key: cord-0042392-spz4vmy5
authors: FENNER, FRANK; BACHMANN, PETER A.; GIBBS, E. PAUL J.; MURPHY, FREDERICK A.; STUDDERT, MICHAEL J.; WHITE, DAVID O.
title: Cultivation and Assay of Viruses
date: 2014-06-27
journal: Veterinary Virology
DOI: 10.1016/b978-0-12-253055-5.50007-4
sha: 06a945a8ff3009e642beb2027444916fac424e82
doc_id: 42392
cord_uid: spz4vmy5

Viruses replicate only within living cells. Some viruses are restricted in the kinds of cells in which they replicate, and a few have not yet been cultivated at all under laboratory conditions. However, most viruses are grown in cultured cells, embryonated hen's eggs, or laboratory animals. In veterinary virology, the natural host animal is used for the cultivation of viruses; indeed the earliest viral assay has been carried out with foot-and-mouth disease virus in cattle. The natural host is still useful for the studies of pathogenesis and immunology, experiments in chemotherapy, and occasionally for diaglostic purposes. However, the in vitro cultivation of viruses in cell cultures is essential for the study of their mode of replication and for diagnostic virology. Cells may be grown in vitro as explants of tissue, such as respiratory or intestinal epithelium, or as cell cultures. Explant cultures are occasionally used for research purposes or for the cultivation of certain viruses, but almost all diagnostic and research work involving viral cultivation is carried out in cell cultures—usually in monolayers, occasionally as suspension cultures. To produce cell monolayers, tissue is cut into small pieces and placed in a medium containing a proteolytic enzyme such as trypsin. After the cells have dispersed into a single-cell suspension, they are washed, counted, and diluted in a growth medium and permitted to settle on the flat surface of a glass or plastic container. Most types of cells adhere quickly and under optimal conditions, they divide about once a day until the surface is covered with a confluent monolayer.

Viruses replicate only within living cells. Some viruses are restricted n the kinds of cells in which they replicate, and a few have not yet been :ultivated at all under laboratory conditions. Fortunately, however, nost viruses can be grown in cultured cells, embryonated hen's eggs, or abora tory animals.

In veterinary virology, the natural host animal can be used for the rultivation of viruses; indeed the earliest viral assay was carried out by ^oeffler and Frosch with foot-and-mouth disease virus in cattle. The latural host is still useful for studies of pathogenesis and immunology, /accine trials, experiments in chemotherapy, and occasionally for diaglostic purposes. However, in vitro cultivation of viruses in cell cultures s essential for the study of their mode of replication (Chapter 4) and for iiagnostic virology (Chapter 13).

Although over 70 years have elapsed since mammalian cells were first jrown in vitro, it is only since the advent of antibiotics that cell culture also referred to as tissue culture) became a matter of simple routine. 39

Aseptic precautions are still essential, but the problems of contamina tion with bacteria, mycoplasmas, fungi, and yeasts are no longer insur mountable. Today, most kinds of animal cells can be cultivated in vitro for at least a few generations, and numerous immortal cell lines have been derived. Since 1949, when Enders, Weller, and Robbins reported that poliovirus could be grown in cultured nonneural cells with the production of recognizable cytopathic changes, hundreds of previously unknown viruses have been isolated and identified in cell cultures. The discovery of the adenoviruses, rhinoviruses, and many others during the 1950s and 1960s was directly attributable to their use, as were the consequent revolution in the diagnosis of viral diseases, the develop ment of vaccines, and the dramatic advances in knowledge of the mo lecular biology of animal viruses.

Cells may be grown in vitro as expiants of tissue, such as respiratory or intestinal epithelium, or as cell cultures. Expiant cultures are occasion ally used for research purposes or for the cultivation of certain viruses, but almost all diagnostic and research work involving viral cultivation is carried out in cell cultures-usually in monolayers, occasionally as sus pension cultures.

To produce cell monolayers, tissue is cut into small pieces and placed in a medium containing a proteolytic enzyme such as trypsin. After the cells have dispersed into a single-cell suspension, they are washed, counted, diluted in a growth medium, and permitted to settle on the flat surface of a glass or plastic container. Most types of cells adhere quickly, and under optimal conditions they divide about once a day until the surface is covered with a confluent monolayer.

Cell culture has been greatly aided by the development of chemically defined media containing almost all the nutrients required for cell growth. The best known of these media, developed by Eagle, is an isotonic salt solution with added glucose, vitamins, and amino acids, buffered at pH 7.4, and containing antibiotics to inhibit the growth of bacteria and fungi. Serum must be added to Eagle's medium, and to most others, to supply additional growth factors, without which most cells will not multiply satisfactorily. In recent years several growth fac tors have been identified, and certain cell lines can now be grown in media that are totally defined chemically. For instance, the Madin-Darby canine kidney cell line has been grown on fibronectin or in poly-lysine-or collagen-coated dishes, in serum-free medium supplemented with hormones including insulin, binding proteins (e.g., transferrin and albumin), and attachment factors. Such serum-free media are particu larly useful for the cultivation of "hybridoma" cells used for the produc tion of monoclonal antibodies, where there is a need to ensure that all the immunoglobulin in the medium is antibody of a single specificity, produced by the hybridoma cells. Defined media also present advan tages for the isolation of viruses that are likely to be neutralized by antibody present in "normal" animal serum, but this is generally not a problem if fetal calf serum is employed. Fetal calf serum (5-10%) is therefore incorporated in the media used for the initial growth of cells in culture. Once the monolayer is established, the "growth medium" is removed, virus is inoculated, and "maintenance medium" containing little or no serum is added to the culture.

Many types of cells undergo only a few divisions in vitro before dying out, whereas others will survive for up to a hundred cell generations and some can be propagated indefinitely. These differences, the nature of which are not fully understood, give us three main types of cultured cells (Plate 3-1).

Primary Cell Cultures. When cultures are established initially from tissue taken directly from animals (often from fetal organs or tissues), they contain several cell types, most of which are capable of only limited growth in vitro-perhaps 5 or 10 divisions at most. This restricts their value, whether for routine diagnostic work or vaccine production, be cause of the high cost and inconvenience of having to obtain fresh tissue each time, as well as lack of consistency from batch to batch. Further more, the donor animals often harbor latent viruses which can confuse diagnosis or contaminate vaccines. Nevertheless, the presence of a di verse range of differentiated cell types in such primary cultures means that they tend to be very sensitive to many animal viruses. In veterinary diagnostic virology, it is common practice to inoculate samples sus pected to contain virus into primary cultures derived from the same species of animal as that providing the samples. virology and vaccine production, but diploid strains have not been much used in veterinary vaccine production.

These are cells of a single type that are capable of indefinite propagation in vitro. Such immortal cell lines originate from cancers, or by spontaneous transformation of a diploid cell strain. Often they no longer bear close resemblance to their cell of origin, as they undergo many mutations during their prolonged culture. The usual indication of these changes is that the cells have lost the specialized morphology and biochemical abilities that they possessed as differenti ated cells in vivo. For example, it is no longer possible to distinguish microscopically between the epithelial cell lines arising from various cells of ectodermal or endodermal origin, or between the fibroblastic cell lines arising from cells of mesodermal origin. Cells of continuous cell lines are often aneuploid in chromosome number, especially if of malig nant origin.

Continuous cell lines derived from monkey (e.g., the Vero cell line), dog (MDCK), cattle (MDBK), pig (PK15), cat (CFK), mouse (L929, 3T3), hamster (BHK-21), rabbit (RK-13), and others are widely used in experi mental and diagnostic virology (see Table 13 -3).

The great advantage of continuous cell lines over primary cell cultures is that they can be propagated indefinitely by subculturing the cells at regular intervals. Like other cells, they retain viability for many years when frozen in serum-containing medium with added dimethylsulfoxide and stored at very low temperature, e.g., in liquid nitrogen (-196°C) or an electric deep-freezer at -70°C or lower. Good laboratories follow the general microbiological precept that the surest way of faithfully maintain ing the characteristics of a cultured cell line is to replace it periodically from frozen stocks.

Some continuous cell lines have been adapted to grow as suspensions of single cells. Such suspension cultures are particularly useful for bio chemical studies of viral replication, because large numbers of identical cells are continuously available for regular sampling and processing.

Various methods have been devised to maximize the surface area to which cells can attach, while keeping the overall size of the vessel and the volume of medium within reasonable bounds. Round bottles can be continuously rolled, or may be filled with glass tubes, glass beads, or spiral plastic film. Perhaps the most useful method for growing cells on a large scale for vaccine production is on plastic or Sephadex beads (microcarriers) maintained in suspension in large fermentation tanks.

Of paramount inportance for virologists is the selection of cell lines that will allow optimal growth of the virus under study. Some viruses replicate in almost any cell line and some cell lines support the replica tion of many different types of viruses. On the other hand, many viruses are quite restricted in the kinds of cells in which they can be isolated from an infected animal. On adaptation by serial passage, however, mutants with somewhat greater growth potential for a given cell line can be selected. A useful general rule is to use a cell strain of low passage number from the same animal species for primary isolation of a virus. Once the virus has been isolated, alternative more convenient cell sub strates may be sought.

Cultured cells serve three main purposes: (1) isolation of viruses from clinical specimens (see Chapter 13), for which purpose a type of cell culture should be selected which is known for its high sensitivity and in which cell abnormality is readily recognized, (2) production of vaccines and antigens for serological diagnosis, for which the principal require ment is for a cell line giving a high yield of virus and free from con taminating agents (see Chapter 14), and (3) biochemical studies of viral replication (see Chapter 4), for which continuous cell lines, preferably growing as suspension cultures, are usually chosen.

The growth of viruses in cell culture can be monitored by a number of biochemical procedures indicative of the intracellular increase in viral macromolecules and virions (see Chapter 4). In addition, there are sim- pier methods that are more commonly used for diagnostic work (see Chapter 13), as outlined below.

Cytopathic Effects. Many viruses kill the cells in which they replicate, so that infected cell monolayers gradually develop visible evidence of cell damage, as newly formed virions spread to involve more and more cells in the culture. These changes are known as cytopathic effects (CPE), and the responsible virus is said to be cytopathogenic. Most cytopathic effects can be readily observed in unfixed, unstained cell cultures, under low power of the light microscope with the condenser racked down and the iris diaphragm partly closed to obtain the required contrast.

A trained virologist can distinguish several types of cytopathic effects, even in unstained, living cultures (Plate 3-2, Table 3 -1). Fixation and staining of the cell monolayer reveals further diagnostic details, notably inclusion bodies (Plate 6-1, Fig. 6-1) and syncytia (Plate 3-2C). These mor phological consequences of viral infection are discussed in Chapter 6.

Hemadsorption and Hemagglutination. Cultured cells infected with orthomyxoviruses, paramyxoviruses, or togaviruses, all of which bud from cytoplasmic membranes, acquire the ability to adsorb erythrocytes. This phenomenon, know as hemadsorption, is due to the incorporation into the plasma membrane of newly synthesized viral protein that binds red blood cells (Plate 3-2D). Hemadsorption can be used to demonstrate infection with noncytopathogenic as well as cytocidal viruses, and can be demonstrated very early, e.g., after 24 hours, when only a small number of cells in the culture are infected. Hemagglutination is a differ ent, though related phenomenon, in which erythrocytes are aggluti nated by free virus (see below). Virions or hemagglutinin may thus be demonstrated in the supernatant fluid of an infected culture.

Immunofluorescence. Newly synthesized intracellular viral antigen can be detected by staining the fixed cell monolayer with specific anti viral antibody which has been labeled with a fluorescent dye, or with an enzyme such as peroxidase. Full details of these techniques are given in Chapter 13.

Interference. The replication of one virus in a cell usually inhibits the replication of another virus (see Chapters 6 and 8). The viruses of rubella and of the common cold were first discovered by showing that infected cell cultures, which showed no cytopathic effect, were nevertheless re sistant to challenge with an unrelated enterovirus. Cell lines have now become available in which these viruses produce cytopathic effects, but interference is still used for the diagnosis of bovine virus diarrhea virus.

Prior to the 1950s, when cell culture began to be widely adopted for the cultivation of viruses, the standard host for the cultivation of many viruses was the embryonated hen's egg (developing chick embryo). The technique was devised by Goodpasture in 1930 and was extensively developed by Burnet over the ensuing years. Nearly all of the viruses that were known at that time could be grown in the cells of one or another of the embryonic membranes, namely the amnion (plus the lung of the chick within the amniotic sac), allantois, chorion, or yolk sac.

Although now largely supplanted by cultured cells, embryonated hen's eggs are still used for the isolation and cultivation of many avian viruses (Table 3 -2). The chorioallantoic membrane and the amniotic sac are convenient and sensitive substrates for the growth of many poxviruses and influenza viruses, respectively, and intravenous inoculation is used for the isolation of bluetongue viruses. Furthermore, the allan tois produces such high yields of certain viruses, notably influenza virus Death of embryo and Newcastle disease virus, that this system is still used by research laboratories and for vaccine production.

Like embryonated eggs, laboratory animals have almost disappeared now from diagnostic laboratories, since cell cultures are so much simpler to handle and much more versatile. Nevertheless, infant mice are still used for the isolation of arboviruses, and the natural host animal is still used for agents that do not grow in cell culture, or when a positive isolation is of critical importance.

Animals are still essential for many kinds of virological research. Ex periments on pathogenic mechanisms and the immune response are commonly carried out in the natural animal host, or in inbred strains of mice. Hamsters and other rodents are widely used in tumor virology, because they are highly susceptible to tumor production by a number of tumorigenic viruses. Finally, since serology looms large in much viro logical research, laboratory animals, usually rabbits, are extensively used for producing antisera. Mice are also commonly used for intraperitoneal implantation of hybridomas which secrete monoclonal anti body as an ascitic fluid.

All scientific research depends on reliable methods of measurement, and with viruses the property we are most obviously concerned with measuring is infectivity. The content of infectious virions in a given suspension can be titrated by infecting cell cultures (or, more rarely, chick embryos, laboratory animals, or the natural host) with dilutions of viral suspensions, then observing for evidence of viral replication. Two types of infectivity assays should be distinguished: quantitative and quantal.

A familiar example of this type of assay is the bacterial colony count on an agar plate. Each viable organism multiplies to produce a discrete clone, recognized as a colony, and the colony count therefore represents a direct estimate of the number of viable bacteria originally plated. The parallel in virology is the plaque assay, using monolayers of cultured cells.

Plaque Assays. In 1952 Dulbecco introduced a modification of the bacteriophage plaque assay into animal virology, and this is now the standard procedure for the quantitàtion of most animal viruses. A series of 10-fold dilutions of a viral suspension is inoculated onto monolayers of cultured cells for an hour or so to allow the virions to adsorb to the cells. The infected cells are then overlaid with medium in an agar or methylcellulose gel, to ensure that the spread of viral progeny is re stricted to the immediate vicinity of the originally infected cell. Hence, each infective particle gives rise to a localized focus of infected cells that becomes, after a few days, large enough to see with the naked eye as an area of cytopathology. To render the plaques more readily visible, the cell monolayers are usually stained with neutral red or crystal violet; the living (uninfected) cells take up the stain and the plaques appear as clear areas against a red or purple background (Plate 3-3). Some viruses, e.g., herpesviruses and poxviruses, will produce plaques even in cell monolayers maintained in liquid medium, because most of the newly formed virions remain cell-associated and plaques form by direct spread of virus to adjacent cells. Infection with a single virus particle is sufficient to form a plaque; the infectivity titer of the original viral suspension is expressed in terms of plaque-forming units (PFU) per milliliter.

Transformation Assays. Some oncogenic viruses do not kill cells but "transform" them (see Plate 6-3), so that they display reduced contact inhibition, and grow in an unrestrained fashion to produce a heaped-up "microtumor" that stands out conspicuously against the background of normal cells in the monolayer. Like malignant cells excised from a tu mor, the transformed cells have also acquired the ability to grow in medium in semisolid agar or methylcellulose. Both properties have been exploited by tumor virologists to assay tumorigenic viruses. [From E. D. Kilbourne, In "Fundamental Techniques in Virology" (K. Habel, ed.), pp. 154, 155. Academic Press, New York; courtesy Dr. E. D. Kilbourne.] Pock Assays. A much older assay, still occasionally used for the poxviruses, is the titration of viruses on the chorioallantoic membrane of the chick embryo. Newly synthesized virus escaping from infected cells spreads mainly to adjacent cells, so that each infecting particle even tually gives rise to a localized lesion, known as pock. The morphology and color of the pock is often characteristic of a particular group of viruses or even a particular mutant.

The second type of infectivity assay is not quantitative but quantal, i.e., it does not measure the exact number of infectious virus particles in the inoculum, but only whether there are any at all. Being an all-or-none assay, it is not as precise as a quantitative assay. Serial (e.g., 10-fold) dilutions of virus are inoculated into several replicate cell cultures, eggs, or animals. Adequate time is allowed for virus to replicate and spread to destroy the whole cell culture, or kill the embryo or animal, as the case may be. Hence, each host yields only a single piece of information, namely, whether or not it was infected by that particular dilution of virus.

The results of typical quantal and quantitative infectivity titrations are IO" 2 C, C, C, C, C c + + + + + 10-3 50, 42, 54, 59, 45 + + + + + IO-4 5, 7, 3, 6, 4 + + -+ + IO" 5 0, 0, 1, 0, 1 -+ 10" 6 0, 0, 0, 0, 0 Titer 10 5 4 PFU per ml 10 5 · 2 TCID 50 per ml ^Inoculum: 0.2 ml. ''Each symbol (+ or -) represents the result in one tissue culture tube or animal. C C, Confluent (uncountable).

given in Table 3 -3. The end point of a quantal titration is taken to be that dilution of virus which infects (or kills) 50% of the inoculated hosts; the titer of the original virus suspension is then expressed in terms of 50% infectious doses (ID 50 or TCID 50 in tissue cultures) or 50% lethal doses (LD 50 ) per milliliter. Statistical procedures must be used to calculate the end point of quantal titrations. Commonly used procedures are those introduced by Reed and Muench, and Karber, details of which can be found in laboratory manuals.

Many viruses contain, in their outer coat, virus-coded proteins capa ble of binding to erythrocytes (Table 3-4) . Such virions can, therefore, bridge red blood cells to form a lattice. This phenomenon, known as hemagglutination (HA), was first described in 1941 by Hirst, who then went on to analyze the mechanism of hemagglutination by influenza virus. The hemagglutinin of influenza is a glycoprotein, which projects from the envelope of the virion (see Plate 26-1). The virus will attach to any species of erythrocyte carrying complementary receptors, which are glycoproteins of a different sort. Hemagglutination by influenza virus and the paramyxoviruses is complicated by the fact that the virions also carry an enzyme, neuraminidase, which destroys the glycoprotein re ceptors on the erythrocyte surface (by removing terminal neuraminic acid) and allows the virus to elute, unless the test is carried out at Rat, mouse, chick Goose; pH critical Goose, 4°C Human ambient or lower temperature. About 10 7 influenza virions are needed to cause agglutination of sufficient erythrocytes to permit the test to be read with the naked eye. Thus, hemagglutination is not a sensitive indicator of the presence of small numbers of virions, but because of its simplicity it provides a very convenient assay if large amounts of virus are available (see Plate 13-3).

Negative staining with potassium phosphotungstate makes it possible to count the number of particles in viral suspensions by electron micro scopy. There are three main approaches. The viral suspension can be mixed with a known concentration of polystyrene latex particles (or with a morphologically distinct virus of known particle count), to provide an easily recognizable marker; the ratio of "unknown" virus particles to latex beads enables its concentration to be determined. Alternatively, a known volume of viral suspension is deposited on the grid by ultra centrifuga tion. Another method is to remove water and salts from a drop of viral suspension hanging from the underside of an ultrathin carbon-coated plastic ("Formvar") film mounted on a copper grid, by diffusion downwards into agar; thus the number of particles in a known volume may be counted. Electron microscope count 10 10 virions Quantal infectivity assay in eggs 10 9 egg ID 50 Quantal infectivity assay in cultured cells 0 10 7 · 8 TCID 50 Quantitative infectivity assay by plaque formation 0 10 8 PFU Hemagglutination assay IO 3 HA units fl In the same cell line.

If a given preparation of viral particles were to be assayed by all of the methods described above, the "titer" would be different in every case. For example, an influenza virus suspension may provide the data set out in Table 3 -5. The difference between the electron microscope and hemagglutination titers reflects merely the difference in sensitivity be tween the two assays. On the other hand, the ratio between the infec tivity titer and the particle count, known as the infectivity to particle ratio, requires deeper analysis. In part, it is explained by the fact that most of the virions visible by electron microscopy are noninfectious, having being inactivated by heat (even at 37°C) or by other mechanisms during extraction and purification, or having been assembled unsatisfac torily in the first place (typically having a defective genome, hence known as "defective" or "incomplete" virus-see Chapter 5). However, the situation is complicated by another factor, namely, the susceptibility of the cell system in which viral infectivity is assayed; here one speaks of the plating efficiency of the virus. Even a fully infectious virion has only a certain chance of successfully negotiating all the barriers it may encoun ter in the course of entering and infecting a cell. The susceptibility of one cell system may be much lower (i.e., resistance much higher) than that of another, hence a given viral preparation may have a lower efficiency of plating in one system than in the other; for example, there is a 1.2 log 10 difference between egg ID 50 and TCID 50 in Table 3-5. FURTHER READING Cooper, P. D. (1967) . The plaque assay of animal viruses. In "Methods in Virology" (K.

Maramorosch and H. Koprowski, eds.), Vol. 3, pp. 244-311. Academic Press, New York.

Culture of Animal Cells: A Manual of Basic Technique

New Developments in Practical Virology

Virus-erythrocyte interactions

Advanced methods in electron microscopy of viruses

Altered cellular morphology resulting from cytocidal virus infection

Transformation assays