Pulmonary Pharmacology & Therapeutics (2000) 13, 127–134

doi.10.1006/pupt.2000.0239, available online at http://www.idealibrary.com on

PULMONARYPHARMACOLOGY& THERAPEUTICS

Importance of �2 Adrenergic Receptor Genotype, Gender andRace on Albuterol-evoked Bronchodilation in Asthmatics

John J. Lima∗†‡, Mohamad H. N. Mohamed†, Timothy H. Self†, Louis V. Eberle§,Julie A. Johnson†‡¶

Departments of †Clinical Pharmacy, ‡Pharmaceutical Sciences and §Medicine, Colleges of Pharmacy andMedicine, University of Tennessee, Memphis, USA

SUMMARY: To determine whether race and gender affect �2 receptor-stimulated bronchodilation, we quantifiedFEV1 and plasma concentrations of albuterol at various times following the oral administration of a single 8-mgdose of albuterol in 15 black and 15 white male and female asthmatics. No important racial or gender differencesin albuterol-evoked FEV1 or percent-predicted FEV1 were evident, although females tended to be more sensitivecompared to males. Pharmacodynamic (PD) models were fitted to data in 19 patients (63%); FEV1 was too erraticto fit in three, and a clockwise hysteresis in the FEV1 vs. albuterol concentration relationship was observed in eightasthmatics. Mean ± SD baseline (E0), maximal FEV1 (Emax) and C50 were: 3.18 + 1.03 l, 4.00 ± 1.12 l, 7.84 ±10.2 �g/l, respectively. �2 receptor genotype was determined in 16 patients. All Arg 16 homozygotes exhibitedproportional FEV1 response vs. plasma albuterol concentration relationships, and thus were fitted by PD models.All those having a poor FEV1 vs. albuterol concentration relationship carried the Gly 16 allele. We conclude thatreceptor genotype, but not race or gender, is an important determinant of albuterol pharmacodynamics.

2000 Academic Press

KEY WORDS: Race, Gender, Genotype, Albuterol, FEV1.

INTRODUCTION and to develop more effective and safer drugs to treatthe disease.

Asthma is a chronic inflammatory disease char- Short acting �2 agonists are used to provide quickacterized by airway hyperresponsiveness and variable relief in patients with mild intermittent to severeairway obstruction.1 It affects 14–15 million people in persistent asthma symptoms, and are the most potentthe United States and is the most common disease bronchodilators currently available.1 Followingof childhood, affecting 4.8 million children. Studies inhalation of short-acting �2 agonists, 60–70% ofreport that asthma prevalence is 10–50% higher,2,3 and patients respond variably, while about a third fail tomortality is three- to four-fold higher3–5 in blacks respond at all7. The mechanisms responsible for thecompared to whites. Hospital admissions for asthma inter-patient variability in response to inhaled �2 agon-are 3–5 times more frequent in blacks than in ists are unclear, and may be related to race andwhites.3,6 Despite advances in our understanding and gender. For example, several studies report that non-treatment of asthma, morbidity and mortality con- pulmonary �2 adrenergic receptor-(�2 AR) mediatedtinue to rise among all ethnic and age groups. Major responses are lower in blacks compared to whites,8–11

goals of research are to better understand asthma, suggesting the bronchodilator response to �2-agonistsmay also differ by race. In comparisons of adultasthmatics, females have greater risk of hos-∗Current address. Center for Clinical Pediatric Pharmacology,pitalization, poorer quality of life and greater asthmaNemours Children’s Clinic, 807 Nira Street, Jacksonville, FL

32207. drug use than males.12,13 However, non-pulmonary �2¶ Author for correspondence. Department of Pharmacy Practice,AR mediated responses are greater, and lymphocyteUniversity of Florida, Box 100486, Gainesville, FL 32610-0486.

Fax: (352) 392 9388. E-mail: johnson@cop.ufl.edu �2 ARs and post-receptor activity are more sensitive

1094–5539/00/030127+08 $35.00/0 2000 Academic Press127

128 J. J. Lima et al

to receptor stimulation in females.14–16 In order to The study protocol was approved by the Universityof Tennessee Institutional Review Board and eachexplore the effects of race and gender on the pul-

monary response evoked by �2 AR stimulation, we subject signed informed, written consent prior to par-ticipation in the study. All studies were conducted inmeasured FEV1, and plasma concentrations at various

times following the administration of a single oral the University of Tennessee General Clinical ResearchCenter (GCRC).dose of albuterol to 15 black and 15 white asthmatics

of both sexes. The pharmacokinetics of albuterol inStudy Protocolthese patients have been published.17 In addition, aStudy subjects reported to the GCRC at 7:00 AM atblood sample was obtained in 16 of the 30 studywhich time they received a standard breakfast. Theysubjects for the purpose of determining the impact ofwere instructed to abstain from caffeine and alcohol�2 AR genotype on response. The results of that studyfor at least 12 h before and throughout the study day,were also recently published.18 The purpose of thisand to avoid all over-the-counter medications for 24 hpaper is to compare albuterol bronchodilator phar-prior to the study. Inhaled corticosteroid therapy �-macodynamics by race, gender and �2 AR genotype.agonists were withheld for 24 and for 12 h respectivelyprior to the study.

An indwelling venous catheter was inserted into theMETHODSforearm for collection of blood samples. Each subjectwas given an 8 mg dose of albuterol oral solution,Patientsfollowed by 4 oz water. Blood samples for de-termination of plasma albuterol concentrations wereFifteen black and 15 white moderate asthmatics (eightcollected at 0, 0.5, 1, 1.5, 2, 2.5, 3, 3.5, 4, 6, 8, 10 andmales and seven females in each group) ranging in12 h after the dose. Blood samples for determinationage from 18 to 50 years participated in the study.of albuterol plasma concentration were collected intoClassification of a patient as having moderate asthmaevacuated blood collection tubes containing heparin.was based on the guidelines of the current ExpertPlasma was separated and removed, and stored atPanel on the Management of Asthma.1 Specifically−20°C until analysis. Subjects received lunch andpatients were required to be receiving short-actingdinner after the fourth and tenth hour time points,inhaled �2 agonists as needed to control symptoms,respectively. All diets were caffeine free, low potassiumand inhaled corticosteroids. They also kept a diary ofand standardized to the extent possible by the GCRCinhaled �2 AR agonist use for 3 days prior to thedietician. Subjects remained supine during most ofstudy. If their use was required >3 times/day on anythe study period.of those 3 days, they were studied at a later date.

Patients taking other asthma medications, systemiccorticosteroids or systemic or oral �2 agonists were Spirometryexcluded from the study. Daily inhaled corticosteroid

All subjects were trained prior to the study in per-use could not exceed the following: 800 �g beclo-forming standard spirometry, based on the guidelinesmethasone, 800 �g triamcinolone, 1000 �g flunisolide,of the American Thoracic Society (ATS).19 Patientsand 600 �g of fluticasone. Black and white subjectswere instructed to ‘blast’ at the very start of thewere matched for inhaled corticosteroid dose. Womenmaneuver with continuous coaching. The coach whowere studied during the second week (follicular phase)trained the patients in the forced expiratory maneuverof their menstrual cycle. Exclusion criteria includedwas used throughout the study. FEV1 was determinedtobacco smoking, any drug therapy for asthma otherby an open circuit testing with a bedside spirometerthan inhaled corticosteroids and inhaled �-agonists,(Spirometrics Flomate, Model 2500) according to pub-the presence of other systemic disease states, knownlished standards.19 The spirometer was calibrated onhypersensitivity to �-agonists, any clinically significanteach study day using a 3 l syringe before the study waselectrocardiographic abnormalities, pregnancy, seruminitiated. The time, date, temperature and barometricpotassium < 4.0 mmol/l (mEq/l) and postmenopausalpressure, and each subject’s race, height and weightand premenopausal women. The presence of exclusionon each study day were included in the spirometer’scriteria were ascertained through history, physicalstudy profile. Knudson’s20 method was selected forexamination, 12-lead electrocardiogram and routinethe calculation of predicted/reference values. Eachlaboratory analysis (SMA-23, complete blood count,subject used a nose clip and performed the spirometryurinalysis). Each woman of child-bearing potentialmaneuvers in the standing position. A minimum ofwas given a urine pregnancy test on the study daythree maneuvers was performed, with a minimumbefore the study was initiated. Subjects were matchedexhalation time of 6 s per maneuver, unless there wasfor age, race, gender, oral contraceptive use (women),an obvious plateau (i.e. no volume change) for atand to the extent possible, for household income,

employment status and educational level. least 1 s, or the subject could not exhale further. The

Factors affecting albuterol-evoked bronchodilation 129

paper printout of the volume-time and flow-volumeE= E0+

(Emax− E0)CpN+ C50

N · CpN (1)displays were subjected to other acceptability criteriaof the ATS. Reproducibility was confirmed by en-suring that the difference of the two largest FEV1 E= E0+

(Emax− E0)Cp+ C50

· Cp (2)values from acceptable maneuvers was less than200 ml. Spirometry was performed just prior to, and E = E0 +�Cp (3)at 1, 2, 3, 4, 6, 8, 10 and 12 h following albuterol

Plots of response vs. plasma concentrations were con-administration.structed in the order in which samples were collectedto detect the presence of hysteresis. If the hysteresis

Determination of �2 AR Genotype was counter-clockwise, a link model (eqn 4)22 and anindirect response model (stimulation of output) (eqn�2 AR genotype was determined by PCR with allele5)23 was used to fit the data.specific oligonucleotide (ASO) hybridization as pre-

viously described.18,21.E=E0+

(Emax− E0)CE+ CE50

· CE (4)

Analytical methods and pharmacokinetics

Details of the methodology employed to quantify CE=F · Dose · ka · keo

Valbuterol plasma concentrations and the phar-macokinetics of albuterol in these patients have been

� e−�z(t−tlag)

(ka−�z)(keo− �z)+

e−ka(t−tlag)

(�z− ka)(keo− ka)�reported.17 Plasma albuterol concentrations werequantified by using HPLC with fluorescence detection.One- and two-compartment models with first-orderinput and a time-lag were fitted to the plasma albuterol dE

dt= kin·�1+ Emax·Cp

Cp+ C50�−kout·E (5)concentration (Cp) vs. time data in each patient.Pharmacokinetic parameters were F/V, ka, �1 and �z

and lag-time, where F is the bioavailability, V is thePharmacodynamic parameters of the link model werevolume of distribution, ka is the first-order absorptionE0, Emax, C50 and keo, where keo is the first-order raterate constant and �1 and �z are the first-order macroconstant describing efflux of albuterol concentrationsconstants describing the initial and terminal slopes ofin the effect compartment (CE).22 Pharmacodynamicdisappearance of Cp vs. time data. The computer fitsparameters of the indirect response model were: kin,were excellent, had high precision and the weightedthe zero-order constant describing albuterol-evokedsum of residual squares (WRSS) were small.17 There-bronchodilation; kout, the first-order rate constantfore, in the present study pharmacodynamic modelsdescribing the disappearance of albuterol-evokedwere fitted to response vs. observed (not fitted) al-bronchodilation; Emax, the maximal albuterol-evokedbuterol concentrations in patients, unless otherwisebronchodilation; and C50, the plasma albuterol con-stated.centration at which maximal albuterol-evoked bron-chodilation is observed. A battery of statistical tools

Pharmacodynamics was used to evaluate the model selected;24 modelselection was based on visual inspection, correlationThe influence of race and gender on the phar-coefficient, mean standard deviation of the fit (squaremacodynamics of albuterol-evoked FEV1 was ex-root of WRSS/degrees of freedom), reality and pre-amined by comparing FEV1 values observed atcision of model parameters and the Akaike in-baseline, and maximal FEV1 values observed fol-formation criteria. A model fit was judged acceptablelowing the administration of albuterol in blacks vs.if the computer-fitted line mimicked the observedwhites and in males vs. females. Changes in FEV1data, had a maximal S value of 0.25, and a minimalevoked by albuterol (maximal FEV1–baseline FEV1)correlation coefficient of 0.8.and % changes in FEV1 ([maximal FEV1-baseline

FEV1]/baseline FEV1∗100) were also compared.

StatisticsPharmacodynamic modeling The sample size was determined by statistical power

analysis. With an alpha of 0.05 and 15 subjects/groupThe Sigmoid Emax model (eqn 1), Emax (eqn 2) or linearwe had greater than 95% power to detect a 1.33(eqn3) models were fitted to the FEV1 vs. plasmastandard deviation difference between groups and 80%albuterol concentration curve in each patient usingpower to detect a 1 SD difference between groups.WinNonLin (WinNonlin, Professional Edition 1.5,

Pharsight Corporation, Palo Alto, California, USA). The FEV1 data were not normally distributed, and

130 J. J. Lima et al

Table 1 Comparison of albuterol-evoked median (range) FEV1 and percent predicted FEV1 by race and gender

FEV1 Race Gender

Black (n=15) White (n=15) Female (n=14) Male (n=16)Baseline, L 2.7 (1.6–4.2) 3.4 (1.9–6.0)∗ 2.7 (2.1–4.0) 3.5 (1.6–6.0)∗Maximal, L 3.1 (1.7–6.4) 3.8 (2.7–7.4) 3.0 (2.4–4.7) 4.0 (1.7–7.4)∗Change, L 0.39 (0.10–2.8) 0.52 (0.11–3.3) 0.31 (0.10–0.77) 0.62 (0.11–3.3)∗Percent PredictedFEV1

Baseline, % 89 (50–117) 100 (50–122) 100 (74–122) 85.5 (50–122)∗Maximal, % 105 (52–167) 109 (70–171) 110 (86–133) 106.5 (51–171)Change, % 12 (2–73) 14 (2–76) 12 (4–24) 13.5 (1–76)

∗P <0.05

therefore statistical comparisons of albuterol phar-macodynamics by race and gender were made by theMann Whitney U test using the program Statistica(StatSoft, Tulsa, Oklahoma, USA). Data are pre-sented as median (range). We used the same approachto detect the effect of receptor genotype on albuterol-evoked FEV1.18 Differences observed in phar-macodynamic model parameters were tested for stat-istical significance using an unpaired t-test.

RESULTS

Effects of race and gender on maximalbronchodilator response

Albuterol-evoked effects on FEV1 by race and genderare shown in Table 1. As anticipated, the baselineFEV1 was lower (P < 0.05) in blacks and females thanin whites and males.19 Percent predicted FEV1 waslower in males than females (P < 0.05) but was notdifferent between blacks and whites. Maximal FEV1

following administration of albuterol, and albuterol-induced change in FEV1 were higher in males thanfemales, but no gender differences were noted in either

2.615

2.9

Albuterol plasma concentrations (ng/ml)

FE

V1

(l)

0 3 6 9 12

2.7

2.8

5.812

6.6

0 2 64 10

6.0

6.2

2.612

3.4

0 2 6 8 10

3.0

3.2

8

6.4

4

2.8

of these parameters when evaluated as percent pre-dicted FEV1. Fig. 1 Individual Pharmacodynamic Model Fits. Model fits of

FEV1 vs. plasma albuterol concentrations in patients #3 (top),#10 (middle) and #13 (bottom). Pharmacodynamic models andparameters are listed in Table 2. The lines are computer-fitted,Pharmacodynamic modelingthe symbols represent observed FEV1 vs. concentration pairs.

Acceptable fits of FEV1 vs. plasma albuterol con-centrations were obtained in 19 of the 30 asthmatics(63%) (Table 2). The Emax and Sigmoid Emax models increase over the fitted E0 and the mean ± C50 was

7.84 ± 10.2 �g/l (median=6.39 �g/l).provided the best fits in 15 patients (Fig. 1), while thelinear model was used to fit data in four patients (data A relatively small counter-clockwise hysteresis was

observed in four of our 30 patients (patients 1, 5, 12not shown). In most patients the fitted and observedvalues for E0 were in excellent agreement; mean ± and 20). The link and indirect models were fitted to

the data in these four patients, but were superior toSD E0 was 3.18 ± 1.03 l. The mean Emax ± SD was4.00 ± 1.12 l, which represented a 26% or 820 ml the Sigmoid Emax or Emax models in only two patients

Factors affecting albuterol-evoked bronchodilation 131

Bronchodilator response was not related to plasmaalbuterol concentrations in 11 (37%) patients, andthus models could not be fitted to the data from thesepatients. When response vs. plasma concentrationswere plotted in the order in which samples werecollected, a clockwise hysteresis was evident in eightof the 11 patients. The data for one of these patientis shown in Fig. 3. This pattern typified the responsevs. concentration relationship in these eight patients.

2.812

3.8

Time (h)

FE

V1

(l)

0 2 6 10

3.0

3.2

4 8

3.6

3.4

Response, concentration pairs were recorded on 64Fig. 2 Link Model Fit. Link model fit of FEV1 vs. time in occasions following administration of albuterol inpatient #5. Model parameters are listed in Table 2. The line is these eight subjects; 64% of these recorded FEV1computer-fitted, the symbols represent observed FEV1 vs. time

values were lower than the value recorded at zero time,pairs.

indicating albuterol evoked a negative bronchodilatorresponse in these patients at various times during the(5 and 12) (Fig. 2). Although the link and indirect12-h study period. The bronchodilator response toresponse models provided superior fits of the dataalbuterol concentrations as a function of time wasover the Emax and Sigmoid Emax models for patienterratic in three asthmatics (data not shown).12, both predicted values C50 of 42 and 41 ng/ml,

respectively. These values are at least four-fold higherthan those observed in other patients (Table 2), and Influence of �2 AR genotype on pharmacodynamicare probably unrealistic. Keo values for patients 5 modelingand 12 were 0.97 and 1.9 hours−1, respectively; Kin

The �2 AR genotype at position 16 of the 16 patientsand Kout for these patients were: 2.5 and 5.7 l/h and0.86 and 2.2 hours−1, respectively. in whom genetic analysis was performed are reported

Table 2 Pharmacodynamic model parameters, �2 adrenergic receptor genotype, race and gender

Patient Model1 Eo, L2 Emax C50 ng/ml N r Codon 16 Race GenderNumber Genotype4

1 E 2.08 2.33 0.785 1 0.88 NP B F2 E 2.83 4.98 1.99 1 0.93 NP W F3 E 2.66 3.28 2.14 1 0.97 NP B F4 SE 1.94 2.61 9.56 6.7 0.94 NP B M5 E∗ 2.7 4.03 2.75 1 0.96 NP B M6 SE 2.93 3.62 6.75 5.2 0.94 Arg W F7 L 3.35 0.98 NP B M8 3.32 NP W M9 SE 3.6 4.51 9.37 1.6 0.95 Arg B M10 SE 5.98 6.53 6.39 5.5 0.98 Arg W M11 3.15 Gly W F12 E∗ 2.64 3.64 42.0 1.4 0.91 H W F13 SE 2.65 2.84 6.43 12 0.97 Arg B F14 1.68 H B M15 SE 4.08 5.04 3.07 8.7 0.90 NP W M16 E 1.92 3.60 10.1 1 0.94 NP W M17 5.33 NP W M18 5.18 Gly W M19 SE 3.37 3.86 4.39 1 0.93 NP W F20 L 3.24 1.4 0.92 Gly B M21 SE 4.25 5.35 6.46 2.7 0.95 Arg B M22 L 3.40 0.81 Gly B F23 2.49 H W F24 2.40 NP B F25 2.67 NP B M26 4.35 H W M27 2.86 H W F28 2.10 Gly B F29 E 3.57 3.74 4.45 1 0.84 NP W M30 L 2.77 0.89 H B F

1E refers to Emax model, SE refers to Sigmoid Emax model, L refers to the Linear model;2Eo, model fitted resting FEV1; if no model was used, Eo is the observed value;3Slope refers to the slope of line describing FEV1 vs. albuterol concentration in the Linear model;4NP = not performed; H=heterozygotes.∗parameters obtained from fitting the link model

132 J. J. Lima et al

albuterol in the FEV1 response. C50 values were notdifferent between blacks and whites.

DISCUSSION

This study revealed several interesting and importantfindings with respect to the pharmacodynamic re-sponse to albuterol. First, we found that the �2 AR

4.812

5.6

Plasma albuterol concentrations (ng/ml)

FE

V1

(l)

0 2 6 10

5.0

4 8

5.4

5.2

genotype at codon 16 was an important determinantof the pharmacodynamic response to albuterol. Spe-Fig. 3 Clockwise Hysteresis. FEV1 vs. plasma albuterol

concentrations plotted in the order in which blood samples were cifically, all the patients who were Arg 16 homozygotesdrawn in patient #17. The arrow indicates the direction of the had predictable albuterol concentration vs. FEV1 re-hysteresis. The hand-drawn line connects consecutive FEV1 vs.

sponse relationships, while this was true in only 36%plasma albuterol concentration pairs.

of carriers of the Gly16 allele. Additionally, all patientsin whom a pharmacodynamic model could not befitted to the data (due to a poor albuterol con-centration-response relationship) carried the Gly16allele. These data suggest that knowledge of a patient’s�2 AR genotype may allow prediction of patients’in Table 2. Of this subpopulation who underwent

genetic analysis, 56% (nine of 16 asthmatics) had responses to albuterol. Specifically, Arg16 homo-zygotes are likely to have a response to albuterolpredictable concentration response relationships, and

thus their data could be described by traditional that is clearly related to drug concentration, whereascarriers of the Gly16 allele are much less likely topharmacodynamic models. As shown in Table 2, all

the patients who were Arg16 homozygotes had pre- exhibit a good or predictable response to �2-agonisttherapy. This speculation however, requires furtherdictable concentration-response data that could be

described by the Sigmoid Emax model. Among those study in a larger number of patients matched forgenotype, gender and race.in the Gly 16 cohort (heterozygotes and Gly16 homo-

zygotes), only 36% (4/11) had predictable con- The mechanism underlying the pharmacogeneticdifference in bronchodilator response to albuterol maycentration–response relationships and could be

modeled. Additionally, all those in whom there was be related to differences in �2 AR density. In-vitrostudies demonstrate the Gly 16 form of the receptor isno predictable concentration–response relationship

(seven asthmatics) carried a Gly16 allele. No re- more sensitive to agonist-promoted down-regulationthan the Arg 16 form.25,26 All our genotyped patientslationship was evident between albuterol-evoked

bronchodilation and �2 receptor genotype at codon with a poor response were carriers of the Gly 16 formof the receptor. Interestingly, in a majority of our27.patients with poor responses to albuterol, FEV1 de-creased (rather than increased) from baseline at vari-

Comparison of pharmacodynamic modeling andous times following albuterol administration. This

model parameters by race and genderresulted in a clock-wise hysteresis in the response vs.concentration relationship. The presence of a clock-Table 3 summarizes the pharmacodynamic model

parameters by gender and race for those asthmatics wise hysteresis implies the development of tolerance,a phenomenon commonly observed following con-whose data were fitted by the Emax or Sigmoid Emax

models. Only 15 of the 30 patients’ data could be tinuous �2 AR stimulation that is thought to bedue to agonist-induced receptor down-regulation.27described by one of these models, and neither race

nor gender influenced the likelihood of patients ex- Because the Gly 16 variant is more sensitive to agonist-induced receptor loss compared to the Arg 16 variant,hibiting a predictable concentration–relationship, de-

scribable by a pharmacodynamic model. No it is possible that the clockwise hysteresis observed inthese patients is due to their being more sensitive tosignificant differences by race or gender were noted

for Eo, Emax, or albuterol-induced change in FEV1, agonist-promoted receptor down regulation.�2 AR expression is also controlled by a 19-aminoalthough with a larger sample size, the model para-

meters Eo and Emax might achieve significance. The acid peptide called the 5′ leader cistron (5′ LC), thatregulates mRNA translation.28 This peptide is encodedmost intriguing finding in the modeled data with

respect to race and gender was the apparent gender by an open reading frame, which is 102 base pairsupstream of the �2 AR coding block. A single nuc-difference in C50. Specifically, the C50 value was nearly

double in males as compared to females (medians: 6.4 leotide polymorphism results in either Arg or Cysat position 19 of the peptide. �2 AR expression invs. 3.3), suggesting that females are more sensitive to

Factors affecting albuterol-evoked bronchodilation 133

Table 3 Comparison of mean ± SD and (median) parameters characterizing albuterol-evoked bronchodilation by gender and race.

GENDER n E0, L Emax, L C50, ng/ml Emax-E0, L Maximal %increase

Male 8 3.5 ± 1.3 4.4 ± 1.2 6.5 ± 2.9 0.92 ± 0.47 33 ± 25(3.6) (4.3) (6.4) (0.94) (26)

Female 7 2.7 ± 0.39 3.5 ± 0.84 3.7 ± 2.5∗† 0.77 ± 0.67 28 ± 25(2.7) (3.6) (3.3) (0.62) (23)

RACEBlack 7 2.8 ± 0.82 3.6 ± 1.1 5.4 ± 3.5 0.72 ± 0.42 25 ± 14

(2.7) (3.3) (6.4) (0.67) (25)White 8 3.4 ± 1.2 4.4 ± 1.1 5.3 ± 2.7† 0.96 ± 0.66 35 ± 31

(3.2) (3.8) (4.5) (0.83) (24)

∗P = 0.088, females vs. males†C50 from patient #12, 43 ng/ml, was omitted from the analysis

individuals bearing Cys (vs. Arg) was two-fold higher in women (which suggests increased receptor sensi-tivity) is also consistent with previous studies thatin airway smooth muscles obtained by rapid autopsy

from individuals without lung disease.28 Moreover, reported women were more sensitive to �-receptorstimulation than men.14–16 The mechanisms underlyinglinkage disequilibrium was detected, which resulted

in 100% of subjects who were homozygous Arg 19 at the gender difference in sensitivity to albuterol arenot clear, and may involve gender differences at the5′ LC being homozygous Gly 16 and Glu 27, while

most who were Cys 19 homozygotes were homozygous �2 receptor or at a post-receptor site.We did not observe any racial differences in responseArg16 and Gln27. Thus a possible explanation for

the greater response to albuterol observed in Arg 16 to albuterol based on either raw data or phar-macodynamic model parameters. This finding washomozygotes is that �2 AR density was higher in

these patients because they either have less agonist- somewhat surprising because compared to whites,asthma morbidity and mortality are higher,3–5 andpromoted receptor down-regulation or higher levels

of �2 AR expression. non-pulmonary �2-adrenergic receptor mediated re-sponses are lower in blacks.8–11 Additionally, the dis-The finding that 37% of our patients had a poor

response to albuterol is consistent with a recent study tribution of the �2 AR codon 27 polymorphism (andthus probably the 5′LC polymorphism) differs be-of 92 asthmatics with acute symptoms, which found

that a third failed to respond to aerosolized albuterol.7 tween blacks and whites.29 Thus, although racialdifferences in albuterol response were anticipated forIn responsive patients, albuterol doses of 2.5, 5 and

7.5 mg increased peak expiratory flow rate (PEFR) a number of reasons, such differences were not ob-served in our study. We believe it is unlikely the lackproportionately from baseline values of 41–73% of

normal. In contrast, PEFR in non-responsive patients of difference by race is due to inadequate sample sizeas our sample size was slightly larger than that ofwas blunted: albuterol maximally increased PEFR

from a baseline value of 31–38%. These patients also other studies that showed racial differences in non-pulmonary �2 AR-mediated responses.9,10 Thus wehad longer hospital stays. Although non-responsive

patients had more severe disease by history, and were conclude that blacks and whites have similar broncho-dilator responses to �2-agonist therapy.characterized by intense airway obstruction at pre-

sentation, these features were not sufficiently selective One limitation of our study is that we genotypedonly about half of our patients. Do our genetic findingsto be predictive.7 The results of our study suggest that

non-responders may carry a Gly 16 allele. Whether apply to our total patient population? Although wegenotyped about half our patients, the percent of theor not knowledge of genotype helps to identify patients

who fail to respond to �2 AR stimulation requires entire patient population who responded predictablyto albuterol (63%) was similar to the percent of geno-further study.

Our study did not suggest important racial or gender typed patients who had predictable responses (56%).The percent of the total population with a clockwisedifferences in the FEV1 response to albuterol. How-

ever, our ability to fully assess the gender differences hysteresis (27%) was similar to the percent of geno-typed patients who had a clockwise hysteresis (25%).in this response may have been confounded by the

fact that the percent predicted FEV1 at baseline was We therefore conclude that our genotyped patientsare reflective of our entire study population.significantly lower in males than females. Thus, al-

though we observed a larger absolute change in FEV1 In summary, our study demonstrates that �2 ARgenotype plays an important role in regulating thein males, this is likely explained by their lower percent

predicted FEV1 at baseline. The finding of lower C50 bronchodilator response to albuterol. The FEV1 vs.

134 J. J. Lima et al

14. Rahman A R, McDevitt D G, Struthers A D, Lipworth B J.albuterol concentration relationship was predictableSex differences in hypokalaemic and electrocardiographic

in all Arg 16 homozygotes. In contrast, two thirds of effects of inhaled terbutaline [see comments]. Thorax 1992;47: 1056–1059.asthmatics that carried the Gly 16 allele had no

15. Wheeldon N M, Newnham D M, Coutie W J, Peters J A,response or responded in a negative fashion to al-McDevitt D G, Lipworth B J. Influence of sex-steroid

buterol. Gender and race had little or no influence on hormones on the regulation of lymphocyte beta 2-adrenoceptors during the menstrual cycle. Br J Clinalbuterol-evoked FEV1.Pharmacol 1994; 37: 583–588.

16. Mills P J, Ziegler M G, Nelesen R A, Kennedy B P. Theeffects of the menstrual cycle, race, and gender on adrenergic

ACKNOWLEDGEMENTS receptors and agonists. Clin Pharmacol Ther 1996; 60:99–104.

17. Mohamad M, Lima J J, Eberle L V, Self T H, Johnson J A.This study was supported in part by a research grantEffects of gender and race on albuterol pharmacokinetics.

from Glaxo-Wellcome, Inc. (Research Triangle Park, Pharmacotherapy 1999; 19: 157–161.18. Lima J J, Mohamed M, Eberle L V, Self T H, Johnson J A.North Carolina, USA), by grant RR00211 and by

Impact of genetic polymorphisms of the �2-adrenergicgrants from the Center for Pediatric Pharmacokineticsreceptor on albuterol bronchodilator pharmacodynamics.

and Therapeutics (University of Tennessee, Memphis, Clin Pharmacol Ther 1999; 65: 519–525.19. ATS1991. Standardization of Spirometry, 1994 Update.Tennessee, USA).

American Thoracic Society. Am J Respir Crit Care Med1995; 152: 1107–1136.

20. Knudson R J, Schroter R C, Knudson D E, Sugihara S.REFERENCES Influence of airway geometry on expiratory flow limitation

and density dependence. Respir Physiol 1983; 52: 113–123.21. Dewar J C, Wilkinson J, Wheatley A, Thomas A, Thomas1. National Asthma Education and Prevention Program. Expert

N S, Doull I, Morton N. et al. The glutamine 27 beta2-Panel Report 2: Guidelines for the diagnosis andadrenoceptor polymorphism is associated with elevated IgEmanagement of asthma. April 1997. Bethesda, Maryland, USlevels in asthmatic families. J Allergy Clin Immunol 1997;Department of Health and Human Services, Public Health100: 261–265.Service, National Institutes of Health, National Heart, Lung,

22. Sheiner L B, Stanski D R, Vozeh S, Miller R D, Ham J.and Blood Institute, publication no. 97–4051.Simultaneous modeling of pharmacokinetics and2. Turkeltaub P C, Gergen P J. Prevalence of upper and lowerpharmacodynamics: application to d-tubocurarine. Clinrespiratory conditions in the US population by social andPharmacol Ther 1979; 25: 358–371.environmental factors: data from the second National Health

23. Jusko W J, Ko H C. Physiologic indirect response modelsand Nutrition Examination Survey, 1976–1980 (NHANEScharacterize diverse types of pharmacodynamic effects [seeII). Ann Allergy 1991; 67: 147–154.comments]. Clin Pharmacol Ther 1994; 56: 406–419.3. NHLBI Working Group. Respiratory diseases

24. Gabrielsson J, Weiner D. Pharmacodynamic concepts. In:disproportionately affecting minorities. Chest 1995; 108:Gabrielsson J, Weiner D, Swedish Pharmaceutical Society,1380–1392.eds Pharmacokinetic and pharmacodynamic data analysis:4. Arrighi H M. US asthma mortality: 1941–1989. Ann Allergyconcepts and applications. Stockholm: SwedishAsthma Immunol 1995; 74: 321–326.Pharmaceutical Press, 1997; 172–248.5. Sly R M, O’Donnell R. Stabilization of asthma mortality

25. Green S A, Turki J, Innis M, Liggett S B. Amino-terminal[see comments]. Ann Allergy Asthma Immunol 1997; 78:polymorphisms of the human beta 2-adrenergic receptor347–354.impart distinct agonist-promoted regulatory properties6. Asthma mortality and hospitalization among children and[published erratum appears in Biochemistry 1994 Nov 29;young adults – United States, 1980–1993. 1996; 45: 350–353.33(47):14368]. Biochemistry 1994; 33: 9414–9419.7. Strauss L, Hejal R, Galan G, Dixon L, McFadden E. R. Jr

26. Green S A, Turki J, Bejarano P, Hall I P, Liggett S B.Observations on the effects of aerosolized albuterol in acuteInfluence of beta 2-adrenergic receptor genotypes on signalasthma. Am J Respir Crit Care Med 1997; 155: 454–458.transduction in human airway smooth muscle cells. Am J8. Johnson J A. Racial differences in lymphocyte beta-receptorRespir Cell Mol Biol 1995; 13: 25–33.sensitivity to propranolol. Life Sci 1993; 53: 297–304.

27. Brodde O E, Brinkmann M, Schemuth R, O’Hara N, Daul9. Johnson J A, Burlew B S, Stiles R N. Racial differences inA. Terbutaline-induced desensitization of human lymphocytebeta-adrenoceptor-mediated responsiveness. J Cardiovascbeta 2- adrenoceptors. Accelerated restoration of beta-Pharmacol 1995; 25: 90–96.adrenoceptor responsiveness by prednisone and ketotifen. J10. Lang C C, Stein C M, Brown R M, Deegan R, Nelson R,Clin Invest 1985; 76: 1096–1101.He H B, Wood M, Wood A J. Attenuation of isoproterenol-

28. McGraw D W, Forbes S L, Kramer L A, Liggett S B.mediated vasodilatation in blacks [see comments]. N Engl JPolymorphisms of the 5′ leader cistron of the human beta2-Med 1995; 333: 155–160.adrenergic receptor regulate receptor expression. J Clin11. Watkins L L, Dimsdale J E, Ziegler M G. Reduced beta 2-Invest 1998; 102: 1927–1932.receptor mediated vasodilation in African Americans. Life

29. Xie H G, Kim R B, Stein C M, Wood A J J. Distribution ofSci 1995; 57: 1411–1416.genetic polymorphisms of �2-adrenoceptor gene between12. Osborne M L, Vollmer W M, Linton K L, Buist A S.black and white populations. Clin Pharmacol Ther 1998; 63:Characteristics of patients with asthma within a large HMO:156–156.a comparison by age and gender. Am J Respir Crit Care

Med 1998; 157: 123–128.13. Prescott E, Lange P, Vestbo J. Effect of gender on hospital

admissions for asthma and prevalence of self-reportedasthma: a prospective study based on a sample of the

Date received: 26 October 1999.general population. Copenhagen City Heart Study Group.Thorax 1997; 52: 287–289. Date accepted: 12 May 2000.