Secretin self-assembles and interacts spontaneously with phospholipids in vitro

Peptides 23 (2002) 201–204

Secretin self-assembles and interacts spontaneously with phospholipids in vitro Salil Gandhia, Israel Rubinsteinb,c,d, Takaya Tsueshitac, Hayat Onyuksela,b,* b

a Department of Bioengineering, University of Illinois at Chicago, Chicago, IL 60612, USA Department of Pharmaceutics and Pharmacodynamics, University of Illinois at Chicago, Chicago, IL 60612, USA c Department of Medicine University of Illinois at Chicago, Chicago, IL 60612, USA d Chicago VA Health Care System West Side Division, Chicago, IL 60612, USA

Received 9 July 2001; accepted 10 September 2001

Abstract Secretin, a 27-amino acid neuropeptide, is a member of the secretin/glucagon/vasoactive intestinal polypeptide (VIP) superfamily of amphipathic peptides. The peptide modulates gastrointestinal and neuronal function and is currently being evaluated for the treatment of autism. However, as most peptides, it has a short circulation half-life. Previously, we have shown that VIP self-assembles in aqueous environment and interacts with a biomimetic phospholipid membrane. These in vitro characteristics increase VIP half-life and bioactivity in vivo. The purpose of this study was to investigate whether secretin exhibits similar properties in vitro by forming micelles in aqueous solution and interacting with phospholipids. Results of this study demonstrated that secretin self-assembles to form micelles in HEPES buffer at 25°C above ⬃0.4 ␮M. Additionally, secretin interacts with a biomimetic phospholipid membrane as indicated from a significant increase in membrane surface pressure (from 25.5 ⫾ 1.3 to 32.5 ⫾ 3.0, P ⬍ 0.05). Importantly, the peptide undergoes conformational transition from predominantly random coil in saline to ␣-helix in the presence of phospholipid, distearoyl-phosphatidylcholine-poly(ethylene) glycol (mol mass 2000) micelles. We suggest that these distinct biophysical attributes could modulate secretin bioactivity in vivo. © 2002 Elsevier Science Inc. All rights reserved. Keywords: Surface tension; Monolayer; Critical micellar concentration; Micelle; Circular dichroism; Neuropeptide; Vasopressin; DSPE-PEG

1. Introduction Secretin, a 27-amino acid neuropeptide, is a member of the secretin/glucagon/vasoactive intestinal polypeptide (VIP) superfamily of amphipathic peptides [3,4]. The peptide modulates gastrointestinal and neuronal function [1,5, 12]. It is currently being evaluated for the treatment of autism. However, as most peptides, it has a short circulation half-life [19]. It is well established that amphipathic peptides undergo conformational transition from predominantly random coil in aqueous solutions to ␣-helix in organic solvents and lipids [11,16]. However, the biological relevance of this phenomenon to improve secretin’s circulation half-life and bioactivity is uncertain. Previous work from our laboratory showed that VIP aggregates in aqueous solution, expresses surface-active

properties and undergoes conformational transition from predominantly random coil in aqueous solutions to ␣-helix in the presence of sterically stabilized phospholipid micelles (SSM) [14,17]. These distinct physico-chemical properties amplified VIP bioactivity in vivo [15,18]. Whether secretin expresses similar biophysical characteristics in the presence of phospholipids is uncertain. Hence, the purpose of this study was to begin to address these issues by determining whether human secretin aggregates in an aqueous solution, and moreover if it interacts with a biomimetic phospholipid monolayer and micelles leading to changes in membrane surface pressure and peptide conformation, respectively. 2. Methods 2.1. Critical micellar concentration (CMC) of secretin

* Corresponding author. Tel.: ⫹1-312-996-2097; fax: ⫹1-312-9960098. E-mail address: [email protected] (H. Onyuksel).

First, the surface tension of HEPES buffer (pH 7.4) contained in a 30-ml custom-made Teflon trough was de-

0196-9781/02/$ – see front matter © 2002 Elsevier Science Inc. All rights reserved. PII: S 0 1 9 6 - 9 7 8 1 ( 0 1 ) 0 0 5 9 6 - 4

202

S. Gandhi et al. / Peptides 23 (2002) 201–204

termined at room temperature (25°C) using a DuNouy surface tensiometer (Fisher Model 21 Tensiomat, Pittsburgh, PA, USA) [14]. Then, 50 ␮l of increasing concentrations of secretin (0.1–1.0 ␮M final concentrations in the trough) dissolved in HEPES buffer (pH 7.4) was injected into the container and surface tensions of the secretin solutions were determined. The concentration above which surface tension no longer significantly decreased was considered as the CMC of secretin [14] 2.2. Effects of secretin on surface pressure Dipalmitoylphosphatidylcholine (DPPC; 10 ␮M) was dissolved in hexane/ethanol (9:1 v/v) and 50 ␮l of the solution were spread over 28 ml HEPES buffer contained in a cylindrical 30-ml custom-made Teflon trough (9 cm diameter) at room temperature (25°C). A DPPC monolayer is formed spontaneously at the buffer/air interface after total evaporation of the organic solvent in 5 min. Surface tension of the monolayer was determined every 5 min for 60 min and every 10 min for 60 min thereafter using a DuNouy surface tensiometer. Surface pressure was defined as the difference in surface tension between the DPPC monolayer and HEPES buffer alone, as described before [21]. Secretin (1.5 ␮M) was injected into the subphase without disturbing the monolayer. Surface tension of mixed monolayer (DPPC with secretin) was measured as outlined above for DPPC monolayer. A nonspecific peptide, [8Arg]-vasopressin (1.5 ␮M) is used as a control [14,20] 2.3. Conformation of secretin in saline and SSM Secretin in phospholipid micelles was prepared using a method previously described in our laboratory [14]. Briefly, poly(ethylene) glycol (mol mass 2000) covalently linked to distearoyl-phosphatidylcholine (DSPE-PEG; 1.0 ␮M) was dissolved in saline and mixed to form phospholipid micelles. The resulting SSM suspension was then incubated with secretin for 2 h at room temperature (25°C) before use. Size of SSM and secretin in SSM was 18 ⫾ 2 nm as determined by quasi-elastic light scattering (Model 380, Nicomp Submicron Particle Sizer; Pacific Scientific, Menlo Park, CA, USA) The secondary structure of secretin was determined by circular dichroism. Spectra were recorded on a JASCO J-700 spectropolarimeter at room temperature (25°C) using a fused quartz cell of 1 cm path length containing saline with secretin (4 ␮M) and secretin (4 ␮M) incubated with SSM (1 mM) for 30 min. A bandwidth of 1.0 nm and a step resolution of 0.5 nm were used to collect an average of 5 accumulations/sample at the near-UV range (190 –260 nm wavelength). The acquired spectra were corrected to the baseline by using saline and empty SSM, and averaged. The peptide spectra were smoothed using the noise reduction function. Data are expressed as % ␣ helix using the equation, % helicity ⫽ [-(␪ ⫹ 4000)/29000] X 100 (␪ defined as

Fig. 1. Critical micellar concentration (CMC) determination of secretin from surface tension measurements. Values are means ⫾ SEM; n ⫽ 3.

ellipticity) and calculated by software Selcon, Softsec ver 1.2 [Softwood, Brookfield, CT] [10]. The concentrations of aqueous secretin and secretin in SSM used in these experiments are based on previous studies in our laboratory and are consistent with concentrations of secretin detected in plasma and tissues [14,17,19]. 2.4. Chemicals and drugs Secretin was obtained from American Peptide Company (Sunnyvale, CA, USA). [8Arg]-vasopressin and HEPES were obtained from Sigma Chemical Co. (St. Louis, MO, USA). DPPC was obtained from Avanti Polar Lipids Inc. (Alabaster, AL, USA). All drugs were prepared and diluted in saline to the desired concentrations on the day of the experiment 2.5. Data and statistical analyses Data are expressed as means ⫾ SEM. Statistical analysis was performed using repeated measures analysis of variance with Neuman-Keuls multiple-range post hoc test to detect values that were different from control values. A P-value ⬍ 0.05 was considered statistically significant

3. Results 3.1. Critical micellar concentration of secretin The effect of secretin on surface tension of the HEPES buffer (pH 7.4) is depicted in Fig. 1. There was a significant

S. Gandhi et al. / Peptides 23 (2002) 201–204

203

Fig. 3. Secondary structure of secretin by CD in saline and in phospholipid micelles. Units for CD are arbitrary.

Fig. 2. Effects of secretin (1.5 ␮m) and [8Arg]-vasopressin (VP; 1.5 ␮m) on surface pressure of a dipalmitoylphosphatidylcholine (DPPC; 10 ␮m) monolayer. Values are means ⫾ SEM; each group, n ⫽ 3. *P ⬍ 0.05 in comparison to DPPC alone. ⫹ P ⬍ 0.05 in comparison to [8Arg]-vasopressin.

concentration-dependent decrease in surface tension from baseline when secretin was injected into the sub-phase (Fig. 1; n ⫽ 3; P ⬍ 0.05). No change in surface tension was recorded above 0.4 ␮M (Fig. 1). This concentration was considered as the CMC of secretin. The actual size of secretin micelles could not be determined because it was below the lower detection limit of the quasi-elastic light scattering apparatus (data not shown).

3.2. Effects of secretin on membrane surface pressure The effect of secretin on surface pressure of a DPPC monolayer is displayed in Fig. 2. The DPPC monolayer alone is unstable at the air/water interface and decomposes shortly after spreading leading to a sustained decline in surface pressure (Fig. 2; n ⫽ 3; P ⬍ 0.05). This may be due to decomposition of the phospholipid by oxidation and hydrolysis. By contrast, injection of secretin into the subphase stabilizes the monolayer as manifested by an initial increase in surface pressure, followed by sustained constant surface pressure of the monolayer. The increase in the surface pressure during the first 20 min of the experiment is the indication of the incorporation of the peptide into the DPPC monolayer over the entire observation period (Fig. 2; each group, n ⫽ 3; P ⬍ 0.05). Unlike secretin, injection of [8Arg]-vasopressin at equimolar concentration (1.5 ␮M) into the sub-phase only abrogated the time-dependent decline in surface pressure (Fig. 2; n ⫽ 3; P ⬍ 0.05), suggesting that VP, unlike secretin, is not incorporated into the phospholipid monolayer but just interacts on the aqueous phospholipid interface to decrease phospholipid decomposition

3.3. Conformation of secretin in saline and SSM As shown in Fig. 3, the conformation of secretin in saline at room temperature was unordered (␣ helix content, 4.0 ⫾ 2.0%; n ⫽ 3). By contrast, secretin in SSM dispersed in saline at room temperature exhibited a significant increase in the proportion of ␣ helix (35.0 ⫾ 6.4%; n ⫽ 3; P ⬍ 0.05 relative to aqueous secretin).

4. Discussion There are three new findings of this study. First, we found that secretin, an amphipathic neuropeptide with a calculated net charge of ⫹1, self-assembles in aqueous solution in vitro. The critical micellar concentration of secretin, ⬃0.4 ␮M, was similar to that of VIP, a 28-amino acid neuropeptide with a calculated net charge of ⫹3 [14]. Second, secretin, at a concentration above CMC, significantly increased the surface pressure of a DPPC monolayer indicating that the peptide interacts directly with phospholipids. The magnitude of this response was similar to that elicited by VIP [14]. The effects of secretin on surface pressure were not related to nonspecific peptide interactions with phospholipids because vasopressin, a structurally unrelated peptide with water solubility higher than that of secretin, had only a modest effect on the stability of the DPPC monolayer. We used [8Arg]-vasopressin as a control because Woodle et al. showed that encapsulating [8Arg]vasopressin in sterically stabilized liposomes prolongs the bioactive effects of the peptide [20]. Lastly, association of secretin with SSM evoked a significant conformational transition of the molecule from predominantly random coil in saline to ␣-helix, the preferred conformation for ligand-receptor interactions [6,16]. Collectively, these data indicate that secretin, like VIP, expresses distinct surface active properties in vitro [2,9,17,21]. Unlike vasopressin, secretin is attracted to and inserts itself into a biomimetic phospholipid monolayer, most likely, by electrostatic and hydrophobic interactions leading to an in-

204

S. Gandhi et al. / Peptides 23 (2002) 201–204

crease in surface pressure. Conceivably, this process alters membrane fluidity and peptide conformation [9,13,17]. The relevance of these novel observations to secretin bioactivity was not explored in this study [3–5]. Nonetheless, these data are consistent with the hypothesis that under certain circumstances, the presence of phospholipids in the extracellular space could constrain secretin into a metastable aggregate (micelles). This phenomenon could then stabilize the peptide and protect it from hydrolytic attack and enzymatic degradation in vivo [9]. Consequently, target cells may be exposed to micellar secretin for a prolonged period of time relative to the monomeric form of the peptide. This, in turn, may amplify secretin bioactivity as we have observed with human vasoactive intestinal peptide (VIP), a member of the secretin/glucagon/VIP superfamily of amphipathic peptides [8,9,15,17]. We have previously shown that VIP interacts with SSM at room temperature in vitro and evokes conformational transition from random coil to ␣-helix. When these experiments were repeated at 37°C, ␣-helical content of the system was increased [17]. We also conducted in vivo experiments using ␣-helix VIP and demonstrated clearly enhanced bioactivity of VIP in the presence of phospholipids [15]. Furthermore, preliminary in vivo experiments with phospholipid-associated secretin at 37°C indicate that secretin bioactivity, in fact, is increased with phospholipid association [7]. The results of this study also suggest that secretin in SSM could interact directly with plasma membrane phospholipids in target cells in a receptor-independent fashion thereby activating intracellular signal transduction pathway(s). The above notwithstanding, conformational transition of secretin molecules in the presence of phospholipids to predominantly ␣-helix may further stabilize the peptide, protect it from hydrolysis and proteolytic inactivation and promote secretin-receptor interactions in target cells [9,15, 17]. Clearly, additional studies using molecular, biophysical, and cell biology techniques are warranted to support or refute these hypotheses. In summary, we found that secretin self-assembles in aqueous solution. In addition, the peptide interacts with a biomimetic phospholipid membrane resulting in a sustained increase in membrane surface pressure. Importantly, interactions of secretin with phospholipid molecules lead to conformational transition of the molecule to ␣-helix relative to disordered, aqueous secretin. We suggest that these distinct biophysical attributes could modulate secretin bioactivity in vivo.

Acknowledgments This study was supported, in part, by the Campus Research Board, University of Illinois at Chicago.

References [1] Bayliss WM, Starling EH. The mechanism of pancreatic secretion. J Physiol 1902;28:325–53. [2] Bodanszky M, Bodanszky A, Klausner YS, Said SI. A preferred conformation in the vasoactive intestinal peptide (VIP). Molecular architecture of gastrointestinal hormones. Bioorgan Chem 1974;3: 133– 40. [3] Carpenter KA, Schiller PW. Aggregation Behaviour and Zn2⫹ Binding Properties of Secretin. Biochemistry 1998;37:16969 –74. [4] Chang CH, Chey WY, Erway B, Coy DH, Chang TM. Modulation of secretin release by neuropeptides in secretin-producing cells. Am J Physiol 1998;275:G192–202. [5] Chez MG, Buchanan CP, Bagan BT, Hammer MS, McCarthy KS, Ovrutskaya I, Nowinski CV, Cohen ZS. Secretin and autism: a twopart clinical investigation. J Autism Dev Disord 2000;30:87–94. [6] Filizola M, Carteni-Farina M, Perez JJ. Conformational study of vasoactive intestinal peptide by computational methods. J Pept Res 1997;50:55– 64. [7] Gandhi S, Onyuksel H, Tsueshita T, Rubinstein I. Increased bioactivity of secretin in sterically stabilized micelles: Implication for therapy. Int Symp Controlled Release 2001;28:572–3. [8] Gao X, Noda Y, Rubinstein I, Paul S. Vasoactive intestinal peptide encapsulated in liposomes: Effects on systemic arterial blood pressure. Life Sci 1994;54:PL247-PL252. [9] Gololobov G, Noda Y, Sherman S, Rubinstein I, BaranowskaKortylewicz J, Paul S. Stabilization of vasoactive intestinal peptide by lipids. J Pharmacol Exp Ther 1998;285:753– 8. [10] Haghjoo K, Cash PW, Farid RS, Komisaruk BR, Jordan F, Pochapsky SS. Solution structure of vasoactive intestinal peptide (11–28)-NH2, a fragment with analgesic properties. Pept Res 1996;9:327–31. [11] Lee S, Iwata T, Oyagi H, Aoyagi H, Ohno M, Anaie K, Kirino Y, Sugihara G. Effects of salts on conformational changes of basic amphipathic peptides from ␤-structure to ␣-helix in the presence of phospholipid liposomes and their channel-forming ability. Biochim Biophys Acta 1993;1151:76 – 82. [12] Li P, Song Y, Lee KY, Chang TM, Chey WY. A secretin releasing peptide exists in dog pancreatic juice. Life Sci 2000;66:1307–16. [13] Noda Y, Rodriguez-Sierra J, Liu J, Landers D, Mori A, Paul S. Partitioning of vasoactive intestinal polypeptide into lipid bilayers. Biochem Biophys Acta 1994;1191:324 –30. [14] Onyuksel H, Bodalia B, Sethi V, Dagar S, Rubinstein I. Surfaceactive properties of vasoactive intestinal peptide. Peptides 2000;21: 419 –23. [15] Onyuksel H, Ikezaki H, Patel M, Rubinstein I. A novel formulation of VIP in sterically stabilized micelles amplifies vasodilation in vivo. Pharm Res 1999;16:155– 60. [16] Robinson RM, Blakeney EW Jr., Mattice WL. Lipid-induced conformational changes in glucagons, secretin, and vasoactive intestinal peptide. Biopolymers 1982;21:1217–28. [17] Rubinstein I, Patel M, Ikezaki H, Dagar S, Onyuksel H. Conformation and Vasoreactivity of VIP in phospholipids: effects of calmodulin. Peptides 2000;20:1497–501. [18] Sejourne F, Rubinstein I, Suzuki H, Alkan-Onyuksel H. Development of a novel bioactive formulation of vasoactive intestinal peptide in sterically stabilized liposomes. Pharm Res 1997;14:362–5. [19] Schutt C. Newsletter Nat. Alliance for Autism Res. 4(1998) 21–25. [20] Woodle MC, Storm G, Newman MS, Jekot JJ, Collins LR, Martin FJ, Szoka FC Jr. Prolonged systemic delivery of peptide drugs by long circulating liposomes: illustration with vasopressin in the Brattleboro rat. Pharm Res 1992;9:260 –5. [21] Xia WJ, Onyuksel H. Mechanistic studies on surfactant-induced membrane permeability enhancement. Pharm Res. 2000;17:612– 8.