No abstract

Electron microscopic study of the H.R. strain of herpes simplex virus in cells of the chicken chorioallantoic membrane (1) suggested that the virus differentiates within the nucleus, where particles possessing a central core and a single limiting membrane are randomly dispersed. In this earlier work the finding that ruptured nuclei were associated with large numbers of intracytoplasmic particles possessing double membranes appeared to be consistent with the hypothesis that the virus acquired a second membrane in the cytoplasm after release from the nucleus. However, cells in tissue culture infected with the H F E M strain of herpes simplex virus and examined by Stoker, Smith, and Ross (2) and herpes B virus examined by Reissig and Melnick (3) subsequently revealed intranuclear particles with double as well as single membranes, thus raising the possibility that development could be completed in the nucleus.Recently, a strain (J.M.) of herpes simplex virus was found to crystallize (4) in a manner analogous to the adenoviruses (5-9). Moreover, the nuclear membranes of cells infected by this strain frequently showed remarkable proliferation. The purposes of this paper are to illustrate and describe the viral crystals as well as the morphology of the cellular response to infection, and to propose an hypothesis concerning the manner of development and the mechanisms whereby virus may gain egress from intact cells. Materials and MetkodsHeLa cells were cultured in a medium consisting of Earle's balanced salt solution which contained 0.5 per cent lactalbumin hydrolysate, 0.25 per cent glucose, 0.1 per cent yeast extract, 0.3 per cent tris(hydroxymethyl)-aminomethane and 20 per cent horse serum. Stable human amnionic cells were cultivated in Eagle's medium to which 20 per cent horse serum * These studies were

Electron microscopic studies of influenza virus have been reported periodically for more than a decade, but the knowledge thus gained concerning the structure and especially the development of this group of viral agents is still rudimentary.Early work by Taylor, Sharp, and associates (I-3) disclosed that strains of swine, type A (PRS) and type B (Lee) virus, isolated by centrifugation from the chorioallantoic fluid of infected chicken embryos, were composed of sphcrical or ovoid particulate units, with average diameters of approximately 78, 78, and 97 m/~, respectively. These investigators also observed that the viral particles apparently possessed an "internal differentiation in the structure, marked by a single region of relatively high density in the individual particles." Mosley and Wyckoff (4) first described elongated or filamentous forms in preparations of the PRS, Weiss, and Lcc strains of virus, noted that these rod-like structures frequently appeared to bc partly segmented into spherical particles having the same dmmeter, and suggested that there was a significant relationship between the two forms. It has since been shown that both filamentous and spherical forms can usually be demonstrated in preparations of influenza virus from infected chorioaUantoic fluids, regardless of the strain employed, although filaments are espccially numerous in recently isolated A strains (5). Moreover, filaments have been seen in tissue cultures of infected chorioallantoic membrane (6), as wcU as in thin sections of infected membrane and mouse lung (7-9). In connection with these latter observations, it has been pointed out that the viral particles appear to develop solely from the surface of cells and that the filaments often show segmcntation into spheres. Consequently, it is scarcely surprising that "the most general interpretation is that the filaments represent an intermediate stage in the multiplication of virus and that the viral par-

Western equine encephalomyelitis (WEE) virus is 40 to 55 m/z in diameter (1) and apparently contains nucleic acid of the ribose type (2, 3). Although electron microscopic examination of whole mounts of tissue cultures infected with a closely related agent (Eastern equine encephalomyelitis virus) has shown viral particles both within and on the surface of cells (4,5), no studies employing thin sections have been previously reported. The purpose of this communication is to illustrate and describe the manner in which WEE virus appears to differentiate within, and gain egress from, infected tissue culture cells as revealed in thin sections by the electron microscope. Similarities in development or release among WEE virus, influenza virus, herpes simplex virus, the virus associated with erythroblastosis of chickens, Rous sarcoma virus and a mouse mammary tumor agent will be discussed. Stable human amnionic cells:A stable line of human anauionic epithelium was kindly supplied by Dr. Katherine Sprnnt. Replicate cultures were prepared in Leighton tubes without coverslip% using Eagle's basal medium with 10 per cent horse serum. The cultures were inocu-* These studies were aided

Although capsids of herpes simplex virus were encountered within phagocytic vesicles, they were more commonly observed free within the cytoplasm. Stages in the release of virus from vesicles were not seen. There appeared to be five distinct steps in the process whereby the virus initiates infection: attachment, digestion of the viral envelope, digestion of the cell wall, passage of the capsid directly into the cytoplasm, and digestion of the capsid with release of the core. Antibody probably interferes with the first two stages. This paper describes and illustrates the probable manner in which herpes simplex virus (Herpesvirus hominis) initiates cellular infection. The three ensuing papers are concerned with the sequential stages of viral development, the effects of blocking DNA synthesis, and the antigenic alteration of host cell components. Since the initial examination of herpes simplex virus in thin sections (21, 22) a confusing nomenclature has arisen, including such descriptive terms as internal body, nucleoid, central body, internal membrane, peripheral coat, second membrane, single membrane form, naked particle, double membrane form, and complete virus. ["Nucleoid" was originally coined to describe the dense, eccentrically placed structure in immature forms of vaccinia and fowl pox viruses (23). It has become an unfortunate misnomer when applied to the dense cores characterictic of the majority of viruses.l Application of the negative staining technique, however, has now provided more precise information about the structure of the virus (33), which indicates that it consists of a core surrounded by a capsid. The capsid is icosahedral in shape and is composed of 162 capsomeres arranged in 5:3:2 axial symmetry. Capsids may or may not be enclosed within an envelope that is devoid of clearly defined subunits. Applying this nomenclature (19) to the appearance of the virus in thin sections, one can recognize a clearly defined core, a capsid (the first or inner membrane), and an envelope (the second or peripheral membrane). These terms will be used henceforth.

Examination of infected cells at sequential intervals after infection revealed that the first viral forms to appear were capsids enclosing cores of low density. Not until the 6th hr were dense cores encountered, and at approximately the same time enveloped virus was seen. Envelopment occurred most frequently in close proximity to the nuclear surface, although the process was also encountered within the nuclear matrix and in the cytoplasm. There was often extensive proliferation of the nuclear membrane. Envelopment of the virus by budding from the cell surface was not observed. It was concluded that enveloped virus consitutes the infectious particle and

The mode of development of herpes simplex virus has been the object of periodic study ever since Lipschiit2, (1) discovered that intranuclear inclusion bodies characteristically appear in the cells of tissues infected by this agent. Interest has centered chiefly on attempts to determine the nature of the nuclear changes but the results have not been concordant. Some observers have concluded that the inclusions actually represent a stage in viral multiplication and are composed at least in part of viral material. Others have inferred that they consist solely of abnormal products of cellular metabolism. The earlier work bearing on this subject has been well reviewed by Van Rooyen and Rhodes (2). More recently, Crouse et al. (3) carried out histocbemical studies which revealed a distinct accumulation of desoxyribese nucleoprotein in early nuclear inclusions. This work was confirmed by Wolman and Behar (4) who considered the findings to be corroborative of the hypothesis that the herpetic inclusion body represents a colony of virus inside a cellular matrix. Scott el al. (5) have extended these observations by a careful study wherein the development of infectious virus was followed in temporal relation to the cytologic changes. These authors concluded that the nucleus probably contains virus during the early formation of the inclusion body. In contrast, Francis and Kurtz (6) and Ackermann and Kurtz (7) reported data on the infectivity of cellular fractions obtained by differential centrifugation which they interpreted as indicating that herpes simplex virus is not associated with the nucleus.A more direct approach to this problem can be made by examining infected tissues in the electron microscope. Morgan et al. (8) have previously reported that altered nuclei could easily be identified by means of the electron micro-* This investigation was supported by grants from

Examination in the electron microscope of ultracentrifuged preparations of vaccinia and fowl pox viruses has shown the particles generally to be brickshaped with dimensions about 210 X 210 X 260 m/~ and 265 X 265 X 330 m/~, respectively (1).In 1942, Green, Anderson, and Smadel (2) noted that va~cinia virus appeared to have a limiting membrane and an internal structure composed of a central zone of greater density, often surrounded by four small bodies. The presence of an inner body was confirmed by Sharp e~ a2. (3). Dawson and McFarlane (4) observed that the central body resisted pepsin digestion, whereas the peripheral area became less opaque to the electron beam. They noted a limiting membrane but suggested that the satellite bodies, described by Green a 02., were artefacts. Internal structure has also been seen by others (5-7), and Peters and Nasemann (8) have illustrated and described a small, dense body located at the periphery of some viral particles, which they consider to represent developmental forms. Bang e~ 02. (9) observed a dense, central body in fowl pox virus and also noted a decrease in density at the periphery of the virus particle after treatment with pepsin.Although vaccinia (10, 11) and fowl pox (12) viruses have previously been identified in sectioned tissues, the relatively thick sections employed and the distortion produced by removal of the embedding plastic with solvents have tended to obscure details of viral structure. More recently, Gaylord and Melnick (13), employing thinner sections without removal of methacrylate, noted several forms of vaccinia virus, including "empty circles" and "circles filled with homogeneous material of low electron density," as well as internal structures described as granules, bars, and dumbbells.In order to obtain further information concerning the structure and development of vaccinia and fowl pox viruses, ultrathin sections of infected * This investigation was supported by grants from