IN PRESS: The Open Enzyme Inhibition Journal Uncompetitive Phospholipase A2 Inhibition by CHEC Sequences including Oral Treatment of Experimental Autoimmune Myeloencephalitis

Catalytic sites of metal dependent phosphatases and phospholipases are characterized by a catalytic histidine (H) and adjacent acidic metal binding residue E (or D). For the secreted phospholipases A2 (sPLA2s), the active sites are stabilized by nearby flanking cystines. CHEC-9 (CHEASAAQC), which displays this organization, is a sPLA2 inhibitor that is an effective treatment for models of traumatic and autoimmune neurodegenerative disorders. We screened modifications of this sequence and found that elimination of the two central alanines followed by cyclization results in another potent sPLA2 inhibitor, CHEC-7 (CHEASQC). CHEC-7, like CHEC-9, had uncompetitive kinetic properties making it ideal for in vivo applications. Subcutaneous injections of CHEC-7 into restrained rats inhibited the transient rise in plasma sPLA2 activity suggesting that this peptide was also effective in vivo. Oral delivery of the peptide attenuated the LPS-induced increases in sPLA2 activity. Thin layer chromatography and mass spectroscopy of representative LPS-treated rats suggested that the peptide also reduced levels of lysophosphatidylcholine, a hydrolytic product of the enzyme reaction. In order to test the efficacy of CHEC-7 for neuroimmune therapies, rats were immunized with spinal cord homogenates to induce experimental autoimmune encephalomyelitis (EAE), followed by daily treatment with oral (1.5 mg/kg) or subcutaneous (0.1, 0.5, and1.5mg/kg) CHEC-7 and CHEC-9 (n=10 for each different condition). The rats were scored for severity of disease applying a standard EAE scale. Control rats all developed disease over the 23-day course of the experiments. Both CHEC-9 and CHEC-7 produced highly significant reductions in disease severity including via oral delivery. The performance of oral CHEC-7 was notable in that this treatment completely prevented disease in half the rats. The results suggested that continued study of modifications of CHEC sequences would provide stable therapeutic peptides for a variety of inflammatory and autoimmune disorders. INTRODUCTION The secreted phospholipase A2 enzymes (sPLA2s) represent unique therapeutic targets because enzyme activity is associated directly with both inflammatory and cell death mechanisms. Since the enzymes are present and active in the circulation, it is not surprising that there has been a significant effort aimed at designing sPLA2 inhibitors for systemic treatment of a number of disorders [1-7]. Several of the effective inhibitors identified so far are of the competitive type containing the indole analogue MeIndoxam, and are targeted toward specific sPLA2 isoforms [1]. The inhibitors may also have nonenzymatic anti-inflammatory effects by blockade of the sPLA2 M-receptor [2]. However, the performance of these molecules in disease models has been variable and clinical success very limited [4-7]. We identified serendipitously a nine amino acid internal fragment of cell survival/anti-inflammatory protein DSEP, and found that this peptide was an inhibitor of systemic sPLA2 activity[3, 4]. The peptide, called CHEC-9, is broad spectrum because it inhibits the diverse enzyme activities in the plasma of rats and humans, dominated by groups II and X sPLA2s, as well as purified enzymes in groups I and III. Kinetic studies showed that the peptide is unique among sPLA2 inhibitors, operating with uncompetitive properties. It is dependent therefore on the levels of substrate and on enzyme activity. In theory, such inhibitors should be well suited to in vivo models especially those with an inflammatory component because there is the potential for cascading effects in which active sPLA2 enzymes and their substrates accumulate [11-13]. In fact, systemic treatment with CHEC-9 is effective in reducing sPLA2 and cell-mediated inflammation, neuron death, and behavioral deficits when applied to in vivo models, including trauma and autoimmune disease (experimental autoimmune encephalomyelitis, EAE). The sequence of CHEC-9 (CHEASAAQC) is interesting because it contains residues that are typical of enzyme catalytic sites in general [5]. The HE (or D) motif is found consistently in active sites of phosphatases and phospholipases and represents a critical participant in catalysis [15, 16]. Also like CHEC-9, catalytic regions of the secreted isoforms of the PLA2s have flanking cystines presumably for stabilization of the active site. It is therefore not surprising that CHEC-9 was identified as an sPLA2 inhibitor given that peptide-based inhibitors are often derived from key motifs in the target molecules (e.g. ref [6]). In addition, the fact that the peptide is active systemically after subcutaneous application suggested that this molecule must have other properties that optimize its stability and unencumbered transport to its targets in vivo. In testing modifications of the CHEC-9 sequence, we sought initially to determine the biological activity of a simple economical modification of CHEC-9 by removing the two central alanines to give CHEASQC (CHEC-7). This modification preserved the basic organization (the HE motif, the flanking cross linked cystines) but changed the hydrophilicity, and effective radius of the cyclic peptide. Since there is considerable current interest in oral formulations of disease modifying drugs for MS [7], we also investigated and compared the feasibility and efficacy of oral delivery of the CHECs, comparing both the original and modified versions of the peptide. We found that CHEC-7 was a potent sPLA2 inhibitor, also uncompetitive, and that both CHEC-7 and CHEC-9 were active after oral delivery. MATERIALS AND METHODS Subjects Female rats of both the Sprague-Dawley (enzyme studies) and Lewis (EAE studies) strains were obtained from Harlan for use in these experiments. Pooled human blood samples were obtained from Bioreclamation, Inc. All specific procedures of this study were approved by both the Institutional Animal Care and Use Committee and by Institutional Review Board of Drexel University. Enzyme assays The assays were conducted at ambient temperature (22–25°) using a Victor 3 fluorescent reader (Perkin Elmer). The sources of secreted phospholipase A2 were either purified enzymes, group Ia (Naja mossambica venom) or group Ib (porcine pancreatic), both obtained from Sigma, or the plasma of rats and humans. For the rat experiments, blood was either obtained from the trunk after decapitation, or for the timed studies, via tail nick after placing the rats in a standard restrainer. Blood samples were treated with citrate-phosphate-dextrose anticoagulant (1:10, Sigma), and plasma prepared by centrifugation, before freezing at -80° until used in the assays. Substrates included 1,2-Bis(1-pyrenebutanoyl)-snglycero-3-phosphocholine (Invitrogen) or 1-Palmitoyl-2-Pyrenedecanoyl Phosphatidylcholine (Caymen Chemical) substrates for all calcium dependent PLA2s with the exception of cPLA2 and PAF-AH. These substrates form phospholipids vesicles in aqueous solutions [13] and produce fluorescent pyrenyldecanoic acid upon hydrolysis, which was measured at 350 nm excitation, 405 nm emission. Another fluorogenic PLA2 substrate, N((6-(2,4-dinitrophenyl) amino) hexanoyl)-2-(4,4-difluoro-5,7dimethyl-4-bora-3a,4a-diaza-s-indacene-3-pentanoyl)-1-hexadecanoyl-sn-glycero-3phosphoethanolamine, triethylammonium salt (PED6, Invitrogen), was also used. The substrates were dissolved in ethanol, and stored at -20° prior to use. Mixed substrates were also prepared in which the pyrene-PCs were mixed in the organic phase with equimolar phosphatidic acid (Avanti Polar Lipids) and dispersed in reaction buffer (see below). Mixed liposomes were prepared using Lipofast (Avestin, Inc.), monitored in a fluorescent microscope, and further dispersed until uniform by freeze-thaw cycles. All substrate solutions were prepared in reaction buffer consisting of 50 mM tris (pH = 7.4), 0.1 M NaCl, 2 mM CaCl2, 0.25% fatty acid-free albumin (Sigma) just prior to use. Kinetic parameters including the properties of CHEC peptides were determined by measuring the initial velocities (Vo) of enzyme reactions (within 2 minutes of initiation). For experiments in which active sPLA2 enzyme concentration was measured in plasma samples from treated rats, we used a single substrate concentration and measured the steady-state rate of the enzyme reaction for 60 minutes. This rate is proportional to the concentration of active enzyme in the plasma if product formation during this period is linear with respect to time [4]. Enzyme activities were expressed relative to baseline or vehicle treatment, or for kinetic studies, fluorescent units (RFU) were converted to product concentration using a pyrenyldecanoic acid (Invitrogen) standard curve. All points on these graphs presented in results were means from 2-3 reactions in individual experiments which were repeated 3 or more times with the same result, i.e., the shapes of the product curves and the direction of change of Km and Vmax was the same following inhibitor treatment. Km and Vmax and r2 were determined with regression software (Prism) and statistical calculations with INSTAT both from Graphpad (San Diego, CA). 2D Thin layer chromatography and Mass spectroscopy These procedures are described in detail in a previous study [8]. Briefly, lipids were extracted from plasma samples using the Folch method and total phospholipids (PL) were determined. Samples were mixed with a fixed amount of N-Acetyl-L-erythro-sphingosine as an internal loading control, dried under nitrogen, and re-suspended in chloroform before spotting on Merck 5631 TLC plates. The plates were developed in 3 directions with drying between each migration [9]. Solvents for the first and second migration were 1) 60:30:8:2, chloroform: methanol: formic acid: H20, and 2) 50:40:7:3, chloroform: methanol: 28% ammonium hydroxide, respectively. For the third migration the plates were developed in diethyl ether. After development, the plates were immersed in 10% CuSO3 and 8% H3PO4 solution for 20 seconds, and charred at 180°C for 20 min. For mass spectroscopy, samples from the Folch-extracted lipid were dried and re-dissolved in chloroform-methanol-300 mM ammonium acetate in water (60:133:7, v/v/v). Lysophosphatidylcholines were analyzed by electrospray ionization triple quadrupole mass spectrometry at the Kansas Lipidomics Research Center, as previously described [10]. Peptide preparation and delivery CHEC-7 (CHEASQC) and CHEC-9 (CHEASAAQC) were synthesized as linear peptides by Celtek, Inc. The peptides were cross-linked in tris-buffered saline (pH 7.5) at 0.25mg/ml overnight. Free sulfhydryls were measured using Ellman’s reagent and were undetectable after incubation. Crosslinking was verified by electrospray mass spectrometry as in previous studies[22]. Peptide concentrations were adjusted for delivery of stated dosages either subcutaneously under the loose skin of the back or p.o. by gavage. Control animals received an equivalent volume of vehicle. Lipopolysaccharide (LPS) treatment and EAE immunization LPS (4mg/kg) was administered subcutaneously under the skin of the upper thigh. For the EAE studies, Mycobacterium tuberculosis H37Ra (MT, Difco) was crushed with pestle and mortar and used to enrich Complete Freund's Adjuvant (CFA, Sigma) at a final concentration of 10 mg/ml. Lyophilized Guinea Pig Spinal Cord Homogenate (Sigma) was weighed and suspended in PBS to yield 40 mg/mL. The homogenate was mixed with equal amount of CFA/MT and emulsified prior to injection into the right hind paw of each rat at a volume of 0.1 ml. Evaluation of the EAE clinical signs We used a standard EAE scoring paradigm: 0Normal behavior, no clinical signs; 1-tail weakness, tail was limp and droops; 2Hind-legs were hypotonic and weak; paresis-wobbly walk, hind legs unsteady or one hind-leg was dragged; 3hind-legs paralyzed, front legs normal; 4 hind-legs paralyzed, front legs weak; 5full paralysis. Experimentally blinded observers, experienced in EAE studies, scored the rats. Statistical analysis Nonparametric statistical tests including Mann-Whitney test and Kruskal-Wallis ANOVA were used throughout the study as indicated. RESULTS Inhibition of plasma sPLA2 activity CHEC-7 was screened in vitro using concentrations between 1 and 100nM applied to reactions of groups Ia, Ib (both at 20nM), and of rat (10%) and human (40%) plasma all with the pyrene substrates. CHEC-9 was used as a positive control in most experiments. CHEC-7 inhibited all these sources of sPLA2, most effectively at peptide concentrations under 25nM and substrate concentrations above 30μM. CHEC-7 inhibition is illustrated in product curves (side by side with CHEC-9), and in a representative Michaelis-Menten plot using human plasma and a range of substrate concentrations (Fig. 1 A,B). Both sets of data were consistent with properties of an uncompetitive inhibitor. For the product curves (Fig. 1A), the small delay initially in the inhibition by the peptides allows for the formation of the enzyme-substrate (ES) complex, which is targeted by uncompetitive inhibitors. Thereafter, the decline of activity persisted for the entire experiment, apparently because of stable binding by the inhibitor to this complex [4]. Furthermore, the data suggested that both CHEC-7 and CHEC-9 have similar potencies ex vivo, both operating at the low nanomolar range for human plasma. CHEC-7 was also effective at nanomolar concentrations for the other sPLA2s tested (data not shown).