Impaired processing of brain proneurotensin and promelanin-concentrating hormone in obese fat/fat mice.

Mice homozygous for the fat mutation exhibit marked hyperpro-insulinemia and develop late onset obesity. The fat mutation was recently mapped to the gene encoding carboxypeptidase E (CpE), a processing enzyme involved in trimming C-terminal paired basic residues from prohormone-derived peptides. The mutation resulted in a loss of CpE activity that correlated with aberrant proinsulin processing. Neurotensin (NT) and melanin-concentrating hormone (MCH) are two neuropeptides that, among other central effects, inhibit food intake. Here, using RIA techniques coupled to reverse phase HPLC, we analyzed the processing products derived from the NT and MCH precursors in the brain of +/fat and fat/fat mice. Compared to control hypothalamic and brain extracts, fat/fat extracts had markedly reduced levels (>80%) of NT and neuromedin N (NN), another active pro-NT-derived peptide. In contrast, they exhibited high concentrations of biologically inactive NT-KR and NN-KR (NT and NN with a C-terminal Lys-Arg extension), two peptides that were undetectable in control extracts. MCH, which is located at the C-terminus of its precursor, was present in 2- to 3-fold higher amounts in fat/fat than in +/fat hypothalamus. Neuropeptide-Glu-Ile, another pro-MCH-derived neuropeptide separated from MCH by an Arg-Arg sequence, was present in amounts similar to those of MCH in control extracts. In contrast, neuropeptide-Glu-Ile was more than 10 times less abundant than MCH in extracts from obese mice. Our data are consistent with a deficit in CpE activity affecting the maturation of both pro-NT and pro-MCH. This suggests that abnormal neuropeptide and hormone precursor processing is a general phenomenon in fat/fat mice and supports the idea that defects in the production of neuropeptide involved in the control of feeding might lead to the development of obesity in these animals.

Neurotensin and neuromedin N brain levels after fornix transection: evidence for an efficient neurotensin precursor processing in subicular neurons.

High levels of neurotensin/neuromedin N precursor mRNA, but few if any NT-positive perikarya have been detected in the dorsal subiculum of the adult rat or human hippocampus. This apparent discrepancy was tentatively ascribed to a lack of precursor mRNA translation or to a poor precursor posttranslational processing in neurons of the hippocampus. Another hypothesis is that in long neuronal pathways, maturation of neuropeptide precursors and derived peptides occurs during axonal transport to terminals, a process which accounts for the lack of peptide detection in cell bodies. In order to test this hypothesis, we performed surgical transection of the fornix to interrupt axonal transport of putative NT/NN products arising from the dorsal hippocampus and measured NT and NN levels in different brain regions. In the mamillary bodies, the main projection area of the dorsal subiculum, NN and NT levels were highly reduced 4 or 14 days after the septo-hippocampal transection which was correlated with a slight increase in NN and NT levels in the dorsal hippocampus and the retrosplenial cortex of 4 days lesioned animals. An increase in hypothalamic NN levels was also detected 14 days after the lesion. These data suggest that the peptide precursor processing can take place during the axonal transport, as shown here for neurotensin and neuromedin N from subicular neurons to their efferent brain areas such as the mamillary bodies.

Effects of thiorphan, bestatin and a novel metallopeptidase inhibitor JMV 390-1 on the recovery of neurotensin and neuromedin N released from mouse hypothalamus.

The effects of the endopeptidase 24.11 ('enkephalinase') inhibitor thiorphan, the aminopeptidase inhibitor bestatin and a novel metallopeptidase inhibitor JMV 390-1 on the K(+)-evoked release of immunoreactive neurotensin and neuromedin N (iNT and iNN) from mouse hypothalamic slices were examined. (JMV 390-1 inhibits several metallopeptidases including endopeptidases 24.11, 24.15 and 24.16, and aminopeptidase N equipotently with Ki values around 50 nM.) Thiorphan increased the recovery of released iNT nearly 2-fold and had no effect on iNN. Bestatin produced a 4-fold increase in iNN recovery and was inactive on iNT. Finally, iNT and iNN recoveries were increased up to 4- and 5-fold, respectively, by JMV 390-1. These results show that in the mouse hypothalamus endopeptidase 24.11 participates with other metalloendopeptidases to the degradation of endogenously released NT while endogenously released NN is principally degraded by aminopeptidase(s).

Calcium-dependent release of neuromedin N and neurotensin from mouse hypothalamus.

Neuromedin N (NN), a hexapeptide, was isolated from porcine spinal cord. Its C-terminal tetrapeptide sequence is identical to that of neurotensin (NT) and it exhibits NT-like effects when injected in the central nervous system. Both peptides were recently shown to be encoded in the same precursor molecule. We have just developed a sensitive and specific radioimmunoassay (RIA) for NN and showed that the peptide central nervous system distribution paralleled that of NT, the highest concentrations being found in the hypothalamus. Using this assay and a specific RIA for NT, we show here that NN and NT were simultaneously released from slices of mouse hypothalamus by K(+)-induced depolarization in a Ca(++)-dependent manner. The ratio of released NN over NT was 0.3 and was identical to the ratio of endogenous NN over NT. For both NN and NT, the releasable peptide pool represented 2% of the endogenous peptide pool. HPLC characterization of the releasable and endogenous immunoreactive material reacting with the NN and NT antisera showed that it coeluted with synthetic NN and NT, respectively. The present data further support the hypothesis that NN acts as a neuromodulator in the central nervous system.

Characterization and autoradiographic distribution of neurotensin binding sites in the human brain.

The characteristics and topographical distribution of monoiodo 125I-Tyr3-neurotensin (NT) binding sites in normal human brain tissue were studied on brain sections and by quantitative autoradiography. Sections at the level of the substantia nigra show a dissociation constant and maximal binding capacity of 4.8 +/- 0.8 nM and 70 +/- 7 fmol/mg protein, respectively. High density of 125I-NT binding sites were mainly found in dopaminergic (DA)-rich areas such as the substantia nigra, the ventral tegmental area, the striatum and the nucleus accumbens, further supporting an interaction between NT and DA neurons in human brain.

The amino acid sequence of radioimmunoassayable neurotensin from bovine intestine.

The amino acid sequence of radioimmunoassayable neurotensin, isolated from bovine small intestinal extracts, has been shown to be the same as that of the peptide originally isolated from bovine hypothalamic extracts. This was accomplished by sequence studies on the intact peptide as well as on its chymotryptic and papain-generated fragments. Thus, neurotensin joins the group of biologically active peptides shown to be present in the same molecular form in both brain and intestine.