Published ahead of print on June 10, 2004, doi:10.1164/rccm.200401-002OC
© 2004 American Thoracic Society doi: 10.1164/rccm.200401-002OC
Enhanced Monocyte Chemoattractant Protein-3/CC Chemokine Ligand-7 in Usual Interstitial PneumoniaDepartment of Pathology, Department of Medicine, Division of Pulmonary and Critical Care Medicine, and Department of Radiology, University of Michigan Medical School, Ann Arbor, Michigan; Mayo Clinic, Scottsdale, Arizona; and Armed Forces Institute of Pathology, Washington, DC Correspondence and requests for reprints should be addressed to Cory M. Hogaboam, Ph.D., Associate Professor, Department of Pathology, University of Michigan Medical School, Rm. 5216B, Med Sci I, 1301 Catherine Rd., Ann Arbor, MI 48109-0602. E-mail: Hogaboam{at}med.umich.edu
Chemokines are increased and may exert effects on both inflammatory and remodeling events in idiopathic pulmonary pneumonia (IIP). Accordingly, we examined the concomitant expression of inflammatory CC chemotactic cytokines or chemokines and their corresponding receptors in surgical lung biopsies obtained at the time of disease diagnosis and pulmonary fibroblasts grown from these biopsies. By gene array analysis, upper and lower lobe biopsies and primary fibroblast lines from patients with usual interstitial pneumonia (UIP), nonspecific interstitial pneumonia, and respiratory bronchiolitis-interstitial lung disease, but not patients without IIP, exhibited CCL7 gene expression. TAQMAN, immunohistochemical, and ELISA analyses confirmed that CCL7 was expressed at significantly higher levels in UIP lung biopsies compared with biopsies from patients with nonspecific interstitial pneumonia, respiratory bronchiolitis-interstitial lung disease, and from patients without IIP. Higher levels of CCL7 were present in cultures of IIP fibroblasts compared with non-IIP fibroblasts, and CCL5, a CCR5 agonist, significantly increased the synthesis of CCL7 by UIP fibroblasts. Together, these data suggest that CCL7 is highly expressed in biopsies and pulmonary fibroblast lines obtained from patients with UIP relative to patients with other IIP and patients without IIP, and that this CC chemokine may have a major role in the progression of fibrosis in this IIP patient group.
Key Words: chemokine • chemokine receptor • idiopathic interstitial pneumonia • idiopathic pulmonary fibrosis Fibrotic lung diseases grouped in the category of idiopathic interstitial pneumonias (IIP) remain difficult to diagnose and a tremendous challenge to treat (1). Surgical lung biopsies (SLBs) are often taken for diagnostic purposes with the primary goal directed toward differentiating usual interstitial pneumonia (UIP) from IIPs with a more favorable prognosis such as nonspecific interstitial pneumonia (NSIP) and respiratory bronchiolitis-interstitial lung disease (RBILD), among others (2). The most common histologic pattern is UIP (also referred to as idiopathic pulmonary fibrosis [IPF] if the disease is idiopathic), for which treatment options are unsatisfactory (1), and patients with this form of IIP have a median survival of less than 3 years after diagnosis (1). Although considerable histopathologic data is gleaned from SLBs pertaining to the degree of cellularity and fibrosis (3), these biopsies may also provide important molecular clues as to the etiopathogensis of IIP. Because fibroblasts are readily grown from IIP SLBs, the abnormal growth and synthetic properties of these cells have also been examined (4).
The pathogenesis of IIP is believed to result, in part, from a miscommunication between inflammatory and structural cells, which is driven by the presence of several soluble factors including cytokines, chemokines, and growth factors (5). The list of chemokines that have been recognized in IIP is large, but little is known about the relative contribution of these mediators in the fibrotic process within the human lung (6). Approximately 94% of all known chemokines fall into two main families that are differentiated by their sequence homology and the position of two conserved cysteine residues in the N-terminus of these small ( Thus, the aim of the present study was to examine SLBs and primary fibroblast lines from patients suspected of having IIP with SLBs and primary fibroblast lines obtained from patients for diseases other than IIP using gene array, quantitative TAQMAN polymerase chain reaction, and ELISA analyses to compare CC chemokine and chemokine receptor expression. Given the recent prior focus on characterizing CXC chemokines in SLBs from patients with IIP and the importance of these factors in angiogenesis during pulmonary fibrosis (8), this study focused instead on the concomitant gene and/or protein expression of inflammatory CC chemokines (CCL1–CCL18) and their respective receptors (CCR1–5 and 8). Many of these CC ligands have been examined individually or in combination with CXC ligands in biologic samples from patients with IIP. The present study expands the findings of previous studies by revealing, for the first time, that monocyte chemoattractant protein-3 or CCL7 (according to its new designation) is increased in the context of UIP and the synthesis of this profibrotic CC chemokine (9) is augmented via CCR5 on UIP fibroblasts. In addition, the data presented herein suggest that the examination of concomitant inflammatory chemokine and chemokine receptor profiles in SLBs by specific gene array, TAQMAN, and specific protein ELISA may provide further valuable information of diagnostic significance in IIP.
Patients The Institutional Review Board at the University of Michigan Medical School approved this study and informed consent was obtained from each patient. The clinical characteristics of the forty-four patients studied are summarized in Table 1. These patients were referred by members of the Michigan End-Stage Lung Disease network with a suspicion of IIP based on a compilation of symptoms, physiologic symptoms, and radiographic findings. Additional detail regarding patient inclusion in this study is provided in an online data supplement.
Isolation and Culture of Primary Pulmonary Fibroblast Lines Pulmonary fibroblasts were grown from upper- and lower-lobe IIP and non-IIP SLBs and each primary fibroblast line was serially passaged five times to yield pure populations as previously described elsewhere (10).
Preparation of RNA and cDNA from SLBs and Pulmonary Fibroblast Lines
Gene Array Membranes
Real-Time TAQMAN Polymerase Chain Reaction Analysis
ELISA Analysis
Immunochemistry
Statistical Analysis
Pathologic and Clinical Features of Patients Studied Forty-four patients were enrolled in the present study and the baseline pathologic and clinical characteristics of this group are summarized in Table 1. Twenty-one of the patients exhibited histologic features consistent with UIP; 20 of these patients had concordant UIP, whereas 1 had NSIP in one lobe and UIP in another lobe (3). The cases of NSIP included two patients with a cellular pattern and four with a fibrotic variant. The non-IIP cases included five patients that were classified as nondiagnostic (i.e., histologically normal), three patients that were diagnosed with hypersensitivity pneumonitis, and one patient that was diagnosed with sarcoid.
Inflammatory CC Chemokine Receptor Gene Expression in SLBs from Patients with IIP and Patients without IIP: Gene Array Analysis
Inflammatory CC Chemokine Ligand Gene Expression in SLBs from Patients with IIP and Patients without IIP: Gene Array Analysis Several inflammatory CC ligands were present on each SuperArray gene array, thereby facilitating a concomitant analysis of these ligands with their corresponding receptors (Figure 2). The following inflammatory CC chemokine gene transcripts were present on the SuperArray: CCL1, CCL2, CCL3, CCL4, CCL5, CCL7, CCL8, CCL11, CCL13, CCL14, CCL15, CCL16, CCL17, and CCL22. The majority of these inflammatory CC chemokine ligand transcripts were present in both SLBs and primary fibroblast lines from all four patient groups. However, one notable exception was CCL7, which was detected in SLBs and primary fibroblast lines from UIP and NSIP patient groups and in SLBs alone from the RBILD patient group. However, gene transcripts for this inflammatory CC chemokine were absent in upper- and lower-lobe SLBs and primary fibroblast lines from the non-IIP group. Thus, this SuperArray analysis suggested that differences in CCL7 expression could differentiate IIP from non-IIP patient groups.
Inflammatory CC Chemokine Ligand Gene Expression in SLBs from Patients with IIP and Patients without IIP: Quantitative TAQMAN Polymerase Chain Reaction Analysis Given the suggestion that CCL7 gene expression differed between IIP and non-IIP SLBs, we turned to quantitative real-time TAQMAN polymerase chain reaction (PCR) and used this technique to explore whether CCL7 was indeed altered in IIP relative to non-IIP SLBs. A group of five other inflammatory CC chemokines (CCL2, CCL3, CCL5, CCL11, and CCL22) that were consistently detected in the SuperArray analysis was also examined in IIP and non-IIP SLBs for comparison. Quantitative TAQMAN PCR analysis was performed on upper and lower SLBs from all patients enrolled in the present study (Figure 3). Transcript levels of the CC chemokines shown were consistently increased twofold or greater in the IIP group relative to the non-IIP group, whereas the majority of other CC chemokine transcripts present on the SuperArray did not show this consistent fold-increase in expression or were not detected using real-time TAQMAN PCR. Higher fold increases in transcript expression (above non-IIP levels) for CCL2, CCL3, CCL5, CCL7, CCL11, and CCL22 were typically observed in SLBs from the UIP and NSIP patient groups compared with biopsies from the RBILD group. In addition, CCL7 and CCL22 gene levels were significantly increased in upper- and lower-lobe SLBs from the UIP patient group compared with the corresponding SLB from the non-IIP group. Thus, the quantitative real-time TAQMAN PCR analysis provided evidence that CCL7 and CCL22 transcript levels were significantly altered in IIP and that the greatest expression of both chemokines was present in UIP SLBs.
Inflammatory CC Chemokine Ligand Protein Expression in SLBs from Patients with IIP and Patients without IIP: ELISA Analysis We next examined whether the changes in gene expression for the CC ligands detected by TAQMAN PCR correlated with changes in protein levels for the same CC ligands. As shown in Figure 4, the expression of CCL2, CCL5, CCL7, and CCL22 in upper- and lower-lung biopsies varied among the IIP and non-IIP patient groups examined. Among the IIP patient groups, CCL2 was more abundantly expressed in SLBs from the patients with UIP and patients with NSIP compared with the RBILD and non-IIP groups, but these differences did not reach statistical significance. However, significant elevations in CCL5 were observed in upper- and lower-lobe SLBs from the NSIP group (compared with the appropriate non-IIP SLB). In UIP SLBs, both upper- and lower-lobe SLBs contained significantly greater levels of CCL7 and CCL22 protein compared with upper and lower SLBs from the non-IIP patient group (Figure 4). Upper- and lower-lobe SLBs from the UIP patient group contained 13.2 ± 5.1 and 12.0 ± 3.0 pg of CCL7/mg of protein, respectively, whereas upper- and lower-lobe SLBs from the non-IIP patient group contained 0.9 ± 0.9 and 1.3 ± 1.3 pg of CCL7/mg of protein. Upper- and lower-lobe SLBs from the UIP patient group contained 166.1 ± 34.6 and 157.7 ± 33.4 pg of CCL22/mg of protein, respectively, whereas upper- and lower-lobe SLBs from the non-IIP patient group contained 56.9 ± 28.5 and 48.9 ± 32.1 pg of CCL22/mg of protein. Accordingly, whole lung levels of CCL7 in UIP SLBs were increased approximately 12- to 13-fold above levels measured in non-IIP SLBs, in contrast to an approximate 3-fold increase in whole-lung levels of CCL22 in UIP SLBs relative to their non-IIP counterparts. Whole-lung levels of immunoreactive CCL3 and CCL11 were consistently below the limits of ELISA detection (data not shown). Thus, these data agreed with the TAQMAN quantitative PCR data showing that SLBs from the UIP patient group are characterized by a markedly increased expression of CCL7 and CCL22.
Immunohistochemical Detection of CCL7 in SLBs from Patients with IIP and Patients without IIP Although further examination of CCL22 was hampered by the lack of a suitable antibody for immunohistochemical analysis of this chemokine, the presence of CCL7 in SLBs was determined using routine immunohistochemical techniques (Figure 5). ELISA analysis suggested that CCL7 was present in greater abundance in UIP and NSIP SLBs compared with other IIP and non-IIP SLBs. The aim of this immunohistochemical analysis was to localize CCL7 protein in IIP and non-IIP biopsies. As shown in Figure 5, CCL7 was localized to interstitial areas of lung samples from UIP (Figure 5B, upper lobe, and Figure 5D, lower lobe) and NSIP (Figure 5F) patient groups, but CCL7 expression appeared to localize more frequently in distinct foci within upper- and lower-lobe samples from patients with UIP (Figures 5B and 5D). Immunoreactive CCL7 was also localized in vascular smooth muscle cells associated with vessels in NSIP (Figure 5F) and RBILD (Figure 5H) biopsies, but CCL7 was not detected in non-IIP control biopsies (not shown). Although both our TAQMAN and ELISA analyses indicated that CCL22 was significantly increased in SLBs from patients with UIP relative to biopsies from patients without IIP, we were unable to localize the expression of this chemokine using immunohistochemical analysis in any of the biopsies we examined. This technical conundrum prevented us from examining in greater detail the identity of the cells that may be sources of CCL22 in UIP. Thus, this data showed that CCL7 was localized to interstitial areas of the lung, particularly in UIP and NSIP SLB biopsies.
Immunocytochemical Analysis of CCR5 Expression by IIP and Non-IIP Pulmonary Fibroblasts SuperArray analysis gene analysis of primary fibroblast lines grown from upper- and lower-lobe biopsies from the RBILD patient group indicated that inflammatory CC chemokine receptor gene expression was nearly entirely absent in these cells. In contrast, all other primary fibroblast lines showed gene expression for CCR1–5 and 8. We therefore next examined whether pulmonary fibroblast lines grown from both IIP and non-IIP biopsies expressed inflammatory CC chemokine receptors at the protein level. Given that CCR5 was present in upper- and lower-lobe fibroblasts in the UIP, NSIP, and non-IIP patient groups and a highly specific and sensitive antibody was available for immunocytochemical detection, we focused on this CC chemokine receptor. Primary pulmonary fibroblast lines from patients with UIP (Figure 6) and from patients with other IIP and without IIP (Figure 7) were examined using routine immunocytochemical techniques to detect the presence of CCR5. Upper- and lower-lobe fibroblasts from patients with UIP exhibited very strong expression of CCR5, and this receptor was observed on the majority of the fibroblasts in culture (Figure 6). Although CCR5 expression was observed in cultures of NSIP (Figure 7B), RBILD (Figure 7D), and non-IIP (Figure 7F), this expression was less dramatic than that observed in cultures of UIP fibroblasts. Taken together, this analysis showed that human fibroblasts from all four patient groups expressed immunoreactive CCR5, thereby highlighting an as yet unexplained disconnect between transcript and protein expression for this chemokine receptor (and possibly other CC receptors) in primary human fibroblast lines. These data also emphasize the importance of concomitant analyses of the CC chemokines and their respective receptors at the transcript and protein levels.
ELISA Analysis of CCL7 Synthesis by Primary Pulmonary Fibroblasts We next determined whether fibroblasts were a source of CCL7, given that the immunohistochemical analysis of SLBs suggested that this chemokine was localized to interstitial areas. The synthesis of CCL7 by primary pulmonary fibroblast lines from upper- and lower-lobe IIP and non-IIP biopsies was analyzed, and the data from this analysis is summarized in Figure 8. The constitutive synthesis of CCL7 by upper- and lower-lobe IIP fibroblasts was consistently greater than that observed in cultures of non-IIP fibroblasts (Figure 8), but no differences were detected among the IIP fibroblast lines examined. When CCL22 at 10 ng/ml was added to cultures of IIP and non-IIP fibroblasts for 24 h, no changes in CCL7 synthesis by these cells was observed (Figure 8). However, the addition of 10 ng/ml of CCL5 significantly increased CCL7 generation by lower lobe UIP fibroblasts (Figure 8) compared with CCL7 generation observed in control cultures of these fibroblasts. The presence of CCL5 in cultures of upper lobe NSIP fibroblasts increased CCL7 generation twofold, but this increase did not reach statistical significance. We saw no evidence that pulmonary fibroblasts from patients with IIP and patients without IIP synthesized CCL5 or CCL22 (data not shown). These data suggest that pulmonary fibroblasts are a prominent source of CCL7, but CCL7 was found in greater abundance in cultures of IIP fibroblasts compared with non-IIP fibroblasts, and the expression of this chemokine by UIP and NSIP fibroblasts was augmented by the presence of CCL5.
IIPs are a heterogeneous group of disorders consisting of several clinicopathologic entities with differing histopathologic patterns, clinical course, response to therapy, and prognosis (2, 13, 14). It is now recognized that accurate diagnosis is required in IIP, particularly in distinguishing various clinicopathologic entities of this disease from UIP, the most common histologic pattern in IIP. With this recognition, an expert, international panel of pulmonologists, radiologists, and pathologists has recently suggested that examination of the SLB in isolation is no longer considered "gold standard" for diagnosis (2). However, SLBs have provided researchers with important clues as to identity of soluble mediators that may account for the subtype differences in IIP. Given that the CC ligands and CC receptors appear to be the altered the greatest during the development of experimental fibrosis (15), the present study focused on this chemokine family in clinical fibrosis. SuperArray molecular analysis showed that CCR1–5 and CCR8 were present in upper- and lower-lobe biopsy samples from the IIP and non-IIP patient groups, but gene expression for these receptors was absent in primary fibroblast lines grown from biopsies of patients with RBILD. Upon further molecular and protein analysis it was observed that CCL7 was significantly increased in SLBs from the UIP patient group. CCL7 was present in interstitial areas of UIP and NSIP SLBs, and primary fibroblast lines from patients with IIP were an excellent source of this chemokine. More importantly, the synthesis of CCL7 was enhanced by CCL5, another chemokine that is prominently expressed UIP SLBs. First identified as a fibroblast-derived chemokine (or fibroblast-induced cytokine [16]), the protein sequence of CCL7 shows 74% identity with CCL2, but secreted CCL7 protein differs from CCL2 in being N-glycosylated (17) and is a more potent activator of basophils (16). Similar to CCL2, CCL7 specifically stimulates the directional migration of T cells as well as monocytes via CCR2 (18). More recently, CCL7 was shown to be a major CC chemokine associated with experimental fibrosis (9). Thus, through the use of gene and protein analyses, the findings from the present study suggest that assessment of CCL7 levels in SLBs and primary fibroblast lines grown from these biopsies may aid in the differential diagnosis of IIP subsets. The observation that CCL5 protein was significantly elevated in NSIP SLBs was a novel and surprising finding in the present study because this CC chemokine and its major receptor, CCR5, have been predominantly associated with sarcoidosis (19). For the ELISA analysis in the present study, the non-IIP SLB protein data was derived from both upper and lower biopsies from nondiagnostic, sarcoid, and hypersensitivity pneumonitis patient groups, but we did not observe any differences in CCL5 levels amongst these three non-IIP groups. Our data may differ from previous studies due to the fact that we analyzed fewer patients with sarcoid or hypersensitivity pneumonitis. NSIP remains an IIP entity that requires better characterization, and encompasses a broad spectrum of features including alveolar wall inflammation (cellular), interstitial fibrosis (fibrotic), and mixed cellular and fibrotic (2). In general, the diagnosis of NSIP is associated with a more favorable prognosis than that of UIP. Differences in levels of cytokines, matrix metalloproteinases, and adhesion molecules have been observed between SLBs from NSIP and UIP patient groups, but the importance of these differences has not been explored. The observation that CCL5 is significantly elevated in NSIP SLBs may provide additional evidence that pulmonary events leading to NSIP differ from that of UIP and other forms of IIP. Further analysis of chemokine and chemokine receptors in SLBs from patients with NSIP is certainly warranted.
Given that the immunostaining of whole-lung biopsies from patients with UIP and patients with NSIP showed that CCL7 was predominately located within interstitial areas of lung, the present study also addressed whether cultured primary pulmonary fibroblasts were a putative cellular source of CCL7. All cultured primary fibroblast lines examined in the present study synthesized CCL7, but it was interesting to note that the presence of CCL5 in cultures of lower-lobe UIP fibroblasts significantly augmented the generation of this CC chemokine. This finding is particularly interesting given the marked increase in CCR5 expression that was observed on UIP fibroblasts; however, it is possible that CCL5 may have augmented CCL7 synthesis via CCR1 and/or CCR3 expression by these cells. CCR5 has been thoroughly characterized as a receptor for CCL3, CCL4, and CCL5; however, it has been shown that CCL7 could bind CCR5 with high affinity without eliciting a functional response (20). However, the latter effect of CCL7 was observed in the context of transfected Chinese Hamster Ovary cells and supraphysiologic concentrations (i.e., Research into the etiopathogensis of IIP has largely focused on elucidating the inflammatory mechanisms that initiate and/or maintain the fibrotic response in the lung, and strong arguments have been presented in support of this research direction (21). However, it is well documented that potent antiinflammatory or immunomodulatory therapies are largely ineffective in the treatment of severe forms of IIP, namely UIP and NSIP (22, 23). These failures have lead to the suggestion that these forms of IIP may be associated more with abnormal wound healing than with inflammation (23). Support for this emerging concept was observed in the present study in that it was shown that CCL7 was elevated in IIP and the pulmonary fibroblast appeared to be a major cellular source of this profibrotic chemokine. We provide evidence to support the diagnostic importance of examining this CC ligand and possibly others (i.e., CCL22) in biopsy samples from patients suspected of having IIP. Thus, further characterization of the patterns of CCL7 expression in IIP may aid in the development of promising anti–chemokine-based therapies in IIP.
The authors thank Robin Kunkel for her artistic assistance during the preparation of this manuscript.
Supported by a grant from the National Heart Lung and Blood Institute P50 HL56402 (C.M.H., S.L.K., K.R.F., F.J.M., and G.B.T.) and the Biomedical Research Council of the University of Michigan (C.M.H.). The Tissue Procurement Core of the University of Michigan Comprehensive Cancer Center also supported these investigations, in part, through grant CA46952. This article has an online data supplement, which is accessible from this issue's table of contents online at www.atsjournals.org Conflict of Interest Statement: E.S.C. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; C.J. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; K.J.C. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; S.L.K. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; H.E. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; F.J.M. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; K.R.F. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; G.B.T. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; T.V.C. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; E.A.K. received research funding from GE Medical Systems, CT Division, in 2003 totaling $40,000 and serves on the GE Medical Advisory Board for CT and serves on the Board of the GE Radiology Research Academic Fellowship receiving $2,500 annually, and received $1,500 for a lecture by GE in 2002; B.H.G. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; W.D.T. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; C.M.H. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Received in original form January 2, 2004; accepted in final form June 7, 2004
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
METHODS
8- to 14-kD) proteins. Presently, there are 16 CXC ligands (CXCL1–16) and 28 CC ligands (CCL1–28) that have been identified and recently ordered using a new classification scheme (
0.05 compared with appropriate lobe from the non-IIP patient group.
200 nM) of CCL7 were required to displace CCR5 agonists (i.e., CCL4) or prevent the binding of R5 strains of HIV (
