There is no general consensus on the definition of adhesive capsulitis on the basis of shoulder motion. In the present study, the diagnosis of adhesive capsulitis required gradual onset of shoulder pain accompanied by simultaneous restriction of active and passive shoulder motion. Deficits in shoulder range of motion in abduction, forward flexion, external rotation, and internal rotation were recorded and were added together to determine the total range of motion deficit. Patients were considered to have adhesive capsulitis if the total range of motion deficit was >50% of that of the opposite shoulder and shoulder pain was present at night. We considered the normal glenohumeral passive range of motion to be 90° of forward flexion, 90° of abduction, 90° of external rotation, and 90° of internal rotation20.
Patients treated at our hospital between June 2006 and December 2008 who met the above criteria for adhesive capsulitis were eligible for enrollment in the study. Patients who had been treated with oral corticosteroids or with shoulder injections were excluded. Patients with adhesive capsulitis secondary to rotator cuff tear, fracture, rheumatoid arthritis, or prolonged immobilization of the shoulder were also excluded.
The ICAM-1 level was evaluated in the joint capsule, fluid, and peripheral blood of patients with adhesive capsulitis and control patients. Markers of inflammation and fibrosis were also measured in normal human synoviocytes cultured in the presence and in the absence of ICAM-1. The study protocol was approved by the institutional review board of our hospital, and informed consent was obtained from the patients.
RNA Extraction and Oligonucleotide Array Analysis of the ICAM-1 Gene in the Glenohumeral Joint Capsule and Synovium
Twenty-six patients were enrolled; seventeen patients (mean, age 52.6 years; range, thirty-seven to sixty-six years) had adhesive capsulitis, and nine control patients (mean age, 40.1 years; range, twenty-five to sixty-four years) had a proximal humeral fracture or anteroinferior shoulder instability (see Appendix). Capsular tissue was harvested from the rotator cuff interval during the release of contractions in the superior and anterior capsule in both the patients with adhesive capsulitis and the controls. The capsular tissue was immediately put into RNAlater (Life Technologies, Grand Island, New York) to protect against RNA degradation. RNA extraction was performed within two days. Oligonucleotide array analysis for ICAM-1 was performed with use of the Oligo GEArray Human Extracellular Matrix and Adhesion Molecules Microarray (SABiosciences, Frederick, Maryland) as previously described21,22.
ICAM-1 Detection in the Glenohumeral Joint Capsule by Immunohistochemical Staining
The tissue samples were harvested from the rotator cuff interval of the seventeen patients with adhesive capsulitis and the nine controls; the control patients had been enrolled in a previous study involving oligonucleotide array and real-time reverse transcription-polymerase chain reaction (RT-PCR) analyses. The harvested capsular tissue was fixed in 10% buffered formalin overnight, washed, dehydrated with use of a graded series of alcohol solutions, and embedded in paraffin. Sections 4 μm in thickness were obtained from the tissue specimens. Hematoxylin and eosin (Sigma-Aldrich, St. Louis, Missouri) staining and immunohistochemical staining for ICAM-1 were performed. Sections were blocked for endogenous peroxidase activity with methanol containing 3% H2O2 and for nonspecific protein binding with Ultra V Block (Lab Vision, Fremont, California) for ten minutes. The primary antibody, mouse monoclonal anti-human ICAM-1 (sc-107; Santa Cruz Biotechnology, Santa Cruz, California), was added and incubation was performed for twelve hours at 4°C. After washing with phosphate-buffered saline solution (PBS), the biotinylated secondary antibody (VECTASTAIN Elite ABC Kit; Vector Laboratories, Burlingame, California) was applied for twenty minutes. Streptavidin-conjugated peroxidase was then added to catalyze chromogen development in 3,3′-diaminobenzidine tetrachloride (DAB).
ICAM-1 Gene Detection in the Glenohumeral Joint Capsule by Real-Time RT-PCR
Quantitative real-time RT-PCR was performed with use of SYBR Green premix reagent and an iCycler thermal cycler (both from Bio-Rad Laboratories, Hercules, California). Calculation of relative messenger RNA (mRNA) levels was based on the 2−ΔΔCT method and was performed with use of the iCycler software23. The value obtained for each gene was normalized against the 18S ribosomal RNA housekeeping gene.
The ICAM-1 primer sequences were 5′-TAT GGC AAC GAC TCC TTC T-3′ for the forward primer and 5′-CAT TCA GCG TCA CCT TGG-3′ for the reverse primer. The 18S primer sequences were 5′-CGG CTA CCA CAT CCA AGG AA-3′ for the forward primer and 5′-GCT GGA ATT ACC GCG GCT-3′ for the reverse primer. The reaction conditions were 95°C for three minutes followed by forty cycles of 94°C for twenty seconds, 58°C for twenty-five seconds, and 72°C for thirty seconds, with a final extension at 72°C for five minutes.
ICAM-1 Detection in Glenohumeral Synovial Fluid by Western Blotting
Synovial fluid (mean volume, 1.3 cm3; range, 0.9 to 2.2 cm3) was aspirated from the shoulders of six patients with adhesive capsulitis and two patients with anteroinferior glenohumeral instability, and was used for the measurement of the ICAM-1 level within the joint. The sample was lysed with ice-cold lysis buffer, and 25 μg of protein was separated with use of 9% SDS-PAGE (sodium dodecyl sulfate-polyacrylamide gel electrophoresis) and was transferred to a nitrocellulose membrane. The membrane was blocked with 5% fat-free dry milk and was treated overnight at 4°C with a 1:1000 dilution of the anti-human ICAM-1 polyclonal primary antibody (sc-7891; Santa Cruz Biotechnology). After washing, the membrane was incubated with a 1:3000 dilution of the peroxidase-conjugated anti-rabbit secondary antibody (Cell Signaling Technology, Beverly, Massachusetts) at room temperature for one hour. Detection was achieved with use of an enhanced chemiluminescence kit (GE Healthcare Bio-Sciences, Piscataway, New Jersey). The western blotting signals were measured by scanning densitometry and normalized against β-actin.
Detection of Soluble ICAM-1 in Peripheral Serum by Enzyme-Linked Immunosorbent Assay (ELISA)
The total level of soluble ICAM-1 was quantified in serum from thirty-two patients (ten male and twenty-two female; mean age, 52.9 years; range, thirty-two to sixty years) with adhesive capsulitis, twenty patients (eleven male and nine female; mean age, 49.8 years; range, thirty-six to fifty-six years) with diabetes mellitus, and fourteen normal controls (seven male and seven female; mean age, 48.3 years, range, thirty-seven to fifty-nine years).
The previously described criteria were used to establish the diagnosis of adhesive capsulitis, and those patients had no other related morbidities. Diabetes was diagnosed with use of the criteria of the American Diabetes Association: (1) a fasting plasma glucose level of >126 mg/dL, (2) a two-hour glucose level of >200 mg/dL on an oral glucose tolerance test, or (3) a random glucose level of >200 mg/dL with typical symptoms associated with diabetes. The control subjects had a fasting glucose level of <100 mg/dL. The diabetic patients and the controls had no shoulder symptoms and no limitation of shoulder motion.
The assay of each sample was performed in duplicate with use of an ELISA kit (EuroClone, Via Figino, Italy) according to the manufacturer’s protocol, and the absorbance was read at 450 nm.
Gene Expression of Cytokines Related to Fibrosis in Cultured Normal Human Synovial Cells After Treatment with ICAM-1 with and without Glucose
A human synoviocyte cell line (ScienCell Research Laboratories, Carlsbad, California) was cultured in synoviocyte medium (ScienCell Research Laboratories) supplemented with 10% fetal bovine serum (FBS). The cells were plated at a density of 4 × 105 cells per dish at the third passage. After serum starvation for twenty-four hours, the cells were divided into three groups: cells without ICAM-1 treatment that served as controls, cells treated only with 10 ng/mL of human recombinant ICAM-1 (R&D Systems, Minneapolis, Minnesota), and cells treated with ICAM-1 and 500 mg/dL of glucose (Sigma, St. Louis, Missouri). Cells were analyzed after twenty-four and seventy-two hours of culture in synoviocyte medium containing 2% FBS.
The genes for cytokines related to fibrosis, including CD44, matrix metalloproteinase (MMP)-9, MMP-14, tissue inhibitors of metalloproteinase (TIMP)-2, connective tissue growth factor (CTGF), and TGF-β1, were measured in each sample with use of real-time RT-PCT (see Appendix). The results were normalized against 18S rRNA. Three replicates were performed, and the results were averaged.
Statistical Analysis
The Student t test and the Mann-Whitney U test were used for comparisons between the adhesive capsulitis and control groups. A p value of <0.05 was considered significant.
Source of Funding
This research was supported by the Basic Science Research Program through the National Research Foundation (NRF) of Korea, funded by the Ministry of Education, Science and Technology (grant 2009-0074794), and by our Catholic Medical Center Research Foundation (program year 2006). The funds were used for the purchase of all chemical reagents and kits for this study.
RNA Extraction and ICAM-1 Gene Detection by Oligonucleotide Array Analysis
The quality of the extracted RNA was excellent; the average optical density (OD) ratio (260 nm/280 nm) of RNA from the joint capsule tissue of the patients with adhesive capsulitis was 1.96, and the average OD ratio of RNA from the controls was 1.89. Several genes for extracellular matrix and adhesion molecules were significantly increased in the oligonucleotide array analysis of the capsular tissue from the patients with adhesive capsulitis compared with the controls (p < 0.05). This included a consistent and significant increase in the gene for ICAM-1 (Fig. 1). The mean relative intensity (and standard deviation) of the ICAM-1 gene was 0.12 ± 0.01 arbitrary units in the patients with adhesive capsulitis compared with 0.09 ± 0.00 in the controls (p = 0.001).
Increased ICAM-1 in the Glenohumeral Joint Capsule and Synovial Fluid of Patients with Adhesive Capsulitis
The real-time RT-PCR data for the capsular tissue showed that the mean quantity of ICAM-1 mRNA was 1.70 ± 0.19 arbitrary units in the patients with adhesive capsulitis and 0.67 ± 0.24 in the controls (p < 0.05, Fig. 2). There were endothelial cells and a few chronic inflammatory cells such as lymphocytes in the perivascular area of the capsule and synovium of the patients with adhesive capsulitis, but there were few inflammatory cells in the tissue from the controls (Fig. 3). Immunohistochemistry for ICAM-1 showed increased staining of ICAM-1 about the endothelial cell layer in the synovium and capsule of the patients with adhesive capsulitis compared with the control patients (Fig. 3). The western blotting results revealed increased levels of the ICAM-1 molecule in the glenohumeral synovial fluid of the patients with adhesive capsulitis compared with the controls. ICAM-1 was visible as a dark band (85-110 kDa) on SDS-PAGE electrophoresis (Fig. 4-A). The mean ICAM-1 concentration in the synovial fluid was 1.70 ± 0.18 arbitrary units in the patients with adhesive capsulitis and 0.48 ± 0.17 in the controls (p < 0.05, Fig. 4-B).
Increased Soluble ICAM-1 in Peripheral Blood of Patients with Adhesive Capsulitis and with Diabetes by ELISA
The soluble ICAM-1 concentration was significantly increased in serum from the patients with adhesive capsulitis (633.22 ± 59.14 ng/mL) and patients with diabetes mellitus (671.25 ± 27.08 ng/mL) compared with the controls (359.86 ± 44.29 ng/mL, p < 0.05, Fig. 5). There was no significant correlation between the serum level and the synovial fluid level of ICAM-1 (r = 0.264, p = 0.736).
Increased Gene Expression of Cytokines Related to Fibrosis in ICAM-1-Treated Normal Human Synovial Cells
The OD (absorbance) was 0.874 in samples from the untreated cells, 0.902 in the cells treated with ICAM-1, and 1.125 in the cells treated with ICAM-1 and high glucose cells at one day of culture. The corresponding values at three days of culture were 1.156, 1.382, and 1.579, respectively. There was greater expression of genes for fibrosis-related cytokines (including CD44, MMP-14, TIMP-2, CTGF, and TGF-β1) in the synoviocytes treated with ICAM-1 only and in the synoviocytes treated with ICAM-1 and high glucose compared with the untreated synoviocytes at one and three days of culture (Fig. 6). Notably, at both one and three days of culture, the synoviocytes treated with only ICAM-1 manifested greater expression of the genes for CTGF and TGF-β1 compared with synoviocytes treated with ICAM-1 and high glucose.
ICAM-1 was increased in the capsular tissue and fluid of the glenohumeral joint as well as the serum of the patients with adhesive capsulitis compared with the control patients. Since the ICAM-1 gene was significantly increased in cultured capsular cells derived from patients with adhesive capsulitis compared with control patients and there were increases in not only the ICAM-1 protein but also the complementary DNA (cDNA) of the ICAM-1 gene in the capsule, it is conceivable that production of ICAM-1 may occur within the glenohumeral joint.
A previous study showed that activation of ICAM-1 is usually associated with shedding of this molecule into the general circulation, resulting in an increase in serum levels of the soluble form. Meliconi et al. suggested that ICAM-1 is expressed by all of the principal synovial cell types and that its wide distribution is associated with elevated soluble ICAM-1 levels in serum of patients with polymyalgia rheumatica18. The mean level of circulating soluble ICAM-1 has been reported to be between 102 and 450 ng/mL19, and it is elevated in association with inflammation, infection, and cancer16,17. In the present study, the mean soluble ICAM-1 concentration was 633.22 ng/mL in the serum of the patients with adhesive capsulitis. This finding suggests that the level of this molecule in peripheral blood may be a useful tool to monitor the status and prognosis of patients with this condition, although the clinical utility of measuring the level of circulating ICAM-1 remains to be established.
It is well known that there is a close relationship between adhesive capsulitis and diabetes24, and many studies have reported upregulation of the ICAM-1 level in patients with diabetes25,26. In the present study, the concentration of soluble ICAM-1 in serum was significantly higher not only in the patients with adhesive capsulitis but also in the patients with diabetes mellitus compared with the normal controls. However, it is also true that many patients with diabetes mellitus who have a high ICAM-1 concentration in the serum show no symptoms of adhesive capsulitis. This phenomenon suggests that adhesive capsulitis may be associated with multiple etiologies besides ICAM-1 elevation.
This study demonstrated that ICAM-1 could induce cell proliferation and an increase in the expression of the cytokines related to fibrosis in a normal human synovial cell line. This effect of ICAM-1 on synovial cells was exaggerated by the addition of high levels of glucose, and this could be one of the reasons for the high prevalence of adhesive capsulitis in the diabetic population. On the other hand, the expression of the growth factors TGF-β1 and CTGF was greater in the synoviocytes treated with only ICAM-1 than in the synoviocytes treated with ICAM-1 and high glucose. This result suggests that the expression of these growth factors was directly influenced by the ICAM-1 molecule rather than by the glucose level. It is well known that TGF-β1 and CTGF play major roles in the fibrosis process, as they accelerate cell growth and differentiation, collagen synthesis, and secretion of fibronectin and TIMP27-29.
This study has some limitations. First, the control specimens were acquired from patients with shoulder instability or proximal humeral fracture. The ideal control tissue for this study would have been harvested from a fresh cadaver with no medical history of a shoulder disorder. However, the loose joint capsule of patients with instability can be thought to represent the opposite of the tense joint capsule of patients with adhesive capsulitis, and the patients with humeral fracture had no history of shoulder impairment before the trauma. Previous studies have also used patients with shoulder instability as controls30,31. Second, the capsular tissue in the present study was harvested from only the rotator cuff interval between the supraspinatus and subscapularis tendons of patients whose adhesive capsulitis was at Neviaser32 stage II or III. Therefore, the results of the oligonucleotide array, RT-PCT, immunohistology, and tissue culture studies represent only this anatomic area, not the entire glenohumeral joint, and they represent only stages II and III of this condition. Many authors agree, however, that the primary intra-articular finding in adhesive capsulitis is the hyperemic synovitis about the rotator cuff interval5-8, and many studies have indicated that the pathology is particularly localized to the tissue of the rotator cuff interval7,33,34. Third, our data do not explain why the ICAM-1 molecule increased in patients with adhesive capsulitis or diabetes mellitus and why ICAM-1 treatment of cells increased the cytokines related to inflammation and fibrosis. In this study, we only observed the ICAM-1 concentration in several tissues and performed experiments to investigate the effect of ICAM-1 on the cultured cells.
In conclusion, ICAM-1 was significantly increased in synovial fluid and capsular tissue of the glenohumeral joint and peripheral blood of the patients with adhesive capsulitis compared with the control patients. The results of this study support the possibility that ICAM-1 can serve as a therapeutic target in the treatment of adhesive capsulitis.
Disclosure: One or more of the authors received payments or services, either directly or indirectly (i.e., via his or her institution), from a third party in support of an aspect of this work. None of the authors, or their institution(s), have had any financial relationship, in the thirty-six months prior to submission of this work, with any entity in the biomedical arena that could be perceived to influence or have the potential to influence what is written in this work. Also, no author has had any other relationships, or has engaged in any other activities, that could be perceived to influence or have the potential to influence what is written in this work. The complete Disclosures of Potential Conflicts of Interest submitted by authors are always provided with the online version of the article.