist paradox. J Clin Invest 112:619–626.
    .
    OpenUrlCrossRefPubMed
  5. ↵
    1. McGraw DW,
    2. Mihlbachler KA,
    3. Schwarb MR,
    4. Rahman FF,
    5. Small KM,
    6. et al.
    (2006) Airway smooth muscle prostaglandin-EP1 receptors directly modulate beta2-adrenergic receptors within a unique heterodimeric complex. J Clin Invest 116:1400–1409.
    .
    OpenUrlCrossRefPubMed
  6. ↵
    1. Liggett SB
    (2006) Cardiac 7-transmembrane-spanning domain receptor portfolios: diversify, diversify, diversify. J Clin Invest 116:875–877.
    .
    OpenUrlCrossRefPubMed
  7. ↵
    1. Liggett SB
    (2007) Long-distance affair with adrenal GRK2 hangs up heart failure. Nat Med 13:246–248.
    .
    OpenUrlCrossRefPubMed
  8. ↵
    1. Liggett SB,
    2. Shah SD,
    3. Cryer PE
    (1989) Human tissue adrenergic receptors do not predict in vivo responses to epinephrine. Am J Physiol 256:E600–E609.
    .
    OpenUrlPubMed
  9. ↵
    1. Thompson MD,
    2. Burnham WM,
    3. Cole DE
    (2005) The G protein-coupled receptors: pharmacogenetics and disease. Crit Rev Clin Lab Sci 42:311–392.
    .
    OpenUrlCrossRefPubMed
  10. ↵
    1. Green SA,
    2. Liggett SB
    (1996) in The Genetics of Asthma, eds Liggett S, Meyers D (Dekker, New York), pp 67–90.
    .
  11. ↵
    1. Fredriksson R,
    2. Schioth HB
    (2005) The repertoire of G-protein-coupled receptors in fully sequenced genomes. Mol Pharmacol 67:1414–1425.
    .
    OpenUrlAbstract/FREE Full Text
  12. ↵
    1. Clark F,
    2. Thanaraj TA
    (2002) Categorization and characterization of transcript-confirmed constitutively and alternatively spliced introns and exons from human. Hum Mol Genet 11:451–464.
    .
    OpenUrlCrossRefPubMed
  13. ↵
    1. Yokomizo T,
    2. Izumi T,
    3. Chang K,
    4. Takuwa Y,
    5. Shimizu T
    (1997) A G-protein-coupled receptor for leukotriene B4 that mediates chemotaxis. Nature 387:620–624.
    .
    OpenUrlCrossRefPubMed
  14. ↵
    1. Lander ES,
    2. Linton LM,
    3. Birren B,
    4. Nusbaum C,
    5. Zody MC,
    6. et al.
    (2001) Initial sequencing and analysis of the human genome. Nature 409:860–921.
    .
    OpenUrlCrossRefPubMed
  15. ↵
    1. Garcia-Blanco MA,
    2. Baraniak AP,
    3. Lasda EL
    (2004) Alternative splicing in disease and therapy. Nat Biotechnol 22:535–546.
    .
    OpenUrlCrossRefPubMed
  16. ↵
    1. Caceres JF,
    2. Kornblihtt AR
    (2002) Alternative splicing: multiple control mechanisms and involvement in human disease. Trends Genet 18:186–193.
    .
    OpenUrlCrossRefPubMed
  17. ↵
    1. Einstein R
    (2003) Splicing 2002: RNA splicing in human pathology. Pharmacogenomics 4:19–22.
    .
    OpenUrlCrossRefPubMed
  18. ↵
    1. Kohno M,
    2. Hasegawa H,
    3. Inoue A,
    4. Muraoka M,
    5. Miyazaki T,
    6. et al.
    (2006) Identification of N-arachidonylglycine as the endogenous ligand for orphan G-protein-coupled receptor GPR18. Biochem Biophys Res Commun 347:827–832.
    .
    OpenUrlCrossRefPubMed
  19. ↵
    1. Liggett SB,
    2. Bouvier M,
    3. Hausdorff WP,
    4. O'Dowd B,
    5. Caron MG,
    6. et al.
    (1989) Altered patterns of agonist-stimulated cAMP accumulation in cells expressing mutant β2-adrenergic receptors lacking phosporylation sites. Mol Pharmacol 36:641–646.
    .
    OpenUrlAbstract
  20. ↵
    1. Biernacki WA,
    2. Kharitonov SA,
    3. Biernacka HM,
    4. Barnes PJ
    (2005) Effect of montelukast on exhaled leukotrienes and quality of life in asthmatic patients. Chest 128:1958–1963.
    .
    OpenUrlCrossRefPubMed
  21. ↵
    1. Funk CD
    (2001) Prostaglandins and leukotrienes: advances in eicosanoid biology. Science 294:1871–1875.
    .
    OpenUrlAbstract/FREE Full Text
  22. ↵
    1. Kuniyeda K,
    2. Okuno T,
    3. Terawaki K,
    4. Miyano M,
    5. Yokomizo T,
    6. et al.
    (2007) Identification of the intracellular region of the leukotriene B4 receptor type 1 that is specifically involved in Gi activation. J Biol Chem 282:3998–4006.
    .
    OpenUrlAbstract/FREE Full Text
  23. ↵
    1. Thorpe JE,
    2. Murlas CG
    (1986) Leukotriene B4 potentiates airway muscle responsiveness in vivo and in vitro. Prostaglandins 31:899–908.
    .
    OpenUrlCrossRefPubMed
  24. ↵
    1. Jackson WT,
    2. Froelich LL,
    3. Boyd RJ,
    4. Schrementi JP,
    5. Saussy DL Jr,
    6. et al.
    (1999) Pharmacologic actions of the second-generation leukotriene B4 receptor antagonist LY293111: in vitro studies. J Pharmacol Exp Ther 288:286–294.
    .
    OpenUrlAbstract/FREE Full Text
  25. ↵
    1. Milne AA,
    2. Piper PJ
    (1995) The leukotriene B4 antagonist, U-75, 302, inhibits antigen-induced airway hyperresponsiveness but not leukocyte accumulation in sensitized guinea-pigs. Br J Pharmacol 114:50.
    .
    OpenUrl
  26. ↵
    1. Seeds EAM,
    2. Kilfeather S,
    3. Okiji S,
    4. Schoupe TS,
    5. Donigi-Gale D,
    6. et al.
    (1995) Role of lipoxygenase metabolites in platelet-activating factor- and antigen-induced bronchial hyperresponsiveness and eosinophil infiltration. Eur J Pharmacol 293:369–376.
    .
    OpenUrlCrossRefPubMed
  27. ↵
    1. Richards IM,
    2. Sun FF,
    3. Taylor BM,
    4. Shields SK,
    5. et al.
    (1991) Contribution of leukotriene B4 to airway inflammation and the effect of antagonists. Ann NY Acad Sci 629:274–287.
    .
    OpenUrlCrossRefPubMed
  28. ↵
    1. Asanuma F,
    2. Kuwabara K,
    3. Arimura A,
    4. Furue Y,
    5. Fleisch JH,
    6. et al.
    (2001) Effects of leukotriene B4 receptor antagonist, LY293111Na, on antigen-induced bronchial hyperresponsiveness and leukocyte infiltration in sensitized guinea pigs. Inflamm Res 50:136–141.
    .
    OpenUrlCrossRefPubMed
  29. ↵
    1. Kent WJ
    (2002) BLAT–the BLAST-like alignment tool. Genome Res 12:656–664.
    .
    OpenUrlAbstract/FREE Full Text
  30. ↵
    1. Fehlbaum P,
    2. Guihal C,
    3. Bracco L,
    4. Cochet O
    (2005) A microarray configuration to quantify expression levels and relative abundance of splice variants. Nucleic Acids Res 33:e47.
    .
    OpenUrlCrossRefPubMed
  31. ↵
    1. Xiang CC,
    2. Kozhich OA,
    3. Chen M,
    4. Inman JM,
    5. Phan QN,
    6. et al.
    (2002) Amine-modified random primers to label probes for DNA microarrays. Nat Biotechnol 20:738–742.
    .
    OpenUrlCrossRefPubMed
  32. ↵
    1. Alexander S,
    2. Mathie A,
    3. Peters J,
    4. MacKenzie G,
    5. Smith A
    (2001) TiPs Nomenclature Supplement 2001. Trends Pharmacol Sci 22(Supp 1):1–103.
    .
    OpenUrlPubMed
  33. ↵
    1. Yabuki Y,
    2. Muramatsu T,
    3. Hirokawa T,
    4. Mukai H,
    5. Suwa M
    (2005) GRIFFIN: a system for predicting GPCR-G-protein coupling selectivity using a support vector machine and a hidden Markov model. Nucleic Acids Res 33:W148–W153.
    .
    OpenUrlCrossRefPubMed
View Abstract
Back to top
Article Alerts
Email Article
Citation Tools
Request Permissions
Alternative splicing of the G protein-coupled receptor superfamily in human airway smooth muscle diversifies the complement of receptors
Richard Einstein, Heather Jordan, Weiyin Zhou, Michael Brenner, Esther G. Moses, Stephen B. Liggett
Proceedings of the National Academy of Sciences Apr 2008, 105 (13) 5230-5235; DOI: 10.1073/pnas.0801319105
del.icio.us logo Digg logo Reddit logo Twitter logo CiteULike logo Facebook logo Google logo Mendeley logo
Proceedings of the National Academy of Sciences: 116 (7)
Current Issue

Submit

Jump to section

You May Also be Interested in

Nonmonogamous strawberry poison frog (Oophaga pumilio). Image courtesy of Yusan Yang (University of Pittsburgh, Pittsburgh).
Putative signature of monogamy
A study suggests a putative gene-expression hallmark common to monogamous male vertebrates of some species, namely cichlid fishes, dendrobatid frogs, passeroid songbirds, common voles, and deer mice, and identifies 24 candidate genes potentially associated with monogamy.
Image courtesy of Yusan Yang (University of Pittsburgh, Pittsburgh).
Active lifestyles. Image courtesy of Pixabay/MabelAmber.
Meaningful life tied to healthy aging
Physical and social well-being in old age are linked to self-assessments of life worth, and a spectrum of behavioral, economic, health, and social variables may influence whether aging individuals believe they are leading meaningful lives.
Image courtesy of Pixabay/MabelAmber.