Association between the Ile164 β2 Adrenergic Receptor Polymorphism and Fatigue in Patients with Rheumatoid Arthritis

Abstract

Introduction

In the present work, the frequency of inherited polymorphisms of the beta 2 adrenergic receptor (β2AR) gene and their association with fatigue in patients with rheumatoid arthritis (RA) was examined.

Methods

An allele-specific polymerase chain reaction was used to determine the common variants of the β2AR at position 16, 27, and 164 in 92 German RA outpatients. Health Assessment Questionnaire (HAQ-DI), Beck Depression Inventory (BDI), Perceived Stress Questionnaire (PSQ-30), Multidimensional Fatigue Inventory (MFI-20) were utilized.

Results

34.7% of German RA patients were diagnosed with associated fatigue. Fatigued patients were more likely to carry the Ile allele at position 164 (OR 7.33, 95% CI 1.09–59.8, p = 0.049). Comparing these risk factors’ contribution to different fatigue dimensions revealed that Ile164 carriers only had significantly higher MFI-20 mean values for general fatigue (p = 0.014) while the clinical difference among other MFI subscales was the largest for mental fatigue (carrier: 8.23, SD: 4.22, noncarrier: 5.67, SD: 1.56, p = 0.089, Cohen’s d = 0.629). Disease activity, perceived stress, and depression were also associated with fatigue with higher mean values for DAS28CRP (p = 0.038), PSQ (p < 0.001), and BDI-II (p < 0.001) in fatigued patients. Physical fatigue was correlated with disease activity (p = 0.009) and depression (p = 0.001) while mental fatigue showed associations with depression (p = 0.001) and perceived stress (p = 0.028).

Conclusion

The discovery study indicates that the Ile164 polymorphism might in contrast to other β2AR polymorphisms affect fatigue levels in RA patients. This association was observed especially with mental fatigue. Further replication studies are warranted to determine further role of β2AR polymorphisms in RA patients.

Introduction

Chronic fatigue is one of the major problems for many patients with rheumatoid arthritis (RA), often severely affecting quality of life. Different dimensions of fatigue, like physical fatigue, reduced activity, reduced motivation, and mental fatigue, as included in the Multidimensional Fatigue Inventory (MFI-20) [1], make it difficult to identify comprehensive pathological contributors as they may differ for each dimension. Furthermore, a clear concept for the background of fatigue has yet to be established.

Multiple variables like inflammation [2], altered neurotransmission and brain activity [2, 3], or hypoactivity of the hypothalamic-pituitary-adrenal axis [4] are likely contributing factors to fatigue. It has also been revealed that chronically fatigued patients show alterations in autonomic regulation [4, 5]. This may be a consequence of altered β-adrenergic receptor (βAR) responsiveness. In the Caucasian American population, a reduced responsiveness could be detected within patients showing symptoms of fatigue and depression [6]. This may be caused by prolonged sympathetic overactivity, as different studies suggest [6–8]. However, reduced βAR responsiveness itself can also contribute to symptoms like fatigue and depression [4]. One possible influence on mechanisms of altered βAR responsiveness is receptor polymorphisms. Especially, β2AR polymorphisms have been reported to be implicated in various diseases, including RA [9–11], and may also play a role in chronic fatigue syndrome (CFS) [12]. The Arg16 receptor polymorphism has been shown to be associated with RA as well as with a younger mean disease onset age [10] in German patients. Another study from northern Sweden additionally reported an association of the Gln27 polymorphism with RA [11]. Of interest, many cells of the immune system express β2AR indicating a close interplay between the autonomic nervous system and the immune system. Additionally, the Gln27 polymorphism was reported to be associated with CFS [12]. As β2AR polymorphisms can lead to alterations in ligand affinity or downregulation [13], they might be implicated in symptoms like fatigue or depression in RA patients.

There are a total of 9 polymorphisms of the β2AR gene, 4 of which lead to a change in the amino acid sequence [13]. Known effects of this structural change include an increased downregulation of the Gly16 receptor compared to the Arg16 variant [13, 14] and less downregulation of the Glu27 compared to the Gly27 variant [15]. However, most of these findings were observed for smooth airway muscle cells. Despite being rare, polymorphisms at position 164 could also potentially be of interest as transfected cells expressing the Ile164 receptor showed less ligand affinity [16]. Data on the 164 polymorphisms, however, are very limited. To reveal a possible connection between fatigue and polymorphisms of the β2AR, we studied polymorphisms in fatigued and nonfatigued RA patients in Germany while evaluating different dimensions of fatigue with questionnaires.

Materials and Methods

Study Groups

DNA from 92 German patients diagnosed with RA who regularly visit the outpatient department for rheumatology of the University Hospital Leipzig was studied for the frequency of β2AR polymorphisms. Patients’ clinical characteristics are shown in Table 1. The same patients were sorted into groups of either fatigued or nonfatigued according to their item scores for general fatigue on the Multidimensional Fatigue Inventory (MFI-20) questionnaire [1]. Age- and sex-dependent cutoff scores were used according to Singer et al. [17] with 75% of the general population in the respective age and gender group being categorized as nonfatigued, based on a German community sample (n = 2,037) by Schwarz et al. [18]. This method limits confounding of the results by age and gender effects as older people and women generally experience higher levels of fatigue [17]. The cutoff scores used are shown in Table 2.

Table 1.

Characteristics of patients with RA

	Patients, n (total, n = 92)
	n	%
Female/male	72/20	21.7/78.3
Mean (SD) age of disease onset, years	52.2 (12.8)
Mean (SD) duration of disease, years	18.8 (8.1)
Rheumatoid factor positive	65	73
Anti-CCP antibodies positive	67	75
Definitive erosions	47	51.1
Mean (SD) CRP	8.26 (16.19)
Mean (SD) DAS-CRP	3.06 (1.09)
Mean (SD) HAQ	0.95 (0.72)
DMARDS	78	84.8
Biologic therapy	18	19.6

Table 2.

MFI-20 cutoff scores

Age-group	≤39	40–59	≥60
Men	≤8	≤10	≤13
Women	≤10	≤11	≤13

Classification from Singer et al. [17] and Schwarz et al. [18].

Determination of β2AR Polymorphisms

DNA Isolation

Genomic DNA was extracted from EDTA-blood leukocytes by the use of standard proteinase K digestion and phenol-chloroform method.

PCR

The three most common polymorphisms of β2ARs at codon position 16, 27, and 164 were studied. Polymorphisms in these positions result in modified amino acid sequences as shown below:

• •codon 16/nucleotide 46: Arg or Gly
• •codon 27/nucleotide 79: Gln or Glu
• •codon 164/nucleotide 491: Thr or Ile

An allele-specific PCR was performed for each polymorphism [10] with a total volume of 50 µL containing 100 µM of each dNTP, 10 pmol of each primer, 0.5 U Taq polymerase (Applied Biosystems, NJ, USA), 0.5 M betaine (Sigma Life Science, St. Lewis, Missouri, USA), and 1 µL genomic DNA. Depending on the primers, 2.5 mM MgCl2 was added for codon 16 and 5 mM MgCl2 for codon 27 and 164.

All PCR reactions started with denaturation at 94°C for 5 min and ended with final extension at 72°C for 5 min (PCR cycler: Eppendorf-Netheler-Hinz GmbH, Hamburg). Cycles in between were specific for each primer: codon 16: the primer pairs were 5′-CTT​CTT​GCT​GGC​ACC​CAA​TA-3′ (sense) for Arg or 5′-CTT​CTT​GCT​GGC​ACC​CAA​TG-3′ (sense) for Gly and 5′-CCA​ATT​TAG​GAG​GAT​GTA​AAC​TTC-3′ (antisense) for Arg and Gly. Temperature cycling was 94°C for 45 s, 59°C for 45 s, and 72°C for 60 s for 35 cycles for Arg and 94°C for 45 s, 60°C for 45 s, and 72°C for 60 s for 34 cycles for Gly. The generated product had a size of 913 bp. Codon 27: the primer pairs were 5′-GGA​CCA​CGA​CGT​CAC​GCA​GC-3′ (sense) for Gln or 5′-GCA​CCA​CGA​CGT​CAC​GCA​GG-3′ (sense) for Glu and 5′-ACA​ATC​CAC​ACC​ATC​AGA​AT-3′ (antisense) for Gln and Glu. Temperature cycling was 94°C for 45 s, 62°C for 45 s, and 72°C for 45 s for 35 cycles. The generated product size was 442 bp. Codon 164: the primer pairs were 5′-TGG​ATT​GTG​TCA​GGC​CTT​AC-3′ (sense) for Thr or 5′-TGG​ATT​GTG​TCA​GGC​CTT​AT-3′ (sense) for Ile and 5′-CAC​AGC​AGT​TTT​CTT​T-3′ (antisense) for Thr and Ile. Temperature cycling was 94°C for 30 s, 59°C for 45 s, and 72°C for 45 s for 29 cycles. The PCR product size was 662 bp.

Separation of PCR Products/Gel Electrophoresis

15 µL of PCR product was stained with GelRed Nucleic Acid Gel Stain (Biotium, Hayward, CA, USA) and separated by size on 2% agarose gel for 25 min at 200 V and 400 mA. Finally, the PCR product was visualized under UV light, photographed, analyzed, and documented (Biozym Scientific GmbH, Oldenburg, Germany).

Clinical Assessment

Disease Activity

We evaluated disease activity with DAS 28, HAQ, anti-CCP antibodies, and Larsen score for radiographs. The Disease Activity Score was calculated with the online DAS28CRP calculator by Flendrie et al. [19]. Values <2.6 indicate remission, 2.6–3.2 low, >3.2–5.1 moderate, and >5.1 high activity [20]. Disability was evaluated with the HAQ for RA [21]. A cutoff value of 1 for determining disability was chosen according to Sokka et al. [22]. We determined anti-cyclic citrullinated peptide antibodies with ELISA (Euroimmun, Germany, cutoff value 5 RE/mL). Hand and wrist radiographs were analyzed for the presence of definitive erosions.

Questionnaires

The Multidimensional Fatigue Inventory (MFI-20) self-report questionnaire was used to evaluate fatigue. It is a tool designed and validated to measure fatigue [1] consisting of 20 items forming 5 scales with 4 items each. The scales include general fatigue, physical fatigue, reduced activity, reduced motivation, and mental fatigue. Each item is scored with 1–5 points making up a total score of 4–20 for each scale with a higher score indicating higher levels of fatigue. For classification of the study groups, cutoff values for the general fatigue scale were used as described earlier.

Due to fatigue being a symptom of fibromyalgia [23], we investigated the prevalence of fibromyalgia in the two study groups as well as a possible connection to β2AR polymorphisms. A questionnaire consisting of the widespread pain index (WPI) and the symptom severity score (SSS) according to the 2010 ACR diagnostic criteria for fibromyalgia [24] (WPI ≥7 and SSS ≥5 or a WPI of 3–6 and SSS ≥9) was used to identify patients with fibromyalgia.

Since fatigue is a common symptom of depression, and psychological factors as well as neural mechanisms may contribute to mental fatigue [3], we investigated depression with the Beck Depression Inventory II (BDI-II) [25]. A total score of 0–63 is calculated by adding up 21 items each scored from 0 to 3 points. A score of 9–13 suggests minimal, 14–19 mild, 20–28 moderate, and 29–63 severe depressive symptoms [25]. We evaluated stress levels with the Perceived Stress Questionnaire (PSQ) [26] because of an association of fatigue with perceived stress [27]. The questionnaire includes 30 items referring to the last 30 days. Each item is answered on a scale from 1 to 4 points. The total score is transformed to an index value between 0 and 1 (total score: 30/90) according to Levenstein et al. [26]. For statistical analysis, patients were classified in low (≤ mean + SD), moderate (> mean + SD, ≤ mean + 2SD), and high (> mean + 2SD) level of stress according to Bergdahl et al. [28].

Statistical Analysis

Categorial clinical data as well as prevalence of β2AR polymorphisms were compared between the study groups using χ² test or Fisher’s exact test if necessary. Odds ratios were calculated with 95% confidence intervals. For comparison of mean values, normal distribution of scale variables was first investigated visually and by using the Kolmogorov-Smirnov test. Normally distributed variables were then compared with t test for independent samples after conducting Levene’s test for equality of variances. Mann-Whitney U test was used for non-normally distributed variables. A p value of <0.05 was considered significant in all tests. Correlation values (Phi and Eta) were calculated depending on scale of the variables. A binary logistic regression was performed with clinical risk factors as independent variables. For some patients, genetic data were missing or invalid at a certain codon position. This led to calculations with a lower total number than 92 patients in some cases. SPSS (SPSS Inc, Chicago, IL, USA, Build 1.0.0.1327) was used for all statistical calculations.

Results

Thirty-two patients (34.8%) fulfilled the criteria for fatigue. Twenty seven (84.4%) of these patients were female. The average age for patients classified as fatigued was 68.3 years (SD 11.1) while the average age for not fatigued patients was 73.4 years (SD 9.5). The most common polymorphisms at position 16, 27, and 164 were Arg16Gly, Gln27Glu, and Thr164Ile, respectively. Table 3 shows a negative association between the Thr164Thr genotype and fatigue (OR 0.14, 95% CI 0.02–1.11, p = 0.049). Accordingly, analysis of the allele carrier status revealed that fatigued RA patients were more likely to carry the Ile allele at position 164 (OR 7.33, 95% CI 0.9–59.8, p = 0.049). Overall, 75 patients (86.2%) were carriers of the Ile164 allele. The analysis of allele frequencies showed no significant associations to fatigue.

Table 3.

β2AR genotype, carrier status, and allele frequency by fatigue in patients with RA

	Fatigued, n (%)	Not fatigued, n (%)	Odds ratio (95% CI)	p value	Phi
β2AR genotype
Arg16Arg	2 (6.3)	4 (6.8)	0.92 (0.16–5.3)	1*	−0.010
Arg16Gly	25 (78.1)	48 (81.4)	0.81 (0.28–2.37)	0.712	−0.039
Gly16Gly	5 (15.6)	7 (11.9)	1.38 (0.4–4.75)	0.747*	0.053
Gln27Gln	3 (9.4)	8 (13.3)	0.67 (0.17–2.73)	0.742*	0.058
Gln27Glu	21 (65.6)	37 (61.7)	1.19 (0.48–2.91)	0.708	0.039
Glu27Glu	8 (25.0)	15 (25.0)	1 (0.37–2.69)	1	0
Thr164Thr	1 (3.2)	11 (19.6)	0.14 (0.02–1.11)	0.049*	0.228
Thr164Ile	21 (67.7)	31 (55.4)	1.69 (0.68–4.25)	0.259	0.121
Ile164Ile	9 (29.0)	14 (25.0)	1.23 (0.46–3.28)	0.683	0.044
Carrier status
Arg16	27 (84.4)	52 (88.1)	0.73 (0.21–2.51)	0.747*	0.053
Gly16	30 (93.8)	55 (93.2)	1.09 (0.19–6.31)	1*	−0.010
Gln27	24 (75)	45 (75)	1 (0.37–2.69)	1	0
Glu27	29 (90.6)	52 (86.7)	1.49 (0.37–6.05)	0.742*	−0.058
Thr164	22 (71)	42 (75)	0.82 (0.31–2.18)	0.683	0.044
Ile164	30 (96.8)	45 (80.4)	7.33 (0.9–59.8)	0.049*	0.228
Allele frequency
Arg16	29 (45.3)	56 (47.5)	0.92 (0.5–1.69)	0.782
Gly16	35 (54.7)	62 (52.5)	1.09 (0.59–2.01)	0.782
Gln27	27 (42.2)	53 (44.2)	0.92 (0.5–1.7)	0.796
Glu27	37 (57.8)	67 (55.8)	1.08 (0.59–2)	0.796
Thr164	23 (37.1)	51 (46.4)	0.68 (0.36–1.29)	0.239
Ile164	39 (62.9)	59 (53.6)	1.47 (0.78–2.77)	0.239

*Fisher’s exact test.

To compare the effects of the Ile164 polymorphism to other contributors to fatigue, we compared the two study groups regarding other possible risk factors as displayed in Table 4. While no effect could be detected for rheumatoid factor, CCP antibodies, CRP, joint erosions, disability, disease duration, prednisolone therapy, DMARD therapy, biologic therapy, or even fibromyalgia, disease activity, perceived stress, and depression were positively associated with fatigue with higher mean values for DAS28CRP (fatigued: 3.38, SD: 1.09, nonfatigued: 2.89, SD: 1.05, p = 0.038), PSQ (fatigued: 0.42, SD: 0.2, nonfatigued: 0.18, SD: 0.13, p < 0.001), and BDI-II (fatigued: 11.94, SD: 6.9, nonfatigued: 6.1 [5.97], p < 0.001) within fatigued RA patients. A regression analysis with these risk factors revealed that stress has the largest effect on fatigue (OR 2,935.53, 95% CI 11.75 to >10⁵) while depression (OR 1.05, 95% CI 0.94–1.17) and disease activity (OR 1.00, 95% CI 0.52–1.90) showed no statistical significance in the model (Table 5). With a Nagelkerke R² of 0.453, this model explains fatigue to a moderate extend.

Table 4.

Possible risk factors by fatigue in patients with RA

	Patients, n (total = 92)
	fatigued	not fatigued
	n (%)	n (%)	odds ratio (95% CI)	p value	phi/eta
Rheumatoid factor positive	21 (72.4)	44 (73.3)	0.96 (0.35–2.58)	0.927	0.010
Anti-CCP antibodies	22 (71)	45 (76.3)	1.32 (0.49–3.51)	0.584	0.058
Definitive erosions	12 (40)	35 (58.3)	0.48 (0.2–1.16)	0.101	0.173
Fibromyalgia	8 (29.6)	7 (12.7)	2.89 (0.92–9.07)	0.075*	0.205
DMARDs	30 (93.8)	48 (80.0)	3.75 (0.78–17.93)	0.126*	0.182
Prednisolone	25 (78.1)	44 (73.3)	1.30 (0.47–3.58)	0.613	0.053
Biologic therapy	6 (18.8)	12 (20)	0.92 (0.31–2.75)	0.886	0.015
Disability (HAQ)	18 (60)	25 (42.4)	2.04 (0.83–4.99)	0.116	0.167
Mean (SD) disease duration	16.82 (5.58)	19.86 (9.11)		0.218*	0.181
Mean (SD) CRP	11.2 (17.05)	6.69 (15.63)		0.115*	0.133
Mean (SD) DAS-CRP	3.38 (1.09)	2.89 (1.05)		0.038	0.218
Mean (SD) PSQ	0.42 (0.2)	0.18 (0.13)		<0.001*	0.590
Mean (SD) BDI	11.94 (6.9)	6.1 (5.97)		<0.001*	0.407

*Fisher’s exact test/Mann-Whitney U test.

Table 5.

Fatigue by clinical risk factors (regression)

	Odds ratio	95% CI	p value
DAS28CRP	1.00	0.52–1.90	0.995
PSQ	2,935.53	11.75–>105	0.005
BDI-II	1.05	0.94–1.17	0.370

Binary logistic regression, Nagelkerke R² = 0.453.

To reveal which aspects of fatigue may be affected by these risk factors, we compared MFI-20 subscale mean values after separating patients regarding the presence of the respective risk factor as presented in Table 6. When comparing Ile164 carriers to noncarriers, a significantly higher mean value was only found for general fatigue (carrier: 11.56, SD: 4.59, noncarrier: 7.92, SD: 4.08, p = 0.014). Despite not being statistically significant, the clinical difference among other MFI subscales was the largest for mental fatigue (carrier: 8.23, SD: 4.22, noncarrier: 5.67, SD: 1.56, p = 0.089, Cohen’s d = 0.629). Besides fatigue, the only other clinical variable associated to a positive Ile164 carrier status was DASCRP28 (carrier: 3.24, SD: 1.07, noncarrier: 2.32, SD: 0.86, p = 0.06). When compared to patients with low disease activity or remission, patients with high or moderate disease activity scored higher mean values for physical fatigue (moderate/high: 12.39, SD: 4.6; low/remission: 9.95, SD: 4.18; p = 0.009) and reduced activity (moderate/high: 11.69, SD: 4.64; low/remission: 9.05, SD: 4.1; p = 0.006) on the MFI-20. Reduced motivation and mental fatigue did not seem to be affected by disease activity. Moderate or high stress levels were positively associated with reduced activity (p = 0.03), reduced motivation (p = 0.033), and mental fatigue (p = 0.028), while no statistically significant association could be shown with physical fatigue. Table 6 also shows the positive association of moderate or severe depression with all MFI-20 subscales (p ≤ 0.005).

Table 6.

MFI-20 subscales by Ile164 carrier status, disease activity, stress, and depression level (mean [SD])

Ile164 carrier
	yes, n = 75	no, n = 13	p value	eta
MFI general	11.56 (4.59)	7.92 (4.08)	0.014*	0.270
MFI physical fatigue	11.36 (4.36)	9.33 (5.16)	0.094*	0.156
MFI reduced activity	10.39 (4.56)	9.5 (4.48)	0.549*	0.068
MFI reduced motivation	9.15 (3.55)	7.67 (3.23)	0.130*	0.145
MFI mental fatigue	8.23 (4.22)	5.67 (1.56)	0.089*	0.219

Disease activity (DAS28CRP)
	low or remission, n = 56	moderate or high, n = 36	p value	eta
MFI general	9.93 (4.59)	12.39 (4.34)	0.016	0.261
MFI physical	9.95 (4.18)	12.39 (4.6)	0.009	0.267
MFI reduced activity	9.05 (4.1)	11.69 (4.64)	0.006	0.289
MFI reduced motivation	8.77 (3.77)	8.86 (3.17)	0.729	0.013
MFI mental fatigue	7.73 (4.11)	8.06 (4.06)	0.699	0.039

Perceived stress
	low, n = 50	moderate/high, n = 10	p value	eta
MFI general	9.94 (4.26)	15 (4.5)	0.002	0.408
MFI physical	10.46 (4.24)	13.2 (4.02)	0.053	0.240
MFI reduced activity	9.66 (4.25)	12.7 (4.22)	0.030	0.262
MFI reduced motivation	8.16 (3.22)	10.9 (3.84)	0.033	0.298
MFI mental fatigue	7.5 (4.23)	10.2 (3.71)	0.028	0.239

Depression
	mild or less, n = 83	moderate/severe, n = 7	p value	eta
MFI general	10.43 (4.5)	15.71 (3.04)	0.005	0.308
MFI physical	10.42 (4.2)	17.14 (3.08)	0.001	0.403
MFI reduced activity	9.58 (4.15)	15.86 (3.89)	0.002	0.381
MFI reduced motivation	8.39 (3.16)	13.57 (3.69)	0.001	0.402
MFI mental fatigue	7.4 (3.76)	13.57 (3.74)	0.001	0.407

All p values calculated with Mann-Whitney U test.

Discussion and Conclusions

The key points emerging from this study are as follows:

• 1.Ile164 carrier status is positively associated with fatigue and could therefore be a possible risk factor.
• 2.Disease activity (DAS), stress, and depression are also associated with fatigue, confirming results of prior studies, while stress seems to have the largest effect.
• 3.While physical fatigue is associated with disease activity (DAS) and depression, a positive Ile164 carrier status seems to affect mental fatigue the most which is also associated with depression and stress.

Prior studies confirmed an involvement of β2AR polymorphisms in pathomechanisms of RA [10, 11]. These findings included a correlation of polymorphisms at amino acid positions 16 and 27 with RA and with an earlier disease onset age. While investigating the effects of these polymorphisms on fatigue however, we did not observe any significant effects. On the molecular level, polymorphisms at amino acid positions 16 and 27 mainly affect downregulation of the receptor [13–15] while polymorphisms at position 164 influence ligand affinity. Effects of polymorphisms at position 164 however have not been extensively investigated in the context of RA before.

With Ile164 carrier status being associated with fatigue, its complete absence in the genotype leading to the Thr164Thr genotype showed a negative association with fatigue and therefore seems to be a protective factor. The observed association of Ile164 carrier status and fatigue can be interpreted with respect to a change in receptor ligand affinity and therefore receptor responsiveness [13]. It has been confirmed that fatigued and depressed patients exhibit less βAR responsiveness [6]. While chronically increased stress levels due to fatigue or depression can lead to downregulation of adrenergic receptors and a decreased responsiveness, it has also been shown that decreased receptor responsiveness itself can cause fatigue [4]. This could also explain the observed association of perceived stress with fatigue. It remains unclear, however, if reduced receptor responsiveness within fatigued patients is the cause or the result of fatigue. Most likely, both pathways play a simultaneous role [6]. The Ile164 β2AR variants decreased ligand affinity and responsiveness could thus explain its association with fatigue. Results of recent studies reporting autoantibodies against β2AR in patients with CFS [29, 30] also support the hypothesis of an impaired β2AR function as a contributor to fatigue. Furthermore, elevated β2AR antibodies have recently been reported in patients experiencing post-COVID-19 fatigue or long COVID [31, 32].

Our findings reveal that a positive Ile164 carrier status and stress may have synergistic effects on fatigue as they both seem to increase mental fatigue while not affecting physical fatigue as much. Considering that stress and depression lead to downregulation and therefore a reduced responsiveness of β2ARs [6, 8] and the Ile164 receptor variant also exhibits a reduced responsiveness [13], these findings suggest that a decreased receptor responsiveness may be involved in pathogenetic pathways of mental fatigue. Studies show that reduced responsiveness of βARs can be detected in patients with depression [33] and anxiety disorders [34, 35] which along with our results indicates that the Ile164 receptor with its reduced ligand affinity may be implicated in psychological symptoms of RA patients including mental fatigue and depression.

Through which biological pathways decreased β2AR responsiveness can lead to symptoms of fatigue, however, remains to be determined. One hypothesis is that β2AR dysfunction leads to impaired modulation of vascular tonus. Physiologically, predominance of β1AR leads to vasoconstriction upon adrenergic activation whereas in brain, skeletal muscle, and heart predominance of β2AR leads to vasodilatation. As muscular and mental fatigue are major components of fatigue symptoms in RA and a reduced heart rate variability [36] along with increased baseline and maximum heart rates [5] have been reported in CFS patients, an impaired β2AR function can be assumed in fatigued RA patients. This would lead to impaired cerebral perfusion which was reported for regional cortical blood flow in CFS patients [37, 38] and could further explain our results regarding mental fatigue. Recent findings also show an association of increased cerebral blood flow and heart rate variability with lower fatigue symptom severity [39]. Reduced vasodilatation in the skeletal muscle as a result of an impaired β2AR function further mediates the release of endogenous vasodilators during exercise [30]. Especially, bradykinin exhibits hyperalgesic effects as it stimulates prostaglandin production which is known to cause hyperalgesia, another symptom commonly associated with fatigue [30, 40].

Another hypothesis is that inflammation largely contributes to fatigue. Chronically ill patients with high levels of inflammation experience more fatigue and depression symptoms [41]. Especially, pro-inflammatory cytokines like IL-1, IL-6, TNFα, and IFNα have been discussed to be involved in fatigue [2]. The expression of β2ARs by immune cells suggests an involvement of sympathetic neurotransmitters in the lymphoid cell activation process [42, 43]. It has also been revealed that lymphocytes of RA patients show a decreased β2AR expression [44]. Receptor polymorphisms could therefore have an effect on immune regulatory processes, potentially through altered ligand affinity or variations in the receptor expression rate mediated by polymorphisms in the 5′ leader cistron [45] which we did not investigate in our study. We observed an association of DAS 28 to fatigue; however, data on correlations between inflammation and fatigue in RA patients seem inconsistent. Disease activity is correlated with fatigue, but after adjusting for pain, it does not seem to be a significant contributor [46].

As group sizes for comparing the effects of the Ile164 polymorphism largely differed and p value results for the non-normally distributed MFI subscale scores may therefore not be very accurate, its involvement in mental fatigue can only be assumed as the clinical difference of its effects on MFI-20 subscales was the largest for mental fatigue while only showing a statistically significant effect for general fatigue. Furthermore, no specific genotype containing Ile164 showed an association with fatigue; only its overall presence or absence with Thr164Thr being the only significant genotype regarding fatigue proved statistically significant. It also remains to mention that genetic distribution of β2AR variants and its effects on RA and receptor responsiveness may be influenced by ethnicity to a great extent. While the Arg16 polymorphism is associated with RA in German patients [10], the Gln27 polymorphism also showed an association with RA in patients from northern Sweden [11]. Furthermore, the correlation of fatigue and depression to a decreased receptor responsiveness was only observed for Caucasian in contrast to African Americans. Thus, further research on the effects of the Ile164 polymorphism with bigger data while considering ethnical differences is necessary to draw definite conclusions.

To conclude, β2-adrenergic receptor responsiveness has complex effects on fatigue and depression within RA patients while the rare Ile164 polymorphism might in contrast to other β2AR polymorphisms affect levels of fatigue, especially mental fatigue, most likely through altered ligand affinity. This study gives further indication of a genetic background of fatigue. Replication studies are warranted to further determine the role of β2AR-polymorphisms in RA patients.

Acknowledgment

We acknowledge the statistical consultation provided by IMISE (Institute for Medical Informatics, Statistics and Epidemiology, Leipzig University).

Statement of Ethics

The study complies with the Declaration of Helsinki. Written informed consent was obtained from all subjects involved in the study. The study was approved by the Ethics Committee of the medical faculty of Leipzig University (N 042/2005).

Conflict of Interest Statement

The authors have no conflicts of interest to declare.

Funding Sources

German Rheumatism Foundation Berlin (Founders: German Society for  Rheumatology and German Rheumatism League),  Project Award 2022 for Olga Seifert. Open Access Publishing Fund of Leipzig University, which is supported by the German Research Foundation within the program Open Access Publication Funding.

Author Contributions

Julian Philipp: data collection, statistical analysis, data interpretation, and manuscript drafting. Christoph Baerwald: contribution to data interpretation and manuscript drafting. Olga Seifert: contribution to conceptualization and manuscript preparation. All authors read the final version of the manuscript and agreed to publication.

Funding Statement

German Rheumatism Foundation Berlin (Founders: German Society for  Rheumatology and German Rheumatism League),  Project Award 2022 for Olga Seifert. Open Access Publishing Fund of Leipzig University, which is supported by the German Research Foundation within the program Open Access Publication Funding.

Data Availability Statement

All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.