 |
Introduction: Cleft Palate, cAMP, and the Major Histocompatibility
Complex |
Our interest in genetic variation affecting
-adrenergic receptors and its potential influence on human
physiology started, perhaps surprisingly, with an observation about
glucocorticoid-induced cleft palate. On the other hand, given the
number of shared metabolic effects of glucocorticoids and the natural
-adrenergic agonists, epinephrine and norepinephrine, perhaps it
should not be surprising. The observation that spurred our interest was
that of Bonner and Slavkin (1975)
, showing that the major
histocompatibility locus of mouse, H-2, greatly influences
susceptibility to glucocorticoid-induced isolated cleft palate. At
about the same time, Meruelo and Edidin (1975) had published evidence
associating mouse liver cAMP levels with variation in the
H-2 genotype. Given the multiple roles of cAMP in
intracellular signaling, we sought to determine whether there was an
association between the H-2-influenced cAMP levels and cleft palate.
We were able to use powerful methods of mouse genetics to perform a
complex genetic/physiological analysis of cAMP levels in palatal
shelves before and after glucocorticoid treatment of the dams (Erickson
et al., 1979). In particular, we used congenic lines of mice in which a
small chromosomal region carrying the genetic allele of interest is
introduced onto another inbred strain by multiple backcrosses. This
allows one to study the influence of genetic variation at a candidate
locus relatively unaffected by all the other genetic differences
between the strains. Such strains were first constructed for
H-2. Our results confirmed an H-2 effect on cAMP
in these palatal shelves, and this effect was influenced by
glucocorticoids. However, these two variables could not explain all the
incidence of cleft palate we found. This effect appeared to be
tissue-specific since H-2 was found to be a determinant of
cAMP strain variations in brain (Hindin and Erickson, 1979), but not
spermatozoa (Erickson et al., 1979) or retina (Heidel and Erickson,
1983).
| |
Variation in  -Adrenergic Receptors Among Inbred Strains of
Mice |
We began a search for genetic variation in -adrenergic
receptors among inbred strains of mice since not all the variation in
tissue cAMP was explained by the H-2 locus in our studies. We turned to measurements of dihydroalprenolol binding to liver membranes to study such variations (Markovac and Erickson, 1983). We
found that liver membranes of the A/J inbred strain bound less dihydroalprenolol than did those of the C57/BL/6J inbred strain in the
absence, but not in the presence, of magnesium. Thus, the ratio of
binding with and without magnesium could be used as a variable to
search for genetic control among recombinant inbred (RI1) strains.
RI strains with their strain distribution patterns of marker genes are
valuable for clarifying the genetic complexity of the control of
-adrenergic activity. RI strains have proved to be powerful genetic
tools in demonstrating genetic independence, linkage, and pleiotropism
(Bailey, 1971; Swank and Bailey, 1973; Taylor, 1978). RI lines are
derived from brother-sister matings, beginning with the
F2 generation obtained from the cross of two inbred strains. These RI lines have a mixed genome from the two parental strains and will only have one or the other parent's alleles
at any one locus, because the majority of RI lines used have been
brother-sister mated for more than 20 generations. Since the assorted
alleles in the RI lines are homozygous at each locus, large numbers of
identical mice were available for the dihydroalprenolol binding
studies. Segregation of the Mg2+-sensitive
dihydroalprenolol binding in 12 RI lines was found, suggesting that it
was very likely to be due to a single gene. We suggested a new gene,
named "Badm", for -adrenergic magnesium effect, for this variation (Markovac and Erickson, 1983).
| |
Variation in Phospholipid Methylation among Inbred Strains of
Mice |
Inasmuch as -adrenergic agonist binding had been shown to
increase phospholipid methylation by methyltransferases I (MTI) and
II, and, conversely, increased membrane methylation appeared to
increase -adrenergic stimulation of cAMP (Hirata and Axelrod, 1980),
we sought to study the genetic effects on these methyltransferases and
their subsequent effects on -adrenergic activation. We found significant differences in membrane MTI when activity was assayed among
inbred strains of mice (Markovac and Erickson, 1985a). The results of
our studies using F1 progeny of high and low
strains suggested a dominance of high MTI activity over low MTI
activity. We also investigated RI lines between the high and low
strains and found that MTI activity must be regulated by at least two major genes. The use of H-2 congenic lines suggested that a
portion of hormone-stimulated methyltransferase I activity is linked to the H-2 complex (Markovac and Erickson, 1985b).
| |
Human -Adrenergic Receptor Variation and the Major
Histocompatibility Complex |
We also sought to determine whether the HLA locus of humans, the
equivalent of H-2 in mice, had a significant effect on
-adrenergic responses. One hundred and fifty-nine individuals were
typed for multiple HLA-A and -B antigens and levels of
isoproterenol-stimulated lymphocyte cAMP. No significant age, sex, or
caffeine effects on the natural log of the lymphocyte cAMP variable (ln
cAMP) were found. A comparison of mean ln cAMP levels between
individuals who carried a particular antigen (homozygous or
heterozygous) and individuals who did not carry the antigen identified
a highly significant decrease in ln cAMP levels associated with the
HLA-B18 antigen. We estimated that 18.9% of the variability in ln cAMP was attributable to the HLA-B18 antigen. In addition, 38% of the variability in ln cAMP was attributable to factors that aggregate in
families that were independent of the HLA-B18 effect. A weaker association of HLA-A10 with lymphocyte cAMP might be due to linkage disequilibrium between HLA-A10 and -B18 (Erickson et al., 1985).
| |
The Cloning of the Human 2-Adrenergic Receptor and Newfound
Variation |
Studies on genetic variation in 2-adrenergic receptors could
focus on genetic variation in the receptor itself, instead of only on
factors that might modify -adrenergic receptor levels, subsequent to
the cloning of the hamster 2-adrenergic receptor in 1986 (Dixon et
al., 1986). The 2-adrenergic receptor turned out to be a member of a
very large family of seven transmembrane domain-containing
G-protein-coupled receptors that includes rhodopsin, the light receptor
of the eye, as one of its members. The human gene was soon cloned and
clinically relevant variation sought. Liggett and coworkers (Reihsaus
et al., 1993) used the sensitive technique of temperature gradient gel
electrophoresis to screen for mutations in a group of patients with
asthma as compared with normal controls. Of nine point mutations (SNPs)
found, only four resulted in amino acid changes, and of these four,
only two were reasonably abundant.
| |
2-Adrenergic Receptor Variation and Asthma-Related
Phenotypes |
It was of interest to study the -adrenergic receptor
polymorphisms in asthma since -adrenergic agonists are one of the
mainstays in the treatment of asthma, implicating -adrenergic
responses in airway constriction. The frequencies of these two receptor polymorphisms, arginine 16 
glycine (Arg16Gly) and glutamine 27 glutamic acid (Gln27Glu) [we will use the nomenclature first described
(the Gly at 16 is more common), or common amino acid, numerical
position of that amino acid, variant amino acid], were compared
between asthmatic patients and normals. No difference in incidence of
these receptor polymorphisms was found between the groups, but the
Arg16Gly variant identified a subset of patients likely to be
steroid-dependent and to require immunization therapy (Reihsaus et al.,
1993). This more severe phenotype was found in homozygous subjects for
Arg16Gly but not in heterozygotes.
| |
A Biochemical Basis for the Phenotypic Effects of the Two Abundant
2-Adrenergic Receptor Variants |
In an important article, Liggett's group studied the potential
basis for the clinical variation found with the amino acid position 16 and 27 substitutions in the 2-adrenergic receptor (Green et al.,
1994). When constructs containing the normal, either single or double
polymorphism, were transfected into Chinese hamster fibroblasts, the
only difference found was variation in the amount of agonist-promoted
down-regulation of receptor expression. The Arg16Gly had increased
down-regulation while the Gln27Glu was resistant to down-regulation,
and the double variant was comparable with the Arg16Gly (Green et al.,
1994). These results were somewhat surprising, given the number of
pathways involved in regulation of 2-adrenergic receptors, not all
of which might expected to be present in Chinese hamster fibroblasts.
These modifying factors include methyltransferases (Hirata and Axelrod,
1980), -arrestin (Ferguson et al., 1996), protein kinase A (Daaka et
al., 1997), -adrenergic receptor kinase (Choi et al., 1997), and
GIT1 (Fremont et al., 1998).
| |
Other Studies of 2-Adrenergic Receptor Variation and
Asthma-Related Phenotypes |
A limitation of the study by Liggett's group (Reihsaus et al.,
1993) was the lack of use of intermediate variables such as IgE, skin
tests, pulmonary function tests, or other quantitative measurements
related to the severity of asthma. Recent genetic association studies
between such intermediate variables and the polymorphisms in the
2-adrenergic receptor have provided more evidence for the importance
of variation at or near the 2-adrenergic receptor (Table
1). Tan et al. (1997) provided data
supporting the notion that the Arg16Gly polymorphism was associated
with down-regulation of the 2-adrenergic receptor. They used a
4-week course of -adrenergic agonist or placebo in
glucocorticoid-dependent asthmatics. Following this, bronchodilator
desensitization was compared by spirometry. The results showed that
homozygous Arg16Gly was associated with greater desensitization.
However, there were only 10 patients in this group being compared with
four patients in the homozygous Arg16 group (Tan et al., 1997).
View this table:
[in this window]
[in a new window]
|
TABLE 1
Studies reporting association of asthma-related traits with genetic
variation in the 2-adrenergic receptor
|
|
Our group's efforts focused on the bronchodilator response to
2-adrenergic agonist given the importance of bronchodilation to
asthma responses (Martinez et al., 1997). We genotyped 269 children who
participated in a longitudinal study of asthma. Spirometry was
performed before and after administration of 180 µg of albuterol, and
a positive response was considered an increase of >15.3% predicted FEV1. There was marked linkage disequilibrium
between the two polymorphisms, with 97.8% of all chromosomes that
carried Arg16 also carrying Gln27 (actually 100% with retyping). When
compared with homozygotes for Arg16Gly, homozygotes for Arg16 were 5.3 times (95% confidence interval, 1.6-17.7) and heterozygotes for Arg16/Arg16Gly were 2.3 times (1.3-4.2) more likely to respond to
albuterol, respectively. Similar trends were observed for asthmatic and
nonasthmatic children, and results were independent of baseline lung
function, ethnic origin, and previous use of antiasthma medication. These results were confirmed by Lima et al. (1999) in a much smaller group of patients and by Kotani et al. (1999) with salbutamol. On the
other hand, we found no association between the Gln27Glu polymorphism
and response to albuterol. This is in contrast to the results of Hall
et al. (1995), who reported decreased responses to the
parasympathomimetic agonist, methacholine, in Gln27Glu homozygotes.
In a study that merely compared nonasthmatic, mild-moderate asthmatics
and fatal/near fatal asthmatics for these polymorphisms, no haplotype
was found to be associated with fatal/near fatal asthmatics (Weir et
al., 1998). However, the combined Arg16Gly/Gln27Glu haplotype that also
showed enhanced down-regulation in Liggett's initial studies (Green et
al., 1994) was associated with moderate asthma as compared with
mild asthma in a division of that subgroup, primarily on the basis of
steroid dependence (Weir et al., 1998). Thus, this group's results
support the initial Reihsaus et al. (1993) data on steroid dependence,
but the lack of a trend from mild to moderate to fatal/near fatal
asthma is not supportive of other results. Finally, the Gln27Glu
variant of the 2-adrenergic receptor was found to be associated with
elevated levels of serum IgE in subjects from asthmatic families (Dewar
et al., 1997), supporting previous data relating increased levels of
cAMP to increased IgE synthesis (Fedyk et al., 1996). A more recent
study could not confirm this finding (Deichmann et al., 1999).
Liggett's group (Turki et al., 1995) also provided evidence for a role
in the Arg16Gly polymorphism in nocturnal asthma. They had previously
shown that the 2-adrenergic receptors in circulating white blood
cells are down-regulated in patients with nocturnal asthma (maximal at
4:00 AM, which does not happen in normals; Szefler et al., 1991). In
line with the biochemical evidence for greater down-regulation of the
Arg16Gly, the frequency of this variant was found to be much higher in
the nocturnal asthma group as compared with asthma patients without
nocturnal exacerbation (Turki et al., 1995).
| |
Other 2-Adrenergic Polymorphisms and Other Phenotypes |
Another polymorphism in the 2-adrenergic receptor produced more
marked effects on the function of the receptor (Green et al., 1993).
The rare Thr164Ile (1.5%) receptor had small decreases in binding
affinity for some 2-adrenergic receptor agonists (determined by
their phosphorylation) with a marked decrease in basal and epinephrine-stimulated adenylyl cyclase activities. This was shown to
be due to defective coupling of the receptor to the stimulatory G-protein, while agonist-promoted sequestration of the receptor was
also impaired (Green et al., 1993). This polymorphism was found to be
associated with poor outcomes in patients with congestive heart
failure. Specifically, the Thr164Ile patients had 1-year survivals of
only 42% compared with 76% for patients with the normal variant at
this position (Liggett et al., 1998).
Large et al. (1997) found a very significant correlation of the
Gln27Glu polymorphism with obesity in women (Table
2). They studied the response of fat
cells obtained by biopsies from these women using terbutaline as a
2-adrenergic specific agonist. They found that the Arg16Gly showed
an increased sensitivity to terbutaline without any change in
2-adrenergic receptor expression. Thus, when studied in human cells
normally expressing the 2-adrenergic receptor, taking advantage of
the different genotypes, sensitivity not related to down-regulation was
found. Hellstrom et al. (1999) found the opposite effect in males. Our
data do not support an association of children's weight with either
Arg16Gly or Gln27Glu (unpublished data), and negative linkage data have
been reported (Takami et al., 1999).
| |
3-Adrenergic Receptor Variation |
While the relationship of the 2-adrenergic receptor to body
weight remains to be clarified, a body of work suggests an association of the 3-adrenergic receptor with weight-related variables. A common
polymorphism at amino acid 64 (Trp64Arg) has been used most frequently
in these studies. This variant has been associated with insulin
resistance (Kawamura et al., 1999) and gestational diabetes (Festa et
al., 1999) and has been implicated in the development of obesity
(Kurabayashi et al., 1996; Hoffstedt et al., 1999; Proenza et al.,
2000), but this result has not been confirmed by many groups (Buettner
et al., 1998; Mitchell et al., 1999; Witchel et al., 2000). Although
the Mitchell group (1999) could not confirm the association of obesity
with the 3-adrenergic receptor, they did find a large Lod (linkage)
score for body mass index with this genetic region. In vitro studies
show that the Trp64Arg polymorphism leads to increased triglyceride
uptake in 3-adrenergic receptor-transfected cells (Gros et al.,
1999). To the extent that morbid obesity can increase breathing
problems, these findings suggest that variations in the 3-adrenergic
receptor could also affect asthma severity.
| |
A Polymorphism in the Leader Peptide of the 2-Adrenergic
Receptor |
The recent identification of a polymorphism in a leader peptide
could provide alternative interpretations to those presented above.
2-Adrenergic receptor mRNA has a 5' leader region within which an
open reading frame encoding a 19-amino acid peptide is found (Kobilka
et al., 1996). This peptide has a direct effect on translation of its
own mRNA (Parola and Kobilka, 1994). A polymorphism for the C-terminal
amino acid of this leader peptide was recently reported (McGraw et al.,
1998). Transfection of cells with constructs coding for the two forms
of the leader peptide, but identical 2-adrenergic genes,
demonstrated that the open reading frame (ORF) Arg19 inhibited
translation twice as effectively as did the ORF Arg19Cys (McGraw et
al., 1998). Importantly, there was complete linkage disequilibrium
between the ORF Arg19Cys polymorphism and the Gln27 polymorphism and
partial linkage disequilibrium with the Arg16Gly polymorphism. Thus,
many of the previous results may reflect different levels of
-adrenergic receptor due to this polymorphism. For instance, in our
study (Martinez et al., 1997), the patients had been withdrawn from
-agonist before the study, so a reasonable mechanism was that the
Arg16Gly subjects were down-regulated due to the exposure to endogenous
catecholamines. We have now investigated the relationship of the new
polymorphism to the bronchodilatory responses previously found
associated with Arg16. Although the linkage disequilibrium leaves a
small number of test cases, we find the leader peptide polymorphism to
be unimportant (Graves et al., in preparation). We anticipate
that there will be many more SNPs in the 5' untranslated region
(putative promoter) of the 2-adrenergic receptor.
| |
Detecting Haplotypes of 2-Adrenergic Polymorphisms |
Our results, and the results of others (see Tables 1 and 2),
strongly indicate the need for methods that can distinguish haplotypes
in heterozygotes, i.e., not depending just on homozygotes to determine
associations of SNPs with disease phenotypes. The Acrydite technology
allows a method to haplotype PCR products in heterozygotes. This
technology depends on the use of "capture" oligodeoxynucleotides
coupled to acrylamide gels using the Acrydite moiety. We have used this
technology, in collaboration with Mosaic Technologies, to haplotype the
Arg16Gly and Gln27Glu27 polymorphisms (Hammond et al., in
preparation). Using a capture oligodeoxynucleotide with the two
SNP variants, it was possible to differentially capture the four
possible oligos using a simple temperature gradient/Acrydite capture
electrophoresis. Thus, PCR products with no matches to the two
positions moved the farthest; there was a differential capture between
the two single polymorphisms in the gradient; and the PCR product with
a match to both SNPs was captured quite high on the gel. Thus, the four
genotypes were clearly distinguishable. This is certainly a very useful
approach for performing haplotype determinations with any SNPs that are
located close together.
| |
Conclusion |
In conclusion, our work started with studies on quantitative
variation in the traits related to -adrenergic signaling, cyclic AMP, and the binding properties of the -adrenergic receptor. Interestingly, these implicated the major histocompatibility locus as a
modifier of -adrenergic signaling. We also identified an as yet
unmapped gene in recombinant inbred lines that influenced the magnesium
sensitivity of antagonist binding by the -adrenergic receptor. Our
work, and the work of others, has advanced to using discrete variations
in the -adrenergic receptor. These polymorphisms, detected as SNPs,
allow very powerful tests of the association of variation in the
-adrenergic receptor with disease phenotypes. Highly significant
associations with a number of phenotypes related to asthma, including
steroid dependence and bronchodilator responsiveness, have been
validated. Associations with obesity and with the outcome for
congestive heart failure (a rare SNP) have not yet been validated. Nonetheless, we believe that this experimental approach has great power
to help identify the many genes that influence common disease, such as
asthma and congestive heart failure, and the genotyped subject's
response to medications.
Abbreviations used are:
RI, recombinant inbred;
SNP, single nucleotide polymorphism;
MT, methyltransferase;
HLA, human
lymphocyte antigen;
ORF, open reading frame;
PCR, polymerase chain
reaction.
![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
Wings for Genetic Research, Steele
Memorial Children's Research, Department of Pediatrics and Molecular
and Cellular Biology (R.P.E.), and Respiratory Sciences Center
(P.E.G.), University of Arizona College of Medicine, Tucson, Arizona
Top
Abstract
Introduction: Cleft Palate,...
-Adrenergic...
