Clustered versus Uniform Display of GALA-Peptides on Carrier

Nov 2, 2017 - In the presence of fluid lipid membranes, the GALA-helix exhibits membrane-active properties that originate from the same hydrophobic fa...
0 downloads 13 Views 1MB Size
Subscriber access provided by READING UNIV

Article

Clustered vs. uniform display of GALA-peptides on carrier nanoparticles: enhancing the permeation of non-charged fluid lipid-membranes Trevan Locke, and Stavroula Sofou Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.7b03706 • Publication Date (Web): 02 Nov 2017 Downloaded from http://pubs.acs.org on November 3, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers

Langmuir is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Subscriber access provided by READING UNIV

and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Langmuir is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

Langmuir

pH ~ 7.4

pH ~ 5.5

GALA peptide

O

N H

O

COOH

COOH

NH H N

O N H

H N O

O N H

H N O

O N H

H N O

O N H

H N O

O N H

H N O

COOH

O N H

H N O

O N H

H N

N H

O

NH O O P O O

H N O

O N H

H N O

COOH

O N H

H N O

N

COOH

O

COOH

O

N H

Na

H O O

O

O

ACS Paragon Plus Environment

O N H

H N O

O N H

H N O

O N H

H N O

COOH

O N H

H N O

O NH2

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Clustered vs. uniform display of GALA-peptides on carrier nanoparticles: enhancing the permeation of non-charged fluid lipid-membranes

Trevan Locke,1 Stavroula Sofou1,2,3, *

1

Departments of Chemical and Biochemical Engineering, and 2Biomedical Engineering,

3

The Rutgers Center for Lipid Research, Rutgers University, 599 Taylor Road,

Piscataway, NJ 08854

* corresponding author: current address: Stavroula Sofou, 116 Maryland Hall, Department of Chemical and Biomolecular Engineering, Johns Hopkins University, 3400 North Charles Street, Baltimore, MD 21218 phone (410) 516-0274; fax (410) 516-5510; email address: [email protected]

Keywords: clustered peptide display, endosomal escape, GALA peptide, phaseseparating lipid membranes, membrane activity

1 ACS Paragon Plus Environment

Page 2 of 32

Page 3 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Abstract GALA-peptide is a random coil in neutral pH; in acidic pH it becomes an amphipathic alpha helix that aggregates in solution, possibly via its hydrophobic facet, that runs along the helix's long axis. In the presence of fluid lipid membranes, the GALA-helix exhibits membraneactive properties that originate from the same hydrophobic facet; these properties make GALA a candidate for inclusion in drug delivery systems requiring permeation of the endosomal membrane to enable drug escape into the cytoplasm. Previous work has shown that uniform functionalization of carrier nanoparticles with GALA-peptides improved their membrane activity and enhanced the endosomal escape of delivered therapeutics. The present study aims to evaluate the potential role of altering membrane activity via cluster-displayed GALA-peptides (for higher local valency) on the surface of carrier nanoparticles. The presentation of GALA-peptides on carrier nanoparticles was designed to also be pH-dependent. The peptide display on the surface of the carrier nanoparticles was uniform in neutral pH; in the acidic endosomal pH, the surface of nanocarriers formed domains (patches) with high local densities of GALA-peptides. The interactions between GALA-functionalized carrier nanoparticles and target lipid vesicles, utilized as endosome membrane surrogates which were used to primarily capture the fluid nature of these membranes, were studied as a function of pH. At endosomal pH values, ranging from 5.5 to 5.0, greatest permeability of target membranes was induced by nanocarriers with clustered and not with uniformly displayed GALA. This enhancing effect had an optimum; at even more acidic pH values, too close approximation of GALA peptides residing within the same patches resulted in preferential intrapatch peptide interactions rather than interactions with the apposing target lipid membranes. This behavior could have the same physicochemical origin

2 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

as the aforementioned GALA-peptide aggregation, observed in solution with lowering pH at increasing peptide concentrations. The findings of this study support the potential of utilizing the clustered display of GALA-peptides on carrier nanoparticles to increase the permeation of fluid membranes used herein to capture this critical physical property of endosomal membranes, and to, therefore, ultimately improve the endosomal escape of delivered therapeutics, enhancing therapeutic efficacy.

3 ACS Paragon Plus Environment

Page 4 of 32

Page 5 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

1. Introduction The present study aims to evaluate the role of display (clustered vs. uniform) of nanoparticle-conjugated GALA-peptides on affecting their membrane activity; i.e. their property of increasing the permeability of endosome surrogate target membranes. GALA is a pHresponsive, thirty amino acid peptide (WEAALA(EALA)2EHLA(EALA)3A)1 whose design was inspired by viral fusion proteins sequences.1, 2 GALA-peptide is a random coil in neutral pH, and in acidic conditions, protonation of the repeated glutamic acid residues, allows the formation of an amphipathic alpha helix. Once formed, the alpha helix has been shown to associate with and to affect the permeability of fluid-phase lipid bilayers. These properties make the GALA-peptide a candidate for inclusion in drug delivery systems that require endosomal escape.3, 4, 5 Previous work has shown that nanoparticles with uniform functionalization with GALA-peptides improved the activity against fluid membranes,

6

enhanced the endosomal escape of delivered

nucleotide contents,7, 8, 9, 10, 11 and increased the transfection efficiency on mammalian cells in culture by 2-3 orders of magnitude.12 The present study aims to evaluate the potential role of clustered GALA-peptides on the surface of carrier nanoparticles in improving permeation of endosome surrogate target membranes. The implications of this design could potentially improve the endosomal escape of delivered therapeutics enhancing, therefore, the therapeutic efficacy.

The patterns of ligand display on the surface of functionalized nanoparticles play a critical role in defining the geometries of ligand-target interactions, and have been extensively studied; for drug delivery applications, usually the aim involves the increase of affinities of functionalized nanoparticles for their targets.13 A popular approach towards this goal has been the chemical organization of ligands in molecular-level clusters14, 15 grafted on the surface of the nanoparticles. These chemically organized clusters of ligands, or 'patches', are thought to be of 4 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 32

molecular dimensions (~2-10 ligands), uniformly distributed over the surface of the carrier nanoparticles and covering several thousands of nm2 (Figure S1 A). Sempkowski et al16 has recently reported a clustered display of ligands on larger 'patches' having sizes occupying a significant fraction (10 to 30 %) of the nanoparticle's total surface (Figure S1 B). These larger patches, containing high local densities of ligands, were demonstrated to enhance the affinity of functionalized nanocarriers to levels not observed by nanocarriers functionalized with the uniformly distributed molecular clusters of ligands.17 In contrast to the uniform surface distribution of chemically organized clusters of ligands, the larger patches form highly heterogeneous nanoparticle surfaces containing a significantly large, densely functionalized patch area of a few thousands of nm2-in-size18 on a nanoparticle surface of otherwise essentially minimal functionalization.

In this study, GALA-peptide was incorporated into gel-phase lipid vesicles (the carrier nanoparticles) in the form of GALA-functionalized lipids. To minimize interactions of the GALA-peptide with the underlying carrier vesicle membranes, the latter were designed to be in the gel-phase at working temperatures. In addition to the pH-responsive conformation of GALApeptide, the underlying lipid membrane patches within the carrier lipid nanoparticles were also designed to be pH-responsive. Briefly, as reported before,19,

20

the carrier nanoparticles' lipid

membranes consist of two types of lipids with different lengths of saturated acyl-tails and with the following headgroups (Figure 1A): a first type of lipid with titratable headgroup (in this study, phosphatidic acid with effective pKa close to 5.0,21 and a second type of lipid with nontitratable, within the pH range of interest, headgroup such as a phosphatidyl choline). The basic idea describing the mechanism of phase-separation and patch formation is the following: in

5 ACS Paragon Plus Environment

Page 7 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

neutral pH, the electrostatic repulsion among negatively charged lipids is expected to largely prevail, forming, therefore, well-mixed lipid bilayer membranes. In acidic pH, the titratable headgroups become protonated, and the intrinsic hydrogen-bonding among phosphatidic acids22 is expected to act as a major attractive force resulting in phase-separated lipid domains (patches).20 To tune the GALA-peptide display on the surface of these nanovesicles, GALApeptide was conjugated on the headgroup of lipids with acyl-tail lengths identical to those of the titratable lipid type (Figure 1C). In neutral pH, the distribution of GALA-functionalized lipids within the vesicles' membrane is relatively uniform. In acidic pH, following formation of lipid patches, preferential partition of the GALA-functionalized lipids, due to the hydrophobic-phase acyl-tail matching, results in higher local densities, and, therefore, in the clustered peptide display.16

Figure 1. (A) Tunable surface topography of gel-phase lipid bilayers comprising the carrier nanoparticles, and of GALA-display using pH as a trigger. The upper (lower) leaflet represents the outer (inner) leaflet of carrier lipid nanoparticles. The co-localization of the GALAfunctionalized lipids (in green/blue) depends on the extent of phase separation of the underlying carrier lipid membrane. (see main text for description of utilization of molecular lipid-lipid 6 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

interactions as a function of pH to form lipid phase-separated domains - shown in blue - and to display GALA-functionalized lipids in clusters). Figure 1. (B) GALA-peptide (in green) is designed to switch conformation from random coil in neutral pH to an amphipathic alpha-helix at pH ~ 5.5. Figure 1. (C) Chemical structure of the GALA-functionalized lipid.

In the present study, the membrane activity of nanoparticle-displayed GALA-peptides was evaluated by monitoring the permeability induced on fluid (target) membranes which were chosen so as to capture this critical physical property of endosome membranes23 as a function of pH. The binding and relative orientation of GALA-peptides with all types of lipid membranes (carrier- and target-lipid vesicles), as well as the interactions between peptides and between lipid vesicles vs. pH were systematically studied in an effort to characterize the observed membrane activities.

7 ACS Paragon Plus Environment

Page 8 of 32

Page 9 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

2. Experimental Section A. Materials All lipids were obtained from Avanti Polar Lipids (Alabaster, AL) and included 1,2distearoyl-sn-glycero-3-phosphocholine (18:0-PC), 1,2-dibehenoyl-sn-glycero-3-phosphocholine (22:0-PC), 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (16:0-PC), L-α-phosphatidylcholine (Egg, Chicken) (EggPC), 1,2-dimyristoyl-sn-glycero-3-phosphate (sodium salt) (14:0-PA), 1,2dipalmitoyl-sn-glycero-3-phosphate

(sodium

salt)

(16:0-PA),

1,2-distearoyl-sn-glycero-3-

phosphate (sodium salt) (18:0-PA), 1,2-dipalmitoyl-sn-glycero-3-phosphoethanolamine-N(succinyl) (sodium salt) (16:0-PE-succinyl), 1,2-distearoyl-sn-glycero-3-phosphoethanolamineN-[amino(polyethylene glycol)-2000]- (ammonium salt) (18:0-PE-PEG), 1,2-dipalmitoyl-snglycero-3-phosphoethanolamine-N-(lissamine rhodamine B sulfonyl) (ammonium salt) (16:0PE-RhD),

and

1,2-dipalmitoyl-sn-glycero-3-phosphoethanolamine-N-(7-nitro-2-1,3-

benzoxadiazol-4-yl) (ammonium salt) (16:0-PE-NBD). GALA was custom synthesized on Rink Amide resin from Anaspec (Fremont, CA). Cholesterol, methanol, chloroform, Sephadex G-50 resin,

calcein,

phosphate

buffered

saline

(PBS)

tablets,

Triton-X

100,

1-

[Bis(dimethylamino)methylene]-1H-1,2,3-triazolo[4,5-b]pyridinium 3-oxid hexafluorophosphate (HATU), N,N-Diisopropylethylamine (DIPEA), petrol ether, ethyl either, N-Methyl-2pyrrolidone (NMP) and dichloromethane (DCM) were purchased from Sigma-Aldrich (Atlanta, GA).

Trifluoroacetic

Triisopropylsilane

acid

(TIPS)

(TFA), were

2,2′-(Ethylenedioxy)diethanethiol

purchased

from

Sigma-Aldrich

(DODT), (Atlanta,

and GA).

Ethylenediaminetetraacetic acid (EDTA) was purchased from ThermoFisher Scientific (Waltham, MA).

8 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 32

B. GALA-lipid conjugation GALA-peptide, > 95 % pure as reported by Anaspec based on HPLC and MALDI-TOF spectrometry, was conjugated to 16:0-PE-succinyl via its N-terminus. Resin bound peptide, lipid, HATU, and DIPEA were mixed in a 1:2:2:4 molar ratio using 5 mL NMP as the solvent. This mixture was allowed to react for 8 hours at room temperature under agitation. The solvent was then drained and rinsed with NMP four times and then with DCM once (5 mL washes) and the resulting conjugate was cleaved from the resin with a cocktail of TFA/water/TIPS/DODT at a 94:2:2:2 mole ratio. After one hour at room temperature, the cleavage cocktail was added to a cold mixture of 40 mL of 1:1 (v/v) ethyl ether:petrol ether. This mixture was centrifuged at 1,700 RCF for 15 minutes, the supernatant was removed, and 40 mL fresh ether was added. Following a second centrifugation and drying under vacuum, the peptide precipitate was dissolved in 40 mL 10% acetic acid. After 10 minutes, the dissolved lipopeptide was frozen and freeze dried. The resulting material was purified in a 7.8 mm x 300 mm C4 column using HPLC (Waters, Milford, MA) with an 85%/15% H2O/ACN to 100% ACN gradient over 1 hour at a flow rate of 5 mL/min equipped with a Waters 2487 Dual λ Absorbance Detector (Waters, Milford, MA). The molecular weight of GALA-functionalized lipid was confirmed via MALDI-TOF spectrometry, using a SCIEX 4800 (Applied Biosystems/MDS SCIEX, Concord, ON, Canada).

C. Lipid Vesicle Preparation and Characterization Lipid vesicles were formed using the thin film hydration method.21 Briefly, lipids (5 mM total lipid) were hydrated (with either PBS, calcein 75 mM, or 1 mM MOPS buffered sucrose, all at pH 7.4 and 300 mOsm), annealed, and extruded 21 times through two stacked 100 nmdiameter polycarbonate membranes at temperatures at least 10oC above the highest transition

9 ACS Paragon Plus Environment

Page 11 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

temperature of the constituent lipids. Then, non-encapsulated contents were removed by size exclusion chromatography with Sephadex G50 eluted with PBS at pH 7.4 and 300 mOsm. Vesicle size distribution was measured using the Zetasizer Nano ZS90 (Malvern Instruments Ltd., Worcestershire, U.K.). Transition temperatures in lipid membranes were determined using differential scanning calorimetry (DSC) by scanning samples (2.5 mM total lipid) from 20 to 80 °C at a scan rate of 8 °C/hour in a Microcal VP-DSC (Malvern Instruments Ltd., Worcestershire, U.K.). Lipid vesicles with uniform and clustered display of GALA-lipids (carrier vesicles) were composed

of

18:0-PC:chol:18:0-PE-PEG:GALA-16:0-PE-lipid

and

18:0-PC:XX-

PA:chol:18PE-PEG:GALA-16PE-lipid at 90:5:3:2 and 63:27:5:3:2 mole ratio, respectively, where XX was either 14:0, 16:0, or 18:0. The above lipid membranes are primarily in gel-phase at the working temperatures of this study. Lipid vesicles (target vesicles) used to capture the fluid-phase characteristic of endosomal membranes were composed of EggPC:chol at 7:3 mole ratio. This composition was chosen to provide a fluid-phase lipid bilayer independent of the working temperature(s) that would exclude explicit charge effects - since headgroups are zwitterionic - to avoid related electrostatic interactions with the GALA-peptide displayed by the carrier lipid vesicles. Target vesicles were not PEGylated to eliminate interference of steric effects on the above interactions. Variations in the lipid compositions of the carrier vesicles, without 18:0-PE-PEG lipid and/or with varying amounts of GALA-16:0-PE-lipid, were also used as indicated. A summary of membrane compositions used in this study is provided in Table 1.

D. Content Release Studies To evaluate the ability of free GALA-peptide or GALA-lipid functionalized carrier vesicles to cause content release from fluid-bilayer target vesicles, the self-quenching relief of

10 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 32

calcein (ex/em 495/515 nm using a Fluorolog-3-22, Horiba Scientific, Edison, NJ) encapsulated in endosomal-membrane analogue vesicles was monitored.19

Change in self-quenching

efficiency, Qt, over time within the pH range from 7.4 to 4.0 was monitored for calceincontaining target vesicles incubated with free GALA-peptide or GALA-functionalized carrier vesicles at 1:25,000 GALA: target lipid mole ratio or 1:100 GALA-lipid: target lipid mole ratio, respectively. Qt was defined as the ratio of the completely relieved fluorescence intensity divided by self-quenched fluorescence intensity at each time point. Triton-X 100 (5% w/v) was added to completely relieve fluorescent quenching. The normalized % change in quenching efficiency was calculated as (Qt-1)/(Qmax-1)*100, where Qmax was the quenching efficiency at t = 0. The rate of release was fitted with a single exponential decay in time. To monitor the effects of GALA-lipid on altering the permeability of the membranes of the supporting carrier vesicles, similar studies were conducted on relief of self-quenching efficiency of calcein encapsulated in carrier vesicles functionalized with GALA-lipid in the absence and presence of target vesicles which did not contain calcein. In all the above studies, pH was adjusted with HCl.

E. Free GALA-peptide binding to Lipid Membranes Lipid vesicles encapsulating 1 mM MOPS and sucrose (300 mOsm) were incubated with free GALA-peptide at approximately 1:100 free GALA peptide:total lipid mole ratio in PBS at different pH values for two hours at 37 °C. Unbound free GALA-peptide was separated by centrifugation at 154,000 x g for 90 minutes at 22 °C. The supernatant, ninety percent of the total volume, was carefully removed using a Pasteur pipette. The remaining volume was removed with the pellet. For quantitation assays, the pH on both fractions was restored to 7.4.

11 ACS Paragon Plus Environment

Page 13 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

On each fraction, lipid content was determined via Stewart’s Assay.24 Peptide content was determined by measuring tryptophan's fluorescence (ex/em: 290/356). Briefly, samples were mixed with ACN to reach a final v/v ratio of 2:1 PBS buffer:ACN, and samples' tryptophan concentration was quantified using a standard curve. Peptide- and lipid-content on the supernatant and pellet were corrected for the 10% liquid volume that was removed with the pellet.

F. Peptide Characterization The fluorescence spectra of the tryptophan residue were acquired in order to investigate tryptophan's local environment upon free GALA-peptide or GALA-functionalized lipid interaction with lipid membranes (ex 290 nm/ em 310-400 nm).

To evaluate the alpha-helicity of GALA, circular dichroism (CD) spectra of samples in PBS at variable pH values (in cells with a 1mm path length) were obtained on a J-710 Spectropolarimeter (Jasco Analytical Instruments, Easton, MD) in a sample chamber flushed with nitrogen at 25°C. The θ values, expressed as degrees centimeter squared per decimole, were recorded, and percent alpha helicity at θ222nm was calculated using an empirical relation derived from completely helical poly-L-lysine25 following spectra correction by background subtraction (of spectra of buffer alone or of lipid-containing buffer). All samples were prepared at final peptide concentration of 0.05 mg/mL and 65 µM total lipid in the case of liposomes containing GALA-lipid.

G. Fluorescence Resonance Energy Transfer (FRET) measurements

12 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 32

To capture the GALA-lipid surface distribution on uniform and patchy carrier vesicles, lipid vesicles composed of 16:0-PC:chol:16:0-PE-RhD:16:0-PE-NBD and 18:0-PC:16:0PA:chol:

16:0-PE-RhD:16:0-PE-NBD

at

mole

ratios

of

94.7:4.7:0.3:0.3

and

66.3:28.4:4.7:0.3:0.3, respectively, were used in order to detect the relative distances between RhD- and NBD-labeled lipids. The fluorescent lipids were chosen to have 16:0-acyl-tails and were used as surrogates of the GALA-functionalized lipids which were also conjugated on lipids with 16:0-acyl tails. The preferential partitioning of fluorescent lipids within phase-separated domains formed primarily by protonated 16:0-PA lipids with lowering pH was expected to increase the extent of FRET. In vesicle suspensions, the value of pH was adjusted with 0.2 N HCl, followed by vesicle incubation at 60oC for two hours. Upon completion of incubation, lipid vesicles were allowed to cool to room temperature and FRET (ex/em: 460/590 nm) from NBDlabeled lipids (ex/em: 460/530 nm) to RhD-labeled lipids (ex/em: 550/590 nm) was measured. The measured FRET intensities were normalized with respect to the corresponding Rhd-lipid fluorescence intensities for two reasons: first, to correct for potential pipetting errors in sample preparation, and, second, to correct for the effects of pH change on the fluorescence intensities of Rhd-lipids.

H. Statistical analysis Results are reported as the arithmetic mean of n independent measurements ± the standard deviation. Student’s t-test was used to evaluate differences in behavior between different forms.

13 ACS Paragon Plus Environment

Page 15 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

3. Results and Discussion A. Characterization of GALA-functionalized lipid and of carrier- and target- lipid vesicles Following purification of GALA-lipid, by HPLC (Figure S2), 80 ± 10% yield by weight was obtained. The molecular weight of the conjugate (MW = 3827 Da) was confirmed, within error of measurement, by the peak at 3814 via MALDI-TOF spectrometry (Figure S3). It is not unexpected for the linear mode of operation, which was used for this measurement, to result in such small errors of mass. Additionally, the peak at 3868, 56 Da heavier than the target MW, could possibly correspond to the uncleaved protective group (C2H9) of one of the seven glutamic acids' side chain. Carrier lipid vesicles, which were designed to be in gel-phase at working temperatures, exhibited sizes of 90 ± 2 nm (PDI 0.08 ± 0.06, n=12) and 67 ± 6 nm-in-diameter (PDI 0.2 ± 0.04, n=12) in the absence and presence of functionalization with GALA-lipid, respectively. Endosomal-analogue (target) vesicles, which were composed of fluid membranes, exhibited sizes of 110 ± 14 nm (PDI 0.12 ± 0.05, n=6). The size of target- and carrier- vesicles was stable at physiologic pH for the duration of studies described herein. For some compositions, acidification of pH resulted in formation of aggregates (Figure S4, S5 and S6). Table 1. Nomenclature and compositions of the carrier-lipid vesicles. Ratios are shown in % mole. Lipid membrane

18:0-PC

14:0-PA

16:0-PA

18:0-PA

16:0GALAlipid

Cholesterol

18:0-PEG

pH-responsive

Clustered 1

63

-

27

-

2

5

3



Clustered 2

63

-

-

27

2

5

3



Clustered 3

63

27

-

-

2

5

3



Uniform 1

90

-

-

-

2

5

3

X

All one domain

87

-

-

-

6

4

3

X

14 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 32

Three compositions of phase-separating carrier-vesicles were studied (clustered 1, 2, and 3) which were functionalized with GALA-lipids for pH-dependent clustered display (Table 1). Across all three compositions, the size of the phase-separated domain was designed to not vary by including the exact same amount of phase-separating lipid (30 % mole of total phospholipid). The composition clustered 1 was chosen to represent the case of greatest partition of GALA-lipid within the phase-separated domain due to the identically chosen lipid acyl-tails between domain-forming lipids (16:0-PA) and GALA-functionalized lipids (GALA-16:0-PE). For compositions clustered 2 and 3, the acyl-tails of domain forming lipids (18:0-PA or 14:0PA) were two carbon atoms longer or shorter than the GALA-lipid (16:0 carbon atoms), respectively. For the pH-independent, uniform display of GALA-lipid on carrier-vesicles, uniform 1 vesicles were included containing GALA-lipid at exactly the same overall molar concentration as all clustered compositions (2 % mole of the total lipid). In an effort to exclude the effect of domain size (patch size), a 'all one domain'-carrier vesicle composition was also studied. The 'all one domain'-carrier vesicles were functionalized throughout the entire vesicle surface at GALA-lipid density equal to the maximum possible local density of GALA-lipid within the phase-separated lipid domains when in clustered display in clustered carrier-vesicles (i.e. with all of the 2 % mole functionalized lipid partitioning in the phase-separated domain). The 'all one domain'-carrier vesicles were not composed of pH-responsive lipid bilayers. The 'all one-domain' lipid vesicles were chosen because they contained on their entire surface the GALA-lipid at high surface densities comparable to the high surface densities observed only locally within the phase-separated domains ('patches') of the 'clustered' lipid vesicles.

15 ACS Paragon Plus Environment

Page 17 of 32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Free GALA-peptide exhibited, as expected,1 pH-dependent shift in alpha helicity, showing maximum helical content at pH 5.5 followed by a slight decrease at pH 4.0 (Table 2). Incorporation of GALA-lipid on carrier-vesicles resulted in almost constant levels of alphahelicity with weak pH dependence (Table 2). These findings were not affected by the presence or absence of PEGylated lipid.

Table 2. Extents of alpha-helicity of GALA-lipids in carrier lipid vesicles and of free GALApeptide as a function of pH. Errors correspond to standard deviations of two independent measurements. Lipid membrane

pH 7.4

pH 6.0

pH 5.5

pH 4.0

Clustered 1 – No PEG

39.4 ± 3.2%

53.5 ± 3.7%

51.2 ± 13.4%

67.7 ± 23.9%

Clustered 1

48.7 ± 3.1%

50.6 ± 4.5%

45.8 ± 0.1%

42.7 ± 5.0%

Uniform 1– No PEG

40.5 ± 6.9%

45.1 ± 2.3%

43.4 ± 2.4%

48.7 ± 8.7%

Uniform 1

43.6 ± 0.4%

46.8 ± 5.1%

46.3 ± 2.6%

50.1 ± 11.9%

free GALA-peptide

37.8 ± 4.0 %

44.9 ± 3.8 %

48.4 ± 1.0 %

38.0 ± 2.5 %

B. Studies on Inducing Content Release from Endosome-surrogate Target-Vesicles Carrier-vesicles exhibiting clustered GALA-lipid display with lowering pH (clustered 1, clustered 2, clustered 3) resulting in phase-separated regions with locally high surface densities of GALA-lipids or carrier-vesicles uniformly functionalized with comparably high surface densities of GALA-lipids ('all one-domain' carrier-vesicles) demonstrated non-monotonic activity in inducing content release from fluid-phase target-vesicles (Figure 2). The maximum release rates were observed at pH 5.5 and 5.0. Contrary, carrier-vesicles uniformly functionalized (uniform 1) with same amounts of GALA-lipids, as the clustered compositions, but resulting in

16 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 32

low surface densities of GALA-lipids, exhibited monotonically increasing release rates with lowering pH (Figure 2). At pH values of 5.5 and greater, clustered carrier vesicle compositions resulted in faster release rates than uniform carrier vesicles. Half lives of calcein's self-quenching relief (indicators of content release) were 12.2 ± 0.6 min (clustered 1) vs. 54.3 ± 23.5 min (uniform) at pH 5.5 (p-value < 0.01). However, at pH 4.0, uniform 1 carrier-vesicles exhibited faster content release rates compared to all other carrier vesicle compositions. Free GALApeptide exhibited monotonic content release similar to that seen from the uniform carrier-vesicles (Figure S7).

Figure 2. Half-lives of release rates (t1/2=ln(2)/krelease) from endosome analogue target-vesicles encapsulating self-quenching concentrations of calcein induced by clustered 1 (white circles), clustered 2 (grey circles), clustered 3 (black circles), uniform 1 (white squares), and 'all one domain' (white diamonds) carrier-vesicles. The lines are guides to the eyes. Error bars correspond to standard deviations of three independent measurements. * indicates p-values