Separation of B and C Type Virions by Centrifugation in Gentle Density Gradients

Murine type B particles were separated from type C (Rauscher leukemia virus) by means of gentle (low-increment rate) density gradients. The best separation was obtained when the density ranged from 1.13 to 1.20 g/cm³ when sucrose was used and from 1.12 to 1.28 g/cm³ with CsCl. The buoyant densities of the B and C particle bands in sucrose were 1.18 and 1.16 g/cm³, respectively. The CsCl gradient gave a better separation with the B particles banding at a density of 1.20 g/cm³ and with the C particle density little different from its value in sucrose.     

Purification of the Kilham Rat Virus

A purification procedure is described for the isolation of Kilham rat virus (RV) from infected suckling hamster kidney and liver suspensions. The procedure involved a combination of sonic treatment, differential centrifugation, butanol-chloroform extraction, agar column flow diffusion, and potassium tartrate density gradient centrifugation. The purified virus retained its infectivity and was specifically neutralized by RV hyperimmune antiserum. Electron micrographs from the RV band (density 1.31 g/ml) showed numerous homogeneous particles approximately 22 mmu in diameter.     

Infectious hepatitis A virus particles produced in cell culture consist of three distinct types with different buoyant densities in CsCl.

Although hepatitis A virus (HAV) released by infected BS-C-1 cells banded predominantly at 1.325 g/cm3 (major component) in CsCl, smaller proportions of infectious virions banded at 1.42 g/cm3 (dense HAV particles) and at 1.27 g/cm3 (previously unrecognized light HAV particles). cDNA-RNA hybridization confirmed the banding of viral RNA at each density, and immune electron microscopy demonstrated apparently complete viral particles in each peak fraction. The ratio of the infectivity (radioimmunofocus assay) titer to the antigen (radioimmunoassay) titer of the major component was approximately 15-fold greater than that of dense HAV particles and 4-fold that of light HAV particles. After extraction with chloroform, the buoyant density of light and major component HAV particles remained unchanged, indicating that the lower density of the light particles was not due to association with lipids. Light particles also banded at a lower density (1.21 g/cm3) in metrizamide than did the major component (1.31 g/cm3). Dense HAV particles, detected by subsequent centrifugation in CsCl, were indistinguishable from the major component when first banded in metrizamide (1.31 g/cm3). However, dense HAV particles recovered from CsCl subsequently banded at 1.37 g/cm3 in metrizamide. Electrophoresis of virion RNA under denaturing conditions demonstrated that dense, major-component, and light HAV particles all contained RNA of similar length. Thus, infectious HAV particles released by BS-C-1 cells in vitro consist of three distinct types which band at substantially different densities in CsC1, suggesting different capsid structures with varied permeability to cesium or different degrees of hydration.     

Heterogeneity of vesicular stomatitis virus particles: implications for virion assembly

Vesicular stomatitis virus (VSV) particles formed at early times after infection contain only one-third the amount of viral glycoportein (G protein), relative to the major internal structural proteins M and N, as is found in particles released later. These "early" particles also have a lower density in equilibrium sucrose gradients than do those formed later; however, the sedimentation velocity and specific infectivity of these two classes of particles are the same. VSV-infected cells also release virus-like particles which sediment considerably faster than authentic virions and contain a higher-than-normal proportion of the VSV G protein relative to internal VSV proteins. These particles have a reduced specific infectivity but a normal density in sucrose gradients. All classes of VSV virions contain a constant proportion of M and N polypeptides. The ratio of G protein to M or N protein, in contrast, can vary over a sixfold range; this implies that an interaction between a precise number of surface G proteins with either of the underlying M and N proteins is not a prerequisite for budding of infectious viral particles from the cell surface.     

Deoxyribonucleotide Pools and Deoxyribonucleic Acid Synthesis in Mouse Embryo Cells Infected with Three Classes of Polyoma Virus Particles

Polyoma virus particles were purified by equilibrium centrifugation in CsCl. Particles from three regions of the density gradient were examined for infectivity, for their ability to induce expanded pools of deoxyribonucleic acid (DNA) precursors, and for their ability to stimulate the synthesis of DNA. The most infectious population of particles, the virions, having a buoyant density of 1.33 g/ml, gave the greatest stimulation of the DNA-synthesizing apparatus of mouse embryo cells. Empty particles at density 1.29 g/ml had no DNA stimulatory activity. A population of particles of intermediate density, referred to as pseudovirions, was also much less active than virions in stimulating DNA synthesis, and the limited stimulatory activity of the latter fraction may be accounted for by its measured contamination with infective particles.     

Properties and origins of infectious rhinovirus type 14 particles of different buoyant densities.

Isopycnic centrifugation of rhinovirus type 14 (RV14), purified from infected HeLa or KB cell cultures, into CsCl gradients resolved two bands of infectious virus particles with buoyant density values of 1.409 +/- 0.007 (H virus) and 1.386 +/- 0.004 (L virus) g/ml. Only H virus was detected by incorporation of radiolabeled uridine into viral RNA, and H virus accounted for the majority of infectivity in gradients. H and L virus could not be differentiated by plaque morphology, extent of neutralization by RV14-specific antiserum, or particle size. Electron microscope studies showed that most L-virus particles were associated with an amorphous material. Treatment of L virus with proteolytic enzymes or rebanding L virus in CsCl gradients resulted in recovery of the majority of infectivity as H virus. Virus purified from cell-free fluids from infected HeLa or KB cell cultures banded only as H virus. HeLa cell cultures challenged with purified H virus and harvested at 3 h postinoculation for virus purification yielded only infectious H virus. Both H and L viruses were detected in cell cultures that had been challenged with purified H virus and harvested at 12 h postinoculation. The data suggest that H virus represents progeny virus, whereas L virus represents sequestered infectious virus particles which become associated with an amorphous material and do not enter into viral replicative processes.     

The origin of nascent single-stranded DNA extracted from mammalian cells.

In vitro cultured bovine liver cells were labelled with radioactive thymidine and dissolved in 0.5% sodium dodecyl sulphate. Centrifugation of the lysate through sucrose gradients in a zonal rotor revealed a slowly sedimenting fraction of preferentially pulse labelled DNA. The DNA of this zone was further analysed by chromatography on hydroxy-apatite, banding in CsCl density gradients, and sedimentation in neutral and alkaline sucrose gradients. It contained besides small amounts of fragmented bulk DNA, single-stranded nascent DNA and single-stranded pre-labelled DNA which could be separated from each other by using BrdU as a density label. The density labelling also revealed small amounts of nascent-nascent DNA duplexes. The slowly sedimenting fraction was practically absent from cell lysates which were prepared in 2 M NaCl - 50 microgram/ml pronase. The results suggest that nascent single-strands and nascent-nascent duplexes are released from the forks of replicating DNA by branch migration. Pre-labelled single strands may be released by the same branch migration. Pre-labelled single strands may be released by the same mechanism, but the in vivo structure from which they originate has yet to be elucidated.     

Equilibrium centrifugation of measles virus

Measles virus has been centrifuged on different density gradients. It sediments at densities of 1,20 g/cm³ in K-tartrate, of 1,18–1,21 g/cm³ in sucrose, 1,19–1,23 g/cm³ in CsCl and 1,19 g/cm³ in metrizamide gradients. Metrizamide reduced measles virus infectivity. In sucrose gradients sometimes more than one infectious peak was observed. Control Vero cells produced particles of the same densities as measles virus peaks. These peaks did contain actin as the major protein. The relevance of this finding in relation to the presence of actin in measles virus is discussed.     

Role of lysine in the replication of reovirus: I. Synthesis of complete and empty virions

Lysine is essential for the replication of infectious reovirus. Omission of lysine from the extracellular medium not only permitted the continued synthesis of structural viral proteins and viral double-stranded ribonucleic acid (RNA), but also caused an enhanced formation of viral structures which were separable by isopycnic sedimentation of CsCl into a top band consisting of empty particles with a buoyant density of 1.29 g/cm³ and essentially free of viral RNA, and two lower bands which were difficult to resolve and had an average buoyant density of 1.37 g/cm³. The lower bands contained most of the viral nucleic acid. The above effects were reversed when lysine was restored early after infection. In contrast, a single band with a buoyant density of 1.38 g/cm³ was obtained from lysine-plus infected cells.     

Use of Zonal Ultracentrifuge Systems for Biophysical Studies of Rhinoviruses

This paper reports the use of zonal ultracentrifuge techniques to conduct biophysical studies of rhinoviruses grown with WI-38 cells. Good clean-out of infectivity from rhinovirus harvests was obtained with the continuous-flow B-V and B-IX rotors. Use of the B-V rotor resulted in the successful concentration of rhinovirus infectivity and antigenicity. Additional purification was achieved by the combined use of continuous-flow centrifugation and isopycnic banding procedures. Two particle sizes were found to be associated with the virus-infected cell harvests. The infectious 22-nm particle banded in density ranges of 1.38 to 1.40 g/cm(3) in CsCl and 1.26 to 1.27 g/cm(3) in potassium citrate. The 8.0 nm capsomere was composed of 2.0 nm subunits and banded with a density of protein at 1.28 g/cm(3) in CsCl. Equivalent sedimentation coefficients of 155 or 185, depending on particle density in sucrose, were calculated from rate zonal experiments by use of the B-IV zonal rotor.     

Herpes Simplex Virus Products in Productive and Abortive Infection III. Differentiation of Infectious Virus Derived from Nucleus and Cytoplasm with Respect to Stability and Size

Approximately 67% of infectivity is associated with the nucleus 8 hr after productive infection of HEp-2 cells with herpes simplex virus. Comparison of nuclear and cytoplasmic infectious virus and macromolecular aggregates labeled with (3)H-thymidine or with (3)H-choline revealed the following. (i) Cytoplasmic infectious virus and macromolecular aggregates banded in CsCl at a density corresponding to enveloped nucleocapsids. The virus was relatively stable; there was only 50% loss of infectivity and only 16% of the virions became disaggregated. (ii) Nuclear macromolecular aggregates banded in CsCl solution at a density corresponding to unenveloped nucleocapsids and, moreover, both the infectious virus and aggregates were highly unstable. (iii) In sucrose density gradients, the nuclear macromolecular aggregates and infectivity sedimented as a single band and migrated more slowly than the corresponding cytoplasmic material. (iv) The infectivity of nuclear and cytoplasmic virus is readily inactivated by digestion with phospholipase C and with pronase. We conclude the following. (i) Cytoplasmic virus consists of enveloped nucleocapsids. (ii) Nuclear virus consists of nucleocapsids covered with lipid or partially enveloped. (iii) The molecular integrity of viral lipids is essential for infectivity. (iv) The envelope protects the nucleocapsid and accelerates adsorption to cells; it is not, however, inherently essential for infectivity.     

Heavy and Light Particles of Adeno-Associated Virus

KB cells coinfected with adenovirus and adeno-associated virus (AAV) yielded two kinds of infectious AAV particles that banded in CsCl at densities of 1.45 and 1.41 g/cm², respectively. The 1.45 band was found to be composed of a heterogeneous group of viral particles that could be subfractionated by velocity sedimentation. The main component from this band had a smaller S value (109) than the main component from the 1.41 band (111S), although both had the same DNA/protein ratio and the same density in metrizamide gradients. Continuous-label experiments showed that early after infection, both components (1.45 and 1.41) were generated in the same amounts, but this was followed by a relative increase in the proportion of the 1.41 component over the 1.45 particles. Pulse-chase analysis failed to demonstrate a precursor-product relationship between these two bands. The slower-sedimenting components from the 1.45 band were unstable in CsCl and were present in a greater proportion early after infection. These particles contained DNA that was enriched for the terminal sequences of the AAV genomes and was accessible to digestion with micrococcal nuclease.     

Characterization and classification of virus particles associated with hepatitis A. I. Size, density, and sedimentation.

Virus-like particles were purified from stools of patients in an epidemic of hepatitis A in Germany. When reference MS-1 chimpanzee pre-inoculation and convalescent sera were used, the close serological relationship of the purified particles to well-known isolates of hepatitis A could be established. On the other hand, the physicochemical characteristics of the particles were determined in parallel to the characteristics of a marker parvovirus (LuIII) and a marker picornavirus (poliovirus type 2). It could be shown that the majority of the hepatitis A-associated particles band at 1.34 g/ml in CsCl and, like poliovirus, sediment at about 160S. In addition, a distinct hepatitis A antigen was observed, which banded at 1.305 g/ml and sedimented between 50 and 90S. A further component accumulated in the density range of between 1.38 and 1.44 g/ml. However, it seemed to be rather labile. Upon reisolation from CsCl and sedimentation in sucrose, it resolved into a 160S, a 90 to 100S, and a 50S form. The size of the 160S particles (27 to 29 nm) could be readily distinguished from that of the parvovirus (22 to 24 nm). It is concluded, therefore, that hepatitis A-associated virus particles are more likely to be classified with the picornaviruses than with the parvoviruses.     

Scanning Electron Microscopic Observation of Differentiation from the Reproductive Short Form to the Infectious Long Form of Holospora obtusa

To identify the surface features of Holospora obtusa during its differentiation from the reproductive short form to the infectious long form, bacteria of four different buoyant densities were isolated by Percoll density gradient centrifugation of homogenates of host cells or isolated macronuclei, and examined with a scanning electron microscope. Bacteria of buoyant density 1.09 g/ml were reproductive short forms as well as cells at various stages in the elongation process including fully elongated ones. Bacteria of buoyant densities 1.11 g/ml and 1.13 g/ml were premature long forms and those of 1.16 g/ml were mature infectious long forms. Bacteria of buoyant density 1.09 g/ml had an entirely rough surface while those of buoyant densities 1.11 g/ml and 1.13 g/ml were smooth and had wale-like stripes on their surface. A small tapered tip was observed at one end of the bacteria of buoyant density 1.13 g/ml. Bacteria of buoyant density 1.16 g/ml had an entirely smooth surface, but one end always showed a rough surface; this locally differentiated surface of the special tip of the infectious long form may be responsible for both the nuclear and species specificities of the infectivity of H. obtusa. These observations indicate that the surface of H. obtusa changes during differentiation and the special tip develops in bacteria of buoyant density 1.13 g/ml.     

Virus-like particles from killer, neutral, and sensitive strains of Saccharomyces cerevisiae.

Procedures were developed for purification of virus-like particles (VLPs) from killer, neutral, and sensitive strains of Saccharomyces cerevisiae. Morphologically similar spherical VLPs measuring 40 nm in diameter were extracted from all three phenotypes. The particles were partially purified by high-speed centrifugation through a layer of CsCl (1.26 g/cm3) and sucrose density gradient centrifugation. Gradient-purified preparations contained three centrifugal species that sedimented at approximately 43, 102, and 162S. The 43S component is considered to be an artifact. Preparations from killer strains contained three double-stranded RNA (ds-RNA) components with molecular weights of 1.19 x 10(6), 1.29 x 10(6) and 2.54 x 10(6). VLPs from neutral and sensitive strains contained only the largest ds-RNA species. VLP preparations were subsequently separated into two major density components by CsCl equilibrium gradient centrifugation. The light component banding at 1.28 to 1.30 g/cm3 was void of nucleic acid, and the heavy component banding at 1.40 g/cm3 contained only the largest ds-RNA species.     

Purification and partial characterization of virus-like particles from Schneider line 2 Drosophila cells

A procedure has been developed for the purification of virus-like particles (VLPs) from Schneider line 2 Drosophila cells. The VLPs were precipitated with polyethylene glycol from the cytoplasmic fraction of lysed cells and further purified by equilibrium centrifugation in CsCl density gradients, in which they band at a density of 1.366 g/ml. Electron micrographs of these preparations revealed polyhedral particles with a diameter of 310–330 Å. We have also found particles of this size in thin sections of the intact cells. Sedimentation of the VLPs through 10–70% sucrose gradients yields a sedimentation coefficient of 235 S. Preliminary studies show that the VLPs contain double-stranded RNA species of 10 S, 14.5 S, 16 S, and 18 S.     

Differential biophysical properties of infectious intracellular and secreted hepatitis C virus particles

The recent development of a cell culture infection model for hepatitis C virus (HCV) permits the production of infectious particles in vitro. In this report, we demonstrate that infectious particles are present both within the infected cells and in the supernatant. Kinetic analysis indicates that intracellular particles constitute precursors of the secreted infectious virus. Ultracentrifugation analyses indicate that intracellular infectious viral particles are similar in size (approximately 65 to 70 nm) but different in buoyant density (approximately 1.15 to 1.20 g/ml) from extracellular particles (approximately 1.03 to 1.16 g/ml). These results indicate that infectious HCV particles are assembled intracellularly and that their biochemical composition is altered during viral egress.     

Replication of Nondefective Parvoviruses: Lack of a Virion-Associated DNA Polymerase

We have examined four of the nondefective parvoviruses for an associated DNA polymerase. Virions were purified from neuraminidase-treated infected-cell lysates by isopycnic centrifugation in CsCl or from infected cell material by CaCl(2) precipitation and centrifugation through sucrose into CsCl. Preparations of bovine parvovirus or Kilham rat virus obtained by the former procedure contained DNA polymerase activity but were not free of contaminating cellular proteins. The latter method produced viral preparations free of contaminating cellular proteins, and no DNA polymerase activity was detected in light infectious particles of H-1, LuIII, bovine parvovirus, or Kilham rat virus. Examination of levels of each cellular DNA polymerase in these preparations from each step of both purification procedures revealed that DNA polymerase beta had a greater tendency to copurify with bovine parvovirus and Kilham rat virus than did DNA polymerases alpha or gamma. Disruption of infectious virions obtained by the second purification method with detergents and sonic treatment did not result in the detection of a DNA polymerase activity. The biological activity and purity of each of the four different viruses obtained by the latter procedure were determined by hemagglutination and infectivity assays, polyacrylamide gel electrophoresis, and electron microscopy. In each case, the virions banding at a density of 1.39 to 1.41 g/cm(2) in CsCl were infectious and contained only the virion structural proteins. DNA polymerase activity was not detected in any of these preparations, and we have concluded that a virion-associated DNA polymerase is not required for productive infection with the nondefective parvoviruses.