Isolation of a tridecapeptide from bovine intestinal tissue and its partial characterization as neurotensin.

Radioimmunoassayable neurotensin (R-NT) has been isolated from acid/acetone extracts of 50 kg of calf small intestine with an overall yield of approximately 15%. The concentration of R-NT in calf intestinal tissue was approximately 35 pmol/g wet weight. Throughout the purification procedures which involved adsorption onto sulfopropyl (SP)-Sephadex, chromatography on Sephadex G-25 and SP-Sephadex, immunoadsorption on neurotensin-antibody Sepharose and high voltage paper electrophoresis, R-NT displayed the chromatographic and electrophoretic properties of neurotensin. R-NT was found to contain a tridecapeptide with the same amino acid composition as neurotensin. This peptide yielded the same products as neurotensin when submitted to digestion by carboxypeptidase A or papain. Its immunological properties were indistinguishable from those of neurotensin and its potency in stimulating hypotension in anesthetized rats was comparable to that of synthetic neurotensin. If the amino acid sequence of this peptide proves to be the same as that of neurotensin, then neurotensin is another biologically active peptide isolated from both brain and intestinal tissues.

and when a peptide was synthesized according to this sequence it was found to be chemically and biologically indistinguishable from the native material (3). The availability of synthetic material made possible the development of a radioimmunoassay for neurotensin (4,5). Using this radioimmunoassay, radioimmunoassayable neurotensin (R-NT) 1 was found not only to have a differential distribution throughout brain (6) but was also present in extracts of small intestinal tissue in approximately the same concentrations as in hypothalamic tissue. Therefore, it seemed feasible as well as of considerable interest to isolate and characterize this material in order to determine if the R-NT was neurotensin itself or a different but closely related peptide( *This research was supported in part by National Institutes of Health Grant 7 ROl AM 16510-01 and Grant 1 ROl AM19428.01. i Supported in part by a grant from "1'Institut National de la were prepared in the laboratory as previously described (5). Carboxypeptidase A and papain were as previously reported (1). Sodium pentobarbital (Nembutal) was obtained from Abbott Laboratories. All reagents were reagent grade from Fisher Chemical Co.   Table I; column size, 14 x 150 cm (resin volume 201); fraction size, 400 ml; eluent, 0.1 M acetic acid; flow rate, 10 ml/min. Immunoreactivity was measured by radioimmunoassay, poly(Glu"', Lys") (PGL)-6 (04) and poly (Glu"', Lys")-4 (m-m) antibodies. Similar results were obtained when the other half of the material was chromatographed on the same conditions. purification steps thus far were performed at the New England Enzyme Center, Tufts University, School of Medicine, Boston, Mass.

First Cation Exchange
Chromatography-This pool was then applied at room temperature to a loo-ml column of sulfoproPY1 Sephadex C-25 equilibrated with 0.1 M acetic acid. The column was eluted with a linear gradient ranging from 0.05 M pyridine/acetate, pH 3.1, to 1 M pyridinelacetate, pH 5.5. The active region was located by radioimmunoassay ( Fig. 2) and pooled. The volume of the pool (370 ml) was reduced to 100 ml by lyophilization.

Second
Chromatography on Sephaden G-25-The IOO-ml pool was applied to a 5-liter column of Sephadex G-25 (fine) equilibrated in 0.1 M acetic acid. The fractions were examined for R-NT (Fig. 3). The active fractions were pooled and the pool lyophilized. http://www.jbc.org/ Downloaded from washed at 4" with 200 ml of phosphate buffer, then three times with 30 ml of an ice-cold solution of 0.01 N NaOH and finally, with 30 ml of phosphate buffer in order to elute any remaining dissociated R-NT. The three washes with 0.01 N NaOH and the last wash with phosphate buffer were pooled and the solution was rapidly neutralized with 2 N HCl. After the three batches of R-NT-containing solution were processed in that manner on the antibody-Sepharose conjugate, the R-NT remaining in the filtrates and 200.ml washes was recycled after concentration by adsorption to SP-Sephadex in the following manner: the filtrates and 200.ml washes were pooled, diluted to 2 liters with distilled water, and the pH adjusted to 3.0 with acetic acid. This solution was applied to a 5.ml SP-Sephadex column equilibrated in 0.1 M acetic acid and eluted with 20 ml of 0.5 M pyridine/acetate buffer, pH 5.5. The eluate was lyophilized, taken up in 40 ml of 0.05 M phosphate buffer, pH 7.4, and submitted to affinity chromatography as described above. All NaOH washes from the four incubations were combined and radioimmunoassayed as well as the final filtrate and the 200.ml wash. Second Cation Enchange Chromatography-The material recovered in the NaOH washes was diluted to 1 liter with distilled water and the pH was adjusted to 3.0 with acetic acid. This solution was applied to a lo-ml sulfopropyl Sephadex column and was eluted with a 2-liter linear gradient ranging from 0.01 to 0.03 M pyridine/acetate, pH 5.5. The fractions were examined for R-NT and the two active regions PI and PII were pooled separately and lyophilized (Fig. 4a). PI was rechromatographed on SP-Sephadex in the same conditions as described above (Fig. 46).
Preparative Paper Electrophoresis-The lyophilized material corresponding to PI1 (about 0.5 mg of protein) was applied to a lo-cm band to Whatman No. 1MM paper and subjected to high voltage electrophoresis at pH 3.5 (Fig. 5). The band was cut into 2.cm strips. The strips were eluted with 0.1 M acetic acid and the eluates were assayed for R-NT. Yields- Table  I summarizes the results of the isolation of R-NT recovered after electrophoresis. A 250,000-fold purification was achieved with an overall yield of approximately 15% for PII. These results are similar to those obtained for  hypothalamic neurotensin using a somewhat different purification procedure (1). Purification-When chromatographed on Sephadex G-25 ( Fig. 1 and Fig. 3) and on SP-Sephadex (Fig. 2j, R-NT was recovered in a mean peak co-eluting with synthetic neurotensin (indicated by an arrow on the figuresj. Further purification of R-NT was obtained using affinity chromatography.

R-NT Content in Intestinal
When 670 nmol of R-NT were chromatographed in several batches (See "Methods") on the immunosorbent, 400 nmol of R-NT (as measured by either antiserum) were recovered yielding 60% for this purification step. The increase in specific activity, although not measured directly, was estimated to be CU. 50 on the basis of data obtained by chromatography of rat intestinal extracts on the same immunosorbent and in the same conditions as described above. When the material post-affinity chromatography was rechromatographed on SP-Sephadex (Fig. 4~) two peaks of R-NT, PI and PII, were obtained. PI1 (250 nmol) chromatographed in the same regions as synthetic neurotensin while PI (60 nmol) was eluted earlier from the column. Amino acid analysis of PI and PI1 (Table II) indicated that although other residues were present, both peaks possessed similar molar ratios of the constituent amino acids in neurotensin. In addition, when PI was rechromatographed in the conditions of Fig. 4a, the immunoreactivity was recovered in the same region as synthetic neurotensin and PI1 (Fig. 4b).
These results suggest that PI represents the same material as PI1 and that it eluted earlier during the initial ion exchange step because of some artifact of the procedure. Finally, when the material contained in PI1 was submitted to high voltage paper electrophoresis, R-NT was recovered only in a single large peak consisting of four regions A, B, C, and D, with Region C having the same electrophoretic mobility as neurotensin (Fig. 5).
Composition- Table  III shows the amino acid compositions determined for the material in Regions A, C, and D. Regions A, C, and D had identical compositions to that of neurotensin except for the values of the molar ratio of leucine which were 1.8, 1.4, and 1.1, respectively. Other residues such as glycine, serine, and alanine, present in PI1 (Table II) were found in very low concentration and are likely to be contaminants. These results suggest that (a) a tridecapeptide having the same amino acid composition as neurotensin is present in Region A; (b) a do-  (2). The free amino acids released when the materials in Regions A, C, and D were treated with carboxypeptidase A for 3 and 180 min are given in Table IV. Leucine and isoleucine were rapidly released in equal amounts (0.6 residue) from the material in Region A. In contrast, 0.35 and only 0.05 residues of leucine were released from the material in Regions C and D, respectively, even after 180 min, whereas isoleucine was released rapidly from both, 0.7 and 0.6 residues appearing after only 3 min. Tyrosine was released more slowly from the materials in each region, as expected if it is adjacent to a prolyl residue as in neurotensin. These results are consistent with the interpretation stated above (see "Composition").
In addition, they indicate that the COOH-terminal sequence of the tridecapeptide (Region A) is the same as that of neurotensin, whereas the COOH-terminal sequence of the dodecapeptide (Region Dj is Tyr-Ile-OH. This accounts for the difference in the amino acid compositions of Regions A and D. Isolation and Amino Acid Composition of Papain Fragments  of Region C-When neurotensin is digested by the endopeptidase papain, the following fragments are obtained. The NH,terminal Fragment P-l (<Glu-Leu-Tyr-Glu-OH), the middle Fragment P-2 (Asn-Lys-Pro-Arg-OH), and the COOH-terminal Fragment P-3 (Arg-Pro-Tyr-Ile-Leu-OH) (2). Such a digestion was performed on neurotensin and R-NT from Region C (because it contained the most material) and the electrophoretie mobilities of the resulting fragments were compared (Fig.  6). Region C yielded four fragments; three of them, PI-l, P'-2, and PI-3 show the same electrophoretic mobilities as P-l, P-2, and P-3, respectively. P'-4 has a mobility slightly higher than that of P'-3. The amino acid compositions determined for these fragments are shown on Table V and are compared to the theoretical compositions of the corresponding fragments obtained from neurotensin. Near integral molar ratios of the constituent amino acids were obtained for each of the fragments; the compositions of P'-1, P'-2, and P'-3 were similar to those of P-l, P-2 and P-3, respectively. The additional Fragment PI-4 was found to have the composition of P-3 minus a leucyl residue; this might account for its slightly higher electrophoretic mobility (Fig. 6). Approximately 3 times as much P'-4 was obtained as P'-3, which is consistent with the interpretation that the R-NT used to generate those fragments (Region C, Fig. 5) consisted of about 40% tridecapeptide and 60% dodecapeptide, and upon treatment with papain these peptides yielded the COOH-terminal fragments, PI-3 and P'-4, respectively.
Biological with an enzyme to substrate molar ratio of 1:50. Of this material, 35 nmol were applied across 6 cm at the origin. The amount of the different peptides eluted from the paper is indicated in nmol. Other materials were spotted as follows. S, standards; P, papain digest of neurotensin (5 nmol); P', papain digest of Region C (5 nmol). Conditions: pH 6.5, 50 min at 86 V/cm; peptides and standards were stained with the chlorine, o-toluidine method.
temic blood pressure of anesthetized rats are compared in Fig.  7. Synthetic neurotensin and Region A at 500 pmol/kg are comparable in inducing hypotension in the rat. In contrast, Region D at a dose about 3 times higher than that of neurotensin and Region A, shows no effect on the systemic blood pressure of the anesthetized rat. DISCUSSION A tridecapeptide (Region A) with the same amino acid composition as that of neurotensin has been isolated from extracts of bovine intestinal tissues. This peptide has immunological, chromatographic, and electrophoretic properties indistinguishable from those of neurotensin. Carboxypeptidase treatment of the isolated peptide indicated that the COOHterminal sequence was the same as that of neurotensin. In addition, this material was as potent as neurotensin in inducing hypotension in anesthetized rats. This strongly suggests that the intestinal tridecapeptide is identical to hypothalamic neurotensin and it will be referred to as intestinal neurotensin. Another peptide which is likely to be the neurotensin (l-12) dodecapeptide (Region D) has also been isolated and separated from the tridecapeptide by high voltage paper electrophoresis. The finding that this dodecapeptide had no detectable effect on the systemic blood pressure of anesthetized rats confirms the requirement of the COOH-terminal region of neurotensin for its biological activity (7). It is reasonable to assume that this dodecapeptide has been generated from intestinal neurotensin prior to the extraction procedure. The degradation of intestinal neurotensin by carboxypeptidases might possibly be the mechanism by which the intestinal neurotensin is biologically inactivated in uioo. The heterogeneity of R-NT was not indicated by the radioimmunoassays performed during the purification procedure. When the materials isolated in Regions A and D were carefully assayed with both poly(Glu6', Lys"')-4 and poly(Glu6', Lys'O)-6 antisera, it was found that the tridecapeptide (Region A) had the same immunological activity as neurotensin, whereas the  (3,9) exhibits several pharmacological properties similar to those of neurotensin, i.e. both induce systemic vasodilation, hypotension, hyperglycemia, and contraction of ileal smooth muscle (10-12).
Another possibility is suggested from what is known about the sialogogic peptide, Substance P. Substance P has been isolated and characterized from both hypothalamic (13) and intestinal tissues (14), a situation which resembles that of neurotensin and intestinal neurotensin. Studies on the localization of Substance P in the gastrointestinal tract indicate that the peptide is mainly localized to the nerve plexuses innervating the gastrointestinal tract (15). In addition, Substance P may participate in chemical neurotransmission (16). Similar studies should be valuable in elucidating the role of both neurotensin and intestinal neurotensin.