Protocols
Experimental and computational surface hydrophobicity analysis of a non-enveloped virus and proteins

https://doi.org/10.1016/j.colsurfb.2017.02.011Get rights and content

Highlights

  • The hydrophobicity of protein surfaces are difficult to calculate and measure.

  • Different experimental and computational approaches were compared.

  • Hydrophobicity using ANS fluorescence and RPC compared to SASA calcs.

  • Porcine parvovirus was determined to be more hydrophobic than proteins.

  • Virus hydrophobicity can be used to determine new methods for virus removal.

Abstract

The physical characteristics of viruses needs to be understood in order to manipulate the interaction of viruses with host cells, as well as to create specific molecular recognition techniques to detect, purify, and remove viruses. Viruses are generally believed to be positively charged at physiological pH, but there are few other defining characteristics. Here, we have experimentally and computationally demonstrated that a non-enveloped virus is more hydrophobic than a panel of model proteins. Reverse-phase and hydrophobic interaction chromatography and ANS fluorescence determined the experimental hydrophobic strength of each entity. Computational surface hydrophobicity was calculated by the solvent exposed surface area of the protein weighted by the hydrophobicity of each amino acid. The results obtained indicate a strong correlation between the computational surface hydrophobicity and experimentally determined hydrophobicity using reverse-phase chromatography and ANS fluorescence. The surface hydrophobicity did not compare strongly to the weighted average of the amino acid sequence hydrophobicity. This demonstrates that our simple method of calculating the surface hydrophobicity gives general hydrophobicity information about proteins and viruses with crystal structures. In the process, this method demonstrated that porcine parvovirus (PPV) is more hydrophobic than the model proteins used in this study. This adds an additional dimension to currently known virus characteristics and can improve our manipulation of viruses for gene therapy targeting, surface adsorption and general understanding of virus interactions.

Graphical abstract

The hydrophobicity of porcine parvovirus (PPV) and several proteins was evaluated with hydrophobic interaction chromatography (HIC), reverse phase chromatography (RPC) and ANS (8-anilino-1-naphthalenesulfonic acid) fluorescence and this was compared to the crystal structure’s solvent accessible surface area (SASA) that was weighted by amino acid hydrophobicity.

  1. Download : Download high-res image (125KB)
  2. Download : Download full-size image

Introduction

Specific molecular interactions govern the attachment of viruses to cells, initiating the viral infection cycle. The understanding of these specific interactions can lead to the creation of unique therapies for viral diseases. In addition, specific chemical interactions can be harnessed to produce a targeted viral gene therapy vector for gene delivery [1], [2], for example, the specific targeting of cancer cells [3]. Virus surface properties can also be exploited to manipulate viruses. Virus removal operations have been designed by absorbing viruses to surfaces [4]. Other uses of virus interactions are in the detection of viruses [5] and in the purification of vaccines or viral gene therapy vectors [6], [7], [8], [9]. To produce a general method to adsorb viruses, for any of these applications, the physical properties of viruses that distinguish them from proteins needs to be clearly identified and quantified.

There are limited data on the isoelectric point of viruses [5], [10], [11]. This is likely due to the difficulty of purifying a virus with a high enough concentration to make the measurement. However, it is generally accepted that viruses are negatively charged at physiological pH. This is shown by the large use of anion exchange chromatography [12], [13], [14], [15] and other positively charged moieties [16], [17] for the adsorption of viruses.

A lesser-known property of viruses is their higher hydrophobicity as compared to many other proteins. Non-enveloped viruses have been found to bind to hydrophobic surfaces. Examples of this include a significant enhanced adsorption of MS2 bacteriophage to hydrophobic sand, compared to protein-coated silica nanoparticles and rotavirus [18] and the removal of porcine parvovirus by adsorption of small, hydrophobic peptides affixed to a chromatographic support [19], [20]. Influenza virus, an enveloped virus, was found to adsorb quickly and with high affinity to a gold surface [21]. Complete xenotropic murine leukemia virus (XMuLV) clearance was achieved under high conductivity with a depth filter, which is large enough for the virus to pass through, suggesting that the hydrophobic interactions strongly contributed to this retrovirus clearance [22]. Bacteriophages MS2 demonstrated a higher removal than ФX174 by ultrafiltration, which may be attributed that the surface of phage MS2 being more hydrophobic than that of phage ФX174. This may imply that an increase in hydrophobicity has the potential to assist virus interactions with the membrane material [23]. Viral hydrophobicity was measured using an octyl Sepharose-4 fast flow resin at pH 7.2 and concluded that MS2 and T4 are hydrophobic viruses while ФX 174 is a hydrophilic virus [24]. Experimentally, hydrophobic interaction chromatography has been used to purify proteins [25], viruses [26] and virus-like particles [27] with minimal theoretical basis. While some work has been conducted on the hydrophobicity of bacteriophages, little work has conclusively determined the hydrophobicity of mammalian viruses and virus-like particles (VLPs).

Other indications of virus hydrophobicity have also been reported. Amino acids and sugars were found to selectively precipitated the human parvovirus B19 [28], PPV [8], and the enveloped Sindbis virus [7], implying that the increased hydrophobicity of the virus as compared to other proteins allowed for the selective precipitation. Polymers created to mimic virus nanoparticles for gene delivery were more efficient when the polymer was more hydrophobic [29]. This may imply that an increase in hydrophobicity has the potential to assist viruses in their mission as gene delivery vehicles.

The hydrophobicity of a protein or a virus is difficult to quantify. The hydrophobic strength of the core of a protein is believed to give the protein structural stability. This is often studied computationally to determine protein folding and stability [30], [31], [32]. Protein fragments can be categorized as globular, surface seeking or transmembrane [33]. However, this computational expensive method is only available for small protein segments. Experimental methods that measure surface hydrophobicity include the parameter 1/m* measured by precipitation in solution [34] and a microbial adhesion to hydrocarbons (MATH) assay [35] that has been used to determine the hydrophobicity of MS2 bacteriophage [36] and rotavirus [18]. The hydrophobic/hydrophilic balance of the phages has been indirectly evaluated from adhesion experiments performed on hydrophobic and hydrophilic self-assembled mono-layers models [37]. An AFM method that measured the hydrophobic interaction forces between a silicon nitride tip coated with ФX174, MS2 and Aichi virus and a hydrophobic sand surfaces demonstrated the hydrophobic interaction of the phages [38]. Other experimental measurements of surface hydrophobicity of proteins is the fluorescent probe ANS, that has been shown to measure the surface hydrophobicity of amyloid aggregates [39], [40] and the separation of proteins using aqueous two phase system (ATPS) as defined by Log K values [41]. However, without a universal hydrophobicity measurement, it is difficult to compare published results, as has been demonstrated [42].

The hydrophobicity of surfaces can be determined by the oscillation of water molecules in molecular dynamic simulations [43]. Others have noted that to have a more quantitative measure of a hydrophobic surface, the cavity formation of the water structure is needed [44]. For proteins, it is more complicated. The surfaces are very heterogeneous and have complex topographies. It has been shown that the size of the hydrophobic patch on a protein is important in the hydrophobic interaction of proteins [45]. A large hydrophobic patch has more energy to control the water network than a lone hydrophobic amino acid that is surrounded by hydrophilic amino acids.

We desire to quantify the hydrophobicity of viruses as compared to a panel of standard proteins. We compared a computational approach based on the surface accessible surface area calculated by the Eisenberg hydrophobicity scale for protein and virus crystal structures to several different experimental methods to measure hydrophobicity. These methods included hydrophobic interaction chromatography, reverse-phase chromatography and ANS fluorescence. Our variety of methods characterized our model virus, PPV, as hydrophobic compared to the panel of proteins explored. The computational approach verified that other viruses also have a highly hydrophobic surface. This hydrophobicity measurement can be used to better understand virus-cell interactions as well as create improved methods to detect, remove, and purify viruses.

Section snippets

Materials

The proteins in this study were, bovine serum albumin, BSA, (Sigma, St. Louis, MO), chicken egg white lysozyme, LYS (CalBioChem, Billerica, MA), bovine fibrinogen, FIB (Sigma, St. Louis, MO), bovine hemoglobin, HEM (Sigma, St. Louis, MO), ribonuclease A, RNAse (Sigma, St. Louis, MO) and human immunoglobulin G, IgG (Equitech-Bio, Kerrville, TX). HPLC grade acetonitrile and trifluoroacetic acid (TFA) were purchased from Sigma (St. Louis, MO) for C18 chromatography. For the phosphate buffer

Experimental measurements of hydrophobicity

We explored three different methods to measure protein and virus hydrophobicity and two methods to calculate the hydrophobicity. This can be seen in Table 2. Due to some methods only working on certain proteins and virus, we show the multiple results and compare them. In Fig. 1, the proteins that were able to be eluted with RPC using acetonitrile and HIC on a butyl column are shown. IgG elution was determined with HIC, but a peak was not found in RPC. It was found that IgG could be eluted from

Comparison of computational and experimental results

There was little correlation for any of the three experimental methods (RPC, HIC and ANS) when compared to the molecular weight of the proteins (using the MW of the individual subunit for PPV) and there was little correlation between the experimental methods and the weighted average hydrophobicity of the amino acid sequence. These correlations can be seen if Fig. S3. Since ANS and HIC are mainly measuring surface hydrophobicity, it was expected that these methods would not correlate with the

Conclusions

We have determined the hydrophobicity of a panel of proteins and of one virus using three experimental methods, hydrophobic interaction chromatography, reverse-phase chromatography and ANS fluorescence. These experimental methods were compared with two computational methods, the weighted average hydrophobicity of the amino acid sequence and the weighted average hydrophobicity of the solvent accessible surface area. We concluded that ANS is the best measure of surface hydrophobicity and the

Acknowledgements

We would like to thank NSF (CBET-1159425, CBET-1125585, and CBET-1451959) for partial funding of this work. The authors would also like to thank Dr. Katherine A. Philips for help troubleshooting the SASA program.

References (67)

  • Y. Zhang et al.

    Selection of downstream steps by analysis of protein surface property: a case study for recombinant human lactoferrin purification from milk of transgenic cow

    Process Biochem.

    (2015)
  • C. Dika et al.

    Isoelectric point is an inadequate descriptor of MS2, Phi X 174 and PRD1 phages adhesion on abiotic surfaces

    J. Colloid Interface Sci.

    (2015)
  • G. Tubio et al.

    Relationship between the protein surface hydrophobicity and its partitioning behaviour in aqueous two-phase systems of polyethyleneglycol-dextran

    J. Chromatogr. B-Anal. Technol. Biomed. Life Sci..

    (2004)
  • R. Dembczynski et al.

    Partitioning of lysozyme in aqueous two-phase systems containing ethylene oxide-propylene oxide copolymer and potassium phosphates

    Food Bioprod. Proc.

    (2013)
  • F. Hachem et al.

    Hydrophobic partitioning of proteins in aqueous two-phase systems

    Enzyme Microb. Technol.

    (1996)
  • C.L. Heldt et al.

    A colorimetric assay for viral agents that produce cytopathic effects

    J. Virol. Methods

    (2006)
  • M.F. Tafur et al.

    Reduction of porcine parvovirus infectivity in the presence of protecting osmolytes

    Antivir. Res.

    (2013)
  • D. Eisenberg et al.

    Analysis of membrane and surface protein sequences with the hydrophobic moment plot

    J. Mol Biol.

    (1984)
  • A.A. Simpson et al.

    The structure of porcine parvovirus: comparison with related viruses

    J. Mol. Biol.

    (2002)
  • J. Holm et al.

    Ligand binding induces a sharp decrease in hydrophobicity of folate binding protein assessed by 1-anilinonaphthalene-8-sulphonate which suppresses self-association of the hydrophobic apo-protein

    Biochem. Biophys. Res. Commun.

    (2012)
  • A. Jungbauer et al.

    Hydrophobic interaction chromatography of proteins: III. Unfolding of proteins upon adsorption

    J. Chromatogr. A.

    (2005)
  • L. Tavel et al.

    Interactions between aroma compounds and beta-lactoglobulin in the heat-induced molten globule state

    Food Chem.

    (2010)
  • A.A. Bhattacharya et al.

    Crystallographic analysis reveals common modes of binding of medium and long-chain fatty acids to human serum albumin

    J. Mol. Biol.

    (2000)
  • R. Diamond

    Real-space refinement of the structure of hen egg-white lysozyme

    J. Mol. Biol.

    (1974)
  • P.L. Sinn et al.

    Gene therapy progress and prospects: development of improved lentiviral and retroviral vectors − design, biosafety, and production

    Gene Ther.

    (2005)
  • A. Lakshminarayanan et al.

    A galactose-functionalized dendritic siRNA-nanovector to potentiate hepatitis C inhibition in liver cells

    Nanoscale

    (2015)
  • Y.R. Liu et al.

    Site-Specific modification of adeno-associated viruses via a genetically engineered aldehyde tag

    Small

    (2013)
  • B. Michen et al.

    Isoelectric points of viruses

    J. Appl. Microbiol.

    (2010)
  • B. Kalbfuss et al.

    Purification of cell culture-derived human influenza A virus by size-exclusion and anion-exchange chromatography

    Biotechnol. Bioeng.

    (2007)
  • M.N. Eglon et al.

    Purification of adenoviral vectors by combined anion exchange and gel filtration chromatography

    J. Gene. Med.

    (2009)
  • T. Iskra et al.

    Development of a modular virus clearance package for anion exchange chromatography operated in weak partitioning mode

    Biotechnol. Progr.

    (2015)
  • W. Riordan et al.

    Design of salt-tolerant membrane adsorbers for viral clearance

    Biotechnol. Bioeng.

    (2009)
  • K. Farkas et al.

    Adsorption of rotavirus, MS2 bacteriophage and surface-modified silica nanoparticles to hydrophobic matter

    Food Environ. Virol.

    (2015)
  • Cited by (52)

    View all citing articles on Scopus
    View full text