Copyright © 1987 by the Genetics Society of America

Perspectives

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

GENE RECOMBINATION AND LINKED SEGREGATIONS IN Escherichia coli

AN article with this title was published in GENETICS
just 40 years ago' (LEDERBERG 1947), following
soon after the first discovery of recombination in
Escherichia coli strain K-12 (LEDERBERG and TATUM
1946). Its appearance coincided with my arrival at the
University of Wisconsin to become an assistant pro-
fessor of genetics. The work had been completed in
E. L. TaTum’s laboratory at Yale University between
March 1946 and June 1947. I then spent the summer
at the Marine Biological Laboratory at Woods Hole
writing my Ph.D. dissertation and the 1947 article?
(which was its most important chapter). These studies
had begun at Columbia University (in the Zoology
Department!) in FRANCIS J. RYAN’S laboratory in July
of 1945. Although his name does not appear in the
authorship, I had benefited enormously from his tu-
telage, encouragement and discipline. Homage to
Francis has been expressed more fully in my own
reminiscences (LEDERBERG 1986, 1987) and by others
as well (MoorE 1964; RavIN 1976). I was equally
fortunate to have had Tatum take me in his labora-
tory and share the then hard-won auxotrophic mu-
tants of E. coli K-12 that greatly facilitated the exper-
iments (LEDERBERG 1977, 1988).

Our first presentation about crossing in K-12 had
been to the Symposium on Heredity and Variation in
Microorganisms at Cold Spring Harbor in July, 1946.
Although it elicited many critical questions, I was most
fortunate to have had such an extraordinary forum in
which to respond to them. With the notable intransi-
gence of Max DELBRUCK aside, I encountered little
further entrenched skepticism about the result, except
from some few who had not participated in that de-
bate. What could be reported that July included:

1. The production of prototrophic recombinants
from the admixture of various auxotrophs. Stringent

‘It will be awkward to cite original sources at every point of historical
attribution: see especially LEDERBERG (1987), BACHMANN (1983), IPPEN-
THLER and MINKLEY (1986), and Jacos and WoLuMAN (1961) for compre-
hensive reviews. An important overview of E. coli and Salmonella genetics
has just been announced (NEIDHARDT 1987).

It is time to correct a typographical error in my 1947 paper: N. T. J.
Bal.ey has pointed out to me that “203” on the first line of Table 5 shouid
be “303.”

Genetics 117: I-4 (September, 1987)

selection allowed the detection of as few as one per
million recombinants and these occurred even from
parents doubly marked to forfend occasional sponta-
neous one-locus reversions.

2. The segregation among selected recombinants of
unselected markers, including auxotrophy (¢.g., pro-
line-less in proline-supplemented medium) and resis-
tance to phage T1.

The fact that TI resistance segregated among pro-
totrophs was an important datum, for it seemed to
rule out additive cell-mixture or nuclear-mixture (het-
erokaryosis) as an artefact. If the prototrophic isolate
was pure and stably sensitive, it could hardly contain
resistant cells. If it were a heterokaryon, it might be
either sensitive or resistant (in fact sensitivity is dom-
inant), but one would not expect a sharp segregation
into two stable categories of prototrophs, pure sensi-
tive and pure resistant. However, not everyone at
Cold Spring Harbor was so persuaded by these genetic
arguments, and I was compelled to promise to do
explicit single-cell isolations. MAX ZELLE helped me
to learn that technique, and it was to do good service
in later studies (ZELLE and LEDERBERG 1951; LEDER-
BERG 1956, 1957; NOssAL and LEDERBERG 1958).

So many new questions were now opened up by
these thrilling observations! How to react? For one
thing, I had to seek another year’s leave from medical
school; who in his right mind would leave the problem
at that stage? The entire project had been motivated
by Avery, MACLEOD and McCarrty’s discovery
(1944) of DNA as the transforming principle in the
pneumococcus. This could not be assimilated as the
chemistry of the gene without a broader base of
genetics in bacteria. We were disappointed, however,
not to find a way to use DNA directly in E. coli
genetics. (It took some years before a witch’s brew
was concocted to condition the cells.) Deoxyribonu-
clease did not influence the K-12 crosses, arguing for
some direct cell-to-cell interaction—perhaps like con-
jugation in ciliates, so brilliantly investigated by T. M.
SONNEBORN (1947; cf. WENRICH 1954). But it was
hard to design direct approaches to the physical mech-
2 J. Lederberg

anism of crossing when it remained such a rare, spo-
radic phenomenon. Even with hyperfertile strains, this
remains something of a difficulty today, compared to
the massive synchronization that facilitates kinetic
studies of viral infection.

The main issues that could be addressed at that
point, and which would contribute to the “Mendeli-
zation” of bacteria, were:

1. What is the range of markers that participate in
crossing? and

2. Are they organized in linkage groups or chro-
mosomes?

The work of 1946-1947 was then devoted to ac-
cumulating a wider panoply of markers and to im-
proving the methods for handling them. Besides the
auxotrophs and virus resistance, sugar fermentation
mutants were particularly attractive: they could be
acquired by visual inspection of the colonies on indi-
cator media (such as eosin-methylene blue agar), and
similar methods could be used to score the segregants
in numbers. The relevant enzymes, especially the di-
saccharases, would also be most readily amenable to
further studies. [The $-p-galactosidase of E. coli K-12
(LEDERBERG 1950) has certainly made its contribu-
tions to our understanding of gene action!|

Table 1 of the 1947 paper lists eight markers, in
addition to another eight auxotrophies used for re-
combinant selection. The former all segregated, and
recombined in every imaginable fashion, but not at
random. The statistics of co-segregation implied a
single linkage group, and permitted the figuring of
the first map, substantially consistent with today’s very
nearly complete mappings of over a thousand markers
(BACHMANN 1983) and the expectation that F. coli K-
12 will have its DNA completely sequenced (about 5
megabase pairs) within this decade. Reverse crosses
were used to show that the segregation ratios were
intrinsic to the locus rather than to the physiology of
the allele. Four-strand crossing-over was also looked
for as prototroph colonies containing two clones of
different crossover classes. These were rare, partly
because of the stringencies of selection. A later study
on microscopically isolated zygotic pairs corroborated
the loss of chromosome segments and gave evidence
of recurrent cycles of recombimation in the exconju-
gant clone (LEDERBERG 1957; ANDERSON and MAZE
1957). Needless to say, these efforts to graft the
classical concepts of chromosomal behavior in meiosis
onto recombination in bacteria have been overtaken
by molecular genetic perspectives.

A great and continuing puzzle was the failure to
find chromosomal aberrations even in heavily irradi-
ated stocks, in contrast to the results typical of Dro-
sophila and other eukaryotes. Yes, deletions (CooK
and LEDERBERG 1962) and, of course, insertions are
now well known, but inversions are rare indeed, apart

from those involved in adaptive gene regulations
(BoRST and GREAVES 1987). I am not aware of any
systematic study of the production of inversions in
bacteria by physical or chemical mutagens. I had
thought at the time that bacteria might lack explicit
enzymatic machinery for heterologous translocational
repairs of broken DNA. One also had to think of
differences in chromosomal organization: there was
no evidence of histones in bacteria. Later, the discov-
ery of species-specific repeated-sequence DNA in eu-
karyotes opened the possibility that this might furnish
homologous stretches for reunion modelled on cross-
ing-over. (Repetitious DNA is far less abundant in
bacteria.) I have looked in vain for published reports
on the distribution of chromosome rearrangements in
interspecific somatic cell hybrids that might test that
hypothesis; it is, however, supported by the correla-
tion of chromosome breakpoints in Drosophila rear-
rangements with repetitious DNA (LEE 1975; of.
Davis, SHEN and JuDD 1987). The matter was of some
consequence in our efforts, starting in the late 1960s,
to design ways of splicing foreign DNA into the bac-
terial genome (LEDERBERG 1969; CIFERRI, BARLATI
and LEDERBERG 1970; SGARAMELLA 1972; EHRLICH,
SGARAMELLA and LEDERBERG 1977; HARRIS-WAR-
RICK and LEDERBERG 1978). To this day we rely
mainly upon interaction of homologous sequences in
designing for DNA integration. The relative paucity
of inversions may also be a perspective of scale: there
is little intergenic spacing in E. coli. Most pairs of
breaks would do potentially lethal intragenic damage
in two places (not to mention the problems of revers-
ing the direction of transcription). SCHMIDT and ROTH
(1983) have discussed these and other contingencies
in connection with the rarity of inversions in Salmo-
nella,

The overall maps of distantly related species, like
Salmonella and E. coli, are remarkably well conserved
despite their large divergence in DNA homology. It
has then been proposed (RILEY and ANILIONIS 1978)
that the large-scale organization of the map in bacteria
may be functionally constrained, as it surely is in
viruses.

Our original linear map started to fall apart when
additional markers were recruited, especially xylose
and maltose fermentation and streptomycin resis-
tance. They just did not fit: in 1951, I published a
map needing three branches (LEDERBERG et al. 1951).
Despite my admonition at the symposium, and in the
caption, that “This diagram is purely formal and does
not imply a true branched chromosome,” the model
was taken as a concrete proposal rather than as a
portent of problems. J. D. WATSON and W. Hayes
(1953) thought that three chromosomes would suit
better than a branched monstrosity. Fortunately,
WOLLMAN and JAcos (1958; reviewed by JACOB and
Perspectives 3

WOLLMAN 1961) soon resolved these and other con-
fusions with their kinetic studies of progressive chro-
mosome transfer, sometimes interrupted in midpas-
sage, and the now accepted circular map. In my 1947
paper, the selective markers had provided operational
termini for the map, as noted in the discussion.

There are some points of tenderness here. Other
data (on persistent heterozygotes with hemizygous
reaches, NELSON and LEDERBERG 1954) implied that
chromosome segments might be deleted from either
parent. For some time (LEDERBERG 1955), I believed
that this post-zygotic elimination (shades of SCIARA or
the Mary Lyon effect!) was an alternative to progres-
sive transfer. That was plainly wrong-headed; but the
data still pertain, although interruptable progressive
transfer is clearly the first-order mechanism. Very
probably, both processes operate: the issue has been
virtually forgotten and has not been cogently ad-
dressed for almost 30 years. (1 left K-12 behind when
I moved from Wisconsin to Stanford in 1959, and
there concentrated on DNA-amenable systems like
Bacillus subtilis.)

It is many years since WOLLMAN and JaAcos illumi-
nated the kinetic mechanism of DNA transfer. We are
approaching the completion of the £. coli map in a
way that portends the sequencing of the human ge-
nome. E. coli today is perhaps exploited almost as
much for its technological potential as for fundamen-
tal studies. Nevertheless, some elementary aspects re-
main unsettled, in particular the precise physical con-
duit of the DNA strand as it is progressively passed
from one cell to the other, and what happens to it on
its way to the formation of recombinants.

JOSHUA LEDERBERG
The Rockefeller University
New York, New York 10021-6399

LITERATURE CITED

ANDERSON, T. F., and R. MAZE, 1957 Analyse de la descendance
de zygotes formés par conjugaison chez Escherichia coli K 12.
Ann. Inst. Pasteur 93: 194-198.

AVERY, O. T., C. M. MACLEOD and M. McCarry, 1944 Studies
on the chemical nature of the substance inducing transforma-
tion of pneumococcal types. J. Exptl. Med. 79: 137-158.

Bacumann, B. J., 1983 Linkage map of Escherichia coli K-12,
Edition 7. Microbiol. Rev. 47: 180-230.

Borst, P., and D. R. GREAVES, 1987 Programmed gene rear-
rangements altering gene expression. Science 235: 658-667.

CrFerRI, O., S. BARLATI and J. LEDERBERG, 1970 Uptake of
synthetic polynucleotides by competent cells of Bacillus subtilis.
J. Bacteriol. 104: 684-688.

Cook, A., and J. LEDERBERG, 1962 Recombination studies of
lactose nonfermenting mutants in Escherichia coli K-12. Ge-
netics 47: 1335-1353.

Davis, P. S., M. W. SHEN and B. H. Jupp, 1987 Asymmetrical
pairings of transposons in and proximal to the white locus of
Drosophila account for four classes of regularly occurring ex-
change products. Proc. Natl. Acad. Sci. USA 84: 174-178.

EHRLICH, S. D., V. SGARAMELLA and J. LEDERBERG, 1977 T4

ligase joins flush-ended DNA duplexes generated by restriction
endonucleases. pp. 261-268. In: Nucleic Acid-Protein Recogni-
tion, Edited by H. J. VocEL. Academic Press, New York.

Harris-Warrick, R. M., and J. LEDERBERG, 1978 Interspecies
transformation in Bacillus: sequence heterology as the major
barrier. J. Bacteriol. 133: 1237-1245.

IpPEN-IHLER, K. A., and E. G. MINKLEY, 1986 The conjugation
system of F, the fertility factor of Escherichia coli. Annu. Rev.
Genet. 20: 593-624.

Jacos, F., and E. L. WOLLMAN, 1961 Sexuality and the Genetics of
Bacteria. Academic Press, New York.

LEDERBERG, J., 1947 Gene recombination and linked segregations
in Escherichia coli. Genetics 32: 505-525.

LEDERBERG, J., 1950 The beta-p-galactosidase of Escherichia coli,
strain K-12. J. Bacteriol. 60: 381-392.

LEDERBERG, J., 1955 Genetic recombination in bacteria. Science
122: 920.

LEDERBERG, J., 1956 Linear inheritance in transductional clones.
Genetics 41: 845-871.

LEDERBERG, J., 1957 Sibling recombinants in zygote pedigrees of
Escherichia coli. Proc. Natl. Acad. Sci. USA 43: 1060-1065.
LEDERBERG, J., 1969 Health in the World of Tomorrow. PAHO/
WHO Lectures on the Biomedical Sciences, 1968. Pan Amer-
ican Health Organization, Scientific Publication No. 175.

Washington, D.C.

LEDERBERG, J., 1977 Edward Lawrie Tatum (1909-1975). Annu.
Rev. Genet. 13: 1-5.

LEDERBERG, J., 1986 Forty years of genetic recombination in
bacteria. A fortieth anniversary reminiscence. Nature 324:
627-628.

LEDERBERG, J., 1987 Genetic recombination in bacteria: a discov-
ery account. Annu. Rev. Genet. 21: 23-46.

LEDERBERG, J., 1988 Edward Lawrie Tatum. Biogr. Mem. Natl.
Acad. Sci. USA. In press.

LEDERBERG, J.,and E.L. Tatum, 1946 Novel genotypes in mixed
cultures of biochemical mutants of bacteria. Cold Spring Har-
bor Symp. Quant. Biol. 11: 113-114.

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.

Lee, C.S., 1975 A possible role of repetitious DNA in recombi-
natory joining during chromosome rearrangement in Drosoph-
ila melanogaster. Genetics 79: 467-470.

Moore, J. A., 1964 Francis Joseph Ryan, 1916-1963. Genetics
50: s15-s17.

NEIDHARDT, F. C. (Editor), 1987 Escherichia coli and Salmonella
typhimurium: Cellular and Molecular Biology, Vols. 1-2. Ameri-
can Society for Microbiology, Washington, D.C.

NELson, T. C., and J. LEDERBERG, 1954 Postzygotic elimination
of genetic factors in Escherichia coli. Proc. Natl. Acad. Sci. USA
40: 415-419.

Nossal, G. J. V., and J. LEDERBERG, 1958 Antibody production
by single cells. Nature 181: 1419-1420.

RAVIN, A. W., 1976 Francis Joseph Ryan (1916-1963). Genetics
84: 1-25.

RiLey, M., and A. ANILIONIS, 1978 Evolution of the bacterial
genome. Annu. Rev. Microbiol. 32: 519-560.

ScuHMipT, M, B., and J. R. RoTH, 1983 Genetic methods for
analysis and manipulation of inversion mutations in bacteria.
Genetics 105: 517-537.

SGARAMELLA, V., 1972 Enzymatic oligomerization of bacterio-
phage P22 DNA and of linear simian virus 40 DNA. Proc.
Natl. Acad. Sci. USA 69: 3389-3393.

SONNEBORN, T. M., 1947 Recent advances in the genetics of
Paramecium and Euplotes. Adv. Genet. 1: 263-358.

Watson, J. D., and W. Hayes, 1953 Genetic exchange in Esche-
richia coli K12: evidence for three linkage groups. Proc. Natl.
Acad. Sci. USA 39: 416-496.
4 J. Lederberg

WENRICH, D. H. (Editor), 1954 Sex in Microorganisms. American transfert de matériel génétique. Ann. Inst. Pasteur 95: 641-
Association for the Advancement of Science, Washington, D.C. 666.
Wo tan, E. L., and F. Jacos, 1958 Sur les processus de conju- ZELLE, M. R., and J. LEDERBERG, 1951 Single-cell isolations of

gaison et de recombinaison chez E. coli. V. Le méchanisme du diploid heterozygous Escherichia coli. J. Bacteriol. 61: 351-355.