Matrix metalloproteinases (MMPs), and in particular gelatinases (MMP-2 and MMP-9), play a key role in cancer progression. However, clinical trials in which MMP inhibitors were tested in cancer patients have been disappointing. Whereas many reasons have been postulated to explain the failure of the clinical trials, lack of inhibitor selectivity was a major limitation. Thus, despite the consensus opinion that MMP-mediated proteolysis is essential for cancer progression and that certain MMPs represent important targets for intervention, effective and selective inhibition of those MMPs remains a major challenge in drug development. We previously reported the first mechanism-based MMP inhibitor, designated SB-3CT, which is a selective gelatinase inhibitor. Here we report that SB-3CT (5-50 mg/kg/d) is a potent inhibitor of liver metastasis and increases survival in an aggressive mouse model of T-cell lymphoma. This study shows that mechanism-based inhibition of gelatinases represents a novel approach to inhibitor design that promises to be a successful anticancer therapy.

Matrix metalloproteinases (MMPs) are zinc-dependent endopeptidases that have been shown to be responsible for many of the proteolytic events leading to cancer progression. The importance of MMPs in cancer, in particular the contribution of gelatinases (MMP-9 and MMP-2) to cancer metastasis and angiogenesis (1), promoted the development of synthetic inhibitors capable of targeting gelatinase activity in tumors (2). Unfortunately, clinical trials in cancer patients with synthetic MMP inhibitors showed lack of therapeutic efficacy and unacceptable side effects (2, 3). Poor inhibitor selectivity has been postulated as one of the major reasons for the disappointing results (35). The synthetic MMP inhibitors tested in the clinic thus far were in their majority broad-spectrum zinc-chelator compounds and often incapable of discriminating among related zinc-dependent proteases (5). Other synthetic MMP inhibitors tested were optimized for the purpose of high-affinity binding (2, 6, 7). However, even when high-affinity was attained, the MMP inhibitors lacked selectivity (7, 8). These compounds also failed to show clinical efficacy and they elicited side effects (5, 9). Therefore, mere high-affinity inhibition as an approach for the design of selective MMP (gelatinase) inhibitors has failed to show promise in treatment of cancer patients. We have described a new approach to MMP inhibition involving covalent mechanism–based inhibition, which has been a point of departure from previous more conventional approaches, and we have focused on inhibition of gelatinases. In this new mechanism of MMP inhibition, we take advantage of a portion of the catalytic mechanism of the targeted gelatinase to generate from the active-site bound inhibitor a reactive chemical entity from an otherwise unreactive functional group, for covalent modification of a residue in the enzyme active site (10). Thereby the principles that impart selectivity in inhibition of gelatinases in this case go beyond mere electrostatic interactions of the classic MMP inhibitors (5, 11, 12). Because the reactive species is formed only within the active site of the targeted enzyme, these inhibitors have great promise for high in vivo selectivity, an aspect that is absent in the traditional high-affinity inhibitors for MMPs (7, 9). The prototype compound, designated SB-3CT, was designed based on these principles (10, 13). Whereas SB-3CT is an effective and selective gelatinase inhibitor, it either does not inhibit or inhibits poorly other MMPs (10) and the closely related zinc-dependent metalloprotease ADAM-17 (tumor necrosis factor-α converting enzyme; ref. 14). Here we tested the antitumor activity of SB-3CT in a mouse model of T-cell lymphoma and show that SB-3CT leads to a significant reduction in the growth and number of experimental liver metastases and to an increase in survival.

Cells, experimental metastasis assays, and SB-3CT treatment. The generation of the mouse lacZ-tagged L-CI.5s T-cell lymphoma cell line has been described (11, 15, 16). Pathogen-free female DBA/2 mice (8-10 weeks old, purchased from Charles River, Sulzfeld, Germany) were inoculated into the tail vein with 5 × 103lacZ-tagged murine L-CI.5s cells, as previously described (16). Treatment with various doses of SB-3CT (5, 12.5, 25, and 50 mg/kg) diluted in 10% DMSO or vehicle alone started 1 hour before tumor cell inoculation and continued daily. SB-3CT was administered i.p. in a volume of 400 μL per dose. Six days after tumor cell inoculation, when macrometastases are already formed (11), the mice were sacrificed and the livers were excised and stained with 5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside (X-Gal; Roche Diagnostics, Penzberg, Germany), as described (16). Blue metastatic foci on the surface of the livers were counted and confirmed by histology.

Survival studies. Tumor-bearing mice treated with 50 mg/kg/d SB-3CT or vehicle alone, as described above, were kept alive until signs of imminent death were evident. Moribund mice were then euthanized. Animal care was done in accordance with the guidelines of the Regierung von Oberbayern (Germany).

Proliferating cell nuclear antigen immunohistochemistry. Expression of the proliferating cell nuclear antigen (PCNA) marker in liver tissue sections isolated from three representative treated and untreated tumor-bearing mice was carried out by immunohistochemistry using a specific anti-PCNA antibody (Novocastra Laboratories Ltd., Newcastle, United Kingdom) and counterstaining with H&E (11). Metastases were photographed and the ratio of PCNA-positive cells per metastasis was scored.

Planimetric analysis. Sections (4 μm) of paraffin-embedded metastasis bearing livers from SB-3CT (50 mg/kg/d) and vehicle-treated mice were stained with H&E and screened for macrometastases. All metastases found in each of random sections of four animals per group were photographed and the areas of the metastases were measured using MetaMorph 4.6 software (Universal Imaging Cooperation, Downingtown, PA). Mean values and SE of all metastases were plotted.

In situ zymography. Mice were inoculated with L-CI.5s cells and treated daily with SB-3CT (50 mg/kg) or vehicle alone as described above. At day 6 after tumor cell inoculation, SB-3CT–treated mice received an additional dose of SB-3CT (50 mg/kg) 1 hour before autopsy and the control mice received vehicle alone (10% DMSO). The livers were then excised, embedded in optimum cutting temperature compound, shock frozen on dry ice, and stored at −80°C until sectioning for in situ gelatin zymography (17). Briefly, liver frozen sections (5 μm) were quickly dried on microslides and overlaid with a solution containing 1% (w/v) agarose, 0.1 mg/mL fluorescein-conjugated DQ gelatin (Molecular Probes, Leiden, Netherlands), and 2.5 μg/mL 4′,6-diamidino-2-phenylindol-dihydrochloride (DAPI; Applichem, Darmstadt, Germany) in PBS. Parallel sections were overlaid with the same solution with the addition of 10 mmol/L 1,10-phenanthroline. The overlaid sections were incubated (12 hours, 20°C) in the darkness and photographed using the AxioVert 135 microscope (Zeiss, Jena, Germany) and visualized with the AxioVision LE 4.2 software (Zeiss).

Statistical analysis. Data were analyzed using the Mann-Whitney rank sum test. Statistical analysis of survival curves was done with the log-rank test.

The T-cell lymphoma L-CI.5s cells form numerous liver colonies when inoculated i.v. into syngeneic DBA/2 mice, a process that has been shown to correlate with enhanced expression of host MMP-9 (11), and thus it mimics certain aspects of human lymphomas (18, 19). This mouse tumor model has been successfully used in the past for rapid screening of MMP inhibitors, providing a measurable and reliable assessment of inhibitor efficacy (11, 12). Therefore, we tested the ability of the gelatinase inhibitor SB-3CT to inhibit experimental metastasis in the L-CI.5s T-cell lymphoma model. To account for our mouse tumor model, we first carried out inhibition kinetic studies of purified mouse MMP-9 with SB-3CT, as described (10), and found that SB-3CT also behaves as a mechanism-based inhibitor of purified mouse MMP-9 with slow-binding onset for inhibition (data not shown) and an inhibition constant (Ki) of 120 ± 40 nmol/L, similar to the human enzyme (10).

Administration of SB-3CT to mice inoculated with L-CI.5s cells caused a dose-dependent reduction in the number of liver metastases, and at the highest inhibitor dose (50 mg/kg/d) a >70% reduction in the number of tumor colonies was observed (Figs. 1 and 2A). This is in contrast with the effects of Batimastat, a broad-spectrum hydroxamate-based MMP inhibitor (20), which led to the promotion of liver metastasis in this tumor model (12). The reduced number of liver colonies in all treatment groups (5 mg/kg/d: 71% ± 5, n = 9; 12.5 mg/kg/d: 66% ± 7, n = 7; 25 mg/kg/d: 56% ± 6, n = 7; 50 mg/kg/d: 27% ± 4, n = 9) compared with the control group was statistically significant (all P < 0.014) with 50 mg/kg/d SB-3CT being the most effective dose (P = 0.002). In this tumor model, SB-3CT was highly effective in inhibiting liver metastasis when the administration of inhibitor began 1 hour before tumor cell inoculation (Fig. 1). In contrast, the number of liver metastasis was not significantly reduced when the inhibitor was administered 1 day after tumor cell inoculation (data not shown), suggesting that SB-3CT may be more effective at inhibiting tumor cell extravasation. At the highest dose, SB-3CT also caused a significant (P < 0.001) reduction in colony size, as determined by gross examination of X-Gal–stained livers (Fig. 2A) and planimetric analysis (Fig. 2B and D) of liver sections. PCNA immunostaining in livers from treated and untreated mice (Fig. 2C) showed a significant decrease in the number of PCNA positive tumor cells per metastases in the livers of SB-3CT–treated animals when compared with control mice (control: 65.1 ± 5.8%, n = 7; SB-3CT: 39.4 ± 4.4%, n = 7; P < 0.05), suggesting that SB-3CT treatment exerts an antiproliferative effect in vivo. To determine if SB-3CT directly inhibits L-CI.5s cell proliferation or cell viability, we examined its effect on in vitro cell growth and viability and found no effect with doses up to 12.5 μmol/L of SB-3CT (data not shown). Because these inhibitor doses are unlikely to be reached in the tissues due to the insoluble nature of SB-3CT and because the liver of treated mice showed lack of evidence of tumor cell necrosis, these data strongly suggest that the effect of SB-3CT on tumor cell proliferation cannot be ascribed to a direct effect on tumor cell proliferation or cytotoxicity and may be mediated by indirect effects on the tumor microenvironment. Indeed, gelatinases, and in particular MMP-9 (21), are known to play a key role in tumor angiogenesis; thus, SB-3CT has the potential to elicit antiangiogenic effects in vivo. However, inhibition of angiogenesis by SB-3CT is not likely to be a factor in its antiproliferative effect in this tumor model as we have previously shown that angiogenesis is not involved in the development and growth of L-CI.5s metastases in highly vascularized liver tissue (11). The in vivo antiproliferative effect of SB-3CT is not unique as other synthetic MMP inhibitors were also shown to inhibit tumor cell growth in experimental models of cancer (2, 7). This is consistent with the growing evidence indicating that MMPs, including gelatinases, exert pleiotropic effects on tumor cell behavior, which are partly mediated by their ability to cleave key bioactive proteins in addition to extracellular matrix components (1, 22).

We also did in situ zymography, using fluorescein-conjugated DQ gelatin as a substrate, to assess the ability of SB-3CT to inhibit gelatinase activity in vivo. As shown in Fig. 3, a representative liver section of vehicle-treated control mice showed punctate green fluorescence at the periphery of a metastatic colony representing gelatinolytic activity (Fig. 3B,, top). Indeed, this activity was specifically inhibited by 1,10-phenanthroline consistent with being MMP mediated (Fig. 3D,, top). In contrast, liver sections of SB-3CT–treated mice showed little or no MMP-mediated gelatinolytic activity even when metastatic lesions of similar size were compared or after addition of 1,10-phenanthroline (Fig. 3B  and D, bottom). The reduced gelatinolytic activity in livers of SB-3CT–treated mice was not due to differential expression of gelatinases because gelatin zymography of liver tissue extracts showed similar levels of both MMP-9 and MMP-2 expression in SB-3CT– and vehicle-treated mice (data not shown). Thus, these studies indicate that SB-3CT inhibits gelatinolytic activity in vivo in metastasis-bearing livers. Because both MMP-2 and MMP-9 are expressed in L-CI.5s liver metastasis (11), the inhibition observed in the in situ zymography assay may be a consequence of the ability of SB-3CT to inhibit both enzymes in vivo, an activity for which it was designed (10). However, we previously showed that development of liver metastasis by L-CI.5s lymphoma cells in mice correlates with a specific induction of host MMP-9 but not MMP-2, indicating that MMP-9 is the predominant gelatinase involved in the dissemination of these T-cell lymphoma cells (11) and a likely, but yet unproven, target of SB-3CT in this tumor model.

L-CI.5s is a highly malignant tumor cell line that causes 100% mortality when inoculated i.v. in mice (15). Therefore, survival studies in this tumor model are a rigorous test of inhibitor efficacy. To this end, mice were inoculated i.v. with L-CI.5s cells as described in Materials and Methods and treated with daily doses of 50 mg/kg/d SB-3CT or with vehicle control until moribund. As shown in Fig. 4, whereas all the mice in the control group died within 11 days after tumor cell inoculation, as expected, 50% of the SB-3CT–treated mice were still alive at that time. Moreover, 20% of the SB-3CT–treated mice survived for 14 days, a statistically significant effect (P = 0.005; log-rank test) under these conditions. Neither apparent toxicity nor changes in body weight were observed between the untreated and treated mice. The observed effects on survival are remarkable considering that SB-3CT is a prototype inhibitor with poor solubility and the aggressive nature of this tumor model, which is well documented. In summary, this is the first study to show the antimetastatic activity of the first mechanism-based MMP inhibitor in an aggressive tumor model. The present results suggest that in light of previous disappointments, this approach to inhibitor design holds the promise of a novel potential recourse in anticancer therapy and deserves further exploration.

Grant support: NIH grants CA-61986 and CA-100475 (R. Fridman) and grant LSHC-CT-2003-503297 ‘Cancerdegradome’ from the European Union Framework Programme 6 (A. Krüger).

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.

The expert technical assistance of Stephanie Laforsch (Institut für Experimentelle Onkologie, München) is gratefully acknowledged. We also thank Dr. Daniel Bonfil (Department of Urology/Pathology, Wayne State University) for comments and suggestions.

1
Egeblad M, Werb Z. New functions for the matrix metalloproteinases in cancer progression.
Nat Rev Cancer
2002
;
2
:
161
–74.
2
Zucker S, Cao J, Chen WT. Critical appraisal of the use of matrix metalloproteinase inhibitors in cancer treatment.
Oncogene
2000
;
19
:
6642
–50.
3
Coussens LM, Fingleton B, Matrisian LM. Matrix metalloproteinase inhibitors and cancer: trials and tribulations.
Science
2002
;
295
:
2387
–92.
4
Matrisian LM, Sledge GW Jr, Mohla S. Extracellular proteolysis and cancer: meeting summary and future directions.
Cancer Res
2003
;
63
:
6105
–9.
5
Pavlaki M, Zucker S. Matrix metalloproteinase inhibitors (MMPIs): the beginning of phase I or the termination of phase III clinical trials.
Cancer Metastasis Rev
2003
;
22
:
177
–203.
6
Hidalgo M, Eckhardt SG. Development of matrix metalloproteinase inhibitors in cancer therapy.
J Natl Cancer Inst
2001
;
93
:
178
–93.
7
Shalinsky DR, Brekken J, Zou H, et al. Broad antitumor and antiangiogenic activities of AG3340, a potent and selective MMP inhibitor undergoing advanced oncology clinical trials.
Ann N Y Acad Sci
1999
;
878
:
236
–70.
8
Brown S, Meroueh SO, Fridman R, Mobashery S. Quest for selectivity in inhibition of matrix metalloproteinases. Curr Top Med Chem 
2004
;
4
:
1227
–38.
9
Hande KR, Collier M, Paradiso L, et al. Phase I and pharmacokinetic study of prinomastat, a matrix metalloprotease inhibitor.
Clin Cancer Res
2004
;
10
:
909
–15.
10
Brown S, Bernardo MM, Zhi-Hong L, et al. Potent and selective mechanism-based inhibition of gelatinases.
J Am Chem Soc
2000
;
122
:
6799
–800.
11
Arlt M, Kopitz C, Pennington C, et al. Increase in gelatinase-specificity of matrix metalloproteinase inhibitors correlates with antimetastatic efficacy in a T-cell lymphoma model.
Cancer Res
2002
;
62
:
5543
–50.
12
Krüger A, Soeltl R, Sopov I, et al. Hydroxamate-type matrix metalloproteinase inhibitor batimastat promotes liver metastasis.
Cancer Res
2001
;
61
:
1272
–5.
13
Kleifeld O, Kotra LP, Gervasi DC, et al. X-ray absorption studies of human matrix metalloproteinase-2 (MMP-2) bound to a highly selective mechanism-based inhibitor. comparison with the latent and active forms of the enzyme.
J Biol Chem
2001
;
276
:
17125
–31.
14
Solomon A, Rosenblum G, Gonzales PE, et al. Pronounced diversity in electronic and chemical properties between the catalytic zinc sites of tumor necrosis factor-α-converting enzyme and matrix metalloproteinases despite their high structural similarity.
J Biol Chem
2004
;
279
:
31646
–54.
15
Krüger A, Umansky V, Rocha M, Hacker HJ, Schirrmacher V, von Hoegen P. Pattern and load of spontaneous liver metastasis dependent on host immune status studied with a lacZ transduced lymphoma.
Blood
1994
;
84
:
3166
–74.
16
Krüger A, Schirrmacher V, von Hoegen P. Scattered micrometastases visualized at the single-cell level: detection and re-isolation of lacZ-labeled metastasized lymphoma cells.
Int J Cancer
1994
;
58
:
275
–84.
17
Frederiks WM, Mook OR. Metabolic mapping of proteinase activity with emphasis on in situ zymography of gelatinases: review and protocols.
J Histochem Cytochem
2004
;
52
:
711
–22.
18
Kuittinen O, Apaja-Sarkkinen M, Turpeenniemi-Hujanen T. Gelatinases (MMP-2 and MMP-9), TIMP-1 expression and the extent of neovascularization in aggressive non-Hodgkin's lymphomas.
Eur J Haematol
2003
;
71
:
91
–9.
19
Sakata K, Satoh M, Someya M, et al. Expression of matrix metalloproteinase 9 is a prognostic factor in patients with non-Hodgkin lymphoma.
Cancer
2004
;
100
:
356
–65.
20
Brown PD. Matrix metalloproteinase inhibitors.
Breast Cancer Res Treat
1998
;
52
:
125
–36.
21
Bergers G, Brekken R, McMahon G, et al. Matrix metalloproteinase-9 triggers the angiogenic switch during carcinogenesis.
Nat Cell Biol
2000
;
2
:
737
–44.
22
Lopez-Otin C, Overall CM. Protease degradomics: a new challenge for proteomics.
Nat Rev Mol Cell Biol
2002
;
3
:
509
–19.