Phage Nanobiotechnology

By Valery A. Petrenko George P. Smith

The Royal Society of Chemistry

Copyright © 2011 Royal Society of Chemistry
All rights reserved.
ISBN: 978-0-85404-184-8

Contents

Chapter 1 The Phage Nanoparticle Toolkit George P. Smith, 1,
Chapter 2 The Roles of Structure, Dynamics and Assembly in the Display of Peptides on Filamentous Bacteriophage Stanley J. Opella, 12,
Chapter 3 Quantitative Analysis of Peptide Libraries Lee Makowski, 33,
Chapter 4 Phage-mediated Drug Delivery Valery A. Petrenko and Prashanth K. Jayanna, 55,
Chapter 5 Imaging with Bacteriophage-derived Probes Susan L. Deutscher and Kimberly A. Kelly, 83,
Chapter 6 Phage-based Pathogen Biosensors Suiqiong Li, Ramji S. Lakshmanan, Valery A. Petrenko and Bryan A. Chin, 101,
Chapter 7 Phage-mediated Detection of Biological Threats Steven Ripp, 156,
Chapter 8 Genetically Engineered Virulent Phage Banks for the Detection and Control of Bacterial Biosecurity Threats François Iris, Flavie Pouillot, Hélène Blois, Manuel Gea and Paul-Henri Lampe, 175,
Chapter 9 Site-directed Chemical Modification of Phage Particles Lana Saleh and Christopher J. Noren, 202,
Chapter 10 Filamentous Phage-templated Synthesis and Assembly of Inorganic Nanomaterials Binrui Cao and Chuanbin Mao, 220,
Chapter 11 Phage Vaccines and Phage Therapy Karen Manoutcharian, 245,
Subject Index, 259,

CHAPTER 1

The Phage Nanoparticle Toolkit

GEORGE P. SMITH

Division of Biological Sciences, University of Missouri, Columbia, MO 65211, USA

1.1 Introduction

The development of phage virions (particles) as novel nanoparticles has been closely tied to phage display technology. Since filamentous phages of the Ff class (wild-type strains fl, fd and M13) are the predominant phage display vectors, so too they have been the predominant type of phage nanoparticle – at least so far. Accordingly, this chapter will concentrate on Ff phages, although it will point out ways in which they differ fundamentally from other phages such as T4. For a summary of Ff phage biology in general, the reader is referred to an excellent review. Here the focus will be more specifically on those aspects of the life cycle that are of direct practical concern to scientists and engineers seeking to develop the virions as nanoparticles.

1.2 Virion Structure and Purification

The wild-type Ff virion is 900 nm long and 6 nm in diameter. Almost all of the nanoparticle’s surface and 87% of its weight is a tubular sheath composed of ~2760 copies of the 50-residue major coat protein pVIII in a geometrically regular array, as depicted in the molecular model in Figure 1.1. Residues in the first third of pVIII are accessible from the outside, whereas the last third of the protein, including four lysines and no acidic side-chains, are exposed in the lumen. Inside the lumen is the single-stranded circular viral DNA and at the tips are five copies each of the minor coat proteins: pVII (33 residues) and pIX (32 residues) at the tip that is extruded first from the cell during assembly, pIII (406 residues) and pVI (112 residues) at the other tip. The length of the virion is dictated by the length of the viral DNA that it contains. If the viral DNA is artificially lengthened or shortened by adding or subtracting segments, the number of pVIII subunits and length of the virion increase or decrease in proportion. The geometry of the pVIII array does not depend on a geometrically specific interaction between the pVIII subunits and the nucleotides of the viral DNA. Instead, packing of the DNA in the lumen of the tubular sheath seems to require only overall electrostatic balance between the positive charges on the four lumenal e-amino groups of the pVIII polypeptide and the negatively charged phosphates. Modifying that balance leads to a compensatory change in ratio of virion length to number of viral DNA nucleotides, without perceptibly changing the geometry of the pVIII array.

Virions can be prepared to a high degree of purity in two complementary steps that are readily scaled to large volumes. First, since virions are released without lysing the host cells (see Section 1.5), simply removing cells by low-speed centrifugation eliminates all but a trace of intracellular components. A second low-speed centrifugation of the culture supernatant dramatically reduces contamination with residual intact cells. This simple, scalable way of removing the vast bulk of intracellular contaminants does not apply to phages such a λ and T4 that are released by lysis rather than extrusion. The second Ff virion purification step is precipitation from the culture supernatant with a low concentration (2% w/v) of poly(ethylene glycol) (PEG); under these conditions, most residual contaminants, including DNA, RNA, proteins and ribosomes, remain in the supernatant after the virions have been sedimented by low-speed centrifugation. Purity can be substantially improved by recentrifuging the pellet after pouring off the bulk of the supernatant and aspirating the residual supernatant. Spherical phages, including λ and T4, can also be precipitated with PEG, but only at a concentration (~10% w/v) that also precipitates many intracellular contaminants such as DNA, RNA and ribosomes.

Ff virions prepared by PEG precipitation from cleared supernatant are sufficiently pure for most purposes. Further purification, if necessary, is usually accomplished by CsCl density equilibrium ultracentrifugation, which accommodates multiple phage clones on a modest scale (up to ~1015 virions), but achieves only modest reductions in contamination with proteins or non-particulate molecules. Size-exclusion chromatography effectively resolves virions from non-particulate proteins, but has very limited capacity (~1013 virions per 30 ml column). Hydroxyapatite chromatography resolves virions from most protein contaminants and is easily scaled to any number of virions but, unlike the other two methods, is not highly reproducible and depends sensitively on the nature of displayed guest peptides. Neither chromatography method conveniently accommodates multiple clones.

The virion is remarkably robust, retaining infectivity after ~10 min of exposure to pH 2.2, pH 12 or 6 M urea, indefinite exposure to 70 °C, and disulfide reduction; exposure to many proteases such as trypsin and chymotrypsin; and other harsh conditions. This robustness facilitates affinity selection, which depends on specific binding of phage to an immobilized target (the ‘selector’) and subsequent release by breaking the bonds between the phage and selector or between the selector and the immobilizing surface. Relatively harsh release conditions can be used without compromising infectivity – the ability of the released phages to be propagated by infecting fresh host bacteria. It should be noted, however, that some release conditions can subtly alter the physical characteristics of the virion even if they do not reduce infectivity.

Virions remain physically intact almost indefinitely at refrigerator temperatures, even if they have not been purified. However, unpurified virions can lose infectivity over a period of years, because two domains of pIII that are necessary for infectivity (see the next section) but not for overall physical integrity are relatively accessible to contaminating proteases. It is therefore best to purify virions that must be stored for more than a few months. Virions can be stored frozen, but they lose a substantial fraction of their infectivity with each freeze–thaw cycle.

1.3 Intrusion

The filamentous phage infection cycle is shown schematically in Figure 1.2. Its major stages will be discussed in this section and the two that follow.

Ff phages infect Escherichia coli cells carrying the F factor plasmid and therefore displaying F pili on their surface. Infection, called ‘intrusion’ in this chapter, is a two-stage process in which two domains, N1 and N2, at the N-terminus of pIII play a critical role; it is summarized in the aforementioned review. In the first stage, domain N2 binds to the tip of the F pilus, which then retracts, drawing the tip of the virion to the cell surface. In the second stage, domain N1 interacts with the cell’s Tol apparatus. This interaction somehow triggers intrusion of the single-stranded viral DNA – the plus strand – into the cytosol and concomitant transfer of the coat proteins into the cell’s inner membrane. When a large number of cells are mixed with a limiting number of virions, infectivity – the number of successfully infected cells per virion – is very high: typically around 50% for wild-type phage. This reflects both the virion’s overall physical robustness (see the previous section) and the efficiency of minus strand synthesis, which converts the infecting plus strand to a double-stranded RF within a few minutes (see Figure 1.2 and the next section). Because the RF is considerably more stable than the viral single strand, once an infection reaches the RF stage it seldom aborts.

Neither the F pilus nor the Tol apparatus plays any role in phage production. Nor are pIII domains N1 or N2 required for phage assembly. Mutant phages in which those domains are mutationally inactivated or deleted altogether are produced in more or less normal yields, although of course they are non-infective.

1.4 DNA Replication Cycle and Gene Expression

Once the viral plus-strand DNA is inside the cell, cellular enzymes synthesize the complementary minus strand starting at an initiation site within a minus-strand origin, one of the functional elements in the intergenic region depicted in Figure 1.3. The result is the circular double-stranded replicative form (RF), which is the template for mRNA transcription and eventually the starting point for further replication. That replication is initiated by nicking of the plus strand at the plus-strand origin by phage protein pII (Figure 1.3) and continues in the rolling circle mode until a complete progeny plus strand has been made. Early in infection, the progeny plus strand serves as template for minus-strand synthesis, just as did the original infecting viral DNA (Figure 1.2). Late in infection and in chronically infected cells, however, the phage single-stranded DNA binding protein pV sequesters most progeny plus strands to form the pV/ssDNA complex, which is the precursor to progeny virion assembly (see Figure 1.2 and the next section).

The phage genes are expressed at very different levels. Some of the proteins in effect act catalytically in phage DNA replication and virion extrusion (see the next section), while five are consumed stoichiometrically by incorporation into the virion (see Section 1.2). The five virion proteins are trafficked to the cell’s inner membrane, from which they are transferred into extruding virions. Two of them, pIII and pVIII, have signal peptides that are cleaved to form the mature polypeptides, whose N-terminal portions are exposed in the periplasm before transfer into virions. Chronically infected cells come to a steady state in which synthesis and consumption of viral DNA and proteins are balanced and do not kill the cell.

1.5 Extrusion of Progeny Virions

Most of the familiar phages, including λ and T4, are assembled in the cytosol of the infected cell and released by cell lysis. Ff phages, in contrast, are continuously extruded through the cell envelope without killing the cell, which continues to divide, albeit at a reduced rate. It is the reduction in growth rate, not lysis, that accounts for plaque formation.

The precursor for progeny virion assembly is the long, thin pV/ssDNA complex (Figure 1.2) produced late in infection and in chronically infected cells. At one tip of the complex the viral DNA forms a hairpin called the packaging signal (PS), one of the functional elements in the intergenic region (Figure 1.3). Assembly is initiated by interaction of the PS with a special pore through the cell envelope fashioned from three phage proteins. The complex’s DNA is extruded through the pore, shedding its intracellular covering of pV molecules back into the cytosol and concomitantly acquiring its extracellular covering of the five coat proteins from the inner membrane. The process is extraordinarily productive; the wild-type yield reaches 1012 virions ml-1 under ordinary laboratory culture conditions and vigorous aeration can boost the yield to 1013 virions ml-1 – equivalent to 250 mg of pVIII molecules per liter.

The assembly process imposes constraints on the peptides or proteins that can be successfully displayed by genetic fusion to coat proteins. Since all the coat proteins are incorporated into the extruding virion from the inner membrane, peptides or proteins that block translocation of the fusion protein through the inner membrane cannot be displayed. Since the virion exits the cell through a narrow pore, large proteins are not efficiently displayed, although amazingly virions with up to 24 50-kDa immunoglobulin Fab domains fused to pVIII have been documented. These constraints are lessened when the virions carry wild-type in addition to recombinant versions of the displaying coat protein. In particular, only very short peptides can be displayed on all copies of pVIII, while proteins as large as the 50-kDa Fab domain can be displayed on a few copies of pVIII when the remaining pVIII subunits are wild-type.

Any substantial imbalance in the multiple convergent pathways of the infection cycle leads to cell killing and absence of phage production. Even subtle impairments, such as those incurred by the presence of a small foreign peptide on a few copies of the pVIII coat protein, can compromise cell viability sufficiently to impose measurable selective pressure against the peptide. The adverse effects of subtle imbalance, or even of gross defects in virion assembly, are largely eliminated by reducing the DNA copy number (the number of double-stranded RF molecules per cell), as will be explained in the final section.

1.6 Display of Guest Peptides

Foreign (‘guest’) peptides and proteins can be fused genetically to the coat proteins and thereby displayed on the virion’s outer surface. This is depicted schematically in Figure 1.4, where the recombinant peptide-bearing proteins and the recombinant genes that encode them are colored red. Usually (but not always) it is the pIII or pVIII coat protein that thus hosts the guest peptide or protein, as summarized in Figure 1.4. N-terminal display on these proteins requires the foreign coding sequence to lie between the coding sequences for the signal peptide and the mature coat protein, without disturbing the reading frame. Small peptides can be fused to all five copies of pIII (type 3 display) or to all copies (thousands of them) of pVIII (type 8 display). In these cases, the recombinant gene III or gene VIII encoding the fusion protein replaces the wild-type gene. Alternatively, peptides or proteins can be displayed on one copy of pIII or on a few dozen copies of pVIII, the remaining pIII or pVIII subunits being the wild-type (type 33 or 88 display). In these cases, the recombinant gene III or VIII is in addition to the corresponding wild-type gene (Figure 1.4). Not depicted in Figure 1.4 are type 3 + 3 and 8 + 8 displays, in which the wild-type and recombinant coat-protein genes are carried on separate genomes present in the same cell. In most cases, the recombinant gene is carried on a phagemid, as will be described in the next section. Display of guest peptides on the other three coat proteins will not be covered in this review.

1.7 The Engineer’s Toolkit

This final section summarizes some of the most common methods for manipulating Ff phage to achieve practical goals such as developing new phage nanoparticles. For some purposes, suitable phage nanoparticles can be created by direct chemical modification of the virion surface, which contains thousands of surface-accessible amines that can be chemically modified without disturbing virion integrity, or even in some cases infectivity. However, most applications exploit the ability to display foreign guest peptides on the virion surface by genetic fusion to the pIII or pVIII coat protein (see the previous section).

The starting point for genetic manipulation of phage is the double-stranded RF (Figure 1.2), which can be purified from stationary-phase cells and used like a typical bacterial plasmid. Numerous vectors with suitable gene-III or -VIII cloning sites and other features have been developed. When necessary, the phage genome can accommodate extra genetic elements in the intergenic region between the PS and the minus-strand origin (Figure 1.3) without perceptibly affecting phage replication. Extra elements can also be tolerated within the plus-strand origin to the right of its two hairpins (Figure 1.3) provided that there is a compensating gene-II mutation. The prime example of an extra genetic element is the recombinant gene III or VIII in type 33 or 88 display (Figure 1.4). In type 3 + 3 and 8 + 8 display, in contrast, the extra recombinant gene III or VIII is carried on a phagemid.

Phagemids are a special class of plasmids that carry the phage intergenic region, spanning the origins of replication and the nearby PS (Figure 1.3). When cells carrying a phagemid are superinfected by phage, the plasmid replicates in the phage mode and one of its strands is extruded in the form of a normal virion; the superinfecting phage in these circumstances is called a ‘helper’. If the phagemid bears a gene encoding the N-terminal domains of pIII, that gene must be repressed before helper superinfection is possible, because those domains block deployment of the F pilus, the receptor for Ff phage infection. In addition, because pIII and pVIII are toxic at high concentrations, expression of gene III or VIII on a high copy number phagemid must generally be greatly down-regulated in order to ensure cell viability even after superinfection has been successfully completed. When the recombinant gene III or VIII in type 3 + 3 or 8 + 8 display is carried on a phagemid, it is included in the phagemid virions released from the cell. This is the design of type 3 + 3 and 8 + 8 libraries, allowing phagemids bearing rare selector-binding peptides to be affinity selected along with the corresponding recombinant gene.

(Continues…)Excerpted from Phage Nanobiotechnology by Valery A. Petrenko George P. Smith. Copyright © 2011 Royal Society of Chemistry. Excerpted by permission of The Royal Society of Chemistry.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.