Selective Inhibition of MMP-2 Does Not Alter Neurological Recovery

Aug 23, 2016 - Matrix metalloproteinase (MMP)-2 knockout (KO) mice show impaired neurological recovery after spinal cord injury (SCI), suggesting that...
0 downloads 0 Views 2MB Size
Download PDF
Letter pubs.acs.org/chemneuro

Selective Inhibition of MMP‑2 Does Not Alter Neurological Recovery after Spinal Cord Injury Ming Gao,†,⊥ Haoqian Zhang,§,⊥ Alpa Trivedi,§ Kiran V. Mahasenan,† Valerie A. Schroeder,‡ William R. Wolter,‡ Mark A. Suckow,‡ Shahriar Mobashery,† Linda J. Noble-Haeusslein,§,∥ and Mayland Chang*,† †

Department of Chemistry and Biochemistry and ‡Freimann Life Sciences Center and Department of Biological Sciences, University of Notre Dame, Notre Dame, Indiana 46556, United States § Department of Neurological Surgery and Physical Therapy and ∥Rehabilitation Science, University of California, San Francisco, San Francisco, California 94143, United States ABSTRACT: Matrix metalloproteinase (MMP)-2 knockout (KO) mice show impaired neurological recovery after spinal cord injury (SCI), suggesting that this proteinase is critical to recovery processes. However, this finding in the KO has been confounded by a compensatory increase in MMP-9. We synthesized the thiirane mechanism-based inhibitor ND-378 and document that it is a potent (nanomolar) and selective slow-binding inhibitor of MMP-2 that does not inhibit the closely related MMP-9 and MMP-14. ND378 crosses the blood-spinal cord barrier, achieving therapeutic concentrations in the injured spinal cord. Spinal-cord injured mice treated with ND-378 showed no change in long-term neurological outcomes, suggesting that MMP-2 is not a key determinant of locomotor recovery. KEYWORDS: MMP-2, ND-378, brain distribution, spinal cord injury

S

compensatory increase in MMP-9 activity,5 and thus the role MMP-2 in SCI has not been conclusively established.

pinal cord injury (SCI) affects approximately 12 000 individuals in the United States every year,1 of whom 5000 die either before reaching the hospital or during hospitalization.2 Life expectancy for persons with paraplegia after SCI at 40 years of age is 27.6 years, significantly below the life expectancy of 39.4 years for those with no SCI.1 The only therapy that has proven efficacious in reducing the extent of paralysis is high-dose methylprednisolone; however, it has to be administered within 8 h of injury.3 Acute SCI involves the initial mechanical injury (primary injury), followed by a biochemical cascade of events that includes disruption of the blood-spinal cord barrier, resulting in cellular damage and cell death (secondary injury).4 Matrix metalloproteinases (MMPs), in particular MMP-9, have been shown to play a detrimental role in acute SCI.5−7 An improvement in locomotor recovery is observed in MMP-9 knockout (KO) mice after SCI and in mice treated with the broad-spectrum MMP inhibitor GM6001 during the acute phase of SCI.6 However, extended broadspectrum MMP inhibition in spinal-cord injured mice failed to support long-term motor recovery,8 suggesting that MMPs, expressed during wound healing, may be critical to the recovery process. MMP-2 activity was observed by gelatin zymography in the wound-healing phase of SCI (7 and 14 days post injury), and spinal-cord injured MMP-2 KO mice showed pronounced long-term neurological impairments.5 Such findings led to the hypothesis that this proteinase may be integral to the recovery process.5 However, the spinal-cord injured MMP-2 KO shows a © XXXX American Chemical Society

Design of selective MMP inhibitors has proven to be challenging due to the high structural similarities of the 23 human MMPs.9 SB-3CT (compound 1) is a prototype mechanism-based thiirane inhibitor, which inhibits MMP-2 and MMP-9 with inhibition constants Ki of 28 ± 7 nM and 400 ± 15 nM, respectively.10 SB-3CT was designed in silico and exhibited high selectivity in inhibition of MMP-2 and MMP9.11 The structure of this prototype inhibitor has been elaborated by syntheses of a few hundred analogs and screening for their breadth of activity among MMPs. These efforts led to inhibitors 2 and 3. The thiirane class of inhibitors would appear to be privileged for treatment of neurological ailments, as these inhibitors penetrate the blood-brain and blood-spinal cord barriers.12 The thiiranes are not metal chelators, as most broadReceived: July 25, 2016 Accepted: August 23, 2016

A

DOI: 10.1021/acschemneuro.6b00217 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

Letter

ACS Chemical Neuroscience Scheme 1. Synthesis of ND-378 (3)

residence time for MMP-2 compared to MMP-8, combined with the 3-fold more potent inhibition of MMP-2, should result in sustained inhibition of MMP-2 even when concentrations of ND-378 are below the Ki value. In order to rationalize the selectivity of ND-378, we carried out molecular docking of the compound to the catalytic sites of MMP-2, MMP-9, and MMP14. The caveat to this analysis is that if a rather large conformational change were to take place on inhibitor binding, we cannot account for it. Furthermore, since an X-ray structure for any MMP bound to a thiirane inhibitor does not exist presently, we have used the complex for MMP-2 bound to the thiirane generated based on quantum mechanics/molecular mechanics (QM/MM analysis).17 However, the results of docking explain a number of features that emerged from the inhibition analysis of ND-378 with the panel of MMPs (Figure 1).

spectrum MMP inhibitors are. The root of the selective inhibition by SB-3CT is the gelatinase (MMP-2 and MMP-9)promoted opening of the thiirane ring that results in a thiolate, which then serves as a picomolar tight-binding inhibitor,13 with reversal of inhibition occurring exceedingly slowly. The thiirane class of inhibitors is not toxic and not mutagenic.14 We recently reported the discovery of ND-336 (compound 2), a watersoluble gelatinase and MMP-14 inhibitor, with Ki values of 85 ± 1 nM for MMP-2, 150 ± 10 nM for MMP-9, and 120 ± 10 nM for MMP-14.14 We discovered that the mere acetylation of the amine in 2 (ND-378, compound 3), led to a selective nanomolar slow-binding inhibitor of MMP-2 that does not inhibit the closely related MMP-9 or MMP-14. This very narrow spectrum for inhibition is unprecedented. Furthermore, we document that ND-378 crosses the blood-spinal cord barrier, achieving therapeutic concentrations in the injured spinal cord. We also observed that administration of ND-378 did not have an effect on locomotor functions in a mouse model of SCI, suggesting that MMP-2 is not a key determinant of locomotor recovery in SCI. For the synthesis of ND-378 (3), we prepared ND-336 (2) from 4-mercaptophenol (4) in seven steps as described previously,14 followed by N-acetylation of 2 using acetyl chloride, in the presence of TEA to afford ND-378 (Scheme 1). ND-378 is a potent slow-binding inhibitor of MMP-2 (Ki 230 ± 10 nM, Table 1); it poorly inhibits other MMPs and Table 1. MMP Kinetic Parameters of ND-378 MMP

kon (×105 μM−1 s−1)

koff (×103 s−1)

Ki (nM)

MMP-2 MMP-9cat MMP-14cat MMP-1cat MMP-3cat MMP-7 MMP-8 ADAM9 ADAM10

3.9 ± 0.05

0.9 ± 0.02

230 ± 10 23% at 50 μM 23% at 50 μM 37% at 100 μM 14% at 100 μM 5% at 100 μM 690 ± 40 31% at 100 μM 8% at 100 μM

linear competitive

Figure 1. Stereo view of ND-378 docked to the MMP-2 catalytic site. The inhibitor is represented in capped sticks, with light gray for carbons, blue for nitrogen, red for oxygen, and yellow for sulfur. Relevant loop amino acid residues are represented in capped sticks, with purple for carbon. The zinc ion is shown in gray sphere representation. The Connolly surface was generated for the protein residues excluding that in the loop region, which covers the cavity. Hydrogen bonds between the inhibitor and the protein are shown as black dotted lines. ND-378, in the absence of steric hindrance, can form favorable hydrogen bonds with the backbone carbonyl oxygen atoms of MMP-2 loop residues. The positions of the bulkier residues in other MMPs, Arg424 of MMP-9 and Gln262 and Met264 of MMP14, are near the 5 o’clock position of the loop.

ADAMs, including MMP-9 and MMP-14. ND-378 inhibits MMP-8 as a linear-competitive inhibitor with a Ki of 690 ± 40 nM. One of the most important factors in sustaining efficacy in vivo is the drug-target complex residence time, the duration in which the drug is physically bound to the target.15 The longer the residence time is, the longer is the duration of pharmacological effect. Residence time can be calculated as the reciprocal of the dissociation rate constant (koff).15 The residence time for ND-378 bound to MMP-2 is 18.2 ± 0.4 min, longer than those for the complexes of MMP-2−TIMP-1 or MMP-2−TIMP-2, which are 7 and 10 min, respectively.16 This is an important finding, as TIMPs are protein inhibitors of MMPs and have evolved for inhibition of these enzymes. In essence, ND-378 is more effective in inhibition of the targeted MMP-2 than are TIMPs. Because of the linear-competitive inhibition of MMP-8 by ND-378, it results in a very short residence time (95%, as confirmed by ultraperformance liquid chromatography (UPLC). Conditions are detailed in the UPLC section. (4-(4-((Thiiran-2-ylmethyl)sulfonyl)phenoxy)phenyl)-methanamine HCl Salt (2). ND-336 (compound 2) was prepared following the procedure described previously.14 N-(4-(4-((Thiiran-2-ylmethyl)sulfonyl)phenoxy)ben-zyl)acetamide (3). Acetyl chloride (43.6 μL, 0.61 mmol) in THF (0.5 mL) was slowly

Figure 4. Effect of ND-378 on catwalk assessment. (A) Stride length, (B) hindpaw base of support, (C) hindpaw swing speed, (D) hindpaw maximum area, and (E) regularity index. There were no significant differences between the ND-378 and vehicle groups (mean + SEM, n = 8 ND-378 and n = 9 vehicle). D

DOI: 10.1021/acschemneuro.6b00217 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

Letter

ACS Chemical Neuroscience

Pharmacokinetic Parameters. The methodology for the calculation of pharmacokinetic parameters listed in Table 2 was the same as reported previously.19 SCI Mouse Model. The procedures involving vertebrate animals for the SCI mouse model were approved by the Institutional Animal Care and Use Committee at the University of California San Francisco. Adult male mice (3 months of age, n = 15 ND-378 and 15 vehicle) were anesthetized and subjected to a moderate spinal cord contusion injury. Briefly, a laminectomy was performed at the T9 vertebral level, and a 2-g weight was dropped 5 cm onto the exposed dura mater. The injured mice were then randomly treated with either ND-378 (25 mg/kg) or vehicle over a period of time when MMP-2 gradually increases in the injured cord. The first dose was given ip at 24 h after injury, with a second dose given sc at 1 h after the first ip dose, and subsequent sc doses once a day for 6 days (total of 8 doses). ND-378 was formulated as described for the PK and brain distribution studies. Behavioral Assessments. All assessments were conducted randomized and blinded to the experimental condition. Grid walking and Catwalk were conducted on animals that showed a score of ≥4 on the Basso Mouse Scale. Basso Mouse Scale (BMS) for Locomotion. Locomotor recovery was assessed in an open field using the BMS scale at 1, 3, and 7 days and weekly thereafter for 6 weeks (n = 14 for ND-378treated group, n = 15 for vehicle group). This scale ranges from 0 (complete paralysis) to 9 (normal movement). Each limb was evaluated separately, the values were then averaged to achieve the final BMS score. Grid Walk. One week prior to euthanasia, each animal was positioned on a grid (0.5 × 0.5 in. spacing), in an open field where overall movement was recorded. We determined the total distance moved and the number of missteps (paw slips through a grid) over a period of 4 min (n = 8 for ND-378-treated group, n = 9 for the vehicle group). Catwalk. Injured mice traversed an illuminated glass platform while a video camera recorded from below. One week prior to euthanasia, mice were trained to cross the walkway. Gait-related parameters such as stride length (the distance between two consecutive paw placements), paw swing speed (this parameter is computed from stride length and swing duration), maximum area (the paw area contacted at the moment of maximal paw-floor contact, during stance), base of support (the distance between the two hind paws), and regularity index (grades the degree of coordination and represents the percentage of normal step sequence patterns) were analyzed for each animal, across three trials (n = 8 for the ND-378-treated group, n = 9 for the vehicle group). Concentrations of ND-378 in Plasma, Brain, and Spinal Cord of SCI Mice. Mice (n = 3 per time point) were subjected to a moderate spinal cord contusion injury, and given the same dose regimen of ND-378 as described above. Terminal blood, brain, and spinal cord were harvested at 0.5, 1, 2, and 4 h after the last dose. The spinal cord was dissected into the injury site and the cervical site. Blood samples were centrifuged to obtain plasma. Brain and spinal cord samples were weighed, immediately flash-frozen in liquid nitrogen, and stored at −80 °C until analysis. Concentrations of ND-378 in plasma, brain, and spinal cord were determined by UPLC with MRM detection relative to internal standard from calibration curves and regression parameters as described above. Statistical Analysis. Data are expressed as mean values ± SEM. Repeated-measure two-way analysis of variance was used for BMS assessments. An unpaired, two-tailed Student’s t test was used for Grid Walk and Catwalk. All the statistical analyses were performed using Prism software (version 6.0; GraphPad).

Systems (Minneapolis, MN). The Km values for MMP-2, MMP-9 and MMP-14 were the same as previously reported by Gooyit et al.19 Inhibitor stock solutions (10 mM) were prepared fresh in DMSO before enzyme inhibition assays. We followed the same methodology for enzyme inhibition studies as reported before by Page-McCaw et al.20 Enzyme inhibition studies were carried out using a Cary Eclipse fluorescence spectrophotometer (Varian, Walnut Creek, CA). ND-378 was stable in the buffers that were used in the kinetic assays. Computational Analysis. MMP-2 protein coordinates were obtained from our previous QM/MM study.17 Coordinates of MMP-9 and MMP-14 were downloaded from the Protein Data Bank (PDB codes of 1GKC and 3MA2, respectively) and prepared using Protein Preparation Wizard via Maestro v 9.3.5 (Schrödinger LLC, NY). ND-378 was prepared using LigPrep v2.55. Molecular docking of ND-378 to the catalytic site of MMPs was carried out with Glide v5.8, implementing 1 Å core restraints of the thiirane group and scored with Standard Precision.21 PK and Brain Distribution Studies. Mice (male CD-1, 6−7 weeks old, ∼30 g body weight, specific pathogen free) were purchased from Charles River Laboratories, Inc. (Wilmington, MA). Mice were fed Teklad 2019 Extruded Rodent Diet (Harlan, Madison, WI) and provided with water ad libitum. Animals were housed in polycarbonate shoebox cages containing Bed-o’Cobs 1/4″ (The Andersons Inc., Maumee, OH)/Alpha-dri (Sheperd Specialy Papers, Inc., Richland, MI) bedding under a 12 h light/12 h dark cycle at 72 ± 2 °F. All procedures involving vertebrate animals for the PK and brain distribution studies were approved by the Institutional Animal Care and Use Committee at the University of Notre Dame. Dosing formulations were sterilized by passage through an Acrodisc syringe filter containing a 2 μm, 13 mm diameter PTFE membrane (Pall Life Sciences, Port Washington, NY). For iv administration, ND378 was dissolved in 61% water/27% propylene glycol/12% DMSO at a concentration of 1.25 mg/mL. For ip administration, ND-378 was dissolved in 45% water/44% propylene glycol/11% DMSO at a concentration of 7.5 mg/mL. For sc administration, ND-378 was dissolved in 30% water/45% propylene glycol/25% DMSO at a concentration of 7.5 mg/mL. Mice were given a single 120 μL iv dose (equivalent to 5 mg/kg) or single 100 μL ip or sc doses (equivalent to 25 mg/kg). Three mice per time point per route of administration were used. At a specific time point, mice were euthanized, and terminal blood was collected by cardiac puncture using heparin. Transcardial perfusion with saline was performed before collection of brain samples. For iv and ip administration, blood and brain were collected at 2, 5, 10, 30, 60, 90, 120, and 240 min. For sc administration, blood and brain were collected at 20, 40, 60, 120, 240, and 480 min. Blood samples were centrifuged to obtain plasma. Brain samples were weighed, immediately flash-frozen in liquid nitrogen, and stored at −80 °C until analysis. Sample Analysis. The preparation of plasma and brain samples as well as calibration curves in control plasma and brain for quantification analysis were the same as described previously.19 Samples were analyzed by UPLC/(+) ESI-multiple-reaction monitoring (MRM) with a reversed phase C18 column (Acclaim RSLC 120 C18, 2.2 μm, 120 Å, 2.1 × 100 mm, Dionex, Sunnyvale, CA). The chromatographic and mass spectrometric conditions were the same as previously reported,19 except for the following: the capillary voltage, cone voltage, extractor voltage, and RF lens voltage were set at 4.6 kV, 25, 3, and 0.1 V, respectively; the cone gas-flow rate was set at 50 L/h (nitrogen). The MRM transitions were 378 → 182 for ND-378 and 300 → 93 for internal standard 5. Quantification of the compounds in plasma and brain was obtained using peak-area ratios of the compounds to the internal standard, and the linear-regression parameters obtained from the calibration curves. The coefficients of determination (R2) were >0.99, and the assays were linear up to concentrations of 100 μM.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. E

DOI: 10.1021/acschemneuro.6b00217 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX

Letter

ACS Chemical Neuroscience Author Contributions

Chemistry: Design, Synthesis, Evaluation (Bräse, S., Ed.), pp 262−286, The Royal Society of Chemistry. (13) Forbes, C., Shi, Q. C., Fisher, J. F., Lee, M., Hesek, D., Llarrull, L. I., Toth, M., Gossing, M., Fridman, R., and Mobashery, S. (2009) Active Site Ring-Opening of a Thiirane Moiety and Picomolar Inhibition of Gelatinases. Chem. Biol. Drug Des. 74, 527−534. (14) Gao, M., Nguyen, T. T., Suckow, M. A., Wolter, W. R., Gooyit, M., Mobashery, S., and Chang, M. (2015) Acceleration of diabetic wound healing using a novel protease-anti-protease combination therapy. Proc. Natl. Acad. Sci. U. S. A. 112, 15226−15231. (15) Copeland, R. A., Pompliano, D. L., and Meek, T. D. (2006) Drug-target residence time and its implications for lead optimization. Nat. Rev. Drug Discovery 5, 730−739. (16) Olson, M. W., Gervasi, D. C., Mobashery, S., and Fridman, R. (1997) Kinetic analysis of the binding of human matrix metalloproteinase-2 and −9 to tissue inhibitor of metalloproteinase (TIMP)-1 and TIMP-2. J. Biol. Chem. 272, 29975−29983. (17) Zhou, J., Tao, P., Fisher, J. F., Shi, Q., Mobashery, S., and Schlegel, H. B. (2010) QM/MM Studies of the Matrix Metalloproteinase 2 (MMP2) Inhibition Mechanism of (S)-SB-3CT and its Oxirane Analogue. J. Chem. Theory Comput. 6, 3580−3587. (18) Davies, B., and Morris, T. (1993) Physiological parameters in laboratory animals and humans. Pharm. Res. 10, 1093−1095. (19) Gooyit, M., Song, W., Mahasenan, K. V., Lichtenwalter, K., Suckow, M. A., Schroeder, V. A., Wolter, W. R., Mobashery, S., and Chang, M. (2013) O-Phenyl Carbamate and Phenyl Urea Thiiranes as Selective Matrix Metalloproteinase-2 Inhibitors that Cross the BloodBrain Barrier. J. Med. Chem. 56, 8139−8150. (20) Page-McCaw, A., Ewald, A. J., and Werb, Z. (2007) Matrix metalloproteinases and the regulation of tissue remodelling. Nat. Rev. Mol. Cell Biol. 8, 221−233. (21) Friesner, R. A., Banks, J. L., Murphy, R. B., Halgren, T. A., Klicic, J. J., Mainz, D. T., Repasky, M. P., Knoll, E. H., Shelley, M., Perry, J. K., Shaw, D. E., Francis, P., and Shenkin, P. S. (2004) Glide: a new approach for rapid, accurate docking and scoring. 1. Method and assessment of docking accuracy. J. Med. Chem. 47, 1739−1749.

⊥

M.G. and H.Z. contributed equally. M.G. synthesized ND378, evaluated its MMP kinetics, and performed the PK and brain/spinal-cord distribution experiments. H.Z. performed the mouse SCI studies. A.T. assisted in the design of the blinding and randomization of the SCI experiments and performed the initial data analysis. K.V.M. performed the computational studies. V.A.S., W.R.W., and M.A.S. performed the in-life portion of the PK and brain-distribution animal studies. L.J.N.H. designed the SCI studies. M.C. conceived and designed the experiments for this project. S.M., L.J.N.-H., and M.C. wrote the manuscript, with input from the authors. Funding

This work was supported by a grant from the Craig H. Neilsen Foundation. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank Sandra Canchola and the Neurobehavioral Core for Rehabilitation Research at UCSF for providing the facilities to conduct the behavioral analyses.

■

REFERENCES

(1) (2011) Spinal cord injury facts and figures at a glance, J. Spinal Cord Med. 34, 620−621.10.1179/204577211X13218754005537 (2) Sekhon, L. H., and Fehlings, M. G. (2001) Epidemiology, demographics, and pathophysiology of acute spinal cord injury. Spine (Philadelphia) 26, S2−12. (3) Bracken, M. B. (2012) Steroids for acute spinal cord injury. Cochrane Database Syst. Rev. 1, CD001046. (4) Mautes, A. E., Weinzierl, M. R., Donovan, F., and Noble, L. J. (2000) Vascular events after spinal cord injury: contribution to secondary pathogenesis. Phys. Ther. 80, 673−687. (5) Hsu, J. Y., McKeon, R., Goussev, S., Werb, Z., Lee, J. U., Trivedi, A., and Noble-Haeusslein, L. J. (2006) Matrix metalloproteinase-2 facilitates wound healing events that promote functional recovery after spinal cord injury. J. Neurosci. 26, 9841−9850. (6) Noble, L. J., Donovan, F., Igarashi, T., Goussev, S., and Werb, Z. (2002) Matrix metalloproteinases limit functional recovery after spinal cord injury by modulation of early vascular events. J. Neurosci. 22, 7526−7535. (7) Zhang, H., Chang, M., Hansen, C. N., Basso, D. M., and NobleHaeusslein, L. J. (2011) Role of matrix metalloproteinases and therapeutic benefits of their inhibition in spinal cord injury. Neurotherapeutics 8, 206−220. (8) Trivedi, A., Hsu, J. Y., Lin, Y., Goussev, S., Gan, J., Topp, K. S., and Noble-Haeusslein, L. J. (2005) The effects of acute and extended inhibition of matrix metalloproteinases on demyelination and functional recovery after spinal cord injury. Int. J. Neuroprt. Neuroregener. 2, 30−38. (9) Fanjul-Fernandez, M., Folgueras, A. R., Cabrera, S., and LopezOtin, C. (2010) Matrix metalloproteinases: evolution, gene regulation and functional analysis in mouse models. Biochim. Biophys. Acta, Mol. Cell Res. 1803, 3−19. (10) Lee, M., Bernardo, M. M., Meroueh, S. O., Brown, S., Fridman, R., and Mobashery, S. (2005) Synthesis of chiral 2-(4phenoxyphenylsulfonylmethyl)thiiranes as selective gelatinase inhibitors. Org. Lett. 7, 4463−4465. (11) Brown, S., Bernardo, M. M., Li, Z. H., Kotra, L. P., Tanaka, Y., Fridman, R., and Mobashery, S. (2000) Potent and selective mechanism-based inhibition of gelatinases. J. Am. Chem. Soc. 122, 6799−6800. (12) Gooyit, M., Peng, Z., Mobashery, S., and Chang, M. (2016) Thiirane class of gelatinase inhibitors as a priviledged template that crosses the blood-brain barrier, in Privileged Scaffolds in Medicinal F

DOI: 10.1021/acschemneuro.6b00217 ACS Chem. Neurosci. XXXX, XXX, XXX−XXX