Adiponectin is Associated with Neutrophils to Lymphocyte Ratio in Patients with Chronic Obstructive Pulmonary Disease

Abstract

Disproportionate systemic inflammation in chronic obstructive pulmonary disease (COPD) is associated with declining lung functions and comorbidities. Neutrophil-to-lymphocyte ratio (NLR) and platelet-to-lymphocyte ratio (PLR) have emerged as valuable markers of the systemic inflammation in COPD. Adiponectin (Acpr30) circulates in serum as complexes of different molecular weight (HMW, MMW, LMW) with multifaceted metabolic and anti-inflammatory properties implicated in airway pathophysiology. We aimed to investigate the association between Acpr30 and its oligomers and the NLR and PLR in COPD patients. Seventy stable COPD patients were enrolled. Acrp30 serum levels and the HMW oligomers as well as hematological parameters and their ratio were evaluated. Both NLR and PLR are associated with lower BMI. Interestingly, total Acpr30 is negatively associated with NLR but not with PLR; after adjusting for age, BMI and FEV1, Acpr30 was independently associated with NLR. Conversely, HMW Acpr30 and HMW/Acpr30 ratio were positively correlated to NLR. The association of Acpr30, HMW Acpr30 and HMW/totalAcpr30 ratio with NLR but not with PLR in COPD patients indicates that Acrp30 oligomerization could represent a biological mechanism interfering with systemic inflammation in COPD. Further studies in larger cohorts of patients are required to confirm these results.

Keywords:

• Adiponectin
• HMW
• NLR
• PLR
• COPD

Introduction

Chronic obstructive pulmonary disease (COPD) is a leading cause of disability, morbidity and mortality worldwide. This is a disease state encompassing chronic bronchitis, small airways dysfunctions and emphysema resulting in progressive airflow limitation [1]. Abnormal inflammatory response within airways and lungs to noxious particles and gases is driven by innate and adaptative inflammatory cells [2,3]. Cardiovascular (CVD) and metabolic disorders are the main COPD comorbid conditions influencing long-term prognosis. Some studies have suggested that the spillover of pro-inflammatory mediators from the [4] lung into the systemic circulation coupled with anti-oxidant depletion resulted in increasing risk for CVD and metabolic conditions in COPD patients [5]. Indeed, in recent years it has become clearer that the humoral interconnection between lung and adipose tissue determines the prognosis and severity of lung disorders, including COPD, by driving the local and systemic inflammatory response [6]. In COPD patients, disproportionate inflammatory response—increased circulating inflammatory cells, cytokines, chemokines and acute phase proteins—have been associated to enhanced pulmonary function deterioration, higher exacerbation risk [7] and development of comorbidities [8]. Amongst the other mediators, the cytokines secreted by adipose tissue can be involved in the lung inflammatory state that in turn can regulate their secretion in a bidirectional interconnection [9]. Adiponectin (Acrp30) is one of the main adipokines involved in the relationship between lung and adipose tissue. Indeed, Acrp30 is increasingly associated with inflammatory pulmonary diseases, such as asthma and chronic obstructive pulmonary disease (COPD), and in critical illness [4,10,11]. In particular, in COPD patients, Acrp30 serum levels are up-regulated and directly associated with disease severity [9]. Although the functional role of Acrp30 up-regulation in COPD is still controversial, it is plausible that the adipose tissue acts by enhancing the production of this adipokine that is a potent anti-inflammatory molecule able to participate in the counteraction of lung impairment. Indeed, it was demonstrated that Acrp30, first recognized for its insulin-sensitizing effects, attenuates inflammatory responses to multiple stimuli in a variety of cell types [12]. Acrp30 is a 30 kDa protein produced by adipose tissue as a monomer but secreted in serum as oligomers of different molecular weight, low (LMW), medium (MMW) and high molecular weight (HMW) with the latter being the most effective oligomers [11–13]. In recent years, based on the rapid availability and relatively negligible costs, the absolute counts of circulating inflammatory cells in the peripheral blood and their ratios have been extensively investigated [14]. Neutrophils-to-lymphocytes ratio (NLR) on peripheral blood reflects the airway inflammation and was associated with impaired survival of erythrocytes with subsequently increased RDW and MPV [15,16]. In a retrospective study conducted in COPD patients, NLR has been previously found to be significantly different among exacerbated COPD (4.28 ± 4.12), stable COPD (2.59 ± 1.79) and healthy controls (1.71 ± 0.65) [17]. Interestingly, a statistically significant negative correlation between NLR and both BMI and fat-free mass index (FFMI) has been reported[18]. In addition, Acrp30 has also been associated with hematological parameters and to NLR in obese patients [19].

The present study aimed to clarify the involvement of Acrp30 in the inflammatory state of COPD by directly comparing hematological parameters and their ratio (recognized as inflammatory markers) to serum Acrp30 levels expression profile in a cohort of COPD patients. Total serum Acrp30, high molecular weight (HMW) and HMW-to-total Acrp30 ratio were evaluated and correlated to hematological parameters.

Methods

Subjects

We recruited 70 stable COPD patients (48 men and 32 women) at Department of Translational Sciences, Università della Campania “Luigi Vanvitelli”. COPD patients aged 50 or more with significant smoking history (pack/years > 10) were included in the study if they were able to sign informed consent. Exclusion criteria were as follows: previous diagnosis of metabolic syndrome; history of any type of cancer other than non-melanoma skin cancer; COPD exacerbation or pneumonia during the previous 8 weeks; hematological disorders; use of systemic glucocorticosteroids during the previous 4 weeks.

The study was approved by the local Ethics Committee of Faculty of Medicine of Università della Campania “Luigi Vanvitelli” and conducted according to ethical principles of the Declaration of Helsinki. All participants gave the informed consent to participate in the study. Blood samples were collected after a 12-hours overnight fasting period and centrifuged to collect serum. Serum aliquots were immediately frozen in liquid nitrogen and stored at −80 °C.

COPD diagnosis

COPD diagnosis was based on clinical history, physical examination and pulmonary function tests (FEV1/FVC below 70% after bronchodilator) [20,21]. The degree of severity of COPD was established using the values of the FEV1% predicted, history of exacerbations and severity of symptoms by the CAT and mMRC questionnaires according to the Global Initiative for Chronic Obstructive Lung Disease (GOLD) recommendation [1]. A stable state was defined as the absence of significant changes in symptoms beyond the expected daily variation, requiring treatment changes during the previous 3 months [22].

Anthropometric and biochemical measurements

The anthropometric and biochemical features of total study participants are reported in the Table 1. The height and weight of patients were measured using standard techniques and the BMI was calculated as body weight (kg)/height² (m²). The concentration of total Acrp30 in serum was measured in all individuals in triplicate by an enzyme-linked immunosorbent assay (ELISA) using a polyclonal antibody produced in-house versus a human Acrp30 amino acid fragment (H2N-ETTTQGPGVLLPLPKG-COOH) as previously described [23]. Each serum sample was tested three times in triplicate.

Table 1. Clinical characteristics in COPD study population (n = 70).

Demographic and clinical parameters	COPD
Age, median [IQR]	67.5 [59.8–73.3]
Sex, n (%)
M/F	48 (68.6)/32 (31.4)
Smoking habit n. (%)
Current	25 (35.7)
Former	45 (64.3)
Pack Years, median [IQR]	19 [16–23]
BMI (Kg/m^2)	26 [24–29]
Acpr30 (ug/μL), median [IQR]	40.1 [34.2–52.7]
HMW Acpr30 (μg/mL), median [IQR]	10.2 [5.7–13.3]
HMW/Acpr30, median [IQR]	0.24 [0.13–0.41]
FEV1, median [IQR]	1.38 [1.05–1.83]
% FEV1 th, median [IQR]	57 [42.8–72]
FVC, median [IQR]	2.5 [2.01–3.00]
% FVC th, median [IQR]	82 [66.5–97]
FEV1/FVC, median [IQR]	0.59 [0.51–0.65]
mMMRC
0–1	31 (44.3)
≥ 2	39 (55.7)
COPD Stage, n (%)
GOLD 1–2	44 (62.9)
GOLD 3–4	26 (37.1)

Statistical analysis

Categorical data were expressed as number and percentage, while continuous variables either as median and interquartile range (IQR) or mean and standard deviation (SD), based on their distribution, which was assessed both by the Kolmogorov-Smirnov Goodness of Fit test and by the Shapiro-Wilk test. Variable comparison at baseline was performed by either the Fisher Exact test or the Chi-Square test for categorical variables. Continuous variables, instead, based on their distribution were tested either by the Student t-test, in the case of a Gaussian distribution. Otherwise, either the non-parametric Mann Whitney U test or Wilcoxon test were used. Bivariate correlations were expressed by Pearson’s correlation coefficient, which assumes a value ranging between −1 and +1. Negative values are expression of an inverse correlation between variables. All correlations were also tested by linear regression models and complemented by appropriate dispersion graphs. All variables that were statistically significant in the univariate analysis were tested for a potential independent association with the outcome of interest. A multivariable logistic regression model, with a stepwise selection method, was performed. A p-value <0.05 was considered statistically significant. Data were analyzed using SPSS software, Version 24 (IBM, Armonk, New York) and STATA 14.0 software (StataCorp. 2015. College Station, TX: StataCorp LP).

Results

Clinical and biochemical parameters

The anthropometric and biochemical characteristics of the 70 COPD patients are reported in Table 1 and 2. Patients were predominantly male (68.6%) with a median pack/year of 19 [IQR 16–23]. 62.9% of COPD patients had mild to moderate airway obstruction (GOLD 1–2) and 37.1% were in more advanced stages (GOLD 3–4). NLR and PLR showed a non-normal distribution with a median value of 4.00 [IQR 2.62–4.85] and 122 [92–210.9], respectively.

Table 2. Laboratoristic findings in COPD study population.

Hematological parameters	COPD (n = 70)
WBC x10^3/μl, median [IQR]	6.1 [3.7–8.49]
Neutrophils x10^3/μl, median [IQR]	7.9 [4.8–56.2]
Neutrophils (%), median [IQR]	49.5 [20.9–67.8]
Lymphocytes x10^3/μl, median [IQR]	2.59 [1.54–29.55]
Lymphocytes (%), median [IQR]	31.8 [21.1–45.9]
Eosinophils x10^3/μl, median [IQR]	0.46 [0.19–2.57]
Eosinophils (%), median [IQR]	2.20 [1.38–4.14]
NLR, median [IQR]	4.00 [2.62–4.85]
RBC x10^6/μl, median [IQR]	5.58 [4.95–13.55]
HGB, median [IQR]	16.1 [14.3–40.3]
HCT, median [IQR]	47.8 [42.2–88.03]
MCV, median [IQR]	96.1 [87.9–218]
PLT x10^3/μl, median [IQR]	159 [114.5–247.2]
PLR, median [IQR]	122 [92–210.9]

Acrp30 correlates to NLR but not to PLR in COPD

Table 3 describes the significant correlations among hematological parameters, Acrp30 and clinical and laboratory parameters in the study population. We found that both NLR and PLR were slightly associated with lower BMI (Pearson Coefficient = −0.240 and −0.329; p = 0.045 and 0.041 respectively). PLR but not NLR showed negative correlation with FEV1/FVC ratio (Pearson coefficient = −0.391; p = 0.014). Acpr30 was negatively correlated with neutrophils (p = 0.01) whereas positive associations were found with lymphocytes and eosinophils (p < 0.001). Further correlations analyses are reported in Supplementary Tables 1–3.

Table 3. Significant correlation between NLR, PLR and adiponectin respectively, with study population characteristics.

	NLR	PLR	Acpr30	p*	p**	p***
BMI	–0.240	–0.329	–0.013	0.045	0.041	0.916
Acpr30	–0.440	0.050	–	<0.001	0.764	–
HMW Acpr30	0.308	0.003	–0.151	0.009	0.984	0.178
HMW/Acpr30 tot	0.419	–0.049	–0.589	<0.001	0.767	<0.001
FEV1/FVC	–0.033	–0.391	–0.237	0.784	0.014	0.049

^šCorrelations are expressed by the Pearson’s correlation coefficient.

Significant values are reported in bold.

Interestingly, NLR but not PLR was associated with both total Acpr30, HMW Acpr30 and HMW/total Acpr30. In particular, negative correlation was found between Acpr30 and NLR (Pearson coefficient = −0.440, p < 0.001), whereas both HMW Acpr30 and HMW/Acpr30 were positively associated with NLR (Pearson coefficient =0.308 and =0.419, p value = 0.009 and <0.001, respectively) (Figure 1).

Figure 1. Linear regression between neutrophils to lymphocytes ratio (NLR) with Acpr30 and HMW/Acpr30.

The final multivariate model showed that after adjusting for age, BMI, and FEV1 the Acpr30 was independently associated with NLR (p < 0.001) (Table 4).

Table 4. Multivariable logistic regression model of Acpr30 expression with a stepwise selection method.

	Standardized coefficients	p-value	95.0% Confidence Interval
	Beta		Lower limit	Upper limit
BMI	–0.124	0.264	–1.206	0.336
NLR	–0.412	0.000	–4.980	–1.472
FEV1	–0.137	0.235	–9.612	2.401
Age	0.174	0.139	–0.076	0.532
NLR	–0.383	0.000	–0.601	–0.164
FEV1	–0.149	0.194	–0.376	0.077
Age	0.164	0.164	–0.068	0.396
NLR	–0.388	0.000	–0.607	–0.169
Age	0.216	0.0054	–0.003	0.435

Significant values are reported in bold.

Discussion

Acrp30 has been extensively associated with pathological lung conditions and critical illness, such as asthma and chronic obstructive pulmonary disease (COPD). Although literature data about the precise functional role of this adipokine are still inconclusive, it is now overall accepted that Acrp30 is mainly involved in counteracting the systemic as well as the lung inflammatory responses. In COPD, the imbalance between anti-inflammatory and pro-inflammatory pathways has been long recognized to influence the course of the disease in both stable and exacerbated phases [24,25].

To further investigate whether Acrp30 participates in pro- and anti- inflammatory balance, in the present study we analyzed total as well as the oligomeric Acrp30 status in COPD patients in correlation to hematologic parameters. We found (i) a significant association between total Acrp30 and NLR; in particular, we found that NLR is negatively correlated with Acrp30 serum levels and positively correlated with HMW Acpr30 and HMW/total Acrp30 ratio; (ii) both NLR and PLR are associated with reduced BMI; (iii) after adjusting for clinical and functional characteristics Acrp30 was independently associated with NLR.

Elevated neutrophil levels in COPD biological samples (sputum and BAL) have been extensively reported and described as a hallmark in COPD pathogenesis [26,27]. Neutrophils activation within the airways perpetrates the inflammation and lead to alveolar damage, small airway disease and enhanced mucus production [28]. While producing local damage, lung-plasma spillover of the neutrophilic inflammation lead to the spread of pro-inflammatory mediators and chemokines potentially resulting in a systemic involvement of COPD.

The inter-organ cross-talk has been widely accepted as a key mechanism regulating the pathogenesis as well as the prognosis of several lung disorders including COPD; in this context, the adipose tissue is one of the main organs playing a role in this cross-talk, mainly through the secretion of adipokines deeply involved in the inflammatory processes. However, at this time poor data are currently available about the cross-link between systemic inflammation and the release of mediators from adipose tissue in chronic diseases with low grade inflammation. Acrp30 is the adipokine that has received a renewed interest because of a clear role in the regulation of immune responses. Indeed, Acrp30 manipulates both innate and adaptative immune response inhibiting them; in vitro studies documented that Acrp30 inhibits neutrophil apoptosis [29], while enhancing the apoptosis and inhibiting the proliferation of antigen-specific T-cell lines [30]. In a murine model of inflammatory bowel disease, Acrp30 promotes both local neutrophil recruitment and activation resulting in exacerbated clinical behavior [31]. Regarding NLR, results from a study involving obese subjects evidenced that this parameter is negatively associated with Acrp30 serum levels while being positively correlated with both IL-1β and TNF-α, markers of pro-inflammatory status [19]. Likewise, we found a negative correlation between NLR and Acrp30 serum levels. A potential explanation for this observation might be based on the following mechanism: in chronic disorders with persistent low-grade chronic inflammation—such as COPD—and in presence of enhanced NLR, adipocytes might reduce the Acrp30 synthesis and release limiting neutrophils anti-apoptotic effects. Unfortunately, longitudinal observational data are not available at this time and cannot confirm this hypothesis.

In addition, low levels of Acrp30 levels—in COPD patients with high NLR—may lead to detrimental effects on cell viability and increased apoptosis in human alveolar epithelial cell line, as we previously reported [32,33].

Interestingly, we found a significant direct correlation between HMW isoform/total Acrp30 and NLR. HMW Acrp30 has the longest half-life and is thought to be the most active oligomer in terms of biological effects. In vitro evidence suggested that HMW oligomers exert marked anti-inflammatory properties through the inhibition of TNF-α [34,35]. One possible explanation of the shift of Acrp30 oligomerization toward the HMW oligomers (in COPD patients with a worse phenotype as documented by high NLR and low FEV) could be represented by an activation of the adipose system to counterbalance the systemic inflammation through inhibiting of the chemokines traffic.

A relevant limitation of the current knowledge in this field is the absence of defined range value for NLR across different groups; for this reason, different cutoff values have been proposed. A range of normality between 0.78 and 3.53 was reported in adult healthy individuals as none of the subjects included in the analysis exceeded this value [36]; however, older adults were excluded and no correction for tobacco oral contraceptive use, ethnic group and sex were performed. In exacerbated COPD, NLR ⩾4 was independently associated with in-hospital mortality with 87% sensitivity and 40% specificity (AUC 0.717), after adjusting for age, sex, anemia and thrombocytopenia [37]. In stable COPD, longitudinal data from a prospective multicenter case control study including 368 moderate to severe COPD showed that a NLR cutoff value of 3.3 predicted mortality with a sensitivity of 85.8% and specificity of 89.7% [38]. Though our original findings may suggest a cross-talk between adipocyte released hormones and systemic inflammation in patients with COPD the presented study has some limitations: firstly, the study population is partially limited with mainly males included; secondly, the inclusive design of the study we did not excluded patients with extremely altered BMI. However, only one patient had BMI < 20 and one patient had BMI > 35. Finally, the present study does not include meaningful outcomes (mortality, exacerbations). Focused researches in this group of patients is required.

Conclusions

This study, for the first time, correlates Acrp30 and, more importantly, HMW/total Acrp30 ration to NLR in COPD patients clearly confirming that this adipokine is involved in COPD disease by acting on two levels, both immune and inflammatory. Our data suggest that, in COPD patients with a worse clinical picture, adipocytes might act reducing Acrp30 synthesis and release. On the other hand, the imbalance toward HMW indicates the need to release the most active form of Acrp30. Future research will determine whether modulation of Acrp30 may be helpful in the treatment of the inflammatory process typical of COPD illness.

Declaration of interest

The authors declare no conflict of interest.

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Additional information

Funding

This research was funded by POR CAMPANIA FESR 2014/2020. project- "iCURE" CUP: B21C17000030007.