вход по аккаунту


Methods of Virus Formation.

код для вставкиСкачать
the usual methods of selection are not suitable for the
isolation of such mutants, or that they are not able to
grow under the usual cultivation conditions. The identification of substances which would inhibit early stages
of lipopolysaccharide biosynthesis in the cell (as d o
some antibiotics in the synthesis of murein) would be
of great theoretical and practical value. Such substances would convert the pathogenic wild types into
non-pathogenic R forms which are known to be readily
eliminated in the organism by macrophages.
Although the biosynthetic pathway of lipopolysaccharides has been extensively elucidated, very little is
known about their degradation in nature. It is surprising that n o enzymes have so far been described which
degrade lipopolysaccharides specifically, since this
would be of extreme value for structural analyses or
for the identification of biologically active regions in
the molecule. We have recently found that amoebae
of Dictyostelium discoideum contain an enzyme system which degrades specifically the lipid A component
of lipopolysaccharides [551. After phagocytosis, lysis,
and digestion of the fed bacteria, these amoebae egest
degraded lipopolysaccharide into the culture medium.
This lipopolysaccharide contains intact 0-specific
chains and core polysaccharide. The lipid A component, however, is quantitatively deacylated. Ester- as
well as amide-bound fatty acids are cleaved off and
the degraded lipopolysaccharide contains nonacylated
glucosamine disaccharide. Amoebae-degraded lipopolysaccharide is non-toxic. It is possible that the inactivation (detoxification) of lipopolysaccharides,
which is observed after incubation with serum, is based
on an analogous reaction with acylases.
Immunochemical and serological studies on lipopolysaccharides have shown that 0 specificity is determined
by defined structures in the 0-specific chains; determinant groups, di- to hexasaccharides, determine the
individual factors of the Kauffmann-White scheme. In
each determinant group one sugar plays the role of
the immunodominant sugar. In R mutants, which lack
specific chains, the terminal sugars of the various incomplete core polysaccharides function as the immunodominant sugars. R mutants exhibit R specificities.
Very recent studies by Rank1571 indicate that under
certain conditions antibodies against lipid A are detectable which, as expected, can cross-react with nonrelated lipopolysaccharides.
All lipopolysaccharides are highly active endotoxins,
irrespective of whether they are derived from S or R
forms. The region responsible for this activity is located in the (KD0)-lipid A part of the molecule. The
polysaccharide part is non-essential.
Received: February 4, 1970
[A 764 IE]
German version: Angew. Chem. 82, 708 (1970)
Methods of Virus Formation
By Gerhard Sauer[*I
Viruses occur in a great variety of shapes and sizes, but for all their diversity in appearance
they possess certain characteristics in common: all viruses contain a single nucleic acid
molecule - deoxyribonucleic acid ( D N A ) or ribonucleic acid ( R N A ) - surrounded by a
protective protein coat. Among other things, the protein coat enables the genetic information stored in the nucleic acid to enter a host cell in a usable state, where it can initiate
the reproduction of identical virus particles. Afrer penetration of the cell the foreign
genetic material of the virus particle first induces the synthesis of macromolecules not
normally present in the cell: the viral nucleic acid undergoes replication and very many
copies are produced, the protein of the virus coat is synthesized, and then these components are assembled to form a new generation of infectious virus particles. Most
viruses also exhibit certain common structural features: their protein coat is built up
from subunits arranged in helical or icosahedral fashion around the genetic material.
1. Conversion of the Virus Genome into a
Functional State
As most viruses consist of nothing but nucleic acid
and a protein coat, they lack the prerequisites to carry
out elementary life processes, i. e. to make constant,
[ * ] Priv.-Doz. Dr. G. Sauer
Institut fur Virusforschung,
Deutsches Krebsforschungszentrum
69 Heidelberg, Thibautstrasse 3 (Germany)
Angew. Chem. internat. Edit.
1 Vol. 9 (1970) No. 9
use of their genetic information. Outside a cell, the
virus particle is only a complex but biologically inert
structure. For their multiplication the viruses require
auxiliary agents that are only available in living cells,
consisting of various enzymes, ribosomes, and transferR N A molecules. The viral nucleic acid only becomes
functional when it penetrates the cell and is freed
from its protein coat.
The cycle of infection, i.e. the period between the
time when a virus particle comes into contact with a
cell and the time when its new infectious progeny are
complete, can be divided into several phases. Adsorption of the virus particles onto the cell is followed by
their penetration of the cell wall and liberation of the
genetic information. When the protective protein coat
is lost from the viral nucleic acid, the virus particle
loses its morphological identity. It is now in the
“eclipse” phase. During this phase occur the transcription and trapslation stages, which lead to the
accumulation of virus components within the cell.
This is followed by maturation of the virus and the
release of a new generation of infectious virus particles.
If cells are infected with a definite amount of virus,
and the infectivity of the nutrient medium surrounding
the cells is determined, the characteristic virus growth
curves are obtained as shown in Figure 1.
Time foilowing infection
Fig. 1. Virus growth curve. The upper part of the figure shows crosssections of vessels containing tissue culture cells covered with viruscontaining nutrient medium. The amount of infectious virus is indicated
by dots. The graph shows the amount of infectious virus during the
course of a virus growth cycle.
Shortly after the addition of the virus suspension to
the cells, the infectivity of the medium gradually
decreases and rises again until finally a plateau is
reached, when the synthesizing ability of the cells is
exhausted. This is frequently associated with death of
the host cells. The initial decrease in the infectivity of
the supernatant is caused by adsorption of the viruses
onto the cells, whereas the subsequent increase is
attributed to the liberation of newly synthesized virus
progeny. We shall first consider the initial part of the
curve, i. e. the phase of adsorption of the virus particles
onto the cell and their penetration of the cell interior.
The first cell-virus contact occurs as a result of molecular brownian motion. The number of irreversibly
adsorbed virus particles per cell is considerably smaller
than that expected from theoretical estimations, since
not every collision between a virus and a cell leads to
a stable contact. The forces binding the virus particles
to the cell membrane are apparently electrostatic in
nature. It is natural to ask whether there are any
specific receptors on the cell membrane responsible for
binding the virus particles, and this conjecture gains
some support when it is considered how exacting
many kinds of viruses are in their choice of host and
even in their choice of the tissue in which they multiply.
One could assume that this selectivity is based on
special processes at the site of contact. Here, however,
the first complications arise: the members of the large
poxvirus group are adsorbed at the same rate by the
very many different cells, even by those in which they
cannot multipJy111. In this case the adsorption is
obviously not based on any specificity. A very different
situation is encountered in the classical example of the
myxovirus group, which includes the influenza virus,
where specific adsorption processes can be observed.
Following its addition to a suspension of erythrocytes,
the influenza virus is adsorbed on the erythrocyte
surface. Binding of the myxoviruses to the periphery
of the erythrocytes gives rise to bridging, as a result
of which the erythrocytes form conglomerations, a
process that is visible macroscopically and has been
termed hemagglutination. By raising the temperature
the virus can be displaced from the erythrocytes. An
interesting point is that erythrocytes agglutinated once
in this way cannot be agglutinated again by the
addition of new virus; this obviously means that their
surface has been changed by contact with the virus.
Fragments possessing hemagglutinating activity called hemagglutinins - can be isolated from the
virus surface 121. In addition, the myxoviruses contain
in their coat an enzyme known as neuraminidase [3,41.
The substrate on which this enzyme acts occurs on the
surface of erythrocytes, as well as in many biological
fluids and on the surface of the natural host cells of
influenza viruses. All these observations indicate that
the contact between myxoviruses and host cells is
brought about as a result of a special enzyme-substrate
relationship. The above remarks show that the processes responsible for the adsorption of viruses onto
cells can vary from one virus group to another 141.
The next step in the cell-virus encounter is penetration.
Certain bacteriophages (bacterial viruses) infect their
hosts by transferring their nucleic acid into the bacterium through an “injection apparatus”. The protein
coat of the bacteriophage does not in this case penetrate the bacterium but is left behind outside the
membrane 151. Such bacteriophages have been studied
very thoroughly and have become a classical example
of how a foreign nucleic acid can be transferred into
a cell.
Electron microscopy has shown that the viruses of
animals behave in quite a different way. They lack
such injection apparatus and, in addition, empty virus
coats have never been observed on cell surfaces.
Everything indicates that animal viruses enter the cells
[l] A. C. Allison and R . C. Valentine, Biochim. biophysica
Acta 40, 393 (1960).
[2] W. Schuyer, Bacteriological Reviews 27, 1 (1963).
[31 G. K . Hirst, J. exp. Medicine 76, 195 (1942).
[41 C. Scholtissek, R . Drreniek, and R . Rott in H . B. Levy:
The Biochemistry of Viruses. Marcel Dekker, New York and
London 1969.
[S] A . D . Hershey and M . Chase, J. gen. Physiol. 36, 39 (1952).
Angew. Chem. internat. Edit. 1 Vol. 9 (1970)
1 NO. 9
in a morphologically intact condition. The cell membrane surrounds the adsorbed virus particle, a vacuole
is formed which moves to the interior of the cell, and
the outer membrane of the cell closes again. In biology
this process is termed phagocytosis. Basically, there is
nothing specific about it, since cells absorb a wide
range of particulate substances, for example metal
salts and latex droplets, in the same way. When the
walls of the vacuole containing the virus particles
break down, these enclosed particles also begin to
Fig. 2. Penetration of viruses into the cell. This electron micrograph
shows the cross-section of a cell infected with herpesviruses. Outside
the cell membrane (left) there is an infectious herpes virus particle
characterized by an outer membrane, a virus coat, and a core. The
identity of the virus particle that has entered the cell (right) is only
revealed by the presence of its core. (This photograph was kindly made
available by Prof. K . Munk and Frau C . Waldrck, magnification:
200 000 x.)
The first stage in the uncoating process is carried out by
the cells. As has been shown by the use of inhibitors
(substances inhibiting protein or R N A synthesis), this
does not require new synthesis of particular proteins
o r even of RNA. Apparently the cells are equipped
with enzymes capable of breaking down the virus
particles to reveal the core, i.e. of removing the outer
virus coat. It is only in the second stage that the viral
DNA is liberated from the core and can then be
broken down by deoxyribonuclease (DNase) into
acid-soluble fragments. This sensitization of the viral
D N A to the action of DNase can be used experimentally to determine the degree and the rate of uncoating161. The principle of such experiments is relatively simple. Purified virus preparations whose DNA
is radioactively labeled for this purpose are used. In
order to analyze the rate and extent of the uncoating
process the cells are disintegrated by ultrasound treatment at various times after infection and the cell
lyzate is treated with DNase. Provided the viral
nucleic acid is within the core it cannot be attacked by
deoxyribonuclease, as it is still protected by its protein
coat. Thus the radioactive components that are
degraded by DNase and broken down into acidsoluble fragments must consist of viral D N A released
from its inner protein coat (Fig. 3).
Figure 2 shows the morphological changes a herpesvirus particle undergoes after penetration into a cell.
The electron micrograph shows a herpesvirus particle
on the outside of the cell and one within the cell. The
intact herpesvirus particle adsorbed onto the cell
surface consists of an outer membrane-like coat
shown in section in the photograph; this coat can be
partly of cellular origin. In the interior of the cell is
the true virus coat, which surrounds the central
structure of the virus particle. The latter consists of a
particular protein and the viral nucleic acid closely
associated with it; in herpesviruses this is DNA. This
inner complex is called the core. Such a core, which
has lost its outer membrane and its coat, is found in
the interior of a cell. This is the first step in the
breakdown of the virus particle leading to liberation
of the viral nucleic acid. The virological term for this
process is “uncoating”. It includes two phases, each
of which has been very carefully investigated using
DNA-containing poxviruses 161. The poxviruses consist
of a double-stranded DNA, the virus coat protein,
and phospholipids. The fate of these three components
in uncoating can be traced precisely in radioactively
labeled virus preparations. The first part of the uncoating process occurs very rapidly and immediately
after adsorption of the virus onto the cell.The phospholipid of the outer virus coat is broken down into acidsoluble degeneration products, which results in particles that are about half the size of the true virus
particles and consist only of D N A and protein. These
are the cores of the virus particles.
When does the viral genome start to function, and
how can this be determined? In discussing this question
we have to distinguish between D N A and RNA
viruses. I n the former the commencement of function
161 W. K . Joklik, J . molecular Biol.
I71 G . Sairer and E. C. Hahn, Z . Krebsforsch. 74, 40 (1970).
8, 263 (1964).
Angew. Chem. internat. Edit. J Vol. 9 (1970)
1 No. 9
core containing DNA
external membrane
Fig. 3. Intracellular liberation of viral DNA from the enclosing protein
coat (uncoating). The figure shows the process in diagrammatic form.
The outer membrane, which can be inissing in many kinds of viruses,
is first removed, followed by the true virus coat. The core is left as a
result of this step. The arrangement of viral DNA within the core is
here represented very schematically. At this stage the D N A is still
protected against DNase. I t is only in the second stage, that of the true
uncoating, that the D N A is freed from the core and exposed to the action
OF DNase.
When the process of uncoating is studied with the
aid of radioactively labeled simian virus (SV 40) 171, it
is found that in the course of a day only about 20 % of
the viruses entering the cells liberate their D N A (the
virus reproduction cycle does not last longer than two
days). This observation shows the inefficiency of virus
infection. It is absolutely untrue that every virus
particle penetrating a cell can initiate an infectious
cycle. Only some of the viruses give rise to infectious
2. The Phase of Information Transfer
is synonymous with synthesis of messenger RNA on
the DNA molecule. The genetic information contained
in the D N A is transcribed onto the messenger RNA,
and it is only after this that the virus-specific proteins
can be synthesized in the cell ribosomes in accordance
with the information contained in the messenger
RNA. Thus, in DNA viruses the transfer of information initially occurs a t the transcription stage (synthesis of messenger R N A on the virus DNA) and
then at the translation stage (production. of virus:
specific proteins in the messenger RNA-ribosome
complex). The RNA-containing viruses behave quite
differently: there is no need for a transcription stage as
their RNA acts directly as messenger RNA. I n this
case the commencement of function of the virus
genome is identical with the production of virusspecific proteins.
The question on the commencement of virus genome
function can be answered most easily by experiments
using D N A viruses, as in this case one only has to
demonstrate the appearance of virus-specific messenger
RNA within the cell. With RNA viruses this can only
be achieved by demonstrating the appearance of
virus-specific protein in the cell, and this presents
certain difficulties. Use of the DNA-RNA hybridization technique can demonstrate the presence of even
a very small amount of virus-specific messenger R N A
in an infected cell. As the sequence of bases in the viral
messenger R N A is an exact mirror image of, and
complementary to, the viral DNA, under suitable
conditions the latter can be bound selectively to the
messenger RNA among all the many other kinds of
R N A present in a cell.
The synthesis of virus-specific messenger RNA requires the presence of an enzyme, a DNA-dependent
RNA-polymerase. This enzyme catalyzes the synthesis
of the messenger RNA molecule. It has recently been
found that many D N A poxviruses carry this enzyme 181. They are thus able to actuate their own transcription. It is not yet known, however, whether this
enzyme occurs in other D N A viruses.
When does the transcription of the viral D N A begin?
For a long time it was believed that this occurred
when the viral D N A was completely freed from its
protein coat by the uncoating process, but here too
some surprising facts emerged from studies of members
of the poxvirus group. It was shown that messenger
R N A is synthesized if just the outer virus coat is
removed, and not the inner protein coat around the
virus D N A [91.
Is all the information in the virus D N A transcribed
at the same time, or does this happen gradually in a
certain order? In the viruses in which this has been
investigated, the transcription is gradual. Thus, for
example, in simian virus 40, only a quarter of the
virus genome is copied as messenger RNA in the
early phase of the infectious cycle. This transcription
[ 8 ] J. R . Kates and B. R . McAusiun, Proc. nat. Acad. Sci. USA
58, 134 (1967).
[ 9 ] J . R . Kates and B. R . McAuslan, Proc. nat. Acad. SCI.USA
57, 314 (1966).
of the “early” messenger RNA is made from t h e viral
D N A after the latter’s penetration of the cell, but
before replication of the DNA [lo]. “Late” messenger
R N A is only transcribed after the initiation of viral
DNA replication (Fig. 4).
virus- DNA
messenger- RNA
Fig. 4. Synthesis of virus-specific messenger RNA on the viral D N A .
The information stored in the virus genome is transcribed in stages. First
o n l y one part of the D N A is translated into “early” messenger RNA.
The remaining sequences of the D N A can only be transcribed at an
advanced stage of the virus infectious cycle when the viral D N A has
replicated. Such “late” messenger R N A contains “early” messenger
RNA and new base sequences.
Obviously replication of the viral DNA is required
before the whole virus genome and not only its early
sequences can be transcribed. If one administers inhibitors of DNA replication and thus prevents multiplication of the viral DNA in the infected cell, only
early messenger RNA is found and the late virus
messenger RNA sequences are missing altogether. Tn
such artificially inhibited cells one can only detect the
presence of the “early” gene functions of the virus.
For most DNA viruses these early gene functions are expressed by the stimulation of a series of enzymes
concerned with D N A metabolism. Thus, in the
affected cells the activity of the enzyme thymidine
kinase is increased severalfold after infection with
most DNA viruses [111. “Late” virus functions require
the presence of “late” virus-specific messenger RNA
in the cells, which, as already mentioned, is dependent
on replication of the viral DNA. Late viral functions
are the synthesis of the virus protein coat and finally,
as the last decisive step in the infectious cycle, assembly
of the virus particle from its components present in the
cell, z.e. virus maturation.
It follows from these remarks that the genetic information stored in the viral genome is transcribed
sequentially and that this transcription is subordinate
to a special control mechanism. It has not yet been
explained what factors control the progressive promulgation of the genetic information, and why the
viral DNA must undergo replication before being
fully transcribed by late virus-specific messenger RNA.
What happens in the promulgation of the genetic
information of RNA viruses? The question is somewhat more difficult to answer experimentally than in
the case of DNA viruses. As regards the enteroviruses, in which this has been investigated, another
set of rules seems to apply. In this case the viral RNA
which also serves here as messenger RNA is translated
not gradually but all at once, in its entirety, into a
single very large protein. This protein is subsequently
[lo] G . Sauer and J . R . Kidwui, Proc. nat. Acad. Sci. USA 61,
1256 (1968).
[ l l ] S. Kit and D. R . Dubbs: Monographs in Virology. Vol. 2.
S. Karger, Basel and New York 1969.
Angew. Chem. internar. Edit. / Vol. 9 (1970)
NO. 9
broken down into several small proteins with different
functions 112,13J. Whether this principle is valid for
all RNA viruses requires further demonstration.
Accumulation of the viral gene products in the infected cell is a function of the virus genome and is not
dependent on whether it is a D N A or an RNA molecule. The amount of these products only increases strikingly after the initiation of the virus D N A o r R N A replication, as shown by Figure 5. As already mentioned,
it is not only the quantity but also the quality of the
virus-specific proteins present in the cell that changes.
No of days after infection
Fig. 5 . Synthesis of specific virus products during a virus growth cycle,
taking simian virus 40 as a n example. In monkey kidney cell culture the
replication cycle takes a little less than two days. The initiation of the
replication of the viral DNA can be divided into two phases. In the
early phase of the infection, even before the start of DNA replication
there is a n increase in the activity of t h e enzyme thymidme kinase
(.)[*].The activityof this enzyme i s subject to regulation,as may beseen
f r o m t h e fall in t h e curve toward the end of the infectious cycle. In the
late phase of infection a t the start of viral DNA synthesis
in t h e cell increases
amount of the virus-specific messenger RNA
strikingly. Mature virus particles are formed a few hours later (0).
Large numbers of copies of the virus nucleic acid are
produced in the infected cells, and a great quantity of
structural protein for the virus coat is synthesized.
These are the precursors, the components of the
newly maturing generation of virus particles, and
their assembly to form infectious virus particles constitutes the end of the virus replication cycle.
3. The Architecture of Viruses
Are there any general principles underlying the structure of virus particles? The possible shape that most
virus particles can assume is limited by the fact that
1121 D . F. Summers and J . V . Maizel, Proc. nat. Acad. Sci.
USA 59, 966 (1968).
[I31 E. D. Kiehn and J . .
J. Virology 5 , 358 (1970).
[ * ] After S . Kit in I . S . Colter and W . Paranchych: T h e Molecular Biology of Viruses. Academic Press, New York 1967.
Angew. Chem. internat. Edit. 1 Vol. 9 (1970) No. 9
only a limited space is available to contain the requisite genetic information. Thus, viruses must be
satisfied with a fairly simple structure, since if this
were complicated it would require many different
kinds of structural proteins; the small amount of
genetic information carried by viruses would certainly
not be sufficient for this purpose.
Just how few opportunities there are for variation in
the shape of virus particles was made clear more than
10 years ago in a review by Crick and Watson at the
symposium on the nature of viruses[141. Crick and
Watson assumed that the nucleic acid of a small virus
particle has a molecular weight of about two million,
which corresponds closely with reality. Such a nucleic
acid is composed of about 6000 nucleotides. As every
three nucleotides determine one amino acid, not more
than 2000 amino acids can be formed altogether and
be united in one protein. Such a protein, however, is
much too small to form a coat around the nucleic acid
of the virus particle. To fulfill this purpose, it can be
calculated that a protein must be at least 1 5 times as
large. All the genetic information of the virus particle
would then have to be used just for the formation of
the protein coat. We know, however, that the virus
nucleic acid has other functions to fulfil as well.
For this reason Crick and Watson concluded that the
coat of a virus particle cannot consist of a single large
protein molecule, but must be formed from many
identical but small protein molecules, as only a minor
section of the nucleic acid is required for the encoding
of small protein molecules. In this way the limited
genetic information is used economically. For the
same reasons Crick and Watson concluded even a t
that time that it seemed unlikely that many small
protein molecules of different kinds could participate
in the construction of the virus coat: each different
protein would require a section of the nucleic acid for
its specification, and we would end up with the same
difficulty. These possibilities are shown schematically
in Figure 6.
Fig. 6. Possible ways of assembling a virus coat. The diagram shows
possible ways in which protein molecules can coat the viral nucleic
acid, shown here by wavy lines. Case A, in which a single large protein
molecule surrounds t h e nucleic acid, is not found. Case C is frequently
found t o occur in small viruses. Here many small, identical, protein
molecules form a coat around the nucleic acid. The coat is only constructed of several different proteins in t h e larger viruses whose nucleic
acid has a higher information content (case B).
According to our present knowledge the above views
of Crick and Watson have proved to be thoroughly
sound. They do, however, need some modification. It
is clear that the structural complexity of the virus coat
[14] F. H . C. Crick and J . D . Watson in C . E . W . Wolstenholme and E . C. P. Millar: Ciba Foundation Symposium on
the Nature of Viruses. Churchill, London 1957.
is directly related to the information present in the
nucleic acid. There are, for example, small viruses,
such as the F2 phages, whose R N A has a molecular
weight of only 1x 106. This phage cannot encode more
than three proteins, one of which is that of the protein
coat of the virus particle [ljI. I n contrast, poxvirus, for
example, has a more complicated structure. Its DNA
has a molecular weight of 1 . 5 ~ 1 0 7 , and so it is not
surprising that the poxviruses have a very complex
architecture, whose subunits are composed not of one
but of several different proteins [161.
These proteins encoded by the virus genome are not
joined in simple fashion to give a membrane-like
envelope, but form small aggregates, as it were the
bricks of the virus coat. After appropriate staining
these bricks o r subunits of the coat appear in the
electron microscope as small cylinders or tubes,
pentagonal or hexagonal in cross-section. These are
termed capsomeres [171. The structure formed by all
the capsomeres is called the capsid, which surrounds
the nucleic acid in the form of a protective coat. Thus,
the virus coat consists of large numbers of radially
arranged capsomeres o r tube-like structures. Some
viruses are in addition surrounded by a membranelike structure, as already described in the previous
section. The mature infectious virus particle is formed
when all these different morphological structures are
assembled together.
The question now arises as to the configuration in
which the morphological subunits and thus the capsomeres are arranged, and we can further inquire as to
whether there are certain basic forms of particle
symmetry that are followed by the most dissimilar
viruses. X-ray crystallography and electron microscopy have shown that most viruses are in fact constructed in one of two ways. They either form helical
Fig. 7. Models of viruses possessing helical and cubic symmetry. The
figure on the left shows a tobacco mosaic virus with a helical arrangement of its monomer proteins; the latter are also the morphological
subunits of its coat. These proteins have been removed from the upper
part to show the spiral configuration of the RNA. The internal axis of
the rod-like particle is hollow [*I. The figure on the right is a model of a
cubic (icosahedral) virus particle. The coat is formed of radially arranged morphological subunits, the capsomeres. The viral nucleic acid
is in the center of the particle [**I.
1151 H . F. Lodish, Nature (London) 220, 345 (1968).
1161 .
. HoIowcrak and W. K. foklik, Virology 33, I17 (1967).
[I71 D . L. D . Caspar, R. Dulbecco, A . Klug, A . Lwoff, M . G . P.
Stoker, P . Tournier, and P . Wildy, Cold Spring Harbor Symposium on Quant. Biol. 27, 49 (1962).
[*I After A. Klug and D. L. D. Caspar, Advances Virus Res. 7,
225 (1960).
[**I After W. Bernhard, Umschau 67, 658 (1967).
tubes or rods or are spherical capsules with cubic
symmetry (Fig. 7).
We shall first consider the group containing the
smaller number of viruses, i.e. the rod-shaped or
helical viruses. The most studied and moFt prominent
member of this group is the tobacco mosaic virus
(TMV). This is an elongated rod whose distinctly
demarcated subunits have a coil-like configuration.
The structure of this virus has been extensively
analyzed in work extending over many years [I*]. TMV
is composed of 2650 chemically identical morphological subunits. Not only is the total number of these
subunits known, but also their molecular weight about 17500. The amino acid sequence of the monomer proteins and thus of these subunits has been
analyzed exactly: they are composed of 158 amino
acid residues. The ribonucleic acid, with a molecular
weight of 2x106, has a coil-like configuration and lies
between coils formed by the subunits. This arrangement implies that the inner axis of the particle is
hollow. Such a virus rod is remarkably stable, and its
RNA is very well protected against the action of
destructive enzymes such as RNase. The stability of
the virus particle is based primarily on the specific
binding forces between the protein molecules. Although the nature of these binding forces is still not
precisely known, it is suspected that they are essentially hydrogen bonds. In addition to the tobacco
mosaic virus there are other plant viruses characterized
by helical symmetry. Among the animal viruses a
helical structure is shown in particular by the myxoviruses; as already mentioned, this group includes the
influenza viruses.
Most kinds of viruses have coats with cubic symmetry.
Several regularly constructed bodies with cubic symmetry are distinguished; incidentally, these regular
solids were already described by Pyflragoras. The
simplest body is the tetrahedron with four sides, and
the next more complicated body is the hexahedron or
cube. In order of increasing complexity, then come
the octahedron, the dodecahedron, and finally the
complicated icosahedron. Of all these bodies the icosahedron is favored by most viruses as the basis of their
construction [191. It is natural to ask why the virus
particles are not constructed according to a more
simple pattern. In the evolution of organisms it is
generally the most economical principle that is preferred, and this in fact is true for the complicated
icosahedron, a regular polyhedron with 20 surfaces
and 1 2 vertices. With its many sides the icosahedron
approaches the shape of a sphere, and it is precisely
because of its many sides that it can be composed of
many bricks or subunits. Consequently, these subunits
and thus the proteins of the virus coat are also very
small. This is a very economical principle, which brings
us back to the idea put forward at the beginning, for
such protein molecules also only require a small part
of the genetic information of the virus particle for
[18} M. A. Lauffer and C. L. Stevens, Advances Virus Res. 13,
1 (1968).
1191 D . L . D . Caspar and A . KIug, Cold Spring Harbor Symposium on Quant. Biol. 27, 1 (1962).
Angew. Chem. internat. Edit.
Vol. 9 (1970) 1 N o . 9
their encoding. As we have already seen, the genetic
information in viruses is generally strictly limited.
It is not altogether easy to furnish proof that a virus
particle is constructed according to the rules of icosahedral symmetry. An icosahedron is composed of a
definite number of structural elements. Thus, in the
analysis of viruses by electron microscopy one attempts
to determine whether the total number of capsomeres
corresponds to this ruler201. It is not only the total
number of capsomeres that must conform to certain
crystallographic rules, but the arrangement of the
capsomeres must also be in agreement with such principles. An icosahedron has several axes of symmetry.
Figure 8 shows schematically an icosahedron from the
aspect of a twofold, a threefold, and finally a fivefold
symmetry axis. If one rotates the icosahedron body
around these axes through 180, 120, o r 72" it will
assume t h e same orientation as at the beginning. The
same must apply to the configuration of the capsomeres in the virus coat for icosahedral symmetry to be
allotted to alvirus. As an example of an icosahedron
Figure 8 gives in diagrammatic form the capsomere
arrangement of bacteriophage'yXl74. This*phage is a
very small virus particle; it consists of a total of
1 2 capsomeres [*I].
B @
Fig. 8. Symmetry properties of an icosahedron. Most cubic viruses are
constructed o n the principle of a n icosahedron, a polyhedron with
12 vertices and 20 surfaces, i . e . the arrangement and the number of its
capsomeres (cf. Fig. 7) follow the rules of icosahedral symmetry. A, B,
and C show two-, three-, and fivefold symmetry axes, respectively.
D shows the capsomere arrangement of a small virus with 12 capsomeres,
indicated simply by circles, and superimposed on this is the fivefold
symmetry axis of the icosahedron.
Not all kinds of viruses have a helical o r icosahedral
shape. There are also viruses which cannot be allotted
to any definite symmetry class. Thus, there is the
complicated T 4 phage with its hexagonal head containing the nucleic acid. Attached to this head is an
[20] A . KIug and D . L . D . Caspar, Advances Virus Res. 7, 225
[21] C. E . Hall, E . C. MacLean, and I . Tessmann, J. molecular
Biol. 1, 192 (1959).
Angew. Chem. infernat. Edit.
1 Vol. 9
No. 9
elaborate apparatus whose purpose is to inject the
nucleic acid into the host cell. Here we have a highly
differentiated morphological structure.
Finally, mention should be made of a problem which
has special relevance to general biology. This is the
question whether the assembly of a virus particle from
its constituent subunits is a self-organizing process,
which occurs automatically and does not depend on
anything other than t h e binding forces between the
protein subunits. Or does it require a special protein,
which controls the temporal and spatial pattern of the
formative processes - in somewhat the same way as
the organizer protein, which is known from the
differentiation processes of higher organisms?
For some years now it has been possible to break down
the tobacco mosaic virus into its monomer protein
subunits and ribonucleic acid, and then to reassemble
these components purely by mixing them in a testtube, i . e . in vitro 122,231. The virus particles produced
in this way do not differ in any way from those
developed normally in a cell, and they prove to be
fully infectious. Even in the absence of the ribonucleic
acid the monomer protein subunits are united by their
binding forces to produce a helical structure, which is
very similar to the mature virus particle. No external
influence is necessary for the assembly of a tobacco
mosaic virus particle, and for this reason the process
is said to be one of "self-assembly".
Recently, however, papers have been published which
indicate that certain RNA phages require another
kind of protein, in addition to their coating protein
and their RNA, in order to be able to assemble an
infectious particle. This protein has been called
maturation protein, as it seems to exert a regulating
function on the assembly of the virus coat. During the
in vitro reconstruction of such phages from their
components the presence of this protein seems to be
necessary to ensure combination of the RNA and the
coat protein to form infectious phages. It has been
shown that only a few molecules of this maturation
protein are necessary for one phage particle 1241. This
observation throws an interesting light on the morphogenesis of such simply formed structures as viruses.
It is just because of the narrow limitations of their
genetic information that viruses offer such excellent
models for the analysis of many processes which are
very difficult to study in higher organisms.
Received: June 15, 1970
[A 780 IE]
German version: Angew. Chem. 82, 723 (1970)
Translated by Express Translation Service, London
[22] G. Schramm, Z . Naturforsch. 2b, 249 (1947).
1231 H . Fraenkel-Conrat and R . C. Williams, Proc. nat. Acad.
Sci. USA 41, 690 (1955).
[241 J . A . Steitz, J. molecular Biol. 33, 923 (1968).
Без категории
Размер файла
827 Кб
formation, method, virus
Пожаловаться на содержимое документа