SCIENCE ON THE MARCH

159

SEX IN BACTERIA AND VIRUSES

HE manifestations of sex are known to

all, but its biological meaning is clear to

few since it can only be understood in terms

of evolution, Evolution by natural selection

takes place because there are present in nat-

ural populations large numbers of different

combinations of hereditary characters, some

. of which have greater chance than others of
survival and propagation.

New characters arise by mutations, that is,
by rare and unpredictable changes of deter-
minants of heredity, or “genes.” One muta-
tion seldom gives rise to an individual better
fit for life than the original. Favorable muta-
tions do, however, arise, if only with low
frequency. How will they be made use of
for the survival and improvement of the

species? As pointed out by Muller in 1932,

in asexual organisms, that is, in organisms
reproducing from one parent only, a mutant
gene will be tested for its survival value only
in combination with the other genes already

present in the line of descendance in which
ithas arisen. If several such favorable muta~
tions arise in different lines, they will compete
with one another and tend to crowd each
other out. Combinations of several favorable
mutant genes can only be formed if the
mutations happen to occur successively within
the same line of descent, necessarily a very
rare event. Much good material for evolution
will thus be wasted.

Sexuality prevents or at least. greatly re-
duces this waste. In the process of sexual,
biparental reproduction the genes of one
individual have at each generation a chance
to be reshuffled with those of another individ-
ual, producing a variety of offspring types.
In natural populations, these are generally all
more or less different from one another and.
from their parents. Reshuffling occurs be-
cause the genes are carried in chromosomes,
nucleoprotein threads present in various num-
ber of pairs in the nucleus of each cell, each
160

member of a pair being contributed by one
parent. The variety of combinations, arising
from the random assortment of the members
of each chromosome pair, is further increased
by exchanges of parts between chromosomes
of each pair within the parents (crossing-
over) in the course of the formation of the
sexual cells, the sperm and the egg. As a
consequence, no set of genes of a grandchild
is identical with any one of those found in
his four grandparents.

Moreover, the very increase in the total
number of gene combinations resulting from
sexual reproduction is an advantage for the
species: if environmental conditions vary,
gene combinations that might previously have
been unfavorable may prove useful in the new
exceptional situation and increase the chances
of survival of the species.

Finally, mutant genes that would be a
handicap in one combination might prove
actually advantageous in other combinations.
Sexuality, so to speak, gives them a chance.
In Muller’s words, “sexuality, through re-
combination, is a means for making the fullest
use of the possibilities of gene mutations.”

Since sex is such a powerful factor of sur-
vival, evolutionists have often wondered how
organisms like bacteria manage to survive
at all if they are, as is generally believed,
completely sexless and only reproduce by
repeated fissions. As pointed out before,
the number of available genetic combinations
will be quite limited, and the chance for the
appearance of more favorable ones rather
small. In part, this limitation might be
compensated for by the large size of bacterial
populations, which allows the occurrence of
even rare mutations, so that a number of
different combinations might actually be
formed within one hereditary line. In the
long run, however, bacteria would probably
be better off if they had some kind of sex.

In 1946 Lederberg and Tatum announced
the results of experiments strongly suggesting
sexlike phenomena in bacteria. From one
original parent strain of Escherichia coli, a
common intestinal bacterium, capable of
manufacturing all its protoplasmic materials
and of growing in a mixture of very simple
chemicals (sugar, ammonia, and inorganic
salts), these authors first isolated several

THE SCIENTIFIC MGNTHLY

“nutritionally deficient” mutants, that is, mu-
tants unable to manufacture some essential
chemical needed for their growth. These
mutants are easily obtained by irradiation of
the parent strain with ultraviolet light. A
mutant can only grow if the chemical or
chemicals which it is unable to produce is
supplied in the culture medium. Indirect
evidence, mainly the analogy with the situa-
tion in the mold Neurospora, had already
suggested that each nutritional deficiency
is tied up with a modification of one gene.

Let us suppose that the original strain
is capable of synthesizing the four substances
A, B, C, and D. A mutant d-B-C+D+
(deficient for compounds A and B as a
result of two mutations, but capable of manu-
facturing C and D) is grown together with
another double mutant with reverse synthetic
abilities, d+B+C-D-— in a medium con-
taining plenty of all four substances A, B, C,
and D. If the cells of the two mutants, while
growing together, undergo some type of
sexual fusion, exchanges of genetic determi-
nants might occur. Among such exchanges
there may be some leading to the formation
of cells 4+B+C+D+. These would be
nutritionally identical with the original strain
(prototrophic, from protos, “original,” and
trophe, “nutrition” and, like the latter, easily
distinguishable because able to grow in the
absence of all four substances A, B, C, and D.

Strains with at least two deficiencies had
to be used to eliminate the possibility of pro-
totrophic organisms being produced by one
mutation, The. chance of two mutations
occurring in the same line is low enough to
be completely negligible.

Lederberg and Tatum actually found pro-
totrophic cells present in mixed cultures of
a number of different mutants with several
deficiencies and interpreted them as proof
that genetic recombination had occurred.
This conclusion was supported by the fact
that, when strains differing by three or more
mutations were grown together, there ap-
peared in the mixed cultures some cells that
showed all possible combinations of the char-
acters of the two parents, indicating that the
transfer could involve different numbers of
genetic determinants.

The mechanism of fusion and gene ex-
SCIENCE ON THE MARCH

change in bacteria is still unknown, and the
results may need careful checking before
their interpretation as genetic recombinations
can be considered as proved beyond doubt.
It seems likely that, if sexual phenomena like
these are actually proved to occur at all in
bacteria, they will be found to be frequent
enough to play an important role in bacterial
evolution. It would, in fact, seem strange
that a mechanism so useful from the evolu-
tionary standpoint, once arisen in a group of
organisms, might remain confined to just a
few strains of bacteria. The very fact of
having been found in one of the few strains
carefully tested would suggest its widespread
occurrence.

Tf bacteria may have sex, what about the
filterable viruses? This might seem to be
asking too much indeed. After all, viruses
are ultramicroscopic entities reproducing in-
side living cells and often thought to be hardly
more than nucleoprotein molecules. And yet
we know that some of them undergo changes
very similar to sex in their genetic conse-
quences. This was discovered in the course
of work on bacteriophages, the viruses that
reproduce inside bacteria somewhat as the
virus of infantile paralysis reproduces in the
cells of the spinal cord of man, or the virus of
influenza in the lungs.

Bacteriophages can undergo a number of
different and independent mutations, and this
writer and Hershey have in recent years de-
scribed a number of them. In 1946 Delbriick
and Bailey infected the same bacterial cells
simultaneously with two bacteriophages car-
rying different mutations. As a result of this
mixed infection, new types of viruses, differ-
ent in their cormmbination of characters from
those that had entered the cells, were pro-
duced. These workers, for example, used
two bacteriophages, T2 and T4, somewhat
related but serologically and otherwise dis-
tinct, which can give mutants called T2r and
T4r. These are easily distinguishable from
T2 and T4 by the peculiar appearance of the
zones of destruction, or “plaques,” that they
cause in a layer of bacteria grown on solid
media. When particles of both viruses T2
and T4r entered the same bacterial cells, the
new phage produced inside these cells con-
sisted of four types: T2, T2r, T4, and T4r.

161

It was necessary to conclude that the property
designated as “r” had either been induced or
else transferred from one phage to the other
in the course of reproduction. Clearly, this
amounted to the same result as that of a
sexual cross, that is, the formation of new
combinations of hereditary characters. These
results were confirmed and extended by
Hershey with the study of several other
mutant properties. They.can be taken to
prove that, whatever the unknown mechanism
of virus growth inside the cell may be, it
must allow for some kind of reshuffling be-
tween the groups of hereditary determinants
brought into the same cell by different virus
particles.

In addition to this, this writer has recently
found that genetic recombinations between
viruses can do more than alter properties of
the above kinds; they can restore their repro-
ductive ability after this has been lost. Bacte-
riophage particles were inactivated by ultra-
violet rays and, as a result, lost their ability
to be transmitted from one bacterium to
another without losing that of penetrating
the host. Two or more such “inactive”
particles were made to enter one bacterium,
and active particles of bacteriophage reap-
peared. We have obtained evidence suggest-
ing that this “resuscitation” depends on the
fact that particles that have been damaged
by radiation in different parts can transfer
portions while inside the same host cell, with
an opportunity for the formation of normally
active particles. In other words, it seems
that a lethal mutation can occur in any one
of a number of parts of a virus particle, each
of these parts being replaceable by transfer
from another virus particle.

It is interesting to remember that more
than ten years ago Berry and Dedrick dis-
covered that the virus causing rabbit fibroma
could be transformed into the virus of
myxoma, a closely related virus of rabbits,
by simultaneous injection of active fibroma
and inactive myxoma virus. This phenom-
enon may find its explanation in some trans-
fer of genetic properties similar to that occur-
ring in bacteriophages.

It would be rash to draw from these results
any general conclusion as to the origin of
sexual processes.and their evolution. It is
162

possible that different mechanisms permitting
genetic recombination have appeared inde-
pendently at various stages of the evolution
of the biological world and that not all such
mechanisms have developed along similar
lines. The frequency of occurrence of genetic
recombinations is likely to be very low in
the case of bacteria and viruses, where re-
production by fission is probably the rule
and conjugation between different individuals
resulting in gene recombination, when pres-
ent, is the exceptional occurrence. It is,

THE SCIENTIFIC MONTHLY

therefore, possible that mechanisms for ge-
netic recombination that could prove too rad-
ical for higher organisms, where they would
be repeated at each generation, might work
successfully in viruses or bacteria. It would
clearly be dangerous to try forcing analogies
from the genetics of plants and animals to
that of bacteria and viruses until the latter
rests upon a more solid factual basis.

S. E. Luria
Department of Bacteriology
Indiana University