J. gen. Virol. 0973), 2i, 533-537
533
Printed in Great Britain
The Mode
of Entry of Vaccinia Virus into L Cells
(Accepted I6 August I973)
SUMMARY
Electron microscopy of L cells infected with vaccinia virus at a high multiplicity
indicated that virus cores entered the cytoplasm, during a 15 min period of adsorption at 37 °C, as a result of fusion between the plasma membrane and the envelopes
of virus particles. It is suggested that for vertebrate poxviruses this, rather than
phagocytosis, could be the effective mode of entry leading to infection.
Recent studies in this laboratory showed that when L cells are infected with vaccinia
virus, RNA and protein synthesis commence with exceptional rapidity, virus-specific
polypeptides being detectable only 2o rain after infection (Metz & Esteban, I972; Esteban
& Metz, 1973). It is generally accepted that when poxviruses infect animal cells phagocytosis
of the particles (' viropexis ') is an obligatory initial step (Dales, 1965; McAuslan, I969).
We have now re-examined by electron microscopy features of the[ adsorption and penetration of vaccinia virus into cultured cells, and our findings indicate the occurrence of
direct fusion between the envelope of attached virus particles and the plasma membrane
of the cell.
L cells at a concentration of I.o × Io7/ml were infected in suspension with vaccinia virus
(strain WR) at a multiplicity of about 5oo particles/cell, for I5 rain at 37 °C in Puck's salineA
with 2o mM-Mg~+ and 2 % and foetal calf serum. Immediately following the adsorption
period, designated zero-time, the cells were diluted to I.O × 1o°/ml in tris-buffered Eagle's
medium (pH 7"2) containing 5 % calf serum. Cultures were shaken at 37 °C, and samples of
2 to 3 ml withdrawn at intervals for fixation. Samples for electron microscopy were taken at
zero-time and at hourly intervals up to 4 h thereafter, together with uninfected controls at
each point. The cells were sedimented, washed briefly by re-suspension in ice-cold Ringer's
solution and pelleted (< 5oo g for I min); they were fixed by re-suspension for 3o rain in
a freshly made ice-cold mixture of glutaraldehyde and osmium tetroxide, washed, and
post-fixed with uranyl acetate according to the schedule of Hirsch & Fedorko (1968). Eponembedded cell pellets were sectioned, mainly as interrupted serial runs to obtain good
sampling, and examined with the electron microscope after staining with uranyl acetate and
lead citrate. In surveying the cells attention was paid particularly to the relationships and
fine structure of vaccinia virus observed (a) as mature extracellular virus particles; (b) as
naked cores within the cell; (c) as virus particles undergoing ingestion, or contained in
cytoplasmic phagosomes.
(a) In all samples fixed from zero-time up to 4 h after infection numerous poxivrus
particles were in close proximity to the external surface of the cells. They exhibited the
usual range of profiles, depending upon plane of section; presenting an internal core,
paired lateral bodies and an external membrane or envelope. The virus envelope, due to the
high contrast obtained with simultaneous glutaraldehyde and osmium fixation, showed
conspicuous triple-layering of the unit membrane with an irregular coating of electrondense material on its outer face. Occasional particles seemed to lie in more intimate contact
with the plasma membrane; only the coating of the virus envelope separating the two
membranes (Fig. I).
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 01:40:40
534
Short communications
l ~
nm
~
(b) Typical poxvirus cores were identifiable in the cytoplasm of all but a few of the cells
examined, at each time point after infection including those taken at zero-time. Evidently
large numbers of virus particles had effectively transferred their cores into the cytoplasm
during the I5 rain period of adsorption. Cores were round, oval or rectangular in outline,
and measured up to 3oo nm in the long axis (Figs. 5, 6). Each possessed a well defined dense
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 01:40:40
Short communications
535
wall with a diffuse zone of radiating projections around it, and the centre was filled with
filamentous material presumed to be virus nucleoprotein. Core profiles in the sections gave
no indication of any regular association with cytoplasmic membranes, vesicles or other
organelles, but in the cells fixed at zero-time their distribution was predominantly marginal,
i.e. most were located close to the plasma membrane or within large pseudopodium-like
extensions of the cell periphery. At the later stages cores were more widely distributed, and
some were found close to the nuclear envelope.
Systematic search was made through a considerable number of apparent virus-cell
attachment sites, to ascertain possible means by which many cores had gained rapid entry
to the cell cytoplasm. In infected cells fixed at zero-time profiles were found indicating clearly
the occurrence of a direct virus to cell fusion phenomenon in which continuity was established, at a point of contact, between the virus envelope and the plasma membrane (Figs. 2,
3, 4). This, in effect, brought the interior of such particles into contact with the peripheral
cell cytoplasm. To avoid the problem of confusing mere superimposition of membranes for
genuine fusion, images that showed only blurring at the point of contact were regarded as
inconclusive; only those in which continuity of the triple layering of the two membranes
could be clearly traced, in the plane of the section, were rated as indicating fusion. An example is seen in Fig. 3, depicting the site of fusion between a vaccinia particle and a projection of the cell margin: the virus core has expanded already into an oval form typical of
naked cores, but the paired lateral bodies are still evident on either side of the core. The
impression was gained that lateral bodies dispersed soon after fusion; traces of only one
remained in the fused particle shown in Fig. 4. Incorporation of the virus envelope into the
cell membrane is suggested by images such as these, but the possibility of its being shed
as an empty vesicle after release of the core is not excluded. In the present survey, images
showing definite fusion were confined to cells fixed immediately after the 15 min adsorption
period. A few less conclusive examples were located in the sample taken r h after infection,
but none were found in the later samples.
(c) In infected cells taken at zero-time no profiles showing vaccinia particles undergoing
phagocytosis were found, nor were poxvirus particles seen within closed phagosomes in the
cell cytoplasm. However, in samples examined from I h to 4 h after infection such profiles
occurred quite often and their frequency appeared to increase somewhat with time. The
intravacuolar vaccinia particle shown in Fig. 7 was evidently ingested along with a C-type
particle from the cell surface: such particles are of course a familiar feature of murine cell
lines, and in the present study they occurred equally in both the control and vacciniainfected cultures.
That animal viruses may release their nucleoprotein into the cytoplasm of susceptible
Figs. 1 to 6. Sections from L cells fixed at zero-time, i.e. immediately after virus adsorption.
Fig. I. A typical intact vaccinia particle in contact with the cell surface.
Fig. 2. Virus attachment site with an indication of localized loss of integrity in closely apposed
portions of the plasma membrane alad a vaccinia envelope; interpreted as commencing membrane
fusion.
Figs. 3, 4. Sections through vaccinia particles fused with the cell membrane; showing continuity of
their envelopes with the plasma membrane, expanded core morphology and (in Fig. 3) paired
lateral bodies.
Figs. 5, 6. Free vaccinia cores in the cell cytoplasm, showing the characteristic oval and rectangular
profiles with external projections.
Fig. 7. An intraphagosomal vaccinia particle, in an L cell fixed following I h of incubation after
virus adsorption.
35
VIR
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 01:40:40
2I
536
Short communications
host cells by a mechanism involving membrane fusion was first proposed for the myxoviruses
(Hoyle, 1962); and this has been amply substantiated by electron microscopy (Meiselman,
Kohn & Danon, 1967; Morgan & Howe, I968; Morgan & Rose, I968; Dourmashkin &
Tyrrell, I97o ). Penetration of cells by direct membrane fusion has now been attributed to
a variety of other enveloped viruses some of which, like myxoviruses, also cause syncytium
formation: they include herpes simplex virus (Morgan, Rose & Mednis, I968), vesicular
stomatitis virus (Heine & Schnaitman, I97I), simian foamy virus (Dermott & Samuets,
1973) and, most recently, an insect pox-like virus (Granados, 1973). Thus, the finding that
vaccinia virus can enter L cells by membrane fusion as well as by phagocytosis brings the
vertebrate poxviruses into line with other major groups in which both modes of entry have
been reported. It does not follow that membrane fusion is necessarily the normal form of
entry leading to infection, but in our view it is quite likely to be so. It may be significant that
at the earliest time of observation, with numerous cores already detectable in the cytoplasm,
there was morphological evidence for fusion but none for phagocytosis, while at later times
the reverse was the case. Phagocytosis is a somewhat non-specific phenomenon, and it
would be surprising not to find a proportion of virus particles from the inoculum being
engulfed and digested along with other particulate matter from the cell surface; whereas it
may be supposed that fusion between a virus envelope and the plasma membrane would
hardly take place other than as a specific event of functional significance. In addition,
penetration by membrane fusion avoids entirely the problem of transferring uncoated virus
cores, intact, from the hostile environment of a phagolysosome into the surrounding cytoplasm. The case for 'viropexis' as an obligatory step in poxvirus infections has rested mainly
on the demonstrability of phagocytic profiles using electron microscopy, and the reversible
arrest of penetration by pre-treatment of cells with sodium fluoride (Dales & Kajioka, 1964).
This does not exclude the direct virus-cell fusion mechanism; and in view of the present find
ings it appears likely that the phenomenon was previously overlooked. A conceivable compromise would be that sometimes, after phagocytosis, poxvirus cores may be transferred
to the cytoplasm prior to lysosomal fusion with the phagosome, or even that the lysosomal
response might be inhibited as occurs following ingestion of certain bacterial pathogens
(Armstrong & Hart, I97I); but there is inadequate evidence to sustain such hypotheses at
present.
National Institute for Medical Research
Mill Hill
London, N.W.7, U.K.
J.A. ARMSTRONG
D.H. METZ
M.R. YOUNG
REFERENCES
ARMSTRONG, J. A. & HART, P. D. 0 9 7 t ) . R e s p o n s e of cultured m a c r o p h a g e s to Mycobacterium tuberculosis,
with observations o n fusion o f lysosomes with p h a g o s o m e s . Journal of Experimental Medicine I34 ,
713-74o.
DALES, S. (1965). Penetration of a n i m a l viruses into cells. Progress in Medical Virology 7, 1-43.
DALES, S. & KAJIOKA, R. (I964). T h e cycle o f multiplication of vaccinia virus in Earle's strain L ceils.
I. U p t a k e a n d penetration. Virology 24, 278-294.
DOURMASHKIN, R. R. & TYRRELL, D. A. J. (I970). A t t a c h m e m o f two myxovirvses to ciliated epithelial cells.
Journal of General Virology 9, 77-88.
DERMOTT, E. & SAMUELS,J. (I973). Electron microscopic observations o n the m e c h a n i s m s o f entry of Simian
F o a m y Virus in Hep-2 cells. Journal of General Virology x9, I35-I39.
ESTtBAN, M. & METZ, D. H. (I973). Early virus protein synthesis in vaccinia virus-infected cells. Journal of
General Virology x9, 2oi-216.
GRANADOS, R. R. (I973). E n t r y o f a n insect poxvirus by fusion o f the virus envelope with the h o s t cell m e m brane. Virology 52, 3o5-3o9.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 01:40:40
Short communications
537
HEINE, J. W. & SCHNAITMAN,C. A. (197I). Entry of vesicular stomatitis virus into L cells. Journal of Virology
8, 786-795.
HIRSCH, J. G. & FEDORKO,M. E. (1968). Ultrastructure of human leukocytes after simultaneous fixation with
glutaraldehyde and osmium tetroxide, and post fixation in uranyl acetate. Journal of Cell Biology 38,
615-627.
HOYLE, L. (1962). The entry of myxoviruses into the cell. CoM Spring Harbor Symposia on Quantitative
Biology 27, I13-12I.
McAUSLAN,B. a. (1969). Biochemistry of poxvirus replication. In The Biochemistry o f Viruses, pp. 361. Edited
by H. B. Levy. New York and London: Dekker.
MEISELMAN,Y., KOHN, A. & DANON, D. (1967)- Electron microscopic study of penetration of Newc~stle Disease
Virus into cells, leadng to formation of polykaryocytes. Journal of Cell Science 2, 71-76.
METZ, D.H. & ESTEBAN, M. (I972). Interferon inhibits vial protein synthesis in vaccinia-infected L cells.
Nature, London 238 , 385-388.
MORGAN, C. & HOWE, C. (1968). Structure and development of viruses as observed in the electron microscope
IX. Entry of para-influenza i (Sendai) virus. Journal of Virology 2, I I22-I 132.
MORGAN, C. & ROSE,H. M. (I968). Structure and development of viruses as observed in the electron microscope
VIII. Entry of influenza virus. Journal of Virology 2, 925-936.
MORGAN, C., ROSE,H. M. & MEDNIS,B. (1968). Electron microscopy of herpes simplex virus I. Entry. Journal of
Virology 2, 5o7-516.
(Received 19 July I973)
55-2
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 01:40:40