Polyvalent Hybrid Virus-Like Nanoparticles with Displayed Heparin

Jun 14, 2018 - Table 1 lists the various mutants, which were generated using standard .... of thiol chips at 30 °C using N-β-maleimidopropionic acid...
0 downloads 0 Views 333KB Size
Subscriber access provided by University of Sussex Library

Brief Article

Polyvalent hybrid virus-like nanoparticles with displayed heparin antagonist peptides Justin M. Choi, Valerie Bourassa, Kevin Hong, Michael Shoga, Elizabeth Lim, Andrew Park, Kazim Apaydin, and Andrew K. Udit Mol. Pharmaceutics, Just Accepted Manuscript • DOI: 10.1021/acs.molpharmaceut.8b00135 • Publication Date (Web): 14 Jun 2018 Downloaded from http://pubs.acs.org on June 19, 2018

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 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 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 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.

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 20 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

Molecular Pharmaceutics

+

+ extended

OSO3‐

heparin assembly

Q coat proteins

+

OSO3‐

OSO3‐

+

+

wild‐type Hybrid particle

ACS Paragon Plus Environment

heparin

OSO3‐

+

OSO3‐

+ Restored coagulation

OSO3‐

Molecular Pharmaceutics 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 2 of 20

1 For Molecular Pharmaceutics

Polyvalent hybrid virus-like nanoparticles with displayed heparin antagonist peptides

Justin M. Choi, Valerie Bourassa, Kevin Hong, Michael Shoga, Elizabeth Lim, Andrew Park, Kazim Apaydin, and Andrew K. Udit*

Department of Chemistry, Occidental College, Los Angeles, CA

*Corresponding Author: Occidental College, Department of Chemistry, 1600 campus Rd, Los Angeles, CA 90041, [email protected], ph: 323-259-2761, fax: 323-341-4912

Abbreviations: VLP, virus-like particle; APTT, activated partial thromboplastin time

ACS Paragon Plus Environment

Page 3 of 20 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

Molecular Pharmaceutics

2 Abstract The potential applications for nanomaterials continue to grow as new materials are developed and environmental and safety concerns are more adequately addressed. Virus-like particles (VLPs) in particular have myriad applications in medicine and biology, exploiting both the reliable, symmetric self-assembly mechanism and the ability to take advantage of surface functionalities that may be appropriately modified through mutation or bioconjugation. Herein we describe the design and application of hybrid VLPs for use as potent heparin antagonists, providing an alternative to the toxic heparin antidote protamine. A two-plasmid system was utilized to generate VLPs that contain both the wild-type coat protein and a second coat protein with either a C- or N-terminal cationic peptide extension (4-28 amino acids). Incorporation of the modified coat proteins varied from 8% to 31%, while activated partial thromboplastin time (APTT) assays revealed a range of heparin antagonist activity. Notably, when examined based on the quantity of peptide delivered due to the varied incorporation rates it appeared that the VLPs largely followed a similar trend, with the quantity of peptide delivered more closely correlating with heparin antagonist activity. The particle with the highest incorporation rate and best anti-heparin activity displayed the C-terminal peptide ARK2A2KA, which corresponds to the Cardin-Weintraub consensus sequence for binding to glycosaminoglycans. Analysis of this particle using heparin affinity chromatography with fraction collection revealed that particles eluting at higher salt concentration had a greater proportion of peptide incorporation. Preliminary dual polarization interferometry experiments further support a strong interaction between this particle and heparin.

Keywords: heparin, protamine, virus-like particle, polyvalent, bacteriophage

ACS Paragon Plus Environment

Molecular Pharmaceutics 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 4 of 20

3 Introduction The application of nanomaterials in biology and medicine, and nanoparticles in particular has become more widespread as researchers have produced myriad particles, and toxicity has been more adequately addressed.1-3 Therapeutic applications have seen good success with a number of nanomaterials currently approved for use by the FDA.4 Particularly attractive for development and use are virus-like particles (VLPs): the ability to use VLPs of varying size, to exploit the varied chemistry of amino acid side chains, perform bio-orthogonal conjugations, and control the position and quantity of specific side chains via mutation make VLPs versatile scaffolds from which one can build a highly specific nanoparticle for targeted applications.5-7 The stability and ease of production of many VLPs further enhances their appeal for biomedical development. A particularly promising VLP for biomedical applications is bacteriophage Qβ, an icosahedral nanoparticle 28 nm in diameter that is composed of 180 copies of a 132 amino acid, 14.1 kDa coat protein.8, 9 The capsid has good stability towards extremes in pH and temperature,10 can tolerate a range of modifications made to its surface either through mutation or chemically post-assembly,11-15 and can be expressed in E. coli in high yield. Our group in particular has examined various applications of Qβ;16-22 current investigations delve into exploring the ability of Qβ variants to antagonize the anticoagulant activity of heparin. Heparin is a naturally occurring glycosaminoglycan derived primarily from porcine tissue as a heterogeneous mixture of polysaccharides with variable molecular weight (5-30 kDa) and degree of sulfation.23, 24 Its anticoagulant activity is due largely to activation of antithrombin, which inactivates the coagulation serine proteases thrombin and factor Xa.25, 26 First put into use clinically in the 1940’s, heparin remains the primary anticoagulant used today. However heparin therapy requires careful monitoring as uncontrolled bleeding can be a serious and potentially

ACS Paragon Plus Environment

Page 5 of 20 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

Molecular Pharmaceutics

4 fatal complication.27-29 To restore normal coagulation heparin is neutralized with protamine, which is available as a mixture of four peptides, each with an approximate composition of 4.5 kDa and 70% arginine.30 While protamine is the only FDA-approved agent for the reversal of heparin anticoagulation, the well-known adverse effects of protamine are significant and varied, including severe anaphylactic reactions and arterial hypertension.31-33 Recognizing that the polyvalent nature of Qβ may interact favorably with heparin chains, we endeavored to explore the ability of this VLP to antagonize heparin anticoagulant activity. Our initial study generated a series of particles that were made cationic relative to the wild-type.21 One of the most effective particles in this study was a single point mutation where a threonine on the coat protein was replaced with an arginine at a solvent-exposed position on the capsid surface (T18R), which ends up being replicated 180 times over the nanoparticle. Follow-up studies further validated the particle with clinically-relevant samples,19 and explored the biophysical properties.16 Two additional particles were also highly effective however they had some significant drawbacks compared to T18R. The first, D14R, was actually the most effective VLP but the particle proved unstable which discouraged follow-up studies. Indeed, this underscores the limits of mutation as the amino acid substitution must be tolerated by the protein scaffold. The second particle was generated by appending polyarginine peptides to the surface of the wildtype capsid via copper-catalyzed azide-alkyne cycloaddition (CuAAC, perhaps the best known of the “click” reactions34). This was a two-step process that first involved converting exposed lysines on the VLP surface to alkynes, followed by CuAAC with the complementary azidepeptide. While feasible, this method does require two conjugation steps with varying efficacy, synthesis of the peptide, and several purification steps (alternative conjugation methods would require a similar process).

ACS Paragon Plus Environment

Molecular Pharmaceutics 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 20

5 These prior experiences suggested that an ideal methodology for generating heparin antagonist nanoparticles would combine the ease of mutation with the versatility of appending peptides to the capsid. Some investigators have described ways of pursuing such modifications.35-37 For Qβ this can be accomplished by exploiting a co-expression methodology first described by Vasiljeva et al11 and further developed by Brown et al.12 Our group has also taken advantage of this to generate Qβ VLPs that bind metals and metallocycles.20, 22 A schematic of the methodology is shown in Figure 1a. A two-plasmid system is utilized, with one plasmid containing the wild-type coat protein and the other a modified coat protein, which in this case has a peptide appended to either the C- or N-terminus. Co-expression of the plasmids yields hybrid VLPs where a certain percentage of the coat proteins in the final capsid have the peptide extension (Figure 1b).

ACS Paragon Plus Environment

Page 7 of 20 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

Molecular Pharmaceutics

6 Materials and Methods All reagents were purchased from Fisher unless otherwise stated. Peptides used for clotting assays were synthesized by GenScript. Protein preparation The wild-type Qβ coat protein was cloned into the pCDF-1b vector with spectinomycin resistance. Coat proteins containing N- or C-terminal extensions were made within the pET-11d vector with ampicillin resistance. Table 1 lists the various mutants, which were generated using standard overlap extension PCR techniques, followed by sequencing to verify each construct. Each mutant-containing pET plasmid was separately co-transformed with the wild-type-pCDF plasmid into E. coli BL21(DE3) cells for protein expression using the protocol provided by the manufacturer. Protein expression proceeded by starting from glycerol stocks of each co-transformed construct. 10 mL of SOB media containing 50 µg/mL spectinomycin and 50 µg/mL ampicillin were inoculated with the transformed cells. After growing overnight in a shaker at 37oC, the 10 mL pre-culture was diluted into 1 L of aspartate media (2 mM magnesium sulfate, 10 µM iron(III) chloride, 4 µΜ calcium chloride, 2 µΜ manganese(II) chloride, 2 µΜ zinc sulfate, 0.4 µΜ cobalt(II) chloride, 0.4 µΜ copper(II) chloride, 0.4 µΜ nickel(II) chloride, 5 mg/mL glucose, 2.5mg/mL aspartic acid, 1.775 mg/mL disodium phosphate, 1.7 mg/mL monopotassium phosphate, 1.34 mg/mL ammonium chloride, 0.355 mg/mL sodium sulfate, 50 µg/mL spectinomycin, 50 µg/mL ampicillin)38 and incubated for 4 hours with shaking at 37ºC. 1 mL of 1 M IPTG was then added, and the culture was left for an additional 8 hours for protein expression. The cultures were then centrifuged with a FiberLite F10S-6x500y rotor at 5,000

ACS Paragon Plus Environment

Molecular Pharmaceutics 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 8 of 20

7 RPM for 10 minutes at 4ºC, the supernatant was decanted, and the pellet was either used immediately or stored at -20oC. Cell pellets were resuspended in 30 mL of 0.1 M sodium phosphate buffer pH 7 containing 0.02% NaN3. One mg each of DNAse, RNAse, and lysozyme were then added and the mixture was sonicated in an ice bath for 2 minutes with a 30% duty cycle. The cell debris was pelleted in a SA-600 rotor at 14,000 RPM for 15 minutes at 4ºC. The supernatant was decanted into a solvent-resistant conical tube and the protein was then precipitated using ammonium sulfate and sodium chloride (0.345g of sodium chloride per 30 mL supernatant, 8.0 g of ammonium sulfate per 30 mL supernatant). The mixture was centrifuged with a SA-600 rotor at 14,000 RPM for 15 minutes at 4°C, and the pellet resuspended in 4 mL of 0.1 M sodium phosphate buffer pH 7. An equivolume mixture of 1:1 chloroform:1-butanol was then added, and the combined solution was moderately vortexed to mix the organic solvent with the aqueous crude protein solution. The mixture was centrifuged at 2,800 RPM for 10 minutes at 8ºC to facilitate phase separation, and then the top aqueous layer was removed with a pipette. The vortex and centrifugation process was repeated with the removed aqueous layer and an equivalent volume of chloroform two additional times. The final aqueous layer was loaded onto a 10-40% sucrose gradient in 0.1 M sodium phosphate buffer pH 7 followed by ultracentrifuging at 25,000 RPM and 4oC for 6 hours with a Beckman SW32 rotor. Overhead illumination of the gradient with a flashlight revealed a diffuse band in the middle of the tube; 3-4 mL of this band were retrieved and then ultracentrifuged at 45,000 RPM for at least 1 hour with a Beckman 50Ti rotor. The pelleted Qβ particles were resuspended in HBS buffer (50 mM HEPES, 100 mM NaCl, pH 7.4). The solution was ultracentrifuged again with the same parameters and resuspended once more in HBS to remove residual sucrose.

ACS Paragon Plus Environment

Page 9 of 20 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

Molecular Pharmaceutics

8 The modified Lowry assay was used to quantify all protein samples. Notably, it was found that HBS resulted in a higher than expected value, a known complication with this assay when using HBS. This was corrected by subtracting the baseline absorption produced by HBS for each protein sample. SDS-PAGE SDS-PAGE was used to determine both the purity of the VLPs and provide quantitative data using the SYPRO stain. Lonza PAGEr® Gold Precast Gels (10 – 20% Tris-Glycine) were used with 1X Tris-Glycine Running buffer (500 mL deionized H2O, 1.5 g Tris base, 7.2 g glycine, 0.5 g SDS). The samples were prepared in 20 µL volumes containing 10-15 µg of the protein, 2 µL of 0.5 M dithiothreitol, and 10 µL of loading buffer (62.5 mM Tris HCl, 2.5% SDS, 0.002% Bromophenol blue, 0.71 M β-mercaptoethanol, 10% glycerol). The samples were incubated in an Eppendorf Mastercycler for 10-15 minutes at 95ºC and then set to cool for 5 minutes before loading into the gel. Fifteen microlitres of the PageRuler Plus Prestained Protein Ladder were also loaded into the gel. The gels were run at 125 V until the dye front neared the bottom of the gel. For coomassie staining the gels were placed in the Coomassie Blue Staining solution while on a rocker overnight and then de-stained with deionized H2O until the ladder and sample bands were visible. For SYPRO staining, the gels were placed in the SYPRO Orange Staining solution (250 mL of 7.5% Acetic Acid, 50 µL of 5000x SYPRO Orange) and incubated at room temperature for 30 minutes. The gel was then washed with 7.5% Acetic Acid for 30 seconds on a rocker before being viewed on a UV-transilluminator at 303 nm. A photo was taken of the gel while on the transilluminator and then analyzed using the LabWorks software to determine the intensity of each fluorescent band. Chromatography

ACS Paragon Plus Environment

Molecular Pharmaceutics 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 20

9 Fast Protein Liquid Chromatography (FPLC) was conducted with a Bio-Rad BioLogic Workstation. For all FPLC runs the column was pre-equilibrated with the running buffer followed by sample injection (typically 300 µL of 1 mg/mL virus). The UV absorption was set to detect eluted protein at 280 nm. Size-exclusion chromatography was performed using Superose 6 size exclusion gel (GE Healthcare). The running buffer was HBS with 0.02% NaN3. Samples were run isocratically at 0.5 mL/min for a total of 60 minutes. Heparin affinity chromatography was performed using a 1 mL HiTrap Heparin column (GE Healthcare). Solutions for chromatography were 20 mM sodium phosphate pH 7 for the running buffer, and 20 mM sodium phosphate pH 7 / 500 mM NaCl for the salt gradient. After a 5 mL isocratic loading step with the running buffer, the concentration of the gradient buffer was increased linearly from 0-50% over 15 mL. The entire run was performed at 0.5 mL/min. When necessary, 1 mL fractions were collected during the elution step followed by evaporation of most of the solvent in a speed-vac. The reduced volume samples were then analyzed by SDS-PAGE with SYPRO staining. Clotting assay The activated partial thromboplastin time (APTT) assay was used to determine clotting times. The desired amount of VLP or peptide in HBS was added to a solution of porcine heparin (sodium salt, ~ 12 kDa, EMD Biosciences) in HBS to give the desired final quantity, and the volume was adjusted to 55 µL with HBS. After 10 minutes, 40 µL of normal human plasma were added (George King Biomedical, Overland Park, KS, USA), followed by 50 µL of platelin reagent (Trinity Biotech, Bray, Ireland). The assay was conducted using the STart4 Viscositybased Detection System (Diagnostica Stago, Parsippany, NJ, USA): the mixture of reagents was

ACS Paragon Plus Environment

Page 11 of 20 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

Molecular Pharmaceutics

10 placed into the cell and allowed to warm to 37oC for 3 min, after which the reaction was initiated by the addition of 10 µL of 50 mM CaCl2 in HBS, and the resulting clotting times were recorded. Protamine sulfate for control reactions was obtained from MP Biomedicals (Solon, OH, USA). Dual polarization interferometry Measurements were kindly conducted by Malvern Instruments using the Analight4D Bio200 workstation. Previously described protocols39 were followed for hydrazide derivatization of thiol chips at 30oC using N-β-maleimidopropionic acid hydrazide (BMPH) followed by heparin treatment. Experiments were carried out with four different concentrations of CCW (0.35, 0.47, 0.7, 1.17, 2.8 mg/mL).

Results and Discussion Prior studies by Fromm,40 Shick,41 and Rullo42 have explored the ability of small cationic peptides to interact with heparin (although coagulation assays were not explicitly examined). Drawing from their work we used a combination of their peptides that either appeared promising, or that we thought would benefit from being tethered to a macromolecular scaffold (e.g., to reduce toxicity). These peptides are shown in Table 1. All peptides were appended to the Cterminus of the Qβ coat protein via PCR, except for NCW which was appended to the Nterminus. A spacer sequence, GSGSG, was placed in between the peptides and coat protein to permit the peptides to more readily extend beyond the capsid surface, potentially facilitating interaction with solution-phase molecules. Notably, peptide Val1, with an intended target sequence of R2GR2, was found to have incorporated several extra amino acids between the spacer and target sequence (likely a consequence of mispriming during PCR). Rather than correct it, since the added amino acids appeared benign with regard to interaction with heparin

ACS Paragon Plus Environment

Molecular Pharmaceutics 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 20

11 (i.e., uncharged) we elected to keep this construct as part of the pool. Peptides NCW and CCW, apart from being appended to opposite ends of the coat protein (N-terminus vs. C-terminus, respectively), are identical in sequence. These peptides are based off of the Cardin-Weintraub consensus sequence XBBBXBX (B = basic, X = hydropathic) for binding to glycosaminoglycans.43 Peptide ArgHel was designed to be a helical peptide with the cationic arginine groups all presented along one face of the helix, which may facilitate interaction with heparin.42 The various particles were expressed and purified according to standard protocols (see Materials and Methods). Particle formation was verified by size-exclusion chromatography (see Supporting Information, Figure S1). Notably, co-expression to achieve cationic peptide incorporation was necessary: preliminary attempts to simply append various cationic sequences to the C- or N-terminus of the coat protein followed by expression of only this modified coat protein did not yield isolable particles. Further, of the constructs listed in Table 1, ArgHel, Val3, and Val8 did not reliably produce isolable VLPs in yields sufficient to either work with or detect. ArgHel and Val8 have the highest charges of the group which may have disrupted assembly. Table 2 lists the nanoparticles that we were able to generate in good yield. Figure 2 shows a protein gel from SDS-PAGE of some of the purified nanoparticles. For all constructs an intense band is seen at 14.1 kDa which corresponds to the wild-type coat protein. An additional band at slightly higher molecular weight for each particle is also observed, representing the modified coat proteins with relative masses that are in accord with the expected peptide extensions. To quantify the amount of peptide-bearing coat protein in each VLP, SDS-PAGE was used in combination with SYPRO staining. The resulting fluorescent signals were integrated to determine the relative ratio of each coat protein per nanoparticle (Figure S2); these results are

ACS Paragon Plus Environment

Page 13 of 20 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

Molecular Pharmaceutics

12 shown in Table 2. Incorporation ratios range from 8% (Val1, GTGTAG2SGSG2TGTAGR2GR2; Val4, A2R3A2RA2) to 31% (CCW, ARK2A2KA). Preliminary attempts to enhance the incorporation ratio by varying the composition of the growth media did not result in notable differences (see Brown et al for discussion on this12). Comparing Tables 1 and 2, looking first at CCW and NCW, relative to the N-termini the C-termini of the coat proteins in the assembled capsid are more solvent exposed (Figure S3) which may facilitate incorporation of peptides appended at this end.44 Comparing CCW to the other constructs, that CCW has a relatively small peptide extension, moderate charge, and some less bulky cationic groups (lysine vs. arginine) all likely contribute to it having the highest degree of incorporation. Activated partial thromboplastin time (APTT) clotting assays were performed to determine efficacy at reversing heparin anticoagulant activity. The assay used normal human plasma supplemented with a platelet surrogate, with clotting initiated by the addition of calcium chloride. Normal clotting under the experimental conditions was 37 seconds. For all APTT assays with heparin, the concentration of heparin was fixed at 45 ng per 150 µL of reaction. Control experiments for some of the peptides are shown in Figure 3 (data for the remaining peptides are in Figure S4). The data for protamine (Figure 3a) shows how acutely sensitive clotting times are to the presence of this peptide; the range of 50-300 pmols over which protamine is an effective heparin antagonist corresponds to just 0.2-1.5 ng. Above this range the well-known effect of prolonged clotting time is observed, believed to be a consequence of disrupting platelet aggregation.45 As may be expected (though not previously assessed40-42) a similar trend was observed for the peptides synthesized for this study (Figures 3b and S4). Notably, both the therapeutic potential and toxicity effects of most of the synthesized peptides appear less acute which may be due to the smaller sizes and reduced positive charges (Table S1)

ACS Paragon Plus Environment

Molecular Pharmaceutics 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 20

13 compared to protamine (~4.5 kDa, ~70% Arg). Peptide Val5 had a more acute response profile which may be attributed to it having the highest positive charge of the group (+6, compared to +4 for the next most positively charged peptide). Clotting assays were performed with the VLPs listed in Table 2. Control APTT experiments with VLPs in the absence of heparin were carried out at the highest concentrations achievable under the assay conditions and with the VLP stock solutions available. These data (Figure S5) show that the VLPs alone did not have any significant effect on normal clotting times. With heparin present, most of the VLPs come close to restoring normal clotting activity at a sufficiently high VLP concentration; exemplary results for clotting assays in the presence of VLPs and heparin are shown in Figure 4a (data for the other VLPs are shown in Figure S6a). The three particles featured represent the range of activity observed, with CCW and Val1 representing the most and least effective particles, respectively, for reversing the anticoagulant effect of heparin. Figure 4b replots the data in Figure 4a, with the x-axis modified to represent the quantity of peptide delivered based on the calculated incorporation ratio (Table 2). Interestingly, plotted in this way the range seen in Figure 4a is significantly reduced, with the three particles appearing to collapse along a common trendline for heparin reversal. Except for Val5, a similar trend was observed for the other VLPs tested (Figure S6b). Further, comparing Figure 3b to Figure 4b, the system appears to benefit from the polyvalent effect: on a per peptide basis, the peptides conjugated to the VLP appear more effective at reversing heparin’s anticoagulant activity compared to the free peptides. The VLP may therefore represent an advantageous delivery vehicle, allowing one to supply various peptides at a lesser therapeutically effective concentration, thereby avoiding the potentially toxic effects of the unconjugated free peptides.

ACS Paragon Plus Environment

Page 15 of 20 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

Molecular Pharmaceutics

14 We focused further on the CCW particle as it was the best performing VLP in the APTT assay, represents the incorporation of a Cardin-Weintraub sequence, and had the highest degree of peptide incorporation (see Figure S7 for MALDI-MS and additional SDS-PAGE). Indeed, at 31% this may be the highest co-expression rate one can expect: the C-termini of the coat proteins come together at a three-fold axis which may hinder multiple peptides at that site, thus one peptide for every three coat proteins may be the maximum tolerated. To further explore the CCW particle we turned to heparin affinity chromatography, performed at 0.5 mL/min and eluting with a linear gradient from 0-250 mM NaCl in 20 mM sodium phosphate pH 7 buffer (Figure 5a). CCW shows a broad elution profile which may be a consequence of a heterogeneous population given that the co-expression methodology would yield a distribution of particles. VLPs eluting earlier may have fewer coat proteins with the appended peptide, whereas those eluting at higher ionic strength presumably possess more than the average number of modified sequences, resulting in stronger binding. To test for this we collected fractions from the chromatography run; the horizontal lines in Figure 5a represent the time points where fractions were taken for SDS-PAGE analysis with fluorescent SYPRO staining for quantification, the results of which are shown in Figure 5b. Quantification of the bands resulted in the following ratios for wild-type to modified coat protein with peptide extension, from earliest to latest fraction: 82:18, 76:24, 74:26, 73:27. Fractions eluting at higher salt concentration do indeed have a greater proportion of the modified coat protein. To further validate the strong interaction of CCW with heparin surfaces, preliminary experiments have been performed with dual polarization interferometry using solution conditions and instrument parameters as previously described.39 The association (ka) and dissociation (kd) rates were obtained as the slope and y-intercept of the linear best fit line (Figure S8) to yield a

ACS Paragon Plus Environment

Molecular Pharmaceutics 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 20

15 dissociation constant (kd/ka = Kd) of 110 nM. For comparison, via surface plasmon resonance, Kd = 2-320 nM for binding of various complement proteins to heparin surfaces.46

Summary The co-expression methodology presents a good compromise between the ease of mutation and the versatility of bioconjugation. This allowed us to readily generate a series of particles displaying heparin antagonist peptides polyvalently. The peptides alone in clotting assays were not fully effective at antagonizing heparin, while high concentrations resulted in prolonged APTT times. The activity of the peptides benefitted from the phage scaffold, permitting full antagonism of heparin activity at a lesser total peptide concentration. Of the particles generated CCW was superior in clotting assays, likely a consequence of maximizing the number of displayed peptides on the surface which in turn may be due to substituting less bulky lysines for arginine. That approximately one third of coat proteins in CCW possessed the peptide extension may be due to the three-fold axis where the C-termini meet, thus imposing a steric constraint. This method of generating potent heparin antagonist particles should be amenable to screening protocols, allowing one to readily explore a library of particles through, for example, directed evolution.

Supporting Information: Table of synthesized peptides, Qβ solvent accessibility profile, exemplary VLP FPLC, exemplary SYPRO quantification, additional APTT clotting assays, SDS-PAGE and MALDI-MS for CCW, dual polarization data for CCW. This material is available free of charge via the Internet at http://pubs.acs.org.

ACS Paragon Plus Environment

Page 17 of 20 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

Molecular Pharmaceutics

16 Funding Sources: Research Corporation, The Camille and Henry Dreyfus Foundation, the Undergraduate Research Center at Occidental College.

Acknowledgement: We thank Dr. Claudia Mujat (Malvern) for performing the dual polarization interferometry experiments.

References 1. Su, H.; Wang, Y.; Gu, Y.; Bowman, L.; Zhao, J.; Ding, M. Potential applications and human biosafety of nanomaterials used in nanomedicine. Potential applications and human biosafety of nanomaterials used in nanomedicine 2018, 38, 3-24. 2. Kumar, P.; Choonara, Y. E.; Khan, R. A.; Pillay, V. The chemo-biological outreach of nano-biomaterials: implications for tissue engineering and regenerative medicine. The chemobiological outreach of nano-biomaterials: implications for tissue engineering and regenerative medicine 2017, 23, 3538-3549. 3. Rother, M.; Nussbaumer, M.; Reggli, K.; Bruns, N. Protein cages and synthetic polymers: a fruitful symbiosis for drug delivery applications, bionanotechnology and materials science. . Chem. Soc. Rev. 2016, 45, 6213-6249. 4. Bobo, D.; Robinson, K. J.; Islam, J.; Thurecht, K. J.; Corrie, S. R. Nanoparticle-based medicines: a review of FDA-approved materials and clinical trials to date. Nanoparticle-based medicines: a review of FDA-approved materials and clinical trials to date 2016, 33, 2373-2387. 5. Sunderland, K. S.; Yang, M.; Mao, C. Phage-enabled nanomedicine: from probes to therapeutics in precision medicine. Phage-enabled nanomedicine: from probes to therapeutics in precision medicine 2017, 56, 1964-1992. 6. Wen, A. M.; Steinmetz, N. F. Design of virus-based nanomaterials for medicine, biotechnology, and energy. Design of virus-based nanomaterials for medicine, biotechnology, and energy 2016, 45, 4035-4438. 7. Ma, Y.; Nolte, R.; Cornelissen, J. Virus-based nanocarriers for drug delivery. Adv. Drug Delivery Rev. 2012, 64, 811-825. 8. Golmohammadi, R.; Fridborg, K.; Maija, B.; Valegard, K.; Liljas, L. The crystal structure of bacteriophage Qβ at 3.5 A resolution. Structure 1996, 4, 543-554. 9. Pumpens, P.; Renhofa, R.; Dishlers, A.; Kozlovska, T.; Ose, V.; Pushko, P.; Tars, K.; Grens, E.; Bachmann, M. F. The true story and advantages of RNA phage capsids as nanotools. The true story and advantages of RNA phage capsids as nanotools 2016, 59, 74-110. 10. Fiedler, J.; Higginson, C.; Hovlid, M.; Kishlukin, A.; Castillejos, A.; Manzenrieder, F.; Campbell, M.; Voss, N.; Potter, C.; Carragher, B.; Finn, M. Engineered mutations change the structure and stability of a virus-like particle. Biomacromolecules 2012, 13, 2339-2348. 11. Vasiljeva, I.; Kozlovska, T.; Cielens, I.; Strelnikova, A.; Kazaks, A.; Ose, V.; Pumpens, P. Mosaic Qβ coats as a new presentation model. FEBS Lett. 1998, 431, 7-11.

ACS Paragon Plus Environment

Molecular Pharmaceutics 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 20

17 12. Brown, S. D.; Fiedler, J. D.; Finn, M. G. Assembly of Hybrid Bacteriophage Q beta Virus-like Particles. Biochemistry 2009, 48, 11155-11157. 13. Rhee, J.-K.; Baksh, M.; Nycholat, C.; Paulson, J. C.; Kitagishi, H.; Finn, M. G. Glycantargeted virus-like nanoparticles for photodynamic therapy. Biomacromolecules 2012, 13, 23332338. 14. Pokorski, J. K.; Hovlid, M. L.; Finn, M. G. Cell targeting with hybrid Qb virus-like particles displaying epidermal growth factor. Chembiochem 2011, 12, 2441-2447. 15. Astronomo, R. D.; Kaltgrad, E.; Udit, A. K.; Wang, S. K.; Doores, K. J.; Huang, C. Y.; Pantophlet, R.; Paulson, J. C.; Wong, C. H.; Finn, M. G.; Burton, D. R. Defining Criteria for Oligomannose Immunogens for HIV Using Icosahedral Virus Capsid Scaffolds. Chem. Biol. 2010, 17, 357-370. 16. Cheong, H.; Lynch, B.; Hollingsworth, W.; Groner, M.; Hong, K.; Polonskaya, Z.; Rhee, J.-K.; Finn, M.; Gale, A.; Udit, A. Heparin binding to an engineered virus-like nanoparticle antagonist. Biomacromolecules 2017, 18, 4113-4120. 17. Groner, M.; Ng, T.; Wang, W.; Udit, A. Bio-layer interferometry studies of a sulfated virus-like nanoparticle with anticoagulant activity. Anal. Bioanal. Chem 2015, 407, 5843-5847. 18. Mead, G.; Hiley, M.; Ng, T.; Fihn, C.; Hong, K.; Groner, M.; Miner, W.; Drugan, D.; Hollingsworth, W.; Udit, A. K. Directed polyvalent display of sulfated ligands on virus nanoparticles elicits heparin-like anticoagulant activity. Bioconjugate Chem. 2014, 25, 14441452. 19. Gale, A. J.; Elias, D. J.; Averell, P. M.; Teirstein, P. S.; Buck, M.; Brown, S. D.; Polonskaya, Z.; Udit, A. K.; Finn, M. G. Engineered virus-like nanoparticles reverse heparin anticoagulation more consistently than protamine in plasma from heparin-treated patients. Thromb. Res. 2011, 128, e9-e13. 20. Udit, A.; Hollingsworth, W.; Choi, K. Metal- and metallocycle-binding sites engineered into polyvalent virus-like scaffolds. Bioconjugate Chem. 2010, 21, 399-404. 21. Udit, A. K.; Everett, C.; Gale, A. J.; Reiber-Kyle, J.; Ozkan, M.; Finn, M. G. Heparin antagonism by polyvalent display of cationic motifs on virus-like particles. ChemBioChem 2009, 10, 503-510. 22. Udit, A. K.; Brown, S.; Baksh, M. M.; Finn, M. G. Immobilization of bacteriophage Qb on metal-derivatized surfaces via polyvalent display of hexahistidine tags. J. Inorg. Biochem. 2008, 102, 2142-2146. 23. Maddineni, J.; Walenga, J. M.; Jeske, W. P.; Hoppensteadt, D. A.; Fareed, J.; Wahi, R.; Bick, R. L. Product individuality of commercially available low-molecular-weight heparins and their generic versions: therapeutic implications. Clin. Appl. Thromb./Hemostasis 2006, 12, 267276. 24. Laremore, T. N.; Zhang, F. M.; Dordick, J. S.; Liu, J.; Linhardt, R. J. Recent progress and applications in glycosaminoglycan and heparin research. Curr. Opin. Chem. Biol. 2009, 13, 633640. 25. Capila, I.; Linhardt, R. J. Heparin-protein interactions. Angew. Chem., Int. Ed. Engl. 2002, 41, 390-412. 26. Pike, R.; Buckle, A.; Le Bonniec, B.; Church, F. Control of the coagulation system by serpins. Getting by with a little help from glycosaminoglycans. FEBS J. 2005, 272, 4842-4851. 27. Warkentin, T. E.; Levine, M. N.; Hirsh, J.; Horsewood, P.; Roberts, R. S.; Gent, M.; Kelton, J. G. Heparin-induced thrombocytopenia in patients treated with low-molecular-weight heparin or unfractionated heparin. N. Engl. J. Med. 1995, 332, 1330-1335.

ACS Paragon Plus Environment

Page 19 of 20 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

Molecular Pharmaceutics

18 28. Krishnaswamy, A.; Lincoff, A. M.; Cannon, C. P. Bleeding complications of unfractionated heparin. Expert Opin. Drug Saf. 2011, 10, 77-84. 29. Cossette, B.; Pelletier, M. E.; Carrier, N.; Turgeon, M.; Leclair, C.; Charron, P.; Echenberg, D.; Fayad, T.; Farand, P. Evaluation of Bleeding Risk in Patients Exposed to Therapeutic Unfractionated or Low-Molecular-Weight Heparin: A Cohort Study in the Context of a Quality Improvement Initiative. Ann. Pharmacother. 2010, 44, 994-1002. 30. Liang, J.; Yang, V.; Vaynshteyn, Y. The minimal functional sequence of protamine. Biochem. Biophys. Res. Commun. 2005, 336, 653-659. 31. Horrow, J. C. Protamine: A review of its toxicity. Anesth. Analg. 1985, 64, 348-361. 32. Kimmel, S.; Sekeres, M.; Berlin, J.; Goldberg, L.; Strom, B. Adverse events after protamine adminisration in patients undergoing cardiopulmonary bypass: risks and predictors of under-reporting. Adverse events after protamine adminisration in patients undergoing cardiopulmonary bypass: risks and predictors of under-reporting 1998, 51, 1-10. 33. Weiss, M. E.; Adkinson, N. F. Allergy to Protamine. Allergy to Protamine 1991, 9, 339355. 34. Hong, V.; Presolski, S.; Ma, C.; Finn, M. Analysis and optimzation of copper-catalyzed azide-alkyne cycloaddition for bioconjugation. Angew. Chem. Int. Ed. 2009, 48, 9879-9883. 35. Servid, A.; Jordan, P.; O'Neil, A.; Prevelige, P.; Douglas, T. Location of the bacteriophage P22 coat protein C-terminus provides opportunities for the design of capsid-based materials. Biomacromolecules 2013, 14, 2989-2995. 36. Patterson, D.; Schwarz, B.; Avera, J.; Western, B.; HIcks, M.; Krugler, P.; Terra, M.; Uchida, M.; McCoy, K.; Douglas, T. Sortase-mediated ligation as a modular approach for the covalent attachment of proteins to the exterior of the bacteriophage P22 virus-like particle. Bioconjugate Chem. 2017, 28, 2114-2124. 37. Chen, Q.; Sun, Q.; Molino, N.; Wang, S.-W.; DBoder, E.; Chen, W. Sortase A-mediated multi-functionalization of protein nanoparticles. Chem. Commun. 2015, 51, 12107-12110. 38. Studier, F. W. Protein production by auto-induction in high-density shaking cultures. Prot. Expression Pur. 2005, 41, 207-234. 39. Popplewell, J. F.; Swann, M. J.; Ahmed, Y.; Turnbull, J. E.; Fernig, D. G. Fabrication of Carbohydrate Surfaces by Using Nonderivatised Oligosaccharides, and their Application to Measuring the Assembly of Sugar-Protein Complexes. Fabrication of Carbohydrate Surfaces by Using Nonderivatised Oligosaccharides, and their Application to Measuring the Assembly of Sugar-Protein Complexes 2009, 10, 1218-1226. 40. Fromm, J.; Hileman, R.; Caldwell, E.; Weiler, J.; Linhardt, R. Pattern and spacing of basic amino acids in heparin binding sites. Arch. Biochem. Biophys. 1997, 343, 92-100. 41. Schick, B. P.; Maslow, D.; Moshinski, A.; San Antonio, J. D. Novel concatameric heparin-binding peptides reverse heparin and low-molecular-weight heparin anticoagulant activities in patient plasma in vitro and in rats in vivo. Novel concatameric heparin-binding peptides reverse heparin and low-molecular-weight heparin anticoagulant activities in patient plasma in vitro and in rats in vivo 2004, 103, 1356-1363. 42. Rullo, A.; Nitz, M. Importance of the spatial display of charged residues in heparinpeptide interactions. Biopolymers 2010, 93, 290-298. 43. Cardin, A. D.; Weintraub, H. J. Molecular modeling of protein-glycosaminoglycan interactions. Arteriosclerosis 1989, 9, 21-32. 44. VIPERdb. http://viperdb.scripps.edu/splot1.php?VDB=1qbe.

ACS Paragon Plus Environment

Molecular Pharmaceutics 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 20 of 20

19 45. Olsson, A.; Alfredsson, J.; Hakansson, E.; Svedjelholm, R.; Berglund, J.; Berg, S. Protamine reduces whole blood platelet aggregation after cardiopulmonary bypass. Scandinavian Cardio. J. 2016, 50, 58-63. 46. Yu, H.; Munoz, E.; Edens, R.; Linhardt, R. Kinetic studies on the interactions of heparin and complement proteins using surface plasmon resonance. Biochim. Biophys. Acta 2005, 1726, 168-176.

ACS Paragon Plus Environment