|
|
||||||||
CUTTING EDGE |








* Department of Protein Engineering,
Department of Immunology,
Department of Protein Chemistry,
Department of Pathology, and
¶ Department of Antibody Engineering, Genentech Inc., South San Francisco, CA 94080; and
|| Department of Bioinformatics, BIOTEC TU, Dresden, Germany
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
The role of chemokines has been best investigated in the immune system. They attract Ag-capturing and APC to tissues like skin and mucosal surfaces, which are primary sites of entry for pathogens to ensure immunosurveillance (3). To elicit an immune response after an infection with a pathogen, chemokines guide APC to secondary lymphoid organs, the focal meeting points of these cells with cells of the adaptive immune response, T and B cells. Furthermore, chemokines regulate homing of lymphocyte subtypes to subcompartments of lymphoid organs, a key function to ensure a coordinated immune response. Besides regulating cell trafficking, some chemokines have also been shown to play a role in processes such as angiogenesis and tumor growth (4, 5).
Many chemokines have been identified by their sequence signature motifs using sequence homology searches in databases (1). However, sequence homology-based methods fail as protein families become more diverse and remote homologues are difficult to identify below 20% sequence identity (6). Threading techniques utilize protein structural information to detect protein compatibility with known protein structures and, because they do not rely on sequence comparison, they are able to identify relationships even if sequence similarity is extremely low (7). In this study, we report the identification of a potential novel chemokine, DMC2 (dendritic cell (DC) and monocyte chemokine-like protein), by threading methods and its functional characterization.
| Materials and Methods |
|---|
|
|
|---|
DMC (AY358433) sequence was previously identified (8) but sequence analysis (BLAST & Pfam) failed to identify any statistically significant sequence homology with any previously characterized protein.
Generating structure-based protein function hypothesis
The fold recognition algorithm ProHit (ProCeryon Biosciences) and a fold library consisting of 7950 representative three-dimensional (3D) protein structures from the Protein Data Bank were used. Sequence-structure alignments were generated with the Smith-Waterman algorithm (9). Two mean force potentials describing the energetic forces between the residues of a fold and between residues and the surrounding solvent were used to calculate the sequence-structure fitness (10, 11). Default ProHit values for gap restrictions were used to control the number, size, and placement of gaps in the query sequence and the fold library entries. The BLOSUM40 amino acid substitution matrix was used for sequence comparison and scoring (12). For ranking and scoring we used: 1) energy score derived from residue-residue and residue-solvent interactions (pair/surf), 2) sequence similarity, 3) a normalized combination of pair/surf and sequence similarity (threading index (Th. Idx.)), and 4) ratio between the fold length (fold length) and the number of aligned residues in the sequence-structure alignment (path length) (fl/pl). Results were ranked by their Th.Idx. fl/pl was used to exclude sequence-structure alignments not covering the full length corresponding to a specific fold and to rule out possible false positives. Based on previous studies (data not shown), hits with values 0.6
fl/pl
1.3 were considered of a higher confidence. Corresponding 3D models were generated, and their respective sequence-structure alignments analyzed (ProHit and InsightII; Accelrys). The structural classification scheme SCOP (13) was used to build up a fold library containing all members of the IL-8-like fold family and to generate structure-based protein function hypothesis.
DMC protein expression and purification
His-tagged DMC was extracted from Escherichia coli inclusion bodies and purified on a Ni-NTA metal-chelate column. The Ni-NTA pool was applied onto a HiLoad 16/60 Superdex 75 gel filtration column (Amersham Biosciences) equilibrated with 20 mM MES (pH 6.0) containing 6 M guanidine HCl. Five milliliters of the Superdex 75 pool was treated with 50 mM DTT at pH 8.0, loaded onto a RP-HPLC Vydac C4 column equilibrated with 0.1% trifluoroacetic acid in water, and eluted with a gradient of acetonitrile (2537%) in 0.1% trifluoroacetic acid. DMC was lyophilized and dissolved in 1 mM HCl before dilution into assay buffer. A mutant form of DMC where all cysteines were converted to serines was generated. This His-tagged version was expressed and subjected to the same purification procedure as the wild-type protein. For baculovirus expression, His-tagged DMC was cloned into the plasmid PH.HIF and transfected into SF9 cells using Lipofectin (Invitrogen Life Technologies). After a 12-h culture in Hanks serum-free medium and 5 days in complete Hanks medium (Invitrogen Life Technologies), SF9 cells were infected with supernatant to generate a viral stock. A second amplification was done by infecting H5 cells in ESF921 medium (Expression System LLC) and DMC was purified using a Ni-NTA column.
Circular dichroism
DMC and CCL5 (R&D Systems) were dissolved in 1 mM HCl and diluted into PBS. Circular dichroism (CD) spectra were obtained in the far-UV range (190290 nm) using quartz cuvettes of 1-mm path length (Aviv model 62DS CD spectrometer; Aviv Associates). DMC was measured at 0.5 µg/ml and CCL5 at 0.1 mg/ml. Data were collected at 2.0-nm intervals with bandwidth 1.0 nm and at 25°C.
Cell isolation, culture, and treatments
Human PBMC were isolated by Hypaque-Ficoll density centrifugation and cultured at 37°C in 5% CO2 at 106/ml in RPMI 1640, 10% FCS, 2 mM L-glutamine, 1 mM sodium pyruvate, penicillin, and streptomycin, either unstimulated or activated with 20 µg/ml LPS (Sigma-Aldrich) for 48 h.
Ab generation and screening
Anti-DMC mAbs were generated by immunizations of mice with recombinant His-tagged DMC. For screening of the Abs, ELISA plates (Immunoplate Maxisorp; Nunc) was coated with 1 µg/ml DMC or with 1 µg/ml His-tagged control protein. After washing, the plate was incubated for 1 h with 50 µl of the test samples, normal mouse serum, anti-DMC Ab, or serum from mice immunized with his-tagged DMC, washed again, and followed by incubation with HRP-conjugated goat anti-mouse IgG Fc (catalog no. A-168; Sigma-Aldrich) for 1 h. The plate was washed, substrate (TMB BioFX Solution, catalog no. TMBT0100-01) was added, and reaction was stopped (TMB stop solution, catalog no. BSTP-0100-01; BioFx) as color developed.
Transwell migration assays and flow cytometry
PBMC transmigrated across 5-µm Transwell migration filters (Corning) for 2.5 h in response to a stimulus in the bottom chamber and were enumerated by flow cytometry. CXCL12 (R&D Systems) was used at 0.5 µg/ml. Subsets of migrated cells were analyzed by preblocking Fc receptors with human IgG (1 µg/106 cells; Sigma-Aldrich), followed by staining with Abs to CD3, CD14, CD11c, and CD16 (BD Biosciences) and analysis on a FACScan. Cells were preincubated for 30 min at 37°C in 5% CO2 with pertussis toxin (List Biological Laboratories). Mouse mAbs against DMC were generated in-house and used at 10 µg/ml. Anti-CCL2 Ab (clone 24822; R&D Systems,) was used at 10 µg/ml.
Northern blot analysis
DMC expression was analyzed with commercially available multiple human tissue Northern blots (BD Clontech).
Immunohistochemistry
Formalin-fixed, paraffin-embedded adult normal and inflamed lung, normal colon, and small intestine tissues were used for immunohistochemistry. Tissue sections were deparaffinized and hydrated. For Ag retrieval, slides were incubated in two rounds of Trilogy Ag retrieval solution (Cell Marque) at 99°C for 30 min. Endogenous peroxidase activity was quenched with 3% H2O2, then blocked with Vector Biotin Avidin Blocking reagents (Vector Laboratories), and the slides were blocked with 10% horse serum. Sections were stained with an in-house generated mouse anti-DMC mAb (clone 3H8) at 10 µg/ml, followed by a biotinylated horse anti-mouse Ab (Vector Laboratories) diluted 1/200 in blocking serum. For detection, Vectors ABC kit was used and metal enhanced diaminobenzidine (Pierce). Sections were counterstained with Mayers hematoxylin.
| Results and Discussion |
|---|
|
|
|---|
DMC is a novel secreted protein of unknown function (8). It is a basic protein (pI
10.9) of 119 aa and an estimated molecular mass of 13,819 Da. The putative signal peptidase I cleavage site is between positions 23 and 24 (Fig. 1A) and it has no potential Asn-linked glycosylation sites. Its chromosomal location is 19:47,624,87647,638,824.
|
|
These findings extend the CXC chemokine signature and may help to identify novel members of the group.
DMC specifically induces migration of monocytes and DCs
To test whether human DMC shows chemotactic activity, we expressed his-tagged DMC in E. coli and performed Transwell migration assays with human PBMC. DMC specifically induced migration of CD14+ monocytes and CD14CD11c+ DCs (Fig. 2A). Other PBMC subtypes, such as CD3+ T cells, CD16+CD3 D14 neutrophils, and NK cells (Fig. 2A), or B cells (data not shown) were not attracted by DMC. As seen for other chemokines, the dose-response curve of monocyte and DC migration to DMC in the migration assay formed a bell-shaped response curve. Because proteins expressed in E. coli are not always folded correctly, we also expressed his-tagged DMC protein in the baculovirus system and confirmed the chemoattraction of monocytes and immature DCs (Fig. 2B). A similar migration activity was observed with a nontagged version of DMC (data not shown). To investigate whether DMC also recruits activated DCs and monocytes, PBMC were stimulated with LPS in cell culture before migration assays. LPS completely inhibited migration of DCs (Fig. 2C) and monocytes (data not shown) to DMC. Furthermore, activation of monocytes with PGE2 and forskolin significantly reduced the response to DMC (data not shown). Heat inactivation of DMC abolished migration of nonactivated DCs and monocytes (data not shown), as did mutation of all cysteines of DMC into serines (Fig. 2D).
|
i protein-coupled receptor.
A panel of monoclonal mouse anti-DMC Abs was generated and selected by binding to DMC but not to the his tag (Fig. 3A). A subset of the Abs was further characterized, and several Abs were found to completely inhibit migration of monocytes and DCs to DMC as shown for clone 3H8 (Fig. 3B). Clone 3H8 is an IgG1 Ab with a
L chain. It also recognized DMC specifically by Western blot (data not shown). Anti-CCL2 Abs did not block migration to DMC (Fig. 3B). Furthermore, DMC protein bound specifically to monocytes and DCs but not to T cells as detected by flow cytometry using the monoclonal anti-DMC Ab and freshly isolated PBMC (data not shown).
|
Circulating blood DCs express a variety of chemokine receptors, CCR1, CCR2, CCR3, CCR5, and CXCR4 (20). They have been shown to respond to the CC chemokines CCL2, CCL8, CCL13, CCL5, and CCL11, which allows their recruitment to sites of inflammation (20). Migration of blood DCs to noninflammatory sites may be regulated by CXCL12, an ubiquitously expressed chemokine (21). CCL20, a chemokine expressed in noninflamed lung and liver, may recruit certain subsets of immature DCs into these tissues (22). However, unlike immature DCs generated in vitro from CD34+ bone marrow cells, blood DCs do not express CCR6 and therefore do not respond to CCL20 (23). We propose that DMC may fulfill this role.
DMC expression
Northern blot analysis using human multitissue blots showed that DMC is expressed in adult trachea, stomach (Fig. 4A), and fetal lung (Fig. 4B). Immunohistochemistry analysis of adult normal lung tissue sections demonstrated that DMC is constitutively expressed on bronchial and bronchiolar epithelium (Fig. 4C), as well as in a subset of alveolar lining cells (Fig. 4E). In addition, DMC expression was also observed in lung tissue from patients with asthma or obstructive pulmonary disease (data not shown). The expression levels and pattern were similar regardless of the inflammation status of the lung, suggesting that chronic inflammatory processes do not regulate DMC expression. However, acute inflammatory conditions have not been investigated yet. Furthermore, DMC expression was also observed in adult, normal small intestine (duodenum) and colon tissue sections (Fig. 4, G and I). Specifically, DMC staining is detected in the villus and some crypt epithelial cells of the small intestine (Fig. 4G) and in colonic epithelial cells, primarily at the luminal side (Fig. 4I). Constitutive expression of DMC in lung, stomach, colon, and small intestine supports a potential role for DMC as a housekeeping chemokine regulating recruitment of nonactivated blood monocytes and immature DCs into tissues. The presence of APCs at mucosal surfaces of these tissues is of biological importance for immunosurveillance of potentially harmful pathogens that may enter the lungs with air intake and the intestinal tract via food intake. If a pathogen invades, it will quickly activate the local APCs. Consequently, the local DCs mature, stop responding to DMC, and migrate to secondary lymphoid organs, where they activate the adaptive immune response to eradicate the pathogen.
|
| Acknowledgments |
|---|
| Disclosures |
|---|
|
|
|---|
| Footnotes |
|---|
1 Address correspondence and reprint requests to Dr. M. Teresa Pisabarro, Department of Bioinformatics, BIOTEC TU Dresden, Tatzberg 47-51, 01307 Dresden, Germany, and Dr. Kerstin N. Schmidt, Department of Immunology, Genentech, Inc., MS 34, 1 DNA Way, South San Francisco, CA 94080. E-mail address: mayte{at}biotec.tu-dresden.de and kerstin{at}gene.com, respectively. ![]()
2 Abbreviations used in this paper: DMC, dendritic cell and monocyte chemokine-like protein; DC, dendritic cell; 3D, three-dimensional. ![]()
Received for publication September 7, 2005. Accepted for publication November 28, 2005.
| References |
|---|
|
|
|---|
Cys mutant. J. Biol. Chem. 271: 3148-3153.
from lung dendritic cells. J. Exp. Med. 186: 825-835.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |