Drug discovery opportunities where loss-of-function alleles of a target gene link to a disease-relevant phenotype often require an agonism approach to up-regulate or re-establish the activity of the target gene. Antibody therapy is increasingly recognized as a favored drug modality due to multiple desirable pharmacological properties. However, agonistic antibodies that enhance the activities of the target enzymes are rarely developed because the discovery of agonistic antibodies remains elusive. Here we report an innovative scheme of discovery and characterization of human antibodies capable of binding to and agonizing a circulating enzyme lecithin cholesterol acyltransferase (LCAT). Utilizing a modified human LCAT protein with enhanced enzymatic activity as an immunogen, we generated fully human monoclonal antibodies using the XenoMouse(TM) platform. One of the resultant agonistic antibodies, 27C3, binds to and substantially enhances the activity of LCAT from humans and cynomolgus macaques. X-ray crystallographic analysis of the 2.45 A LCAT-27C3 complex shows that 27C3 binding does not induce notable structural changes in LCAT. A single administration of 27C3 to cynomolgus monkeys led to a rapid increase of plasma LCAT enzymatic activity and a 35% increase of the high density lipoprotein cholesterol that was observed up to 32 days after 27C3 administration. Thus, this novel scheme of immunization in conjunction with high throughput screening may represent an effective strategy for discovering agonistic antibodies against other enzyme targets. 27C3 and other agonistic human anti-human LCAT monoclonal antibodies described herein hold potential for therapeutic development for the treatment of dyslipidemia and cardiovascular disease.
LCAT is intimately involved in HDL maturation and is a key component of the reverse cholesterol transport (RCT) pathway which removes excess cholesterol molecules from the peripheral tissues to the liver for excretion. Patients with loss-of-function LCAT mutations exhibit low levels of HDL cholesterol and corneal opacity. Here we report the 2.65 A crystal structure of the human LCAT protein. Crystallization required enzymatic removal of N-linked glycans and complex formation with a Fab fragment from a tool antibody. The crystal structure reveals that LCAT has an alpha/beta hydrolase core with two additional subdomains that play important roles in LCAT function. Subdomain 1 contains the region of LCAT shown to be required for interfacial activation, while subdomain 2 contains the lid and amino acids that shape the substrate binding pocket. Mapping the naturally occurring mutations onto the structure provides insight into how they may affect LCAT enzymatic activity.
Title: Genetic Predisposition for Variable Response to Anticholinesterase Therapy Anticipated in Carriers of the Butyrylcholinesterase Atypical Mutation Loewenstein Y, Schwarz M, Glick D, Norgaard-Pedersen B, Zakut H, Soreq H Ref: In Enzyme of the Cholinesterase Family - Proceedings of Fifth International Meeting on Cholinesterases, (Quinn, D.M., Balasubramanian, A.S., Doctor, B.P., Taylor, P., Eds) Plenum Publishing Corp.:471, 1995 : PubMed
Normal butyrylcholinesterase (BuChE), but not several of its common genetic variants, serves as a scavenger for certain anti-cholinesterases (anti-ChEs). Consideration of this phenomenon becomes urgent in view of the large-scale prophylactic use of the anti-ChE, pyridostigmine, during the 1991 Persian Gulf War, in anticipation of nerve gas attack and of the anti-ChE, tacrine, for improving residual cholinergic neurotransmission in Alzheimer's disease patients. Adverse symptoms were reported for subjects in both groups, but have not been attributed to specific causes. Here, we report on an Israeli soldier, homozygous for 'atypical' BuChE, who suffered severe symptoms following pyridostigmine prophylaxis during the Persian Gulf War. His serum BuChE and recombinant 'atypical' BuChE were far less sensitive than normal BuChE to inhibition by pyridostigmine and several other carbamate anti-ChEs. Moreover, atypical BuChE demonstrated 1/200th the affinity for tacrine of normal BuChE or the related enzyme acetylcholinesterase (AChE). Genetic differences among BuChE variants may thus explain at least some of the adverse responses to anti-ChE therapies.
To explore the molecular basis of the biochemical differences among acetylcholinesterase (AChE), butyrylcholinesterase (BCHE) and their alternative splicing and allelic variants, we investigated the acylation phase of cholinesterase catalysis, using phosphorylation as an analogous reaction. Rate constants for organophosphate (DFP) inactivation, as well as for oxime (PAM)-promoted reactivation, were calculated for antibody-immobilized human cholinesterases produced in Xenopus oocytes from natural and site-directed variants of the corresponding DNA constructs. BCHE displayed inactivation and reactivation rates 200- and 25-fold higher than either product of 3'-variable AChE DNAs, consistent with a putative in vivo function for BCHE as a detoxifier that protects AChE from inactivation. Chimeric substitution of active site gorge-lining residues in BCHE with the more anionic and aromatic residues of AChE, reduced inactivation 60-fold but reactivation only 4-fold, and the rate-limiting step of its catalysis appeared to be deacylation. In contrast, a positive charge at the acyl-binding site of BCHE decreased inactivation 8-fold and reactivation 30-fold. Finally, substitution of Asp70 by glycine, as in the natural 'atypical' BCHE variant, did not change the inactivation rate yet reduced reactivation 4-fold. Thus, a combination of electrostatic active site charges with aromatic residue differences at the gorge lining can explain the biochemical distinction between AChE and BCHE. Also, gorge-lining residues, including Asp70, appear to affect the deacylation step of catalysis by BCHE. Individuals carrying the 'atypical' BCHE allele may hence be unresponsive to oxime reactivation therapy following organophosphate poisoning.
Title: Engineering of human cholinesterases explains and predicts diverse consequences of administration of various drugs and poisons. Schwarz M, Glick D, Loewenstein Y, Soreq H Ref: Pharmacol Ther, 67:283, 1995 : PubMed
Title: Site-directed mutagenesis of active site residues reveals plasticity of human butyrylcholinesterase in substrate and inhibitor interactions Gnatt A, Loewenstein Y, Yaron A, Schwarz M, Soreq H Ref: Journal of Neurochemistry, 62:749, 1994 : PubMed
In search of the molecular mechanisms underlying the broad substrate and inhibitor specificities of butyrylcholinesterase (BCHE), we employed site-directed mutagenesis to modify the catalytic triad residue Ser198, the acyl pocket Leu286 and adjacent Phe329 residues, and Met437 and Tyr440 located near the choline binding site. Mutant proteins were produced in microinjected Xenopus oocytes, and Km values towards butyrylthiocholine and IC50 values for the organophosphates diisopropylfluorophosphonate (DFP), diethoxyphosphinylthiocholine iodide (echothiophate), and tetraisopropylpyrophosphoramide (iso-OMPA) were determined. Substitution of Ser198 by cysteine and Met437 by aspartate nearly abolished activity, and other mutations of Ser198 completely abolished it. Tyr440 and Leu286 mutants remained active, but with higher Km and IC50 values. Rates of inhibition by DFP were roughly parallel to IC50 values for several Leu286 mutants. Both Km and IC50 values increased for Leu286 mutants in the order Asp < Gln < Lys. In contrast, cysteine, leucine, and glutamine mutants of Phe329 displayed unmodified Km values toward butyrylthiocholine, but up to 10-fold decreased IC50 values for DFP, iso-OMPA, and echothiophate. These findings add Tyr440 and Phe329 to the list of residues interacting with substrate and ligands, demonstrate plasticity in the active site region of BCHE, and foreshadow the design of recombinant BCHEs with tailored scavenging properties.
The acetylcholine hydrolysing cholinesterases control the termination of cholinergic signalling in multiple tissues and are targets for a variety of drugs, natural and man-made poisons and common insecticides. Molecular cloning and gene mapping studies revealed the primary structure of human acetyl- and butyrylcholinesterase and localized the corresponding ACHE and BCHE genes to the chromosomal positions 3q26-ter and 7q22, respectively. Several different point mutations in the coding region of BCHE were found to be particularly abundant in the Israeli population. Analytical expression studies in microinjected Xenopus oocytes have demonstrated that the biochemical properties of cholinesterases may be modified by rationalized site-directed mutagenesis and in chimeric ACHE/BCHE constructs. These properties are differently altered in the various allelic BCHE variants, conferring resistance to several anti-cholinesterases, which may explain the evolutionary emergence of these multiple alleles. At the clinical level, abnormal expression of both ACHE and BCHE and the in vivo amplification of the ACHE and BCHE genes has been variously associated with abnormal megakaryocytopoiesis, leukemias and brain and ovarian tumors. Moreover, antisense oligonucleotides blocking the expression of these genes were shown to interfere with hemocytopoiesis in culture, implicating these genes in cholinergic influence on cell growth and proliferation.
Bicuculline methiodide (BMI), a gamma-aminobutyrate (GABA) antagonist, is a powerful convulsant agent when injected into the cerebral ventricles, amygdala, hippocampus, thalamus, neocortex, and deep prepiriform cortex in rats. In contrast, bilateral microinjection of BMI into the rat striatum confers protection against seizures induced by the cholinergic agonist pilocarpine (380 mg/kg, i.p.), with an ED50 of 94 fmol (range 45-195 fmol). No topographical variation in the anticonvulsant action of BMI was detected throughout rostrocaudal and dorsoventral aspects of the striatum. The anticonvulsant action of BMI in the striatum was reversed by coadministration of the GABA agonist muscimol or by blocking GABA-mediated inhibition in either the substantia nigra pars reticulata or in the entopeduncular nucleus. The results show that blockade of GABA-mediated inhibition in the striatum has a powerful anticonvulsant effect in the pilocarpine model, suggesting that GABAergic transmission in the striatum modulates the seizure propagation in the forebrain.
Increasing doses of pilocarpine, 100-400 mg/kg, were given intraperitoneally to mice and the resulting behavioral, electroencephalographic and neuropathological alterations were studied. No behavioral phenomena were observed in mice treated with the lowest dose of pilocarpine. Occasional tremor and myoclonus of hindlimbs were found in animals which received pilocarpine in a dose of 200 mg/kg. At doses of 300, 325 and 350 mg/kg, pilocarpine produced a sequence of behavioral alterations including staring spells, limbic gustatory automatisms and motor limbic seizures that developed over 15-30 min and built up progressively into a limbic status epilepticus lasting for several hours. The highest dose of pilocarpine, 400 mg/kg, was generally lethal to mice. Pilocarpine produced both interictal and ictal epileptiform activity in the electroencephalogram (EEG). The earliest EEG alterations appeared in the hippocampus and then spread to cortical areas. EEG seizures started 10-15 min after injection of large doses of pilocarpine, 300-350 mg/kg. Ictal periods lasted for 1-2 min, recurred every 5-10 min and were followed by periods of depression of the EEG activity. By 30-45 min paroxysmal activity resulted in a status epilepticus. Examination of frontal forebrain sections with light microscopy revealed a widespread damage to several brain regions including the hippocampus, amygdala, thalamus, olfactory cortex, neocortex and substantia nigra. Scopolamine, 10 mg/kg, and diazepam, 10 mg/kg, prevented the development of convulsive activity and brain damage produced by pilocarpine. The results emphasize that excessive and sustained stimulation of cholinergic receptors can lead to seizures and seizure-related brain damage in mice. It is proposed that systemic pilocarpine in mice provides a useful animal model for studying mechanisms of and therapeutic approaches to temporal lobe epilepsy.
Behavioural, electroencephalographic and neuropathological responses to increasing doses of pilocarpine (100-400 mg/kg) administered intraperitoneally to rats were studied. At the dose of 400 mg/kg pilocarpine produced a sequence of behavioural alterations including staring spells, olfactory and gustatory automatisms and motor limbic seizures that developed over 1-2 h and built up progressively into limbic status epilepticus. Smaller doses showed different threshold for these behavioural phenomena but a similar time course of development. The earliest electrographic alterations occurred in the hippocampus and then epileptiform activity propagated to amygdala and cortex. Subsequently electrographic seizures appeared in both limbic and cortical leads. The ictal periods recurred each 5-15 min and were followed by variable periods of depression of the electrographic activity. The sequence of electrographic changes correlated well with the development of behavioural phenomena. Histological examination of frontal forebrain sections revealed disseminated, apparently seizure-mediated pattern of brain damage. Neuropathological alterations were observed in the olfactory cortex, amygdaloid complex, thalamus, neocortex, hippocampal formation and substantia nigra. Pretreatment of animals with scopolamine (20 mg/kg) and diazepam (10 mg/kg) prevented the development of convulsive activity and brain damage. These results show that systemic pilocarpine in rats selectively elaborates epileptiform activity in the limbic structures accompanied by motor limbic seizures, limbic status epilepticus and widespread brain damage. It is suggested that a causative relationship between excessive stimulation of cholinergic receptors in the brain and epileptic brain damage may exist.
