A total of 30 OA and 52 LC patients, together with 26 healthy volunteers participated in the study. Blood samples were drawn from anterocubital vein in the morning.

Protocol was approved by the Clinical Research Ethics Committee of Alicante Department of Health, General Hospital (Alicante, Spain). All participants signed informed consent before enrolment, and the study was performed according to the Declaration of Helsinki.

Patients from Primary Care of Alicante General Hospital Department of Health with knee OA were included in this study.

The inclusion criteria were as follows: Ages ranging from 50 to 80 years; diagnosis of knee OA according to the criteria established by the American College of Rheumatology (ACR) using history, physical examination and radiographic findings, knee X-rays in the last 12 months and a Kellgren-Lawrence (KL) OA grade of 2 or more, based on the radiological severity (grade 1, questionable narrowing of joint space and possible osteophytic lipping; grade 2, definite osteophytes and possible narrowing of joint space; grade 3, moderate multiple osteophytes, definite narrowing of joints space, some sclerosis, and possible deformity of bone contour; and, grade 4, large osteophytes, marked narrowing of joint space, severe sclerosis, and definite deformity of bone contour) (40).

Patients with infections, inflammatory diseases, malignancies or patients using α- or β-adrenergic receptor agonists or antagonists were excluded from the study.

Patients from Liver Unit of Alicante Department of Health, General Hospital with LC were included in this study.

The inclusion criteria were as follows: Ages ranging from 40 to 80 years; diagnosis of cirrhosis according to the criteria established by Spanish Association for the Study of the Liver (AEEH) and the Biomedical Research Network Center for Liver and Digestive Diseases (CIBERehd) (41): Either liver biopsy or unequivocal clinical, imaging and biochemical findings (e.g., complete blood cell count, serum chemistries, liver function test, and coagulation studies).

None of the patients had an established transjugular intrahepatic portosystemic shunt (TIPS). Patients with systolic blood pressure <100 mmHg bradycardia (heat rate <50/min), obstructive airway disease or other contraindications for treatment with propranolol were excluded from the study.

Patients were classified in two groups: Under prophylaxis or treatment of first upper gastrointestinal bleeding episode (‘primary prophylaxis’) or prophylaxis of recurrent bleeding (‘secondary prophylaxis’). As non-bleeding varices are generally asymptomatic, high hepatic venous pressure gradient (HVPG) is the clinical gold standard for risk of formation prediction (>12 mmHg) or rebleeding (HVPG >20 mmHg) and to predict the response to β blocker antagonists during treatment of portal hypertension (>20% fall from baseline of portal pressure). Hence HVPG clinical routinely monitoring allow the identification of patients who will be effectively protected against the risk of bleeding (labeled as ‘responder’) and those who, by not achieving a HVPG reduction by 20% of baseline or to ≤12 mmHg, present a very high risk of bleeding (‘non-responders’). LC patients clinical data is presented in Table I.

In patients with previous variceal bleeding, investigations were made at least 7 days after the complete recovery of bleeding.

A total of 26 healthy controls were recruited through blood donors from the same geographical areas as patients, and were matched to patients according ethnicity (at least 2 generations from the same area). None of the subjects in the healthy control group had any clinically significant abnormality shown by routine history, physical examination, or laboratory tests.

PBMC were isolated from EDTA venous blood by Percoll density gradient centrifugation. Activation of β2AR leading to an increase in the intracellular level of cyclic adenosine monophosphate (cAMP, pmol/ml/10⁶ cells) by increasing adenylate-cyclase (AC) activity was evaluated in duplicates by cAMP determination using a competitive Enzyme Immunoassay (EIA) according to laboratory procedures and manufacturer guidelines (Cayman Chemical Company, Ann Arbor, MI, USA).

Cells were counted in a Coulter Counter and with a Neubauer chamber. Viability was determined by trypan blue exclusion and ranged between 94–98%. Cells were incubated as described by Sondergaard et al (42). Aliquots of 1×10⁶ PBMC were rinsed with Phosphate-buffered saline (1× PBS, pH 7.4, containing 135 mM NaCl, 5.4 mM KCl, 1.4 mM MgSO₄, 1.4 mM CaCl₂, 0.18 mM sodium phosphate) and 3% (wt/vol) of bovine serum albumin (BSA) fraction V (Sigma-Aldrich; Merck KGaA, Darmstadt, Germany) (PBS-BSA) at room temperature. Phosphodiesterases were inhibited by preincubation at 37°C for 30 min with 1 mM isobutyl-methyl-xanthine (IBMX) (Sigma-Aldrich; Merck KGaA). Then, cells were stimulated with different concentrations of isoproterenol (ISO) ranging from 10⁻⁹ to 10⁻⁴ mM or vehicle for 15 min. Stimulation was stopped at 100°C water bath for 3 min. Samples were centrifuged for 5 min at 3500 rpm and disrupted by sonication 3×15 sec (model SM25E-MT; Branson Ultrasonics Corporation, Geneva, Switzerland). Lysates were immediately frozen and stored at −80°C until EIA analysis was performed as described previously.

On the day of the assay, samples were thawed at room temperature for 25–30 min and centrifuged at 4°C at 3,500 rpm for 10 min to remove insoluble material. cAMP was measured in the supernatant using a EIA. cAMP increase was calculated by subtraction of values determined in IBMX preincubated samples. All assays were performed in duplicate.

Blood samples from OA and LC patients and healthy controls were collected and genomic DNA was extracted from isolated PBMC using QIAamp^® DNA Midi Kit according to manufacturer's instructions.

Genomic DNA was genotyped for SNPs within the β2AR gene locus. Three SNPs with high frequency of polymorphism in the human population (>20% prevalence) and located within the coding region for gene were chosen. The known functional SNP rs1042713, rs1042714 and rs1800888 are well-studied common non-synonymous SNPs (43,44).

β2AR genotype at positions 16, 27 and 164 was determined by polymerase chain reaction (PCR) and sequenced by Thermosequenase Primer Cycle kit (Amersham Pharmacia Biotechnology, Piscataway, NJ, USA). PCR was performed in a final volume of 25 µl containing 100–200 ng DNA, 0.2 mM of each dNTP, 1X de reaction buffer (50 mM KCl and 20 mM Tris-HCl, pH 8.4), 1.5 mM MgCl₂, 1 U Taq DNA polimerase, 10% DMSO and 200 nM of each primer. Temperature cycling was 94°C for 30 sec, 61°C for 45 sec, and 72°C for 45 sec for 30 cycles. In total, 10 µl of the PCR products were visualized on a 1% agarose gel stained with ethidium bromide. Computer analysis of all SNP combinations in the human EST database (dbEST release 030405, 6,053,112 human entries) was performed using BLAST (National Library of Medicine, Bethesda, MD, USA). Complete nucleotide sequence of β2AR gene was used (NM 000024) to design the primers (Primer 3; UCSD, San Diego, CA, USA).

To determine whether cAMP concentration changes in PBMC from patients was associated with variations in immunodetectable β2AR protein, quantitative Western blotting analysis was performed.

After treatment with isoproterenol, cells were lysed and protein was extracted using RIPA buffer (50 mM Tris-HCl pH 7.4; 150 mM NaCl, 1 mM EDTA, 0.25% Na-deoxycholate, 1% NP-40, 1 mM PMSF, 1 mM sodium orthovanadate, 1 µg/ml leupeptin, 1 µg/ml aprotinin, and 1 µg/ml pepstatin). Protein concentration was determined by BCA assay and samples were separated on 12% SDS-PAGE gels, and transferred onto Hybond-enhanced chemiluminescence (ECL) nitrocellulose membranes. Membranes were probed with antibodies against β2AR (R11E1, sc-81577; Santa Cruz Biotechnology, Inc., Dallas, TX, USA) of human origin. Protein bands were observed using ECL and specific bands were detected with X-film according to procedures of Proteomics and Genomics Division, Research Technical Facility of University of Alicante (Alicante, Spain).

Spot detection and quantification was performed using Progenesis Same Spots software according to manufacturer's instructions (Nonlinear USA,. Inc., Durham, NC, USA). Two individual gel replicates from each subject were used for the analysis. Results were expressed as relative arbitrary units (AU) according to procedure previously described (45), using as standard samples from 4 healthy subjects. GAPDH was used as endogenous control.

Quantitative data is expressed as mean ± standard deviation (SD). Differences between groups were analysed using the T-Student or non-parametric Mann-Whitney U-test according to normality. Qualitative variables are expressed as frequency or percentage and differences between groups were evaluated using χ² test. A comparison of independent single variables between the groups was calculated by one way analysis of variance (ANOVA) followed by Turkey's procedure. When normality test failed, Kruskal-Wallis one-way ANOVA on ranks was used. A two-tailed P<0.05 was considered to indicate a statistically significant difference.

Thirty OA patients (age, 70±8 years; female, 78%; median (min, max) OA duration, 6 (1–29) years) were included in this study. A total of 12% were diabetic, 14% hypertensive, 8% dislipemic and 9% BMI >30 kg/m². Mainly category of KL radiological severity was 3–4 (90%) grades and 15 (54%) patients required knee joint replacement (76% KL grade 4). Regular current use of analgesic (acetaminophen, dipyrone), and non-steroidal anti-inflammatory drugs (NSAIDs: Ibuprofen, naproxen or celecoxib) was very common (21 and 50%, respectively), followed by chondroitin sulphate (12%), glucosamine (6%) and tramadol (2%).

A total of 52 LC patients (age, 54±11 years; females, 13.5%) were submitted to the hospital clinic for HVPG determination and were included in the study. Alcohol was actively consumed by 11.5% of patients. Scores for Model for End-stage Liver Disease (MELD) were 11±4 and Child-Pugh 7±2, respectively. In total, 37% patients were previously treated with beta-blockers and 50% exhibited clinical response to propranolol (46).

A total of 26 healthy human volunteers (aged 59±13 years; female 50%) participated in the present study. They had normal blood cell counts, normal liver and kidney function test results, and normal findings on physical examination.

PBMC from control group stimulated with different concentrations of isoproterenol (10⁻⁹ to 10⁻⁴ mM), showed a significant increase in cAMP production in a dose-dependent manner with a maximum response between 10⁻⁵ to 10⁻⁴ mM (Fig. 1A and B). Isoproterenol induced a significantly lower increase in cAMP concentration in OA and LC patients (44±28 and 14±15.5 pmol/ml/10⁶ cells respectively) vs. controls (90±66 pmol/ml/10⁶ cells, P<0001) at isoproterenol 10⁻⁵ mM stimulus (Fig. 1A and B, respectively).

OA patients with severity KL grade 4 (n=17) presented a smaller response than lower KL severity grades (n=6) (38±21 and 61±42 pmol/ml/10⁶ cells at isoproterenol 10⁻⁵ mM stimulus, respectively, P=0,06) (Fig. 1A). We observed a relevant (67%) and significant reduction in cAMP increase in KL grade 4 with knee replacement patients (n=12) compared with non-surgery patients (n=5) (30±14 vs. 52±23 pmol/ml/10⁶ cells, respectively, P=0046).

Cirrhotic patients have a significant decrease in β2AR-mediated AC activity stimulation, similar in patients with primary or secondary prophylaxis and in responder or non-responder cirrhotic patients (Fig. 1B).

Table II shows the genotypes frequencies, alleles, and carrier states at amino acid positions 16, 27 and 164 of the β2AR gene.

The distribution of expected and observed frequencies of the different genotypes at the different amino acid positions followed the Hardy-Weinberg equilibrium both in patients and controls. Minor allele frequencies in our Caucasian Spanish sample from Alicante Department of Health, General Hospital were Gly16 (allele G, 0, 24-0, 44), Glu27 (allele G, 0, 19-0, 38) and Ile164 (allele T, 0, 00-0, 15).

Carriage of Arg16, Gln27 and Ile164 was more prevalent in controls, OA and LC patients. Even though we observed some differences in allele frequency i.e., a higher frequency of G allele (Gly16) in LC patients, especially in responders; however, none of these differences were statistically significant probably due to the reduced sample size.

In global, subjects with any of the β2AR SNPs analyzed shown a non-significant decrease in cAMP increase (mean ± SD) at isoproterenol 10⁻⁵ mM vs. wild type (WT) (Fig. 2): In controls (94±70 vs. 85±64 pmol/ml/10⁶ cells, P=0.753), OA (52±31 vs. 34±19 pmol/ml/10⁶ cells, P=0.054) and LC patients (15±17 vs. 13±16 pmol/ml/10⁶ cells, P=0.718) (Fig. 3A). The prevalence of the different genotypes did not differed between patients according to any clinical variable in both models of chronic inflammation (Fig. 3B and C). Combination of Arg16⁺-Gln27⁺ shown a decreased cAMP increase stimulus in controls (n=5), OA (n=13) and cirrhotic (n=6) (50±32, 38±21 and 14±19 pmol/ml/10⁶ cells, respectively, P=0.094) (data not shown). The bigger decrease AC stimulus was evidenced for the Arg16⁺-Gln27⁺ in controls (48% β2AR agonism stimulus blocked) and for Arg16⁻-Gln27⁺ in OA (51% agonism stimulus blocked). In cirrhotic (13% agonism stimulus blocked) the stimulus was the same for both Arg16⁻-Gln27⁺ or Arg16⁺-Gln27⁺.

Western blot detection of β2AR showed a decreased expression above 0.411-fold for OA and 0.845-fold for LC compared to healthy controls (data not shown).