Special Viewpoints

Bacterial Sex: Playing Voyeurs 50 Years Later

See allHide authors and affiliations

Science  08 Aug 2003:
Vol. 301, Issue 5634, pp. 802-803
DOI: 10.1126/science.1085154


The concept of chromosomes with a ring structure was born during the early studies of bacterial sexuality, and the discovery of fertility factors— episomes or plasmids—provided much later the key tools for gene cloning and biotechnology. But the plasmid-mediated transfer of antibiotic and other resistances, as well as pathogenicity, has served bacteria well in their own adaptive evolution.

Although in the shadow cast by the 50th anniversary of DNA structure, the half-century anniversary of the discovery of the nature of bacterial sexuality must not pass unnoticed. In 1953, L. L. Cavalli-Sforza, J. Lederberg, and E. M. Lederberg (1) and W. Hayes (2) published in the same issue of the Journal of General Microbiology the identification of the F (fertility) factor as a transmissible agent that determines bacterial sexuality. As an hommage to these and other pioneers of bacterial genetics, which has made possible a detailed analysis of prokaryotic chromosomes and other genetic elements, as illustrated in the articles by D. J. Sherratt (3) and J. Hacker (4), we present a short historical account of their discoveries and a visualization of the accomplishment of the Escherichia coli sexual act using immunofluorescence microscopy.

The Amazing Story of the Discovery of BacterialSexuality

Progress in classical genetics was largely limited by the generation time of the test organism. For this reason, bacteria became the preferred model system, and consequently, the first proof that DNA is the genetic material was obtained from transformation experiments with Pneumococcus in 1944. However, this bacterium was inconvenient to use because it did not grow in synthetic media, which limited the number of useful genetic markers. J. Lederberg and E. Tatum (5) chose E. coli K-12, which can grow in synthetic medium, and isolated the so-called biochemical (auxotrophic) mutants. Then they mixed cultures of two different auxotrophs and obtained prototrophs (cells growing on unsupplemented medium). Initially, cellular fusion was thought to be responsible for prototrophic growth. Later, the unidirectional transfer of genetic material was demonstrated by mating streptomycin-sensitive and streptomycin-resistant cells (6). The existence of the subcellular agent called F, fertility or sex factor, responsible for genetic transfer was demonstrated by the observation that F+ character can be transmitted from F+ to F cells, without involving the bacterial chromosome (1, 2). The choice of E. coli K-12 strain by Lederberg and Tatum was critical for this amazing discovery. We now know that, by chance, this strain harbored the conjugative plasmid that, unlike the majority of F-like plasmids, was a naturally occurring mutant transferring at elevated frequency. Furthermore, it contained insertion sequences, which facilitated its integration into the host chromosome.

The nature of chromosome transfer was elucidated by studying conjugational transfer from Hfr (high frequency of recombination) donor strains (6, 7). Matings between Hfr and F strains with multiple genetic markers showed that there was a hierarchy in transfer efficiencies of different markers. The interpretation was that Hfr injects its chromosome to F from a genetically defined chromosomal point; the rationale was that the closer the marker to the injection point, the higher the production of recombinants. This hypothesis was proven by F. Jacob and E. Wollman (7) by the famous “blender experiment,” i.e., by interrupting mating at different time points. They found that the Hfr character is transferred as the last marker, whereas the F factor is transferred rapidly. The discovery of F carrying the host lac gene, which can transform F+ into an Hfr after its integration into the chromosomal lac gene, led to the conclusion that an F+ strain has the F factor in an autonomous state called an “episome,” whereas in Hfr strains, the F is integrated into the chromosome. Analyzing the characteristics of the transfer origin and its direction in different Hfrs, Jacob and Wollman concluded that the chromosome of E. coli exists as a circle. Their hypothesis was later proven by J. Cairns' famous autoradiograph picture (8).

The first hypothesis about the mechanism of F-mediated DNA transfer was provided by F. Jacob and S. Brenner in their “Replicon” theory (9). They proposed that every autonomously replicating unit has its own initiator (protein) and replication origin (DNA site). Furthermore, they imagined that the start of the Hfr chromosome transfer is initiated by the interaction between the F initiator and its origin and is followed by DNA replication. Therefore, the mating should be dependent on DNA replication. However, they found that the inhibition of DNA synthesis by various means had no effect on mating. Parallel to these studies, they tried to isolate a mutant of the chromosomal DNA replication initiator among conditional lethal mutants. According to their theory, even after inactivation of the host initiator, chromosomal replication driven by the F-factor initiator should resume during mating. Although they could isolate the host initiator mutant, dnaAts (10), they failed to show specific conjugation-dependent chromosomal synthesis. The first elements of an answer were provided in 1963 by W. Kunicki-Goldfinger from Warsaw, who found an increment of single-stranded DNA in conjugated bacteria, which suggested that the chromosomal transfer occurs without DNA synthesis, simply by separating two DNA strands and transferring only one strand to F.

Visualization of the Bacterial Sex by Immunofluorescence Microscopy

Even today, most genetic studies of bacterial mating depend on the analysis of the clonal descendents of the recombinant cells, i.e., colony formation. Using immunofluorescence microscopy, we have peeped into the bacterial mating act in single cells (Fig. 1). The transferred Hfr (Fig. 1A) or F+ (Fig. 1B) DNA was visualized as yellow foci of the hemimethylated DNA binding protein, SeqA, in the (green) mated F cells (see legend to the figure). The appearance of these foci present yet another proof that the methylated donor (dam+) DNA is transferred as single-stranded and copied in the methylation-deficient (dam) recipient cell, which results in a stably hemimethylated DNA bound by SeqA. Up to 75% of F cells show evidence of successful sex (see legend).

Fig. 1.

The GFP (green fluorescent protein)–producing F cells were distinguished from the donor (Hfr or F+) cells by a monoclonal antibody against GFP (green fluorescence). The transfer of the donor chromosome into F cells was detected indirectly by using an antiserum against SeqA protein (red fluorescence). This protein has a high affinity for hemimethylated DNA. Consequently, in a wild-type strain, SeqA forms foci around the replication forks where the newly synthesized hemimethylated DNA is transiently produced (11). However, in DNA methylation-deficient dam mutants, the SeqA protein is diffuse (11). When a single-stranded DNA is transferred from Hfr (A) or F+ (B) dam+ donor strains into Fdam GFP cells and copied in the absence of methylation activity, it becomes a permanently hemimethylated double-stranded DNA, to which the SeqA protein binds. The appearance of SeqA foci [yellow, in (A) and (B)] exclusively in the mated F (green) cells signals the presence of transferred DNA. Independent cultures of each parental strain show (black) cells with only red SeqA foci [HfrH in (C)] or (green) cells without SeqA foci [Fdam in (D)]. The mating was performed by mixing donor and recipient cells (1:1) on a Millipore filter for 100 min at 30°C before processing for microscopy as described (11).


It is difficult to overstate the importance of the discovery of bacterial sex for the development of molecular genetics and biology. Even the ring structure of the bacterial chromosome was first revealed by the study of conjugation. Conjugation, transduction, and transformation are still basic tools for cloning, characterizing, and sequencing all genes and genomes. But bacterial sex is extremely important to bacteria themselves. In the evolutionary past, horizontal gene transfer and recombination have shaped bacterial genomes, which appear as complex mosaics composed of genes from different lineages, species, and genera. One no-less-important evolutionary role of bacterial sex is the selective advantage provided by the plasmid-mediated transfer of resistance to antibiotics and heavy metals, as well as pathogenicity genes. Formidable sexual promiscuity has given bacteria a unique advantage over other creatures because it provides an awesome mechanism for ongoing adaptive evolution—a sort of permanently and rapidly evolving communal genome.

References and Notes

View Abstract

Stay Connected to Science

Navigate This Article