Biochemical Studies Of Neurons And Other Cell Types

Child Health And Human Development

Investigators

Linked publications & trials

Abstract

Vasoactive intestinal peptide (VIP) has neurotrophic and growth-promoting actions that are mediated through glia-derived substances. A major goal of this section is to identify these mediators of VIP-regulated development and study their mechanism of action. Previous studies have indicated that one of the protective proteins released by VIP is a femtomolar-acting protein: activity dependent neurotrophic factor (ADNF). A continuing priority of this section is to characterize ADNF at the protein and genetic level and to develop therapeutic agents based on peptides derived from ADNF. The investigation of ADNF has been particularly challenging due to its low abundance, pH-sensitive instability, high aggregational properties, poor aqueous solubility and molecular complexity. Previously, small amounts of the ADNF were isolated biochemically with the use of four chromatographic procedures. This methodology, although essential in defining the ADNF concept, was time consuming and resulted in significant loss of product during isolation. Based on antiserum made against ADNF-derived peptides, an affinity column has been developed during the past year that is capable of isolating purified ADNF-like protein in a single step from conditioned medium obtained from VIP-stimulated astrocyte cultures. This alternative procedure has significantly increased the availability of ADNF protein for structural and pharmacological studies. Although initially thought to be a single 14 kDa protein, further structural analysis has indicated that ADNF exists as a complex of at least three components. Additional information on the ADNF complex has been obtained with MALDI (Matrix Assisted Laser Desorption/Ionization) and SELDI (Surface Enhanced Laser Desorption/Ionization) mass spectrometric/time of flight analyses. These techniques determined that the molecular weight of the ADNF complex was 14,590 Daltons. Furthermore, both the MALDI and SELDI spectra of ADNF exhibited an unusually broad peak molecular range in excess of 1000 Daltons, an observation consistent with the conclusion that ADNF is an array of proteins or glycoproteins that are stably associated to form a functional unit. Further support for this molecular model is the separation of three component peaks of ADNF by N-CHO capillary electrophoresis. The three peaks isolated from the N-CHO column exhibit unique monotonic dose responses differing widely in potency. The biological activity of these peaks is inhibited by a distinct group of the antisera generated against ADNF peptides. Based on its relative stability, component 2 of the ADNF complex was chosen for additional MALDI studies which demonstrated a molecular mass of 13,757 Daltons. Thus, the breadth of the MADLI spectrum for ADNF may be attributed to the existence of multiple components that are similar in size and tightly associated. These studies form the basis of our working model of ADNF and suggest the possibility that multiple gene products or post-transnational modifications may explain its structure. The nature of ADNF was further investigated with a continued search for biologically active peptide fragments of ADNF components and with an effort to obtain more antibodies directed against tryptic peptides. These studies revealed that of the nine tryptic digest peptides discovered, four were found to exhibit neuroprotection from apoptotic death produced in cerebral cortical cultures treated with tetrodotoxin. As measured by both neuronal cell counts and carboxy fluorescein diacetate, potent and efficacious neuroprotection were observed after treatment with these four peptides. The size of the peptides ranged from 10-30 amino acids. After searches in available data bases, all nine of these peptides appear unique with no apparent homology to any known protein. In addition, back translation of these peptide sequences did not reveal significant identities or similarities in nucleotide sequences within any database. Three new antibodies were generated this year and investigated for their possible interaction with ADNF. When used to treat cerebral cortical cultures, all three antibodies decreased neuronal survival by 20-35% in comparison to controls. The pre-immune sera had no effect on neuronal survival. In addition, all three new antisera blocked the survival-promoting activity of the biochemically purified ADNF, supporting the hypothesis that antisera interact with important components of the ADNF complex. These new antisera will be used in further attempts to expression clone the components of the ADNF complex. One anti-ADNF peptide was used to assess ADNF immunoreactivity in sections of the neonatal brain. These studies revealed that the most dense localization of ADNF-like protein was the brain stem, particularly in cells of the trigeminal ganglion, reticular formation and parabrachial/facial nuclei. In fibers, the greatest abundance of ADNF-like protein was found in the hindbrain, with significant amounts localized in the caudate putamen, olfactory tubercle, and to a lesser extent in the cerebral cortex and hippocampus. Control studies conducted under the same conditions with pre-immune serum did not reveal immunoreactivity. Previous studies have shown that short peptides derived from ADNF and a pharmacologically related protein (activity dependent neuroprotective protein, ADNP) exhibit neuroprotection at femtomolar concentrations. The protective action of these peptides is observed in cultures treated with clinically relevant toxins including glutamate, beta amyloid peptide, iron, hydrogen peroxide and gp120, the HIV envelope protein. The pharmacology of ADNF-9 (agonist to ADNF) and NAP (agonist to ADNP) has been further explored. Surprisingly, ADNF-9 and NAP were found to be fully active in the D-isomer form as well as in the L-isomer form. This observation was initially made in studies of apoptotic death produced by tetrodotoxin in cerebral cortical cultures. Importantly, treatment with the combination of D-ADNF-9 and D-NAP significantly increased the range of concentrations at which the peptides exhibited effective neuroprotection. The studies with D-isomer peptides were extended to the prevention of embryonic death in a model of fetal alcohol syndrome. These studies indicated that both intraperitoneally and orally administered D-ADNF-9 was efficacious in preventing fetal death produced by prenatal alcohol treatment. These studies indicate that the peptides from these glial proteins exhibit protective properties through non-chiral mechanisms. Furthermore, the discovery of orally active forms of these peptides increases their potential as lead compounds for drug development. The growth-promoting action of VIP is evident in the early postimplantation period of embryogenesis. VIP receptors are localized to the neural tube at this stage of development. Explanted neural tubes have been used as a model system to study VIP-mediated regulation of gene expression and neurotrophin release. The GEArray system was utilized to monitor signal transduction pathways and cytokines in control and VIP-stimulated neural tubes. The following cDNA's were up regulated after VIP treatment: GRO1, BAX, p21 and p57. A number of cytokines were down regulated after VIP treatment: GCSF, IL-10, TGF-B1, TNF-alpha, TNF-beta and p53. This constellation of changes mediated by VIP demonstrate the complexity of cytokine and pathway genes regulated by VIP at this critical period of development. PACAP-38, a peptide that belongs to the VIP family of peptides, was found to inhibit the growth of embryonic day 9.5 mouse embryos in culture. Thus the stimulatory action VIP on growth is not a general feature of this group of peptides.