key: cord-0252693-o8oapxpz authors: David-Ferreira, J. F.; Manaker, R. A. title: AN ELECTRON MICROSCOPE STUDY OF THE DEVELOPMENT OF A MOUSE HEPATITIS VIRUS IN TISSUE CULTURE CELLS date: 1965-01-01 journal: J Cell Biol DOI: nan sha: 616d2117cb63fecdaa5ae2901cca9d1b610dd799 doc_id: 252693 cord_uid: o8oapxpz Samples taken at different intervals of time from suspension cultures of the NCTC 1469 line of mouse liver—derived (ML) cells infected with a mouse hepatitis virus have been studied with the electron microscope. The experiments revealed that the viruses are incorporated into the cells by viropexis within 1 hour after being added to the culture. An increasing number of particles are found later inside dense cytoplasmic corpuscles similar to lysosomes. In the cytoplasm of the cells from the samples taken 7 hours after inoculation, two organized structures generally associated and never seen in the controls are observed: one consists of dense material arranged in a reticular disposition (reticular inclusion); the other is formed by small tubules organized in a complex pattern (tubular body). No evidence has been found concerning their origin. Their significance is discussed. With the progression of the infection a system of membrane-bounded tubules and cisternae is differentiated in the cytoplasm of the ML cells. In the lumen of these tubules or cisternae, which are occupied by a dense material, numerous virus particles are observed. The virus particles which originate in association with the limiting membranes of tubules and cisternae are released into their lumen by a "budding" process. The virus particles are 75 mµ in diameter and possess a nucleoid constituted of dense particles or rods limiting an electron transparent core. The virus limiting membrane is sometimes covered by an outer layer of a dense material. In the cells from the samples taken 14 to 20 hours after inoculation, larger zones of the cell cytoplasm are occupied by inclusion bodies formed by channels or cisternae with their lumens containing numerous virus particles. In the samples taken 20 hours or more after the inoculation numerous cells show evident signs of degeneration. Samples taken at different intervals of time from suspension cultures of the NCTC 1469 line of mouse liver--derived (ML) cells infected with a mouse hepatitis virus have been studied with the electron microscope. The experiments revealed that the viruses are incorporated into the cells by viropexis within 1 hour after being added to the culture. An increasing number of particles are found later inside dense cytoplasmic corpuscles similar to lysosomes. In the cytoplasm of the cells from the samples taken 7 hours after inoculation, two organized structures generally associated and never seen in the controls are observed: one consists of dense material arranged in a reticular disposition (reticular inclusion); the other is formed by small tubules organized in a complex pattern (tubular body). No evidence has been found concerning their origin. Their significance is discussed. With the progression of the infection a system of membrane-bounded tubules and cisternac is differentiated in the cytoplasm of the ML cells. In the lumen of these tubules or cisternae, which are occupied by a dense material, numerous virus particles are observed. The virus particles which originate in association with the limiting membranes of tubules and cisternac arc released into their lumen by a "budding" process. The virus particles are 75 m# in diameter and possess a nucleoid constituted of dense particles or rods limiting an electron transparent core. The virus limiting membrane is sometimes covered by an outer layer of a dense material. In the cells from the samples taken 14 to 20 hours after inoculation, larger zones of the cell cytoplasm are occupied by inclusion bodies formed by channels or cisternae with their lumens containing numerous virus particles. In the samples taken 20 hours or more after the inoculation numerous cells show evident signs of degeneration. Acute hepatic disease of mice resulting from virus infection was described in 1951 by Gledhill and Andrews (36) . Shortly thereafter, Nelson (58) , in the course of serial transmission of a spontaneous leukemia in Princeton mice by means of splenic implants, detected a virus responsible for the induction of hepatic lesions in his experimental mice and for the concomitant loss of the transplanted leukemic cells. These and subsequent reports of virus-induced acute hepatitis (6, 45, 55, 56) in experimental mice permit the conclusion that the specific agents responsible for the disease are widely disseminated in mouse populations in a latent state, and that activation follows the systemic stress imposed during experimental procedures. Since some strains of virus produce pathological lesions in the mouse liver similar to those found in human hepatitis (44) , and because trans-mission a n d study of the h u m a n disease is beset with difficulties, the s t u d y of hepatitis in mice h a s received increased attention, D u r i n g recent years, the biology a n d p a t h o l o g y of the disease h a s been u n d e r investigation, a n d some electron microscopic observations h a v e b e e n p u b l i s h e d (51, 69, 72) . T h e s e studies d e m o n s t r a t e d the presence of virus in m o u s e liver a n d described the ultras t r u c t u r a l alterations o c c u r r i n g in affected h e p a t i c cells. N e i t h e r the site n o r the m o d e of virus replication has b e e n reported, I n a n effort to clarify s o m e of the obscure points of the cell-virus relationship, we u n d e r t o o k a n electron microscope s t u d y of virus-cell interaction e m p l o y i n g c u l t u r e d cells. T h e m o r p h o l o g i c a l c h a n g e s observed in infected cells d u r i n g the first 24 h o u r s postinfection are described. CELLS: T h e N C T C 1469 line of mouse liver-derived cells was obtained from Dr. Virginia Evans (30, 43) and maintained in serial subculture in our laboratory. T h e cells are herein designated M L cells. T h e M L cells were adapted in our laboratory to growth on Eagle's minimal essential m e d i u m (28) supplemented with l0 per cent unfiltered horse s e r u m previously heated 30 minutes at 56°C, and containing 50 #g of k a n a m y c i n per ml of complete m e d i u m . T h e same m e d i u m proved satisfactory for suspended M L cell cultures in spinner flasks. VIRUS: T h e A59 mouse hepatitis virus isolated by M a n a k e r et el. (50) was available for this study. T h e virus pool used had undergone 25 passages in M L cells, 7 passages in L 929 cells, and 3 further passages in M L cells. Fluids from infected cultures induce acute hepatitis in mice. In the course of this investigation, another morphologically different particle, herein termed the V L particle, was observed budding from cell mere-branes. No evidence of overt disease which m i g h t be attributed to this agent was detected in mice inoculated with control M L cell culture fluids, nor did this agent induce intracellular change in the control cultures comparable with those observed in cells infected with the hepatitis virus. Hepatitis virus titration was m a d e in M L cells, which are destroyed by this agent. Twenty-four hours after tube cultures containing 375,000 cells were infected with 5.6 X 102 tissue culture (T.C.) ID.~0 of virus, large syncytia involving more t h a n half of the cell sheet were evident. A eytopathic response was evident within 3 days in those tubes that received virus following inoculation of the limiting dilution. O n this basis a titer of 5.6 X 10 ~ T.C. ID56 was determined by the method of Reed and M u e n c h for the virus pool used in this study. M L cells were grown in T-60 flasks at 36.5 to 37°C. Heavy cultures were shaken lightly to dislodge the cells. 100 ml of growth m e d i u m in which was suspended 3.3 X 10 ~ cells was introduced into a spinner flask and incubated at 36.5 to 3 7°C T h e culture was infected with 8 X 106 T,C. ID,o of virus. A similar uninfected suspension culture provided a control. At intervals over a period of 24 hours, beginning 2 hours after addition of virus, cells were withdrawn from the cultures for electron microscopic examination. These samples were compared with cells removed from the control cultures at the same times. Samples taken from the controls or from the infected cultures were centrifuged at 1000 RPM for 10 minutes and the pellets obtained fixed for 1 hour in c h r o m e -o s m i u m tetroxide fixative (15) , in i per cent o s m i u m tetroxide phosphate-buffered at p H 7.4, or in 2 per cent potassium p e r m a n g a n a t e (48, 52) . After fixation the blocks were dehydrated in increasing concentrations of ethanol and embedded in Epon 812 mixtures (49) . T h e thin sections were cut on an L K B ultratome with a d i a m o n d knife and picked up on Formvar-coated grids. In order to improve the contrast we double-FIGUaE 1. Portion of the nucleus (iV) and cytoplasm of an M L cell from a non-infected culture. In the cytoplasm are seen mitochondria (m) dense corpuscles (L) and the Golgl apparatus (G). The Golgi apparatus is surrounded by numerous small vesicles. The arrow on the cell surface indicates a VL-partMe of the type habitually seen in these cells. M ~4,000. 3. Virus-like particle budding from the cell membrane of a non-infected M L cell. This particle, which is similar to the "immature" C particles observed in the murine leukemias, has its outer membrane covered by small spicules. )< 1~0,000. T h e M L cells (Fig. 1 ) are round or oval and approximately 80 to 120 # in diameter. They possess one or two nuclei centrally located which present some variation in size and shape from cell to cell. T h e nuclear envelope as usual is formed by two membranes with pores, a n d the nucleoprotein consists of a granular or filamentous component randomly distributed. In each nucleus one or two nucleoli are habitually seen. In the cytoplasm, the Golgi complex is apparent generally in a juxtanuclear position (Fig. 1 ). Mito-chondria, ribosomes, rough membranes of the endoplasmic reticulum, and occasionally small lipid inclusions are diffusely scattered through the cytoplasm. Another cytoplasmic component observed in variable n u m b e r in almost all the cells is round or oval dense corpuscles of different dimensions ( Fig. 1 ). They are limited by a m e m b r a n e and have a dense, generally homogeneous matrix in which myelin-like structures are sometimes observed. In some cells fine fibrils 70 to 75 A in width have been observed in the cytoplasmic matrix. Their a m o u n t a n d distribution is variable. They can be irregularly distributed among the cell organelles, but sometimes they are concentrated into oriented bundles near the nuclear membrane. In a small percentage of the cells, dense granules 200 A in diameter with the staining characteristics of glycogen granules are observed scattered a m o n g the cytoplasmic organelles. At some points of the plasma m e m b r a n e a peculiar differentiation 1200 to 1600 A long is frequently observed, characterized by a denser aspect of the cell membrane which is covered on the cytoplasmic side by a dense fibrous material. Some of these formations are observed invaginated into the cytoplasmic matrix ( Fig. 2 ). Special cytoplasmic vesicles presenting a limiting membrane with the same constitution as these infoldings have been observed near the cell surface and in other regions of the cell. All the transitions between the infoldings and the cytoplasmic vesicles have been found (Fig. 4) . They are similar to the "cell pits" described in liver cells (68) and in the thymus of chickens with myeloblastosis (25). Although we have seen these special vesicles in various regions of the cytoplasm in our material, they are observed in significant amounts near the Golgi zone. Frequently we have noted vesicles of this type with their membrane partially disrupted and their contents lying free in the cytoplasm (Fig. 4) . Another peculiarity seen in the controls as well as in the infected cells is the presence of virus-like particles attached to or in the process of budding from the plasma membrane (Figs. 1 and 3). They are spherical in shape with an average diameter of 100 m#. Their nucleoid, 60 m# in diameter, is electron-transparent, and it is limited by an electron-opaque membrane. The space between this internal membrane and the outer membrane is occupied by a dense material, and sometimes in this region an intermediate membrane is identifiable. The outer surface of the particle is covered by "spicules" similar to those referred to by Zeigel (76) in the particles of the chicken pancreas agent. These virus-like particles are seen in almost all the cells observed, one to four particles per cell section, generally budding from the plasma membrane. Occasionally similar particles are also seen inside cytoplasmic vesicles. The early association between the hepatitis virus and the ML cells was observed in the samples taken within 1 hour after the culture inoculation. In these cells the virus is present in the outer surface of the plasma membrane sometimes partly enclosed by a cell process or inside cytoplasmic vesicles near the cell surface. In the samples taken 2 to 3 hours after the culture inoculation virus particles are found in increasing numbers inside dense cytoplasmic corpuscles (Fig. 5) . These dense bodies are spherical and range in size from 400 to 700 m~. They are bounded by a membrane in which, when cut normally to its surface, the components of the unit membrane are recognizable. In the preparations fixed with potassium permanganate the matrix of the dense bodies is less dense and their content of virus particles more evident (Fig. 14) . The number of virus particles per dense body varies greatly; some have just one or two particles, but others are completely packed with them. ~Ihe appearance of the particles inside these corpuscles is also variable (Fig. 5) . In some only intact particles are seen, but in others the particles are partially disintegrated (Fig. 5 ). Dense bodies containing structures similar to virus membranes are also observed. During the first 5 hours after infection the changes observed in the ML cells, besides those already described, are slight increases in the number and size of the pseudopodia, in the number of intracytoplasmic vesicles and vacuoles, and in the number of dense bodies with virus. 3-he nucleoli of the infected cells are frequently hyper-trophied and present dense spots on the nucleolonema (Fig. 6) . In the cells from the samples taken 7 hours and more after the inoculation two organized structures never noticed in the controls or, to our knowledge, in any type of normal cell are observed in the cytoplasm (Fig. 7) . These structures, almost always associated, are of two type: one, that we have called the reticular inclusion, is composed of threads of a dense filamentous material disposed in a reticular pattern and was the first to be noticed; the other, named tubular body, consists of membrane-limited tubules arranged in a very complex pattern. The reticular inclusion (Fig. 8) , although it varies in size and shape from cell to cell, is usually round or oval and is between 1 to 2 # in diameter. Its threads are 250 to 400 A in width and are composed of a dense matrix wherein are dispersed dense granules approximately 35 A in diameter (Fig. 9) . We have observed one to three of these structures per cell section. Always present in the vicinity of the reticular inclusion and generally disposed around it are several vacuoles with a diameter of about 200 m~ (Fig. 8) . They have a dense limiting membrane 50 A thick in which at some points the three layers of the unit membrane are recognized. These vacuoles are generally empty but sometimes there is seen in their interior a coiled filament 30 A in width (Fig. 8 a) . The tubular body (Figs. 10 and 11), which is composed of tubules with a medium diameter of approximately 160 to 250 A, has a round or oval outline in the thin sections and is between 1 to 1.7 in diameter. The form of the whole inclusion is spherical. The elements ot the tubular body are frequently in continuity with a system of membrane-limited tubules and cisternae that develop in the cytoplasm of the infected cells (Fig. 10) . With the progression of the infection the system of tubules develops and occupies progressively larger zones of the cytoplasm. In the insert one of the vesicles associated with the reticular inclusion is shown at higher magnification. X 64,000. 64 THE JOURNAL OF CELL BIOLOGY -VOLUME ~4, 1965 The tubules or cisternae are 310 to 370 A wide (Figs. 10 and 12) , are limited by smooth membranes, and their lumens are occupied by a dense substance. Free in the cytoplasmic matrix separating the cisternae are numerous dense granules 150 to 200 A in size interpreted as ribosomes. Virus particles are seen inside the cisternae or in contain numerous virus particles are observed in the cytoplasm. In these stages the reticular inclusion and the tubular body are seen less frequently. Also the Golgi apparatus, which is so evident in the uninfected ceils, is not apparent. ~[he possibility exists that it has been transformed during the virus infection. the process of budding from the limiting membranes into the cisternae (Fig. 12) . The virus particles have a circular outline and a median diameter of 75 m#. Their nucleoid, 55 m# in diameter, is formed by dense granules and rods composing a ring which limits a central electrontransparent space. The limiting membrane of the virus, 30 A thick, is separated from the nucleoid by a clear space 80 A in width. The outer surface of the virus is covered by a layer of moderately dense material. In the cells taken from the samples 14 to 20 hours after the virus inoculation, large inclusions consisting of cisternae (Fig. 13 ) whose lumens In preparations fixed with potassium permanganate the reticular inclusion, tubular body, and the system of tubules and cisternae are clearly seen and their relations are evident (Figs. 14 and 15). Inside the cisternae the virus particles are observed but their nucleoid is not seen (Fig. 15 ). In these preparations we sometimes noted between the two components of the nuclear membrane round formations which might correspond to virus particles. ~[his point is not yet clear and needs further investigation. In the samples taken 20 hours or more after the inoculation numerous cells present evident signs of degeneration (Fig. 16 ). Their nuclei have very formed by closely interwoven, membrane-limited tubules (tubular body). Note its relation with the double-membraned system developed in the cytoplasm of these cells after inoculation of the hepatitis virus. )< 60,000. In its periphery are seen severa] incompletely formed viruses (arrows). X 60,000. dense marginated chromatin, and their cytoplasm shows a complete lack of organization. In the cytoplasm of the degenerated elements (Fig. 17) accumulations of dense granules 150 A in diameter (ribosomes?) and inclusions having the appearance of lipids are observed. Virus particles are seen inside vacuoles or cisternae or lying free. Around these dying cells are frequently seen masses of cytoplasm originating from their disruption (Fig. 16) . They contain abundant dense granules. Before discussing the data concerning the infected cells, we need to comment on three points from the observations made on the controls: (a) the presence of virus-like particles in these cells; (b) the localized plasma membrane specializations, their infoldings, and related cytoplasmic vesicles; (c) the bundles of 70 to 75 A fibrils existing in the cytoplasm of some of the cells. (a) The virus particles observed attached to or budding from the plasma membrane of the ML cells are, by their morphology and process of formation, similar to the "immature" C particles described in the murine leukemias (16) . Not long ago, Dales and Howatson (13) identified morphologically similar particles in their cultures of Earle's strain L cells. Specific biological activity could not be associated with these entities. In our experiments the particles observed apparendy did not interfere with multiplication of the mouse hepatitis virus in the ML cells. However, their presence in this cell line provides an example of the caution which must be exercised in drawing conclusions from observations of viruslike particles in cultured cells employed in the search for new viruses. As Bernhard (3) points out, "the morphologist is ahead of the biologist and has many unlabeled virus particles to sell." (b) The localized differentiations observed in the plasma membrane of the ML cells as well as their related invaginations and special cytoplasmic vesicles are morphologically identical with the formation reported by Roth and Porter (68) in hepatic cells and by de-~fh6 et al. (25) in blastlike elements in the thymus of chickens with myeloblastosis. We have recently observed similar structures in several other types of cells. These observations all suggest that such formations have a general significance in cell physiology. A dynamic interpretation provides two possibilities: they may represent different steps of a process of incorporation into the cell, or they may be related to a phenomenon of cellular excretion. Roth and Porter (68) , on the basis of their observations on the mosquito oocyte, concluded that the membrane differentiations and their infoldings are special sites for protein uptake. However, additional investigation is required to permit precise conclusions concerning the functional significance of these structures. For this purpose tissue cultured cells provide a good study system. A point of interest that emerges from the present observations is the localization of these special vesicles in the Golgi zone. Although the significance of this observation can at the moment be only a matter of speculation, a possible relationship is indicated between these vesicles and the function of the Golgi apparatus. (c) The fibrils 75 A in diameter observed in bundles in the cytoplasm of some of the ML cells (controls and infected) are identical to the fibrils described in normal or virus-infected cells by several authors (14, 24, 54) . Dales and Siminovitch (14) have suggested that similar fibrils which they described in L cells could be a response to injurious stimuli. Another plausible hypothesis about their significance is that of De Petris et al. (24) who suggest that they are contractile in nature and probably related to cell movements. The data reported herein show that the hepatitis virus enters the cell by phagocytosis. 3-his mechanism of cell penetration by virus was first suggested by Fazekas (32) demonstrated that it is by viropexis that the vaccinia virus (10, 14) , adenovirus (9), and influenza virus (11) are incorporated into cells infected in vitto. The same mechanism has been reported by Heine et al. (42) in the virus of avian erythroblastosis, by Zamboni and Biberfeld (75) in the polyoma virus, and by Mussgay and Weibel (57) in the Newcastle disease virus. Tanaka et al. (73) who studied a mouse hepatitis virus in vitto also concluded that it is by engulfment that this virus is incorporated by the cells. The data available at present suggest that viropexis is a general mechanism of cell penetration by a virus and probably one of the first steps in cell infection. However, from the morphological evidence, we cannot eliminate the possibility that, simultaneously with the entry of intact particles, the cell is also penetrated by disassembled "in fectious material" more directly related to the initiation of the cellular infection. According to the observations presented, after the mouse hepatiti.s virus enters the cell by viropexis, a process which generally does not involve more than two particles, it is observed inside dense cytoplasmic corpuscles. The number of virus particles in each corpuscle is variable, but most corpuscles contain large numbers. This indicates that after incorporation the virus is concentrated inside these bodies. The localization of virus particles in dense cytoplasmic corpuscles is an observation reported after cellular infection with different types of virus. Bonar et al. (5) and de-Th6 et al. (26) described in avian myeloblasts, cultivated in vitro, the presence of virus in dense corpuscles. Haguenau et al. (40) noted dense bodies with virus in Rous sarcoma cells cultivated in vitro. Dales (9) reported the presence of adenovirus inside dense cytoplasmic corpuscles in HeLa cells infected in vitro. The process of accumulation into cytoplasmic bodies is not a special reaction of the cell to the entry of a virus, and has been repeatedly reported after cellular phagocytosis of different types of extraneous material (2, 31, 41, 60, 61, 64) . Evidence has been accumulated that the dense cytoplasmic bodies, generally described as being associated with cell phagocytosis, are identical to Straus's phagosomes (70, 71) or de Duve's lysosomes (20, 21, 59) . The concentration of virus in the dense cytoplasmic corpuscles raises the question of how they enter and are accumulated in these corpuscles. One can speculate that vesicles, containing virus, are incorporated in preformed dense corpuscles by fusion of their limiting membranes, but another possibility is that several phagocytotic vesicles containing particles fuse together and a dense corpuscle is subsequently formed by progressive concentration of the dense matrix in the vesicle. A similar mechanism has been postulated by Farquhar and Palade (31) for the dense bodies containing ferritin observed in the visceral epithelium of rats previously inoculated with a solution of ferritin. The most interesting observations on hepatitis virus-infected ML cells were made 7 hours after their exposure to the agent. At this time, two formations, the reticular inclusion and the tubular body, were noticed. The reticular inclusion, the first of the two to appear, resembles in some characteristics the cytoplasmic inclusions observed in other virus infections. For example, inclusions of a filamentous substance were reported to develop in L cells after their infection with mengovirus (Dales and Franklin, 12) or with vaccinia virus (Dales, 10) . It is in these areas that the immature vaccinia virus particles were later observed. Dales calls the dense material viroplasm and interprets these inclusions as "factories" where the viral nucleoprotein is synthesized. Dense inclusions of the same type were described long ago with different names (matrix area, ground substance, viroplasm) in cells infected with viruses, especially those from the pox group (4, 27, 33, 35, 47, 53, 62, 67, 74) . In all these observations it is evident that the final virus formation is related to the dense cytoplasmic inclusions. In our experiments the relation between the inclusions observed and the virus is difficult to establish because the virus formation is never observed in these areas. Recently Kuhn and Harford (46) found cytoplasmic inclusions in HeLa cells infected with parainfluenza virus and observed that these inclusions consist of dense areas of filaments and granules which they claim to be viral ribonucleoprotein. They never observed virus particles in these areas and they suggest that the absence of complete virus can be explained by the hypothesis that the "viral capsid is only acquired when the viral nucleoprotein passes through the cell membrane." In our material we can also hypothesize that the reticular inclusion is the site where the viral nucleoprotein replicates, although the final formation of the virus takes place in the tubules and cisternae by budding from their membranes. Another possibility is that the reticular inclusion is a precursor material from which the tubular body originates. The significance of the vesicles observed in close relationship with the reticular inclusion is not clear. Similar vesicles have been observed in the vicinity of the cytoplasmic inclusions in vaccinia virus-infected cells (10) as well as near the dense inclusions of parainfluenza virus-infected cells (46) . Dales interpreted the vesicles noted in the vaccinia virus-infected cells as virus "shells" remaining in these areas after the release of the DNA core. In our material this interpretation, although not excluded, is unlikely because of the great differences in size between the vesicles and the hepatitis virus. The tubular body, which is usually found in the vicinity of the reticular inclusion, is a peculiar structure which, to our knowledge, has not been previously detected in normal or infected cells. Its origin is obscure, but it is evident that it is always connected with the system of tubules and cisternae differentiated in the infected cells. It can be interpreted as the center from which the development of this membranous system is initiated. As a tentative interpretation of the significance of the reticular inclusion, the tubular body, and their relationship to each other, we can speculate that the reticular inclusion represents the fabric where the synthesis of the viral nucleoprotein takes place and that the tubular body is the center from which is developed the membrane-limiting tubules and cisternae where the final formation of the virus is realized. When we compare the ML cells from the controls with cells sampled 14 to 20 hours after virus inoculation, the most impressive element is the existence of the developed system of membrane-limited tubules and cisternae in the cyto-plasm of the infected cells. The appearance of membranous structures, vesicular or tubular, in the cytoplasm of virus-infected cells is a phenomenon that has been reported in the literature by several authors (4, 12, 39, 54) . Dales and Siminovitch (14) interpreted these formations as "an example of cellular response in a variety of animal cells which is elicited not only by virus, but also by toxic agents," and compared them with the membranous structures appearing after different experimental conditions. The membrane-limited tubules and cisternae described in our experiments are much more regular formations and have been interpreted from their morphology as belonging to the agranular reticulum (7, 63) . Unfortunately, the physiological significance of the agranular reticulum, which has been lately discussed in several papers (8, 29, 65) , is not yet clear. In our experiments the tubules and ~isternae which contain a dense substance elaborated or accumulated therein are intimately :rel~ed with the virus formation. As the virus particles bud from the cisternae limiting membranes and are released into the lumens, their outer membranes become covered by the dense substance present in the cisternae. The formation of virus or virus-like particles in vesicles or cisternae of the endoplasmic reticulum is a process reported several times in the last few years. Friedlaender and Moore (34) , and Adams and Prince (1) observed such formation in the endoplasmic reticnlum of the Ehrlich ascites tumor cells. Dalton and Fdix (17) described virusparticles inside intracytoplasmic vesicles in S 37 tumor cells. Type A particles were reported inside or budding from the limiting membranes of the endoplasmic reticulum in plasma cell tumors (19) and in a fibroblastic sarcoma (38) . In murine leukemias it was established that the virus forma- Fixation with potassium permanganate. The relationship between the tubular body (Tb), reticular inclusion (Ri), and associated vacuoles (V) is clearly seen. A dense body conraining virus particles is also evident (L). X 35,000. FmUUE 15. Portion of the cytoplasm of an ML oell infected with mouse hepatitis virus. Fixation with potassium permanganate. In the upper part of the micrograph a tubular body (Tb) and its relation to the system of the cytoplasmic tubules are clearly seen. Inside the tubules small vesicles corresponding to virus particles are indicated by an arrow. X 3~,000. THE JOURNAL OF CELL BIOLOGY-VOLUME ~4, 1965 FIOURE 16. An infected ML cell presenting evident signs of degeneration. In the cytoplasmic fragments are observed accumulations of dense granules, and inclusions with the appearance of lipids. ;< 20,000. tion takes place in the limiting membranes of the intracytoplasmic channels of the megakaryocytes (18, 22, 23) . In all these cases it was not established that there was any difference in the organization and a m o u n t of the endoplasmic reticulum (granular or agranular) before and after the presence of the virus. Only in the murine leukemias is it known that the system of intracytoplasmic channels from which the virus buds is a normal component of the megakaryocytes. More recently Granboulan and Wicker (37) , reporting the formation of a simian virus in hamster fibroblasts cultivated in vitro, inferred that the appearance of the virus nucleoid "est accompagnfie par la formation de systemes m e m b r a n e u x trds developp~s, plus au moins associ~s avec les grains R N P cytoplasmiques." In our experiments it is evident that the tubules and cisternae of the endoplasmic reticulum are developed after the entry of the virus. They grow with time following infection and are intimately associated with the virus formation. In the light of this observation it seems appropriate to conclude that the hepatitis virus, once integrated into the cell, is able to originate the differentiation of a cellular system never observed in the controls. Speculating about the differentiation of this system, we can surmise either that it is a result of information tramferred to the cell by the virus material or that it is the development of a cell potentially which is activated after the virus infection. An electron microscopic study of the morphology and distribution of the intracytoplasmic "virus-like" particles of Ehrlich ascites tumor cells A suggestion as to the nature of the lysosome granules The detection and study of tumor virus with the electron microscope Les formes intracytoplasmiques du virus fibromateux de Shope. Etudes de coupes ultrafines au microscope electronique Ultrastructure of avian myeloblasts in tissue culture Viral hepatitis associated with transplantable mouse leukemia. I. Acute hepatic manifestations following treatment with urethane or methylformamide Studies on the fine structure of the mammalian testis Differentiation of the spermatids in the cat (Fells domestica) The normal fine structure of opossum testicular interstitial cells An electron microscope study of the early association between two mammalian viruses and their hosts The uptake and development of vaccinia virus in strain L cells followed with labeled viral deoxyribonucleic acid, or Attachment and penetration of influenza virus A comparison of the changes in the fine structure of L cells during single cycles of viral multiplication, following their infection with the viruses of mengo and encephalomyocarditis Virus-like particles in association with L strain cells The development of vaccinia virus in Earle's strain cells as examined by electron microscopy A chrome-osmium fixative for electron microscopy Micromorphology of murine tumor viruses and of affected cells The electron microscopy of normal and malignant cells An electron microscopic study of a series of murine lymphoid neoplasms Some ultrastructural characteristics of a series of primary and transplantable plasma cell tumors of mice Intracellular distribution patterns of enzymes of rat liver tissue Lysosomes, a new group of cytoplasmic particles Electron microscope study of a cell free induced leukemia of the mouse: A preliminary report Further electron microscope studies of a mouse leukemia induced by cell-free filtrates Filamentous structures in the cytoplasm of normal mononuclear phagocytes Ultrastructural characters of the thymus in myeloblastosis and of the adenosinetriphosphatase activity of thymic cells and associated virus Ultracytochemical study of virus and myeloblast phosphatase activity, or A study with the electron microscope of the skin tumour of molluscum contagiosum Amino acid metabolism in mammalian cell cultures Observations on the fine structure of lutein cells The growth in vitro of massive culture of liver cells Segregation of ferritin in glomerular protein absorption droplets Regeneration of virus receptors in mouse lungs after artificial destruction Observation, pendant la phase muette du developpement intracellulaire du virus du fibrome de Shope, de corps d'inclusion diffus, sans virus corpusculaires, correspondant avec la pr6sence d'un antigone soluble Occurrence of bodies within endoplasmic reticulum of Ehrlich ascites tumor cells Intracellular forms of pox viruses as shown by the electron microscope (vaccinia, ectromelia, molluscum contagiosum), a r A hepatitis virus of mice Etude ultrastructurale du d6veloppement d'un virus simien latent Pr6sence de particules d'aspect viral dans un sarcome greffable de la souris provoqu6 par le m6thylcholanthr~ne Reduplication of nuclear membranes in HeLa cells infected with adenovirus Ultrastructure du virus du sarcome de Rous in vitro Electron microscopy of HeLa cells after the ingestion of colloidal gold Virus of avian erythroblastosis. VII. Ultrastructure of erythroblasts from the chicken and from tissue culture Establishment of a clone of mouse liver cells from a single isolated cell The effect of a murine hepatitis virus on the liver. An anatomic and histochemical study Problems in the production of leukemia with cell-free extracts Electron microscopic examination of cytoplasmic inclusion bodies in cells infected with parainfluenza virus Electron microscopy study of mouse liver infected by Ectromelia virus Permanganate, a new fixative for electron microscopy Improvements in epoxy resin embedding methods A hepatitis virus complicating studies with mouse leukemias Viral hepatitis in mice: An electron microscopy study Permanganate fixation of plant cells Structure and development of viruses observed in the electron microscope. II. Vaceinia and fowl pox viruses Electron microscopic observations on the development of herpes simplex virus, o r A new member of hepatoencephalitis group of murine viruses A new member of hepatoencephalitis group of murine viruses Early stages of infection with Newcastle disease virus as revealed by electron microscopy Acute hepatitis associated with mouse leukemia. 1. Pathological features and transmission of the disease Lysosomes and the physiology and pathology of cells Electron microscopy of lysosome-rich fractious from rat liver Uptake and transfer of particulate matter from the peritoneal cavity of the rat Studies on the growth of viruses ectromelia and vaccinia in strain L cells and I-IeLa cells Studies on the endoplasmic reticulum. II The endoplasmic reticulum Studies on the endoplasmic reticulum. V. Its form and differentiation in pigment epithelial cells of the frog retina The use of lead citrate at high pH as an electron-opaque stain in electron microscopy, or Structure and development of viruses as observed in the electron microscope. IV. Echo virus type 9 Specialized sites on the cell surface for protein uptake Electron and fluorescence microscopy of mouse hepatitis virus (MHVI) Colorimetric analysis with N, ~-Dimethyl-p-phenylenediamine of the uptake of intravenously injected horseradish peroxidase by various tissues of the rat Rapid cytochemical identification of phagosomes in various tissues of the rat and their differentiation from mitochondria by the peroxidase method An electron microscope study of viral hepatitis in mice Electron microscopic study on the cultured liver cells infected with the mouse hepatitis virus. A preliminary report Le d~veloppe-merit intracellulaire du reovirus observ6 au microscope 61ectronique The early stages of polyoma virus infection in mice subcutaneous tissue, 5th Internat Morphological evidence for the association of virus particles with the pancreatic acinar cells of the chick Part of this work was supported by a grant from the Calouste Gulbenkian Foundation (Lisbon, Portugal). The authors wish to thank Dr. A. J. Dalton for his advice and encouragement throughout the course of this work. Thanks are also due to Mrs. Karin David-Ferreira and Miss Virginia Kearns-Preston for their valuable technical assistance.