N. Ling
Salk Institute for Biological Studies
Network
Latest external collaboration on country level. Dive into details by clicking on the dots.
Publication
Featured researches published by N. Ling.
Biochemical and Biophysical Research Communications | 1983
Frederick Esch; N. Ling; Peter Bohlen; S Y Ying; Roger Guillemin
The two major protein components of bovine seminal plasma, PDC-109 and BSP I, have been purified by gel filtration, partition chromatography and reverse-phase high performance liquid chromatography from an 86% ethanol precipitate of bovine seminal plasma ejaculate. The complete 109-residue amino acid sequence of PDC-109 has been established by automated Edman degradation of the intact peptide as well as its proteolytic digestion and cyanogen bromide cleavage fragments. The 12,774 dalton structure has two structurally similar domains of 38 and 41 amino acids, each containing two disulfide bonds.
Biochemical and Biophysical Research Communications | 1980
N. Ling; Frederick Esch; D. Davis; Maluz Mercado; M. Regno; Peter Bohlen; Paul Brazeau; Roger Guillemin
The synthesis of ovine hypothalamic somatostatin-28 (Ser-Ala-Asn-Ser-Asn-Pro-Ala-Met-Ala-Pro-Arg-Glu-Arg-Lys-Ala-Gly-Cys-Lys-Asn-Phe-Phe-Trp-Lys-Thr-Phe-Thr-Ser-Cys-OH) has been accomplished by solid phase methodology. The structure of the synthetic material was verified by: (1) direct sequence analysis with a Beckman 89°C sequencer, (2) correlation of the amino acid analyses of the isolated tryptic peptide fragments with their theoretical compositions, and (3) comparison, using high performance liquid chromatography, of the synthetic methionine-sulfoxide and methionine-sulfone modified NH2-terminal peptides (residues 1–11) with the corresponding tryptic fragment from somatostatin-28.
Clinical Endocrinology | 1976
Wylie Vale; Catherine Rivier; Marvin R. Brown; Juhani Leppäluoto; N. Ling; Mike Monahan; Jean Rivier
It is now accepted that the hypothalamic regulation of the anterior pituitary gland is mediated by hypothalamic hypophysiotrophic hormones, often referred to as hypothalamic releasing and release inhibiting factors or hormones. To date three of these neurohormones have been characterized, synthesized and shown to have unequivocal hypophysiotropic activities. Shown in Table 1 are the structures and principal hypophysiotrophic actions of these three peptides : TRF (thyrotrophin releasing factor), LRF (luteinizing hormone releasing factor) and somatostatin. In addition to their effects on the pituitary, each of these hormones has been reported to have direct actions on the central nervous system to modify behaviour. Furthermore, the wide distribution of both TRF and somatostatin throughout the central nervous system has led some to propose that they might possess physiological neurotrophic roles. Somatostatin is also distributed throughout and influences the function of the gastrointestinal tract and pancreas. A more thorough discussion of the physiological and clinical significances of the hypothalamic hypophysiotrophic hormones is beyond the scope of this report; for review see Vale et al. (1975b), Gerich & Guillemin (1976), Besser et al. (1974), Vale & Rivier (1975) and other articles in this volume by Drs Fink, Besser and Forsham. These peptides are useful diagnostic tools with considerable therapeutic potential even though their applications are limited by factors such as short duration of action and lack of complete specificity of effects. By now several laboratories have synthesized hundreds of analogues of these hormones to look for peptides that would possess useful biological properties: such as being more potent, having a longer duration of action, expressing different specificity patterns, or behaving as antagonists of a native hypothalamic hormone. These analogues also provide useful tools for structure-function relationship studies in which the various determinants of biological activity can be investigated. The resultant biological activity of a peptide can reflect variations in distribution, metabolism, receptor affinity or intrinsic activity (ability to generate a signal following receptor interaction). Also other more subtle changes may exist. For example, Roth et. al. (1975) have reported that analogues of insulin have been found which do not exhibit the negative co-operativity Correspondence: Dr W. Vale, The Salk Institute for Biological Studies, La Jolla, California 92037, U.S.A.
Experimental Biology and Medicine | 1984
Roger Guillemin; Fusun Zeytin; N. Ling; Peter Bohlen; Frederick Esch; Paul Brazeau; Bertrand Bloch; William B. Wehrenberg
Conclusions In conclusion, the long-sought growth hormone-releasing factor has been characterized and sequenced in several species (human, porcine, bovine, and murine). The noteworthy achievement with GRF has been the rapidity with which the sum of information has been gathered. In the space of 12 months most of the pioneering studies on the mechanism of action of the peptide in vivo and in vitro were described. Immunocytochemical mapping of GRF neurons was carried out. Structure-function studies were initiated. Clinical trials were started and confirmed the potent GH-releasing activity of GRF in man. The effect of the peptide on specific GH mRNA levels was described and molecular cloning was used to establish the structures of human (pancreas) preproGRF. All the hypothalamic releasing factors which had been postulated in the early fifties as humoral regulators of the secretion of each pituitary hormone have now been characterized.
Current topics in molecular endocrinology | 1976
Wylie Vale; Catherine Rivier; Marvin R. Brown; Lana Chan; N. Ling; Jean Rivier
Anterior pituitary cell function can be influenced by neural peptides (including the hypothalamic regulatory hormones, HRH), neurotransmitters and peripheral hormones. Because of the pituitary gland’s close (vascular) coupling to the central nervous system, the secretory rates of its hormones are highly dynamic and easily modified by experimental circumstances. It is often difficult to determine if responses seen in vivo are due to direct effects on the pituitary or are mediated through extrapituitary mechanisms. In vitro assa;ys of the hypophysiotropic substances offer isolation from such indirect effects. It has been the aim of biologists in this field to develop an in vitro assay that would be technically simple, sensitive, reliable, accurate and valid. With the availability of radioimmunoassays to the pituitary hormones, the preparations of pituitary tissues themselves became the limiting factor to the achievement of these goals.
Biochemical and Biophysical Research Communications | 1973
N. Ling; Roger Burgus; Jean Rivier; Wylie Vale; Paul Brazeau
Abstract A peptide has been isolated from the extracts of ovine hypothalamic tissues that inhibits the secretion of somatotropin (growth hormone). The primary structure of this peptide, named somatostatin, has been determined to be H-Ala-Gly- Cys-Lys-Asn-Phe-Phe-Trp-Lys-Thr-Phe-Thr-Ser-Cys -OH. The sequence -Asn-Phe-Phe-Trp-Lys- was confirmed by direct mass spectrometry, using an acetylated and permethylated tryptic digest of somatostatin.
Brain Research | 1984
Robert Benoit; Peter Bohlen; Frederick Esch; N. Ling
Two neuropeptides reacting with antibodies directed against the C-terminal region of somatostatin 28(1-12) were purified to homogeneity from rat brain extracts. Amino acid analysis revealed that the larger peptide (8 kdaltons) consisted of 76 amino acids. Microsequencing established its aminoterminal structure as: Ala-Pro-Ser-Asp-Pro-Arg-Leu-Arg-Gln-Phe-X-Gln-Lys. The 8 kdalton somatostatin 28(1-12)-like peptide corresponds to the whole prosomatostatin molecule without Arg-Lys-somatostatin-14. The smaller peptide (5 kdaltons) consists of 44 amino acids and is generated after cleavage of a Leu-Leu bond at position 56-57 of pre-prosomatostatin. Both 8 kdalton and 5 kdalton somatostatin 28(1-12)-like peptides contain somatostatin 28(1-12) at their C-termini. The 4 most abundant neuropeptides derived from pre-prosomatostatin (pre-proSS) and presently characterized are: somatostatin-14, somatostatin 28(1-12), somatostatin 28 and pre-proSS25-100.
Analytical Biochemistry | 1976
N. Ling; Juhani Leppäluoto; Wylie Vale
Abstract Iodination of thyrotropin releasing factor (TRF) with chloramine-T and NaI at pH 7.5 produces both mono- and diiodo-TRF as well as unreacted TRF and I − . The four compounds can be separated by ion-exchange chromatography on SP-Sephadex C-25, using a stepwise gradient of 0.01 m MH 4 OAc at pH 3.5, 5.2 and 7.2. The monoiodo-TRF was characterized by nuclear magnetic resonance, mass spectra, and elemental analysis as [Im-5-iodo-His 2 ]-TRF while the diiodo-TRF was similarly characterized as [Im-2,5-diiodo-His 2 ]-TRF. Both iodination products showed no thyrotropin releasing activity at doses up to 1 μg in the in vivo mouse bioassay. The specific binding of mono- and diiodo-TRF was measured against three different TRF antisera. In the absence of TRF, antisera A and C bound more diiodo-TRF than monoiodo-TRF while antiserum B seemed to bind both tracers equally. In all cases steeper curves were produced when diiodo-TRF was displaced by synthetic TRF.
Biochemical and Biophysical Research Communications | 1975
N. Ling; Wylie Vale
Summary [(Nα-Me)Leu7]LRF, (pGlu-His-Trp-Ser-Tyr-Gly-(Nα-Me)Leu-Arg-Pro-Gly-NH2), and [D-Ala6,(Na-Me)Leu7]LRF, (pGlu-His-Trp-Ser-Tyr-D-Ala-(Nα-Me) Leu-Arg-Pro-Gly-NH2), were synthesized by solid-phase methodology. The peptides were assayed by the in vitro system against LRF and found to have ca. 102% and 560% the potency of LRF respectively. The biological results are interpreted in terms of the conformational aspects of LRF.
Biochemical and Biophysical Research Communications | 1979
Solange Lavielle; N. Ling; Paul Brazeau; Robert Benoit; T. Wasada; D. Harris; R. Unger; Roger Guillemin
Abstract The synthesis by solid-phase methodology of two glycosylated analogs of somatostatin [Glc-Asn 5 ]-SS and [NAcGlc-Asn 5 ]-SS is described. These two analogs have been biologically tested on the secretion of pituitary growth hormone, pancreatic glucagon and insulin. The results show that glycosylation of somatostatin on the Asn 5 residue decreases by a hundred fold the inhibition activity on GH release when tested in vitro . In vivo , since the activity is similar to somatostatin the carbohydrates are probably removed by some enzymatic reaction and thus liberate the full activity of somatostatin.