Copyright © 1989 by the Genetics Society of America

Perspectives

Anecdotal, Historical and Critical Commentaries on Genetics
Edited by James F. Crow and William F. Dove

REPLICA PLATING AND INDIRECT SELECTION OF BACTERIAL MUTANTS:
ISOLATION OF PREADAPTIVE MUTANTS IN BACTERIA BY SIB SELECTION

ISTORICALLY and conceptually, the two
themes in these titles (LEDERBERG and LEDER-
BERG 1952; CAVALLI-SFORZA and LEDERBERG 1956)
are complexly intertwined. Replica plating is a homely
methodology to ease the technical burdens of screen-
ing large numbers of bacterial colonies for mutants to
use in genetic analysis. This was sometimes easy. For
example, phage or drug resistance could be readily
obtained with positive selection using phage or a drug,
respectively, But many other kinds of mutants were
desired, and replica plating facilitated their discovery.
Furthermore, the broader philosophical question re-
mained open, whether the resistance was a preadap-
tive or a postadaptive change: did it emerge by natural
selection of rare preexisting mutants or did the agent
somehow induce the heritable change? I elaborate
first on this second broader question and its resolution
by indirect (sib) selection.

Luria and DELBRUCK (1943) introduced a carefully
thought out quantitative kinetics into the study of
bacterial mutation. The jackpot theory (LuRIA 1984)
was a consequence of clonal expansion. Spontaneous
mutations, in a culture started from a small inoculum,
would occur rarely among the few cells present at
early stages of exponential growth. But when this
happened they would have a disproportionate number
of mutant offspring. Hence, while the distribution of
mutational events should follow a Poisson distributior.
over a series of similar cultures, the distribution of
mutant cells would show occasional jackpots. Uncited
by and perhaps unknown to Luria and DELBRicK,
YANG and BRUCE WHITE (1934) had previously noted
such fluctuations and remarked that they signified the
spontaneous occurrence of “rough” mutants in Shi-
gella.

Nevertheless, the theoretical rigor of LurIA and
DELBRUCK’s analysis and its concurrence with the
quantitative data were for many the decisive stimulus
to look seriously at mutation in bacteria, to think
about a bacterial genetics. In my own development,
this paper was one of the key impulses to inquire

Genetics 121: 395-399 (March, 1989)

about genetic recombination in bacteria (LEDERBERG
1987).

By 1950 it was no longer controversial among ge-
neticists. Pace periodic selection, the monotonic in-
crease of mutant number with time in long-term che-
mostat cultures observed by Novick and SzILARD
(1950) sealed lacunae in the 1943 experiments. (Oth-
erwise, one could have entertained some labored hy-
pothetical alternatives like uncontrolled environmen-
tal variation from tube to tube, or a subset of that:
temporal fluctuations within a culture in its overall
propensity to turn resistant on exposure to a selective
agent.) Nevertheless, many, perhaps most, readers of
the 1943 article did not understand its abstruse math-
ematical argument and would respond either with
uncritical acceptance or uncritical rejection. Notable
for the latter, the last holdout, was the prestigious Sir
CYRIL HINSHELWOOD, President of the Royal Society
of London, who denounced every assertion of genes
in bacteria in favor of an extended reaction network
model of biological continuity and adaptivity (Hin-
SHELWOOD 1946; DEAN and HINSHELWOOD 1957). As
inappropriate as these network models have proven
to be for the fundamental elements of genetic struc-
ture, they have been revived for the explanation of
developmental switches, where one gene product rein-
forces its own synthesis and represses an alternative
gene’s and vice versa (DELBRUCK 1949; PTASHNE
1986).

Meanwhile, bacterial genetics had grown to the
point of needing ever larger libraries of mutants, most
of which bore biochemical defects and were not read-
ily amenable to positive selection. Today we apply
many ingenious tricks to this game (VINOPAL 1987).
In 1948, the penicillin method (LEDERBERG and ZIN-
DER 1948; Davis 1948) partly answered this need by
selectively killing cells actively growing in minimal
medium. Penicillin could exert positive selection in
favor of auxotrophic mutants. But many bacterial
strains were relatively recalcitrant to the feeble mu-
tagens than available, and even after penicillin selec-
396 J. Lederberg

tion there was still the tedious task of screening thou-
sands of colonies for the 1% or so that might be
growth-factor dependent. In addition, the escalation
of recombination studies imposed the equally tedious
task of classifying vast numbers of individual recom-
binant colonies to score them on a series of growth
factors, sugars, drugs and bacteriophages. For the first
several years of my work at the University of Wiscon-
sin, starting in the fall of 1947, I was deeply preoc-
cupied with these technical and doctrinal issues and
eager to follow other leads.

L. SzILARD and A. Novick, at the University of
Chicago, faced similar problems in scoring the phe-
notypes of an abundance of colonies. In February
1951, at one of the monthly phage seminars that
SzILARD had organized, they remarked that they had
been using multipronged inoculators, even a wire
brush, for a primitive kind of what I later called replica
plating (Novick 1972). This was not very satisfactory
owing to the poor resolution available with that ma-
terial.

For a couple of years prior to that point, possibly as
the result of a serendipitous accident, it had been
common practice in our laboratory to make impres-
sions of the dark and light colonies on eosin-methylene
blue agar plates simply by pressing a piece of paper
onto the agar surface and then mounting this under
cellulose tape in our laboratory notebooks. This was
far more convenient and cheaper than photography
for retaining a permanent record of the actual colony
distribution when this was manifest in a pigment dif-
ference. We had the idea of trying to use such prints
as inocula for fresh plates, but there was far too much
smearing on the paper to allow this to be useful (as
was also found by N. Visconti at the Cold Spring
Harbor Laboratory). (Have no fear about the biolog-
ical hazard in these notebooks: we have never been
able, alas, to recover viable organisms from these
impressions, and we do not advocate such a procedure
for spore formers or pathogens. Even so, this practice
is probably disallowed under present-day regulations.)

Meanwhile, HowARD NEWCOMBE (1949) evoked a
still more graphic image of the clonal expansion of
spontaneous mutants. Allowing a lawn of bacterial
growth on an agar surface, he would count the num-
ber of mutational events by directly spraying the
mature plate with phage. Each individual resistant
cell, or the clone clustered around it, would be scored
as a single colony. However, if the plate were respread
prior to selection, there was always a substantial in-
crease in the number of mutant colonies: this was a
direct translation of the clonal expansion of the indi-
vidual mutational stem cells. Sometime thereafter, I
elaborated on his experiment by making a single
streak of growth rather than a two-dimensional lawn.
I would then move a spreader perpendicular to the

stroke, and each clone would then be represented by
a line of resistant colonies along the direction of the
second stroke. These findings engendered still more
intense preoccupation with the imagery of what was
happening to mutant clones buried within the bacte-
rial population. If only there were some constructive
method to sample those clones prior to exposing them
to the selective agent!

Perhaps the multipoint sampling technology of
NOVICK and SZILARD could be applied to this broader
problem as well as to the tedium of colony scoring!
But: how to improve upon the poor resolution and
handling properties of the wire brush? Ep TaTuM had
taught me to use a beakerful of sterilized tooth picks,
one by one, for colony picking; that saved the time
needed to flame a platinum loop between picks. The
brush was conceptually an ordered array of tooth-
picks. What might be a functional equivalent?

Paper was unsatisfactory: its lateral capillarity and
its compression of the colonies distorted and broke
up the original growth pattern. It occurred to me that
some fabric with a vertical pile would be an analog of
the paper on one hand and the wire brush on the
other, and I soon collected a wide variety of remnants
from the local dry goods shops to put them to empir-
ical tests. (The predictable myth that I invaded my
wife’s wardrobe for this purpose is pure fantasy.) Also
helpful were books on fabric structure (like STRONG
1947) which helped me to focus on cotton velveteen
as the most desirable material. (Nylon velvets were
then far more expensive and their stiffer fibers caused
some problems.) The cotton velveteen has become
quite standardized and the material can, if necessary,
be conveniently laundered and resterilized for re-
peated use. The experimental trials soon confirmed
that one could rapidly enrich the proportion of resist-
ant mutants by fishing the growth at the point on the
original plate where a replica demonstrated the pres-
ence of a resistant clone. Without difficulty, one can
exclude about 99% of the irrelevant growth on a
bacterial lawn, which offers the prospect of a 100-fold
enrichment at every cycle. This means that one can
achieve a pure culture of resistant organisms in three

or four cycles based on selection applied to the sibling

clones that have been exposed to the selective agent.
This procedure thus accomplished a constructive
proof that had been so long elusive, namely, the
preadaptive initiation of these rare resistant mutant
clones. One can argue that this need never have been
in doubt, but here for the first time was a general
procedure that could be applied to any such proble-
matical situation.

JAMES CROw, my colleague in the Wisconsin Ge-
netics Department, enjoyed being able to analogize
bacterial indirect selection with sibling selection as
practiced in dairy and poultry husbandry. Bulls and
Perspectives 397

roosters are selected by indices of milk and egg pro-
duction by their sisters and daughters (LusH 1945).
At a time when the genetic basis of altruism is in
question (MICHOD 1982), we can reflex on a system
whereby replica-platefuls of clones, sensitive and re-
sistant alike, were relegated to the autoclave once they
had given the locational cues for their indirectly se-
lected cousins.

As satisfying as this demonstration was, I lamented
that it depended on a new technology (that of replica
plating). An Euclidean bent has always led me to seek
for proofs that came closer to the use of nothing more
than a straight edge and a compass. Furthermore, it
was difficult to quantitate the progress of indirect
selection on agar plates. It was difficult to maintain
rigorous control over the proportion of cells that were
transferred during replica plating or the proportion
that could be plucked by going back to the source
plate. There was no reason, I thought, that the same
methodology could not be used with an array of test
tubes taking the place of the agar plate, with pipettes
transferring well defined volumes from tube to tube
in place of the velvet transfer. This was the burden of
CAVALLI-SFORZA and LEDERBERG (1956): during one
of several visits to our laboratory, CAVALLI took an
interest in following the quantitative aspects of selec-
tion using that methodology. If one starts, say, with
100 tubes so adjusted that a mutational event will
have occurred in only one of them, one can find out
which tube that was by testing a sample of each for
drug resistance. Discarding the other 99 tubes would
in principle then give a 100-fold enrichment of the
proportion of mutant cells within that tube compared
to the overall population. Then one knows what di-
lution to make of the culture in that tube for a second
cycle of inocula and can recycle from there with
progressive enrichment. CAVALLI found that this pro-
cedure worked very well, although this progress of
selection was often not as rapid as predicted by the
first-order theory. This could be explained by the
observed growth lags of the resistant mutants and thus
offers no great problem.

Replica plating has grown into a major industry, its
progeny including the various blots around the com-
pass—Northern, Southern, etc.—as well as its direct
application to a number of microbiological problems.
Sib selection by serial dilution has been used in several
genetic engineering enterprises where a special pro-
ducing clone is a needle in the haystack but its prod-
ucts can be sensitively detected (KEDES et al. 1975;
NAGATA ¢ al. 1980).

A 1989 perspective on postadaptive mutation: The
concept of the gene as immutable in the course of
hereditary transmission (MORGAN 1926) was an ideal-
ization that played a constructive role in the emer-
gence of Mendelian genetics. As HALDANE (1949)

pointed out, this view taken to an extreme contradicts
the understanding of the gene as a material substance
(we would now say as DNA). We must have an open
mind about evolutionary specializations where meta-
bolic alterations can target the DNA itself. This might
sometimes lead to postadaptations, that is, adaptive
genetic changes specifically induced by an environ-
mental stress. As specialized evolutionary develop-
ments, one does not expect them to be a routine
occurrence. They have been hard to find and authen-
ticate, with one generic exception: lysogenizing vi-
ruses typically confer immunity to the lytic function
of the virus, a subset of lysogenic inductions (HERSHEY
1971; PrASHNE 1986). The frequency of “processed
pseudogenes” in eukaryotic genomes is particular tes-
timony to a history of RNA insertions (see pp. 448ff.
in DARNELL, LODISH and BALTIMORE 1986). With site-
directed mutagenesis we know today how to expose,
or even directly use, DNA sequences so as to achieve
mutation by intelligent design.

To turn to less specific responses, mutational storms
hypothetically related to insertional transposons may
also be mediated by RNA in train of environmental
stress (MCDONALD 1983). In addition, contemporary
with the development of replica plating, my own
laboratory had concluded that UV mutagenesis was
probably a secondary physiological response during
recovery and DNA repair (LEDERBERG eft al. 1951).
We know now how environmental DNA damage can
evoke the “SOS” response (WALKER 1987) leading to
several categories of genetic instability. One could
argue that UV resistance was a postadaptive mutation,
but this is to ignore the broad-ranging nonspecificity
of the mutations that occur during the SOS response.

It is too early to answer many questions that have
been raised during the past few months about posta-
daptive mutational responses claimed to occur in gly-
cosidase-deficient Escherichia coli mutants incubated
in the presence of lactose or salicin (CAIRNS, OVER-
BAUGH and MILLER 1988; CaIRNs e al. 1988; HALL
1988). There are many pitfalls in the exclusion of
artefacts in mutagenesis experiments (LEDERBERG
1948) and each has entrapped unwary investigators in
the course of microbial genetics. Such artefacts aside,
cells starved for carbon but receiving a trickle of
nutrient through spontaneous or allospecific enzy-
matic hydrolysis of the substrate are in a metabolic
state that requires critical examination. As a reducing
sugar, unmetabolized lactose might well be expected
to react with the DNA of such mutants (LEE 1987).
Even if it should be verified that lactose is a mutagen
for lac” mutants, its specificity, i.e., lactose at the lac
locus vs. salicin at the 4gi locus in a common genetic
background, should be corroborated before further
evolutionary speculation.

Specific DNA alterations are achieving higher cred-
398 J. Lederberg

ibility in a role in epigenesis, in reaction against long-
held dogmas of the uniformity of the genome of
somatic cells (LEDERBERG 1958). Segmental DNA ex-
cisions responding to an environmentally induced,
site-specific DNA recombinase are associated with ter-
minal heterocyst differentiation in Anabaena (HAS-
ELKORN et al. 1987). A similar story has just been
reported for the terminal differentiation of the
mother ceil during sporulation in Bacillus subtilis
(STRAGIER et al. 1989)

In microbial genetics, two other phenomena also fit
the paradigm of directed mutation. Bacteria can be
cured of many plasmids under the influence of acri-
dine dyes (HtrorA 1960) or other chemical and phys-
ical agents (TREVORS 1986). Acridine dyes alter mi-
tochondria in yeast and also remove kinetoplasts from
trypanosomes, and streptomycin ablates chloroplasts
from Euglena and other green plants (reviewed in
LEDERBERG 1952). Acridine sensitivity of bacterial
plasmids depends on the presence of specific DNA
sequences (WECHSLER and KLINE 1980). LEDERBERG
and ST. CLAIR (1958) showed that E. coli cells could
be converted en masse into spheroplasts with penicillin
in hypertonic media, and that these spheroplasts could
be propagated as wall-deficient clones in agar but
promptly reverted in the absence of penicillin. LAND-
MAN (1968) has reported, however, that wall-deficient
“L forms” of B. subtilis induced by lysozyme were
clonally propagated as L-forms in the absence of ly-
sozyme, although they would revert in solid media.
This is evidently an extranucleic event and deserves
further study as a unique instance in bacteria of mor-
phogenetic continuity of a cytoplasmic organelle, so
long the focus of study in the genetics of Paramecium
(Sapp 1987).

These exceptions notwithstanding, reinforcing the
Darwinian model of adaptive resistance mutation in
bacteria bolstered the eventual discard of instructional
theories of induced enzyme formation (MONOD 1956,
1966) and antibody formation (LEDERBERG 1989).
Any heuristic can be treacherous, but a Darwinian
explanation is the first I would seek in explaining a
biological enigma. I do not insist that it will always
last, but it has had enormous power in bringing us to
our present understanding.

JOSHUA LEDERBERG
The Rockefeller University
New York, New York 10021

LITERATURE CITED

Cairns, J., J. OVERBAUGH and S. MILLER, 1988 The origin of
mutants. Nature 335: 142-145.

CaIrns, J., ef al., 1988 Origin of mutants disputed. (Correspond-
ence of several authors pertaining to CAIRNS, OVERBAUGH and
MILLER 1988). Nature 336: 525-528.

CAVALLI-SFORZA, L. L., and J. LEDERBERG, 1956 Isolation of

preadaptive mutants in bacteria by sib selection. Genetics 41:
367-381.

DARNELL, J., H. LopisH and D. BALTIMORE, 1986 Molecular Cell
Biology. Scientific American Books, New York.

Davis, B. D, 1948 Isolation of biochemically deficient mutants of
bacteria by penicillin. J. Am. Chem. Soc. 70: 4267.

Dean, A. C. R., and C. HINSHELWOOoD, 1957 Aspects of the
problem of drug resistance in bacteria, pp. 4-24 in Drug
Resistance in Micro-Organisms (Ciba Foundation Symposium),
edited by G. E. W. WOLSTENHOLME and C. M. O’CONnNoR. J.
& A Churchill, London.

DeLprtck, M., 1949 Enzyme systems with alternative steady
states, pp. 33-4 in Unités Biologiques Douées de Continuite Géne-
tique (International Symposium CNRS No. 8). Editions du
CNRS, Paris.

HALDANE, J. B. S., 1949 In defence of genetics. Mod. Q. 4: 194-
202.

Hat, B. G., 1988 Adaptive evolution that requires multiple
spontaneous mutations. I. Mutations involving an insertion
sequence. Genetics 120: 887-897.

HASELKORN, R., J. W. GOLDEN, P. J. LAMMERS and M. E. MULLIGAN,
1987 Rearrangement of nif genes during cyanobacterial het-
erocyst differentiation. Phil. Trans. R. Soc. B 317: 173-181.

Hersuey, A. D. (Editor), 1971 The Bacteriophage Lambda. Cold
Spring Harbor Laboratory, Cold Spring Harbor, N.Y.

HINSHELWOOD, C. N., 1946 The Chemical Kinetics of the Bacterial
Cell. Clarendon Press, Oxford.

Hirota, Y., 1960 The effect of acridine dyes on mating type
factors in Escherichia coli. Proc. Natl. Acad. Sci. USA 46: 57-
64.

Kepes, L. H., A. C. Y. CHANG, D. HOUSEMAN and S. N. COHEN,
1975. Isolation of histone genes from unfractionated sea ur-
chin DNA by subculture cloning in £. coli. Nature 255: 533-
538.

LanpMaNn, O. E., 1968 Protoplasts, spheroplasts and L-forms
viewed as a genetic system, pp. 319-332 in Microbial Proto-
plasts, Spheroplasts and L-Forms, edited by L. B. GuZE. Williams
& Wilkins, Baltimore.

LEDERBERG, J., 1948 Problems in microbial genetics. Heredity 2:
145-198.

LEDERBERG, J., 1952 Cell genetics and hereditary symbiosis. Phys-
iol. Rev. 32: 403-430.

LEDERBERG, J., 1958 Genetic approaches to somatic cell variation:
summary comment. J. Cell. Comp. Physiol. 1: 383-402.

LEDERBERG, J., 1987 Perspectives: gene recombination and linked
segregations in Escherichia coli. Genetics 117: 1-4.

LEDERBERG, J., 1989 Reflections on Darwin and Ehrlich: the
ontogeny of the clonal selection theory of antibody formation.
Ann. NY Acad Sci. (in press).

LEDERBERG, J., and E. M. LEDERBERG, 1952 Replication plating
and indirect selection of bacterial mutants. J. Bacteriol. 63:
399-406.

LEDERBERG, J., and J. St. CLarr, 1958 Protoplasts and L-type
growth of Escherichia coli. J. Bacteriol. 75: 143-160.

LEDERBERG, J., and N. D. ZINDER, 1948 Concentration of bio-
chemical mutants of bacteria with penicillin. J. Am. Chem. Soc.
70: 4267-4268.

LEDERBERG, J., E. M. LEDERBERG, N. D. ZINDER and E. R. LIvELy,
1951 Recombination analysis of bacterial heredity. Cold
Spring Harbor Symp. Quant. Biol. 16: 413-443.

Ler, A. T., 1987 The nonenzymatic glycosylation of DNA by
reducing sugars in vivo may contribute to the DNA damage
associated with aging. Age 10: 150-155.

Luria, S. E., 1984 A Slot Machine, a Broken Test Tube: an Autobiog-
raphy. Harper & Row, New York.

Luria, S. E., and M. DELBRUCK, 1943 Mutations of bacteria from
virus sensitivity to virus resistance. Genetics 28: 491-511.
Perspectives 399

Lusu, J. L.,1945 Animal Breeding Plans, Ed. 3. lowa State College
Press, Ames.

McDonaLp, J. F., 1983 The molecular basis of adaptation. A
critical review of relevant ideas and observations. Annu. Rev.
Ecol. Syst. 14: 77-102.

MicuHop, R. E., 1982 The theory of kin selection. Annu. Rev.
Ecol. Syst. 13: 23-55.

Monon, J., 1956 Remarks on the mechanism of enzyme induc-
tion, pp. 7-28 in Enzymes: Units of Biological Structure and
Function, edited by O. H. GArBLER. Academic Press, New
York.

Monon, J., 1966 From enzymatic adaptation to allosteric transi-
tions. Science 154: 465-483.

Morcan, T. H., 1926 The Theory of the Gene. Yale University
Press, New Haven, Conn.

Nacata, S., H. Tatra, A. HALL, L. JOHNsRUD, M. STREULI, J.
Ecsopi, W. BOoLi, K. CANTELL and C. WEISSMANN,
1980 Synthesis in E. coli of a polypeptide with human leuko-
cyte interferon activity. Nature 284: 316-320.

Newcompe, H. B., 1949 Origin of bacterial variants. Nature 164:
50.

Novick, A., 1972 Introduction, pp. 389-392 in The Collected
Works of Leo Szilard. Part IV. Scientific Papers, edited by B. T.
FELD and G. W. SziLarb. MIT Press, Cambridge, Mass.

Novick, A., and L. SzILarD, 1950 Experiments with the chemo-

stat on spontaneous mutations of bacteria. Proc. Natl. Acad.
Sci. USA 36: 708-719.

PTASHNE, M., 1986 A Genetic Switch: Gene Control and Phage 3».
Cell Press and Blackwell Scientific Publ., Palo Alto, Calif.
Sapp, J., 1987 Beyond the Gene. Cytoplasmic Inheritance and the
Struggle for Authority in Genetics. Oxford University Press.
STRAGIER, P., B. KUNKEL, L. Kroos and R. Losick, 1989
Chromosomal rearrangement generating a composite gene for

a developmental regulatory protein. Science (in press).

STRONG, J. H., 1947 Fabric Structure. Chemical Publishing Co.,
New York.

TrREvors, J. T., 1986 Plasmid curing in bacteria. FEMS Microbiol.
Rev. 32: 149-157.

VinopaL, R. T., 1987 Selectable phenotypes, pp. 990-1015 in
Escherichia coli and Salmonella typhimurium: Cellular and Molec-
ular Biology, Vol. 2, edited by F. C. NEIDHARDT. American
Society for Microbiology, Washington, D.C,

WALKER, G. C., 1987 The SOS response of Escherichia coli, pp.
1346-1357 in Escherichia coli and Salmonella typhimurium: Cel-
lular and Molecular Biology, Vol. 2, edited by F. C. NEIDHARDT.
American Society for Microbiology, Washington, D.C.

WECHSLER, J., and B. C. KLINE, 1980 Mutation and identification
of the F plasmid locus determining resistance to acridine orange
curing. Plasmid 4: 276-280.

YANG, Y. N., and P. BRUCE WHITE, 1934 Rough variation in V.
cholerae and its relation to resistance to cholera-phage (type A).
J. Pathol. Bacteriol. 38: 187-200.