Protein pin arrays assessed interactions between alphaB crystallin and 12 regulatory proteins, including EGF, FGF-2, IGF-1, NGF-beta, TGF-beta, VEGF, insulin, beta-catenin, caspase-3, caspase-8, Bcl-2, and Bcl-xL, which are important in cellular differentiation, proliferation, signaling, cytoskeletal assembly, and apoptosis. FGF-2, NGF-beta, VEGF, insulin, and beta-catenin had strong interactions with human alphaB crystallin peptides, and the alphaB crystallin interactive sequences for these proteins were identified. The seven remaining proteins (EGF, IGF-1, TGF-beta, caspase-3, caspase-8, BCl-2, and Bcl-xL) did not interact with alphaB crystallin. The alphaB crystallin sequences that interacted with FGF-2, NGF-beta, VEGF, insulin, and beta-catenin overlapped with sequences that selectively interact with partially unfolded proteins, suggesting a common function for alphaB crystallin in chaperone activity and the regulation of cell growth and differentiation. Chaperone assays conducted with full-length alphaB crystallin and synthetic alphaB crystallin peptides confirmed the ability of alphaB crystallin to protect against the aggregation of FGF-2 and VEGF, suggesting that alphaB crystallin protects these proteins against unfolding and aggregation under conditions of stress. This is the first report in which sequences involved in interactions with regulatory proteins, including FGF-2, NGF-beta, VEGF, insulin, and beta-catenin, were identified in a small heat shock protein.

The functions of the interactive sequences in human alphaB crystallin that are involved in chaperone activity and complex assembly of small heat shock proteins need to be characterized to understand the mechanisms of action on unfolding and misfolding proteins. Protein pin arrays identified the hydrophobic N-terminal sequence (41STSLSPFYLRPPSFLRAP58) and the polar C-terminal sequence (155PERTIPITREE165) as interactive domains in human alphaB crystallin, which were then deleted to evaluate their importance in complex assembly and chaperone activity. Size exclusion chromatography determined that the complexes formed by the deletion mutants, Delta41-58 and Delta155-165, were larger and more polydisperse than the wild-type (wt) alphaB crystallin complex. In chaperone assays, the Delta41-58 mutant was as effective as wt alphaB crystallin in protecting partially unfolded betaL crystallin and alcohol dehydrogenase (ADH) and significantly less effective than wt alphaB crystallin in protecting unfolded citrate synthase (CS) from aggregation. Chaperone activity did not correlate with complex size but corresponded with the amount of substrate protein unfolding. The results confirmed the importance of N-terminal residues 41-58 in selective interactions with completely unfolded substrates. Poor solubility and limited or no chaperone activity for the three substrates characterized the Delta155-165 deletion mutant, which demonstrated the importance of C-terminal residues 155-165 in maintaining the solubility of unfolded substrates in a manner independent of the amount of substrate protein unfolding. The results presented in this report established that interactive domains in the N- and C-termini of human alphaB crystallin are important for the recognition, selection, and solubility of unfolding substrate proteins.

The 1.5-benzodiazepine (clobazam), the 1,4-benzodiazepine (diazepam), and two nonbenzodiazepine antiepileptic drugs (phenobarbital and valproate) were evaluated in mice and rats with a battery of well-standardized anticonvulsant test procedures. The results obtained indicate that clobazam and valproate exhibit a wider range of experimental anticonvulsant activity than either diazepam or phenobarbital. Except for clobazam by the maximal electroshock seizure (MES) test in rats, clobazam and valproate are effective in nontoxic doses against MES and all four chemically induced seizures (Metrazol, bicuculline, picrotoxin, and strychnine). Clobazam is effective by the MES test in rats only in doses that exceed the median minimal toxic dose. Phenobarbital is effective against all of the above tests, but minimal toxic doses must be employed to prevent strychnine seizures. Diazepam, on the other hand, is effective in nontoxic doses against seizures induced by Metrazol, bicuculline, and picrotoxin, but protects animals from maximal electroshock and strychnine seizures only when given in toxic doses. When compared on the basis of protective indices (PI = TD50/ED50) calculated from intraperitoneal data, the PIs for clobazam were 1.6 to 13 times higher than those for diazepam. Overall, except for the MES test in rats, the PIs for clobazam were from 1.5 to 44 times higher than those for any of the other three substances. With respect to the MES test in rats, the PI for clobazam was 10.8 times higher than that for diazepam; however, the PIs for phenobarbital and valproate were 3.5 and 4.4 times higher, respectively, than that for clobazam. These data suggest that the spectrum of anticonvulsant activity for the 1,5-benzodiazepine (clobazam) is superior to that for the 1,4-benzodiazepine (diazepam). Also, the broad experimental profile of anticonvulsant activity of clobazam agrees well with its reported broad clinical efficacy.