key: cord-0041814-yh5ajvxc authors: BUCHANAN, J. S.; MADELEY, C. R. title: Studies on Herpesvirus scophthalmi infection of turbot Scophthalmus maximus (L.) ultrastructural observations date: 2006-04-07 journal: J Fish Dis DOI: 10.1111/j.1365-2761.1978.tb00033.x sha: 3b425a472b08657ac0bbfb73d986ec81a0e8fd57 doc_id: 41814 cord_uid: yh5ajvxc Abstract. Recent heavy mortalities amongst O+ group turbot at a fish farm were found to be associated with a herpes‐type viral infection of the epithelia of the skin and gills. The morphology of the virus is described with ultrastructural observations on its morphogenesis and release from infected cells. Herpes-like viruses have been described from channel catfish, Ictalurus punctatiis (Rafinesque) by Wolf & Darlington (1971) ; from carp, Cyprinus carpio L, by Schubert (1966) and from fresh water salmonids, by Wolf & Taylor (1975) , Wolf, Herman, Darlington & Taylor (1975) and Wolf, Sano & Kimura (1975) , In a preliminary note (Buchanan, Richards, Sommerville & Madeley 1978) we reported the presence of a herpes-like virus in skin and gill lesions of farmed turbot Scophthalmus maximus (L.) and that similar lesions are also found less frequently in wild fish. described the histopathology of this disease as seen in the light niicroscope and this paper describes the electron microscopic findings, including stages in the replication of the virus. Details of the infected and control fish used in this study have already been published . Wet preparations of skin scrapings were examined by phase contrast light microscopy for the presence of the characteristic giant cells and those showing a high proportion were fixed in 2-5% glutaraldehyde in 0-05 M cacodylate buffer overnight. After thorough rinsing in buffer solution they were post-fixed in 1% osmium tetroxide m 0-05 M sym-coUidine buffer for 1 h at 4°C followed by dehydration in a graded series of alcohols and embedding in epoxy resin (EMscope Ltd.) . Sections were cut on a Reichert OMU 3 ultramicrotome, stained with uranyl acetate and lead citrate or with 1% potassium permanganate and examined in a JEOL lOOC electron microscope. Similar preparations were made from control turbot from which giant cells were absent. The blocks were trimmed on an LKB Pyramitome and 0-2 fim sections were cut and stained with toluidine blue. These were examined under phase and differential interference phase optics to locate the giant cells and serial 500 A sections were taken for examination by electron microscopy. Scrapings of skin from infected and normal fish were suspended in Eagle's tissue culture medium. Approximately 1 ml of skin scrapings and medium were transferred to a ten Broek homogenizer and 2 ml of distilled water were added. After homogenization the resultant slurry was clarified and the supernatant concentrated by further centrifugation at 100,000 g for 30 min. The pellet was resuspended in 2 drops ol EM diluent (0-1% Bacitracin) and mixed with an equal volume of negative stain. One drop of this mixture was applied to a carbon-Formvar coated grid and the surplus drawn off with the edge of a filter paper. The grid was then examined in a Philips oU electron microscope at an accelerating voltage of 80 kV or in a JEOL lOOC at an accelerating voltage of 100 kV. The negative stains used were 3% potassium phosphotungstate (PTA) and 4/o ammonium molybdate, both at pli 7. Initially, a variety of treatments were used remove mucus from the skin scrapes but these were later found not to be necessary- Examination of 0-2/an sections of epoxy resin embedded skin scrapings by difiere tial interference phase contrast and phase contrast light microscopy showed that , ,\ (h) Serial section of (a). This transmission electron micrograph shows that the granulesâ re clumps ol' enveloped viruses mostly embedded in an electron-dense matrix. The nucleus ( / ^t ains only unenveJoped viruses. An unaffected host cell at top loft ]ias complex cyto]jlasmic prothat intcrdigitate with the giant cell (x6000), . ( c) Detail from (b) above. Note the nuclear envelope, Desmosomes are shown (arrowed) at right. Tlie cytoplasm of this giant cell contains tlie l-emnants of liost cell organelles su mitochondria and rough endo])lasmic reticulum. Tho majority of clumps of virus are bound by unit membranes (x 12,000). Ig^, (d) These virus aggregates are details from the same giant coll and show the finely graniua . (,ron dcn,s(; matrix surrounding tlie virus particles. Some of the particles ap])ear to be dccayec ^^^ĝ raded. In some the capsids and cores are absent and the outer envelope appears broken open, of membrane of unlinown origin occupy the clump at lower left (x 125,000). giant cells seen by light microscopy and described by were polynucleate (Fig. la) . These giant cells were fbrmed by the dissolution of the cj'toplasmic membranes between cells with, the consequent formation of syncytia. Subsequently, the nuclei within the.se syncytia fused to form a single giant nucleus which was usually oval with a maximum width of 60 yim and length of 100 /(m. Numerous osmiopliilic inclusion bodies of between 0-4 and 0-9 /(m were scattered throughout the nucleoplasm (Fig. Id) , The nuclei were surrounded by a layer of cytoplasm 10-30 /.(m thick. The whole giant cell was therefore 70-130 /nn in diameter and was approximately 1000 times the volume of the surrounding Malpighian cells of the epidermis. Serial section through individual giant cells enabled a picture of their ultrastructure to he built uj). Details of the junction of giant cells with surrounding Malpighi^ĉ ells are shown in Fig. ] b and c. Tho cytoplasm of the giant cells varied between vacuolato and highly granular in aiDpearance and the cytoplasmic membrane showed multiple interdigitations with that of the overlying Malpighian cells. Desmosomes were frequently observed at points along the junction between giant cells and Malpighian cells (Fig. lc) . Nuclei similar to those of the surrounding Malpighi^iĉ olls were occasionally found in the cytoplasm of these giant cells with their envelopes intact. Electron mi croscopy of tliin sections showed that tlie nuclei of giant cells containe finely granular material in which dense paracrystalline arrays of particles, 100 nm in diameter, were emhedded. These resembled herpesvirus capsids with hexagonô r spherical outlines (Figs lc, 2b & c, 3b & c) . These dense clumps of virus corresponded with the nuclear inclusions visible by light microscopy and illustrated in Fig'- ^^' Approximately Jialf of these particles contained electron-dense cores and both the outer layer and coire were found to correspond in size and shape to particles seen nn egatively stained preparations from aifected fish. In thin section the core appearec either as an electron-dense bar or as an electron-dense ring surrounding an electrontranslucent centre, Capsids both with and without cores were arranged at the peiiphery of tlie nucleus adjacent to the nuclear envelope and enveloped particle were seen in large numbers immediately adjacent to the cytoplasm. Direct buddmt, was not observed but virus particles in cytoplasmic vacuoles were invariably enve oped and it was probable tliat t]ie envelope was acquired from t]ie inner membran of the nuclear envelope. Occasionally, the nuclear membrane appeared to ruptur or to disintegrate so that unenveloped virus was seen in tho cytoplasm; the capsi then appeared to acquire envelopes by budding into cytoplasmic vacuoles (Fig-' ^"'^' The envelope surrounding the capsid.in cisternae immediately adjacent to the nucleai envelope was often pear shaped or 'tailed'. These perinuclear cistornae also containe 'hour glass' shaped envelopes without contained capsids but bearing the 18 nm long spikes associated with the outer surface of enveloped viruses (Fig-3c ;-Ê nveloped virus particles in vacuoles furtlier from the nucleus were more regu and tliese more bizarre forms were not seen. Fig, 4 this electron graph shows a herpes simplex virus particle and its envelope (x 200,000), Stained with 2% PI A, 1 ^( h) Detail of the junction between nucleus and cytoplasm of the giant cell shown in Fig, 2 . 1 ^ ŝ ome evidence that a process of budding of particles from the nucleus through the inner lai" t he nuclear envelope gives rise to the enveloped viruses in the perinticlear cisternae. In place inner membrane of the nuclear envelope appears to have disintegrated and virus particles may their envelope by budding into cytoplasmic vacuoles (x 25,000), 2 (c) Further detail of the junction between nucleus and cytoplasm of the giant cell shown m ^^N ote the presence of numbers of pear shaped or tailed particles in the perinuclear cisterna ' diameter of the enveloped viruses with their fringe of spikes is 200-220 nm. The mean diameter o capsids within tho nucleus is 100 nm (x 125,000). Virus appeared to be released from the cell at the periphery of the cytoplasm either singly or in membrane-bound aggregates (Figs lb & 2b) . The cytoplasm of giant cells contained very large amounts of virus most of which appeared to lie within membranebound vacuoles. These were so numerous that it seems possible that they were cytoplasmic channels connecting the perinuclear cisternae to the exterior (Fig. 2b) . The virus could be released by migration via these intracytoplasmic tubules from the nucleus to the exterior. A proportion of giant cells contained membrane-bound aggregates of enveloped capsids embedded in a granular matrix (Fig. lb & c) . These clumps of virus were scattered throughout the cytoplasm and corresponded in size and shape to the osmiophilic granules seen by phase contrast light microscopy and illustrated in Fig. la . Frequently, these clumps included enveloped viruses that appeared to be inidergoing degradation. The particles were often swollen with the outer envelope broken and the capsid either absent or poorly defined. Mature enveloped virions showed 5 concentric zones surrounding the coi-e. The outer zone consisted of 18 nm long spikes which were attached to a trilaminar membrane forming the envelope or second zone (Fig. 3c) . The envelope was separated from the capsid by a third, electron translucent layer containing various amounts of fibrillar material. The capsid (4t]i zone) was seen as an electi-on-dense ring that was frequently hexagonal in outline and this was separated from the core by an electrontranslucent space (5th zone). The diameter of the enveloped particle, including spikes. Was between 200 and 220 nm. Negatively stained preparations of skin scrapes with or without partial purification by centrifugation showed numbers of enveloped and unonveloped viruses closely resembling members of the herpesvirus group, but with some differences (Fig. 4a-e) . The unenveloped capsids had a mean diameter of 100 nm (range 85-110 nm) with subunits of about 9 nm in diameter. Where stain had penetrated into the middle of a particle tlie subunits appeared as hollow tubes and this gave a castellated edge to the periphery of the capsid (Fig. 4a) . Surface details were not seen clearly enough to establish the exact structure but the arrangement of subiniits was very similar to that of the members of the herpes group. Despite careful searching and the use of standard staining conditions, details of the surface structure were never as clearly as they are on herpes simplex virus (Fig. 3a) . Enveloped particles had a surface of i^etal shaped spikes resembling those of coronaviruses. Their 18 nm length the spacing between them were the same as the spikes seen in thin sections of Slant cells. Frequently, the envelope was drawn out into a tail giving a tadpole b .tfi."f''' appearance to the particle. An intact envelope appeared to be impervious to the stain and this prevented the surface details of the capsid being seen (Fig. 4a ). Attempts are being made to culture Herpesvirus scophthalmi but, to date, no cell line has been found that will support the growth of this marine virus in vitro. Until this can be done full characterization of this new virus will not be possible. Nevertheless, the evidence we present indicates that the pathognomonic giant cells are virus 'factories', and that the virus produced has many of the characteristics of the herpesvirus group. This evidence consists of: (1) the formation of giant cells, syncytia and giant nuclei; (2) the development of intranuclear inclusions; (3) the assembly of capsids in the nucleus; (4) the envelopment of capsids at the inner nuclear membrane; (5) the architecture of the virus as shown by negative staining. These points will be discussed in relation to known members of the herpesvirus group. ( The primary cytopathic effect we have observed was that of virus-induced cell fusion. Polykaryocytosis is characteristic of herpesviruses (Pereira 1961; Roizman 1978) . Ultrastructural evidence of cellular fusion came from observations of desmosomes at the junctions between giant cells and the surrounding Malpighian cells (Pig. lc). Cook & Stevens (1970) found isolated desmosomes in the cytoplasm of cells infected with varicella zoster virus and interpreted this as evidence of virus-induced cell fusion. Further, apparently normal Malpighian cell nuclei were found within the cytoplasm of these giant cells but without the envelope of the giant cell nucleus. This suggested that nuclear recruitment was taking place and that giant cell nuclei were formed by the fusion of several nuclei following dissolution of the common membranes between the Malpighian cells. This is similar to the social behaviour of cells in cell (e) Tailed, enveloped virus. The envelope was frequently drawn out into a long tail during the preparative processes for negative staining. All the above electron micrographs are at a magnification of 200,000. cultures infected with herpesviruses described by Roizman (1978) and is compatible with observations made by Wolf & Darlington (1973) who carried out sequential studies of fish cell cultures inoculated with Herpesvirus salmonis. ( Many of the features observed from phase contrast light microscopy of resmembedded skin scrapings match those already described in a previous paper by on the histopathology of this disease. For example, the osmiophilic granules seen in the cytoplasm of some giant cells (Fig. la) corresponded m size and number to the Feulgen positive and occasionally PAS positive granules seen by light microscopy. Our ultrastructural observations showed that these granules consisted of membrane-bound clumps of enveloped viruses embedded in a finely granular electron-dense matrix. Similarly, the dense basophilic and Feulgen positive coarse granules described as one type of intranuclear inclusion could be correlated with dense paracrystalline arrays of unenveloped capsids enclosed within the nucleus of giant cells (Fig. 2a & c) . The second type of intranuclear inclusion which consisted of eosinophilic, Feulgen negative bodies could be correlated with utoastructural observations of areas in the nucleus devoid of virus particles and chromatin and containing only highly dispersed granular material. It seems probable that these inclusions were areas where former intense viral replication had exhausted the contents or a nucleus. Similar intranuclear inclusions in cell monolayers infected with herpes simplex virus have been described as scars left by earlier viral multiplication (Dulbecco & Eisen 1973) . These intranuclear inclusions were considered by Pereira (1961) to be the most characteristic of the cellular lesions caused by vii-uses of the herpes group. ( Herpesvirus replication has been reviewed in detail by Roizman & Spear (19/•' ->' • Our observations of the maturation sequence closely resemble those described from ultrastructural studies of other herpesviruses infecting animal cells (Smith 19bo, Pinkerton, Sun, Henson & Neff 1964; Shipkey, Erlandson, Bailey, Babcock& Southam 1967; Nii, Morgan & Rose 1968; Stackpole ]969; Cook & Stevens 1970; Nazerian & Witter 1970; Nayak 1971; Wolf & Darlington 1971) . Incomplete capsids were often found in infected nuclei. Some appeared to be empt} while others contained cores which were bar-or ring-shaped. Envelopment of the capsid was closely similar to that described by Wolf & Darlington (1971) for channel catfish virus. It occurred at the inner lamellae of the nuclear envelope and by budding into cytoplasmic vacuoles. Our observations support the concept proposed by v* o and Darlington that envelopment can occur wherever the unenveloped capsid en counters host cell membrane. Further evidence in support of this maturation sequenc came from observations of enveloped and partially enveloped virus particles, som having tail-like structures, in the perinuclear cisternae. There seems to be general agreement that within the herpesvirus group envelopment takes place at the inner nuclear membrane but opinions differ regarding the release of virus from cells. Epstein (1962) showed that herpesvirus was released by budding at the cell membrane and Morgan, Rose, Holden & Jones (1959) and Nii, Morgan & Rose (1968) described the release of herpesvirus as a process of reverse phagocytosis. The mode of release through cytoplasmic channels was first described by Schwartz & Roizman (1969) and confirmed by Strandberg & Aurelian (1969) and Jasty & Chang (1972) . Subsequent studies by Fong, Tenser, Hsuing & Gross (1973) showed that mature nuclear virus particles were first released into perinuclear cisternae and then travelled through cytoplasmic channels to the extracellular space. The results of the present study indicate that this method was the one by which Herpesvirus scophthalmi was released from giant cells. It seems likely that the cytoplasmic channels shown in Fig. 2b were modified areas of pre-existing endoplasmic reticulum (ER). Extracellular virus particles were frequently observed to be aggregated in membrane-bound clumps some of which contained numerous enveloped virus particles bound together in a finely granular matrix. Similar aggregations were observed in the cytoplasm of some giant cells and these bore a close resemblance to those described by McGavran & Smith (1965) who showed that clumping was due to virus-cell interaction. They noted that host cell lysosomes surround, engulf and obscure many cytomegaloviruses within the cytoplasm. Similar observations were made by Reubner, Hirano, Slusser, Osborn & Medearis (1966) of cytomegaloviruses embedded in electron-dense material having many of the characteristics of lysosomes suggesting that a host-cell reaction against virus produced in its own nucleus may have taken place. They considered that this reaction could be pathogenetically important in long term infections. The present study has shown that membrane-bound aggregates of virus in giant cells show the virus in differing stages of dismemberment and decay as though under attack from lysosomal enzymes. Further cytochemical studies are required to determine the nature of the matrix that surrounds the viruses. Negative staining has shown that the appearance of the virus accords closely with that of other members of the herpesvirus group as illustrated in comparative studies by Madeley (1972) and Roizman & Furlong (1974) . However, the outer layer of spikes on the envelope appears to be unique to Herpesvirus scophthalmi in that they are approximately twice as long as those reported by Wildy, Russell & Home (1960) The other architectural components of the virus were typical of the herpesvirus group as a whole. In a review, Roizman & Spear (1971) noted that there is general agreement that the typical herpes virion consists of a core 25-30 nm in diameter, a capsid consisting of 162 capsomeres and an envelope (with or without spikes) and thus can be defined as an enveloped nucleocapsid. The present study has shown that the turbot virus possessed a core having a diameter of 25-30 nm and that was compatible with a toroid when seen in thin section. The core was bound by a capsid that was frequently hexagonal in outline and having a mean diameter of 100 nm and a definite substructure of hollow capsomeric units. These were less well defined than in most herpesviruses but they were comparable in size and number to the characteristic hollow herpesvirus capsomeres. The total diameter of mature enveloped particles was 200-220 nm as compared with 160-180 nm for the herpesvirus group as a whole but this was due to the fringe of spikes surrounding the envelope. The relative ease of diagnosis and the unfortunate abundance of infected fish suggest that this new virus would provide an ideal subject for comparative studies within the herpesvirus group. A herpes-type virus from the turbot, Scophthalmus maximus, L. The Veterinary Record Replication of varicella zoster virus in cell culture; an ultrastructura study Herpesviruses. In: Microbiology Observations on the mode of release of herpesvirus from infected HeLa eel s Ultrastructural studies on the envelopment and release of guinea-pig herpes-like virus in cultured cells Release of infectious bovine rhinotracheitis virus from productively infected bovine kidney cells: An electron microscopic study A thin section study of the morphogenesis of Aujeszky's disease virus in synchronously infected cell cultures Ultrastructural, cytochemical and microchemical observations on cytomegalovirus (salivary gland virus) infection of human cells in tissue culture Virus Morphology, 179 pp Electron microscopical observations on the development of herpes simplex virus Isolation and characterization of a herpesvirus from leukemic guinea-pigs Cell free transmission and in vivo replication of Marek's disease virus Electron microscopy of herpes simplex virus 11. Sequence of development The cytopathic effect of animal viruses Human cytomegalovirus: Intranuclear virussynthesizing bodies Cytomegalovirus infection. Viral ultrastructure with particular reference to the relationship of lysosomes to cytoplasmic inclusions Studies on Herpesvirus scophthalmi infection of turbot Scophthalmus maximus L.; Histopathological observations The herpesviruses The replication of herpesviruses Herpesviruses: current information on their composition and structure The infective agent in Carp Pox Concerning the egress of herpes simplex virus from infected cells: electron and light microscope observations Virus biographies II. Growth of herpes simplex virus in tissue culture Physical and biological ohservations of herpesvirus Replication of canine herpesviruses II. Viral development and release in infected dog kidney cells Herpes-type virus of the frog renal adenocarcinoma I. Virus development in tumor transplants maintained at low temperatures The morphology of herpes virus Channel catfish virus: a new herpesvirus of ictalurid fish Salmonid viruses: effects of Herpesvirus salmonis in rainbow trout Herpesvirus Disease of Sahnonids. Fish Disease Leaflet 44 of the U.S. Dept. of the Interior, Fish and Wildlife Service, Division of Cultural Methods Research Salmonid viruses: a syncytium forming agent from rainbow trout We wish to thank Mr T. Forrest of the Biology Department, University of Stirling, for his assistance in the preparation of the electron micrographs and Mr G. Mace oi Goldexi Sea Fish Farms for the provision of turbot. Dr Buchanan is supported by a grant from the Natural Environmental Research Council.