Interrelationship of Circulating Matrix Metalloproteinase-9, TNF-α, and OPG/RANK/RANKL Systems in COPD Patients with Osteoporosis

Abstract

Previous studies have shown that matrix metalloproteinase-9 (MMP-9) and its cognate inhibitor TIMP-1, inflammatory cytokine TNF-α, and the OPG/RANK/RANKL system may each play individual roles in the pathogenesis of osteoporosis in patients with COPD. In the present study, we investigated the interrelationships of these factors in male COPD patients with and without osteoporosis. The serum levels of MMP-9, MMP-9/TIMP-1 ratio, TNF-α, RANKL, OPG, and the RANKL/OPG ratio were higher in COPD patients with osteoporosis than in individuals with normal or low bone mineral density (BMD) (N = 30, all P < 0.05 or < 0.01). The lung function FEV₁%Pre and the BMD of the lumbar spine and femoral neck were found to be negatively correlated with MMP-9 serum level (r = −0.36, P < 0.05, r = −0.58, P < 0.001, and r = −0.62, P < 0.01, respectively), RANKL serum level (r = −0.21, P < 0.05, and r = −0.25, P < 0.05, and r = −0.26, P < 0.05, respectively), and RANKL/OPG ratio (r = −0.23, P < 0.05, r = −0.33, P < 0.05, and r = −0.38, P < 0.05, respectively). However, they had no correlation with TIMP-1, TNF-α, OPG, or RANK. The MMP-9 serum level was found to be positively correlated with TNF-α level (r = 0.35, P < 0.05) and RANKL/OPG ratio (r = 0.27, P < 0.05) but not associated with RANKL. These results suggest that MMP-9, TNF-α, and the OPG/RANK/RANKL system may be closely interrelated and may play interactive roles in pathogenesis of osteoporosis in COPD.

Keywords :

• Bone mineral density
• osteoporosis
• COPD
• matrix metalloproteinase
• pathogenesis

Abbreviations
COPD	=	Chronic obstructive pulmonary disease
BMD	=	Bone mineral density
DXA	=	Dual X-ray absorptiometry
MMPs	=	Matrix metalloproteinase
TIMPs	=	Tissue inhibitors of metalloproteinase
OPG	=	Osteoprotegerin
RANK	=	Receptor activator of NF-κB
RANKL	=	RANK ligand
FEV1	=	Forced expiratory volume in one second
FVC	=	Forced vital capacity
BMI	=	Body mass index
CAT	=	COPD assessment test
GOLD	=	The Global Initiative for Chronic Obstructive Lung Disease

Introduction

Chronic obstructive pulmonary disease (COPD) is a common chronic respiratory disease characterized by airflow limitation, It has a high and increasing prevalence and mortality. By 2020, COPD may become the third most common fatal disease in the world (1). The systemic complications associated with chronic obstructive pulmonary disease (COPD) include osteoporosis, loss of skeletal muscle mass and function, systemic inflammation, catabolic intermediary metabolism, impaired glucose tolerance, and an increased risk of cardiovascular disease, all of which cause considerable morbidity and mortality (2–3). Osteoporosis, a condition characterized by low bone mass and bone microarchitectural deterioration which predisposes to fragility fracture, is one of the most important systemic complications of COPD. The incidence of osteoporosis among COPD patients is significantly higher than in the general population (4–5) and this association is not limited to those with severe COPD. In a study performed by Duckers et al., over 80% of the male COPD patients with predominantly mild to moderate airflow ­obstruction (GOLD I and II) had osteopenia or osteoporosis (6). Likewise, Sin et al. found that, in non-Hispanic Caucasians, airflow obstruction was associated with decreased bone mineral density. Thirty-three percent of female patients with severe COPD suffered from osteoporosis, while almost all women with relatively mild airway obstruction suffered from osteopenia (7). Clinical sequelae of osteoporosis, including thoracic compression fractures and kyphosis, can further reduce vital capacity, thereby further compromising respiratory status in patients with COPD (8). The mechanisms underlying the pathogenesis of osteoporosis in COPD are not yet fully clear. However, the pathophysiology appears to be at least partially independent of exposure to glucocorticoids. For example, Bolton et al. reported a high prevalence of osteoporosis in patients with COPD who were not receiving long-term oral corticosteroid therapy (9).

Matrix metalloproteinases (MMPs), a group of calcium-activated zinc ion endopeptidases widely found in a variety of connective tissues, play an essential role in the degradation of the extracellular matrix and basement membrane enzyme. They are differentially expressed during different stages of COPD development (10). The TIMPs are endogenous natural inhibitors of MMPs. The airway MMP-9/TIMP-1 ratio is considered a biomarker of airway tissue destruction and of the dynamic equilibrium of repair (11). MMP-9 is produced by osteoclasts and osteoclast precursor cells, the monocytes also. It is essential for recruitment of preosteoclasts into bone and their local migration into the diaphysis (12). Specific expression of MMP-9 in osteoclasts may play an important role in osteoclastic bone resorption, thereby promoting osteoporosis. In contrast, TIMP-1 prevents bone resorption (13).

TNF-α is a multifaceted cytokine and a well-recognized systemic inflammatory factor related to the occurrence and development of a variety of the systemic complications of COPD, including osteoporosis. TNF-α is also an important regulator of bone metabolism and remodeling. It stimulates osteoclast differentiation in a synergistic manner (14). A study by Pinto-Plata showed that levels of the inflammatory markers IL-6 and TNF-α became higher and the levels of markers of injury and repair became lower as the disease progressed (GOLD 1 vs. 4) (15). Bon et al. also showed a clear correlation between many serum inflammatory cytokines, including IL-1, IL-6, and TNF-α, and the severity of emphysema as assessed by high-resolution CT (16). Levels of IL-1, IL-6, TNF-α, and other cytokines have been shown to stimulate osteoclast differentiation (17). Inflammatory cytokines can also directly act on osteoclasts and their precursor cells, which increases osteoclast activity, causing bone destruction and bone loss (18).

In addition, these inflammatory cytokines can interact with the osteoprotegerin (OPG)/RANK/ receptor activator of nuclear factor kappa B ligand (RANKL) system and so regulate bone metabolism (19). Osteoprotegerin (OPG) is an endogenous receptor antagonist of RANKL. It blocks the binding of RANKL to RANK, so inhibiting osteoclast development and bone resorption. RANKL and OPG are the promoting and inhibitory factors of osteoclast differentiation, respectively. They maintain a certain ratio within the body. Imbalances in this ratio can disrupt bone metabolism, leading to reduced bone mass/osteoporosis among other bone diseases (14,19–20). Lastly, TNF-α is involved in the recruitment of neutrophils and is associated with the upregulation of MMP-9 and TIMP-1 (21–23).

To our knowledge, no studies have evaluated MMP-9/TIMP-1, TNF-α, and the OPG/RANK/RANKL system in patients with COPD. We hypothesized that MMP-9 and its inhibitor TIMP-1, inflammatory cytokine TNF-α, and the OPG/RANK/RANKL system may play interactive roles in the pathogenesis of osteoporosis in COPD patients. In this study, the correlations among serum MMP-9 and its inhibitors, specifically TIMP-1, TNF-α, and the OPG/RANK/RANKL system were assessed, and their associations with impairment of lung function and bone mineral density in patients with COPD was evaluated.

Methods

Subjects

Ninety male patients diagnosed with clinically stable COPD were enrolled from May 2010 to May 2012. Female patients were excluded from the study to avoid the influence of postmenopausal osteoporosis. COPD was diagnosed according to the criteria issued by the Global Initiative for Chronic Obstructive Lung Disease (GOLD) (2).

These ninety patients were divided into three groups based on BMD: those with COPD and normal BMD (T-score > −1.0), those with COPD with osteopenia (T score −1.0 to −2.4), and those with COPD and osteoporosis (T-score ≤ −2.5). Each group contained 30 patients. All patients were more than 40 years old and all were current or former smokers. The exclusion criteria were as follows: neoplastic, metabolic, or inflammatory disease; cardiac failure; current ischemic symptoms; inhaled, oral, or intravenous corticosteroid treatment within the last 3 months; and use of weight-lowering drugs. We excluded subjects with known osteoporosis and those being treated for osteoporosis, including those taking bisphosphonates because such treatments were likely to be confounders. On the recruiting day, all patients completed the life quality assessment using the COPD Assessment Test (CAT). Total scores ranged from 0 to 40. Patients with CAT scores of 0–10 were categorized as mildly affected, those with scores of 11–20 as moderately affected, those with of 21–30 as severely affected, and those with scores of 31–40 as very severely affected (24).

The study was approved by the local ethics committee of the First People's Hospital of Foshan. All subjects provided informed consent before joining the study.

Anthropometry and lung function

Patient height and weight were recorded and used to calculate body mass index (BMI) using the following formula: BMI = weight/height (kg/m²). Pulmonary function tests were performed by an experienced technician using a Master Screen Body spirometer (Jaeger, Germany), with quality control performed according to standards issued by the American Thoracic Society (ATS). Before spirometry, patients abstained from short-acting inhaled bronchodilators for 6 hours and from long-acting bronchodilators for 12 hours. Forced expiratory volume in 1 second (FEV1), forced vital capacity (FVC), and their ratio (FEV1/FVC) were determined in all subjects from the best of three valid attempts.

Dual-energy X-ray absorptiometry

The BMD of lumbar spine (L1–L4) and bilateral femoral neck was measured using dual X-ray absorptiometry (Hologic Inc., Waltham, MA, U.S.). The instrument had a coefficient of variation (CV) of area under 0.52%, and CV of BMD under 0.40%, as indicated by daily calibration against a human vertebral phantom. BMD is here expressed in g/cm² and a T-score, which was used for diagnosis according to the World Health Organization guidelines as follows: normal BMD: T score greater than −1 at both sites (lumbar spine and FN); osteopenia: T score less than or equal to −1 but greater than −2.5 at either site; osteoporosis: T score less than or equal to −2.5 at either site (25).

Enzyme-linked immunosorbent assay (ELISA)

Ninety patients were divided into three groups of 30 patients each based on BMD: those with COPD and normal BMD, those with COPD with osteopenia, and those with COPD and osteoporosis. Fasting blood samples were collected by venipuncture the next morning. These were centrifuged at 1000 × g for 5 min at room temperature to produce serum samples, which were stored at −80°C until analysis. High-sensitivity MMP-9 (pro-MMP-9), TIMP-1, and TNF-α levels were determined using sandwich ELISA kits (Catalog No. DMP900; DTM100; DTA00C; R & D Systems, Inc. Minneapolis, MN, U.S.). The intra- and inter-assay coefficients of variation were both under 10%, and the sensitivity was 0.156 ng/ml for MMP-9, 0.08 ng/ml for TIMP-1, and 5.5 pg/ml for TNF-α. OPG/RANK/RANKL concentrations were measured using ELISA kits (Catalog No. CSB-E04692h, E13539h, E05125h; Cusabio Biotech, U.S.), with a limit of detection at 0.078 ng/ml for OPG, 4.69 pg/ml for RANK, and 1.95 pg/ml for RANKL. Assays were performed in accordance with the manufacturer's instructions. (ELISA was performed in triplicate.)

Statistics

Data are expressed as the mean ± SD and analyzed using the Statistical Package for the Social Sciences, SPSS13.0 (SPSS Inc, Chicago, IL, U.S.). Data from the three experimental groups were analyzed with normally distributed test(shapiro-wilk test), one-way ANOVA with a post-hoc Tukey's test, and Pearson's correlation analysis. The relationships among continuous parameters were evaluated using multiple linear stepwise regression analysis. A p-value < 0.05 was considered statistically significant.

Results

Demographic parameters, pulmonary function, and quality of life scores (Table 1)

The age, BMI, smoking index, FEV1/FVC, FEV1%Pre, and CAT scores of the COPD patients were compared between all groups of patients. COPD with normal BMD vs. COPD with low BMD, COPD with normal BMD vs. COPD with osteoporosis, and COPD with low BMD vs. COPD with osteoporosis were evaluated. As shown in Table 1, COPD patients with osteoporosis had significantly lower BMI (P < 0.01) and higher CAT scores (P < 0.05) than COPD patients with normal BMD. There was no significant difference in age, smoking index, FEV1/FVC, or FEV1%Pre. As expected, based on the study design, the lumbar spine and femoral neck BMD was significantly lower in the groups of COPD with low BMD (P < 0.05) and COPD with osteoporosis (P < 0.01) than in COPD with normal BMD.

Table 1.  Basic characteristics of three groups of patients with COPD.

Characteristics	COPD with normal BMD (N = 30)	COPD with osteopenia (N = 30)	COPD with osteoporosis (N = 30)
Age (years)	64.5 ± 9.2	66.5 ± 7.2	72.5 ± 9.6
FEV1/FVC	58.2 ± 9.5	54.6 ± 12.7	52.3 ± 6.8
FEV1%Pre	59.1 ± 18.4	55.9 ± 12.5	51.6 ± 17.5
Smoking index (packs/year)	29.2 ± 20.8	33.8 ± 22.4	36.9 ± 18.6
BMI	27.6 ± 3.1	24.3 ± 3.4*	21.4 ± 4.1**
CAT score	14.9 ± 8.5	20.7 ± 3.9*	25.3 ± 5.9*
BMD (lumbar spine) (g/cm^2)	1.05 ± 0.06	0.87 ± 0.07*	0.79 ± 0.03**^^
BMD (femoral neck) (g/cm^2)	0.93 ± 0.09	0.84 ± 0.20*	0.68 ± 0.03**^^

*P < 0.05,**P < 0.01 vs COPD with normal BMD.

^P < 0.05,^^P < 0.01 vs COPD with low BMD.

Serum level of MMP-9/TIMP-1, TNF-α, and OPG/RANK/RANKL (Table 2, Figure 1–2)

Differences in MMP-9 serum level were detected among the three groups (P < 0.05). The MMP-9 serum levels increased successively from COPD patients with normal BMD to those with low BMD to those with osteoporosis. However, no difference in serum levels of TIMP-1 was detected among the three groups. The MMP-9/TIMP-1 ratio increased successively from COPD patients with normal BMD to those with COPD and low BMD to those with COPD and osteoporosis. A difference in MMP-9/TIMP-1 ratio was detected between COPD patients with normal BMD and those with osteoporosis (P < 0.05).

The serum level of TNF-α differed among the three groups (P < 0.05), increasing successively from those with COPD and normal BMD to those with COPD and low BMD to those with COPD and osteoporosis. The COPD patients with osteoporosis had the highest serum level of TNF-α (P < 0.05).

The serum levels of OPG were higher in COPD patients with osteoporosis than in COPD patients with normal BMD (P < 0.05), but there was no difference in OPG serum level between either the COPD patients with osteoporosis and those with low BMD or between the COPD patients with low BMD and those with normal BMD. No difference existed in the level of RANK among these three groups.

The RANKL serum level was markedly higher in COPD patients with osteoporosis than in those with normal or low BMD (P < 0.05), but no difference was observed between COPD patients with low BMD and those with normal BMD.

The RANKL/OPG ratio was higher in COPD patients with osteoporosis than in those with normal or low BMD (P < 0.05), but no difference was detected between COPD patients with low BMD and those with normal BMD.

Correlations of BMD with MMP-9/TIMP-1, TNF-α, and OPG/RANK/RANKL serum levels

Pearson analysis revealed negative correlations between MMP-9 serum levels and the lumbar spine BMD (r = −0.58, P < 0.001) and femoral neck BMD (r = −0.62, P < 0.01); between RANKL serum levels and the lumbar spine BMD (r = −0.25, P < 0.05) and femoral neck BMD (r = −0.26, P < 0.05); and between the RANKL/OPG ratio and the lumbar spine BMD (r = −0.33, P < 0.05) and femoral neck BMD (r = −0.38, P < 0.05). Other cytokines, such as TIMP-1, TNF-α, OPG, and RANK, did not correlate with the lumbar spine or femoral neck BMD.

Within the COPD groups, multiple linear stepwise regression analyses were performed with either lumbar spine BMD or femoral neck BMD as the dependent variable,and MMP-9, TIMP-1, TNF-α, and OPG,RANK,RANKL as independent variables. The level of RANKL and MMP-9 (P < 0.05) were both predictive for lumbar spine BMD with an adjusted R2 = 0.212. The level of MMP-9 and the ratio of RANKL/OPG (P < 0.05) were both predictive for femoral neck BMD with an adjusted R2 = 0.519. (Table 3)

Table 2.  Serum levels of MMP-9/TIMP-1, TNF-α, and OPG/RANK/RANKL in patients with COPD.

	COPD with normal BMD (N = 30)	COPD with osteopenia (N = 30)	COPD with Osteoporosis (N = 30)
MMP-9 (ng/ml)	32.54  ± 10.06	49.14  ± 11.34*	81.26  ± 20.61**^
TIMP-1 (ng/ml)	178.35 ± 25.95	188.14 ± 33.01	210.05 ± 38.34
MMP-9/TIMP-1	0.215  ± 0.124	0.278  ± 0.106	0.401  ± 0.207*^
TNF-α (ng/ml)	21.14  ± 3.57	33.58  ± 7.36*	42.63  ± 8.76*^
OPG (pg/ml)	239.35 ± 57.24	260.35 ± 46.37	272.63 ± 69.52*
RANK (pg/ml)	79.38  ± 20.16	75.37  ± 18.27	71.55  ± 36.76
RANKL (pg/ml)	439.56 ± 75.37	518.46 ± 48.58	582.33 ± 67.32*^
RANKL/OPG	1.88   ± 0.39	2.05   ± 0.62	2.39   ± 0.51*^

*P < 0.05, **P < 0.01 vs COPD with normal BMD.

^P < 0.05 vs COPD with low BMD.

Table 3.  Multiple regression analyses for Bone mineral density in COPD patients

		R^2	β	SE	P
Lumbar spine BMD	constant	1.658	0.219	0.000
	RANKL	0.179	−0.001	0.000	0.030
	mmp-9	0.033	−0.014	0.012	0.034
Femoral neck BMD	constant	1.393	0.096	0.000
	mmp-9	0.380	−0.006	0.001	0.000
	RANKL/OPG	0.139	−0.130	0.046	0.006

Correlation of FEV1%Pre with serum MMP-9/TIMP-1, TNF-α, and OPG/RANK/RANKL levels

Pearson analysis revealed that FEV1%Pre was negatively correlated with serum of MMP-9 (r = −0.36, P < 0.05), RANKL (r = −0.21, P < 0.05), and RANKL/OPG ratio (r = −0.23, P < 0.05) but not other cytokines (TIMP-1, TNF-α, OPG, RANK).

BMD, FEV1%Pre, and other parameters of the correlation analysis

Pearson analysis revealed a significant correlation between FEV1%Pre and the lumbar spine BMD (r = 0.42, P < 0.01) and femoral neck BMD (r = 0.48, P < 0.01); MMP-9 was positively correlated with TNF-α (r = 0.35, P < 0.05) and RANKL/OPG ratio (r = 0.27, P < 0.05) but did not correlate with RANKL.

Discussion

In the present study, COPD patients with osteoporosis showed high levels of serum MMP-9, which was negatively correlated with lung function FEV1%Pre, and lumbar spine and femoral neck BMD. These results were consistent with the reports by Higashimoto et al. and Olafsdottir et al (26–27). Both observed a negative correlation between the serum level of MMP-9 and FEV1. The results of the present work are also consistent with those of a study by Bolton et al., who reported that MMP-9 levels were higher in patients with osteoporosis than in healthy subjects but that TIMP-1,2 levels were not (9). The present results are also consistent with data published by Qin et al., which show that, in postmenopausal osteoporotic women, serum MMP-9 levels increased as BMD decreased (28). Nyman J.S. et al. analyzed the tibial metaphyseal trabecular bones of 16-week-old wild-type female mice and found defective MMP-9 to be associated with increased density of trabecular connectivity. When MMP-9 and MMP-2 are overexpressed, the vertebral trabecular bone becomes thinner and the number of trabecular bones decreases. This manifests as osteoporosis (29).

Our findings also indicate that the MMP-9/TIMP-1 ratio is higher in COPD patients with osteoporosis than in those with normal BMD. It was also negatively correlated with lumbar spine and femoral neck BMD. This could be due to the high level of MMP-9 expression in activated osteoblasts during osteoporosis. In contrast, TIMP-1 was not higher in those with osteoporosis. The MMP-9/TIMP-1 ratio imbalance may increase osteoclast cell activity but it may also weaken the osteoblasts, leading to irreversible bone loss and so causing osteoporosis. Our results suggest that the MMP-9/TIMP-1 ratio imbalance is associated with BMD and may play a role in the development of osteoporosis secondary to COPD.

The concentrations of inflammatory cytokines, including IL-1β, IL-6, and TNF-α, are increased in COPD patients (6, 25, 28). This causes bone loss as the cytokines affect osteoclast activity (24, 30, 31). In the present study, the serum levels of the inflammatory cytokine TNF-α were found to be higher in COPD patients with osteoporosis than in COPD patients with normal or low BMD. They were also positively correlated with MMP-9 (r = 0.35, P < 0.05). These results confirm the presence of systemic inflammation in COPD. They are also consistent with recent reports stating that the level of TNF-α is directly correlated with the levels of the bone resorption marker beta-Crosslaps(betaCL) and MMP-9 (32).

However, our results showed that only the serum level of MMP-9, not that of TNF-α, had any significant correlation with the lumbar spine or femoral neck BMD or with FEV1%Pre. This suggested that TNF-α may not be directly involved in the bone loss or impaired pulmonary functions observed in COPD patients. It has been reported that TNF-α can induce MMP-9 in normal human bronchial epithelial cells via the NF-kappa B-mediated pathway (33). Because TNF-α did not show any correlation with lung function or the BMD of the lumbar spine or femoral neck, we speculate that TNF-α may not directly impair lung function or cause bone loss but may instead induce or potentiate MMP-9 in osteoclasts, leading to osteoporosis in COPD patients. This should be verified in further studies.

The OPG/RANK/RANKL system plays crucial roles in bone metabolism and the pathogenesis of osteoporosis in patients with COPD (34). RANKL and OPG are synthesized by stromal cells and osteoblasts, which control osteoclast differentiation, maturation, and activation. OPG is a secreted glycoprotein that lacks transmembrane domains. It is expressed in a wide range of tissues, but most highly in those of the bones, heart, lungs, kidney, stomach, small intestine, skin, brain, and spinal cord. OPG knockout mice show significant reductions in bone mass, but mice that over-express OPG show fewer osteoclasts than usual and increased bone mass (35).

RANK is present in osteoclast precursor cells and the surface of the mature osteoclasts. It is the only natural receptor for RANKL. Upon binding with RANK, RANKL mediates the differentiation of osteoclast cell precursors into mature osteoclasts, promotes the activation of mature osteoclasts, and inhibits apoptosis of osteoclasts, so promoting bone resorption. OPG is a decoy receptor for RANKL. Specifically, it is an endogenous receptor antagonist of RANKL, which binds to three subtypes of RANKL, thus neutralizing all of RANKL's biological effects and blocking the binding of RANKL to RANK. This inhibits osteoclast cell development and bone absorption (35–37). In this way, RANKL and OPG promote and inhibit osteoclast differentiation, respectively. Imbalances in the levels of RANKL and OPG can disrupt bone metabolism, resulting in reduced bone mass or osteoporosis (19). In this way, the RANKL/OPG ratio is closely related to osteoclastogenesis and ultimately affects BMD and bone strength. The present study showed that serum levels of RANKL, OPG, and RANKL/OPG were significantly higher in COPD patients with osteoporosis than in those without osteoporosis. Results also showed the lung function FEV1%Pre, lumbar spine BMD, and femoral neck BMD to have negative correlations with the serum level of RANKL and RANKL/OPG ratio but no correlation with other cytokines (TIMP-1, TNF-α, OPG, or RANK). These results are consistent with those of a study published by Duckers et al., which showed that serum levels of OPG were significantly higher in COPD patients with low BMD or osteoporosis than in COPD patients with normal BMD. Serum levels of OPG were also found to be negatively correlated with hip BMD (6). Similar results were observed in the present work. Specifically, both the serum level of OPG and the RANK/OPG ratio were higher in COPD patients with osteoporosis than in those with normal BMD (P < 0.05), suggesting an imbalance in the OPG/RANK/RANKL system in COPD patients, which increases bone resorption, which may lead to osteoporosis.

Limitations of this study

Firstly, We choosed male patients, female patients were excluded from the study to avoid the influence of postmenopausal osteoporosis, though female patients with COPD have been reported have a higher incidence of osteoporosis; Secondly, the type of COPD may have associations with BMI and potentially inflammatory markers. Specifically, chronic bronchitis patients usually having greater BMI than patients with mostly emphysema. However, we have not enough study subjects to have appropriate power to do so; Thirdly, in the general population, the causes of osteoporosis include age, smoking, gender, vitamin D deficiency, hypogonadism, reduced physical activity, and sarcopenia, et al (38). We did not evaluate other possible causes of osteoporosis in COPD in the current study. These associations suggest a possible role for MMP-9 and OPG/RANK/RANKL in the pathogenesis of osteoporosis in COPD, but that further work specifically evaluating the mechanism is needed.

Conclusion

In conclusion, the present study demonstrated that the level of MMP-9, MMP-9/TIMP ratio, TNF-α, RANKL, OPG, and RANKL/OPG ratio were all higher in COPD patients with osteoporosis than in those with normal BMD or osteopenia. The BMD and FEV1%Pre were negatively correlated with MMP-9, RANKL, and the RANKL/OPG ratio but not with TIMP-1, TNF-α, OPG, or RANK. Positive correlations were observed between FEV1%Pre and BMD and between MMP-9 and TNF-α. Because each of these factors, including MMP-9, TNF-α, and the OPG/RANK/RANKL system, play important roles in the occurrence and development of osteoporosis in COPD patients, the interrelationships among them suggest that they may collectively or synergistically mediate both bone and lung parenchymal destruction. To our knowledge, this is the first study that included MMP-9/TIMP-1, TNF-α, and the OPG/RANK/RANKL system, the three major critical factors of bone metabolism, to identify their correlations and the cause-and-effect interrelationships inwith bone density in COPD patients. These results suggest that these factors may be important in the pathogenesis of the systemic complications associated with COPD. Further studies will be needed to determine the exact mechanisms of the processes underlying COPD-related bone loss.

Competing Interests

The authors declare that they have no competing ­interests. This study was supported by National Natural Science Foundation of China (81270087, 81270089); National Program on Key Basic Research Project (973 Program, 2012CB518200).

Figure 1.  Comparison of Serum levels of MMP-9/TIMP-1, TNF-α among patients with COPD. (*P < 0.05)

Figure 2.  Comparison of Serum levels of OPG/RANK/RANKL among patients with COPD. (*P < 0.05)