Identification and neuroprotective properties of NA‐184, a calpain‐2 inhibitor

Abstract

Our laboratory has shown that calpain‐2 activation in the brain following acute injury is directly related to neuronal damage and the long‐term functional consequences of the injury, while calpain‐1 activation is generally neuroprotective and calpain‐1 deletion exacerbates neuronal injury. We have also shown that a relatively selective calpain‐2 inhibitor, referred to as C2I, enhanced long‐term potentiation and learning and memory, and provided neuroprotection in the controlled cortical impact (CCI) model of traumatic brain injury (TBI) in mice. Using molecular dynamic simulation and Site Identification by Ligand Competitive Saturation (SILCS) software, we generated about 130 analogs of C2I and tested them in a number of in vitro and in vivo assays. These led to the identification of two interesting compounds, NA‐112 and NA‐184. Further analyses indicated that NA‐184, (S)‐2‐(3‐benzylureido)‐N‐((R,S)‐1‐((3‐chloro‐2‐methoxybenzyl)amino)‐1,2‐dioxopentan‐3‐yl)‐4‐methylpentanamide, selectively and dose‐dependent inhibited calpain‐2 activity without evident inhibition of calpain‐1 at the tested concentrations in mouse brain tissues and human cell lines. Like NA‐112, NA‐184 inhibited TBI‐induced calpain‐2 activation and cell death in mice and rats, both male and females. Pharmacokinetic and pharmacodynamic analyses indicated that NA‐184 exhibited properties, including stability in plasma and liver and blood–brain barrier permeability, that make it a good clinical candidate for the treatment of TBI.

NA‐184 is a novel calpain‐2 inhibitor, which provides neuroprotection when administered after TBI.

Abbreviations:

CCI

controlled cortical impact

C2I

calpain‐2 inhibitor

HEK Cells

human embryonic kidney ells

PBS

phosphate buffer saline

LTP

long‐term potentiation

SBDP

spectrin break‐down product

SILCS

site identification by ligand competitive saturation

TBI

traumatic brain injury

TUNEL

terminal deoxynucleotidyl transferase dUTP nick end labeling

1. INTRODUCTION

Traumatic brain injury (TBI) is a significant public health problem in the United States. There were over 220 000 TBI‐related hospitalizations in 2019 and 69 473 TBI‐related deaths in 2021, making it the 8th largest cause of death (https://www.cdc.gov). Among the different types of TBI, penetrating traumatic brain injuries produce the worst outcomes and highest mortality rates. ¹ TBI induces immediate neuropathological consequences, including neurodegeneration ² and axonal damage. ³

Numerous reviews have discussed the role of calpain in neurodegenerative diseases, ⁴ , ⁵ including stroke ⁶ , ⁷ and TBI. ⁸ , ⁹ Consequently, numerous studies have attempted to reduce neurodegeneration in both stroke and TBI using calpain inhibitors. ¹⁰ , ¹¹ Results of these studies have been ambiguous, as some studies using the first generation of calpain inhibitors in TBI reported beneficial effects, ¹² while other studies did not. ¹³ , ¹⁴ Importantly, all these studies used nonselective calpain inhibitors, which did not differentiate calpain‐1 (https://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=2336) and calpain‐2 (http://www.guidetopharmacology.org/GRAC/FamilyDisplayForward?familyId=731), the major calpain isoforms in the brain. Several reasons could account for the failure to develop clinical applications with such inhibitors, including their lack of specificity/potency/selectivity, ¹⁵ and the incomplete knowledge regarding the functions of calpain‐1 and calpain‐2 (aka μ‐ and m‐calpain) in the brain. Work from our laboratory over the last 10 years has revealed that calpain‐1 and calpain‐2 play opposite roles in both synaptic plasticity and neuroprotection/neurodegeneration. ¹⁶ Thus, calpain‐1 activation is required for theta burst stimulation‐induced long‐term potentiation (LTP) and is neuroprotective. ¹⁷ , ¹⁸ On the other hand, calpain‐2 activation limits the magnitude of LTP and is neurodegenerative. ¹⁷ , ¹⁸ These findings could explain the failure of the previous studies to convincingly demonstrate the role of calpain in neurodegeneration and the lack of clear efficacy of the previously tested calpain inhibitors, which did not discriminate between calpain‐1 and calpain‐2.

We previously found that the peptidyl α‐ketoamide, Z‐Leu‐Abu‐CONH‐CH₂‐C₆H₃(3,5‐(OMe)₂), ¹⁹ referred to as C18 (aka C2I or NA‐101), with a reported 100‐fold difference in K _i values for calpain‐2 vs calpain‐1, did enhance the magnitude of TBS‐induced LTP ¹⁸ and facilitated learning and memory in mice. ²⁰ It also provided significant neuroprotection in both a mouse model of acute glaucoma ²¹ and a mouse model of TBI. ²² However, in all our experiments, this compound exhibited an inverted U‐shape dose response curve, with low doses providing learning facilitation and neuroprotection and high doses eliciting the opposite effects, which reflected the inhibition of calpain‐2 and calpain‐1, respectively. Based on this finding, we initiated a program of compound optimization to identify a calpain inhibitor more selective for calpain‐2 than calpain‐1 and with better properties for further drug development for the treatment of TBI.

2. MATERIALS AND METHODS

2.1. Molecular dynamics simulation

The crystal structures of rat calpain‐1 (PDB 2R9C) and rat calpain‐2 (PDB 3BOW) are mutated in silico to human's using Molecular Operating Environment (MOE) ²³ to be consistent with the experimental assay. CHARMM36 force field ²⁴ was used for all simulations. CHARMM36 force fields for deprotonated cysteine and protonated histidine were used for the catalytic site. ²⁵ The ketoamide warhead was re‐parameterized previously. ²⁵ All the MD simulation systems in explicit solvent were prepared by using the CHARMM‐GUI. ²⁶ Each system was solvated into a rectangular water box consisted of CHARMM TIP3P water molecules ²⁷ and 150 mM KCl, with an edge distance of 10 Å. All the MD simulations were performed using NAMD ²⁸ under periodic boundary conditions at a constant temperature of 300 K and pressure of 1 atm (NPT ensemble). ²⁹ A smoothing function was applied to van der Waals forces between 10 and 12 Å. The solvated complexes were minimized and equilibrated using a stepwise procedure set up by the CHARMM‐GUI.

To assist our medicinal chemistry campaign, we used Site Identification by Ligand Competitive Saturation (SILCS) program ³⁰ to rank and prioritize chemical modification. Our previous study showed that docking the P1′ fragments produced higher ranking accuracy among reversible covalent analogs than docking the whole ligand in the calpain catalytic site. ²⁵ Hence, the binding affinity ranking was done following the protocol described in. ²⁵

2.2. Analog synthesis

Synthesis of analogs of C2I was performed using traditional methods for peptidomimetics. All materials were obtained from commercial sources. The purity of each compound was checked by HPLC and ¹H NMR and mass spectrometry. We are reporting below the synthesis of two of these analogs, which represent the most extensively studied members of the family of molecules synthesized in this project.

2.2.1. Preparation of NA‐112

Compound NA‐112 was prepared in 10 steps from commercially available raw materials as summarized in Scheme 1.

SCHEME 1.

10‐step reaction for NA‐112 synthesis.

The final product was purified by column chromatography on silica gel, followed by recrystallization from dichloromethane to provide an off‐white solid (98.4% purity by LC/MS). The product exists as a mixture of diastereomers, with the (S)‐Leu center fixed.

2.2.2. Preparation of NA‐184

Compound NA‐184 was prepared in 11 steps from commercially available raw materials as summarized in Scheme 2.

SCHEME 2.

11‐step reaction for NA‐184 synthesis.

The final product was purified by column chromatography on silica gel, followed by precipitation from a mixture of MeTHF and MTBE to provide an off‐white solid (99.1% purity by HPLC). The product exists as a mixture of diastereomers, with the (S)‐Leu center fixed.

2.2.3. Calpain assays

In vitro calpain assays

The hydrolysis of the fluorogenic substrate Suc‐Leu‐Tyr‐AMC by calpain‐1 and calpain‐2 was performed as previously described, ³¹ with minor modifications. Briefly, purified calpain‐1 from human or porcine erythrocytes (8 μg, Millipore) or human or rat recombinant calpain‐2 produced by bacterial expression as described by ³² (plasmids were generous gifts from Dr. Peter Davies from Queen's University, Ontario, CA) was incubated with Suc‐Leu‐Tyr‐AMC (0.5 mM) in 60 mM imidazole‐HCl buffer, pH 7.3, containing 5 mM CaCl₂, 5 mM cysteine, 2.5 mM β‐mercaptoethanol, and different concentrations the various molecules (ranging from 0 to 20 μM). The reaction was initiated by adding the enzyme and continued at 30°C for 15 min, while the fluorescence of 7‐amino‐4‐methylcoumarin (Ex 380 nm/Em 450 nm) was monitored every 30 s in a POLARstar Omega fluorescence microplate reader (BMG Labtech). The rate of hydrolysis (increase in fluorescence/min) was determined from the linear portion of the curve. The IC₅₀ values were obtained by adjusting data from each experiment into a sigmoidal dose–response curve. The K _i values were calculated from the average of the IC₅₀ values and from a single substrate concentration by using a K _i calculator tool for fluorescence‐based competitive binding assays (http://sw16.im.med.umich.edu/software/calc_ki/). The Km values of Suc‐Leu‐Tyr‐AMC used for the K _i calculation were 4.74 and 2.21 mM for calpain‐1 and calpain‐2, respectively, as previously reported. ³¹

Ex‐vivo calpain assays

Mouse pons‐cerebellum was used to provide a source of endogenous calpain‐1 and calpain‐2, as these regions exhibit the highest levels of calpain activity. ³³ Membrane fractions were prepared by centrifugation of homogenates and were resuspended in 60 mM imidazole–HCl buffer, pH 7.3, containing 5 mM cysteine, 2.5 mM β‐mercaptoethanol. Calpain‐1 activity was measured by adding 20 μM calcium and the fluorogenic substrate (Suc‐Leu‐Tyr‐AMC, 0.5 mM), while total calpain (calpain‐1 and calpain‐2) was measured in the presence of 2 mM calcium. Calpain‐2 activity was calculated from the difference in activity between total calpain and calpain‐1 activity. Alternatively, membrane fractions were prepared from pons‐cerebellum from calpain‐1 knock‐out mice to measure calpain‐2 activity.

Calpain assay using human cells

We used HEK cells lacking either calpain‐1 or calpain‐2 (generous gift from Dr. Sandra Cooper, University of Sydney, Australia) to test the efficacy of NA‐112 and NA‐184 to inhibit calpain‐mediated spectrin truncation. Cells were homogenized in calpain assay buffer and incubated in the presence of various concentrations of inhibitors and in the absence or presence of calcium (2 mM). Aliquots were processed for Western Blots with spectrin antibodies (1:1000, MAB1622, EMD Millipore).

2.3. Various protease assays

Commercially available kits were purchased to test the effects of calpain inhibitors on various cysteine proteases, i.e., cathepsin B and L and caspase‐3, serine proteases, i.e., thrombin and kallikrein, and metalloproteases, i.e., ACE and MMP‐8. Assays were conducted according to the manufacturer's instructions.

2.4. CCI model of TBI

2.4.1. Animals

Animal use in all experiments followed NIH guidelines and all protocols were approved by the Institution Animal Care and Use Committee of Western University of Health Sciences. Calpain‐1 KO mice on a C57Bl/6 background were obtained from a breeding colony established from breeding pairs generously provided by Dr. Chishti (Tufts University). C57Bl/6 mice were purchased from Jackson Labs and were the corresponding wildtype (WT). Six–seven‐month‐old Sprague–Dawley rats (males and females) were also used for the CCI model of TBI.

The CCI model was established in mice following the previously described protocol. ²² Mice (3‐month‐old, 25–30 g) or rats were anesthetized using isoflurane and fixed in a stereotaxic frame with a gas anesthesia mask. A heating pad was placed beneath the body to maintain body temperature around 33–35°C. The head was positioned in the horizontal plane. The top of the skull was exposed, and a 5‐mm craniotomy was made using a micro drill lateral to the sagittal suture and centered between Bregma and Lambda. The skull at the craniotomy site was carefully removed without damaging the dura. The exposed cortex was hit by a pneumatically controlled impactor device (AMS‐201, Amscien). The impactor tip diameter was 3 mm, the impact velocity was 3 m/sec, and the depth of cortical deformation was 0.5 mm. After impact, the injured region was sutured using tissue adhesive (3 M) and mice were placed in a 37°C incubator until they recovered from anesthesia. In sham surgery, animals were sutured after craniotomy was performed. The selective calpain‐2 inhibitors were injected intraperitoneally (ip) at various doses and control animals were injected with the vehicle (5% DMSO in PBS).

2.5. Immunohistochemistry (IHC)

Twenty‐four h after TBI, mice or rats were anesthetized and perfused intracardially with freshly prepared 4% paraformaldehyde in 0.1 M phosphate buffer (PB, pH 7.4). After perfusion, brains were removed and immersed in 4% paraformaldehyde at 4°C for post‐fixation, then in 15% and 30% sucrose at 4°C for cryoprotection. Three coronal frozen sections (20 μm thick) in each brain at Bregma −0.58, −1.58, −1.94 were prepared. To determine in situ calpain activation, sections were co‐stained with rabbit anti‐spectrin breakdown product (SBDP) (1:500, a gift from Dr. Saido, Riken, Japan) antibodies. Sections were first blocked in 0.1 M PBS containing 5% goat or donkey serum and 0.3% Triton X‐100 (blocking solution) for 1 h, and then incubated with primary antibodies prepared in blocking solution overnight at 4°C. Sections were washed three times in PBS (10 min each) and incubated in Alexa Fluor 488 goat anti‐rabbit IgG, prepared in blocking solution for 2 h at room temperature. After three washes, sections were mounted with a mounting medium containing DAPI (Vector Laboratories). For the quantification of SBDP levels, three 637 × 637 μm areas near the lesion site in each section were imaged using confocal microscopes (Nikon C1 and Zeiss LSM‐880) and analyzed. In each area, the regions proximal (0–170 μm from the lesion edge) and distal (>170 μm from the lesion edge) to the impact site were separately outlined using the “freehand selection” function of ImageJ, and mean fluorescence intensity (MFI) of SBDP was measured. Data in all three sections from the same brain were averaged.

TUNEL staining. Brains were collected at 0, 6, 24 and 72 h after TBI. Coronal frozen sections (20 μm thick) at Bregma 1.54, 0.50, −0.58, −1.58, −1.94 and −2.30 were prepared. TUNEL staining was performed using the ApopTag in situ apoptosis detection kit (S7165, Millipore). Sections were visualized under confocal microscopy (Nikon). All TUNEL positive nuclei surrounding the lesion area in 6 sections of each brain were counted using the “analyze particles” function in ImageJ. To separately analyze cell death in the region proximal and distal to the impact site, the regions proximal (0–170 μm from the edge of impact site) and distal (>170 μm from the edge of impact site) to the impact site were outlined using the “freehand selection” function of ImageJ. TUNEL positive cells in these two regions were separately counted. Image acquisition and quantification were done by two examiners in a blind fashion.

2.6. Pharmacokinetics and pharmacodynamics

The pharmacokinetic study of NA‐184 was carried out in CD1 mice by intravenous injection via tail vein with a dose of 10 mg/kg in a liposomal formulation (CD1 mice were used for easier tail vein injection, as these mice are white and not black as the C57/BlL6 mice). Blood samples were collected before dosing (0 time) and at time points of 1, 5, 15, 30 min, 1, 2, 4, 8, 16, 24 h post dose. Approximately, 0.8 mL of blood were collected into heparinized tubes via cardiac puncture. Plasma was separated by centrifuging for 5 min at 3,000 g. Whole brains were collected, weighed, and mixed with 3× (by volume) PBS buffer to homogenize. All samples were kept at −80°C before further processing.

The concentration of NA‐184 was determined using an HPLC‐MS/MS method with C2I as the internal standard. Briefly, NA‐184 was extracted using a liquid–liquid extraction method. For each 100 μL of plasma samples or tissue homogenate, 10 μL of the internal standard solution (1000 ng/mL) was added. After mixing, 400 μL of methyl tert‐butyl ether was added to each sample mixture. The samples were then mixed using a vortex mixer for 3 min followed by centrifuging at 5000 rpm for 3 min. The upper clear solvent layer was transferred into a clean glass tube and evaporated under the nitrogen blow until dry. Then 0.1 mL of 80% acetonitrile was added to each tube and vortexed for 1 min. The mixture was centrifuged at 10 000 rpm for 5 min and 5 μL injected into the HPLC‐MS/MS system for analysis.

The LC/MS/MS system consisted of an API 3200 LC/MS/MS system (Sciex) and two Shimadzu LC‐20AD Prominence Liquid Chromatograph pumps equipped with an SIL‐20A Prominence autosampler (Shimadzu). Chromatography was carried out using a Zorbax SB C18 column (150 × 2.1 mm, 5 μm, Zorbax, Agilent) which was proceeded with a SB‐C18 Guard Cartridges (12.5 × 2.1 mm, Zorbax, Agilent, Santa Clara, CA, USA).

Typical mass spectrometric conditions included gas 1, nitrogen (40 psi); gas 2, nitrogen (40 psi); ion spray voltage, 5000 V; ion source temperature, 450°C; curtain gas, nitrogen (25 psi). Multiple reaction monitoring (MRM) scanning in positive ionization mode was used to monitor the transition of m/z 531.201–398.191 for NA‐184 and 528.149–241.100 for C2I. Gradient elution was carried out consisting of acetonitrile (A) and aqueous buffer (B) (0.1% formic acid containing 2 mM ammonium acetate) with a flow rate of 0.3 mL/min. The elution began at 20% of mobile phase A and maintained for 1.5 min. The ratio was increased to 90% in 0.5 min and kept for 6 min. Afterwards, the gradience was returned to 20% of A and balanced for 1.5 min. The temperatures of analytical column and autosampler were both set at room temperature.

This method showed a good linearity for concentration ranging from 2 to 500 ng/mL with average accuracy between 85% and 115%. The lower limits of quantification (LLOQ) for plasma samples and tissue samples were set as 2 and 5 ng/mL, respectively.

2.7. Nomenclature of targets and ligands

Key protein targets and ligands in this article are hyperlinked to corresponding entries in http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS Guide to PHARMACOLOGY, ³⁴ and are permanently archived in the Concise Guide to PHARMACOLOGY 2019/20. ³⁵

3. RESULTS

3.1. Molecular modeling, synthesized molecules, and the screening of novel calpain inhibitors

Using the calpain activity assay, we previously determined that the S‐S diastereoisomer of C2I is the active, as compared to the inactive R‐S isomer. Using the S‐S diastereoisomers, we computed the selectivity of known ketoamide calpain inhibitors between calpain‐1 and calpain‐2 using free energy perturbation. ²⁵ The correlation between computed and experimental selectivity provided a validation of our in silico protein models. These protein models were then used to rank aromatic fragments at P1′ position using SILCS docking protocol (see the Methods section for details). The SILCS FragMaps and docking scores provided a general structure–activity relationship at P1′ site, in which additional substitution is favored at para‐position of phenyl group in calpain‐1 P1′ site, but at meta‐position in calpain‐2 P1′ site; hydrogen bond donor and acceptor were present at ortho‐position in calpain‐2, but no density was found at ortho‐position in calpain‐1 P1′ site. ¹⁶

Based on these analyses, about 130 analogs of C2I (see Table 1 for a subset of these molecules) were synthesized, and tested for their selectivity and potency against human calpain‐1 and calpain‐2. Results from a subset of these compounds are shown in Table 2.

TABLE 1.

Various analogs of the original calpain‐2 inhibitors C2I, aka NA‐101 .

R1	R2	X	No.	MW	cLogP a	Drug‐likeness a
		O	C2I	527.62	2.42	−13.7
		O	11	623.52	2.77	−0.61
		NH	12	526.63	3.33	−1.15
		CH2	13	525.65	4.08	−2.36
		NH	8	593.09	2.99	1.22
		CH2	9	640.65	2.97	−5.47
		NH	14	514.6	3.50	−2.49
		CH2	15	513.61	4.25	−3.70
		NH	16	496.61	3.40	−1.15
		CH2	17	495.62	4.15	−2.36
		NH	18	527.62	2.33	−1.15
		CH2	19	526.63	3.08	−2.36
		NH	20	551.64	3.16	−5.43
		CH2	21	550.6	3.91	−6.63
		NH	22	513.59	2.62	−1.33
		CH2	23	512.61	2.62	−1.27
		NH	24	521.62	3.23	−5.43
		NH	25	483.57	2.69	−1.33
		NH	53/ NA‐184	531.05	2.62	2.57

^a cLogP and druglikeness were calculated using OSIRIS Datawarrior software version 5.5.0. For druglikeness, among 3300 traded drugs, 80% of the drugs have a positive druglikeness value.

TABLE 2.

K _i and IC₅₀ for calpain‐1 and calpain‐2 of some of the original calpain‐2 inhibitors C2I, aka NA‐101.

Compound	K i for human calpain‐1 (nM)	K i for rec human calpain‐2 (nM)	Ratio 1/2	IC50 for mouse calpain‐1 (nM)	IC50 for mouse calpain‐2 (nM)	Ratio 1/2
C2I	124 ± 6	55 ± 7	2.3	1130	44	26
12/NA‐112	208 ± 37	72 ± 20	2.9	1376	106	13
13	134 ± 26	56 ± 9	2.4	1176	78	14.9
8	29 ± 9	43 ± 18	0.7	322	118	2.7
9	86 ± 16	115 ± 19	.8	1788	196	9.1
11	N/A	5500 ± 1064
14	72 ± 15	72 ± 3	1.0	178	92	1.9
15	470 ± 47	103 ± 15	4.5	3582	113	31.7
16	77 ± 15	47 ± 6	1.7	1936	210	9.2
17	152 ± 31	54 ± 5	2.8	2049	85	24.1
19	61 ± 17	62 ± 7	1.0	442	214	2.1
20	252 ± 47	137 ± 43	1.6	274	114	2.4
21	146 ± 28	133 ± 25	1.1	984	97	10.1
22	221 ± 19	139 ± 30	1.4	254	340	0.7
23	63 ± 16	43 ± 11	1.5	409	498	0.8
28	312	646	0.5	772	1622	0.5
30	65 ± 15	110 ± 6	0.6	168	112	1.5
32	291	197	1.5	3039	203	15
36	613	503	1.2
39	13	86	0.2
40	1607	1712	0.9	1740	1967	0.9
41	826	1370	0.9	675	1314	0.5
53/NA‐184	309 ± 50	50 ± 2	4.9	2826	134	21

Note: Representative list of K _i of calpain inhibitors tested against purified human calpain‐1 and recombinant human calpain‐2 and their IC₅₀ against endogenous brain calpain‐1 and calpain‐2. Results are means ± SEM of three experiments or results of a single experiment. Highlighted compounds were then selected to be tested in the mouse TBI model.

In addition, the selectivity and potency of these compounds were tested against mouse calpain‐1 and calpain‐2 activity in cerebellar membrane preparations. Examples for this type of analysis are shown in Figure 1, where compounds 12 and 53 (aka NA‐112 and NA‐184, respectively) were compared. Interestingly, while the range of values for K _i ratios of calpain‐1 over calpain‐2 was not wide, the ratios for the IC₅₀ values of calpain‐1 over calpain‐2 showed a much wider range.

FIGURE 1.

Selectivity of NA‐112 and NA‐184 for calpain‐2 vs calpain‐1. (A) NA‐112 structure. (B) NA‐184 structure. (C, D) Inhibition of mouse calpain‐1 and calpain‐2 by NA‐112 (C) and NA184 (D). Membrane fractions from WT or calpain‐1 knock‐out (KO) mice were prepared. Calpain‐1 activity was measured in membrane fractions from WT mice in the presence of 20 μM calcium. Calpain‐2 activity was measured from membrane fractions from calpain‐1 KO mice in the presence of 2 mM calcium. Results are means ± S.E.M. of two experiments.

3.2. Effects of calpain‐2 inhibitors on the TBI and comparison of NA‐112 and NA‐184 in the mouse model of TBI

We focused further testing on molecules for which the ratio of the IC50 for calpain‐1 over calpain‐2 was larger than 10, as these molecules were likely to be more selective for calpain‐2 than for calpain‐1. These molecules were then tested in the CCI mouse model of TBI. Compounds were administered ip 1 h after TBI, and the animals were sacrificed 24 h later. Calpain‐1 and calpain‐2 activities were analyzed using cerebellar/pons membranes and the number of degenerating cells around the lesion site was analyzed by TUNEL assay. While compound 15 exhibited a high IC₅₀ ratio it had no effect on either brain calpain‐1 or calpain‐2 or on cell death when administered at 0.1 or 1.0 mg/kg (not shown). It also showed no effect on neuronal damage at doses up to 10 mg/kg (not shown). Our interpretation for the lack of effect of compound 15 is that this molecule does not cross the blood–brain barrier, as it showed good efficacy in the mouse model of acute glaucoma when injected intraocularly (not shown).

As both NA‐112 and NA‐184 have higher selectivity for calpain‐2 than for calpain‐1, detailed dose–response curves for NA‐112 and NA‐184 on brain calpain‐1 and calpain‐2 activity and cell death 24 h after TBI were performed in mice and results are shown in Figure 2. The EC₅₀ for calpain‐2 was 0.11 mg/kg for NA‐112 and 0.43 mg/kg for NA‐184. NA‐112 inhibition of calpain‐1 was first detected at 1 mg/kg and was close to 40% at 10 mg/kg. NA‐184, however, inhibited less than 10% of calpain‐1 activity at 10 mg/kg. The effects of the 2 compounds on cell death were consistent with their effects on calpain‐2 and calpain‐1. Thus, low doses of NA‐112 prevented cell death at doses up to 1 mg/kg, but higher doses were less effective, in agreement with the inhibition of calpain‐1, as previously reported for NA‐101. ²² In contrast, NA‐184 resulted in a dose‐dependent inhibition of TBI‐induced cell death with an EC₅₀ of 0.13 mg/kg, a maximal effect reached at 1 mg/kg and no further increases at doses up to 10 mg/kg. Based on these results, we decided to pursue the pre‐clinical development of NA‐184 for the treatment of TBI. Note that we have obtained excellent neuroprotective results with NA‐112 in a mouse model of epilepsy, ³⁶ and will therefore consider the potential preclinical development of NA‐112 for epilepsy.

FIGURE 2.

Effects of NA‐112 and NA‐184 on calpain activity and cell death after CCI in mice. WT Mice were subjected to CCI and were injected ip 1 h later with various doses of NA‐112 or NA‐184. Twenty‐four h later, mice were sacrificed, and calpain activity was measured in pons/cerebellum membrane fractions. (A, B) Calpain‐1 activity represents the activity measured in the presence of 20 μM calcium, while calpain‐2 activity represents the difference in activity measured in the presence of 2 mM calcium and 20 μM calcium. (C, D) The number of TUNEL‐positive cells was determined in tissue sections from the forebrain of the same animals. Results are means ± SEM of 4–5 animals.

3.3. Effects of NA‐184 in the CCI model of TBI in both male and female mice and rats

To further determine the effects of NA‐184 as a potential treatment for TBI, we tested its effects in the CCI model of TBI in male and female mice and in male and female rats (Figure 3). Animals were injected with NA‐184 (1 mg/kg, ip) 1 h after TBI (for rats, the NA‐184 injection was repeated after 8–10 h) and were sacrificed 24 h later. Calpain‐1 and calpain‐2 activity were assayed in membrane fractions from the cerebellum and cell death in the cortical area surrounding the trauma was analyzed with the TUNEL assay. NA‐184 administration had no effect on calpain‐1 activity (Figure 3A) but resulted in 50%–70% inhibition of calpain‐2 activity in both male and female mice and rats (Figure 3B). The treatment also resulted in a significant reduction (between 30% and 70%) of cell death in the cortical area surrounding the lesion site (Figure 3C). The levels of cell death were plotted as a function of the relative activity of calpain‐2 for all the animals (Figure 3D); there was a highly significant correlation (r ² = 0.84) between the levels of calpain‐2 activity and the level of cell death [similar to what we previously reported for NA‐101 ²² ], further supporting the conclusion that calpain‐2 activity is responsible for cell death following TBI.

FIGURE 3.

Effects of NA‐184 on calpain activity and cell death in both male and female rats and mice following TBI. WT male and female mice and rats were subjected to CCI and were injected 1 h later with 1 mg/kg NA‐184 (ip; rats were injected twice at 1 and 8 h after CCI). Twenty‐four h later, mice and rats were sacrificed, and calpain activity was measured in pons/cerebellum membrane fractions. Calpain‐1 activity represents the activity measured in the presence of 20 μM calcium (A), while calpain‐2 activity represents the difference of activity measured in the presence of 2 mM calcium and 20 μM calcium (B). The number of TUNEL‐positive cells was determined in tissue sections from the forebrain of the same animals (C). (D) Correlation between number of dead cells and calpain‐2 activity levels in all rats and mice. Results are means ± SEM of 4–5 animals. * p < .05; ** p < .01 (two‐tailed unpaired Student's t‐test).

3.4. Selectivity, stability, and epimerization of NA‐184

Several key features of NA‐184 are reported in Table 3. Recombinant human calpain‐2 was used in the initial screening of the newly synthesized inhibitors as well as an artificial substrate. The recombinant human calpain‐2 includes the full‐length human large subunit of calpain‐2 but a truncated small subunit. We thought it would be important to determine the inhibitory potency of NA‐184 against full‐length human calpain‐2 and when using an endogenous protein target. Therefore, lysates from HEK cells lacking either calpain‐1 or calpain‐2 were used to test the potency of NA‐184 to inhibit the truncation of spectrin by either calpain‐2 or calpain‐2, respectively (Figure 4A,B). Under these conditions, we found an EC₅₀ for NA‐184 of 1.6 nM (Figure 4C) and no inhibition of human calpain‐1 at concentrations up to 1 μM (Figure 4D).

TABLE 3.

Key properties of NA‐184.

Molecular weight	531 g × mol−1
cLogP3	2.62
Drug‐likeness 3	2.5692
Solubility in PBS	14 μg/mL
Total Surface Area	416
Relative PSA	0.25908
In vivo IC50 against human calpain‐2	1.60 nM
In vivo Inhibition against human calpain‐1	No inhibition up to 10 μM
Eurofin Safety Panel (44 receptors/enzymes)	CCK1: 97% at 10 μM Kappa Opiate Receptor: 57% at 10 μM μ Opiate Receptor: 58% at 10 μM

Note: cLogP and drug‐likeness were calculated using OSIRIS Datawarrior software version 5.2.1. cLogP value of compound is the calculated logP, which is the logarithm of its predicted partition coefficient between n‐octanal and water. The cLogP is calculated as the sum of contributions of every atom based on its atom type. Datawarrior uses 368 atom types and more than 5000 experimental logP values as training set to optimize the contribution values associated with the atom types. The drug‐likeness is calculated as the total scores of substructure fragments that are present in the molecule. Datawarrior uses 5300 distinct substructure fragments with associated drug‐likeness scores. A positive value states that the molecule contains predominantly fragments which are frequently present in commercial drugs.

FIGURE 4.

Effects of NA‐184 on human calpain‐1 and calpain‐2. HEK cells deleted of calpain‐1 (Calp1 KO; A, C) or calpain‐2 (Calp2 KO; B, D) were lysed in calpain assay buffer as described in the Materials and Methods section. They were incubated in the presence of 2 mM calcium and the indicated concentrations of NA‐184 for 30 min at 30°C. Aliquots of the homogenates were processed for western blots with a spectrin antibody. Intensities of the native spectrin band and the calpain‐mediated breakdown product (SBDP) were quantified and the ratio SPDP/spectrin calculated. Results were normalized to the values found in the absence of calpain‐2 inhibitors. Results are means ± S.E.M. of 2–4 experiments.

The effects of NA‐184 in the Eurofins safety panel, which contains 44 receptors/enzymes, were also evaluated. The only significant effects of NA‐184 were a 97% inhibition of the CCK1 receptor, a 57% inhibition of the K opiate receptor and a 58% inhibition of the μ opiate receptor, all at 10 μM. Inhibition of various cytochrome P450 enzymes, by 10 μM NA‐184 was also tested and resulted in 50 to 65% inhibition of CYP2C9, CYP2C19, and CYP2D6.

The effects of NA‐184 on various proteases are shown in Table 4. Only cathepsin L and B were significantly inhibited, with NA‐184 Kis of 1.9 and 220 nM, respectively. NA‐184 did not, however, exhibit significant inhibition on a variety of other cysteine‐, serine‐ or metallo‐proteases at concentrations up to 10 μM.

TABLE 4.

Selectivity of NA‐184 against other proteases.

Protease	IC50	K i
Cathepsin B	303 ± 3 nM	220 ± 3 nM
Cathepsin L	2.6 nM	1.9 nM
Caspase‐3		>21.1 μM
S20 proteasome		>10 μM
Thrombin	No inhibition up to 10 μM
Kallikrein	No inhibition up to 10 μM
ACE	No inhibition up to 10 μM
MMP‐8	No inhibition up to 10 μM

Note: Commercially available kits for the various proteases were used to determine NA‐184 inhibition against these various proteases. Results are means ± S.E.M. of 2–3 experiments.

NA‐184 is a mixture of two diastereoisomers, as it has 2 chiral centers, the S‐S and the R‐S isomers. We separated these 2 isomers and determined, using the calpain assay, that the S‐S isomer was the active isomer while the R‐S isomer was completely inactive (not shown). It was therefore important to determine whether the 2 diastereoisomers could epimerize in buffer and/or plasma. The active S‐S‐diastereoisomer was preincubated in PBS for various periods of time and the remaining inhibitory activity determined. Two hundred nM was selected as the starting concentration since it represents the K _i of NA‐184 against calpain‐2. As can be seen in Figure 5A, the inhibitory effect of the active S‐S‐NA‐184 rapidly decreased with a half‐life of about 15 min. Conversely, when we preincubated the inactive R‐S‐NA‐184 in PBS and then tested its inhibitory activity against calpain‐2, the inactive molecule became active with a half‐life of about 15 min (Figure 5B). Similar effects were found when the 2 compounds were incubated in plasma instead of PBS. Such a rapid epimerization of NA‐184 supports the clinical development of the mixture of the two diastereoisomers. These results are similar to those found with other ketoamide‐based calpain inhibitors. ³⁷ , ³⁸

FIGURE 5.

Epimerization of NA‐184 in PBS. (A) The active form of NA‐184 (S‐S‐NA‐184, 200 nM) was incubated in PBS at room temperature for the indicated periods of time. Aliquots were then added to the calpain‐2 assay buffer and calpain‐2 activity determined. (B) The inactive form of NA‐184 (R‐R‐NA‐184, 100 nM) was incubated in PBS at room temperature for the indicated periods of time. Aliquots were then added to the calpain‐2 assay buffer and calpain‐2 activity determined. Results are means ± S.E.M. of two experiments.

To test the stability in mouse plasma and liver homogenates, NA‐184 was incubated in plasma and liver homogenates and aliquots were tested after various periods of time for their ability to inhibit calpain‐2 (Figure 6A,B). The half‐life of NA‐184 in plasma was calculated to be about 3.3 h, while it was about 17.6 h in liver homogenates. These data were extended with studies performed at Charles River Laboratories, which confirmed the relatively short half‐life in plasma from different species, including humans, with the exception of the mouse plasma (Table 5).

FIGURE 6.

Stability of NA‐184 in plasma and liver homogenates. NA184 (0.4 mM) in hydroxypropyl‐ß‐cyclodextrin (400 mg/mL) was diluted 8 times in freshly prepared WT mouse plasma (A) or freshly prepared mouse liver homogenate (B) (final concentration in plasma or liver homogenate: 50 μM). The mixture was incubated at 37°C for the indicated periods of time. At the indicated time point, 1 μL of the mixture was added to 99 μL of calpain‐2 assay solution and calpain‐2 activity was determined. As a control, 1 μL of plasma alone was subjected to the calpain‐2 assay and its hydrolysis rate was set as 100%. Results are means ± SEM of three experiments.

TABLE 5.

NA‐184 stability in plasma from various species.

Species	T 1/2 (h)	% Remaining at 120 min
CD‐1 mouse	1.0	95.8
Sprague–Dawley rats	3.2	69.1
Beagle dog	3.4	67.8
Cynomolgus monkey	2.2	54.5
Human	3.7	67

Note: Analysis of NA‐184 stability of NA‐184 in plasma from various species was performed by Charles River Laboratories.

3.5. Pharmacokinetics of N‐184 in the plasma and brain

We determined the pharmacokinetics properties of NA‐184 in C57/Bl6 mice. The plasma and tissue concentrations versus time curves following iv injection of 10 mg/mL are shown in Figure 7. A rapid distribution phase was observed followed by a relatively slow elimination phase with an average elimination half‐life of 5.15 h. The typical pharmacokinetic parameters were calculated using a noncompartmental analysis (Table 6). Moreover, NA‐184 could rapidly be distributed into brain tissue as the peak concentration was observed at the initial time points (Figure 7B). Similarly, a rapid elimination phase in the brain was followed by a slow phase of elimination with a half‐life of approximately 3 h (Figure 7B).

FIGURE 7.

Pharmacokinetics of NA‐184 in plasma and brain after iv injection. NA‐184 (10 mg/mL in liposomes) was injected in the tail vein of CD1 mice. Mice were sacrificed at various time points, plasma (A) was obtained from retro‐orbital blood collection, and whole brains (B) were collected. NA‐184 was analyzed as described in the Materials and Methods section. Results are expressed in ng/ml and are means ± S.E.M. of 2 experiments.

TABLE 6.

Pharmacokinetic parameters of NA‐184 in mouse plasma following iv injection (10 mg/kg).

Parameter	Value
Elimination T 1/2 (h)	5.15
CL (Ml/h/kg)	480.8
V d (L/kg)	3.56
AUC0–∞ (ng/mL·h)	20 800

Note: Parameters for the pharmacokinetics study were obtained by fitting the data from Figure 7 to a noncompartmental model.

These pharmacokinetic data are useful for understanding the action of NA‐184 within the central nervous system, which is essential for therapeutic and safety assessments in the future.

Finally, preliminary studies for in vitro toxicity have been performed and indicated that NA‐184 was neither genotoxic nor mutagenic at concentrations up to 50 μM (not shown).

4. DISCUSSION

Starting from Compound 18, aka C2I or NA‐101, we generated about 130 analogs by modifying different aspects of the molecule, mostly on the P1′ and the P3 site. The selectivity of these molecules for calpain‐2 versus calpain‐1 depended significantly on the source of the enzymes used in the calpain assay. When using purified human calpain‐1 and recombinant human calpain‐2, the ratios of the Kis against calpain‐1 over calpain‐2 were not greatly different, although some molecules exhibited 10‐fold higher selectivity for calpain‐1 and some, 5‐fold greater selectivity for calpain‐2, thereby resulting in a 50‐fold range. The search for better selectivity for calpain‐2 than for calpain‐1 was facilitated by testing the molecules against mouse brain calpain‐1 or calpain‐2, which was further facilitated by using the calpain‐1 KO mouse. This allowed us to identify several molecules with a ratio of the IC50s for calpain‐1 over calpain‐2 greater than 10. These molecules were then tested in the in vivo mouse model of TBI and led us to the identification of 2 molecules, NA‐112 and NA‐184, which provided a very significant degree of protection against cell death at relatively low doses, between 0.1 and 1.0 mg/kg. These molecules also exhibited a relatively high degree of selectivity for calpain‐2 over calpain‐1. In particular, in vivo administration of NA‐184 at up to 10 mg/kg did not result in in vivo inhibition of calpain‐1. NA‐184 EC50 against calpain‐2 was 0.43 mg/kg and its EC50 for neuroprotection was 0.13 mg/kg; moreover, NA‐184 was equally effective in both male and female rats and mice in the CCI model of TBI. There was a very good correlation between calpain‐2 inhibition and protection against neuronal death, which is consistent with our previous reports using different calpain‐2 inhibitors ²² , ³⁹ , ⁴⁰ and supports the conclusion that calpain‐2 is the calpain isomer responsible for triggering neuronal damage following TBI. This conclusion is further supported by our study using selective calpain‐2 deletion in excitatory neurons of the forebrain, which resulted in decreased levels of cell death and lesion volume in the CCI model of TBI. ³⁹ , ⁴⁰

While the half‐life of NA‐184 in plasma was relatively short (about 3 h), it is notable that even 24 h after injection, brain calpain‐2 was still significantly inhibited up to 60%–70% following TBI. At this time point, NA‐184 levels in the brain could no longer be detected. However, it is important to stress that these molecules are likely to form a reversible covalent bond with the active site of calpain and their in vivo dissociation is probably slow enough that some molecules are still bound to the enzyme at 24 h. Moreover, it is also probably the case that following TBI, the penetration of these molecules across the blood–brain barrier is facilitated, ⁴¹ resulting in higher brain NA‐184 concentrations. Because NA‐112 appeared to inhibit calpain‐1 at concentrations above 1 mg/kg and loses its neuroprotective effects, we decided to pursue the preclinical development of NA‐184 for the treatment of TBI.

NA‐184 has several features, which make it a good clinical candidate for TBI treatment. It is very potent against human calpain‐2 when tested against spectrin degradation with an EC₅₀ of 1.6 nM and does not show inhibition of human calpain‐1 at concentration up to 1 μM. Based on our dose–response results in the mouse TBI model where maximal protection was achieved at a dose of 1 mg/kg, we anticipate that this should be the dose to be used in the clinic. This converts into a dose of 0.08 mg/kg in human, i.e., 0.15 μM. ⁴² This concentration is therefore not likely to interfere with the CCK1, kappa and mu opiate receptors, which we found in the Eurofins panel. NA‐184 could potentially interfere with Cathepsin B and L, based on the in vitro assays. At this point we do not know if NA‐184 does interfere with these enzymes in vivo, and how it would interfere with the cleavage of their endogenous targets. This issue of selectivity for calpains versus cathepsins has been previously discussed. ¹¹ In this paper, the authors discussed the concept that it is not essential to have a calpain‐specific or a cathepsin‐specific inhibitor to achieve clinical success in TBI or other neurodegenerative conditions, as both cathepsin and calpain are involved. While cathepsin S has been shown to be involved in the pathological consequences of TBI, ⁴³ we did not determine the effects of NA‐184 on this cathepsin. Importantly, the protection against mild or severe TBI provided by systemic administration of NA‐112 or NA‐184 is similar to that provided by deletion of calpain‐2 in excitatory neurons of the forebrain. ²² , ⁴⁰ These results strongly support our conclusion that NA‐112 and NA‐184 exert their neuroprotective effects by inhibiting calpain‐2 rather than cathepsins. Moreover, Cathepsin L or Cathepsin B KO mice do not exhibit any significant neurological problems. ⁴⁴ Furthermore, Abbvie recently completed a Phase I clinical trial with their calpain inhibitor, ABT‐957, aka alicapistat, which is not selective for calpain‐1 or calpain‐2, although it is more selective for calpain that for other cysteine proteases. ⁴⁵ They found that twice‐daily dosing for 2 weeks did not result in significant adverse effects in both healthy elderly and patients with mild to moderate Alzheimer's disease, ⁴⁶ suggesting that a semi‐chronic calpain‐2 inhibition might not result in adverse effects. Finally, it is worth stressing that, for the treatment of TBI patients, a relatively short‐term treatment (7 days) with NA‐184 is planned, which should minimize potential side‐effects.

In conclusion, we have identified a potent calpain‐2 inhibitor, NA‐184, which produces highly significant neuroprotection when administered after TBI in male and female mice and rats. NA‐184 has a good drug profile and we are in the process of completing the pre‐IND studies in order to initiate clinical trials for the treatment of concussion/TBI.

AUTHOR CONTRIBUTIONS

MB directed the work, obtained the funding, and wrote the manuscript. YW generated most of the experimental data. XB supervised the experimental work and wrote the manuscript. YL performed molecular dynamics simulation and modeling. ZW performed the analysis of the PK studies. ZK, AS, ES, DL, HK, and GL performed the synthesis of the inhibitors. KF, PB, SM, and GC analyzed the data and guided the experimental work.

CONFLICT OF INTEREST STATEMENT

MB, XB, YW, and YLL are cofounders of NeurAegis, Inc., a start‐up company developing calpain‐2 inhibitors for the treatment of various neurodegenerative disorders.

ETHICS STATEMENT

Authors acknowledge that the submitted work is original and has not been published elsewhere. Authors acknowledge that animal research was performed according to the NIH‐approved procedures and following reviews by the WesternU IACUC committee. Experiments were performed blind with experimenters unaware of animal treatment.

Baudry M, Wang Y, Bi X, et al. Identification and neuroprotective properties of NA‐184, a calpain‐2 inhibitor. Pharmacol Res Perspect. 2024;12:e1181. doi: 10.1002/prp2.1181

DATA AVAILABILITY STATEMENT

Data are available upon reasonable request from the corresponding author.