Asthma affects around 6-31% of Indian children [1]. Short-acting beta-2 agonists (SABA) form the mainstay of treatment of acute exacerbations. However, a significant heterogeneity exists in response to inhaled SABA, of which 70-80% may have a genetic basis [2]. SABA act by binding to the beta 2 adrenergic receptor (ADRB2), coded by an intron-less gene on chromosome 5. A single nucleotide polymorphism (SNP) at nucleotide position 46 of this gene (AGA to GGA) substitutes the 16th amino acid of the translated protein chain from arginine to glycine and alters receptor function. This p.Arg16Gly polymorphism has been the subject of intense research. Conflicting data exist regarding effect of this polymorphism on salbutamol response particularly with regard to different ethnicities. There is a paucity of Indian data, especially in children. Thus, our first objective was to determine an association between Salbutamol response and p.Arg16Gly polymorphism of ADRB2 gene in children with asthma with acute exacerbation. The second was to determine prevalence of these genotypes in the asthmatic and non-asthmatic population.

Clinical details and CAS were recorded in a pre-designed structured proforma. All enrolled children underwent spirometry using a portable spirometer (Superspiro MK2 Micro Medical Ltd, UK),　as per standard technique. The　absolute and percentage predicted values　of following parameters were recorded: FEV1　(Forced Expiratory Volume-1 second); FVC (Forced Vital Capacity) PEFR (Peak Expiratory Flow Rate) FEF50　(Forced Expiratory Flow 50). Thereafter, they received salbutamol 100 µg 2 puffs with MDI (metered dose inhaler) with spacer every 10 min for a total of three doses followed immediately by repeat spirometry. The highest FEV which was available from all adequate curves produced during baseline and post treatment spirometry were recorded. Percentage increase in actual FEV was measured using the formula [(FEV (post-bronchodilator) – FEV baseline)/FEV baseline] X 100. Positive bronchodilator response was considered if percentage increase in FEV1 was ³15%.

For DNA analysis, 3-5 mL blood was collected in EDTA vacutainers and stored at 4ºC. DNA was extracted using the phenol chloroform method [7]. A 168 bp region flanking the p.Arg16Gly polymorphism region was amplified using Polymerase Chain reaction (PCR). The forward primer: 5' GCC TTC TTG CTG GCA CCC CAT 3' (21 bases) and reverse primer: 5' CAG ACG CTC GAA CTT GGC CAT G 3' (22 bases) were used. PCR reactions were carried out in 25 µL mix containing 2.5 µL of 10X PCR buffer, 2.5 µL of 2 mM dNTPs, 1.0 µL of 10 µM forward and reverse primer each, 0.25 µL of 3 U/µL Taq polymerase, 16.75 µL of sterile water and 1 µL of extracted DNA (100 ng/mL). The above mix was kept in a thermal cycler with set temperature conditions repeated for 35 cycles. Initial denaturation was done at 94ºC for 2 minutes followed by denaturation at 94ºC for 40 seconds, annealing at 64ºC for 40 seconds, extension at 72ºC for 50 seconds and final extension at 72ºC for 5 minutes. A blank, with all components except the DNA template, was run simultaneously with each run of PCR as control. The PCR products were checked for adequate DNA amplification by a run at 150 volts in a horizontal gel electrophoresis system. The underlined bases in both the forward and reverse primers were modified from A to C to create restriction sites for the NcoI enzyme which was later used in restriction digestion of the PCR products. The forward primer creates an NcoI restriction site on Gly16 PCR product but not on Arg16 PCR product. The reverse primer creates a restriction site on both, thus serving as a control to assess whether restriction digestion was complete.

Data were collected in a structured proforma and managed using MS Excel software. Statistical analysis was done using STATA 11. The comparison of salbutamol responsiveness of children with different ADRB2 genotypes (AA/AG/GG) was done using the Pearson Chi square test. P value less than 0.05 was considered statistically significant. Hardy-Weinberg equilibrium analysis of ADRB2 SNP at 46th nucleotide position was also performed [8].

Results

During the study period, 246 children were screened for inclusion in to the study. Of these, 120 results were finally available for analysis (Fig. 1). The baseline characteristics of the study population are tabulated in Table I.

Fig. 1 Study flow for screening, enrolment and stratification into responders and non-responders.

Of the analyzed 120 children with asthma, 42.5% were labelled as responders. Comparison of responders and non-responders shows that they were are ell-matched in characteristics which could have confounded the response to salbutamol (Table I).

TABLE I Baseline Charactersitics of Responders Versus Non Responders 

Variable	Responders	Non-	P
	(n=51)	responders	value
		(n=69)
Age; mean (SD), y	9.3 (2.12)	9.1 (2.37)	0.76
Female gender, n (%)	11 (21.6)	18 (26.1)	0.56
Height; mean (SD), cm	131.3 (13.1)	131.0 (15.3)	0.90
Weight; mean (SD), kg	27.08 (8.47)	26.89 (10.1)	0.91
Age at onset; mean (SD), mo	46.11 (36.6)	55.95 (42.4)	0.19
Severity asthma, n (%)			0.85
Intermittent	06 (11.7 )	09 (13.0)
Mild persistent	26 (50.9 )	30 (43.5)
Moderate persistent	18 (35.3 )	29 (42.0)
Severe persistent	01 (01.9 )	01 (01.4)
Family history of allergy, n (%)	42 (60.9 )	35 (68.6)	0.38

No significant association was found between the ADRB2 genotypes and salbutamol-responsiveness (P= 0.55) (Table II). However, a trend towards greater bronchodilator responsiveness was seen amongst those who were carrying the A/A polymorphism as compared to G/G homozygotes. The median of change in percentage predicted FEV1 in the three ADRB2 genotypes was 14.5 for A/A, 12.5 for A/G and 7.7 for G/G genotypes, respectively.

TABLE II	ADRB2 Genotype and Bronchodilator Response in the Study Population

ADRB2 Genotype	Responders	Non-responders,
	no. (%), n = 51	no. (%), n = 69
A/A	14 (27.4 )	15 (21.7 )
A/G	27 (52.9 )	35 (50.7 )
G/G	10 (19.6 )	19 (27.5 )
P values for comparisons between A/A vs G/G, A/G vsG/G, A/A vs A/G and G/G vs A/A + A/G were all >0.05.

No significant association was found between the studied ADRB2 genotypes and the severity of asthma (P= 0.39) or family history of asthma (P=0.25). In the 120 children with asthma, the frequency of AA, GG, AG genotypes had prevalence of 24.2%, 24.2% and 51.7% respectively and were in equilibrium as per the Hardy Weinberg law (c2= 0.13).

In this observational study done amongst children with asthma aged 5 to 15 years with acute exacerbation of asthma, no association was found between the ADRB2 genotype (Arg16Gly) and salbutamol-responsiveness defined as the percentage change in FEV1 ³15%.

Green, et al. [9] suggested that several SNPs of the ADRB2 gene significantly alter ADRB2 receptor down regulation. Unfortunately, the multitude of studies subsequently done have shown vastly discordant results. Early studies by Martinez, et al. [3] showed better bronchodilator response in asthmatics with A/A genotype of ADRB2 compared to the G/G genotype (P=0.05). Later, similar results were seen in several studies [10-12]. A study by Choudhary, et al. [13] showed better salbuta-mol response in those with AA genotype in Puerto Ricans but not in Mexican highlighting ethnic differences. In contrast a few studies have shown absolutely opposite results [14]. The only Indian study done showed a better salbutamol response in those possessing the G/G ADRB2 genotype [15]. Some larger studies have reported lack of any such association [16-20].

This study is one of the few studies which have researched the response to inhaled SABA during an acute exacerbation of asthma. Most other studies have focused on cohorts of stable asthma patients [3,10-12,15,16,18]. This has direct implications for finding the best personalized treatment for acute asthma attacks that are responsible for hospitalization and mortality. Percentage change in actual FEV was chosen as the main outcome measure as it is the most objective and immediate outcome. It has been taken as the study end-point in several studies, making the comparison of results easier. A suitably high cut-off of ³15% was taken as a meaningful response to rule out any measurement variability [21]. Though no significant association could be established in the current study, a trend towards a greater positive bronchodilator response amongst the Arg16Arg homozygotes has been seen. A larger sample size could prove or disprove such an association. Importantly, the results of our study have not shown concordance with the only other Indian study on the subject [15], though it involved a Southern Indian population. This ethnic difference of the populations enrolled may explain the variability in the results obtained, highlighting the need of conducting these studies in various ethnic groups.