Therapeutic advantage of anti-VAP-1 over anti-α4 integrin antibody in concanavalin a-induced hepatitis

Authors

  • Woo-Yong Lee,

    1. Calvin, Phoebe & Joan Snyder Institute for Chronic Diseases, University of Calgary, Calgary, AB, Canada
    2. Department of Physiology and Pharmacology, Faculty of Medicine, University of Calgary, Calgary, AB, Canada
    Search for more papers by this author
  • Marko Salmi,

    1. MediCity Research Laboratory, University of Turku, Turku, Finland
    2. Department of Medical Biochemistry and Genetics, University of Turku, Turku, Finland
    3. National Institute of Health and Welfare, Turku, Finland
    Search for more papers by this author
  • Margaret M. Kelly,

    1. Calvin, Phoebe & Joan Snyder Institute for Chronic Diseases, University of Calgary, Calgary, AB, Canada
    2. Department of Physiology and Pharmacology, Faculty of Medicine, University of Calgary, Calgary, AB, Canada
    3. Department of Pathology and Laboratory Medicine, Faculty of Medicine, University of Calgary, Calgary, AB, Canada
    Search for more papers by this author
  • Sirpa Jalkanen,

    1. MediCity Research Laboratory, University of Turku, Turku, Finland
    2. National Institute of Health and Welfare, Turku, Finland
    3. Department of Medical Microbiology and Immunology, University of Turku, Turku, Finland
    Search for more papers by this author
  • Paul Kubes

    Corresponding author
    1. Department of Physiology and Pharmacology, Faculty of Medicine, University of Calgary, Calgary, AB, Canada
    • Calvin, Phoebe & Joan Snyder Institute for Chronic Diseases, University of Calgary, Calgary, AB, Canada
    Search for more papers by this author

  • Potential conflict of interest: Nothing to report.

  • Supported by the Canadian Association of Gastroenterology/Crohn's and Colitis Foundation of Canada Fellowship Award.

Address reprint requests to: Paul Kubes, Ph.D., Department of Physiology & Pharmacology, Faculty of Medicine, University of Calgary, HRIC 4AA16, 3330 Hospital Drive NW, Calgary, AB, T2N 4N1, Canada. E-mail: pkubes@ucalgary.ca; fax: 403-270-7516.

Abstract

Hepatitis induced by concanavalin A (Con A) in mice is well known to be a T-lymphocyte-mediated injury. It has been reported that T helper (Th)1 and Th2 lymphocytes use α4 integrin and vascular adhesion protein (VAP)−1, respectively, to adhere within the hepatic sinusoids. Therefore, we investigated whether inhibition of these molecules ameliorates or worsens the Con A-induced hepatic injury in vivo. Vehicle or antibody to α4 integrin or VAP-1 was intravenously administered 30 minutes before Con A administration. In control mice Con A markedly increased the serum alanine aminotransferase (ALT) level in a dose-dependent manner, and induced a massive infiltration of CD3, particularly interleukin (IL)−4 producing CD4 T cells and liver injury. Both parameters were reduced by anti-VAP-1 antibody despite antibody only blocking the adhesion, not the amine oxidase activity of VAP-1. Both activities of VAP-1 were eliminated in VAP-1-deficient mice and both Con A-induced liver injury and CD4 T-cell infiltration were eradicated. In contrast to anti-VAP-1, anti-α4 integrin antibody reduced interferon-gamma (IFN-γ)-producing CD3 T cells but this worsened Con A hepatitis, suggesting inhibition of a suppressor cell. Con A induced the recruitment of CD49d+ monocytic myeloid-derived suppressor cells (MDSCs) and regulatory T cells (Tregs) into the liver. Anti-α4 integrin dramatically blocked the influx of MDSCs but not Tregs. Conclusion: Our findings show that VAP-1 and α4 integrin have opposing effects in Con A-induced hepatic injury, which is associated with blocking the recruitment of CD4 lymphocytes and monocytic MDSCs, respectively. Moreover, these data provide the rationale for a potential therapeutic approach to target adhesion molecules in autoimmune hepatitis. (Hepatology 2013;58:1413–1423)

Abbreviations
ALT

alanine aminotransferase

Con A

concanavalin A

FOXP3

forkhead box P3

HA

hyaluronic acid

IBD

inflammatory bowel disease

MDSC

myeloid derived suppressor cell

MNC

mononuclear cell

MPO

myeloperoxidase

MS

multiple sclerosis

Siglec

sialic acid-binding Ig-like lectin

SSAO

semicarbazide sensitive amine oxidase

Treg

regulatory T cell

VAP-1

vascular adhesion protein-1.

Autoimmune hepatitis is a progressive chronic inflammatory disease of the liver characterized by a loss of self-tolerance.[1] Although immunosuppressants are widely used to inhibit autoimmune responses, serious side effects or resistance to a standard immunosuppressive therapy has been shown in many patients.[2, 3] Although T-cell-mediated immune responses play an important role in the development and progression of autoimmune, viral, and alcoholic hepatitis,[4-6] the underlying mechanisms are still unclear. Concanavalin A (Con A)-induced hepatic injury is a well-established murine experimental model of T-cell-mediated autoimmune or viral hepatitis that shares several pathological features with the disease in humans.[7, 8] The Con A model also serves to elucidate mechanisms of infiltration and activation of T cells in the liver that are critical for the development of human autoimmune and viral hepatitis.[4, 5] This injury is associated with both T helper (Th)1 and Th2 cell recruitment which release a large amount of interferon-gamma (IFN-γ) and interleukin (IL)-4, respectively. In a previous study, we clearly showed that Th1 and Th2 cells use α4 integrin and VAP-1, respectively, and adhere in the liver microvasculature in Con A-mediated liver inflammation.[9] However, how these adhesion molecules contribute to the pathogenesis of the hepatic injury was not fully elucidated.

Vascular adhesion protein-1 (VAP-1) is an endothelial cell molecule that is rapidly translocated from the intracellular storage vesicles to the endothelial cell surface upon inflammation.[10-12] It contributes to several steps in the extravasation cascade and mediates trafficking of lymphocytes, granulocytes, and monocytes to various sites of inflammation. In Con A-induced hepatitis, however, VAP-1 had significant specificity affecting only Th2 but not Th1 or granulocyte recruitment. VAP-1 has unique features distinct from other conventional adhesion molecules because, besides being an adhesin, it is also an enzyme. It catalyzes oxidative deamination of primary amines and produces hydrogen peroxide, aldehyde, and ammonium.[13] The endproducts of the enzymatic activity are highly potent inflammatory mediators and can up-regulate other adhesion molecules, such as E- and P-selectin, ICAM-1, VCAM-1, and MadCAM-1.[14-16] Here we used anti-VAP-1 antibody that can block just adhesion or a VAP-1 knockout system that can block both adhesion and enzymatic activity and demonstrate that both functions may contribute to Con A-induced liver injury. Our data also reveal that blocking Th1 cells with α4 integrin antibody results in worsening of disease, but because of the lack of cell-type specificity this might be due to blocking the recruitment of important regulatory cells, namely, myeloid derived suppressor cells (MDSCs). MDSCs are a heterogeneous population of cells that regulate liver inflammation[17-19] and are in various intermediate stages of myeloid cell differentiation.[20] Here we report that blocking α4 integrin causes the lack of monocytic MDSC recruitment in Con A-induced acute hepatitis and the subsequent exacerbation of injury, raising some concerns about blocking adhesion molecules with broad cellular inhibitory effects.

Materials and Methods

Mice

Con A was purchased from Sigma-Aldrich (Oakville, Ontario, Canada). Male BALB/c and C57BL/6 mice were purchased from the Jackson Laboratory. VAP-1 deficient mice on a 129S6 background have been described.[21] Vap-1−/− mice in C57BL/6 background were produced by crossing Vap-1−/− mice (in 129S6 background) with C57BL/6 wild-type animals and then backcrossing the animals for 10 generations.[21] Foxp3gfp mice were gifts from Alexander Y. Rudensky (University of Washington, Seattle, WA).[22] All mice were maintained in a specific pathogen-free, double-barrier unit at the University of Calgary (Calgary, AB, Canada). The protocols used were in accordance with the guidelines drafted by the University of Calgary Animal Care Committee and the Canadian Council on the Use of Laboratory Animals. Mice were used between 6 and 10 weeks of age.

Con A-Induced Hepatitis and Treatment of Blocking Antibodies Anti-α4 Integrin and Anti-VAP-1

Con A (0, 13 mg/kg, 15 mg/kg, or 20 mg/kg of mouse body weight) was intravenously administered to male BALB/c, C57BL/6, or Vap-1−/− mice for 8 hours or 24 hours before analysis. We chose 15 mg/kg of Con A for all subsequent experiments to ensure that the mice developed significant and reproducible liver injury, but were still well enough to subsequently endure anesthesia, surgery, and intravital microscopy. For untreated mice, 100 μL of sterile saline was injected. To investigate the role of α4 integrin and VAP-1 in the Con A induced-hepatitis, 100 μg of anti-α4 integrin (clone PS/2) or cocktail of 7-88 and 7-106 (50 μg each) were intravenously pretreated at 30 minutes prior to Con A administration.[9, 23] A semicarbazide sensitive amine oxidase (SSAO) inhibitor SZE5302 [(1S,2S)−2-(1-methylhydrazino)−1-indanol, also known as BTT-2052, a gift from Dr. Ferenc Fülöp from the University of Szeged, Szeged, Hungary][24] was administered by way of an intraperitoneal route at doses of 50 mg/kg. Vehicle (sterile physiological saline) injections served as negative controls.

Liver Enzyme

Blood was obtained by cardiac puncture at the time of sacrifice for analysis of serum alanine aminotransferase (ALT) as an index of hepatocellular injury. Measurements of serum ALT were made using a commercially available diagnostic kit (Biotron Diagnotics, Hemet, CA). The results are expressed as units per liter of serum.

Liver-Derived Lymphocyte Isolation and Flow Cytometric Analysis

Liver-derived lymphocytes were isolated from mice using a method previously described.[25, 26] Briefly, livers were excised and finely minced in a digestive medium containing 0.05% collagenase (Worthington Biomedical, Lakewood, NJ) and 0.002% DNase I in Hank's buffered salt solution (HBSS) (Invitrogen Canada, ON, Canada). After gentle agitation at 37°C for 30 minutes, the concentrate was passed through a 40-μM nylon filter and washed twice with ice-cold phosphate-buffered saline (PBS) (pH 7.4) and centrifuged at 300 g for 10 minutes. Lymphocytes were purified by a 37%/70% Percoll gradient. Lymphocytes were washed in cold PBS and counted in 0.4% trypan blue using a hemocytometer. Only mononuclear cells (MNCs) were gated in FSC versus SSC flow cytometric plots. The absolute number of MNCs or each subpopulation was standardized by the weight of liver tissue analyzed.

To investigate the alteration of lymphocyte population induced by Con A, flow cytometric analysis was used. Liver lymphocyte single-cell suspensions from each group of mice were stained with PE-conjugated anti-CD4 (H129.19, BD Pharmingen), FITC-conjugated anti-CD8a (53-6.7, BD Pharmingen), and PerCP-conjugated anti-CD3 (145-2C11, BD Pharmingen) to distinguish the recruited T-cell populations. For intracellular cytokine analysis, liver lymphocyte suspensions from each experimental group of animals were incubated with 10 μM brefeldin A in CO2 incubator for 3 hours at 37°C. The cell suspension was stained with PerCP-conjugated anti-CD3 (145-2C11, BD Pharmingen), washed with cold PBS, and resuspended in Cytofix/Cytoperm (BD Bioscience) for 30 minutes on ice. Samples were washed in Perm/Wash (BD Bioscience) and then stained with PE-conjugated anti-IFN-γ (XMG1.2, eBioscience) or PE-conjugated anti-IL-4 (11B11, BD Pharmingen).

To measure the number of recruited regulatory T cells (Tregs), FOXP3 was detected after cell permeabilization using an anti-FOXP3 antibody (FJK-16s; eBioscience) according to the manufacturer's protocol after staining with PerCP-conjugated anti-CD4 antibody (RM4-5, BD Pharmingen).[27] Monocytic MDSCs were defined as being mononuclear CD11b+Gr-1dim cells that coexpressed CD49d as described (Supporting Fig. 1).[28]

Spinning Disk Confocal Intravital Microscopy

Briefly, Foxp3gfp+ mice were anesthetized by intraperitoneal injection of a mixture of 10 mg/kg xylazine hydrochloride (MTC Pharmaceuticals, Cambridge, ON) and 200 mg/kg ketamine hydrochloride (Rogar/STB, London, ON). The right jugular vein was cannulated to administer additional anesthetic cocktail solution. Body temperature was maintained at 37°C using an infrared heat lamp. Mice were placed in a right lateral position on an adjustable microscope stage. A lateral abdominal incision along the costal margin to the midaxillary line was made to exteriorize the liver, and all exposed tissues were moistened with saline-soaked gauze to prevent dehydration. The liver was prepared for in vivo microscopic observation. Briefly, the liver was placed on the pedestal of a microscope and continuously superfused with warmed bicarbonate-buffered saline (pH 7.4). The liver surface was then covered with a coverslip to hold the organ in position. The liver microvasculature was visualized using a spinning disk confocal microscope and images were acquired with an Olympus BX51 upright microscope using a ×10/0.30 UplanFL N and ×20/0.45 LUCplanFL N objectives as described.[29-31] The microscope was equipped with a confocal light path (WaveFx, Quorum, Guelph, ON) based on a modified Yokogawa CSU-10 head (Yokogawa Electric, Tokyo, Japan). Foxp3gfp+ mice were used to visualize Foxp3gfp+ Tregs in the liver. A 488-nm excitation laser (Cobolt, Stockholm, Sweden) was used in rapid succession and images were visualized with the appropriate bandpass filter (Semrock, Rochester, NY). The typical exposure time for excitation wavelengths was 0.6-0.8 seconds. A 512 × 512 pixel back-thinned electron-multiplying charge-coupled device camera (C9100-13, Hamamatsu, Bridgewater, NJ) was used for green fluorescence detection. Volocity acquisition software (PerkinElmer, Waltham, MA) was used to drive the confocal microscope. Sensitivity settings were 200-220, and autocontrast was used. Images were captured at 16 bits/channel in RGB. Only the green channel using brightest point settings was exported in .jpg or .avi format. The behavior of green fluorescent protein (GFP)-expressing cells in the hepatic microvasculature was assessed.

Histology

For histological analysis the livers were excised at 8 hours or 24 hours after Con A administration, fixed in 10% formaldehyde, and prepared for microscopic assessment using standard methods (hematoxylin-eosin staining). Necroinflammatory features of autoimmune hepatitis were blindly evaluated by a pathologist (M.K.) on liver sections. An inflammatory activity was separated into none, minimal, mild, moderate, and severe, which corresponds to a numerical grade of 0 through 4, respectively.[32]

Statistical Analysis

All data are shown as mean ± standard error of the mean (SEM). Data were analyzed using standard statistical analysis (analysis of variance [ANOVA] with Bonferroni's correction for multiple comparisons where appropriate; GraphPad Software, San Diego, CA). Statistical significance was set at P < 0.05.

Results

Antivascular Adhesion Protein-1 Antibody Ameliorates, but Anti-α4 Integrin Antibody Worsens the Hepatic Injury Induced by Con A

To investigate hepatic injury after Con A administration, serum ALT levels were measured. As shown in Fig. 1A, ALT levels were significantly elevated in a dose-dependent manner at 8 hours after Con A administration. Mice are exquisitely sensitive to even a small increase in dose of Con A. At a dose of 13 mg/kg of mouse body weight only minor hepatic injury was induced, while a dose of 15 mg/kg markedly increased the serum ALT level. The dose of 20 mg/kg caused extreme damage, so we chose the intermediate dose of 15 mg/kg for further investigation.

Figure 1.

Effects of anti-α4 integrin, anti-VAP-1 antibodies, and SSAO inhibitor on serum ALT level at 8 hours and 24 hours after Con A treatment. (A) Serum ALT activity was measured at 8 hours after 13, 15, or 20 mg/kg of Con A. (n = 4-6 per group; *P < 0.05, **P < 0.01 versus vehicle-treated group by one-way ANOVA with Bonferroni's correction). (B) The effect of anti-α4 integrin and anti-VAP-1 on the liver injury at a Con A dose of 15 mg/kg (n = 5-7 per group; **P < 0.01 versus vehicle-treated group, ++P < 0.01 versus Con A-treated control by two-way ANOVA with Bonferroni's correction). (C) The effect of SSAO inhibitor on the serum ALT level at 8 hours after Con A administration (n = 4-6 per group; *P < 0.05, **P < 0.01 versus Con A-untreated group by one-way ANOVA with Bonferroni's correction).

The increase in serum ALT level at 8 hours after Con A (15 mg/kg) administration was attenuated by an anti-VAP-1 antibody that is known to inhibit adhesive functions but not to interfere with the enzymatic activity of VAP-1.[9] However, the attenuation was only apparent at 8 hours (70% attenuated) but not at 24 hours (Fig. 1B). Consistent with this observation, histological analysis revealed that Con A-induced portal and lobular inflammation was minimally attenuated in anti-VAP-1-treated livers at 8 hours (statistically not significant) (Fig. 2A-D, Table 1). At 24 hours of hepatic injury, blocking lymphocyte recruitment into the liver was not sufficient to affect the overall injury markedly (Fig. 2E-H, Table 2). Interestingly, some spatial difference occurred. The histological score for periportal inflammation was lower, as was the necrosis of the two rows of hepatocytes adjacent to the space of Disse (interface) while the lobular inflammation and necrosis were not reduced.

Figure 2.

Representative H&E staining at 8 hours (A-D) and 24 hours (E-H) after Con A (15 mg/kg) administration. (A,E) Sterile saline was treated as a vehicle to untreated group. (B,F) Con A was intravenously administered. Anti-VAP-1 (C,G) or anti-α4 integrin (D,H) antibody was pretreated at 30 minutes before Con A injection (original magnification, ×100).

Table 1. Histological Evaluation for 8-Hour Posttreatment of Con A (15 mg/kg)
GroupInflammationNecrosisSum of Score
PortalPeriportalLobularInterfaceLobular
  1. N= 4-6 per group;

  2. a

    P < 0.05,

  3. b

    P < 0.01 versus vehicle-treated group; ++P < 0.01 versus Con A-treated control group by one-way ANOVA with Bonferroni's correction.

Vehicle-treated0.0±0.00.0±0.00.0±0.00.0±0.00.0±0.00.0±0.0
Con A-treated      
No treatment1.3±0.3a0.0±0.01.3±0.2a0.0±0.01.8±0.2b4.5±0.3b
Anti-VAP-10.9±0.2a0.3±0.21.1±0.1b0.2±0.21.4±0.3b3.9±0.2b
Anti-α4 integrin1.3±0.2b0.8±0.2b,++1.5±0.2b0.0±0.01.9±0.3b5.6±0.4b
Table 2. Histological Evaluation for 24-Hour Posttreatment of Con A (15 mg/kg)
GroupInflammationNecrosisSum of Score
PortalPeriportalLobularInterfaceLobular
  1. N = 4-6 per group;

  2. a

    P < 0.05,

  3. b

    P < 0.01 versus vehicle-treated group; +P < 0.05, ++P < 0.01 versus Con A-treated control group by one-way ANOVA with Bonferroni's correction.

Vehicle-treated0.0±0.00.0±0.00.0±0.00.0±0.00.0±0.00.0±0.0
Con A-treated      
No treatment1.8±0.3b1.8±0.3b0.8±0.31.1±0.4b1.1±0.3a6.5±0.4b
Anti-VAP-11.8±0.2b0.8±0.2a,+1.6±0.2b0.0±0.0++2.0±0.0b6.2±0.6b
Anti-α4 integrin1.8±0.2b1.8±0.2b1.2±0.4a1.3±0.2b1.2±0.4a7.3±1.1b

VAP-1 functions as an SSAO in addition to an adhesin.[13, 15] The catalytic activity of VAP-1 has been invoked in the induction of various liver diseases.[14, 15] Therefore, we examined whether an enzymatic inhibitor for SSAO can attenuate the hepatic injury derived by Con A. Although there was no statistically significant difference in serum ALT level between the vehicle-treated and SSAO inhibitor-treated group (Fig. 1C), the values were 2,007 ± 391 for vehicle and Con A and 1,433 ± 332 for SSAO inhibitor and Con A.

Inhibiting α4 integrin, however, not only did not inhibit injury, there was a very significant almost 2-fold further increase in injury. Furthermore, anti-α4 antibody induced periportal inflammation that was not induced by Con A alone (Table 1). At 24 hours after Con A, the lobular inflammation was more severe with α4-integrin treatment (Fig. 2E-H, Table 2) and serum ALT levels were higher, particularly at 8 hours, to Con A alone (Fig. 1B). However, these blocking antibodies themselves did not cause any liver damage and systemic inflammation as shown in serum ALT and lung myeloperoxidase (MPO) level (Supporting Fig. 2).

Decreased Recruitment of CD4 Lymphocytes Was Associated With Beneficial Role of VAP-1 in Con A Hepatitis

The liver is known to have a large amount of resident mononuclear cells including T cells, natural killer (NK) cells, and NKT cells. A further significant infiltration of mononuclear cells including NK cells and CD3+ lymphocytes and a decrease in NKT cells were noted with Con A (Supporting Fig. 3). Although the mononuclear cell values for anti-α4 and anti-VAP-1 antibody-treated mice were lower, they did not reach significance perhaps because only some subpopulations were decreased while others stayed the same or even increased (Fig. 3A). Indeed anti-VAP-1 antibodies attenuated the increase in CD4+ cells (Fig. 3B), but did not influence the NK and NKT cells (Supporting Fig. 3). There was no change in the number of CD8 T cells (Fig. 3B). In addition, Con A administration increased the intracellular production of Th1 cytokine IFN-γ and Th2 cytokine IL-4 in the CD3 T cells (Fig. 3C,D). Anti-VAP-1 attenuated the increase in the intracellular IL-4 expression but not IFN-γ production (Fig. 3C,D). Although we previously reported that recruitment of exogenous IFN-γ producing Th1 cells was abolished by α4-integrin antibody, anti-α4 integrin dramatically increased endogenous IFN-γ production from the remaining CD3+ cells (Fig. 3D), suggesting perhaps that α4 integrin antibody affects a suppressor cell type that increases IFN-γ production.

Figure 3.

Effects of anti-α4 integrin and anti-VAP-1 antibodies on Con A-induced lymphocyte recruitment in the liver. Absolute number of mononuclear cells (MNCs, A), CD4 and CD8 T lymphocytes (B) were counted by flow cytometric analysis at 8 hours after Con A. Effects of anti-α4 integrin and anti-VAP-1 antibodies on the percentage of intracellular IL-4 (C) and IFN-γ (D) producing CD3 lymphocytes at 8 hours after Con A. Single cell suspension was incubated with 10 μM brefeldin A in CO2 incubator for 3 hours at 37°C (n = 3-6 per group; *P < 0.05, **P < 0.01 versus vehicle-treated group; +P < 0.05 versus Con A-treated control group by one-way ANOVA with Bonferroni's correction).

Monocytic Myeloid-Derived Suppressor Cells Not Tregs Are Inhibited by α4-Integrin in Con A-Induced Hepatic Injury

As the integrin α4 subunit is one of the surface molecules on Tregs, we hypothesized that blocking α4 integrin abolished the recruitment and beneficial roles of Tregs into liver in the Con A-induced hepatitis model. In both intravital microscopy (Fig. 4A,B; Supporting Video 1-4) and flow cytometry (Fig. 4C) using Foxp3gfp mice, anti-α4 integrin did not affect the increased recruitment of Tregs derived by Con A hepatitis, making Treg inhibition by α4-integrin an unlikely regulator of the hepatic injury derived by Con A.

Figure 4.

Effect of anti-α4 integrin on Treg and monocytic MDSC recruitment increased by Con A treatment. Con A (15 mg/kg) was administered at 8 hours prior to flow cytometry or intravital microscopy. The vasculatures of Foxp3gfp+ mouse livers were visualized using spinning disk confocal intravital microscopy at 4 hours and 8 hours (A,B) after vehicle or Con A treatment. CD4+FOXP3+ Treg (C) and CD49d+CD11b+Gr-1dim MDSC (D) recruitment into the liver in response to Con A in mice treated with vehicle (UT) or anti-α4 integrin were analyzed by flow cytometry. (B-D) (n = 3-6 per group; *P < 0.05, **P < 0.01 versus vehicle-treated group; +P < 0.05 versus Con A-treated control group by one-way ANOVA with Bonferroni's correction).

Recently, Haile et al.[28] have shown that CD49d (α4 integrin) is a specific marker for MDSCs and CD49d-expressing MDSCs are mainly monocytic and more potent suppressors than CD49d-negative MDSCs, which are of neutrophil origin. Therefore, we examined whether Con A can recruit the MDSCs into the liver and anti-α4 integrin can block the increase of MDSC recruitment derived by Con A administration. Interestingly, Con A increased 3-fold the CD49+ monocytic MDSC recruitment, and this increase was abolished in the anti-α4 integrin pretreated mice, making MDSCs a possible regulator of the hepatic injury derived by Con A and α4 integrin (Fig. 4D). It is worth noting that α4 integrin antibody reduced the number of MDSCs under basal conditions but this did not cause inflammation.

Con A-Induced Hepatic Injury Was Attenuated in VAP-1-Deficient Mice

To block both the adhesin and oxidant capacity of VAP-1, Con A (15 mg/kg) was intravenously administered to C57BL6 wild-type or Vap-1−/− mice. As shown in Fig. 5A, at 8 hours after Con A administration the serum ALT level was significantly increased but this increase was markedly lower (by 80%) in Vap-1−/− mice. CD4 cells were increased in the Con A-treated C57BL6 wild-type livers (Fig. 5B). The increase was significantly attenuated in the Vap-1−/− mice. In line with the result of anti-VAP-1 pretreatment, intracellular CD3 IL-4 production in VAP-1 deficient liver was decreased compared to wild-type CD3 IL-4 production (Fig. 5C). No change in IFN-γ production was noted (Fig. 5C). Interestingly, VAP-1-deficient mouse liver had twice as many CD4+FOXP3+ Tregs basally as wild-type mouse liver and slightly more Tregs were recruited into the Con A-treated VAP-1-deficient liver than into the Con A-treated wild-type liver (Fig. 5D), suggesting that VAP-1 may be a negative regulator of Tregs.

Figure 5.

Decreased hepatic injury and lymphocyte recruitment in Con A-treated Vap-1−/− mouse liver. Serum ALT level (A) and hepatic CD4 T cell number (B) in C57BL6 wild-type (WT) and Vap-1−/− mice at 8 hours after Con A (15 mg/kg) administration (n = 3-4 per group; **P < 0.01 versus vehicle-treated group; +P < 0.05, ++P < 0.01 versus Con A-treated WT group by one-way ANOVA with Bonferroni's correction). (C) IL-4 and IFN-γ producing CD3 T cells isolated at 8 hours after Con A treatment in WT and Vap-1−/− mice. Histograms are representative of 2-3 independent experiments. (D) CD4+FOXP3+ Treg recruitment into the liver at 8 hours after vehicle (UT, left panel) or Con A treatment (right panel) in WT (upper panel) or Vap-1−/− (lower panel) mice. Flow cytometry dot plots are representative of 2-3 independent experiments.

Discussion

There is increasing evidence that recruitment of certain leukocytes are closely related with autoimmune disease. Integrin-mediated leukocyte recruitment and/or signaling was shown to contribute to psoriasis, multiple sclerosis (MS), and inflammatory bowel disease (IBD) in mouse and human. However, the liver was different from skin, gut, and brain in that integrins were often not involved and other adhesion molecules were implicated. Indeed, anti-adhesion was demonstrated for VAP-1 in hepatic sinusoids in a model of hepatitis.[9] Adhesion molecules such as CD44/hyaluronic acid and Siglec-10/VAP-1 have also been implicated in interactions between leukocytes and endothelium in liver.[33-36] McDonald et al.[33] have shown that CD44 was responsible for reversible neutrophil adhesion to hyaluronic acid (HA) within liver sinusoids in vivo and the disruption of CD44-HA interactions reduced liver injury in response to bacterial lipopolysaccharides. The CD44-dependent adhesion was reversible and not replaced by other adhesive mechanisms or physical trapping, suggesting a key role for molecular adhesion in liver sinusoids.[33] Moreover, Siglec-10, a member of the family of sialic acid-binding Ig-like lectins (Siglecs), was first identified as a leukocyte surface ligand and a substrate for VAP-1, which is important in lymphocyte recruitment in liver.[35, 36] These findings are crucial for the development of new antiinflammatory therapy rather than traditional anti-integrin therapy.

Herein, we provide further evidence that VAP-1 is a good target to ameliorate the autoimmune-induced hepatic injury by selectively regulating recruitment of CD4 Th2 but not Th1 lymphocytes. We analyzed the effect of anti-VAP-1 therapy for lymphocyte recruitment to the liver using a well-established Con A-induced hepatitis model.[4, 5, 7, 8] Con A treatment caused acute liver injury as assessed by increased serum transaminase level as early as 4 hours (data not shown) after intravenous administration and continued until 24 hours. Con A treatment is not specific for liver and causes injury elsewhere, including lung, where many neutrophils are recruited but anti-VAP-1 was ineffective (Supporting Fig. 4), indicating that VAP-1 does not affect neutrophil recruitment in lung or as previously reported in liver.[9]

VAP-1 functions as both an adhesion molecule and a generator of oxidants. Our data demonstrate that blocking a specific adhesion molecule holds some therapeutic promise, while inhibitors of the mono-oxidase activity showed marginal (25%) benefit. However, inhibiting both functions of VAP-1 reduced Con A-induced inflammation to the greatest extent, suggesting that targeting both functions of VAP-1 could have optimal benefit. The attenuated liver injury in anti-VAP-1-treated mice and VAP-1-deficient mice correlated with reduced CD4+ T cell recruitment and reduced numbers of IL-4 producing T cells. This is in line with the essential role of CD4+ T cells in induction of liver injury derived by Con A but our data point toward a Th2 subset.

α4 integrin has been targeted successfully in both MS and IBD with the caveat that there is likely the inhibition of not just, the disease-causing T cells but also inhibition of cells critical for preventing viral disease. Herein, we show that α4 integrin did not work in Con A-induced hepatitis but rather exacerbated symptoms perhaps by blocking MDSCs. MDSCs are a heterogeneous family of cells that are in various stages of myeloid cell differentiation.[37] In mice, MDSCs are phenotypically characterized as CD11b+GR1+ coexpressing cells that represent a mononuclear CD49d+ MDSC subpopulation which strongly suppresses T-cell proliferation and activation through inducible NOS in tumor-bearing mice[28] but also in T-cell-mediated diseases. Importantly, our results show that Con A causes the recruitment of monocytic MDSCs into the liver, which can be inhibited by anti-α4 integrin leading to increased IFN-γ production in CD3 T cells. Although we previously blocked exogenous Th1 trafficking to the liver with anti-α4 integrin, the net effect of inhibiting α4 integrin recruitment of endogenous cells including MDSCs caused more injury and increased recruitment of some cell types. Moreover MDSCs may stimulate Treg function and expansion[38] and Tregs are potent inhibitors of monocytic cells including effector T cells.[39, 40] Although blocking α4 integrin did not modify the number of Tregs in Con A-induced injured liver, we cannot exclude the possibility that a reduced number of MDSCs might cause a failure in stimulating Tregs and subsequent inhibitory cytokine production. Interestingly a 2-fold higher proportion of Tregs physiologically resides in the liver of VAP-1-deficient mice than of wild-type mice, which is augmented further in acute inflammation derived by Con A, suggesting that increased Tregs could also be contributing to benefits in VAP-1-deficient mice. Although SSAO inhibition alone did little to reduce inflammation blocking SSAO and adhesion of VAP-1 was optimal. It has been reported that an SSAO inhibitor of VAP-1 reduced the recruitment of Gr-1+CD11b+ myeloid cells into tumor vasculature and attenuated the growth of tumor indicating that the SSAO activity of VAP-1 may be responsible for the recruitment of at least some leukocytes into the tumor.[41]

In conclusion, VAP-1 plays a critical role in recruitment of CD4 Th2 cells and inhibition of or lack of VAP-1 causes a decline in IL-4-producing T cells and subsequent improvement of the disease state. In various disease states ranging from sepsis to viral and autoimmune hepatitis, a very significant number of lymphocytes are recruited into the liver sinusoids. The strategy of inhibiting an inappropriate accumulation of inflammatory cells in liver microvasculature could improve the pathological state of a number of inflammatory diseases. Our data suggest that targeting VAP-1 has promise for the development of a potential antiinflammatory therapy. Anti-α4 integrin exacerbates the injury derived by Con A, which may be due to the inhibition of monocytic MDSCs, or indirect effects upon Tregs. The end result is that the blockade of antiinflammatory cells outweighs inhibition of inflammatory cells leading to injury making VAP-1 a better target than α4-integrin at least in this model of autoimmune hepatitis.

Acknowledgment

We thank Dr. M.N. Ajuebor (LSU Health Science Center) for the liver-derived lymphocyte isolation technique, the University of Calgary Flow Cytometry Facility and L. Kennedy for their assistance with the flow cytometric analysis. We also thank C. Badick for excellent technical support and the Live Cell Imaging Facility funded by the Canada Foundation for Innovation and Dr. P. Colarusso for training and assistance related to microscopy. This work was supported by the Canadian Association of Gastroenterology/Crohn's and Colitis Foundation of Canada Fellowship Award (to W-Y. L.), the Canadian Institutes of Health Research (to W-Y. L. and P.K.), the Canada Research Chairs Program and the Alberta Heritage Foundation for Medical Research (to P.K.).

Ancillary