lac Operon Positive Control and Catabolite Repression

Positive control means a regulatory protein is required to turn a gene on, in contrast to negative control, where a repressor must be removed to permit transcription. The lac promoter uses both. The Lac repressor (LacI) exerts negative control: it binds the operator and sterically blocks polymerase, and this block is relieved when allolactose (or synthetic IPTG) binds the repressor and triggers a conformational change that lowers its DNA affinity. But even with the repressor gone, the lac promoter is a poor promoter — its −35 and −10 hexamers deviate from the consensus sequence recognized efficiently by sigma-70 RNA polymerase holoenzyme. Left alone, derepressed lac transcribes at only a small fraction of its potential rate.

CAP (catabolite activator protein, also called CRP for cAMP receptor protein) supplies the missing activation. CAP is a homodimeric DNA-binding protein that, when bound to its allosteric ligand cyclic AMP (cAMP), binds a specific DNA site centered at position −61.5 relative to the lac transcription start site, just upstream of the promoter's −35 element. CAP-cAMP bound there bends the DNA sharply (roughly 90°) and makes direct protein-protein contact with the alpha subunit of RNA polymerase, helping recruit and stabilize polymerase at the promoter. This raises transcription roughly 20- to 50-fold above the CAP-independent level. The cAMP signal itself is the second half of the story. Intracellular cAMP concentration falls when glucose is available and rises when glucose is scarce. The enzyme adenylate cyclase, which synthesizes cAMP from ATP, is inhibited (directly and indirectly) when glucose is being transported into the cell; correspondingly, cAMP accumulates when glucose is absent. Without cAMP, CAP cannot bind DNA productively, activation fails, and lac transcription — even fully derepressed — stays low. This is catabolite repression: the "repression" of an inducible catabolic operon by a preferred carbon source (glucose), acting not by directly silencing the gene but by depriving the activator of its cofactor.

Step 1 — Baseline promoter weakness. The lac promoter's −10 element (TATGTT) and −35 element (TTTACA) differ from the E. coli consensus (TATAAT and TTGACA), and the spacing between them is suboptimal. Sigma-70 holoenzyme therefore binds and isomerizes to open complex inefficiently at this promoter alone. This built-in weakness is not a design flaw; it is what makes CAP-dependent activation necessary and biologically meaningful. A "good" promoter would need no help and could not be regulated positively in this way.

Step 2 — Glucose signal sets cAMP levels. When glucose is transported into the cell via the phosphotransferase system (PTS), the sugar-specific transporter component EIIAGlc becomes largely dephosphorylated (its phosphate is donated down the PTS phosphorelay to incoming glucose). Dephosphorylated EIIAGlc fails to activate adenylate cyclase, so cAMP synthesis drops and cAMP is pumped out of the cell; this contributes to a phenomenon called inducer exclusion as well (dephospho-EIIAGlc also directly inhibits the lactose permease LacY, reducing lactose uptake). When glucose is absent, EIIAGlc stays phosphorylated, adenylate cyclase is stimulated, and cAMP accumulates — often cited around 10-fold higher than in glucose-replete cells, though the exact fold-change varies with growth conditions.

Step 3 — cAMP binds CAP and switches its conformation. CAP is a 209-residue polypeptide that dimerizes; each monomer has an N-terminal cAMP-binding/dimerization domain and a C-terminal helix-turn-helix DNA-binding domain. Binding of cAMP (two molecules per CAP dimer, though only one may be needed for full activity at physiological concentrations) reorients the DNA-binding domains into a configuration that fits the major grooves of the CAP DNA site, a roughly palindromic 22 bp sequence (consensus half-site 5′-AAATGTGATCT-3′-like sequence, actual lac site: TGTGAGTTAGCTCACTCATT, imperfect palindrome).

Step 4 — CAP-cAMP binds DNA and bends it sharply. The complex binds centered near position −61.5 of the lac promoter. Binding induces a pronounced DNA bend, on the order of 80-90°, wrapping the DNA around the protein. This bend is not incidental architecture — a specific surface of CAP, called activating region 1 (AR1), is thereby positioned to contact the C-terminal domain of the RNA polymerase alpha subunit (α-CTD), while a second surface (activating region 2, AR2, used at some Class II promoters) can contact the α-N-terminal domain or sigma region. At lac, which is a Class I CAP-dependent promoter (CAP site upstream of the −35 element), CAP works primarily through AR1-α-CTD contact.

Step 5 — Recruitment and stabilization of RNA polymerase. The α-CTD of RNA polymerase, tethered by a flexible linker to the rest of the enzyme, can swing to find its own weak DNA contacts upstream of a promoter (UP elements) or, at CAP-dependent promoters, dock instead onto the CAP surface. This protein-protein bridge increases the local concentration and dwell time of RNA polymerase at the promoter, raising the rate of closed-complex formation and subsequent isomerization to the transcription-competent open complex. The net functional outcome: transcription initiation frequency rises manyfold — commonly cited as 20- to 50-fold for lac — compared to a promoter with LacI removed but no CAP-cAMP bound.

Step 6 — Combinatorial output. The operon's transcription rate is therefore jointly set by two switches:

This is why lac transcription is not a simple binary switch but an AND-like gate: full induction needs lactose (removes repression) and glucose starvation (enables activation). The result is a physiologically sensible rule — use lactose only when it is present and nothing better is available.

Step 7 — Why this matters functionally. Growing E. coli on a glucose+lactose mixture produces diauxic growth: cells consume glucose first, pause briefly (the diauxic lag) while re-arranging gene expression, then resume growth on lactose once lac enzymes are induced. The pause reflects the time needed to synthesize new enzymes once cAMP rises and CAP activates the operon — a direct physiological signature of catabolite repression in action.

1. Monod's diauxic growth curves (1940s). Jacques Monod first characterized diauxic growth: E. coli grown on a mixture of glucose and another sugar (such as lactose) shows two distinct exponential growth phases separated by a lag, rather than simultaneous use of both sugars. This observation — that a "better" sugar is used preferentially and delays use of a second sugar — predated any molecular mechanism and set the empirical puzzle that catabolite repression would eventually explain.

2. Discovery that cAMP reverses catabolite repression — Makman and Sutherland; Ullmann, Pastan, and colleagues (1960s-1970s). Makman and Sutherland showed E. coli intracellular cAMP levels drop sharply when glucose is added to the medium and rise when glucose is removed, correlating cAMP directly with the presence/absence of the repressing sugar. Building on this, Ira Pastan, Agnes Ullmann, Jon Beckwith and colleagues showed that adding exogenous cAMP to glucose-grown cells could relieve catabolite repression and restore near-normal induction of lac and other catabolite-sensitive operons — the "cAMP rescue" experiment. This was the key causal link: it showed cAMP was not merely correlated with derepression but was sufficient to cause it, and that catabolite repression works through lowering cAMP rather than through some independent glucose-sensing mechanism.

3. Isolation of CAP and demonstration that cAMP acts through it — Zubay, Schwartz, Beckwith (1970); Emmer, de Crombrugghe, Pastan, Perlman (1970). Two groups independently purified an E. coli protein required, together with cAMP, for lac transcription in vitro. Geoffrey Zubay and colleagues developed a cell-free coupled transcription-translation system programmed with lac DNA (from a transducing phage carrying the lac operon) and showed that β-galactosidase synthesis in this extract absolutely required both cAMP and a specific protein factor — CAP — in addition to derepression of LacI. Extracts from crp mutant bacteria (lacking functional CAP) failed to support cAMP-stimulated lac transcription even when supplied with exogenous cAMP, directly demonstrating CAP is the obligatory mediator of the cAMP signal, not a redundant or parallel pathway.

4. Genetic identification of crp mutants (Class I catabolite-repression mutants). Isolation of E. coli mutants unable to induce lac (and other catabolite-sensitive operons) even in the absence of glucose, but rescuable by nothing (unlike cAMP-synthesis mutants, which were rescued by exogenous cAMP), mapped to a single gene, crp (catabolite repressor protein, i.e., the gene for CAP). This distinguished two classes of catabolite-repression-defective mutants:

• cya mutants (defective adenylate cyclase, cannot make cAMP) — rescued by adding exogenous cAMP.
• crp mutants (defective CAP itself) — not rescued by cAMP, because the receptor for the signal is broken.

This complementation logic — "does adding the small molecule fix the phenotype?" — is a classic genetic method for placing a gene product upstream or downstream of a signal in a pathway, and it unambiguously placed CAP downstream of cAMP and as the direct effector.

5. Direct DNA-binding and footprinting studies (1970s-1980s). DNase I footprinting of CAP-cAMP on lac promoter DNA (Majors 1975; Simpson 1980) showed a specific, cAMP-dependent protected region centered near −61, distinct from the operator and the −35/−10 elements. Gel mobility and circular permutation assays (Wu and Crothers, 1984) demonstrated the sharp CAP-induced DNA bend directly, by showing that DNA fragments with the CAP site at different positions relative to their ends migrated anomalously in a way consistent with a fixed bend angle rather than a uniform curve.

6. Structural and mutational dissection of the activating surface — Ebright and colleagues (1990s). Systematic alanine-scanning mutagenesis of CAP identified a discrete cluster of surface residues (around residue 156-164, defining "activating region 1," AR1) that, when mutated, abolished transcriptional activation at Class I promoters without affecting CAP's ability to bind DNA or bend it. Complementary mutations in the RNA polymerase α-subunit C-terminal domain (α-CTD) that specifically blocked activation (without blocking basal transcription) mapped the reciprocal contact surface on polymerase. This "mutant-cycle" or double-mutant analysis — showing that a CAP point mutation's defect could be specifically suppressed by a compensating α-CTD point mutation — provided strong genetic evidence for a direct, specific protein-protein contact, later confirmed structurally by co-crystal structures of CAP-DNA-α-CTD complexes. Together, these experiments moved the field from a phenomenological observation (diauxie) to a fully dissected molecular mechanism: glucose → lowers cAMP → inactive CAP → no α-CTD recruitment → weak transcription, with every arrow supported by an independent line of genetic or biochemical evidence.

Example 1 — Predicting phenotypes of lacI, crp, and cya mutants. Consider four E. coli strains grown in minimal medium with lactose as the sole carbon source (no glucose present). Predict whether each grows on lactose:

The distinguishing test between crp⁻ and cya⁻ is exactly the historical genetic experiment described above: add cAMP to the medium. A cya⁻ strain resumes near-normal induction (its defect is upstream, in signal production); a crp⁻ strain does not respond at all (its defect is in the receptor itself, downstream of the signal). This logic — "can I bypass the block by adding the product of the missing step?" — is a general and powerful genetic tool for ordering pathway components.

Example 2 — Quantifying the fold-activation from CAP, given competing inputs. Suppose in a reporter-gene assay measuring β-galactosidase activity (in Miller units) from the lac promoter, you observe the following under four combinations of IPTG (mimics lactose, releases repressor) and glucose (lowers cAMP):

Interpretation:

• Comparing A vs C isolates the effect of releasing the repressor (IPTG) when CAP-cAMP is fully active (no glucose): activity rises from 5 to 2100, a 420-fold induction — this combines both repressor release and full CAP activation.
• Comparing C vs D isolates the effect of adding glucose (lowering cAMP, inactivating CAP) while the repressor stays released (IPTG present in both): activity falls from 2100 to 65, a ~32-fold drop, which is attributable specifically to loss of CAP-cAMP activation. This number matches the textbook estimate that CAP-cAMP activation alone typically provides a 20-50-fold boost.
• Comparing A vs B (repressor bound in both cases; only 5 vs 3 units) shows that glucose has almost no further effect when the operon is already repressed — consistent with the idea that catabolite repression matters biologically only when the operon could otherwise be induced. This is exactly what you'd expect from an AND-gate: if one input (repressor released) is already "false," the state of the second input barely matters, because output is already near the floor.
• The residual "leak" values (2-5 units) reflect basal transcription from an uninduced, unactivated promoter — never exactly zero, because repressor binding and CAP absence reduce but don't perfectly abolish RNA polymerase access.

The molecular explanation for catabolite repression emerged from a productive rivalry and collaboration among several laboratories, but Ira Pastan and Agnes Ullmann, working with Jon Beckwith and in dialogue with the Monod-Jacob group at the Pasteur Institute, were central to establishing that cAMP was the missing signal and CAP its receptor. Pastan's laboratory at the NIH pursued cAMP's role in bacterial gene regulation after cAMP had already been established (by Earl Sutherland, in mammalian systems) as a universal "second messenger"; extending this concept to bacterial catabolite repression connected two fields that had previously seemed unrelated — hormone signaling in animals and sugar metabolism in bacteria. Agnes Ullmann, working between Pastan's group and the Pasteur Institute, helped show that exogenous cAMP could reverse catabolite repression in intact cells, the pivotal "rescue" experiment that made the causal chain concrete. Geoffrey Zubay's development of an efficient in vitro coupled transcription-translation system, capable of faithfully reproducing CAP- and cAMP-dependent activation of lac transcription outside the cell, gave the field a biochemical assay to purify CAP and dissect its requirements directly, complementing the genetic approach.

Understanding activator-dependent promoters and catabolite repression has practical reach well beyond E. coli physiology. In biotechnology, the lac promoter and its CAP-dependence are exploited (and deliberately bypassed) in expression vectors: many industrial and laboratory protein-expression systems use lac-derived promoters (such as tac or trc, hybrids retaining CAP-independence engineered in), specifically to avoid the complication of needing low-glucose conditions for strong expression, since fermentation media are often rich in glucose. Knowing precisely why native lac is glucose-sensitive lets bioengineers design around that sensitivity. In microbiology and infectious disease, catabolite repression-like regulatory logic controls virulence gene expression in several pathogens, where CAP/CRP homologs integrate nutrient status with the decision to express toxins or invasion factors, since many pathogens preferentially use host-preferred carbon sources before committing to a costly virulence program; CRP-family regulators are studied as potential targets for anti-virulence strategies that do not simply kill bacteria (avoiding strong selection for resistance) but instead disarm regulatory decision-making. In agriculture and industrial fermentation, catabolite repression governs which sugars are consumed preferentially by industrial microbes fermenting mixed sugar streams (e.g., from plant biomass containing both glucose and other sugars), and engineering strains to relieve catabolite repression (so that multiple sugars are used simultaneously rather than sequentially) is a major goal in biofuel and bio-based chemical production, directly improving fermentation yield and speed. Conceptually, the CAP-cAMP-lac system remains the pedagogical prototype for combinatorial gene control: the idea that a single promoter can integrate multiple independent regulatory inputs (an activator and a repressor, each responsive to a distinct signal) to compute an output more sophisticated than either input alone. This same logic — combinatorial binding of multiple transcription factors at cis-regulatory regions, each reading a distinct cellular or environmental state — scales up dramatically in eukaryotic enhancers and gene regulatory networks, the subject taken up directly in Combinatorial Control and Gene Networks.

Ask the tutor: How would the logic of this promoter change if the CAP binding site were moved to overlap the −35 element instead of sitting upstream of it?