Overview of Microbial Genetics

The bacterial genome is the complete set of genetic information in a prokaryotic cell: typically one main chromosome — a covalently closed circular double-stranded DNA molecule — plus zero or more plasmids, smaller extrachromosomal DNA circles (or, in some species, linear elements) that replicate independently. The chromosome is not naked in the cytoplasm; it is supercoiled and organized by DNA-binding proteins (such as HU and H-NS, functional but non-histone analogs) into a dense, irregularly shaped body called the nucleoid, occupying perhaps a sixth of the cell's volume without any surrounding membrane. Because there is no nuclear envelope separating DNA from ribosomes, prokaryotic gene expression is characterized by coupled transcription-translation: RNA polymerase transcribes messenger RNA (mRNA) while ribosomes simultaneously attach to the still-growing transcript and begin translating it into protein. This coupling has no counterpart in eukaryotic cells, where transcription is nuclear and translation is cytoplasmic, separated by RNA processing and nuclear export. Bacterial genomes are strikingly gene-dense. In E. coli, roughly 88% of the chromosome codes for protein or stable RNA, versus about 1.5% coding sequence in the human genome. Genes are frequently clustered into operons — a promoter, an operator, and two or more structural genes transcribed together as a single polycistronic mRNA. This lets a bacterium switch on an entire metabolic pathway (for example, the enzymes needed to catabolize lactose) with one regulatory decision rather than many, a theme fully developed in Bacterial Gene Regulation. Genome organization also differs from eukaryotes in the near-absence of introns in most bacterial genes; the sequence transcribed is, with few exceptions, the sequence translated, colinear from start codon to stop codon. The flow of genetic information in bacteria follows the same central dogma as all cellular life — DNA → RNA → protein — but executed with prokaryote-specific machinery: a single multi-subunit RNA polymerase (versus three separate polymerases in eukaryotes), ribosomes built from 30S and 50S subunits (forming a 70S ribosome, versus the eukaryotic 80S), and translation initiation guided by a Shine-Dalgarno sequence rather than a 5′ cap. Understanding this architecture — chromosome, nucleoid, operon, coupled expression — is the foundation for everything that follows in microbial genetics: how mutations arise and are detected (Mutation in Bacteria), how genes move between cells (Horizontal Gene Transfer), and how expression is dynamically tuned (Bacterial Gene Regulation).

Step 1 — Genome packaging into the nucleoid. The E. coli chromosome, if stretched out, would be about 1.4 mm long — roughly 700 times the length of the cell that contains it. This is achieved through negative supercoiling (introduced by the enzyme DNA gyrase, a type II topoisomerase) and organization into ~50 independent topological loops or "domains" anchored by architectural proteins (HU, H-NS, Fis, IHF). Supercoiling is not just for compaction: negative supercoils store torsional energy that makes local DNA strand separation easier, facilitating both replication and transcription initiation.

Step 2 — Replication initiation and semiconservative synthesis. Chromosomal replication begins at a single fixed origin, oriC</strong>, where the initiator protein DnaA binds AT-rich repeats and locally melts the double helix. Helicase (DnaB) is loaded, unwinding DNA at each of two replication forks that move outward in opposite directions around the circular chromosome — bidirectional replication — until they meet at a terminus region roughly opposite oriC (ter sites bound by Tus protein). At each fork, DNA polymerase III holoenzyme synthesizes new DNA continuously on the leading strand and discontinuously as Okazaki fragments on the lagging strand, each fragment initiated by a short RNA primer (made by primase, DnaG) later removed and replaced by DNA polymerase I, with DNA ligase sealing the nicks. Each daughter chromosome ends up with one parental strand and one newly synthesized strand — the semiconservative pattern proven by Meselson and Stahl (below). A typical E. coli chromosome replicates in about 40 minutes; cells growing faster than that overlap successive rounds of replication so that daughter cells inherit chromosomes already partway copied.

Griffith's transformation experiment (Frederick Griffith, 1928). Griffith worked with two strains of Streptococcus pneumoniae: a virulent, capsulated "S" (smooth) strain that killed mice, and a non-virulent, non-capsulated "R" (rough) strain that did not. Injecting mice with live R alone caused no disease; injecting heat-killed S alone caused no disease. But injecting live R mixed with heat-killed S bacteria killed the mice — and live S bacteria were recovered from the corpses. Some "transforming principle" from the dead S cells had permanently converted living R cells into S cells. Griffith did not know that principle was DNA, but he had discovered heritable transformation.

Avery, MacLeod, and MacCarty (1944). Oswald Avery, Colin MacLeod, and Maclyn McCarty set out to identify Griffith's transforming principle biochemically. They purified the active material from heat-killed S cells and subjected it to enzymes that specifically destroy protein (proteases), RNA (RNase), or DNA (DNase). Only DNase treatment destroyed the transforming activity; protease and RNase treatment left it intact. Furthermore, the purified material had a chemical composition (elemental ratios, UV absorption) consistent with DNA, not protein. This was the first direct biochemical demonstration that DNA is the hereditary material, a conclusion so at odds with the then-prevailing assumption that proteins (with their 20 amino acids) were too "complex" a molecule to carry genetic information that many biologists remained skeptical for nearly a decade.

Hershey and Chase (1952). Alfred Hershey and Martha Chase settled the argument using bacteriophage T2, a virus that infects E. coli and consists of only DNA packaged inside a protein coat. They grew separate batches of phage in media containing either radioactive sulfur (³⁵S, which labels protein but not DNA — sulfur is present in the amino acids cysteine and methionine but not in nucleotides) or radioactive phosphorus (³²P, which labels DNA but not protein, since phosphate is in the nucleic acid backbone but essentially absent from protein). Each labeled phage stock was allowed to infect fresh, unlabeled bacteria, and the mixtures were agitated in a blender to shear phage coats off the cell surfaces, then centrifuged to separate the heavier bacterial cell pellet from the lighter phage-coat supernatant. The result: most of the ³²P (DNA label) was found inside the bacterial pellet, while most of the ³⁵S (protein label) stayed in the supernatant with the empty phage "ghosts." Because progeny phage were produced from the infected bacteria, the genetic material that entered the cell and directed new phage production was DNA, not protein.

Meselson and Stahl (1958). Matthew Meselson and Franklin Stahl tested three competing models of DNA replication — conservative (the original double helix stays intact, and an entirely new one is made), dispersive (both strands are cut and mixed with new DNA throughout), and semiconservative (each new double helix has one old strand and one new strand). They grew E. coli for many generations in medium containing the heavy nitrogen isotope ¹⁵N, so that all cellular DNA became "heavy." They then switched the bacteria abruptly to normal ¹⁴N ("light") medium and sampled DNA after successive rounds of replication, separating DNA by density using cesium chloride equilibrium density-gradient centrifugation. After one generation in light medium, all DNA banded at a single intermediate density — inconsistent with the conservative model, which predicts one heavy band and one light band. After two generations, DNA banded in two positions: half at the intermediate density, half at the fully light density, with no heavy DNA remaining — inconsistent with the dispersive model, which predicts a single, ever-lightening band, and confirming that each strand of the original heavy duplex is conserved intact and paired with a newly made light strand. The experiment, often called "the most beautiful experiment in biology" (a phrase popularized by science writer John Cairns), directly demonstrated semiconservative replication.

Example 1 — Estimating replication time and overlapping replication rounds. E. coli replicates its ~4.6 × 10⁶ base pair chromosome bidirectionally from a single origin, with each of the two replication forks synthesizing DNA at roughly 1,000 base pairs per second. Since the two forks move in opposite directions and together must copy the whole circle, each fork needs to synthesize about half the genome:

This matches the empirically measured ~40-minute C period for chromosome replication. But E. coli can divide every 20 minutes under rich growth conditions — faster than one round of replication takes to complete. The resolution: cells growing this fast initiate a new round of replication at oriC before the previous round has finished, so a single chromosome can have multiple replication forks in progress simultaneously ("multifork replication"). A daughter cell is thus born already carrying partially replicated chromosomes, allowing the 20-minute division time to be shorter than the 40-minute replication time — a strategy unavailable to organisms without a simple circular chromosome and a single well-defined origin.

Example 2 — Interpreting a Meselson-Stahl-style density gradient after three generations. Suppose bacteria fully labeled with ¹⁵N (heavy) are shifted to ¹⁴N (light) medium and allowed to divide exactly three times, and total DNA is banded by CsCl density-gradient centrifugation. Using semiconservative replication logic: each round of replication, every existing DNA strand serves as a template for exactly one new (light) strand, while the pre-existing strand itself is passed on unchanged.

At generation 3, the gradient shows two bands: an intermediate (hybrid) band containing 25% of the DNA and a fully light band containing 75% — with no heavy band at all, since every original heavy strand, however many generations pass, is always paired with a newly made light strand and never regenerates a fully heavy duplex. A student who predicts a shrinking heavy band at generation 3 has implicitly (and incorrectly) reasoned as if replication were dispersive; the defining semiconservative signature is that the heavy band disappears completely after just one generation and never reappears, while the hybrid band's proportion — not its density — shrinks by half each generation thereafter.

Matthew Meselson (b. 1930) trained as a chemist at Caltech under Linus Pauling before turning to molecular biology, and it was there, alongside Franklin Stahl, that he designed the 1958 density-labeling experiment that confirmed semiconservative DNA replication — a mechanism James Watson and Francis Crick had proposed as a natural consequence of the double helix's base-pairing structure in 1953, but which required direct physical proof. Meselson and Stahl's insight was methodological as much as conceptual: rather than trying to see replication happen, they realized that the mass of DNA molecules — measurable with exquisite precision by equilibrium centrifugation in a cesium chloride density gradient — would carry an interpretable signature of how strands were distributed across generations, provided the parent DNA could first be made distinguishably "heavy" by isotopic labeling. The experiment took about a year of painstaking centrifugation runs and became a textbook model of how a well-chosen physical measurement can decisively distinguish between mechanistic hypotheses that look identical at the level of the genetic code itself. Meselson went on to a long career at Harvard working on DNA repair, recombination, and — notably — the scientific investigation of the 1979 Sverdlovsk anthrax outbreak, helping establish it as an accidental release from a Soviet bioweapons facility rather than a natural outbreak, tying his molecular biology back to public health and biosecurity.

Understanding bacterial genome organization underlies modern antibiotic development: drugs like rifampin target bacterial RNA polymerase directly, and fluoroquinolones target DNA gyrase, exploiting molecular differences between prokaryotic and eukaryotic replication/transcription machinery to kill bacteria while sparing human cells. Genetic engineering and biotechnology depend on bacterial plasmids as the workhorse vectors for cloning; the same coupled transcription-translation and simple operon logic (as in the classic lac operon) let engineers build inducible expression systems used to manufacture recombinant proteins, from insulin to vaccine antigens. In clinical microbiology, recognizing that resistance genes often live on mobile plasmids rather than the chromosome explains why resistance can spread between unrelated bacterial species almost instantly — a phenomenon fully explored in Horizontal Gene Transfer — reshaping how hospitals track and contain multidrug-resistant organisms. In synthetic biology, engineers exploit operon architecture to build genetic circuits (toggle switches, oscillators) using the same promoter-operator logic bacteria use naturally. And in evolutionary and public health surveillance, whole-genome sequencing of bacterial chromosomes and plasmids — made tractable because prokaryotic genomes are small, circular, and gene-dense — lets epidemiologists trace outbreak transmission chains in near real time, connecting the basic architecture described here directly to the study of Mutation in Bacteria and the spread of resistance.

Ask the tutor: How does multifork replication let a bacterium divide faster than its own chromosome takes to replicate?