The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sizing, I. D.
Right arrow Articles by Rennert, P. D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sizing, I. D.
Right arrow Articles by Rennert, P. D.
The Journal of Immunology, 2007, 178: 2249-2261.
Copyright © 2007 by The American Association of Immunologists, Inc.

Epitope-Dependent Effect of Anti-Murine TIM-1 Monoclonal Antibodies on T Cell Activity and Lung Immune Responses

Irene D. Sizing, Véronique Bailly, Patricia McCoon, Wenjie Chang, Sambasiva Rao, Lourdes Pablo, Rachel Rennard, Meghan Walsh, Zhifang Li, Mohammad Zafari, Max Dobles, Leticia Tarilonte, Steven Miklasz, Gerard Majeau, Kevin Godbout, Martin L. Scott and Paul D. Rennert1

Biogen Idec Incorporated, Cambridge, MA 02142


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
The TAPR locus containing the TIM gene family is implicated in the development of atopic inflammation in mouse, and TIM-1 allelic variation has been associated with the incidence of atopy in human patient populations. In this study, we show that manipulation of the TIM-1 pathway influences airway inflammation and pathology. Anti-TIM-1 mAbs recognizing distinct epitopes differentially modulated OVA-induced lung inflammation in the mouse. The epitopes recognized by these Abs were mapped, revealing that mAbs to both the IgV and stalk domains of TIM-1 have therapeutic activity. Unexpectedly, mAbs recognizing unique epitopes spanning exon 4 of the mucin/stalk domains exacerbated immune responses. Using Ag recall response studies, we demonstrate that the TIM-1 pathway acts primarily by modulating the production of TH2 cytokines. Furthermore, ex vivo cellular experiments indicate that TIM-1 activity controls CD4+ T cell activity. These studies validate the genetic hypothesis that the TIM-1 locus is linked to the development of atopic disease and suggest novel therapeutic strategies for targeting asthma and other atopic disorders.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
Asthma is a multigenic heritable disease that is subject to environmental influence. Asthma and other atopic disorders are increasingly common chronic diseases in developing countries (1). Although the specific factors underlying the increase in asthma rates are not well understood, it has been hypothesized that the rise in asthma prevalence is linked to improved hygiene and to a dramatic drop in exposure to viral infections (2, 3). Such epidemiological observations provide a foundation for the hypothesis that the recent increase in asthma is due to a disruption in the normal induction of TH1 antiviral immunity, leading to a pathogenic shift to predominant TH2 immunity, which underlies atopic disease (4).

Effector CD4+ T cells provide the cellular framework supporting TH1 or TH2 immunity. TH1 T cells control cytotoxic responses that combat intracellular infections by producing proinflammatory cytokines including IFN-{gamma} and TNF. TH2 T cells produce cytokines that control extracellular infections, including IL-4, IL-5, IL-6, and IL-13. Dysregulation of TH2 cytokine production underlies the development of atopic disorders, including atopic dermatitis and asthma (4). In the asthma setting, TH2 cytokines drive eosinophil activation, IgE production, IgE-mediated mast cell activation and degranulation, and accumulation of mononuclear cells and granulocytes into lung interstitial space. T cells and activated granulocytes that traffic into the lung tissue continue to secrete TH2 cytokines, chemokines, and effector molecules, thereby fostering chronic lung inflammation (5).

Genetic linkage studies have highlighted the association of TH2 cytokines with the development of asthma. Particular attention has focused on a specific region of chromosome 5 (5q23–35) that contains a large number of cytokine pathway genes, including IL-9, IL-12p40, IL-4, IL-5, and 1L-13, and many other relevant genes (5). A recent congenic mouse study identified an additional locus at 5q23–35, adjacent to the cytokine loci, with significant linkage to asthma development. This locus, termed TAPR2 (T cell and airway phenotype regulator), cosegregated with IL-4 and IL-13 production by T cells and the development of airway hypersensitivity in congenic analyses (6). The locus contained a newly described gene family called the T cell, Ig domain, and mucin domain (TIM) family. Moreover, analysis of genetic variation in human atopic patients demonstrated that TIM-1 gene variability contributed to disease susceptibility in hepatitis A virus (HAV)-seropositive patients and in the general population (7, 8). Because TIM-1 functions as a primate HAV cellular receptor (9), this suggested that the activity of this protein underlies the inverse correlation of HAV incidence and atopic disease described in epidemiological studies (2). TIM-1 is also a renal epithelial cell protein that is overexpressed and shed in urine after various insults to the kidney (10, 11). The role of TIM-1 expression in kidney is unknown although it is tempting to speculate that it relates to the development of renal inflammation that occurs after kidney injury.

In this study, we have used anti-TIM-1 mAbs to modulate the activity of TIM-1 in immune response and asthma models. Our results show that TIM-1 activity influences the expression of TH2 cytokines by activated CD4+ T cells. The data support the genetic hypothesis inherent in the elucidation of the TAPR locus in mouse and TIM-1 genetic variation in human patients, and, furthermore, suggest mechanistic models for regulating TIM-1 activity. Finally, our examination of the domains of TIM-1 suggests novel therapeutic strategies for treating asthma and other atopic disorders by targeting specific features of the TIM-1 protein.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosures
 References
 
Mice

Six- to 8-wk-old BALB/c female mice were purchased from The Jackson Laboratory. Fischer rats were purchased from Charles River Laboratories. All animal experiments were approved by the Biogen Idec Institutional Animal Care and Use Committee.

Fusion proteins and Abs

Full-length mouse TIM-1 clones were obtained from BALB/c spleen RNA by RT-PCR amplification. The mouse TIM-1 extracellular domain, consisting of the IgV, mucin, and stalk domains, was amplified from mouse splenic RNA by RT-PCR using TIM-1-specific primers containing restriction enzyme sites (5'-GCGGCCGCTCTAGAATGAATCAGATTCAAGTCTTCATTTCAGGCCTCA and 3'-GTCGACGCCCTTAGTAGGGTTTTTCTGCGGC) and standard methods. This PCR product was cloned in frame with a human IgG-Fc construct in the pV90 vector, which is a pGEM 4Z-based vector. The resulting TIM-1-Fc construct was transfected into Chinese hamster ovary (CHO) cells and stable cell lines were selected based on growth in nucleoside-deficient medium.

A stable CHO line expressing 0.07 mg/ml TIM-1-Fc was established and used to generate supernatants for purification. TIM-1-Fc was purified using a protein A-Sepharose column in PBS equilibrated to pH 7. The column was washed with PBS (pH 5.0) and the fusion protein was eluted with PBS (pH 2.8) and the buffer was neutralized with 0.5 M sodium phosphate (pH 8.6). The eluted protein was submitted to a second purification step to isolate the intact form from aggregates and a clipped form that contaminated the sample. This second step was done on a column of Phenyl Fractogel equilibrated in PBS. The intact dimeric TIM-1-Fc bound to the resin and was eluted with 65% ethylene glycol, leaving the large aggregates on the column. The dimeric TIM-1-Fc sample was adjusted to pH 5 and further purified over a cation exchange column (Fractogel SE) eluted with 225 mM NaCl. Pools corresponding to the expected peak size for TIM-1-Fc were collected. Lastly, the purified protein was dialyzed against PBS (pH 7).

The clipped portion of murine TIM-1-Fc did not bind to the Phenyl Fractogel column equilibrated in PBS and was recovered in the flow-through. N-terminal sequence analysis showed that this protein was the product of an N-terminal truncation that cleaved the entire IgV domain and a small part of the mucin domain ({Delta}1–136aa), which we refer to herein as TIM-1-mucin/stalk-Fc.

The clipped form (lacking TIM-1 residues 1–136) was collected and submitted to a partial digestion with endo-AspN (1.3 µg of enzyme/800 µg of protein). Under these conditions (low enzyme concentration and nondenaturing buffer), the Fc portion of the protein stayed intact and linked to the last 21 residues of the TIM-1 extracellular domain, thereby encompassing the majority of the stalk region, which is ~30 aa long. The resulting fragment is referred to herein as the TIM-1-stalk-Fc fusion protein.

Murine TIM-1-IgV-Fc was cloned using TIM-1-specific primers containing specific restriction enzyme sites (5'-GCGGCCGCTCTAGAATGAATCAGATTCAAGTCTTCATTTCAGGCCTCA and 3'-ACTAGTGTCGACTGGTTTAACTTGCAATGAAAAGGTCACTTTCTGA), to amplify the TIM-1 IgV domain with Taq polymerase (Fisher Scientific) and the following amplification conditions: 95°C x 1 min, 25 x (95°C x 30 min, 68°C x 1 min), 68°C x 10 min. The resulting PCR product was cloned into a pCR 2.1 TOPO vector (Invitrogen Life Technologies) to make PEM088. A TIM-1-IgV-Fc fusion construct for 293EBNA expression was made by ligating the NotI-SalI TIM-1-IgV fragment from PEM088 plus the SalI-NotI human IgG1 Fc fragment from EAG409 into vector pNE001 to make PEM102. The NotI fragment encoding TIM-1-IgV-Fc was subcloned from PEM102 into the CHO expression vector pV90 digested with NotI to make the plasmid RR163. Murine TIM-1-IgV-Fc was expressed from a stable CHO cell line grown in spinner flasks and purified on protein A-Sepharose as described for TIM-1-Fc. The protein was submitted to a second purification step on an anion exchange resin (Fractogel trimethylaminoethyl) that bound any aggregates while the target protein was collected in the flow-through.

WEHI 7.1 cells were electroporated with the full-length murine TIM-1 cDNA cloned into the expression vector pCGC, along with a puromycin expression plasmid for selection. Cells were selected in puromycin-containing medium (RPMI 1640/10% FBS), tested for TIM-1 expression by FACS analysis, then sorted on a MoFlo (BD Biosciences) to isolate high-expressing clones.

Fischer rats were immunized with the TIM-1-Fc protein using standard procedures and, after several boosts to increase the anti-TIM-1 titer, the spleens were harvested and fused with SP2/0 to create the hybridomas. Individual clones were derived and screened by ELISA for reactivity with TIM-1-Fc, TIM-2-Fc, and purified human IgG. Clones reactive with TIM-1 but not TIM-2 or human IgG were selected for subcloning and further characterization by ELISA and by FACS analysis using 293 cells transfected with full-length TIM-1 or TIM-2 cDNAs.

Analysis of purified proteins

Five micrograms of various purified proteins were loaded per lane and resolved under nonreduced or reduced conditions using Novex 4-20% tris-glycine gels (Invitrogen Life Technologies). The gel was stained with Coomassie blue. For Western blot analyses, 5 ng of fusion proteins TIM-1-Fc, TIM-1-IgV-Fc, TIM-1-mucin/stalk-Fc, and TIM-1-stalk-Fc were loaded into wells of 4–20% SDS-PAGE gels (Invitrogen Life Technologies), electrophoresed, and blotted to 0.2-µm polyvinylidene difluoride (Immunoblot; Bio-Rad) for 4 h at 4°C at 0.4 amps in Towbin buffer (25 mM Tris, 192 mM glycine (pH 8.3) plus 20% methanol). Blots were blocked with 5% nonfat dry milk in TBS-N (TBS (pH 7.5) containing 0.2% Nonidet P-40) and stained by standard methods using 1 µg/ml indicated Ab followed by 1/1000 HRP-conjugated goat anti-rat secondary Ab (Jackson ImmunoResearch Laboratories) and detection using ECL (Amersham Biosciences). All Ab dilutions were made in 3% nonfat dry milk in TBS-N and all wash steps were done in TBS-N.

mAb characterization

ELISA plates (Nunc MaxiSorp) were coated overnight at 4°C with 5 µg/ml goat anti-human IgG-Fc-specific antisera (Jackson ImmunoResearch Laboratories) and then washed with ELISA wash buffer (PBS/0.05% Tween 20) before being blocked for 1 h at room temperature with blocking buffer (PBS/1% BSA). The blocked plates were incubated with TIM-Fc fusion proteins at a concentration of 1 µg/ml for 1 h at RT, washed, then blocked again for 1 h at room temperature before loading anti-TIM-1 mAbs for 2 h at room temperature, washing again, and then incubating with mouse anti-rat IgG (H + L)-HRP conjugate (Zymed Laboratories) diluted 1/5000 in blocking buffer. mAbs were titated across a 7 log concentration range to generate specific binding curves.

Peptides were prepared and stored in DMSO at 10 mg/ml, then diluted 1/1000 in carbonate coating buffer to coat the ELISA plate. mAbs were incubated with the peptide-coated plates, which were washed and developed as described above.

A BIAcore 2000 biosensor system (BIAcore) was used to study the binding of various rat anti-mouse TIM-1 mAbs to immobilized TIM-1-Fc and/or TIM-1-IgV-Fc. All experiments were performed at 25°C with a 10 µl/min flow rate. Each experiment was performed using HEPES-buffered saline buffer (10 mM HEPES, 150 mM NaCl, 3.4 mM EDTA, 0.005% P20 surfactant at pH 7.4). The same solution was used both as running buffer and as sample diluent.

The CM5 chip (BIAcore) surface was first activated with N-hydroxysuccinimide/N-ethyl-N'-(3-diethylaminopropyl)-carbodiimide hydrochloride (BIAcore). TIM-1-Fc or TIM-1-IgV-Fc, diluted to 30 µg/ml in 10 mM acetic acid (pH 5), was then injected. The unreacted groups of the chip’s dextran matrix were then blocked once with 30 µl and again with 15 µl of ethanolamine-HCl (pH 8.5). This resulted in a surface density of ~1500 resonance units for the experiments. The chip was regenerated with five 20-µl injections of 1 mM formic acid to establish a reproducible and stable baseline.

For the experiment, rat anti-mouse TIM-1 mAbs were diluted to 30 µg/ml in diluent buffer. For each run, 100 µl of one of the rat anti-mouse TIM-1 mAb was injected over the surface of the chip. Immediately after each injection, the chip was washed with 300 µl of the diluent buffer and regenerated between experiments with three injections (30, 20, and 10 µl) of 1 mM formic acid. After regeneration, the chip was equilibrated with the diluent buffer.

RT-PCR analyses

Cell lines were cultured in "complete" RPMI (containing 10% FBS (JRH Biosciences) and antibiotic/antimycotic solution (Invitrogen Life Technologies). Primary cells were isolated from mouse spleen or lymph nodes (LN) using specific negative selection columns (Cedarlane Laboratories). RNA was isolated from cells or tissues using the homogenizer method (Omni International) and TRIzol buffer (Invitrogen Life Technologies). TIM-family mRNAs were amplified using the First Strand RT-PCR kit (Invitrogen Life Technologies). Mouse TIM-1 was amplified following the RT step using specific primers (5'-CACTCCTCCAACATCTACACACACATGG and 3'-CTTAGAGACACGGAAGGCAACCACGCT). Mouse TIM-3 was amplified following the RT step using specific primers (5'-GTCTTACCCTCAACTGTGTCCTG and 3'-GTGTCTCTGAACCATT TCTCTCC).

Costimulation assays and lymphocyte activation

Thymocytes (2 x 106 cells/ml) were activated with 10 ng/ml PMA (Sigma-Aldrich) or with 10 µg/ml soluble anti-CD3 plus 20 U/ml murine IL-2 (R&D Systems). Splenocytes (2 x 106 cells/ml) were stimulated with 50 ng/ml PMA and 500 ng/ml ionomycin (both from Sigma-Aldrich). Cells were harvested at 24 and 48 h after stimulation and washed with FACS buffer (PBS/1% BSA/0.05% sodium azide). Cells (1 x 106) were incubated with mAb (2 or 20 µg/ml 1H9.9, 20 µg/ml 1H8.2, 20 µg/ml 3A2.5, or 5 µg/ml biotinylated 1C11.11) in 100 µl of FACS buffer containing 10% mouse serum for 1 h on ice. The cells were washed with 1 ml of FACS buffer, collected by centrifugation, then incubated for 30 min on ice with a 1/200 dilution of PE-coupled goat anti-rat IgG Ab or streptavidin-FITC (Jackson ImmunoResearch Laboratories). PE-coupled mAbs against mouse CD4 and B220 Ags (BD Biosciences) were used at a 1/250 dilution for costaining T and B cells. After washing, cells were resuspended in 75 µl of PBS, then fixed with 75 µl of 1% paraformaldehyde in PBS before analysis. FACS was performed on a FACScan and analyzed using FlowJo software (BD Biosystems).

CD4+ T cells isolated from LN using an affinity column (Cedarlane Laboratories) were cultured in complete RPMI in the presence of immobilized anti-CD3 mAb (1 µg/ml, clone 145-2C11; BD Biosciences) and soluble anti-CD28 mAb (1 µg/ml, clone 37.51; BD Biosciences). Splenocytes were cultured with soluble anti-CD3 only. Anti-TIM-1 mAbs were added at a final concentration of 10 µg/ml. After 24–72 h of incubation, the supernatants were harvested and analyzed for cytokine expression using the TH1/TH2 and Inflammation CBA kits (BD Biosciences). Parallel cultures were pulsed with 1 µCi of tritiated thymidine per well (Amersham Biosciences) for 8 h, then harvested and counted using the Microbetajet system (Wallac).

Keyhole limpet hemocycanin (KLH) response

BALB/c were immunized s.c. between the shoulders with 1 mg of KLH (Sigma-Aldrich) emulsified in 50 µg of CFA (Sigma-Aldrich). Five or 6 days later, the mice were sacrificed and the draining LN (brachial and axillary) were harvested. CD4+ T cells were isolated using affinity columns (Cedarlane Laboratories) and 200,000 cells/well were cultured in RPMI 1640 medium containing 10% FBS and varying concentrations of KLH in the presence of irradiated splenocytes (1000–1500 rad) from naive (untreated) syngenic mice. Each experimental condition was represented by three to four identical wells. Forty-eight hours later, the cell culture supernatants were harvested for analysis using CBA beads (BD Biosciences) and ELISA kits (R&D Systems). Parallel cultures were pulsed with 1 µCi of tritiated thymidine per well (Amersham Biosciences) for 8 h, starting at the 64th hour of culture, then harvested and counted using the Microbetajet system (Wallac). Mice treated with anti-TIM-1 mAbs were dosed i.p. with 200 µg of reagent 24 h before immunization and again on days 2 and 4. Eight mice per cohort were used in all experiments. In all of the in vivo experiments, nonspecific rat IgG2a mAb (BD Biosciences) was used as the isotype control.

OVA-induced lung inflammation and recall assays

Six- to 8-wk-old female BALB/c mice were given i.p. injections of 100 µl of 0.5 mg/ml OVA (grade V; Sigma-Aldrich) mixed with 100 µl of ImjectAlum (Pierce) on days 1 and 7. For the recall response assay, mice were rested for 14 days, at which time the spleens were removed, and lymphocytes were isolated for the recall assay (see below). For the lung inflammation model, mice were immunized twice as described above. Three weeks after the second injection, mice were exposed for 20 min daily for 3 days to an aerosol of 1% OVA in PBS using an ultrasonic nebulizer (Devilbiss). Two days after the final nebulization session, the mice were sacrificed for analysis. Bronchial lavage (BAL) fluid was collected via tracheotomy using three washes with PBS containing 0.1% BSA and 0.02 mM EDTA. BAL cells were pelleted using a cytospin and coated slides (Shandon) and then air dried and stained with Hema3 stain per the manufacturer’s instructions (Fisher Scientific) for identification of different cell populations. For analysis of relative total cellularity, a minimum of six fields were counted and the mean was determined. The lung tissue was harvested into neutral-buffered formalin for routine histology, or was snap frozen in TRIzol for subsequent RNA isolation. Draining (bronchial) LN were harvested and pooled for isolation of mononuclear cells. Five hundred thousand cells per well were placed into culture in RPMI 1640/10% FBS with varying concentrations of OVA. Each experimental condition was represented by three to four identical wells. Seventy-two hours later, the supernatants were harvested and cells were pulsed as described above. Supernatants were analyzed using CBA beads (BD Biosciences) and IL-13 ELISA (R&D Systems). Mice treated with anti-TIM-1 mAbs for full coverage were dosed i.p. with 200 µg of reagent 24 h before the first immunization and again on days 2, 4, 6, and 8 after. Mice were also given a single dose of 500 µg i.p. before the first nebulization session. Mice treated with anti-TIM-1 mAb 4A2.2 during the nebulization phase only were given a single dose of 500 µg before the first nebulization session. Eight mice per cohort were used in all experiments.

Immunohistochemistry (IHC)

Spleens obtained from immunized mice were fixed in 10% buffered formalin for 6 h followed by fixation in 70% ethanol. Fixed spleens were paraffin embedded and 6-µm sections were cut. For immunostaining, the slides were first deparaffinized in 100% xylene, followed by repeated washes in varying concentrations of ethanol and a final wash in distilled water. The sections were treated with 30% hydrogen peroxide in methanol to kill endogenous peroxide. After rinsing in PBS, the sections were unmasked by treating with pepsin (Zymed Laboratories) followed by a wash in PBS. An avidin/biotin block (Vector SP-2100; Vector Laboratories) was applied followed by blocking with 0.5% casein in PBS. The sections were then stained overnight with primary Ab (anti TIM-1 mAb 1H9.9 or anti-CD3 polyclonal sera; DakoCytomation)) at 4°C. The following day, the slides were washed in PBS and incubated for 1 h with biotinylated F(ab')2 of mouse anti-rat Ab (Jackson ImmunoResearch Laboratories). The slides were then incubated in avidin-HRP using Vectastain elite (Vector Laboratories). The HRP was developed using the Vector SK-4100 kit (Vector Laboratories) according to the manufacturer’s recommendations. The stained sections were examined using a Zeiss microscope (Zeiss) and the images were captured using an axiocam and analyzed using Openlab software.

CD4+ T cell response model

Five- to 6-wk-old mice were immunized with OVA/alum as described above on days 0 and 7. Different cohorts were pretreated with anti-TIM-1 mAb 1H8.2 or rat IgG2a control mAb, starting on day –5 and then dosed every 3–4 days with 200 µg i.p. Mice were sacrificed on day 14 and spleens were removed. Splenocytes were purified into CD4+ T cell and CD4– APC populations using enrichment or depletion magnetic beads (Miltenyi Biotec). Ex vivo cultures were set up in quadruplicate using 200,000 CD4+ T cells and 300,000 CD4– APC (irradiated at 1000 rad) per well from different treatment groups in the presence of Ag starting at 150 µg/ml, with serial dilutions of 1/3. Culture supernatants were collected at 48 h to assess cytokine production and the plates were pulsed with 1 µCi of tritiated thymidine for 6–8 h. Eight mice per cohort were used in all experiments.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosures
 References
 
Characterization of TIM-1 and related fusion proteins

TIM family members were named as T cell membrane glycoproteins with IgV and mucin domains (6). A distinct domain of ~30 aa containing N-linked glycosylation sites is present between the mucin domain and the transmembrane domain, and is referred to herein as the stalk domain. Two of the N-linked glycosylation sites in this domain are evolutionarily conserved across mouse, rat, nonhuman primate, and human proteins. Using standard PCR and cloning techniques, constructs for expression in CHO cells of the full-length extracellular domain and IgV domain-only TIM-1 fused to a human Fc{gamma} fragment were generated and are referred to as TIM-1-Fc and TIM-1-IgV-Fc, respectively. TIM-2-Fc, TIM-3-Fc, and TIM-4-Fc were generated using similar techniques. These fusion proteins demonstrated electrophoretic mobilities consistent with a heavy glycosylation state. TIM-1-Fc proteins appeared as large diffuse bands by SDS-PAGE, migrating at much larger molecular mass than predicted, and presented wide peaks by size-exclusion chromatography (data not shown). TIM-Fc proteins are susceptible to proteolysis, and a small amount of clipped fragments missing portions of one or more of the TIM-IgV domains were seen; these fragments accounted for <10% of any given preparation. Specific truncated forms of TIM-1 were generated from clipped full-length protein, as detailed in Materials and Methods, to yield highly purified TIM-1-mucin/stalk-Fc and Tim-1-stalk-Fc proteins (data not shown).

Characterization of mAbs to mouse TIM-1

B cell hybridomas were generated by fusing splenocytes of rats immunized with the TIM-1-Fc fusion protein. A panel of rat mAbs (all IgG2a) specific for murine TIM-1 was identified by ELISA analysis against TIM-1-Fc, TIM-2-Fc, TIM-3-Fc, TIM-4-Fc, or human IgG1-coated plates and FACS screening of 293 cells transfected with TIM-1 (data not shown). mAbs were further characterized by domain-specific Western blot and ELISA analyses. All of the anti-TIM-1 mAbs bound specifically to purified full-length TIM-1 protein in Western blot, surface plasmon resonance (SPR), and ELISA analyses. ELISA values are shown to illustrate and quantitate the results (Table I), which were consistent with the other analyses (data not shown). None of the anti-TIM-1 mAbs cross-reacted with murine TIM-2, TIM-3, or TIM-4 (data not shown).


View this table:
[in this window]
[in a new window]

  		Table I. ELISA analysis of the binding of anti-TIM-1 mAbs to TIM-1 domainsa

 
mAbs 1H9.9 and 4A2.5 bound with high affinity to the fusion protein containing only the IgV domain (Table I), showing that their epitopes resided within this domain. SPR analysis was used to demonstrate that these two mAbs did not cross-block and therefore recognize distinct epitopes (data not shown). mAbs 3A2.5 and 1H8.2 failed to bind to the fusion protein encoding the IgV domain only (Table I) and therefore appeared to require the presence of the mucin and/or stalk domains. mAb 1H8.2 was shown by ELISA and Western blot analyses to bind within the mucin domain, with residual activity observed with the stalk protein, suggesting that the epitope partially is contained in the stalk. Of interest, mAbs 1H8.2 and 5D1.1 failed to bind to a full-length form of TIM-1 corresponding to the DBA-1 allele defined in the TAPR analysis (Table I), suggesting that part of the epitope of those mAbs is encoded by exon 4 (i.e., within aa 164–184). mAb 3A2.5 bound fully within the stalk domain downstream of exon 4 (Table I).

Overlapping 18-mer peptides were used in ELISA to more precisely determine the epitope of mAbs 3A2.5, 5D1.1, and 1H8.2. This analysis confirmed that the epitope for mAb 3A2.5 lay within the stalk domain of TIM-1and revealed that the epitope was within the region delineated by two overlapping peptides covering aa 185–211 of the BALB/c form of the protein (Table II). Additional analyses revealed that 1H8.2 and 5D1.1 recognized epitopes upstream of 3A2.5, encompassing four to five exon 4-encoded amino acids as well as additional downstream amino acids (Table II). mAbs recognizing the IgV, mucin, and stalk domains were thus identified and characterized.


View this table:
[in this window]
[in a new window]

  		Table II. Binding of anti-TIM-1 mAbs to specific TIM-1 peptides by ELISAa

 
Expression of TIM-1 in mouse lymphoid cells and the effect TIM-1 Ab treatment on lymphocyte activation in vitro

Expression of TIM-1 has been reported on activated lymphocytes (6). We used RT-PCR and FACS analyses to confirm and extend these observations. Mouse thymocytes, splenic T and B cell populations, and A20 B lymphoma cells contained TIM-1 RNA (Fig. 1a). The WEHI 7.1 cell line was negative for TIM-1 mRNA expression, and therefore we used this line for stable cell line production. Expression of TIM-1 by stably transfected WEHI 7.1 cells was confirmed using mAbs 4A2.2 and 1H9.9, as compared with an isotype control mAb (Fig. 1b). Activation of thymocytes and splenocytes increased TIM-1 mRNA expression (Fig. 1a), and we confirmed TIM-1 protein expression in thymocytes using mAbs 1H9.9, 3A2.5, and 1H8.2, as compared with A20 cells (Fig. 1c). Finally, FACS analysis of mouse splenocytes activated for 48 h with PMA and ionomycin revealed distinct T and B cell populations that were positive for TIM-1 staining (Fig. 1d). These experiments demonstrate that TIM-1 is expressed on the cell surface of transfected cells and activated lymphocytes, and also show that the mAbs used in our studies are capable of binding cell surface-expressed TIM-1 protein.


Figure 1
View larger version (38K):
[in this window]
[in a new window]

  		FIGURE 1. Murine lymphoid cell expression of TIM-1. a, Left panel, RT-PCR analysis of freshly isolated splenocyte cell populations: B, B220+ B cells; 4', CD4+ T cells; 8', CD8+ T cells. Right panel, RT-PCR analysis of activated lymphocyte populations: A, two identical samples from A20 cells (used as a positive control), and 1–3', freshly isolated (lane 1) or activated (lanes 2 and 3) mouse thymocytes. Thymocytes were activated for 48 h with PMA (lane 2) or anti-CD3 and (lanes 3, 4, and 5); 4,5', freshly isolated (lane 4) or activated (lane 5) splenocytes. Splenocytes were activated for 48 h with PMA and ionomycin. b, FACS analysis of stably transfected WEHI 7.1/TIM-1 cells stained with anti-TIM-1 or isotype control mAb (blue). Staining of untransfected cells is shown as the shaded histogram, staining of transfectants with secondary Ab alone is shown in red. c, Left, FACS analysis showing staining of A20 cells with rat anti-TIM-1 mAbs 1H9.9 (black), 1H8.2 (red), and 3A2.5 (blue) and with rat IgG2a isotype control (shaded); right, activated thymocytes stained with anti-TIM-1 mAbs 1H9.9 (black) and 1H8.2 (red) or with rat IgG2a isotype control (shaded). Thymocytes were activated for 48 h with anti-CD3 and IL-2. d, FACS analysis of TIM-1 expression on CD4+ and B220+ splenocytes activated with PMA and ionomycin as detected with anti-mouse TIM-1.

 
Because TIM-1 was expressed on activated T cells and B cells, we tested whether the anti-TIM-1 Abs would affect lymphocyte activation. Splenocytes or LN CD4+ T cells were plated in 96-well plates with or without anti-CD3 Ab or anti-CD3 plus anti-CD28 Abs, respectively. Anti-TIM-1 mAbs were added at 2 µg/ml and cellular proliferation and cytokine secretion were compared with control (no treatment, isotype control mAb, or polyclonal human IgG) cultures. No difference in cellular proliferation or cytokine production was detected when anti-TIM-1 mAbs were added to the cultures, compared with controls (data not shown). Therefore, modulation of the TIM-1 pathway using these mAbs did not affect naive T cell activation in vitro.

TIM-1 expression in lung tissue and bronchial LN after induction of lung inflammation

Because the TAPR locus, and TIM-1 in particular, have been genetically linked to the development of asthma, we examined whether TIM-1 expression could be detected in the inflamed lung and lung draining LN. BALB/c mice were injected with chicken OVA/alum twice (on days 0 and 7), rested for 3 wk, and then exposed each day for 3 days to OVA aerosol using a nebulizer. We used RT-PCR to examine TIM-1 expression in lung tissue and draining (bronchial) LN. TIM-1 message was induced in both bronchial LN and lung tissue within 24 h after nebulization (Fig. 2, a and b). Because variants of both TIM-1 and TIM-3 were possibly linked to airway pathology in the initial study (6), we also looked for changes in TIM-3 message in these same tissues. In contrast to the enhancement of TIM-1 mRNA levels, TIM-3 mRNA levels were not modulated after challenge with OVA aerosol in either lung tissue or bronchial LN (Fig. 2, a and b). The level of TIM-2 mRNA was up-regulated in a manner similar to TIM-1 (12).


Figure 2
View larger version (108K):
[in this window]
[in a new window]

  		FIGURE 2. Expression of TIM-1 and TIM-3 during lung inflammation induced with OVA aerosol. Specific mRNA levels were assessed by RT-PCR. a, Induction of TIM-1, but not TIM-3, mRNA in bronchial LN 24 and 48 h after a 3-day aerosol challenge with OVA. b, Induction of TIM-1 and expression of TIM-3 mRNA in lung tissue 24 and 48 h after a 3-day aerosol challenge with OVA. c, IHC staining of TIM-1 in spleen. Mice were immunized on days 1 and 7 with OVA in alum, and organs were harvested on day 14. Sections from paraffin-embedded spleen were stained for TIM-1 or T cells (anti-CD3). The isotype control is a nonspecific rat IgG2a mAb. Clusters of TIM-1+ cells were identified near the border between the T and B cell areas in the immunized spleen (arrow).

 
We examined expression of TIM-1 in the spleen of immunized mice by IHC. mAb 1H9.9 brightly stained fixed and paraffin-embedded A20 cells, and therefore this mAb was chosen for IHC (data not shown). Mice were immunized with OVA in alum on days 1 and 7, and the spleens were harvested on day 14. TIM-1 expression was absent in the nonimmunized control spleen, but strong and localized in spleens from immunized mice (Fig. 2c). In particular, TIM-1-positive cells appeared to cluster near the border of T and B cell zones (Fig. 2c), a region likely to harbor activated T cells. In some spleen sections, a few TIM-1-positive cells were found near the central arteriole. Staining of TIM-1-positive cells was limited to a few T cell zones in each spleen examined. Double staining protocols will be required to identify the cell types that are TIM-1 positive in spleen. Furthermore, staining of bronchial LN and lung tissue harvested following nebulization did not consistently identify TIM-1-positive cells in these tissues, and further characterization of bronchial LN and lung harvested at different times following aerosol challenge will be required.

Effect of TIM-1 mAb 1H8.2 on the CD4+ T cell response to Ag in vivo

We began studying the activity of anti-TIM-1 mAbs using 1H8.2, which recognizes an epitope that lies partly within the region of the mucin domain encoded by exon 4. The presence of the exon 4-encoded mucin domain region reportedly varies at the genomic level between BALB/c and DBA/1 mice, as noted in the genetic analysis of the TAPR locus (6). We first assessed the activity of mAb 1H8.2 using the KLH Ag recall assay. BALB/c mice were treated with 1H8.2, rat IgG2a control mAb, or PBS, then immunized with KLH/CFA and 6 days later the draining LN were excised. LN CD4+ T cells were isolated and restimulated ex vivo with purified KLH in the presence of irradiated whole splenocytes isolated from untreated mice. Sixty hours after ex vivo stimulation, cellular proliferation was measured by incorporation of tritiated thymidine, and we found that mAb 1H8.2 dramatically increased T cell proliferation in response to KLH challenge ex vivo (Fig. 3a). Forty-eight hours after ex vivo stimulation, we measured cytokines produced in the cultures from cells treated with mAb 1H8.2 and found that the cultures contained higher levels of TH2-associated cytokines than did the control cultures, including 4-fold increases in the levels of IL-4, IL-5, and IL-10, and a 3-fold increase in IL-13 (Fig. 3b). Notably, the level of IFN-{gamma} also increased 6-fold (Fig. 3b). Levels of TNF, IL-2, and IL-6 were slightly elevated (2-fold). Cytokine values measured in the supernatants collected from the cultures incubated with 66 µg/ml KLH, a concentration representing a plateau point for both control and 1H8.2 proliferation curves; this point was used so that a large number of cytokines could be examined.


Figure 3
View larger version (22K):
[in this window]
[in a new window]

  		FIGURE 3. Recall response of draining LN CD4+ cells to KLH Ag. a, Proliferation of CD4+ cells isolated from control and 1H8.2-treated mice in response to challenge with KLH ex vivo. CD4+ T cells were cultured with varying concentrations of KLH in the presence of irradiated splenocytes, then pulsed with tritiated thymidine. b, Increased expression of TH2 cytokines by CD4+ T cells isolated from 1H8.2-treated mice in response to Ag challenge. Culture supernatants were harvested after 48 h of incubation with 100 µg/ml KLH. The experiments using mAb 1H8.2 were performed four times. Results shown are representative of these experiments. SEs of triplicate wells containing pooled samples are shown. *, p < 0.0001.

 
Effect of anti-TIM-1 mAb 1H8.2 on OVA-induced lung inflammation

Next, we examined the effect of anti-TIM-1 Abs on the induction of lung inflammation using the OVA model. mAb 1H8.2 induced robust eosinophil influx in the BAL of treated mice, such that the percent eosinophils was more than double that of controls after nebulization with Ag aerosol (Fig. 4a). Modest changes in the percentage of neutrophils and lymphocytes, which constituted a smaller fraction of BAL cellularity, were also noted (Fig. 4a), although these numbers were highly variable. Total cell number recovered from BAL increased as the total percentage of granulocytes increased; however, there was considerable variation in recovery of BAL in this experiment, making this measure of inflammation difficult to quantify (data not shown). Therefore, we used IHC to visualize cellular infiltration. Staining of lung paraffin sections from 1H8.2-treated mice showed large mononuclear infiltrates containing numerous CD3+ cells (Fig. 4b). H&E staining showed extensive inflammation scattered throughout the lung sections (Fig. 4b). Bronchial LN cells isolated from 1H8.2-treated mice and challenged with OVA ex vivo proliferated more and expressed higher levels of TH2-associated cytokines than did control cultures (Fig. 4, c and d). In particular, very high levels of IL-5 and IL-13 were produced, as compared with controls, although levels of IL-4, IL-6, and IL-10 were also elevated. IFN-{gamma} and TNF levels were also increased, although overall, levels of these cytokines were low. Cytokine values were measured in the supernatants collected from the cultures incubated with 100 µg/ml OVA, a concentration representing maximum proliferation for both control and 1H8.2-derived cultures; this point was used so that a large number of cytokines could be examined. In an independent experiment, cytokine values were measured across a wide range of OVA concentrations, yielding similar results (Fig. 5), except that at the very highest concentration no differences in the levels of TNF and IFN-{gamma} were observed. In the same model, mAb 5D1.1, whose epitope overlaps that of 1H8.2, had similar effects on BAL cellularity and mononuclear cell infiltration of lung tissue (data not shown).


Figure 4
View larger version (54K):
[in this window]
[in a new window]

  		FIGURE 4. Lung inflammation and Ag responses to OVA in anti-TIM-1-treated mice. Mice were immunized with OVA, rested, and then challenged with OVA aerosol. mAb-treated mice were dosed for full coverage, i.e., during the immunization and challenge phases. a, Cellularity in BAL of control or mAb-treated mice following OVA challenge; *, p = 0.008 relative to isotype control, using the test of mean equivalence. Cellularity was assessed by differential staining of BAL cytospins. b, Large cellular infiltrates containing numerous CD3+ cells in lung tissue of mice treated with mAb 1H8.2. c, Hyperproliferative response of LN cells from 1H8.2-treated mice to ex vivo stimulation with OVA Ag. d, Robust induction of TH2 cytokines by LN cells from 1H8.2-treated mice to ex vivo stimulation with 100 µg/ml OVA Ag. c and d, Means and SEs of quadruplicate (c) or triplicate (d) wells containing pooled samples are shown. **, p = 0.0007; *, p ≤ 0.0001.

 

Figure 5
View larger version (10K):
[in this window]
[in a new window]

  		FIGURE 5. Lung inflammation and Ag responses to OVA in anti-TIM-1-treated mice. Mice were immunized with OVA, rested, and then challenged with OVA aerosol. mAb-treated mice were dosed for full coverage, i.e., during the immunization and challenge phases. Induction of cytokine secretion by LN cells from 1H8.2-treated mice to ex vivo stimulation with various concentrations of OVA Ag. Cells were derived from the following cohorts of mice as labeled: {blacksquare}, 1H8.2-treated; {diamondsuit}, isotype control treated; {blacktriangleup}, negative control (PBS nebulized). **, p ≤ 0.005; *, p ≤ 0.0001.

 
Immune responses enhanced by 1H8.2 treatment are due to an effect on T cell activity

An ex vivo model was used to determine whether the impact of modulating TIM-1 function was dependent upon the T cell population, the APC compartment, or both. Mice were immunized with OVA in alum in the presence of 1H8.2 treatment or isotype control treatment. Splenocytes from each cohort of mice were pooled and divided into two samples: a highly enriched CD4+ population (>94% pure) and a CD4-depleted, irradiated, APC sample. Combinations of 200,000 treated or control CD4+ T cells with APC were restimulated with OVA ex vivo, and cell proliferation and cytokine production were measured (Table III). The response elicited in this assay was lower than that seen in the LN recall assay (Fig. 4d) in which a total of 500,000 LN cells from mice both immunized and then given three nebulization challenges were used. This likely reflects the much greater immune response generated after five exposures to Ag. Alternatively, the different extent of immune response and impact of 1H8.2 treatment may reflect different percentages of TH2 cells present in these different lymphoid organs after these different Ag challenge protocols. Nonetheless, the data (Table III) show that CD4+ T cells isolated from 1H8.2-treated mice were sufficient to induce the increase in cell proliferation and TH2 cytokine production seen in the OVA model, while this same activity was not demonstrated using the APC sample from 1H8.2-treated mice. These results demonstrate that the CD4+ T cell population was responsible for the altered immune response resulting from modulating TIM-1 activity.


View this table:
[in this window]
[in a new window]

  		Table III. Ex vivo response to OVA of CD4+ T cells and APC after Ag priminga

 
Identification of therapeutic anti-TIM-1 mAbs

The exacerbating activity of mAbs 1H8.2 and 5D1.1 on immune responses in the KLH and OVA assays led us to examine mAbs that recognized epitopes in other domains to assess their activity. In the OVA model, full coverage administration of mAb 3A2.5 reduced the percentage of eosinophils in the BAL (Figs. 4a and 6a) and reduced cellular proliferation and production of TH2-associated cytokines in the bronchial LN recall assay (Fig. 6, b and c). Total BAL cell recovery was variable. The percentage of neutrophils and lymphocytes was also variable and not statistically different from that of controls (Figs. 4a and 6a). In additional experiments, BAL recovery was further standardized to reduce this variability (see below). As noted above, mAb 3A2.5 recognizes the stalk domain of TIM-1 and, in particular, binds to a region containing a conserved N-linked glycosylation site (Tables I and II). We next examined the effect of mAbs directed to the IgV domain of TIM-1, focusing on mAbs 1H9.9 and 4A2.2, since these were shown in SPR analyses to recognize distinct epitopes within the IgV domain (data not shown). In multiple experiments, mAb 1H9.9 had no effect on lung inflammation, BAL cellularity, or the expression of cytokines by lung draining LN cells (Fig. 4a and data not shown). In contrast, mAb 4A2.2 markedly impacted lung responses in the OVA model, including reduced BAL cellularity as measured by cell number per field as well as percent eosinophils and lymphocytes (Fig. 7a). Inflammation of lung tissue as assessed by H&E staining and mucus production in the large airways was also reduced (Fig. 7b). Proliferation and dramatically reduced expression of TH2 cytokines by lung draining LN cells was also observed (Fig. 7c and data not shown). The cytokines shown (IL-4, IL-5, IL-10, and IL-13; Fig. 7c) were the most robustly expressed in this assay. A modest decrease in IL-6 secretion was also measured (roughly 2-fold), while secretion of TNF and IFN-{gamma} was unchanged; however, expression of these cytokines was very low (e.g., <10 pg/ml for IL-6 and IFN-{gamma}, data not shown).


Figure 6
View larger version (29K):
[in this window]
[in a new window]

  		FIGURE 6. Blockade of lung inflammation with anti-TIM-1 mAb. a, Cellularity in BAL of control or 3A2.5-treated mice following OVA challenge. b, Reduced proliferative response of LN cells from 3A2.5-treated mice to ex vivo stimulation with Ag. c, Reduced induction of TH2 cytokines by LN cells from 3A2.5-treated mice to ex vivo stimulation with Ag. b and c, The mean and SEs of triplicate wells containing pooled samples are shown. *, p < 0.0001 relative to isotype control, using the test of mean equivalence.

 

Figure 7
View larger version (52K):
[in this window]
[in a new window]

  		FIGURE 7. Anti-TIM-1 mAb 4A2.2 reduces lung inflammation and associated lung pathology. The negative control refers to mice immunized with PBS. a, Relative total cellularity and percent cell counts in BAL after challenge using the OVA lung inflammation model (n = 8/cohort). Mice were dosed with mAb for full coverage as described in the text. *, p < 0.0001; **, p < 0.0005; ***, p < 0.002, relative to isotype control, using the test of mean equivalence. b, Periodic acid-Schiff (PAS) staining of paraffin-fixed lung sections to stain mucus. Mice treated with isotype control mAb showed extensive mucus staining, compared with negative control mice, and mice treated with mAb 4A2.2. H & E staining at lower power is used to show the extent of cellular infiltration around the large airways. c, Reduction in the expression of cytokines by lung draining LN cells from mice treated with mAb 4A2.2 in response to soluble OVA ex vivo. Cytokines were measured in cell culture supernatants 48 h after the addition of OVA. The mean and SEs of triplicate wells containing pooled samples are shown. *, p < 0.0001; **, p < 0.001, relative to the isotype control, using the test of mean equivalence. The results are representative of three independent experiments.

 
BAL cellularity data was consistently impacted more by 4A2.2 treatment than by 3A2.5 treatment, although this difference was not statistically significant (Fig. 7a). SPR analysis indicated a faster off rate for mAb 3A2.5 than for mAb 4A2.2, and pharmacokinetic analysis showed a distinct difference in mAb half-life in vivo (3A2.5 at 1.9 days and 4A2 at 6 days). These physical properties may account for the apparent increased efficacy of 4A2.2 relative to 3A2.5, at least at the level of BAL cellularity. Alternatively, mAb access to the membrane proximal stalk region may be different from access to the presumably extended IgV domain.

We used 4A2.2 in additional studies to compare the effect of full coverage and nebulization phase treatment protocols on lung inflammation and pathology using the OVA model. Mice were immunized with OVA/alum as in the previous studies, but treatment with anti-TIM-1 mAb 4A2.2 was delayed until just before the first nebulization procedure. Administration of mAb 4A2.2 using this protocol resulted in reduced lung inflammation as measured by relative BAL cellularity and percent eosinophils and lymphocytes in BAL (Fig. 8a). Reduced lung inflammation as assessed by H&E staining of mononuclear cell infiltrates and reduced lung pathology as measured by mucus production in the large airways was also observed (Fig. 8b). In addition, draining LN cells produced less TH2 cytokines upon restimulation with OVA ex vivo (Fig. 8c), similar to levels seen in the previous experiments.


Figure 8
View larger version (46K):
[in this window]
[in a new window]

  		FIGURE 8. Treatment just before the nebulization phase with anti-TIM-1 mAb 4A2.2 reduces lung inflammation and lung pathology. The negative control refers to mice immunized with PBS. a, Percent cell counts and relative total cellularity in BAL after challenge using the OVA lung inflammation model (n = 8/cohort). Mice were treated just before the nebulization phase with mAb as described in the text. *, p < 0.0001; **, p = 0.005, relative to the isotype control, using the test of mean equivalence. b, Periodic acid-Schiff staining of paraffin-fixed lung sections to stain mucus. Mice treated (Figure legend continues) with isotype control mAb showed extensive mucus staining, which was reduced in mice treated with mAb 4A2.2. H & E staining at lower power is used to show the extent of cellular infiltration around the large airways. c, Reduction in the expression of cytokines by lung draining LN cells from mice treated with mAb 4A2.2 in response to soluble OVA ex vivo. Cytokines were measured in cell culture supernatants 48 h after the addition of OVA. The mean and SEs of triplicate wells containing pooled samples are shown. *, p < 0.0001, relative to the isotype control, using the test of mean equivalence. The nebulization the phase-dosing experiment was done twice, with very similar results.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosures
 References
 
One hallmark of asthma is inflammation in the lung. This critical process, triggered by antigenic insult and influenced by underlying environmental and genetic factors precedes and is responsible for disease pathology. A significant and compelling body of evidence supports the view that asthma-related lung inflammation is caused by dysregulated TH2 cytokine expression, although some recent mouse and human studies have suggested that this view of the disease is oversimplified (13, 14). This reassessment is underscored by the clinical failure of therapeutics targeting select TH2 cytokines (15). Such clinical results may however reflect the redundant functions of the TH2 cytokines, each of which is capable of supporting a diverse array of cellular responses (15), or the fact that the critical cytokine pathway has yet to be targeted in clinical trials. In this regard, therapeutics that modulate secretion of multiple cytokines, including, for example, IL-13, may be of particular interest. The multifactorial nature of the asthma response is reflected in the abundance of genetic loci reportedly associated with asthma phenotypes (5). Perhaps the most frequently identified asthma loci are in the region defined by 5q23–35, which contains the TH2 cytokine locus, and many other genes as well as key regulatory elements (5, 16). This region includes the TAPR locus containing the TIM gene family. Genetic analyses have shown that the TH2 phenotype is regulated by the TIM locus in mice (6). Importantly, a study of human patients showed that TIM-1 allelic variation influenced susceptibility to atopic disease in HAV-seropositive individuals (7).

We have prepared and characterized novel TIM-1-specific mAbs that have allowed us to examine cellular TIM-1 expression during immune responses. These experiments have demonstrated TIM-1 expression by thymocytes, activated B and T cells, and activated splenocytes in vivo. Furthermore, we have used the well-characterized model of OVA-induced lung inflammation to show that the TIM-1 pathway influences TH2 cytokine expression and the extent of lung inflammation. The OVA model is TH2 cytokine dependent, as shown by the many studies elucidating the critical role of TH2 pathways in OVA-induced inflammation, eosinophilia, IgE production, airway hyperresponsiveness, and airway remodeling (17). Importantly, because B cells do not play a role in mediating the immune response in the OVA model (18), we have been able to use this model to focus on the role of TIM-1 expression on T cells. Using this model, we have shown that modulation of TIM-1 function using specific mAbs profoundly influenced the development of lung inflammation. Exacerbation of lung inflammation, including influx of large numbers of eosinophils, Ag-specific hyperproliferation of lymphocytes, and robust expression of TH2 cytokines, was demonstrated using mAbs whose epitopes include a portion of exon 4, the exon responsible for the allelic variation described in the original TAPR study (6).

We were able to use the exacerbating activity of mAb 1H8.2 to demonstrate that the TIM-1-dependant hyperactivity observed in the OVA immune response was dependant on CD4+ T cells. Thus, CD4+ T cells isolated from immunized, 1H8.2-treated mice were capable of inducing the hyperactive immune response in culture. As seen in the prior experiments, this hyperactive immune response was characterized by expression of TH2 cytokines by CD4+ T cells in response to Ag stimulation.

mAb 1H8.2-mediated hyperproliferation and increased expression of TH2 cytokines was also seen in the KLH recall assay, demonstrating that the TIM-1 pathway influenced the immune response regardless of the adjuvant used. The increase in TH2 cytokine expression was not accompanied by decreased IFN-{gamma} expression; indeed, IFN-{gamma} levels were also somewhat elevated. Therefore, modulation of TIM-1 does not appear to impact the TH2 immune response by down-regulating the TH1 response; it appears instead that the TH2 response is preferentially supported, allowing its robust expansion even in the presence of a TH1-enhancing adjuvant such as CFA.

Importantly, we have also demonstrated down-regulation of lung inflammation and pathology using TIM-1-specific mAbs targeting distinct epitopes of the protein. Specifically, treatment with mAbs to the stalk domain (3A2.5) and IgV domain (4A2.2) inhibited OVA-induced lung inflammation and caused a large reduction in the level of TH2 cytokines produced by draining LN cells in the recall assay. Furthermore, dosing with mAb 4A2.2 just before nebulization was capable of blocking disease development, showing that anti-TIM-1 therapy is operative at the effector phase of the immune response. These data suggest that the mAb impacted T cell activity after initial priming and activation had taken place, i.e., during the immunization phase, and therefore that mAb 4A2.2 effects the activity of effector T cells. Our data suggest that the activity targeted is cytokine secretion, but other activities (trafficking, interaction with other cells) may also be affected. Pathological change to the lung tissue was also impacted, as shown by reduced mucus production in the lung airways. Because the baseline degree of inflammation and lung pathology observed in our experiments was milder than we would have expected, we are currently modeling anti-TIM-1 treatment using a more intense OVA-induced lung inflammation model that produces up to a 70% eosinophil influx in BAL. We are also extending these studies to airway hyperresponsiveness measurements using anti-cytokine mAbs (e.g., anti-IL-5, anti-IL-13) to benchmark the extent of the therapeutic effect in the acute OVA model as well as in the DO11.10 transfer model of lung inflammation.

With regard to the distinct in vivo activities of the TIM-1-specific mAbs reported thus far, it is interesting to consider how these effects reflect epitope specificities (Fig. 9a and Table I). Of note, mAbs 1H8.2 and 3A2.5 recognize distinct epitopes, each containing a conserved N-linked glycosylation site (Fig. 9 and Tables I and II), and these mAbs have opposing activities in the OVA inflammation model. Recently, it was reported that an anti-TIM-1 mAb that recognized an epitope in the IgV domain also exacerbated lung inflammation in a tolerance model (19). In contrast, another report identified an anti-TIM-1 mAb that blocked OVA-induced lung inflammation (20), similar to the results reported here with mAbs 3A2.5 and 4A2. We have analyzed the binding of the blocking mAb (20) to our domain-specific proteins and found it recognized the IgV domain (data not shown). Thus, it appears that mAbs targeting both the IgV or mucin/stalk domains of TIM-1 can exacerbate or block immune responses. These results suggest that TIM-1 secondary structure is complex and may involve the physical interaction of the IgV and mucin/stalk domains, perhaps involving O-linked or N-linked sugar residues. For example, 1H8.2 and 3A2.5 may differentially affect N-linked carbohydrate use by a coreceptor, ligand, or TIM-1 itself (Fig. 9 and Table II). In our hands, recombinant TIM-1 (and TIM-4) fusion proteins bind promiscuously to a wide variety of cell lines, and this has led us to be very cautious interpreting in vivo results generated with such molecules. Such promiscuous binding may be reflective of carbohydrate-mediated interactions of fusion proteins with cell surface moieties (V. Bailly, P. McCoon, and P. D. Rennert, unpublished observations).


Figure 9
View larger version (21K):
[in this window]
[in a new window]

  		FIGURE 9. a, Model of TIM-1 structural features that may be influenced by mAbs recognizing different epitopes. The TIM-1 IgV domain may be involved in multiple interactions, including ligand binding, coreceptor engagement, or protein folding. The TIM-1 mucin domain contains numerous O-glycosylation sites that could interact with other cell surface proteins such as galectins. The TIM-1 stalk domain contains two N-glycosylation sites that are conserved phylogenetically from rodent to primate. N-linked carbohydrates may influence coreceptor binding, protein folding, or other interactions. b, Sequences of the epitopes defined for mAbs 1H8.2 and 3A2.5 are shown. Each mAb binds to an epitope containing a conserved N-glycosylation site. Because overlapping peptide analysis was used to generate the epitopes, their actual 5' and 3' boundaries are not precisely known; therefore, the minimum epitope is shown.

 
TIM-4 has been reported to be a ligand for TIM-1 (21); however, we have been able to detect only very weak interaction of TIM-1 and TIM-4 by ELISA, and no detectable interaction by FACS or SPR analyses (data not shown). Several lines of evidence suggest that ligand binding of TIM-1 transduces a positive proliferative signal to T cells. Umetsu et al. (19) have shown that an anti-TIM-1 mAb, but not its Fab, can activate T cells in vitro, and another recent in vitro study has demonstrated that TIM-1 engagement increases transcription of IL-4 via the NFAT/AP1 pathway through activation of a conserved tyrosine phosphorylation site (22). Of additional note is the observation that defined TIM family-interacting proteins represent disparate protein families, including a semaphorin (TIM-2 ligand) and a galectin (TIM-3 ligand) in addition to TIM-4 (21, 23, 24). This suggests that multiple binding partners may exist for each TIM family member, and further suggest that both cis- and trans-interactions may be involved in regulating TIM protein activity. Galectins, for example, could very well interact at the cell surface with N-linked or O-linked sugars attached to the stalk and mucin domains of TIM-1, leaving the IgV domain free to interact with ligand(s) in trans. Such structural complexity provides a variety of hypotheses to be tested regarding the activity of specific anti-TIM-1 mAbs (Fig. 9a). Whether any of the proposed structural features outlined here also influence HAV binding to the human TIM-1 protein is not known, but this would appear to be an area for further productive investigation.

In summary, genetic analyses in mouse strains and in human patient populations have suggested that TIM-1 is an atopic disease susceptibility gene. Accumulating functional data support this genetic hypothesis. In this study, we have shown that the TIM-1 pathway influences CD4+ T cell activity in response to pulmonary Ag challenge by controlling the expression of multiple TH2 cytokines, and thereby controls the extent of the immune response, lung inflammation, and tissue pathology. We have identified a novel strategy for modulating TIM-1 activity to blunt the expression of TH2 cytokines and block lung inflammation. The epitope-mapping data suggest that candidate therapeutic anti-human TIM-1 mAbs will require careful analysis. However, the influence that the TIM-1 pathway has on effector T cell responses and associated pathology suggests that this pathway will be a productive target for the treatment of asthma, atopic dermatitis, and other TH2-driven diseases.


   Acknowledgments
 
We thank Jose Sanchez for FACS analyses; Chris Tonkin, Brittney Coleman, Michelle McAuliffe, and Richard Tizard for DNA sequence analysis; Carmen Young, Raymond Boyton, and Susan Foley for N-terminal sequence and mass spectrometry analyses; Greg Thill for mammalian cell line development; Cheryl Black for analyzing pharmacokinetic data; the pathology group for superb tissue staining, and Debbie Zhuang for excellent technical assistance during her internship.


   Disclosures
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
References
 
All authors are or were employees and stockholders of Biogen Idec, Incorporated.


   Footnotes
 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1 Address correspondence and reprint requests to Dr. Paul D. Rennert, Biogen Idec Incorporated, 12 Cambridge Center, Cambridge, MA 02142. E-mail address: paul.rennert{at}biogenidec.com Back

2 Abbreviations used in this paper: TAPR, T cell and airway phenotype regulator; TIM, T cell Ig domain and mucin domain; HAV, hepatitis A virus; IHC, immunohistochemistry; SPR, surface plasmon resonance; CHO, Chinese hamster ovary; KLH, keyhole limpet hemocyanin; BAL, bronchial lavage; LN, lymph node. Back

Received for publication August 17, 2006. Accepted for publication December 5, 2006.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 

  1. Woolcock, A. J., J. K. Peat. 1997. Evidence for the increase in asthma worldwide. Ciba Found. Symp. 206: 122-134. [Medline]
  2. Matricardi, P. M., F. Rosmini, L. Ferrigno, R. Nisini, M. Rapicetta, P. Chionne, T. Stroffolini, P. Pasquini, R. D’Amelio. 1997. Cross sectional retrospective study of prevalence of atopy among Italian military students with antibodies against hepatitis A virus. Br. J. Med. 314: 999-1003.
  3. Shirakawa, T., T. Enomoto, S. Shimazu, J. M. Hopkin. 1997. The inverse association between tuberculin responses and atopic disorder. Science 275: 77-79. [Abstract/Free Full Text]
  4. Umetsu, D. T., J. J. McIntire, O. Akbari, C. Macaubas, R. H. DeKruyff. 2002. Asthma: an epidemic of dysregulated immunity. Nat. Immunol. 3: 715-720. [Medline]
  5. Maddox, L., D. A. Schwartz. 2002. The pathophysiology of asthma. Annu. Rev. Med. 53: 477-498. [Medline]
  6. McIntire, J. J., S. E. Umetsu, O. Akbari, M. Potter, V. K. Kuchroo, G. S. Barsh, G. J. Freeman, D. T. Umetsu, R. H. DeKruyff. 2001. Identification of Tapr (an airway hyperreactivity regulatory locus) and the linked Tim gene family. Nat. Immunol. 2: 1109-1116. [Medline]
  7. McIntire, J. J., S. E. Umetsu, C. Macaubas, E. G. Hoyte, C. Cinnioglu, L. L. Cavalli-Sforza, G. S. Barsh, J. F. Hallmayer, P. A. Underhill, N. J. Risch, et al 2003. Immunology: hepatitis A virus link to atopic disease. Nature 425: 576[Medline]
  8. Gao, P. S., R. A. Mathias, B. Plunkett, A. Togias, K. C. Barnes, T. H. Beaty, S. K. Huang. 2005. Genetic variants of the T-cell immunoglobulin mucin 1 but not the T-cell immunoglobulin mucin 3 gene are associated with asthma in an African American population. J. Allergy Clin. Immunol. 115: 982-988. [Medline]
  9. Kaplan, G., A. Totsuka, P. Thompson, T. Akatsuka, Y. Moritsugu, S. M. Feinstone. 1996. Identification of a surface glycoprotein on African green monkey kidney cells as a receptor for hepatitis A virus. EMBO J. 15: 4282-4296. [Medline]
  10. Ichimura, T., J. V. Bonventre, V. Bailly, H. Wei, C. A. Hession, R. L. Cate, M. Sanicola. 1998. Kidney injury molecule-1 (KIM-1), a putative epithelial cell adhesion molecule containing a novel immunoglobulin domain, is up-regulated in renal cells after injury. J. Biol. Chem. 273: 4135-4142. [Abstract/Free Full Text]
  11. Bailly, V., Z. Zhang, W. Meier, R. Cate, M. Sanicola, J. V. Bonventre. 2002. Shedding of kidney injury molecule-1, a putative adhesion protein involved in renal regeneration. J. Biol. Chem. 277: 39739-39748. [Abstract/Free Full Text]
  12. Rennert, P. D., T. Ichimura, I. D. Sizing, V. Bailly, Z. Li, R. Rennard, P. McCoon, L. Pablo, S. Miklasz, L. Tarilonte, J. V. Bonventre. 2006. TIM-2 gene-deficient mice reveal a novel mechanism for the regulation of TH2 immune responses and airway inflammation. J. Immunol. 177: 4311-4321. [Abstract/Free Full Text]
  13. Dahl, M., K. Dabbagh, D. Liggitt, S. Kim, D. B. Lewis. 2004. Viral-induced T helper type 1 responses enhance allergic disease by effects on lung dendritic cells. Nat. Immunol. 5: 337-343. [Medline]
  14. Heaton, T., J. Rowe, S. Turner, R. C. Aalberse, N. de Klerk, D. Suriyaarachchi, M. Serralha, B. J. Holt, E. Hollams, S. Yerkovich, K. Holt, P. D. Sly, J. Goldblatt, P. Le Souef, P. G. Holt. 2005. An immunoepidemiological approach to asthma: identification of in-vitro T-cell response patterns associated with different wheezing phenotypes in children. Lancet 365: 142-149. [Medline]
  15. Alvarez, D., R. E. Wiley, M. Jordana. 2001. Cytokine therapeutics for asthma: an appraisal of current evidence and future prospects. Curr. Pharm. Des. 7: 1059-1081. [Medline]
  16. Fields, P. E., G. R. Lee, S. T. Kim, V. V. Bartsevich, R. A. Flavell. 2004. Th2-specific chromatin remodeling and enhancer activity in the Th2 cytokine locus control region. Immunity 21: 865-876. [Medline]
  17. Kips, J. C., G. P. Anderson, J. J. Fredberg, U. Herz, M. D. Inman, M. Jordana, D. M. Kemeny, J. Lotvall, R. A. Pauwels, C. G. Plopper, et al 2003. Murine models of asthma. Eur. Respir. J. 22: 374-382. [Abstract/Free Full Text]
  18. Maclean, J. A., A. Sauty, A. D. Luster, J. M. Drazen, G. T. De Sanctis. 1999. Antigen-induced airway hyperresponsiveness, pulmonary eosinophilia, and chemokine expression in B cell-deficient mice. Am. J. Respir. Cell Mol. Biol. 20: 379-387. [Abstract/Free Full Text]
  19. Umetsu, S. E., W.-L. Lee, J. J. McIntire, L. Downey, B. Sanjanwala, O. Akbari, G. J. Berry, H. Nagumo, G. J. Freeman, D. T. Umetsu, R. H. DeKruyff. 2005. TIM-1 induces T cell activation and inhibits the development of peripheral tolerance. Nat. Immunol. 6: 447-454. [Medline]
  20. Encinas, J. A., E. M. Janssen, D. B. Weiner, S. A. Calarota, D. Nieto, T. Moll, D. J. Carlo, R. B. Moss. 2005. Anti-T-cell Ig and mucin domain-containing protein 1 antibody decreases TH2 airway inflammation in a mouse model of asthma. J. Allergy Clin. Immunol. 116: 1343-1349. [Medline]
  21. Meyers, J. H., S. Chakravarti, D. Schlesinger, Z. Illes, H. Waldner, S. E. Umetsu, J. Kenny, X. X. Zheng, D. T. Umetsu, R. H. DeKruyff, et al 2005. TIM-4 is the ligand for TIM-1, and the TIM-1-TIM-4 interaction regulates T cell proliferation. Nat. Immunol. 6: 455-464. [Medline]
  22. de Souza, A. J., T. B. Oriss, K. J. O’Malley, A. Ray, L. P. Kane. 2005. T cell Ig and mucin 1 (TIM-1) is expressed on in vivo-activated T cells and provides a costimulatory signal for T cell activation. Proc. Natl. Acad. Sci. USA 102: 17113-17118. [Abstract/Free Full Text]
  23. Kumanogoh, A., S. Marukawa, K. Suzuki, N. Takegahara, C. Watanabe, E. Ch’ng, I. Ishida, H. Fujimura, S. Sakoda, K. Yoshida, H. Kikutani. 2002. Class IV semaphorin Sema4A enhances T-cell activation and interacts with Tim-2. Nature 419: 629-633. [Medline]
  24. Zhu, C., A. C. Anderson, A. Schubart, H. Xiong, J. Imitola, S. J. Khoury, X. X. Zheng, T. B. Strom, V. K. Kuchroo. 2005. The Tim-3 ligand galectin-9 negatively regulates T helper type 1 immunity. Nat. Immunol. 6: 1245-1252. [Medline]



This article has been cited by other articles:


Home page
 J. Immunol.Home page
A. J. de Souza, J. S. Oak, R. Jordanhazy, R. H. DeKruyff, D. A. Fruman, and L. P. Kane
T Cell Ig and Mucin Domain-1-Mediated T Cell Activation Requires Recruitment and Activation of Phosphoinositide 3-Kinase
J. Immunol., May 15, 2008; 180(10): 6518 - 6526.
[Abstract] [Full Text] [PDF]

Home page
 Int ImmunolHome page
M. Mizui, T. Shikina, H. Arase, K. Suzuki, T. Yasui, P. D. Rennert, A. Kumanogoh, and H. Kikutani
Bimodal regulation of T cell-mediated immune responses by TIM-4
Int. Immunol., May 1, 2008; 20(5): 695 - 708.
[Abstract] [Full Text] [PDF]

Home page
 BloodHome page
S. Nakae, M. Iikura, H. Suto, H. Akiba, D. T. Umetsu, R. H. DeKruyff, H. Saito, and S. J. Galli
TIM-1 and TIM-3 enhancement of Th2 cytokine production by mast cells
Blood, October 1, 2007; 110(7): 2565 - 2568.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sizing, I. D.
Right arrow Articles by Rennert, P. D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sizing, I. D.
Right arrow Articles by Rennert, P. D.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS