Vor. 30, 1944 BACTERIOLOGY: S. E. LURIA 393

A GROWTH-DELA YING EFFECT OF UL7 A VIOLET RADIATION
ON BACTERIAL VIRUSES

By S. E. Luria

BACTERIOLOGICAL LABORATORIES, INDIANA UNIVERSITY*
Communicated October 20, 1944

Ultraviolet and ionizing radiations are known to inactivate bacterial
viruses (bacteriophages)."? The inactivated virus no longer multiplies on
sensitive bacteria or causes lysis. In the case of ultraviolet radiation, it
has been found in some instances that the inactivated virus still retains the
ability to be adsorbed by sensitive bacteria, and to interfere both with
their viability and with their ability to support growth of active virus.? .
This residual property of irradiated virus is itself destroyed by higher doses
of radiation,

Radiation is supposed to produce mutations in viruses. In the course
of attempts to increase by irradiation the rate of occurrence of mutations
affecting the host range of bacterial viruses,5 we encountered a new type of
physiological, non-hereditary effect of ultraviolet radiation, namely, a
growth-delaying effect, which will be described in the present communica-
tion.

Bacterial viruses grow by infecting living bacterial cells of sensitive.
strains and multiplying inside the cells. After a fairly constant interval
of time, the cells are lysed and liberate a large number of virus particles.*
We can describe the process in terms of a series of parameters: rate of
virus adsorption by bacteria; ‘‘constant period,” i.e., minimum interval
between infection and lysis; ‘‘rise period,” i.e., spread in the values of this.
interval for individual cells; ‘‘burst size,” i.e., average yield of virus par-
ticles per lysed cell. These parameters are easily reproducible under given
conditions for each virus-host system. Methods for their determination
have been described in detail.’

The radiation effect with which we are dealing was discovered when we
studied by these methods the growth of the fraction of irradiated virus that
had survived ultraviolet irradiation. Filtrates of various bacterial viruses,
all active on a common host, Escherichia col B, were irradiated in thin
layers (average about 0.02 mm.) with ultraviolet light containing about
80% of 2537 A. radiation. The virus surviving various times of irradiation
was then titrated and, within a few minutes, growth experiments of this
surviving virus were performed. Our conditions of irradiation, although
not such as to eliminate all screening effects by components of the medium,
were, however, accurate enough to give reproducible amounts of inactiva-
tion.

- The three viruses studied, a, y and 77, are known to be unrelated on the
394 BACTERIOLOGY: S. E. LURIA Proc. N. A. S.

basis of serological and cross-resistance tests. Viruses a and T7 consist of
small particles, virus y of large particles, as shown by electron micrography
and x-ray sensitivity tests.* ° Upon ultraviolet irradiation, virus y proved
the most sensitive, virus T77 more resistant, virus a very resistant. For
the same percentage of inactivation, the relative doses for the three viruses,
measured in time of exposure, were in the approximate ratios of 1:4: 10.

In the growth experiments using the virus surviving irradiation, ad-
sorption by sensitive bacteria proved normal. When the constant periods
were measured, however, they were found to be longer than for normal
virus. The rise period was also longer than normal, whereas the burst size
was normal. The values for the constant period are shown in table 1.

TABLE 1
“ GONSTANT PERIOD OF.
TIME OF GROWTH ON Escherichia

IRRADIATION, RESIDUAL col B (IN NUTRIENT

MIN. acriviry, % BROTH AT 37°C.), MIN.
Virus « 0 100 13
1 20 18
3 2 25
6 0.3 31
Virus + 0 100 21
0.5 1 - 24
Virus T7 0 100 13
1 7 23
5 0.2 , 28

Rise period and burst size were only determined in a few cases, because it is
difficult to run complete growth experiments with low initial virus titers.
We had to avoid the use of concentrated virus suspensions, in order to ex-
clude the possibility of interference effects from the inactive virus par-
ticles: all experiments were done with bacteria in excess of the total amount
of virus (active + inactive). The increase in rise period, however, was
clearly shown in all cases by changes in the slope of the growth curve of
virus.

- As a whole, the effect appears to consist of a delay in virus liberation,
without change in virus yield per bacterium. Not only is the minimum
time for lysis longer, but the intervals between infection and lysis for
individual bacteria are also more variable, as shown by the increase of the
rise period. Both constant period and rise period are progressively more
affected, for each virus, by increasing doses of radiation.

Similar results were obtained if the irradiated virus was diluted and
stored at 5°C. for one day between irradiation and the growth experiment.
The effect proved, however, strictly non-hereditary: the offspring of the
irradiated virus, when tested in its turn, grew on sensitive bacteria in the
same way as normal virus.
Vor. 30, 1944 BACTERIOLOGY: S. E. LURIA 395

It is seen in table 1: that the growth-delaying effect of radiation, while
increasing with the dose of radiation, does not parallel the inactivation
effect on the various viruses: for comparable amounts of inactivation,
growth delay is pronounced for the slowly inactivated viruses a and T7,
barely evident for the very sensitive virus y. It appears that the growth-
delaying effect is more closely related, for various viruses, to the total time
of irradiation, that is, to the number of quanta absorbed per unit of volume.
Since the delay effect must be brought about by the action of quanta which
have not acted lethally, and since it is a typically progressive effect, in-
creasing in intensity with increasing doses of radiation, one is led to con-
clude that it results from a cumulative effect of ultraviolet quanta on the
body of the virus particles.

Upon further irradiation, these particles would then become inactivated,
Inactivation is likely to be caused, not by the cumulative effect of a larger
number of quanta, but by a single “‘effective hit,” as is true in the case of
ionizing radiation, and as seems suggested by Gates’s results with ultra-
violet light.1_ A discussion of the mechanism of inactivation is, however,
beyond the scope of the present paper.

The discovery of a physiological effect of non-lethal ultraviolet quanta
on the particles of bacterial viruses raised an interesting question as to the
occurrence of a similar effect for x-rays. It has been conclusively proved® ®
that in the case of x-rays nearly each “‘hit,” that is, each elementary act of
absorption of radiation in a particle of a bacterial virus, is effective in
producing inactivation. A comparison of the ‘‘sensitive volumes” with
the actual volumes of the particles shows that the ratio of hits to effective
hits is not larger than 3:1 or 4:1, and is very likely close to 1:1. Under
such conditions, we should hardly expect any non-lethal effect of radiation
energy absorbed by the particles, comparable to the effect described above
for ultraviolet light.

This expectation was proved to be correct in a series of experiments in
which we exposed viruses a and y to doses of x-rays (200 kv.) sufficient to
reduce their titers to 0.1-5 per cent, and then studied the growth of the
surviving virus. In all cases the growth was completely normal: constant
period, rise period and burst size were like those of non-irradiated controls.
It is clear that no delaying effect on growth is produced by x-rays, as
expected in view of the high yield of inactivation. by each act of absorption.

The effect of ultraviolet light described in this paper is an example of a
non-lethal, non-genetic alteration of virus particles produced by radiation,
The change involved affects the time interval between adsorption of.a
virus particle by a sensitive host and liberation of new virus. It is im-
possible to decide at present which of the various steps involved in the
process leading to virus liberation is affected, whether it is the speed of
penetration into the bacterial cell, or the rate of multiplication inside the
396 BACTERIOLOGY: S. E. LURIA Proc. N. A. 5S.

cell, or the reactions responsible for cell lysis. A delay in any one or more
of these steps would account for the result. It is worth recalling that non-
lethal, growth-delaying effects of ultraviolet and ionizing radiation are of
familiar occurrence in the case of bacteria, yeasts, protozoa and cells
of higher organisms. The growth-delaying effect here described may well
be the expression of the same phenomenon in bacterial viruses.

The actual nature of the structural change causing growth delay is
also open to speculation. Since bacterial viruses seem to be composed
of nucleoproteins ™ 1" a study of the effectiveness of ultraviolet light of .
different wave-lengths, differently absorbed by specific components of
nucleoproteins, might throw some light on this question.

In conclusion, it is seen that virus particles, in spite of their small size
and probably simple composition, can undergo, under the influence of
ultraviolet radiation, structural changes of various kinds. Of these, some
affect the reproducing capacity of the virus; others even destroy the ability
of the virus to be adsorbed by the host cells and to interfere with the
growth of other virus particles; other changes may hereditarily affect some
of the properties of the virus particles; finally, changes of the kind here
described alter only the rate of some reactions involved in virus reproduc-
tion, in a strictly individual, non-hereditary manner.

The growth-delaying effect of radiation, detected for bacterial viruses,
may well occur with other groups of viruses. In most of them, however,
such an effect would be difficult to detect, since only for bacterial viruses
can we study an isolated step of growth of a given group of virus particlés,
Radiation effects of this type may be of some relevance in the study of virus
attenuation, partial neutralization, vaccine production and similar prob-
lems.

Summary.—Ultraviolet radiation, besides inactivating bacterial viruses,
produces a delay in the growth of the surviving virus particles on a sen-
sitive host; the delay increases with increasing doses of radiation. This
effect is non-hereditary, and differently pronounced for different viruses.
It appears to be due to the cumulative effect of the quanta absorbed by the
virus particles before their inactivation. No similar effect was found on
virus particles which had survived x-ray irradiation. This is explained by
the fact that, for ionizing radiation, nearly each act of absorption by a
vitus particle is effective in producing inactivation: the surviving particles
are likely not to have absorbed any radiation at all.

* Part of the experiments described in this paper were done in the summer of 1944 in
the Department of Genetics of the Carnegie Institution of Washington, Cold Spring
Harbor, New York. ,

1 Gates, F. L., Jour. Exptl. Med., 60, 179 (1934).

2 Wollman, E., and Lacassagne, A., Ann. Inst. Past., 64, 5 (1940).

3 Luria, S. E., and Delbriick, M., Arch. Biochem., 1, 207 (1942).

4 Gowen, J. W., Cold Spring Harbor Symp., 9, 187 (1941).
VoL. 30, 1944 BACTERIOLOGY: E. H. ANDERSON 397

5 Luria, S. E., Genetics, in press.

6 Delbriick, M., Advances in Enzymology, Vol. 2, 1 (1942).

1 Delbriick, M., and Luria, S. E., Arch. Biochem., 1, 111 (1942).

* Luria, S. E., and Anderson, T. F., these ProceEpincs, 28, 127 (1942)
9 Luria, S. E., and Exner, F. M., Ibid., 27, 370 (1941).

% Schlesinger, M., Biochem. Z., 264, 6 (1938).

11 Northrop, J. H., Jour. Gen. Physiol., 21, 335 (1938).