Acta physiol. scand. 1978. 102. 143-147 From the Department of Pharmacology, Karolinska Institutet, Stockholm, Sweden, a n d the Institute of Biomedical Research, The University of Texas Austin, Texas, USA
Indirect vascular actions of (Gln4)-neurotensinin canine adipose tissue BY
SUNEROSELL,AKEROKAEUS, DINGCHANGand KARLFOLKERS Received 16 June 1977
Abstract ROSELL,S., A. ROKAEUS, D. CHANGand K. FOLKERS. Indirect vascular actions of (Oh4)neurotension in canine adipose tissue. Acta physiol. scand. 1978. 102. 143-147. The vasoconstrictor action of the tridecapeptide (Gln4)-neurotensin has been studied in subcutaneous adipose tissue in the inguinal region of anesthetized dogs. Close intra-arterial infusion of (Gln4))-neurotensin, 30-120 pmol x kg-I b.wt. x min-', elicited similar vasoconstrictions in the adipose tissue on the infusion side and on the contralateral side. This suggests that (Gln4)-neurotensin must enter the general circulation before it can elicit vasoconstriction. Removal of parts of the gastrointestinal tract did not change the vasoconstrictor response. Thus, there is no indication of release of vasoactive substances from the gastrointestinal tract by (Gln4)-neurotensin. Infusion into the portal vein elicited the same vasoconstriction in adipose tissue as the same dose administered i.v. It is suggested that the vasoconstrictor action in adipose tissue is not caused by (G1n')-neurotensin per se. Instead, vasoactive substance(s) may be formed from (Gln')-neurotensin.
Neurotensin is a tridecapeptide which has been isolated from bovine hypothalamus (Carraway and Leeman 1973). The amino acid sequence has been established and the synthesis decribed (Carraway and Leeman 1975 a, 1975 b). Recently Folkers et al. (1976) suggested that the acid analog (Gln4)-neurotensin rather than neurotensin may be the naturally occurring peptide. In dogs, neurotensin and (Gln4)-neurotensin infused i.v. (5 pmol x kg-' x min-l, or more) have pronounced effects on certain functions in the gastrointestinal tract. Thus, they inhibit the gastric motor activity (Anderson et al. 1977) and the gastric acid secretion (Anderson et al. 1976). The neurotensins also have vasoconstrictor action in some vascular beds, including denervated adipose tissue (Rosell et al. 1976). The vasoconstriction is delayed in onset, occurring about 5 min after the beginning of i.v. infusion. Therefore, it was suggested that this effect and possibly other actions of neurotensins, may be mediated by a neurotensin metabolite or by the release of some vasoactive sbbstances(s) (Rosell et al. 1976). In the present investigation these possibilities as far as the vasoconstriction in adipose tissue is concerned have been analyzed experimentally. 143
144
SUNE ROSELL,
AKE ROKAEUS, DING CHANG AND KARL FOLKERS
Methods The experiments were performed on 19 female mongrel dogs weighing 9-25 kg, anesthetized with sodium pentobarbital, 30 mg/kg with supplement as necessary. Tracheotomy was performed, and the dogs were ventilated with a Braun Melsungen model 74052 respirator. To prevent coagulation, heparin (2500 IE/kg) was administered. Subcutaneous adipose tissue in the left and the right inguinal regions was isolated from skin and other surrounding tissues (Rosell 1966). This procedure provides two almost identical adipose tissue preparations, each supplied by one artery, one vein and one mixed nerve. The tissues were acutely denervated. Blood flow was measured using a silicon-filled drop recorder inserted into the arterial circuit of the tissue. Systemic blood pressure was measured in a carotid artery with a Statham P23Ac transducer and recorded together with blood flow on a Grass model 7B polygraph. Vasoconstriction was calculated in peripheral resistance units (PRU mmHg 4 m l - ' ~min-' x 100 g-'). To obtain a measure of the total vasoconstrictor effect of infused drugs in adipose tissue the area under the peripheral resistance time curve was calculated using the trapezoidal rule. Drugs were administered by close intra-arterial infusions via a side arm in the arterial line just proximal to the drop counter. Infusions into the portal vein were made via a polyethylene cannula inserted through a mesenteric vein. (Gin4)-neurotensin (M.W. 1674) was dissolved in saline.
Results Adipose tissue blood flow. Infusion of (Gln4)-neurotensin intraarterially (30-1 20 pmol x kg-I x min) to the inguinal subcutaneous adipose tissue induced vasoconstriction after a latency of about 5 min. As is evident from Fig. 1, La. infusion of (Gln4)-neurotensin on one side induced similar vasoconstrictions on the infusion side and contralateral side as far as latency, degree and duration of vasoconstriction are concerned. These results indicate that (Gln4)-neurotensin must enter the general circulation before inducing vasoconstriction in adipose tissue. Partial removal of the gastrointestinal tract. One possible mechanism for the vasoconstrictor action of (Gln4)-neurotensin is that the polypeptide releases vasoactive substances. In order to examine this possibility, abdominal organs were extirpated in acute expts. and the effects of these procedures on the vasoconstriction in subcutaneous adipose tissue were investigated. (Gln4)-neurotensin was infused i.v. before and after the extirpations of various
TIME rn,"
m 7.
Fig. 1. Experimental record illustrating blood pressure and blood flow in actuely denervated subcutaneous adipose tissue. Blood flow was measured in both the left (upper record) and the right (lower record) adipose tissue preparations. (Gln4)-neurotensin, 120 pmol x kg-l x min-l) was infused into the artery supplying the right adipose tissue.
145
INDIRECT ACTIONS OF NEUROTENSIN
-
BLOOD PRESSURE
(8
..
-
adipose t ISSUB
TIME min
I""
Fig. 2. Experimental record illustrating blood pressure and blood flow in acutely denervated subcutaneous adipose tissue. (Gln4)-neurotensin, 30 pmol X. kg-' X. min-I. was infused intravenously (upper record) into the portal vein (lower record).
abdominal viscera. The reference pattern was unaffected by splenectomy (2 expts.), pancreatectomy (3 expts.), removal of duodenum, jejunum, ileum or colon individually. However, in those expts. where the whole small intestine was removed the responses were diminished although the latency and vasoconstrictor pattern was unchanged. Infusion of (Gln4)-neurotensin into the portal vein. (Gin*)-neurotensin infused into the portal vein in doses ranging between 3&120 pmol x kg-l x min-l, produced vasoconstriction in adipose tissue which was of the same magnitude as the vasoconstriction following i.v. administration. Moreover, the latency and the duration of the vasoconstriction were similar irrespective of the routes of administration (Fig. 2). The vasoconstriction, expressed as the area under the peripheral resistance time curve, was not significantly different from that following the same dose administered intravenously (Student's t-test for paired observations, n = 6 ) (Fig. 3). Adipose tissue
I
Fig. 3. Changes in vascular resistance in subcutaneous adipose tissue in one dog following infusion of (Gln*)-neurotensin, 60 pmol x kg-1 x min-' intravenously or into the portal vein. I6 - 185872
mq
761214
146
SUNE ROSELL,
AKE
ROKAEUS, DING CHANG AND KARL FOLKERS
Discussion Previously reported experiments have shown that in doses above 18 pmol x kg-l x min-1 i.v. both neurotensin and (Gln4)-neurotensin elicit a delayed vasoconstriction in denervated canine subcutaneous adipose tissue and to a lesser degree in the skin and small intestine. Due to the long latency of about 5 min following i.v. administration, it was suggested that the vasoconstriction in adipose tissue was indirectly mediated. The vasoconstriction was not blocked by adrenergic a-receptor blocking agents or by bilateral ligation of the renal pedicles. Thus, the vasoconstriction does not seem to be mediated by catecholamines or by effects on the renin-angiotensin system or other renal mechanisms. Furthermore, the vasoconstrictor action does not seem to be mediated via the central nervous system since infusion of neurotensins into the vertebral artery produced smaller effects than similar doses administered i.v. (Rosell ef al. 1976). The present expts. were aimed at further testing the hypothesis of an indirect action of (Gln4)-neurotensin. Apart from a slight initial vasodilatation, close intra-arterial infusion of (Gln4)-neurotensin to the inguinal adipose tissue did not induce any changes in peripheral resistance that could be ascribed to a direct action of the peptide. Instead i.a. infusions of (Gln4)-neurotensin on one side produced similar vasoconstrictions on the infusion side and contralateral side. This result indicates that (Gln')-neurotensin must enter the general circulation before it can elicit vasoconstriction in the adipose tissue. Therefore it is possible that (Gln4)-nebrotensinreleases some vasoconstrictor substance or, alternatively, is metabolized to one or more peptides with vasoconstrictor properties. Both these alternatives have been considered in the present series of expts. The results of expts. in which abdominal organs or parts of the gastrointestinal canal were extirpated did not indicate any release. The reduction of the vasoconstrictor response following extirpation of large parts of the gastrointestinal tract may be a consequence of the decrease in the portal blood flow and thus an altered distribution of the infused (Gln4)-nebrotensin. The finding that infusion of (Gln4)-neurotensin infused into the portal vein elicits the same vasoconstriction as the same dose administered i.v. indicates that the peptide is not inactivated in the liver. Intead, (Gln4)-neurotensin may be activated to vasoactive substances by e.g. cleavage of the neurotensin molecule to partial sequences. Presumably, such a conversion does not take place in the liver since one would then have expected a greater vasoconstriction after infusion into the portal vein than after i.v. infusion. The reason being that following i.v. infusion only about 25% of the total dose will be distributed to the liver whereas 100% of the dose will pass the liver after intraportal administtation. Whether some of the partial sequences of neurotensin have vasoconstrictor properties is not known. Carraway and Leeman (1975 c) have tested partial sequences of neurotensin on several biological parameters and found that COOH-terminal sequences of 5 or more amino acids in length have full intrinsic activity. Whether formation of these peptides may occur in plasma or in some other tissue remains to be investigated. Apart from their vasoactivity in small doses in adipose tissue and the gastrointestinal tract, neurotensins are potent inhibitors of gastric antral pressure motility and gastric acid secretion stimulated by pentagastrin or a test meal (Andersson et al. 1976, 1977, Rosell
INDIRECT ACTIONS OF NEUROTENSIN
147
et al. 1976). At higher doses, neurotensins produce elevation of blood glucose (Carraway,
Demers and Leeman 1973, Rosell er al. 1976). Whether or not some or all of these effects are directly mediated by neurotensins or are due to metabolites remains to be determined. This investigation was appropiately supported by grants from the Swedish Medical Research Council (3518), Ake Wibergs Stiftelse and from the Robert A. Welch Foundation. Heparin was generously supplied by AB Vitrum, Sweden.
References ANDERSON,S., D. CHANG,K. FOLKERS and S. ROSELL,Inhibition of gastric acid secretion in dogs by neurotensin. Life Sciences 1976. 19. 367-370. ANDERSON,S., S. ROSELL,U. HJELMQIJIST, D. CHANGand K. FOLKERS, Inhibition of gastric and intestinal motor activity in dogs by (Gln4)-neurotensin. Acta physiol. wand. 1977. 100. 231-235. The isolatiation of a new hypotensive peptide. Neurotensin from bovine CARRAWAY, R. and S. E. LEEMAN, hypothalami. J. biol. Chem. 1973. 248. 6854-6861. The amino acid sequence of a hypothalamic peptide. Neurotensin. J. CARRAWAY, R. and S. E. LEEMAN, biol. Chem. 1975 a. 250. 1907-1911. CARRAWAY, R. and S. E. LEEMAN, The synthesis of neurotensin. J. biol. Chem. 1975 b. 250. 1912-1918. Structural requirements for the biological activity of neurotensin, a new CARRAWAY, R. and S. E. LEEMAN, vasoactive peptide. In: Peptides: Chemistry, Structure and Biology, edited by R Walter and J. Meinenhofer. Ann. Arbor Science Publ. 1975 c. Hyperglycemic effect of a hypothalamic peptide. Fed. Proc. CARRAWAY, R., L. DEMERS and S. E. LEEMAN, 1973. 32. 211. FOLKERS, K., D. CHANG,S. LEEMAN and C. BOWERS,Proc. nut. Acad. Sci. (Wash.) 1976. 73. No. 11. 3833. ROSELL,S., Release of free fatty acids from subcutaneous adipose tissue in dogs following sympathetic nerve stimulation. Acta physiol. scand. 1966. 67. 3 17-322. D. CHANGand K. FOLKERS, Cardiovascular and metabolic actions of neurotensin ROSELL,S., E. BURCHER, and (Gln4)-neurotensin. Arta physiol. scand. 1976. 98. 484-491.
Indirect vascular actions of (Gln4)-neurotensinin canine adipose tissue BY
SUNEROSELL,AKEROKAEUS, DINGCHANGand KARLFOLKERS Received 16 June 1977
Abstract ROSELL,S., A. ROKAEUS, D. CHANGand K. FOLKERS. Indirect vascular actions of (Oh4)neurotension in canine adipose tissue. Acta physiol. scand. 1978. 102. 143-147. The vasoconstrictor action of the tridecapeptide (Gln4)-neurotensin has been studied in subcutaneous adipose tissue in the inguinal region of anesthetized dogs. Close intra-arterial infusion of (Gln4))-neurotensin, 30-120 pmol x kg-I b.wt. x min-', elicited similar vasoconstrictions in the adipose tissue on the infusion side and on the contralateral side. This suggests that (Gln4)-neurotensin must enter the general circulation before it can elicit vasoconstriction. Removal of parts of the gastrointestinal tract did not change the vasoconstrictor response. Thus, there is no indication of release of vasoactive substances from the gastrointestinal tract by (Gln4)-neurotensin. Infusion into the portal vein elicited the same vasoconstriction in adipose tissue as the same dose administered i.v. It is suggested that the vasoconstrictor action in adipose tissue is not caused by (G1n')-neurotensin per se. Instead, vasoactive substance(s) may be formed from (Gln')-neurotensin.
Neurotensin is a tridecapeptide which has been isolated from bovine hypothalamus (Carraway and Leeman 1973). The amino acid sequence has been established and the synthesis decribed (Carraway and Leeman 1975 a, 1975 b). Recently Folkers et al. (1976) suggested that the acid analog (Gln4)-neurotensin rather than neurotensin may be the naturally occurring peptide. In dogs, neurotensin and (Gln4)-neurotensin infused i.v. (5 pmol x kg-' x min-l, or more) have pronounced effects on certain functions in the gastrointestinal tract. Thus, they inhibit the gastric motor activity (Anderson et al. 1977) and the gastric acid secretion (Anderson et al. 1976). The neurotensins also have vasoconstrictor action in some vascular beds, including denervated adipose tissue (Rosell et al. 1976). The vasoconstriction is delayed in onset, occurring about 5 min after the beginning of i.v. infusion. Therefore, it was suggested that this effect and possibly other actions of neurotensins, may be mediated by a neurotensin metabolite or by the release of some vasoactive sbbstances(s) (Rosell et al. 1976). In the present investigation these possibilities as far as the vasoconstriction in adipose tissue is concerned have been analyzed experimentally. 143
144
SUNE ROSELL,
AKE ROKAEUS, DING CHANG AND KARL FOLKERS
Methods The experiments were performed on 19 female mongrel dogs weighing 9-25 kg, anesthetized with sodium pentobarbital, 30 mg/kg with supplement as necessary. Tracheotomy was performed, and the dogs were ventilated with a Braun Melsungen model 74052 respirator. To prevent coagulation, heparin (2500 IE/kg) was administered. Subcutaneous adipose tissue in the left and the right inguinal regions was isolated from skin and other surrounding tissues (Rosell 1966). This procedure provides two almost identical adipose tissue preparations, each supplied by one artery, one vein and one mixed nerve. The tissues were acutely denervated. Blood flow was measured using a silicon-filled drop recorder inserted into the arterial circuit of the tissue. Systemic blood pressure was measured in a carotid artery with a Statham P23Ac transducer and recorded together with blood flow on a Grass model 7B polygraph. Vasoconstriction was calculated in peripheral resistance units (PRU mmHg 4 m l - ' ~min-' x 100 g-'). To obtain a measure of the total vasoconstrictor effect of infused drugs in adipose tissue the area under the peripheral resistance time curve was calculated using the trapezoidal rule. Drugs were administered by close intra-arterial infusions via a side arm in the arterial line just proximal to the drop counter. Infusions into the portal vein were made via a polyethylene cannula inserted through a mesenteric vein. (Gin4)-neurotensin (M.W. 1674) was dissolved in saline.
Results Adipose tissue blood flow. Infusion of (Gln4)-neurotensin intraarterially (30-1 20 pmol x kg-I x min) to the inguinal subcutaneous adipose tissue induced vasoconstriction after a latency of about 5 min. As is evident from Fig. 1, La. infusion of (Gln4)-neurotensin on one side induced similar vasoconstrictions on the infusion side and contralateral side as far as latency, degree and duration of vasoconstriction are concerned. These results indicate that (Gln4)-neurotensin must enter the general circulation before inducing vasoconstriction in adipose tissue. Partial removal of the gastrointestinal tract. One possible mechanism for the vasoconstrictor action of (Gln4)-neurotensin is that the polypeptide releases vasoactive substances. In order to examine this possibility, abdominal organs were extirpated in acute expts. and the effects of these procedures on the vasoconstriction in subcutaneous adipose tissue were investigated. (Gln4)-neurotensin was infused i.v. before and after the extirpations of various
TIME rn,"
m 7.
Fig. 1. Experimental record illustrating blood pressure and blood flow in actuely denervated subcutaneous adipose tissue. Blood flow was measured in both the left (upper record) and the right (lower record) adipose tissue preparations. (Gln4)-neurotensin, 120 pmol x kg-l x min-l) was infused into the artery supplying the right adipose tissue.
145
INDIRECT ACTIONS OF NEUROTENSIN
-
BLOOD PRESSURE
(8
..
-
adipose t ISSUB
TIME min
I""
Fig. 2. Experimental record illustrating blood pressure and blood flow in acutely denervated subcutaneous adipose tissue. (Gln4)-neurotensin, 30 pmol X. kg-' X. min-I. was infused intravenously (upper record) into the portal vein (lower record).
abdominal viscera. The reference pattern was unaffected by splenectomy (2 expts.), pancreatectomy (3 expts.), removal of duodenum, jejunum, ileum or colon individually. However, in those expts. where the whole small intestine was removed the responses were diminished although the latency and vasoconstrictor pattern was unchanged. Infusion of (Gln4)-neurotensin into the portal vein. (Gin*)-neurotensin infused into the portal vein in doses ranging between 3&120 pmol x kg-l x min-l, produced vasoconstriction in adipose tissue which was of the same magnitude as the vasoconstriction following i.v. administration. Moreover, the latency and the duration of the vasoconstriction were similar irrespective of the routes of administration (Fig. 2). The vasoconstriction, expressed as the area under the peripheral resistance time curve, was not significantly different from that following the same dose administered intravenously (Student's t-test for paired observations, n = 6 ) (Fig. 3). Adipose tissue
I
Fig. 3. Changes in vascular resistance in subcutaneous adipose tissue in one dog following infusion of (Gln*)-neurotensin, 60 pmol x kg-1 x min-' intravenously or into the portal vein. I6 - 185872
mq
761214
146
SUNE ROSELL,
AKE
ROKAEUS, DING CHANG AND KARL FOLKERS
Discussion Previously reported experiments have shown that in doses above 18 pmol x kg-l x min-1 i.v. both neurotensin and (Gln4)-neurotensin elicit a delayed vasoconstriction in denervated canine subcutaneous adipose tissue and to a lesser degree in the skin and small intestine. Due to the long latency of about 5 min following i.v. administration, it was suggested that the vasoconstriction in adipose tissue was indirectly mediated. The vasoconstriction was not blocked by adrenergic a-receptor blocking agents or by bilateral ligation of the renal pedicles. Thus, the vasoconstriction does not seem to be mediated by catecholamines or by effects on the renin-angiotensin system or other renal mechanisms. Furthermore, the vasoconstrictor action does not seem to be mediated via the central nervous system since infusion of neurotensins into the vertebral artery produced smaller effects than similar doses administered i.v. (Rosell ef al. 1976). The present expts. were aimed at further testing the hypothesis of an indirect action of (Gln4)-neurotensin. Apart from a slight initial vasodilatation, close intra-arterial infusion of (Gln4)-neurotensin to the inguinal adipose tissue did not induce any changes in peripheral resistance that could be ascribed to a direct action of the peptide. Instead i.a. infusions of (Gln4)-neurotensin on one side produced similar vasoconstrictions on the infusion side and contralateral side. This result indicates that (Gln')-neurotensin must enter the general circulation before it can elicit vasoconstriction in the adipose tissue. Therefore it is possible that (Gln4)-nebrotensinreleases some vasoconstrictor substance or, alternatively, is metabolized to one or more peptides with vasoconstrictor properties. Both these alternatives have been considered in the present series of expts. The results of expts. in which abdominal organs or parts of the gastrointestinal canal were extirpated did not indicate any release. The reduction of the vasoconstrictor response following extirpation of large parts of the gastrointestinal tract may be a consequence of the decrease in the portal blood flow and thus an altered distribution of the infused (Gln4)-nebrotensin. The finding that infusion of (Gln4)-neurotensin infused into the portal vein elicits the same vasoconstriction as the same dose administered i.v. indicates that the peptide is not inactivated in the liver. Intead, (Gln4)-neurotensin may be activated to vasoactive substances by e.g. cleavage of the neurotensin molecule to partial sequences. Presumably, such a conversion does not take place in the liver since one would then have expected a greater vasoconstriction after infusion into the portal vein than after i.v. infusion. The reason being that following i.v. infusion only about 25% of the total dose will be distributed to the liver whereas 100% of the dose will pass the liver after intraportal administtation. Whether some of the partial sequences of neurotensin have vasoconstrictor properties is not known. Carraway and Leeman (1975 c) have tested partial sequences of neurotensin on several biological parameters and found that COOH-terminal sequences of 5 or more amino acids in length have full intrinsic activity. Whether formation of these peptides may occur in plasma or in some other tissue remains to be investigated. Apart from their vasoactivity in small doses in adipose tissue and the gastrointestinal tract, neurotensins are potent inhibitors of gastric antral pressure motility and gastric acid secretion stimulated by pentagastrin or a test meal (Andersson et al. 1976, 1977, Rosell
INDIRECT ACTIONS OF NEUROTENSIN
147
et al. 1976). At higher doses, neurotensins produce elevation of blood glucose (Carraway,
Demers and Leeman 1973, Rosell er al. 1976). Whether or not some or all of these effects are directly mediated by neurotensins or are due to metabolites remains to be determined. This investigation was appropiately supported by grants from the Swedish Medical Research Council (3518), Ake Wibergs Stiftelse and from the Robert A. Welch Foundation. Heparin was generously supplied by AB Vitrum, Sweden.
References ANDERSON,S., D. CHANG,K. FOLKERS and S. ROSELL,Inhibition of gastric acid secretion in dogs by neurotensin. Life Sciences 1976. 19. 367-370. ANDERSON,S., S. ROSELL,U. HJELMQIJIST, D. CHANGand K. FOLKERS, Inhibition of gastric and intestinal motor activity in dogs by (Gln4)-neurotensin. Acta physiol. wand. 1977. 100. 231-235. The isolatiation of a new hypotensive peptide. Neurotensin from bovine CARRAWAY, R. and S. E. LEEMAN, hypothalami. J. biol. Chem. 1973. 248. 6854-6861. The amino acid sequence of a hypothalamic peptide. Neurotensin. J. CARRAWAY, R. and S. E. LEEMAN, biol. Chem. 1975 a. 250. 1907-1911. CARRAWAY, R. and S. E. LEEMAN, The synthesis of neurotensin. J. biol. Chem. 1975 b. 250. 1912-1918. Structural requirements for the biological activity of neurotensin, a new CARRAWAY, R. and S. E. LEEMAN, vasoactive peptide. In: Peptides: Chemistry, Structure and Biology, edited by R Walter and J. Meinenhofer. Ann. Arbor Science Publ. 1975 c. Hyperglycemic effect of a hypothalamic peptide. Fed. Proc. CARRAWAY, R., L. DEMERS and S. E. LEEMAN, 1973. 32. 211. FOLKERS, K., D. CHANG,S. LEEMAN and C. BOWERS,Proc. nut. Acad. Sci. (Wash.) 1976. 73. No. 11. 3833. ROSELL,S., Release of free fatty acids from subcutaneous adipose tissue in dogs following sympathetic nerve stimulation. Acta physiol. scand. 1966. 67. 3 17-322. D. CHANGand K. FOLKERS, Cardiovascular and metabolic actions of neurotensin ROSELL,S., E. BURCHER, and (Gln4)-neurotensin. Arta physiol. scand. 1976. 98. 484-491.
