Effects of menopause and fat mass in asthmatic inflammation

Abstract

Introduction

Female hormones and obesity have an impact on women with asthma. We aimed to describe how these components affect asthma inflammatory processes.

Methods

Sex hormones [FSH, LH, estradiol (E2), estrone (E1), testosterone and Δ4 androstenedione (A4)] and serum IL1β, IL13, IL17a, IL-5, IL6, TNF-a were measured from 11 to18 pre- and postmenopausal women with asthma.

Results

Premenopausal normal weight women revealed higher levels of IL5 and IL17a than obese women on both days of the menstrual cycle (IL5: D1: 6.4 vs 1.4 pg/ml, p = .036 and D14: 3 vs 1.4 pg/ml, p = .045 and IL17a: D1: 13.7 pg/ml vs 10.6 pg/ml and D14: 12.4 pg/ml vs 10.6 pg/ml, p = .009, respectively). In premenopausal women on D1, Δ4 Androstenedione was positively correlated with IL1β (p = .016, r = 0.733), whereas on D14, Estradiol with IL1β (p = .009, r = −.768) and TNF-a with Testosterone (p = .004, r = −0.816), and Δ4 Androstenedione (p = .002, r = −0.841) negatively. In postmenopausal women, TNF-a was negatively associated with FSH (p = .004, r = −0.638), but positively with Testosterone (p = .025, r = 0.526) and IL10 also positively with Estradiol (p = .007, r = 0.610).

Conclusion

Obesity shows a protective role in asthma through the suppression of IL5 and IL17. Estrogens seem to inhibit Th1 and Th2 inflammation, while androgens have a dual role with negative and positive correlations with neutrophilic biomarkers.

Keywords:

• Asthma
• obesity
• menopause
• sex hormones

Introduction

Sex and obesity have a strong association with asthma occurrence. Boys present more often and with more severe forms of asthma than girls before puberty, but this correlation is reversed in adult life (1,2). This pattern is also being observed in hospitalizations because of asthma in kids and teenagers but is three-fold higher among women aged between 20 and 50 years than men and changes after menopause to become equal to that of men (1,3). Additionally, the incidence of asthma after menopause has been connected with decreased levels of estrogen in combination with a protective effect of elevated FSH and LH levels (4,5). It has been established that the beginning of menopause leads to a low systemic inflammatory status, which is reflected by increased serum levels of proinflammatory cytokines related with the Th1 cascade (6), while asthma onset during menopause seems to present with pronounced severity and frequent exacerbations (7). However, the literature on this theme is conflicting, and a logical exegesis is lacking.

The management and therapy of asthma in obese asthmatics are often difficult mainly because of increased systemic inflammation linked to adipose tissue hormones such as adiponectin and leptin (8,9). Obesity has not been ‘causally’ associated with asthma but increasing body mass index has been shown to be related with more asthma symptoms and severity and this association, for unknown reasons, seems to be more powerful for women compared to men (10). Since obesity is a pro-inflammatory systemic metabolic state (11), it is widely assumed that an aggravation of airway inflammation is perhaps a dominant contributor to the association between these two conditions, and the severity of asthma. Actually, numerous unbiased hierarchical principal component analyses have identified severe asthma, which is characterized as late-onset and non-allergic, to cluster uniquely with obesity (12,13). On the other hand, it has been shown that leptin and adiponectin regulation in asthma depends on mechanisms which are both obesity and non-obesity related although obesity seems to have a bigger influence in active systemic inflammation compared to asthma per se (14). The exact pathophysiological pathway that bridges asthma and obesity has not been described and seems to be even more complicated when an extra parameter as sex hormones is added to the equation.

The aim of the present study was to describe inflammatory and functional characteristics of pre- and post-menopause female asthmatics and to examine possible associations of both Th1 and Th2 inflammatory markers with sex hormones and obesity.

Materials and methods

Study design

We included patients with previously diagnosed asthma, followed up in asthma clinics of 2nd Respiratory Medicine Department, University of Athens, “Attikon” Hospital, Chaidari, Greece. All patients provided detailed medical history and their demographic data, age of menarche and/or menopause, age of asthma onset and asthma treatment were recorded. Diagnosis of asthma was based on the GINA guidelines (15). All patients included in the study were in stable condition and optimally treated. Optimization of treatment was based on clinical and functional criteria and the results of Asthma Control Test (ACT).

In our study, both premenopausal and postmenopausal patients were included. The menstrual cycle was counted from the first day of the menstrual cycle to the first day of the next. All premenopausal women who participated in the study had a stable cycle of 28 days (16). This was evaluated based upon the menstrual history documentation of the patients during the last 12 months, using mobile applications (Flo Period Tracker & Calendar^©, Period Tracker - Period Calendar^© and Clue Period & Cycle Tracker^©). For premenopausal women, all hormonal measurements were performed on day 1 and were repeated on day 14 of the menstrual cycle. All patients confirmed that their menstruation started on day 28 by telephone contact. Two patients who did not present menstruation on the expected day were excluded from the study. Postmenopausal women underwent one single measurement. The postmenopausal population was confirmed to be at least 24 months without menses.

Patients with a history of hysterectomy with or without oophorectomy, gynecological or hormonal pathology, using hormonal therapy (contraception or hormonal replacement therapy), and those who were pregnant or lactating were excluded from the study. Moreover, those with severe asthma under biological therapy, those who had experienced an exacerbation of asthma or respiratory infection during the last 2 months, those with a diagnosis of other respiratory disease, concomitant malignancy or severe heart, liver, renal or collagen disease were also excluded. History of smoking was also an exclusion criterion. The study protocol was approved by the ethics committee of Attikon Hospital and all subjects provided written informed consent [Protocol Number: 839/16-10-13].

Study measurements

Lung function and FeNO

Forced expiratory volume in 1 s (FEV₁) and forced vital capacity (FVC) were measured using Master Screen Body (Viasys Healthcare, Jaeger, Hoechberg, Germany) according to the American Thoracic Society guidelines and were evaluated by asthma specialists (17).

FeNO was measured using a portable NO analyzer (NIOX MINO, Aerocrine, Solna, Sweden). MINO measurements are performed at a fixed flow rate of 50 ml/sec, are expressed as parts per billion (ppb) and are in clinically acceptable agreement with measurements provided by a stationary analyzer according to the ATS guidelines (18).

Blood measurements

Blood samples were collected from study participants for the measurement of LH, FSH, estradiol (E2), estrone (E1), progesterone, androstenedione (D4), testosterone were measured by ELISA (DIASOURCE Belgium), while inflammatory mediators IL-1β, IL4, IL5, IL6, IL13, IL10, IL17, TNFa were measured by a multiplex bead array assay (Luminex) using a commercial kit (MILLIPLEX MAP Human TH17 Magnetic Bead Panel, SEVEN PLEX Parameters, Millipore, USA) in blood serum. Pre-menopause women underwent two measurements on D1 and D14 while women after menopause only one measurement was performed.

BMI and DEXA measurements

In all patients, the Body Mass Index (BMI) was calculated as kilograms/meters² (kg/m²). Overweight or obese women were those with BMI over 25 or 30, respectively. Dual-energy X-ray absorptiometry (DEXA) (Hologic QDR series Discovery W densitometer of fan beam technology, Hologic, Bedford, MA, USA, provided with software APEX, version 3.2) was used to assess fat distribution, including whole and regional body composition and fat-free mass. DEXA provides estimates by measuring the body’s absorbance of X-rays at two different energies, given that fat, bone mineral and fat-free soft tissue have different absorption properties. Patients were positioned for regional and whole-body scans, according to the manufacturer’s protocol. Thus, they were instructed to lay in a supine position on the scanner table with straightened legs and with their arms close to the body and to remain still. In parallel with bone density, whole-body composition analysis provided data on total Fat Mass (FM) and its regional (trunk, arms, legs, head) distribution, with an average time of measurement of approximately 7 min (QDR bone densitometer with fan-beam technology, software version for Windows XP 12.3, Hologic Discovery-W, Bedford, MA, USA).

Statistical analysis

Categorical variables are presented as n (%), whereas numerical variables are presented as mean ± standard deviation (SD) or median (interquartile ranges) for normally distributed and skewed data, respectively. Normality of distributions was checked with Kolmogorov–Smirnov test. Statistical comparisons between groups were performed with one-way analysis of variance (ANOVA) for normally distributed data and Kruskal–Wallis tests for skewed data, accompanied by appropriate post hoc tests for multiple comparisons (Bonferroni and Dunn’s, respectively). Differences in numerical variables between two groups were evaluated with unpaired t-tests and Mann–Whitney U-tests for normal and skewed data, respectively, whereas comparisons of proportions were performed using chi-square tests. Correlations were performed using Spearman correlation coefficient. All p-values <.05 were considered statistically significant. Data were analyzed using SPSS 17.0 for Windows (SPSS Inc., Chicago, IL, USA), and graphs have been created using GraphPad Prism 6 (GraphPad Software, Inc La Jolla CA USA).

Results

Clinical data and markers of inflammation between pre- and post-menopausal asthmatic women

Twenty-nine patients with controlled asthma were included in the study. 11 (38%) were premenopausal with a mean age of 39 years and 18 (62%) were postmenopausal with a mean age of 65 years. Demographic characteristics of study participants are summarized in Table 1. Postmenopausal patients had a higher BMI and lower lung function parameters as expected by age. However, Asthma Therapy (OCS use, ICS dose), Total Body Fat (%), Fat Mass Index, Android to Gynoid Ratio, Trunk to Limb Fat Ratio or Appen, Lean Mass to Height² showed no significant difference between the two groups. No significant difference was observed on inflammatory biomarkers between pre-menopausal and post-menopausal women.

Table 1. Demographic and functional characteristics and levels of inflammatory biomarkers.

Parameter	Patients of reproductive age	Postmenopausal patients	p-value
	Ν = 11	Ν = 18
Age	39 [34–41]	65 [59–69]	<.001
BΜΙ	25.3 [21.0–26.4]	35.0 [25.0–41.1]	.048
OCS	1 (9.1%)	4 (22.2%)	.364
High ICS dose	10 (90.9%)	17 (94.4%)	.715
FEV1% pred D1	100 [79–102]	66 [55–89]	.039
FVC % pred D1	100 [80–108]	77 [65–99]	.047
FEV1/FVC D1	80 [74–84]	72 [65–75]	.014
ACT D1	21 [10–24]	18 [11–24]	.596
FeNO D1 (ppb)	16 [11–51]	28 [18–46]	.373
Total body fat %	41 [34-52]	47 [40-51]	.342
Fat Mass Index	12 [8–30]	16 [10–23]	.713
Android/gynoid ratio	0.9 [0.8–1.0]	1.0 [0.8–1.1]	.391
Trunk to Limp fat ratio	0.8 [0.6–1.0]	1.0 [0.8–1.1]	.118
Appen. lean mass/height^2 (kg/m^2)	7.2 [6.1–10.6]	7.7 [5.9–8.6]	.806
FEV1% pred D14	98 [79–104]	66 [55–89]	.072
FVC % pred D14	99 [89–116]	77 [65–99]	.051
FEV1/FVC D14	77 [75–82]	72 [65–75]	.040
ACT D14	20 [12–24]	18 [11–24]	.759
FeNO D14 (ppb)	19 [12–53]	28 [18–46]	.521
IL10 D1 (pg/ml)	1.1 [1.0–2.3]	1.3 [1.0–1.6]	.867
IL13 D1 (pg/ml)	70 [47–90]	78 [53–100]	.811
IL17a D1 (pg/ml)	11.5 [10.5–13.9]	11.2 [10.5–12.2]	.631
IL1b D1 (pg/ml)	2.2 [1.8–3.6]	1.9 [1.7–2.2]	.124
IL5 D1 (pg/ml)	1.7 [1.3–6.7]	2.4 [1.5–3.8]	.486
IL6 D1 (pg/ml)	2.5 [2.3–3.2]	2.5 [1.0–3.6]	.544
TNF-a D1 (pg/ml)	4.5 [3.8–5.4]	7.1 [4.3–9.1]	.058
IL10 D14 (pg/ml)	1.2 [0.9–2.2]	1.3 [1.0–1.6]	.689
IL13 D14 (pg/ml)	79 [49–106]	78 [53–100]	.944
IL17a D14 (pg/ml)	11.1 [10.5–12.5]	11.2 [10.5–12.2]	.981
IL1b D14 (pg/ml)	2.1 [1.9–3.0]	1.9 [1.7–2.2]	.133
IL5 D14 (pg/ml)	1.6 [1.3–3.8]	2.4 [1.5–3.8]	.286
IL6 D14 (pg/ml)	2.5 [2.2–3.3]	2.5 [1.0–3.6]	.689
TNF-a D14 (pg/ml)	4.6 [4.0–5.1]	7.1 [4.3–9.1]	.051

Abbreviations: FEV₁ = Forced expiratory volume in 1 s; FVC = forced vital capacity; ACT = Asthma Control Test; FeNO = fractional concentration of exhaled nitric oxide; TNF-a = tumor necrosis factor alpha; IL = interleukin; BMI = body mass index; ICS = inhaled corticosteroids; OCS = oral corticosteroids; FSH = follicle-stimulating hormone; LH = luteinizing hormone; D1 = Day 1; D14 = Day 14.

Bold indicates statistical significance.

A further analysis of postmenopausal patients according to the asthma onset (pre or post menopause), showed no significant differences in lung function asthma treatment, inflammatory markers or gender hormones (data shown in Table 2).

Table 2. Comparison of postmenopausal patients according to age of asthma onset.

	Asthma onset before menopause	Asthma onset after menopause	p-value
	N = 8	N = 10
Age	63.5 (51.2, 60.5)	64.4 (60.7, 69.2)	.573
Age of menopause	44.5 (29.2, 49.5)	49.5 (44.7, 53.2)	.128
BΜΙ (kg/m^2)	36.8 (24.0, 46.3)	32.0 (27.7, 38.6)	.829
OCS N (%)	2 (25)	2 (20)	.800
High ICS dose N (%)	8 (100)	9 (90)	.357
FEV1% pred	57.5 (55.0, 101.9)	72.5 (55.0, 89.2)	.696
FVC % pred	78.5 (62.1, 93.5)	75.1 (67.0, 101.7)	.965
FEV1/FVC	70.2 (59.8, 73.8)	73.0 (64.6, 77.7)	.633
ACT	17.5 (10.0, 23.0)	18.0 (10.5, 25.0)	.965
FeNO (ppb)	31.5 (20.2, 49.0)	23.7 (14.5, 44.2)	.485
Total body fat %	46.4 (20.2, 49.0)	48.2 (41.3, 50.7)	.959
FFMI (kg/m^2)	19.7 (14.4, 21.9)	16.9 (15.6, 20.9)	.645
Fat Mass Index	17.0 (9.6, 24.1)	15.9 (10.7, 20.9)	.798
Android/gynoid ratio	1.01 (0.81, 1.12)	0.97 (0.84, 0.99)	.442
Trunk to Limp fat ratio	1.00 (0.75, 1.11)	0.97 (0.82, 1.12)	.959
Appen. lean mass/height^2 (kg/m^2)	8.3 (5.8, 9.2)	7.1 (5.8, 8.3)	.505
IL10 (pg/ml)	1.3 (1.1, 1.6)	1.3 (1.3, 3.6)	.897
IL13 (pg/ml)	83.9 (45.4, 118.6)	69.4 (53.1, 87.5)	.573
IL17a (pg/ml)	11.1 (10.1, 12.6)	11.1 (10.5, 12.5)	.515
IL1b (pg/ml)	1.8 (1.7, 2.2)	1.9 (1.7, 2.2)	.829
IL5 (pg/ml)	2.7 (1.6, 5.5)	2.0 (1.5, 3.0)	.460
IL6 (pg/ml)	2.5 (1.4, 3.3)	2.5 (0.7, 5.2)	.897
TNF-a (pg/ml)	7.1 (4.4, 9.5)	7.5 (4.0, 10.1)	.965
Testosterone (ng/ml)	0.07 (0.03, 0.17)	0.4 (0.03, 0.3)	.965
Δ4 Androstenedione (ng/ml)	0.3 (0.1, 1.0)	0.3 (0.1, 0.9)	.829
Estrone (E1) (pg/ml)	20.7 (0.5, 4.)	25.5 (2.3, 119.5)	.515

Abbreviations: FEV₁ = Forced expiratory volume in 1 s; FVC = forced vital capacity; ACT = Asthma Control Test; FeNO = fractional concentration of exhaled nitric oxide; TNF-a = tumor necrosis factor alpha; IL = interleukin; BMI = body mass index; ICS = inhaled corticosteroids; OCS = oral corticosteroids; FFMI = fat-free mass index; FSH = follicle-stimulating hormone; LH = luteinizing hormone; D1 = Day 1; D14 = Day 14.

Bold indicates statistical significance.

Comparison of pre and post-menopausal asthmatic patients according to body weight

A comparison according to body weight revealed higher levels of serum IL5 in normal weight women compared to overweight/obese on both days of the menstrual cycle median (IQR) (D1: 6.4 (1.7, 34.0) vs 1.4 (0.9, 1.7) pg/ml, p = .036 and D14: 3.0 (1.6, 28.9) vs 1.4 (1.1, 1.7) pg/ml, p = .045, respectively). The same profile showed IL17a with 13.7 (12.1, 14.7) pg/ml vs 10.6 (10.1, 10.9) on D1 and 12.4 (11.4, 13.1) pg/ml vs 10.6 (10.3, 10.8) pg/ml on D14 for normal weight and overweight/obese women respectively, p = .009 for both comparisons. On the other hand, in postmenopausal women the division in overweight/obese and normal weight, revealed no statistical significant difference in inflammatory markers and sex hormones levels. Differences between overweight/obese and normal weight asthmatic women (pre and post menopause) are presented on Tables 3 and 4.

Table 3. Comparison between obese and non-obese women of reproductive age.

	Women with normal weight	Overweight or obese women	p-value
	N = 5	N = 6
Age	39 [35–40]	38 [29–48]	.783
BΜΙ (kg/m^2)	21.0 [19.7–22.7]	26.2 [25.3–49.0]	.006
FEV1% pred D1	96 [74–104]	102 [79–102]	.528
FVC % pred D1	97 [80–104]	102 [82–115]	.347
FEV1/FVC D1	81 [73–85]	78 [73–83]	.465
ACT D1	22 [9–25]	21 [10–24]	.745
FeNO D1 (ppb)	46 [14–66]	14 [6–32]	.117
IL10 D1 (pg/ml)	1.9 [1.2–4.8]	1.0 [1.0–1.1]	.117
IL13 D1 (pg/ml)	80 [52–248]	68 [44–77]	.251
IL17a D1 (pg/ml)	13.7 [12.1–14.7]	10.6 [10.1–10.9]	.009
IL1b D1 (pg/ml)	2.6 [2.1–6.5]	1.9 [1.7–2.4]	.117
IL5 D1 (pg/ml)	6.4 [1.7–34.0]	1.4 [0.9–1.7]	.036
IL6 D1 (pg/ml)	2.5 [1.6–13.9]	2.5 [2.5–2.5]	1.000
TNF-a D1 (pg/ml)	4.5 [3.7–6.7]	4.3 [3.7–5.1]	.602
FEV1% pred D14	93 [77–98]	104 [79–105]	.117
FVC % pred D14	95 [85–102]	115 [82–119]	.175
FEV1/FVC D14	81 [71–84]	77 [73–81]	.602
ACT D14	19 [11–25]	21 [12–23]	.916
FeNO D14 (ppb)	31 [19–74]	12 [7–46]	.149
IL10 D14 (pg/ml)	1.9 [1.1–3.9]	1.0 [0.8–1.2]	.076
IL13 D14 (pg/ml)	83 [65–232]	52 [35–99]	.251
IL17a D14 (pg/ml)	12.4 [11.4–13.1]	10.6 [10.3–10.8]	.009
IL1b D14 (pg/ml)	2.1 [2.0–4.4]	2.1 [1.8–2.7]	.528
IL5 D14 (pg/ml)	3.0 [1.6–28.9]	1.4 [1.1–1.7]	.045
IL6 D14 (pg/ml)	3.2 [1.9–12.6]	2.5 [1.4–2.5]	.147
TNF-a D14 (pg/ml)	4.2 [3.8–4.7]	4.6 [3.9–6.8]	.341

Abbreviations: FEV₁ = Forced expiratory volume in 1 s; FVC = forced vital capacity; ACT = Asthma Control Test; FeNO = fractional concentration of exhaled nitric oxide; TNF-a = tumor necrosis factor alpha; IL = interleukin; BMI = body mass index; ICS = inhaled corticosteroids; OCS = oral corticosteroids; FSH = follicle-stimulating hormone; LH = luteinizing hormone; D1 = Day 1; D14 = Day 14; DEXA = dual-energy X-ray absorptiometry.

Bold indicates statistical significance.

Table 4. Comparison between obese and non-obese postmenopausal women.

	Women with normal weight	Overweight or obese women	p-value
	N = 4	N = 14
Age	64 [46–70]	65 [59–69]	.750
Age of menopause	41 [36–52]	48 [45–53]	.165
BΜΙ (kg/m^2)	23 [22–24]	37 [29–45]	.003
FEV1% pred	57 [43–98]	73 [56–89]	.366
FVC % pred	79 [68–118]	75 [64–99]	.595
FEV1/FVC	57 [50–69]	74 [67–78]	.019
ACT	21 [16–23]	16 [9–25]	.453
FeNO (ppb)	49 [18–49]	26 [17–38]	.228
IL10 (pg/ml)	1.3 [1.1–1.6]	1.3 [0.9–2.2]	.710
IL13 (pg/ml)	69 [51–87]	81 [51–113]	.671
IL17a (pg/ml)	10.5 [10.3–11.6]	11.4 [10.6–12.9]	.242
IL1b (pg/ml)	1.8 [1.7–1.9]	2.0 [1.7–2.3]	.221
IL5(pg/ml)	2.6 [1.6–5.0]	2.3 [1.5–3.8]	.915
IL6 (pg/ml)	1.7 [0.4–2.5]	2.5 [1.6–3.9]	.155
TNF-a (pg/ml)	4.5 [3.7–8.5]	7.9 [6.0–9.5]	.339
Estrone (Ε1) (pg/ml)	97 [18–1379]	16 [0–63]	.135
Τestosterone (ng/ml)	0.03 [0.03–0.11]	0.08 [0.03–0.25]	.277
Δ4 Androstenedione (ng/ml)	0.19 [0.11–0.33]	0.49 [0.14–1.02]	.242

Abbreviations: FEV₁ = Forced expiratory volume in 1 s; FVC = forced vital capacity; ACT = asthma control test; FeNO = fractional concentration of exhaled nitric oxide; TNF-a = tumor necrosis factor alpha; IL = interleukin; BMI = body mass index; ICS = inhaled corticosteroids; OCS = oral corticosteroids.

Bold indicates statistical significance.

Correlations of sex hormones with inflammatory markers

A positive correlation was found in premenopausal women on Day 1 between IL13 and LH (p = .036, r = 0.700), Estrone and IL1β (p = .016, 0.732) and Δ4 Androstenedione and IL1β (p = .016, r = 0.733), whereas on Day 14 there was a negative correlation between Estradiol and IL1β (p = .009, r = −0.768) and TNF-a with Testosterone (p = .004, r = −0.816), Estrone (p = .044, r = −0.644) and Δ4 Androstenedione (p = .002, r = −0.841), (Tables 5 and 6).

Table 5. Association of sex hormones with inflammatory biomarkers in women of reproductive age D1.

	IL-10	IL13	IL17α	IL1β	IL5	IL-6	TNF-α
FSH	p = .433	p = .488	p = 1.000	p = .286	p = .527	p = .217	p = .332
	r = .300	r = .267	r = .000	r= −.400	r = .244	r= −.456	r= −.367
LH	p = .139	p = .036	p = .286	p = .381	p = .260	p = .963	p = .488
	r = 0.533	r = 0.700	r = 0.400	r = 0.333	r = 0.420	r = 0.018	r= −0.264
Progesterone	p = .831	p = .848	p = .286	p = .406	p = .351	p = .143	p = .154
	r = 0.830	r = 0.050	r = 0.400	r = 0.317	r = 0.353	r = 0.529	r = 0.517
Τestosterone	p = .417	p = .683	p = .194	p = .748	p = .349	p = .603	p = .458
	r= −0.310	r= −0.159	r= −0.477	r = 0.126	r= −0.354	r = 0.202	r= −0.285
Estradiol (E2)	p = .966	p = .831	p = .286	p = .088	p = .192	p = .064	p = .332
	r = 0.017	r = 0.083	r = 0.400	r = 0.600	r = 0.479	r = 0.639	r = 0.367
Estrone (E1)	p = .431	p = .178	p = .119	p = .016	p = .309	p = .219	p = .629
	r = 0.281	r = 0.413	r = 0.525	r = 0.732	r = 0.359	r = 0.426	r= −0.175
Δ4 Androstenedione	p = .425	p = .174	p = .162	p = .016	p = .298	p = .366	p = .293
	r = 0.285	r = 0.467	r = 0.479	r = 0.733	r = 0.366	r = 0.321	r= −0.370

Abbreviations: TNF-a = Tumor necrosis factor alpha; IL = interleukin; FSH = follicle-stimulating hormone; LH = luteinizing hormone; D1 = Day 1.

Bold indicates statistical significance.

Table 6. Association of sex hormones with inflammatory biomarkers in women of reproductive age D14.

	IL-10	IL13	IL17α	IL1β	IL5	IL-6	TNF-α
FSH	p = .603	p = .803	p = .328	p = .907	p = .827	p = .143	p = .906
	r = 0.188	r = 0.091	r= −0.345	r = 0.043	r= −0.080	r= −0.498	r = 0.043
LH	p = .580	p = .751	p = .556	p = .249	p = .723	p = .243	p = .551
	r = 0.200	r= −0.115	r= −0.212	r= −0.402	r= −0.129	r= −0.407	r = 0.215
Progesterone	p = .556	p = .829	p = .907	p = .392	p = .880	p = .579	p = .093
	r= −0.212	r= −0.079	r= −0.042	r= −0.305	r = 0.055	r = 0.200	r = 0.558
Τestosterone	p = .511	p = .150	p = .651	p = .343	p = .853	p = .901	p = .004
	r = 0.236	r= −0.491	r= −0.164	r = 0.335	r= −0.067	r= −0.045	r=−0.816
Estradiol (E2)	p = .446	p = .405	p = .511	p = .009	p = .506	p = .313	p = .575
	r= −0.273	r= −0.297	r = 0.236	r=−0.768	r= −0.239	r= −0.356	r = 0.202
Estrone (E1)	p = .701	p = .803	p = .328	p = .464	p = .987	p = .735	p = .044
	r = 0.139	r= −0.091	r = 0.345	r = 0.262	r= −0.006	r = 0.123	r=−0.644
Δ4 Androstenedione	p = .150	p = .405	p = .701	p = .115	p = .322	p = .579	p = .002
	r = 0.491	r= −0.297	r = 0.139	r = 0.530	r = 0.350	r = 0.200	r=−0.841

Abbreviations: TNF-a = Tumor necrosis factor alpha; IL = interleukin; FSH = follicle-stimulating hormone; LH = luteinizing hormone; D14 = Day 14.

Bold indicates statistical significance.

In postmenopausal women a negative correlation was observed between TNF-a and FSH (p = .004, r= −0.638) and between IL17a and LH (p = .016, r= −0.577), and a positive correlation between TNF-a and Testosterone (p = .025, r = 0.526) and between IL10 and Estradiol (p = .007, r = 0.610) (Table 7).

Table 7. Association of sex hormones with inflammatory biomarkers in postmenopausal women.

	IL-10	IL13	IL17α	IL1β	IL5	IL-6	TNF-α
FSH	p = .238	p = .798	p = .242	p = .234	p = .531	p = .541	p = .004
	r= −0.293	r= −0.065	r= −0.291	r= −0.296	r = 0.158	r= −0.154	r=−0.638
LH	p = .197	p = .598	p = .016	p = .165	p = .656	p = .943	p = .072
	r= −0.319	r= −0.133	r=−0.577	r= −0.342	r = 0.113	r= −0.018	r= −0.434
Progesterone	p = .579	p = .932	p = .527	p = .766	p = .848	p = .772	p = .218
	r = 0.140	r= −0.022	r= −0.160	r= −0.076	r = 0.049	r = 0.073	r = 0.305
Τestosterone	p = .567	p = .948	p = .977	p = .958	p = .804	p = .289	p = .025
	r = 0.145	r = 0.017	r= −0.001	r = 0.013	r= −0.063	r = 0.265	r = 0.526
Estradiol (E2)	p = .007	p = .144	p = .101	p = .202	p = .371	p = .418	p = .071
	r = 0.610	r = 0.358	r = 0.399	r = 0.316	r = 0.224	r = 0.203	r = 0.436
Estrone (E1)	p = .956	p = .925	p = .539	p = .450	p = .948	p = .138	p = .132
	r = 0.014	r= −0.024	r= −0.155	r= −0.190	r= −0.017	r= −0.363	r= −0.368
Δ4 Androstenedione	p = .584	p = .839	p = .525	p = .340	p = .707	p = .404	p = .165
	r= −0.138	r = 0.052	r = 0.160	r = 0.239	r= −0.095	r = 0.210	r = 0.342

Abbreviations: TNF-a = Tumor necrosis factor alpha; IL = interleukin; FSH = follicle-stimulating hormone; LH = Luteinizing Hormone.

Bold indicates statistical significance.

Discussion

In this study, we found no difference in inflammatory biomarkers between pre and post menopause asthmatic women. Premenopausal normal weight women revealed higher levels of IL5 and IL-17a compared to obese women on both days of the menstrual cycle. Furthermore, a positive correlation was found in premenopausal women between IL13 and LH on D1, and between Estrone and IL1β and Δ4 Androstenedione and IL1β on Day 14, whereas there was a negative correlation between Estradiol and IL1β, between TNF-a and Testosterone, and between Estrone and Δ4 Androstenedione on D14. Finally, in postmenopausal women a negative correlation was observed between TNF-a and FSH and between IL17a and LH and a positive correlation between TNF-a and Testosterone and between IL10 and Estradiol. To our knowledge, this is the first study that examines the role of inflammatory cytokines representative for both eosinophilic and neutrophilic inflammation in association with sex hormones and menopause.

A significant difference in the levels of IL5 between overweigh/obese and normal weight premenopausal asthmatics, leaving aside the day of their menstrual cycle, indicates a lower Th2 inflammatory status in premenopausal women with increased BMI and interestingly this difference was not observed in postmenopausal women. It has been previously reported that female asthmatics have higher levels of IL5 compared to males (19) and that sputum IL5 levels seem to be increased in luteal phases of menstrual cycle although no specific difference has been established concerning menopause in women (20). The finding of the lower levels of IL5 in obese premenopausal women is in accordance with the recent study of Bantulà et al., who showed that asthmatic patients with obesity or healthy controls have lower levels of IL5 compared to non-obese asthmatics (14). Moreover, according to Marijsse et al., levels of sputum IL5 were positively associated with the presence of obesity in patients with asthma. Although this association could not be related directly with sputum eosinophilia, rather with higher concentration of eosinophils in the airway submucosa of obese asthmatics (21). Furthermore, the absence of the difference between overweight/obese and normal weight post-menopausal women probably means that in the absence of estrogens body weights does not affect the Th2 inflammatory process.

IL1β, a cytokine of the NLRP3 inflammasome-mediated responses, plays a role in the pathogenesis of severe neutrophilic asthma by contributing to airway remodeling processes, as well as the polarization of IL17 producing Th17 cells (22–24). The negative association of IL1β and Estradiol in our study is in accordance with the findings of Cheng et al., who showed that mRNA expression of IL1β is suppressed by Estradiol (E2) in ovalbumin-challenged mice (25). On the other hand, the study of de Oliveira et al., also in murine models, suggested that Estradiol stimulates the release of IL1β from BAL cells (26), describing the complex effect of estrogens on airway inflammation. Furthermore, IL10 is an anti-inflammatory cytokine expressing its effects through the suppression of mast cells, eosinophils and IgE production (27,28), while in parallel inhibits the proliferation of Th1 cells (29). The positive association between IL10 and Estradiol in our postmenopausal women could be supported by other studies such as Kanda et al., in which mononuclear cells of patients with Systemic Lupus Erythematosus treated with Estradiol resulted in increased levels of IL10 (29) and De Oliveira et al., who gave evidence that Estradiol enhances the release of IL10 by bone marrow cells in allergic mice (26). In contrast to the above studies, Girón-González et al. showed that the addition of Estradiol in supernatants of T-cell cultures did not affect the concentration of IL10, but this study has the limitation of exclusively in vitro observations (30). In summary, the positive association of Estradiol with IL10 enhances the theory of the protective activity of estrogens against both Th1 and Th2 airway inflammation.

According to Wegienka et al., Estrone in urine samples of premenopausal asthma patients seems to have a positive correlation with T regular cells in blood (31). In our study, Estrone was associated positively with IL1β on the 1st day of menstrual cycle and negatively with TNF-a on the 14th day of menstrual cycle. To our knowledge, these correlations have not been presented before; and hence we cannot interpret them appropriately. The role of estrogens in suppressing airway inflammation is well established (32), but most studies use Estradiol measurements, since this is the most potent estrogen in women of reproductive age, thus the data for Estrone are limited.

Testosterone can decrease Th2 airway inflammation by reducing IL-13 expression and eosinophilic infiltration in the airways (33). There is evidence that TNF-a contributes to steroid-resistant neutrophilic inflammation in asthma (34). Furthermore, androgens are known to prevent the production of TNF-a on monocyte macrophages (35). That could explain the negative association of TNF-a with Testosterone and Δ4 Androstenedione in premenopausal women. Furthermore, a study in women with polycystic ovary syndrome showed that women with hyperandrogenenmia had higher levels of Δ4 Androstenedione and neutrophils in peripheral blood (36). The positive correlation of Δ4 Androstenedione and IL1β on the 1st of menstrual cycle in our study, in combination to the ability of IL1β to induce neutrophilic inflammation (22), provide a possible mechanism for Δ4 Androstenedione in stimulating neutrophilic inflammation. It has also been suggested that TNF-a increased in women after surgical oophorectomy (37) and TNF-α-producing Th1 cells correlated positively with age in postmenopausal women (38). In the same study of Han et al., FSH levels in postmenopausal women showed a negative association with Th1 cells (38), so that a negative correlation between FSH and TNF-a could be speculated. Although Testosterone levels seem to be low after menopause, Maggio et al., found that this hormone associated positively with TNF-a levels after age adjustments pointing at a role for testosterone in Th1 inflammatory processes (39).

IL17a and IL17F have been implicated in asthma pathogenesis and in particular in neutrophilic inflammation (40). IL17a signaling is also known to promote metabolic dysfunction and obesity (41). These data explain the fact that IL17a levels are lower in obese premenopausal compared to normal weight women regardless of the day of their menstrual cycle. Considering all the above it seems that obesity has a prophylactic effect on IL5 and IL17-mediated inflammation in premenopausal women.

The main limitation of our study is the evaluation of menstrual cycle stability, which was based upon the documentation that the patients had done during the last year for their own use. The ovulation was not confirmed with a hormonal measurement, but we consider it took place on day 14 since all the premenopausal women presented menstruation on day 28 as confirmed by telephone contact. Another reason for varying results between our study and others may be the timing of the blood samples. The timing of blood sampling during the menstrual cycle is very important, since hormone concentrations fluctuate on a daily basis and consequently cytokines at the mid follicular phase may differ in concentrations from early or late follicular phase. We have performed all measurements early in the morning of both days of the menstrual cycle and also in post-menopausal women in an effort to be as precise as possible and to avoid diurnal variability of the measurements. A major problem needs to be addressed in studies with sex hormones is the standardization of sampling period. Finally, our study included a small number of pre- and post-menopausal women and that could also limit the power of our results. Larger studies are needed in order to confirm these association and interpret them in clinical practice.

Conclusion

Obesity may be involved in asthma pathogenesis and seems to have a prophylactic effect on female asthma patients through the suppression of IL5 and IL17. Estradiol seems to function as a Th2 and Th1 inflammatory suppressor in asthma through the stimulation of IL10 and negative association with IL1β. However, androgens have a dual role in premenopausal women suppressing Th1 inflammation with the negative correlation between Testosterone and TNF-a on day 14 of the cycle, but enhancing neutrophilic cascade through the positive association of Androstenedione with IL1β on day 1. After menopause also two opposite Th1 effects are expressed, because FSH seems to antagonize TNF-a, whereas Testosterone raises its levels.

Abbreviations
FEV1:	=	Forced expiratory volume in 1 s
FVC:	=	Forced vital capacity
ACT:	=	Asthma Control Test
Th1:	=	T helper cell type 1
Th2:	=	T helper cell type 2
FeNO:	=	Fractional concentration of exhaled nitric oxide
TNF-a:	=	Tumor necrosis factor alpha
IL:	=	Interleukin
BMI:	=	Body mass index
ICS:	=	Inhaled corticosteroids
OCS:	=	Oral corticosteroids
FFMI:	=	Fat-free mass index
FSH:	=	Follicle-stimulating hormone
LH:	=	Luteinizing hormone
D1:	=	Day 1
D14:	=	Day 14
DEXA:	=	Dual-energy X-ray absorptiometry

Declaration of interest

The authors report no conflicts of interest. The authors alone are responsible for the content and writing of this article.

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.