₂-Adrenergic Receptor Pharmacogenetics

STEPHEN B. LIGGETT

Departments of Medicine (Pulmonary) and Molecular Genetics, University of Cincinnati College of Medicine, Cincinnati, Ohio

	INTRODUCTION

TOP INTRODUCTION ESTABLISHED CONCEPTS CLINICAL STUDIES OF AGONIST... FUTURE DIRECTIONS CONCLUSIONS IMPORTANT QUESTIONS REFERENCES

₂-Adrenergic receptors (₂ARs) expressed on airway smooth muscle are the targets for -agonists used in the treatment of bronchospasm. Studies have shown that this receptor is polymorphic within the human population and that some of these polymorphic receptors have different pharmacologic properties. In this review three areas are addressed regarding these polymorphisms: established concepts, clinical studies that need confirmation, and future directions.

	ESTABLISHED CONCEPTS

TOP INTRODUCTION ESTABLISHED CONCEPTS CLINICAL STUDIES OF AGONIST... FUTURE DIRECTIONS CONCLUSIONS IMPORTANT QUESTIONS REFERENCES

Genetics

The human ₂AR is encoded by an intronless gene, with a coding block consisting of 1,239 nucleic acids. Within this region there are nine known loci that vary within the normal population (1). The nucleic acid changes within the context of the codons are shown in Figure 1, which depicts the amino acid sequence and proposed membrane topography of the ₂AR. As might be predicted, the single nucleotide polymorphisms (SNPs) are randomly located throughout the sequence. As shown, five of these do not result in a change in the encoded amino acid. However, four polymorphisms result in variation in the encoded residues at amino acids 16, 27, 34, and 164. We initially discovered these by direct dideoxy sequencing of polymerase chain reaction (PCR) products derived from overlapping primers designed to span the open reading frame. Subsequent studies have failed to reveal any additional nonsynonymous polymorphisms, nor have we found any variants consisting of deletions of truncations of the receptor. The original description of the human gene has been used to define the "wild-type" sequence, although it is clear that in some cases this is not the most frequent allele (Table 1). The most common polymorphisms are in the amino terminus of the receptor at amino acid position 16, where Arg or Gly can be found, and position 27, where Gln or Glu is common. A rare variant has been found at position 34, with an allele frequency of < 1%. At position 164, Thr (wild type) or Ile can be found, although the latter is somewhat uncommon (heterozygous state has been found in 3-5% of the population). Although not extensively studied, we have found that the frequencies of the position 16 and 27 variants differ between white, black, and Asian populations. In one report, we found that the allele frequencies for Gly-16 were 0.61, 0.50, and 0.40 for the three respective ethnic groups, and for Gln-27 they were 0.57, 0.73, and 0.80. 

View larger version (36K): [in this window] [in a new window]   Figure 1.   Amino acid sequence of the human 2AR and its 5' leader cistron. Shown are the location of polymorphisms in the DNA sequence that result in variation in amino acids at the indicated positions. The darkened circles indicate codons where the DNA sequence is variable but does not result in variation in the encoded amino acid.

In our initial case-control study, we found no difference in the frequencies of these polymorphisms between a heterogeneous asthmatic group and a nonasthmatic cohort (1). This lack of association with asthma has been reported by others as well (2). One study, carried out in children, did report a significant association with childhood asthma and the Gln-27 genotype. However, given that the ₂AR gene is localized to chromosome 5q 31-32, and microsatellite markers in this region have been linked to asthma, bronchial hyperresponsiveness, or atopy in family studies, it is generally accepted that this finding with the ₂AR polymorphism is due to its being in linkage disequilibrium with a nearby disease-causing gene.

We have begun to investigate the 5' untranslated region of the ₂AR for polymorphism (5). In this region a 19-amino acid peptide is encoded by a 5' leader cistron sequence (6). This peptide has been shown to modify ₂AR translation and thus the level of cellular expression. Critical elements of the peptide include arginines in its carboxy-terminal region (6). As shown in Figure 1, a single nucleotide polymorphism has been identified in the last codon of the cistrons, such that Arg or Cys can be present with allele frequencies of 0.37 and 0.63, respectively. Neither allele is more common in subjects with asthma (S. B. Liggett, unpublished observation, 1999).

As is subsequently discussed, the preceding polymorphism of the 5' leader cistron and those of the coding block clearly alter receptor expression, function, or regulation. With the three most common loci, we have been able to assess the relative frequencies of the various genotypic combinations (Table 2). As can be seen, several combinations are present at reasonable frequencies, some are uncommon, and others were not found. From Table 2 several other observations can be made. First, it appears that within the coding block, the Arg-16/Glu-27 genotypic combination is rare. This is due to linkage disequilibrium at these two close loci (4). So, when Arg-16 is present the likelihood of Gln being at position 27 is high. When Gly-16 is present, Glu-27 is more likely than Gln-27, but both combinations are observed. These results were obtained by methods that do not distinguish from which parental chromosome the polymorphisms arise. Thus haplotypes can only be inferred. To determine true haplotypes, specific techniques need to be employed. It would then be known exactly which two receptors, taking into account all polymorphic loci, are encoded. Such haplotyping has not been extensively performed to date, but as discussed below may be an important consideration.

Finally, one report has identified additional polymorphisms in the 5' upstream region (7). These occur at nucleotides 20, 367, 468, 654, 1023, 1343, and 1429.

Functional Consequences

Before embarking on additional clinical studies, our approach was to determine if these polymorphisms altered receptor function (5, 8). If so, then hypothesis-based clinical studies could be rationally designed. The results from such clinical studies are likely to be much more convincing if they are backed by results from cell- or transgenic mouse-based mechanistic studies. This scenario is in contrast to what might occur in association studies with a polymorphism of unknown relevance to protein function.

The methods we have used for determining the functional relevance of these polymorphisms have been delineated in detail (8, 9) and reviewed elsewhere (11). The approach has been to mimic the polymorphisms by site-directed mutagenesis, then to express the variant receptor in host cells that lack AR expression. Pharmacologic and biologic properties are then assessed. Typically, the effects of these polymorphisms on receptor expression, agonist- and antagonist-binding affinities, physical and functional coupling to stimulatory G protein (G_s), receptor trafficking, and receptor regulation by agonist are assessed. The results of these types of studies carried out in Chinese hamster fibroblasts expressing the different human polymorphic receptors of the ₂AR are summarized in Table 3.

The Ile-164 receptor has an approximately threefold decreased affinity for isoproterenol, epinephrine, and norepinephrine (8). More dramatic is the decrease in basal and agonist-stimulated adenylyl cyclase activities (Figure 2). The latter series of studies suggested that the conformation of the receptor is altered such that spontaneous toggling to the activated state, and agonist-stabilized activation, is significantly decreased in the Ile-164 receptor. Consistent with this notion, agonist competition studies in the absence of guanine nucleotide showed that the Ile-164 receptor failed to display high-affinity agonist-receptor-G_s interactions (8). The Ile-164 receptor has been expressed in the hearts of transgenic mice and a similar coupling defect has also been observed in receptor signaling (14). In addition, physiologic studies of these mice examining inotropic and chronotropic responses revealed that the Ile-164 receptor is substantially impaired in vivo. We have shown that this polymorphism has a significant disease-modifying effect in congestive heart failure (15), where ₂AR function may be particularly important since ₁AR are downregulated. Individuals with an Ile-164 polymorphism had an increased risk of death or transplant compared with those with Thr-164 with a relative risk of 4.81 (p < 0.001).

View larger version (16K): [in this window] [in a new window]   Figure 2.   Functional consequences of the Thr-to-Ile polymorphism of the human 2AR.

In contrast to the above, polymorphisms at positions 16 and 27 do not alter agonist affinity or coupling to G_s (9). However, agonist-promoted trafficking is altered. As shown in Figure 3A, long-term, agonist-promoted downregulation of receptor number is enhanced when Gly is present at position 16, and is absent when Glu is at position 27 (with Arg at position 16). This difference in downregulation is due to alterations in receptor degradation after the internalization step (9). Interestingly, when both polymorphisms are present (Gly-16/Glu-27), the downregulation is similar to that of Gly-16 receptors, suggesting that this locus may dominate the phenotype when Gly is present. These studies have also been carried out in human airway smooth muscle cells, where endogenous expression of receptors (i.e., not a transfection-based study) was examined in primary culture of cells from various individuals with different ₂AR genotypes. As shown in Figure 3B, downregulation was greater in Gly-16 cells compared with Arg-16 cells, and was minimal in cells expressing the Glu-27 receptor.

View larger version (29K): [in this window] [in a new window]   Figure 3.   Downregulation (DR) phenotypes of the amino-terminal polymorphisms. Transfected CHW cells (A) or airway smooth muscle cells (B) expressing the indicated polymorphic receptors were exposed to vehicle or isoproterenol (1 µM) in cell culture for 24 h and receptor density was determined by radioligand binding.

The polymorphism in the 5' leader cistron was found to alter receptor expression. This was examined by constructing vectors having identical sequence in the 5' untranslated region and the coding block except at nucleotide 47. COS-7 cells were transfected with these constructs and expression was quantitated by radioligand binding. As shown, there was a consistently greater ₂AR expression that occurred with the 5' LC Cys-19 construct as compared with the 5' LC Arg-19 construct. Similar experiments, carried out by cotransfection of a firefly luciferase gene to control for transfection efficiency, again showed greater expression of the 5' LC Cys-19 construct. Importantly, ₂AR mRNA levels were the same in these cells, pointing toward differences in protein translation rather than mRNA transcription, consistent with the role of the leader cistron peptide.

Scott and coworkers have reported evidence that four of the polymorphisms in the 5' upstream region are within the first ~ 500 bp of the 5' untranslated region, which may be of functional significance in regard to receptor expression (7). It is not clear, however, whether most of this control is due to the aforementioned polymorphism (5) at nucleotide 47, or whether there are contributions from the polymorphisms at nucleotides 20, 367, and 468.

	CLINICAL STUDIES OF AGONIST RESPONSIVENESS NEEDING CONFIRMATION

TOP INTRODUCTION ESTABLISHED CONCEPTS CLINICAL STUDIES OF AGONIST... FUTURE DIRECTIONS CONCLUSIONS IMPORTANT QUESTIONS REFERENCES

As indicated above, the phenotype of the common polymorphisms in the coding block, as assessed in cells, is one of differences in downregulation. Clinically, agonist-promoted downregulation may result in tachyphylaxis, which is defined as a waning of effectiveness of an agonist over time during repetitive administration. If polymorphisms are differentially downregulated in vivo by endogenous catecholamines (as they are in tissue culture), then the initial response to therapy could also be affected. Several studies examining the effect of ₂AR polymorphisms on agonist responsiveness have been reported.

Martinez and coworkers (2) examined the bronchodilating response to the administration of a single dose of albuterol in children (mean age, 10 yr). The study group consisted of 191 unaffected children and 78 children with a history of wheezing. Of this latter group 37 had a diagnosis of asthma. In the asthmatic group, 60% of those with homozygous Arg-16 had a positive (> 15.3% increase in FEV₁) response to albuterol, compared with 13.3% who were homozygous for Gly-16. While borderline significant (p = 0.05 for trend), it should be noted that the homozygous Arg-16 group consisted of only five patients. In addition, the actual changes in FEV₁ were not reported, and so the magnitude of the effect is not clear. Children with a history of wheezing failed to have a significant difference, while unaffected children again showed a greater percentage of positive responses in the homozygous Arg-16 group (p = 0.04). When the groups were analyzed together, the odds ratio (adjusted for asthma and wheezing) for having a positive bronchodilator response to albuterol was 5.3 for homozygous Arg-16 children (p = 0.007). No associations were noted for the polymorphism at position 27. This study suggests that the initial response to albuterol in children is affected by ₂AR genotype at position 16, although the data concerning children with asthma suffer from a small sample size. The results suggest that the Gly-16 receptor may be chronically downregulated (presumably by endogenous catecholamines), thereby resulting in a depressed response to exogenous agonist.

Tan and coworkers (16) examined relationships between tachyphylaxis to 4 wk of treatment with formoterol and ₂AR polymorphisms at positions 16 and 27. Those who were homozygous for Arg-16 displayed essentially no bronchodilator desensitization, while those who were homozygous for Gly-16 were found to have a 46% loss of responsiveness. A similar degree of tachyphylaxis was noted in heterozygotes. No association was found with the position 27 polymorphisms. Of note, the homozygous Arg-16 group consisted of only four patients, and within the Gly-16 group there was a significant amount of variation in bronchodilator desensitization (from a low of 0% to a high of 160%). Besides the small number of patients and the degree of variability within the groups, another concern raised by this study is its lack of agreement with other studies of tachyphylaxis. In most studies, and in general clinical experience, only a subset of patients displays tachyphylaxis to -agonists, and this group does not represent the majority of patients. On the basis of the above study, those with the most common allele (Gly-16) undergo significant tachyphylaxis, representing 80% of subjects.

Ohe and coworkers (17) used a restriction fragment polymorphism assay, which identifies silent polymorphisms at codon 175, to study subjects from four families. The response to albuterol was examined in 46 individuals. Those who were heterozygous or homozygous for the 2.3-kb allele had an approximately twofold greater responsiveness to this agonist compared with those homozygous for the 2.1-kb allele. It would be interesting to know the genotypes at amino acids 16 and 27, and at position 19 of the 5' leader cistron, in this cohort to compare these results with those of others.

Sears and colleagues (18) have reported the genotypes of patients they originally studies in 1990, where they studied the effects of regular versus as-needed fenoterol in a group of 64 subjects with asthma. The earlier report showed a deterioration in "asthma control" with regular fenoterol as compared with as-needed fenoterol. Clinically, more exacerbations, decreases in FEV₁ and morning peak expiratory flow (PEF), and increased responsiveness to methacholine were noted during regular agonist treatment. These investigators have now determined the ₂AR genotype at positions 16 and 27 in 61 of these subjects to determine whether changes in these variables in the earlier study were associated with ₂AR polymorphisms. Two of 10 markers of asthma control were significantly associated with genotype: subjects homozygous for Gly-16 had no decrease in bronchial responsiveness to methacholine with regular treatment, and those homozygous for Glu-27 had no increase in evening PEF during regular treatment. The study is somewhat complicated by the fact that during regular treatment the majority of patients experienced loss of control, so a comparable number of patients who did not experience this outcome was not available. Nevertheless, the results suggest that ₂AR genotyping has value in predicting these two outcomes during chronic -agonist treatment. It is not altogether clear, however, from the recombinant cell studies how these polymorphisms affect these two parameters under the conditions of the protocol. Additional in vitro studies with fenoterol may reveal agonist-specific phenotypes. Additional clinical studies, which include the 5' leader cistron polymorphism and any other upstream variants, are necessary to further assess these findings.

Other studies have investigated the potential for ₂AR polymorphisms to act as disease modifiers. Associations have been found with nocturnal asthma (19), bronchial hyperresponsiveness (20), increased IgE levels (21), and asthma severity (3). These results bring to mind the complexity of considering how polymorphisms of drug targets affect clinical efficacy. Since such polymorphisms may also affect the disease itself, it may be acting in two ways to ultimately alter drug response its disease-modifying effect and its pharmacogenetic effect.

	FUTURE DIRECTIONS

TOP INTRODUCTION ESTABLISHED CONCEPTS CLINICAL STUDIES OF AGONIST... FUTURE DIRECTIONS CONCLUSIONS IMPORTANT QUESTIONS REFERENCES

Interactions between ₂AR Polymorphisms

Although the 5' leader cistron and coding block polymorphisms can occur in various combinations, little is known about how they interact since most studies have studied isolated polymorphisms within an otherwise common "wild-type" sequence. Other 5' untranslated polymorphisms, more recently described, further complicate the picture. Cell-based studies of the common combinations need to be carried out. This will be extremely helpful in providing a sound mechanistic basis for the results of clinical studies in which ₂AR polymorphisms at multiple loci are determined. Within this context determination of the true haplotype will provide the greatest amount of genetic information. Knowing the genotype or haplotype at a few of these loci may provide as much power to predict responsiveness as knowing all of the polymorphisms, given that there is relatively small genetic distance between them. This concept needs to be explicitly tested, however. The fact that several of the preceding clinical studies found associations with position 16 or position 27 polymorphisms but not both suggests that several genotypes will indeed need to be ascertained.

Differential Agonist Effects

A number of -agonists in common use today differ significantly in structure and/or pharmacologic properties. In vitro studies, as well as clinical trials, need to be carried out with representative agents from each class. It is interesting to consider that the effects of these polymorphisms may be agonist specific, and that knowing the ₂AR genotype may help in selecting a drug with the most favorable predicted response. Alternatively, all -agonists may have the same phenotype, and the presence of certain polymorphisms may indicate the need to use alternative therapeutic agents acting at different targets.

Clinical Response to Agonists

As indicated above, control studies with larger patient populations, different levels of asthma severity, and different -agonists need to be carried out to confirm the existing studies. Trials examining the initial response as well as potential tachyphylaxis are necessary. A consensus on a method for defining a loss of clinical efficacy would be helpful since tachyphylaxis is currently studied by a variety of different protocols. Tachyphylaxis to the bronchodilator and bronchoprotective effects of -agonists needs to be considered.

Pharmacoethnogenetics

As indicated above, there appear to be significant differences in the frequencies of the polymorphisms between ethnic cohorts. It is intriguing to consider whether polymorphisms of the ₂AR may be the basis of differences in asthma phenotype or response to therapy between racial groups. This needs to be explicitly tested in clinical trials.

Molecular Basis of the Phenotypes

The ₂AR is one of the most extensively studied receptor drug targets in regard to the effects of polymorphic variation on receptor phenotype. Nevertheless, there are a number of questions that remain as to the precise manner by which these specific variances alter receptor function. Knowing more about how these polymorphisms alter function may lead to the development of new -agonists whose functions are not affected by polymorphic variation.

	CONCLUSIONS

TOP INTRODUCTION ESTABLISHED CONCEPTS CLINICAL STUDIES OF AGONIST... FUTURE DIRECTIONS CONCLUSIONS IMPORTANT QUESTIONS REFERENCES

The ₂AR is polymorphic within the human population. Cell and transgenic mouse studies have shown that some of these polymorphisms significantly alter function expression, or regulation of the receptor. Most, but not all, clinical studies to date have shown that ₂AR polymorphisms act as disease modifiers or are a determinant of responsiveness to -agonist therapy. Additional studies will be helpful in establishing the predictive role of these polymorphisms in asthma management. Finally, other polymorphisms of asthma drug targets have and continue to be found. Thus, the practice of physicians obtaining an "asthma polymorphism panel" on patients to further characterize the disease and optimize therapy is quite feasible.

	IMPORTANT QUESTIONS

TOP INTRODUCTION ESTABLISHED CONCEPTS CLINICAL STUDIES OF AGONIST... FUTURE DIRECTIONS CONCLUSIONS IMPORTANT QUESTIONS REFERENCES

• Are ₂-adrenergic receptor polymorphisms the basis of differences in asthma phenotype or response to therapy among racial groups?
• Will knowledge of the ₂-adrenergic receptor genotype assist in selecting the drug with the most favorable outcome? Or will all ₂-adrenergic agonists have the same phenotype and the presence of certain polymorphisms indicates the need for alternative therapies directed toward different targets?
• Will precise knowledge of ₂-adrenergic receptor phenotype variation lead to new ₂-adrenergic agonists whose actions are unaffected by polymorphic variation?
• Is there a potential for, and a value of, genetically tailored therapy?

	Footnotes

Correspondence and requests for reprints should be addressed to Stephen B. Liggett, M.D., University of Cincinnati College of Medicine, 231 Bethesda Avenue, Room 7511, P.O. Box 670564, Cincinnati, OH 45267-0564. E-mail: stephen. liggett@uc.ed; Internet: http://receptors.med.uc.edu

	References

TOP INTRODUCTION ESTABLISHED CONCEPTS CLINICAL STUDIES OF AGONIST... FUTURE DIRECTIONS CONCLUSIONS IMPORTANT QUESTIONS REFERENCES

1. Reihsaus, E., M. Innis, N. MacIntyre, and S. B. Liggett. 1993. Mutations in the gene encoding for the ₂-adrenergic receptor in normal and asthmatic subjects. Am. J. Respir. Cell Mol. Biol. 8: 334-339 .

2. Martinez, F. D., P. E. Graves, M. Baldini, S. Solomon, and R. Erickson. 1997. Association between genetic polymorphisms of the beta2-adrenoceptor and response to albuterol in children with and without a history of wheezing. J. Clin. Invest. 100: 3184-3188 [Medline].

3. Weir, T. D., N. Mallek, A. J. Sandford, T. R. Bai, N. Awadh, J. M. FitzGerald, D. Cockcroft, A. James, S. B. Liggett, and P. D. Pare. 1998. ₂ Adrenergic receptor haplotypes in mild, moderate and fatal/ near fatal asthma. Am. J. Respir. Crit. Care Med. 158: 787-791 [Abstract/Free Full Text].

4. Dewar, J. C., A. P. Wheatley, A. Venn, J. F. J. Morrison, J. Britton, and I. P. Hall. 1998. ₂-Adrenoceptor polymorphisms are in linkage disequilibrium, but are not associated with asthma in an adult population. Clin. Exp. Allergy 28: 442-448 [Medline].

5. McGraw, D. W., E. T. Donnelly, M. G. Eason, S. A. Green, and S. B. Liggett. 1998. Role of ARK in long-term agonist-promoted desensitization of the ₂-adrenergic receptor. Cell Signal. 10: 197-204 [Medline].

6. Parola, A. L., and B. K. Kobilka. 1994. The peptide product of a 5' leader cistron in the beta 2 adrenergic receptor mRNA inhibits receptor synthesis. J. Biol. Chem. 269: 4497-4505 [Abstract/Free Full Text].

7. Scott, M. G. H., C. Swan, A. P. Wheatley, and I. P. Hall. 1999. Identification of novel polymorphisms within the promoter region of the human ₂ adrenergic receptor gene. Br. J. Pharmacol. 126: 841-844 [Medline].

8. Green, S. A., G. Cole, M. Jacinto, M. Innis, and S. B. Liggett. 1993. A polymorphism of the human ₂-adrenergic receptor within the fourth transmembrane domain alters ligand binding and functional properties of the receptor. J. Biol. Chem. 268: 23116-23121 [Abstract/Free Full Text].

9. Green, S., J. Turki, M. Innis, and S. B. Liggett. 1994. Amino-terminal polymorphisms of the human ₂-adrenergic receptor impart distinct agonist-promoted regulatory properties. Biochemistry 33: 9414-9419 [Medline].

10. Green, S. A., J. Turki, P. Bejarano, I. P. Hall, and S. B. Liggett. 1995. Influence of ₂-adrenergic receptor genotypes on signal transduction in human airway smooth muscle cells. Am. J. Respir. Cell Mol. Biol. 13: 25-33 [Abstract].

11. Liggett, S. B.. 1995. Functional properties of human ₂-adrenergic receptor polymorphisms. News Physiol. Sci. 10: 265-273 . [Abstract/Free Full Text]

12. Liggett, S. B. 1996. The genetics of ₂-adrenergic receptor polymorphisms: relevance to receptor function and asthmatic phenotypes. In S. B. Liggett and D. A. Meyers, editors. The Genetics of Asthma. Marcel Dekker, New York. 455-478.

13. Liggett, S. B. 1996. Molecular and genetic basis of ₂-adrenergic receptor function and regulation. In P. Barnes, M. Grunstein, A. Leff, and A. Woolcock, editors. Asthma. Lippincott-Raven, New York. 299-311.

14. Turki, J., J. N. Lorenz, S. A. Green, E. T. Donnelly, M. Jacinto, and S. B. Liggett. 1996. Myocardial signalling defects and impaired cardiac function of a human ₂-adrenergic receptor polymorphism expressed in transgenic mice. Proc. Natl. Acad. Sci. U.S.A. 93: 10483-10488 [Abstract/Free Full Text].

15. Liggett, S. B., L. E. Wagoner, L. L. Craft, R. W. Hornung, B. D. Hoit, T. C. McIntosh, and R. A. Walsh. 1998. The Ile 164 ₂-adrenergic receptor polymorphism adversely affects the outcome of congestive heart failure. J. Clin. Invest. 102: 1534-1539 [Medline].

16. Tan, S., I. P. Hall, J. Dewar, E. Dow, and B. Lipworth. 1997. Association between beta 2-adrenoceptor polymorphism and susceptibility to bronchodilator desensitisation in moderately severe stable asthmatics. Lancet 350: 995-999 [Medline].

17. Ohe, M., M. Munakata, N. Hizawa, A. Itoh, I. Doi, E. Yamaguchi, Y. Homma, and Y. Kawakami. 1995. Beta₂ adrenergic receptor gene restriction fragment length polymorphism and bronchial asthma. Thorax 50: 353-359 [Abstract/Free Full Text].

18. Hancox, R. J., M. R. Sears, and D. R. Taylor. 1998. Polymorphism of the ₂-adrenoceptor and the response to long-term ₂-agonist therapy in asthma. Eur. Respir. J. 11: 589-593 [Abstract].

19. Turki, J., J. Pak, S. Green, R. Martin, and S. B. Liggett. 1995. Genetic polymorphisms of the ₂-adrenergic receptor in nocturnal and non-nocturnal asthma: evidence that Gly 16 correlates with the nocturnal phenotype. J. Clin. Invest. 95: 1635-1641 .

20. Hall, I. P., A. Wheatley, P. Wilding, and S. B. Liggett. 1995. Association of the Glu27 ₂-adrenoceptor polymorphism with lower airway reactivity in asthmatic subjects. Lancet 345: 1213-1214 [Medline].

21. Dewar, J. C., J. Wilkinson, A. Wheatley, N. S. Thomas, I. Doull, N. Morton, P. Lio, J. Harvey, S. B. Liggett, I. S. Holgate, and I. P. Hall. 1997. The glutamine 27 ₂ adrenoceptor polymorphism is associated with elevated immunoglobulin E levels in asthmatic families. J. Allergy Clin. Immunol. 100: 261-265 [Medline].