| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Burnham Institute for Medical Research, La Jolla, California, USA
1Correspondence: Burnham Institute for Medical Research, 10901 North Torrey Pines Rd., La Jolla, CA 92037, USA. E-mail: strongin{at}burnham.org
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: IDDM • monoclonal antibody • CD44 • MT1-MMP
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. The pathogenesis of IDDM involves the activation of autoimmune T cells, followed by their homing into the pancreatic islets. In the islets, T cells directly destroy insulin-producing ß cells (2)
. The expression of cell surface adhesion receptor CD44 is elevated in activated T cells. CD44, via its interactions with endothelial hyaluronan, mediates T cell adhesion to the endothelium and the subsequent transmigration events (3)
. There is considerable evidence that CD44 is a target of MT1-MMP proteolysis in tumor cells. MT1-MMP cleavage releases the extracellular domain of CD44 from cell surfaces and inactivates the CD44 cell receptor function (4
5
6
7)
. Invasion-promoting MT1-MMP, a multifunctional membrane-tethered enzyme, functions in cancer cells as one of the main mediators of pericellular proteolytic events (8)
and directly cleaves cell surface receptors (9
10
11
12
13
14)
.
We recently discovered that clinically tested inhibitors of matrix metalloproteinases (MMPIs) can efficiently control the functional activity of T cell MT1-MMP and significantly delay the onset of adoptively transferred diabetes in NOD mice (15)
. We have determined that both the induction of the intrinsic T cell MT1-MMP activity and the shedding of cellular CD44 followed the adhesion of highly diabetogenic, insulin-specific, CD8-positive, Kd-restricted T cells of the TGNFC8 clone (IS-CD8+ T cells) (16
, 17)
to the matrix. Conversely, inhibition of these events by AG3340/Prinomastat [3(S)-2,2-dimethyl-4[4-pyridin-4-yloxy)-benzensulfonyl]-thimorpholine-3-carboxylic acid hydroxamate], a potent, subnanomolar range hydroxamate MMPI that has been widely used in clinical trials in cancer patients (18
19
20)
, impeded the transmigration of diabetogenic T cells into the pancreas and delayed the onset of adoptively transferred diabetes in NOD mice (15)
. It is known that the Ki values of AG3340 against MMP-2, MMP-3, MMP-13, and MT1-MMP are 
100 pM, 300 pM, 40 pM, and 200 pM, respectively, whereas other individual MMPs are significantly less sensitive to AG3340 inhibition (e.g., the Ki for MMP-1 and MMP-7 are 10 nM and 55 nM, respectively).
To extend and validate these observations and to determine whether there is a mechanistic rationale for clinical trials of MMPIs in IDDM in humans, we have confirmed a causal link between MT1-MMP and CD44 shedding in adherent T cells. We have also determined that low dosages of AG3340 exerted protective effects in diabetic NOD mice that, similar to humans, developed the disease spontaneously. Our results suggest that the pharmacological inhibition of T cell MT1-MMP by the readily available hydroxamate MMPIs has the potential for therapeutic translation and for clinical trials in humans, and that these trials will result in a favorable outcome for the IDDM patients.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
were maintained in Click’s medium supplemented with 5% FCS, 2 x 10–5 M ß-mercaptoethanol, 20 mM penicillin-streptomycin, 3 mg/ml L-glutamine, and 5 U/ml recombinant murine interleukin (IL)-2 (21)
. Every 3 wk IS-CD8+ cells were mixed with irradiated NOD splenocytes (2000 Rad) loaded with the L15YLVCGERG23 insulin B chain peptide (10 µg/ml) (16)
. The animal treatment protocols were reviewed and approved by the Burnham Institute review committee.
Newly diabetic NOD mice
In our experiments we used 4- to 5-month-old NOD female mice that spontaneously developed diabetes. The onset of spontaneous diabetes was identified by assessing urine glucose (Glc) levels with Diastix strips (Fisher, Pittsburgh, PA, USA). Mice with urine Glc levels > 1000 mg/dl for three consecutive days were considered diabetic. According to numerous publications (e.g., ref. 22
). Glc levels in the urine closely follow those in the blood. The continuing measurement of Glc in urine is a widely accepted method used to follow the development of diabetes in NOD mice (23)
. After development of diabetes, insulin (15–20 U/kg; one injection in every 2 or 3 days) was injected s.c. in mice. Control animals (6 mice/group) received insulin alone, while an experimental group (5 mice/group) received insulin s.c. jointly with AG3340 i.p. (1 mg/kg; one injection in every 2 or 3 days). AG3340 (molecular mass=421 Da) was a kind gift from Dr. Peter Baciu (Allergan, Irvine, CA, USA). Injections were continued for 40 days, then mice were sacrificed. Leukocytes and granulated ß cells were stained with H&E and aldehyde fuchsin, respectively, in the sections of paraformaldehyde-fixed, paraffin-embedded, pancreata. Islets (
100/mouse) were scored as follows: 0, no lesions; 1, peri-insular leukocytic aggregates, periductal infiltrates; 2, <25% islet destruction; 3, >25% islet destruction; and 4, totally destroyed islets. Assessment and scoring of the islets were done by a "blinded" observer. To identify hormone-producing cells, the sections were stained with the antibodies to insulin (Linco Research, St. Charles, MO, USA) and glucagon (DacoCytomation, Carpinteria, CA, USA), followed by a species-specific secondary horseradish peroxidase-conjugated antibody (Ab) and a 33,3'-diaminobenzidine substrate.
Fluorescent tracing and morphometric analysis
For trafficking studies, IS-CD8+ T cells were incubated at 1 x 107 cells/ml for 30 min at 37°C in the dark in the complete Click’s medium containing 5% FCS and 0.0075 mg/ml of fluorescent dye 1,1'-didodecyl-3,3,3',3'-tetramethylindocarbocyanine perchlorate (DiI; Molecular Probes, Eugene, OR, USA). After incubation, cells were washed three times with PBS to remove excess DiI. Labeled IS-CD8+ cells (1x107 cells) were i.v. injected in 0.2 ml of PBS into irradiated (725 Rad, 24 h in advance) NOD mice. Mice were sacrificed 24 h after injection of DiI-labeled cells. Where indicated, the function-blocking murine Ab IM7.8.1 (BD Biosciences, San Jose, CA, USA) against CD44 and an Ab isotype control (0.1 mg Ab/animal), and AG3340 (1 mg/kg) were each injected i.v in NOD mice 30 min before the i.v. injection of DiI-labeled IS-CD8+ T cells. In 24 h the spleen and pancreata were excised and fixed in 0.1 M periodate-lysine-paraformaldehyde phosphate buffer. The organs were next sucrose-saturated, freeze-molded in OCT compound (Sakura Finetek USA, Torrance, CA, USA), and freeze-sectioned. Cryostat sections 7 µm-thick of the spleen and entire pancreas were prepared at 60 µm intervals using a Leica CM1900 cryotom. The distribution of DiI-labeled CD8+ cells within the spleen and islets was examined using a fluorescent microscope. At least 100 individual islets per mouse, 4–5 mice per each experimental group, were examined. The characteristic morphology of the islets is easily identified on the cryosections. DiI-labeled cells were counted by a blinded observer. The position of each IS-CD8+ labeled cell was determined relative to the islet boundary. The labeled cells localized within the islet boundary were considered to be inside, whereas the labeled cells adjacent to the islet but outside of the islet boundary were considered to be "at the entrance." This method was described in detail in our earlier work (21
, 24)
.
To visualize pancreatic microvessels, cryosections of the pancreas were fixed for 5 min in acetone at –20°C, then stained with an R-phycoerythrin-labeled Ab (red; Pharmingen, San Diego, CA, USA) to CD105 (endoglin; an endothelial marker). We also stained the section with an FITC-labeled Ab CA50 (green; clone CAM17.1, RDI, Concord, MA, USA) to pancreatic mucin (an exocrine pancreatic marker).
To identify the positions of the DiI-labeled T cells relative to the pancreatic microvasculature, we used lectin from Lycopersicon esculentum (tomato) (Sigma, St. Louis, MO, USA). DiI-labeled IS-CD8+ T cells (1x106/mouse) were injected i.v. into NOD mice. In 4 h mice were sacrificed. The pancreas was perfused through the lower aorta with 2–3 ml of an FITC-labeled lectin solution (a 1:200 dilution in PBS supplemented with 2 U of heparin). In 15 min the pancreata were excised, flattened firmly between two glass slides, and examined under a fluorescence microscope.
Western blot
IS-CD8+ cells were surface biotinylated for 1 h at 4°C with 0.1 mg/ml of sulfo-NHS-LC biotin (Pierce, Rockford, IL, USA). After washes to remove biotin, the labeled cells were allowed to adhere in serum-free medium to plastic coated with 2% type I collagen/gelatin for 4 h or were kept in suspension. Under these experimental conditions, the vast majority of the cells became attached to gelatin. Where indicated, tissue inhibitor-1 of matrix metalloproteinases (TIMP-1), tissue inhibitor-2 of matrix metalloproteinases (TIMP-2) (100 ng/ml each), and AG3340 (50 µM or 21 µg/ml) were added to the cells. The cells were then lysed with 50 mM N-octyl-ß-D-glucopyranoside in PBS supplemented with 1 mM CaCl2, 1 mM MgCl2, and a protease inhibitor cocktail containing phenylmethylsulfonyl fluoride (1 mM) and aprotinin, pepstatin, and leupeptin (1 µg/ml each). Biotin-labeled CD44 was captured from the cell lysate and from the medium aliquots on streptavidine-agarose beads. The captured samples were examined by Western blot with the CD44 (clone IM7.8.1) Ab to determine the released, soluble CD44 ectodomain in the medium samples and the residual, membrane-anchored cellular CD44 in the cell lysates.
MT1-MMP-dependent MMP-2 activation and gelatin zymography
IS-CD8+ cells (1x106) were either allowed to adhere for 4 h in serum-free, unsupplemented Click’s medium to plastic coated with 2% gelatin or kept in solution. In 18 h, media samples (30 µl each) were withdrawn and analyzed by gelatin zymography (25)
to identify the proteolytic activity and the activation status of MMP-2 naturally synthesized by IS-CD8+ cells. Where indicated, the cells were supplemented with external purified pro-MMP-2 (20 ng), TIMP-1 and TIMP-2 (100 ng/ml each), and AG3340 (50 µM or 21 µg/ml). Pro-MMP-2 was isolated from a conditioned medium of p2AHT2A72 cells derived from an HT1080 fibrosarcoma cell line sequentially transfected with the E1A and the MMP-2 cDNAs (26)
.
Chromium-51 release cytotoxicity assay
Islets of Langerhans were isolated by the collagenase method from the pancreas of 5- to 7-wk-old NOD mice. The islets were also purified by discontinuous gradient centrifugation on Histopaque 1119 (Sigma) (27)
. The purified islets were dissociated by using Cell Dissociation Buffer (Sigma) to obtain a single-cell suspension. The single-cell suspension was transferred in 0.2 ml of complete Click’s medium containing 5% FCS and labeled for 2 h at 37°C with 51Cr (100 µCi; MP Biomedicals, Irvine, CA, USA). After labeling, the cells were washed three times in PBS to remove any unbound label. The labeled cells (1x104 cells in 0.2 ml of Click’s medium with 5% FCS /well of 96-well plate) were cocultured for 4 h with increasing concentrations of the splenic cells isolated from NOD mice and the effector IS-CD8+ cells to reach the effector-to-target ratio from 12.5:1 to 100:1. Cells were then collected by centrifugation and the amounts of 51Cr released by the labeled islets cells were counted in the supernatant fraction using a gamma counter. Where indicated, IS-CD8+ cells were coincubated for 24 h at 37°C in complete Click’s medium containing 5% FCS supplemented with either the CD44 Ab (10 µg/ml) and AG3340 (50 µM) or left untreated, then used in the cytotoxicity assays. The Ab and AG3340 were then included in the assay medium where indicated. The spontaneous release of the label was measured in the wells that contained the target cells alone. In addition, Nonidet P-40 was added to a final concentration of 1% and the cell lysates were assayed for 51Cr activity to determine the total amount of the cell-incorporated label. Cytotoxicity of the effector cells was calculated using the following formula: specific cytotoxicity (%) = [the label released in the presence of effector cells – spontaneously released label]/[total cell label – spontaneously released label].
We also used the splenocytes as effector cells in the cytotoxicy assays. For these purposes, the first group of freshly diabetic, 5-month-old NOD mice received insulin alone for 2 wk. The second group of animals received insulin and AG3340 (1 mg/kg). The spleen was isolated from both groups of mice. The total population of the splenic cells from the second group was cultivated for 2 wk in the medium with AG3340 (50 µM). Splenic cells isolated from the first group of mice were cultivated in a medium without AG3340. To enrich the population of the ß cell-specific T cells, the splenocyte cultures were stimulated once a week by coincubation with the NOD pancreatic islets (21)
. The splenocytes were then used to target the islet cells in the 51Cr-release cytotoxicity assays.
Monoclonal antibodies and flow cytometry
DiI-labeled IS-CD8+ cells, the total cell population isolated from the spleen and the pancreatic lymph nodes (both obtained from NOD mice) were stained with the CD44 (clone IM7.8.1), the CD3 (clone 17A2), the CD4 (clone L3T4), and the CD8 (clone 53–6.7) monoclonal antibodies (all from BD Biosciences), followed by staining with fluorescein isothiocyanate- or phycoerythrin-conjugated secondary Ab (BD Biosciences), then analyzed on a FACScan flow cytometer (Becton Dickinson, Mountain View, CA, USA).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. To determine whether MT1-MMP represents a drug target in spontaneous IDDM and to confirm and extend our observations, we set out to determine whether CD44 is the major mediator involved in the adherence of diabetogenic IS-CD8+cells. To quantitatively assess the role of CD44 in the homing of T cells, NOD mice were irradiated at 725 Rad. In 24 h, a function-blocking Ab against CD44 and AG3340 was injected in mice. After 30 min, this injection was followed by i.v. injection of IS-CD8+ T cells labeled with a fluorescent dye DiI. Mice were sacrificed 24 h after injection of the cells. The pancreata were excised and freeze-sectioned. The distribution of DiI-labeled IS-CD8+ cells within the islets was examined using a fluorescent microscope. DiI-labeled cells were counted within the area relevant to each individual islet. Figure 1
A shows that IS-CD8+ cells efficiently homed inside the islets of the control mice, while the blocking of the CD44 functionality reduced the efficiency of T cell accumulation into the pancreatic islets.
|
Our data are consistent with earlier observations that the injection of anti-CD44 antibodies before cell transfer of diabetogenic splenocytes induced considerable resistance to diabetes in recipient mice (28)
. We thus confirmed that CD44-mediated adhesion plays a highly significant role in the homing of T cells to the islets. AG3340 (by inhibiting the MT1-MMP-mediated proteolysis of CD44 in the adherent T cells) also caused a 50% reduction in T cell homing. Representative images show the main difference between anti-CD44 and AG3340: the first repressed the homing of DiI-labeled cells inside the islets, whereas the second incapacitated the adherent T cells on the pancreatic endothelium boundary at the islet’s entrance.
In our additional experiments, the pancreatic sections were stained with the FITC-conjugated mucin Ab (an exocrine pancreatic marker) and with PE-conjugated CD105 Ab (an endothelial marker) to visualize the islet endothelium. A representative islet, the boundaries of which are marked by the microvasculature (stained in red) and surrounded by the pancreatic acinar cells (stained in green), is shown in Fig. 1B
.
To identify the relative position of the injected IS-CD8+cells compared with the pancreatic vasculature and to confirm adhesion of the diabetogenic T cells to the pancreatic endothelial cell layers, we injected i.v. DiI-labeled IS-CD8+cells in NOD mice. A 4 h period was sufficient to establish the adhesion of the injected T cells (21)
. Mice were sacrificed, then perfused through the lower aorta with an FITC-labeled lectin solution. This lectin from Lycopersicon esculentum (tomato) binds endothelial cells and visualizes the microvasculature (29)
. A representative micrograph of the pancreas showing adherent, yellow, DiI-labeled IS-CD8+cells associated with the pancreatic microvasculature (green) appears in Fig. 1B
.
The CD44 Ab and AG3340 did not affect the functionality of IS-CD8+ cells
To exclude the adverse effect of the CD44 Ab and AG3340 on the functionality and viability of IS-CD8+ cells, we determined the number of DiI-labeled IS-CD8+ cells in both the spleen and the pancreatic lymph nodes of NOD mice. To identify the labeled cells in the spleen, we used flow cytometry and microscopic analyses of spleen cryosections (Fig. 2
). The microscopic analyses determined that neither the CD44 Ab nor AG3340 affected the number of DiI-labeled IS-CD8+ cells trafficked to the spleen compared with mice that received PBS alone. The quantitative flow cytometry analysis confirmed these results. Thus, the DiI-labeled IS-CD8+ cells represented about one-third of the total population of CD8+ splenocytes in the control and experimental mice. Statistical analyses confirmed the significance of the FACS results (Fig. 2)
. Our results suggested that neither AG3340 nor the CD44 Ab affected the survival and the proliferation of the injected labeled T cells. Because the homing of T cells to the spleen does not involve lymphocyte-endothelial adhesive interactions and because our previous observations determined that IS-CD8+ cells do not significantly proliferate, at least for 96 h postinjection (21)
, we concluded that both AG3340 and the CD44 Ab targeted only the adhesion to the endothelium and the ensuing transmigration/homing of the DiI-labeled cells. We also examined lungs of the mice: no difference in the number or distribution of DiI-labeled IS-CD8+ cells was observed between mice that received the Ab control, anti-CD44, or AG3340 (not shown).
|
To corroborate our conclusion, we also determined, using flow cytometry, the number of DiI-labeled IS-CD8+ cells that were capable of homing into the pancreatic lymph nodes of the experimental and control mice. The percentage of labeled cells in the lymph node cell population was 3.3 ± 0.3% in control mice. The CD44 Ab reduced the percentage of the labeled cells 3-fold (1.1±0.2%; P=0.007). The percentage of the labeled cells in AG3340-treated mice was 2.4 ± 0.4% (P=0.04). These parameters are similar to the homing pattern of IS-CD8+ observed in the islets, and they support our hypothesis that the CD44 Ab predominantly targets the adhesion of T cells to the endothelium whereas AG3340 largely affects the postadhesive transmigration of T cells.
We also determined that neither the CD44 Ab nor AG3340 exhibited any unexpected adverse effects on the cytotoxicity of IS-CD8+ cells (Fig. 3
). For these purposes, pancreatic islets were isolated from NOD mice, then labeled with 51Cr. The labeled islet target cells were coincubated with the IS-CD8+ effector cells. After coincubation, the levels of 51Cr released by the target cells were measured in the extracellular milieu. Figure 3
shows that both the CD44 Ab and AG3340 did not significantly affect the cytotoxic efficiency of IS-CD8+ cells.
|
To assess whether AG3340 has any effect on the ß cell-specific T cell reactivity in mice, we determined the cytotoxic efficiency of the total splenocyte population isolated from NOD mice. A 2 wk incubation in the presence of AG3340 had no effect on the cytotoxicity of the splenocytes or their ability to target the pancreatic islets (Fig. 3)
. Overall, we concluded that neither the CD44 Ab nor AG3340 affected the functionality of IS-CD8+, including homeostasis, proliferation, and survival.
A causal link between MT1-MMP and CD44 shedding
We next determined the significance of MT1-MMP activity in the shedding of CD44 and identified a causal link between the two in the adherent IS-CD8+ cells. For these purposes, IS-CD8+ cells were surface biotinylated with membrane-impermeable biotin, and the labeled cells were then either allowed to adhere to a gelatin-coated plastic or were kept in solution. The cells were then lysed and biotin-labeled CD44 was captured from the cell lysate and from the medium aliquots on streptavidine-agarose beads. The captured samples were examined by Western blot to measure both the quantities of the released soluble CD44 ectodomain in the medium samples and the residual, membrane-anchored cellular CD44 in the cell lysates. In addition, media samples were analyzed by gelatin zymography to identify the activation status of MMP-2 that is naturally synthesized by IS-CD8+ cells. MMP-2 is an enzyme known to be directly activated by MT1-MMP (9
, 26)
. Where indicated, cells were supplemented with external purified pro-MMP-2. TIMP-2 (a potent inhibitor of MT1-MMP), TIMP-1 (a poor inhibitor of MT1-MMP), and AG3340 were each added to the cell samples to distinguish the role of MT1-MMP from the putative effect imposed by the other individual cell surface-associated proteases (30)
(Fig. 4
).
|
Consistent with our previous observations (15)
, endogenous MT1-MMP was latent in nonadherent cells whereas the adhesion of T cells induces the activation of MT1-MMP, subsequent activation of MMP-2, and cleavage of CD44. Thus, only after their adhesion to gelatin, were IS-CD8+ cells capable of activating MMP-2. Nonadherent cells did not activate MMP-2. In agreement, release of the CD44 fragments into the medium was detected only in adherent IS-CD8+ cells. CD44 remained intact in nonadherent cells. AG3340 and TIMP-2 each fully blocked both the activation of MMP-2 and the shedding of CD44 in adherent cells. In contrast, TIMP-1, a poor inhibitor of MT1-MMP but a potent inhibitor of other MMPs, and ADAMs, (A Desintegrin And Metalloproteinase) (31
32
33)
, had no effect on MMP-2 activation. TIMP-1 demonstrated an insignificant inhibition of CD44 proteolysis. These results confirmed that MT1-MMP is the main individual mediator of CD44 shedding in T cells. In tumor cells, other proteases (presumably, ADAMs) (5)
are also involved in the proteolysis of CD44. According to the differential effects imposed by TIMP-1 and TIMP-2 on the shedding of T cell CD44, the combined effect of the proteases other than MT1-MMP is insignificant in diabetogenic, adherent T cells (Fig. 4)
.
It was determined earlier by Kajita et al. (34)
that coexpression of MT1-MMP and mutant CD44 lacking the MT1-MMP processing site did not result in shedding or promote migration of breast carcinoma cells, suggesting that the processing of CD44 by MT1-MMP is critical in the migratory stimulation. In agreement, our results suggest that the interaction of MT1-MMP with CD44 is highly important to migratory stimulation, transmigration, and subsequent homing of T cells to the pancreatic islets.
Inhibition of MT1-MMP proteolysis of CD44 in spontaneous IDDM
To determine its efficiency in spontaneous, newly developed IDDM, we injected AG3340 in female NOD mice that had already spontaneously developed the disease. After the onset of diabetes (Fig. 5
, day 0), control animals received insulin alone. The experimental group received insulin jointly with AG3340. The dose of AG3340 we used was an order of magnitude lower than the lowest dosage used in the phase I trials in cancer (18)
. Even at the low concentration we used in our animal experiments, the resulting levels AG3340 in the blood of NOD mice (1 mg/kg or 20 µg/mouse) were many orders of magnitude higher relative to the Ki value of this inhibitor. These levels of the inhibitor were sufficient for the inhibition of the readily accessible, T cell surface-associated protease.
|
In the course of our 1.5 month experiments, Glc levels were measured in the urine of mice of both groups (Fig. 5)
. After the development of diabetes, injection of either insulin alone or insulin with AG3340 returned Glc levels to normal in diseased NOD mice. When insulin injections were terminated, Glc levels began to rise in the animals that received insulin alone. When the concentrations of Glc approached critical levels, the mice were sacrificed. The severity of glucosuria was manifested less in the animals that received insulin jointly with AG3340, suggesting a partial reversal of the disease by the MMPI. To confirm this suggestion, we examined the pancreata of the animals, which were all sacrificed on day 44.
Thus, the pancreata were excised, sectioned, and the islets were analyzed and graded according to the observed insulitis. The sections were also stained with antibodies to insulin and glucagon to identify the functionality of the surviving and the newly formed islets (35)
.
According to our observations, a short-term treatment with AG3340 was not sufficient to return diabetic NOD mice to normoglycemia. There was, however, a promising and statistically significant reduction in the severity of insulitis in diabetic mice that received AG3340 compared with that in control (Fig. 6
). AG3340 caused an increase in the number of intact islets and the islets with limited peri-islet insulitis. In addition, AG3340 caused a recovery of insulin-producing ß cells, and possibly a de novo formation of the islet-like structures in the pancreatic parenchyma. These small, regenerating islets were observed only in animals that received the inhibitor; similar structures were not found in control mice. These islet-like structures were free from mononuclear infiltration, and produced insulin (Fig. 6)
and glucagon (not shown), thus providing evidence of the functional recovery of the hormone-secreting 
and ß cells. Identification of either of these insulin-producing ß cell structures as de novo formed islets or actually new islet structures, however, is beyond the scope of our work. In contrast, there was an intensive mononuclear infiltration, evident destruction of the islets, and a strongly reduced hormone production in the untreated control.
|
In addition, we analyzed the pancreatic islets (100/mouse) and scored the intensity of insulitis of the islets in the animals that received insulin alone and AG3340 jointly with insulin. Our unbiased analysis determined that a 40 day treatment with AG3340 was favorable for the IDDM animals. Figure 6
shows that AG3340 significantly increased the number of intact islets and islets with minor insulitis compared with the animal that received insulin alone.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 37)
, which is a subnanomolar range inhibitor of MT1-MMP, caused diabetes protection by effectively controlling an islet destructive autoimmunity and stimulating the functional recovery of the insulin-producing ß cells and, potentially, the regeneration of the pancreatic islets. It is seems unlikely that either AG3340 or related MMPIs may be used for the prophylactics of IDDM because these cancer drugs cannot be given chronically to an at-risk population that is predominantly comprised of children. The results of our in vitro experiments and animal studies in diseased NOD mice, however, imply that clinically tested hydroxamate MMPIs are capable of treating or slowing the progression of IDDM and that they potentially represent a basis for a therapy of IDDM. We confirmed that MT1-MMP proteolysis of the T cell surface CD44 adhesion receptor (a major receptor involved in the homing of the diabetogenic T cells) directly regulated the homing of T cells into the pancreas. T cell MT1-MMP (a major sheddase of T cell CD44) is activated after adhesion of T cells to the matrix. This event stimulates the shedding of T cell CD44, transmigration of the adherent cells through the endothelial cell layers, and the homing of T cells to the pancreatic islets. Conversely, the inhibition of MT1-MMP activity rescues CD44 from proteolysis and immobilizes the adherent diabetogenic, cytotoxic T cells on the vascular endothelium, thus preventing T cell homing into the islets.
It is possible that the MMPIs may exert effects in vivo not only by inhibiting T cells MT1-MMP, but also by repressing the activity of the islet-derived MMPs. Thus, it was shown in the model of the onset of type 2 diabetes that MMP-2, MMP-12, and MT1-MMP were up-regulated in diabetic male and high fat-fed female Zucker diabetic fatty rats compared with their nondiabetic lean counterparts. Consistent with our results, a broad-range MMPI preserved ß cell mass presumably by affecting the turnover of islet extracellular matrix (ECM) molecules (38)
. Even though the mechanisms of protective effects and the relative importance of the individual targets of MMPIs in IDDM and type 2 diabetes are not completely understood and their precise identification will require extensive study, it is obvious that AG3340, an anticancer hydroxamate MMPI, exerted clinically relevant effects in a mouse model of IDDM. In addition, our data suggested that AG3340 did not affect the functionality of either IS-CD8+ T cells or the splenocytes, thereby confirming that the observed effects of AG3340 are directly linked to the CD44-MT1-MMP axis.
We conclude that pharmacological inhibition of MT1-MMP by the readily available anticancer hydroxamate drugs, including AG3340, will result in a favorable outcome for IDDM patients. Because the inhibitors readily access cell surface-associated T cell MT1-MMP in the blood, low concentrations of inhibitors are required in IDDM. In contrast, cancer treatment requires high concentrations of the inhibitors. It appears that required dosages of the MMPIs in IDDM will be below the levels that will cause side effects (18)
. Because of the low dosages of the inhibitors, we believe that bioavailability and the unbalanced therapeutic ratio of MMPIs observed in cancer therapy will not be problematic in IDDM patients. Our results have the potential for therapeutic translation and provide a solid foundation and a scientific, mechanistic rationale for clinical trials of the T cell MT1-MMP-targeting MMPIs in IDDM human patients. The most recent publications on reversal of type 1 diabetes in NOD mice strengthen our results and suggest that IDDM can be reversed and that sufficient endogenous ß cell mass can be restored to cure diabetic NOD mice (39
40
41)
. Conversely, we suspect that MMPIs, including AG3340, represent the most readily available, realistic instrument to control the transmigration of the diabetic T cells to the islets and reverse, or at least to alleviate, the symptoms of IDDM in humans.
|
ACKNOWLEDGMENTS |
|---|
Received for publication March 27, 2006. Accepted for publication April 21, 2006.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  J.-A. Cho, P. Osenkowski, H. Zhao, S. Kim, M. Toth, K. Cole, A. Aboukameel, A. Saliganan, L. Schuger, R. D. Bonfil, et al. The Inactive 44-kDa Processed Form of Membrane Type 1 Matrix Metalloproteinase (MT1-MMP) Enhances Proteolytic Activity via Regulation of Endocytosis of Active MT1-MMP J. Biol. Chem., June 20, 2008; 283(25): 17391 - 17405. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Y. Savinov, D. V. Rozanov, and A. Y. Strongin Specific Inhibition of Autoimmune T Cell Transmigration Contributes to beta Cell Functionality and Insulin Synthesis in Non-obese Diabetic (NOD) Mice J. Biol. Chem., November 2, 2007; 282(44): 32106 - 32111. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
