lac Promoter: Structure and Induction Explained

By Dr. Zubair Khalid, DVM, MS, PhD · 2026-09-20

The lac promoter is the DNA sequence in Escherichia coli where RNA polymerase binds to begin transcription of the lactose utilization genes. It is a weak sigma-70 promoter that is switched on only when two conditions are met: the sugar glucose is scarce, and a lactose-derived inducer is present to remove the LacI repressor.

Two sentences are enough to define it, but the lac promoter has shaped how molecular biologists think about gene regulation for more than sixty years. It gave us the first working model of a repressor, the first example of positive control by an activator, and the first inducible expression system that a bench scientist could buy in a kit. Almost every inducible bacterial expression vector in use today, from pET to pTrc to pGEX, borrows at least one part from this promoter. Understanding its architecture explains why some inductions work beautifully and others leak, stall, or fail outright.

What the lac promoter actually is

A promoter is the DNA element that recruits RNA polymerase and sets the transcription start site. The lac promoter sits just upstream of the lacZ gene in the lac operon and drives a single polycistronic mRNA that carries lacZ, lacY, and lacA. The promoter itself is a compact piece of DNA, roughly 80 base pairs of regulatory sequence, but it is crowded with binding sites. The core promoter elements, the catabolite activator protein (CAP) site, and the main lac operator all overlap or sit within a few helical turns of each other. That crowding is not accidental. It is the physical basis of the logic.

The lac promoter is often called a "weak" promoter, and that label is precise. Weak means that in the absence of CAP, RNA polymerase binds it poorly and initiates transcription at a low rate. The wild-type lac promoter is not designed for maximum output. It is designed for regulation, and regulation requires headroom. A promoter that is already firing at full speed cannot be turned up much further by an activator.

Architecture of the lac promoter

The -35 and -10 hexamers

Bacterial sigma-70 promoters are recognized by two short sequences upstream of the transcription start site (the +1 position). The -35 element sits about 35 base pairs upstream, and the -10 element sits about 10 base pairs upstream. The sigma factor of RNA polymerase makes sequence-specific contacts with both.

The consensus -35 element is TTGACA and the consensus -10 element is TATAAT. The wild-type lac promoter deviates from both. Its -35 region reads TTTACA, a single mismatch, and its -10 region reads TATGTT, which differs from consensus at two positions. Those mismatches reduce the intrinsic affinity of RNA polymerase for the promoter. The spacing between the elements also matters. A 17 base pair spacer is optimal, and the lac promoter is close to that but not perfect.

The practical consequence is a low basal rate of transcription. When CAP is absent, the lac promoter fires at a small fraction of its activated level. The exact fold-activation depends on the assay and growth condition, but the textbook figure is on the order of 20 to 50 fold stimulation by CAP-cAMP.

The upstream CAP site

The catabolite activator protein, also called the cyclic AMP receptor protein or CRP, binds a DNA site centered roughly 61.5 base pairs upstream of the transcription start site [1]. The CAP site is an inverted repeat, and CAP binds it as a homodimer. The consensus CAP site is 5'-AAATGTGATCTAGATCACATTT-3', and the lac promoter carries a near-consensus version [2].

When cyclic AMP (cAMP) binds CAP, the complex bends the DNA and contacts RNA polymerase directly. The activating region of CAP maps to amino acids 156 to 162, a surface loop that makes protein-protein contact with the polymerase [3]. This contact is the reason CAP works. CAP does not simply clear the DNA or recruit polymerase by mass action. It physically touches the enzyme and stabilizes the transcription complex.

The mechanistic details are well worked out. CAP stabilizes the closed complex, where polymerase is bound but the DNA is not yet melted, and it accelerates the isomerization to the open complex, where the DNA strands separate around the -10 region. Kinetic analysis shows that the acceleration of open complex formation accounts for most of the transcriptional activation [4]. A half-turn insertion between the CAP site and the polymerase site abolishes the stabilization, which confirms that the two proteins must sit on the same face of the DNA helix to interact [4].

The -45 region

Between the CAP site and the core promoter sits a 13 base pair sequence from -38 to -50, sometimes called the -45 region [1]. This region is not a classic promoter element, but it modulates both the CAP-independent and CAP-dependent activity of the lac promoter. Random mutagenesis of this region produced a 26-fold range in CAP-independent expression, and the strongest variants required an intact RNA polymerase alpha subunit for activity in vitro, which is the signature of an upstream DNA recognition element that contacts the alpha C-terminal domain [1]. Mutant promoters also varied in their response to CAP, from 2-fold to the wild-type level of about 22-fold [1]. The -45 region is a reminder that the lac promoter is not just two hexamers and a CAP site. It is a tuned sequence.

The operator and the overlap problem

The main lac operator, O1, overlaps the promoter. It is centered near +11, which places it downstream of the transcription start site but still within the region that RNA polymerase must occupy. When the LacI repressor tetramer binds O1, it physically blocks polymerase from forming a productive open complex. Repression and activation are therefore competing for overlapping real estate.

E. coli has three lac operators: O1, the main operator, and two weaker pseudo-operators, O2 and O3. O2 sits downstream within lacZ, and O3 sits upstream near the CAP site. The three operators cooperate. Removing O2 or O3 alone reduces repression by about 2 to 3 fold, but removing both reduces repression more than 50 fold. O1 alone represses only about 20 fold [5]. The cooperative effect comes from DNA loop formation. A tetrameric LacI can bind two operators at once and hold the intervening DNA in a loop, which raises the effective local concentration of repressor and tightens repression. A dimeric LacI that cannot loop represses the operon only weakly [5].

There is also a physical interaction between CAP and LacI at the promoter. The ternary complex with both proteins bound is 4 to 11 fold more stable than predicted from their independent binding affinities, even though the two proteins have opposite effects on transcription [6]. The CAP and repressor sites are separated center-to-center by 72 base pairs, which is 6.9 turns of B-form DNA, so both proteins sit on the same face of the helix. DNA curvature appears to facilitate the contact [6]. This is one of the more counterintuitive facts about the system: the activator and the repressor help each other bind, and the logic still works because the inducer removes the repressor and the activator then drives transcription.

How induction works

The glucose side: cAMP and CAP

E. coli prefers glucose. When glucose is abundant, the cell keeps cAMP low, CAP is mostly in its cAMP-free form, and the lac promoter runs at its weak basal rate. When glucose runs out, cAMP rises, CAP binds cAMP, and the CAP-cAMP complex binds the upstream site and activates transcription. This is catabolite repression, and it is the reason the lac operon is sometimes described as glucose-sensitive.

The logic is economical. There is no point in building lactose-metabolizing enzymes when glucose, a better carbon source, is available. The cell waits until glucose is depleted before investing in the lactose machinery.

The lactose side: inducer and LacI

The second input is the inducer. In the natural system, the inducer is not lactose itself but allolactose, a minor side product of the LacZ enzyme. Allolactose binds LacI and changes its conformation so that it releases the operator. Once the operator is free, RNA polymerase can access the promoter, and if CAP-cAMP is also present, transcription proceeds.

The system has a chicken-and-egg problem that it solves elegantly. Lactose does not induce well on its own because the cell needs LacZ to make allolactose, and LacZ needs to be expressed to be present. A small amount of basal transcription from the leaky promoter produces enough LacZ to kickstart the loop. This positive feedback is part of why the lac operon can show bistable, all-or-nothing behavior in single cells, especially at low inducer concentrations [7].

IPTG as a gratuitous inducer

Isopropyl beta-D-1-thiogalactopyranoside, universally called IPTG, is the standard laboratory inducer. It is a gratuitous inducer, meaning it binds LacI and relieves repression but is not metabolized by the cell. The sulfur in place of the glycosidic oxygen blocks hydrolysis by LacZ, so the inducer concentration stays constant throughout the experiment. That is the practical advantage over lactose, which is consumed and whose effective concentration drops as the culture grows.

IPTG uptake is more complicated than the textbook cartoon suggests. At low concentrations, IPTG enters cells mainly through the lactose permease LacY. At higher concentrations, it can cross the membrane independently of LacY [8]. This matters for induction kinetics and for population heterogeneity. Single-cell analysis shows that bimodal induction, where some cells are fully induced and others are not, occurs at roughly ten-fold lower IPTG concentrations than with the alternative inducer TMG [8].

There is a further wrinkle. The transacetylase LacA, the third enzyme of the operon, influences apparent induction. In a strain carrying lacA, fluorescence from a Plac-gfp reporter decreased after a few hours, while in a lacA deletion strain it continued to rise [8]. The interpretation is that LacA activity can reduce the effective IPTG concentration below the inducing threshold, which means the third gene of the operon feeds back on the induction state [8]. This is a good example of why the lac operon is more than a simple switch.

A table of components

The induction decision path

The following flowchart shows how a cell decides whether to transcribe from the lac promoter, given the two environmental inputs.

flowchart TD
    A[Glucose level] --> B{Glucose low}
    B -->|No| C[cAMP low]
    B -->|Yes| D[cAMP high]
    C --> E[CAP inactive]
    D --> F[CAP cAMP active]
    G[Inducer level] --> H{Inducer present}
    H -->|No| I[LacI bound to operator]
    H -->|Yes| J[LacI releases operator]
    E --> K[Weak basal transcription]
    I --> K
    F --> L{Promoter accessible}
    J --> L
    L -->|Yes| M[High transcription]
    L -->|No| K

The key point is that both inputs must be favorable. High cAMP without inducer leaves LacI on the operator. Inducer without high cAMP removes the repressor but leaves a weak promoter. Only the combination produces strong transcription.

Worked regulation example

Consider a culture of E. coli growing in minimal medium with 0.2 percent glycerol as the carbon source, plus 1 mM IPTG. Glycerol does not repress cAMP, so cAMP is high, CAP is active, and the CAP site is occupied. IPTG enters the cells, binds LacI, and causes the repressor to release the operator. With CAP bound and the operator free, RNA polymerase initiates at a high rate. The culture expresses lacZ, lacY, and lacA.

Now switch the carbon source to 0.2 percent glucose and keep the IPTG. Glucose transport lowers cAMP, CAP loses its cAMP, and the activator leaves the CAP site. LacI is still inhibited by IPTG, so the operator is free, but the promoter is now in its weak basal state. Transcription drops sharply. The cell has effectively decided that lactose metabolism is not worth the investment while glucose is available.

Now remove the IPTG but keep the glycerol. cAMP stays high and CAP stays bound to the CAP site. LacI is not inhibited, so it occupies O1, loops in O2 or O3, and blocks polymerase. Transcription is repressed. The CAP site being occupied is not enough. Repression dominates when the operator is bound.

The final case is the fully induced state: low glucose and IPTG present. This is the configuration used in almost every recombinant protein expression experiment. It is also the configuration that produces the highest expression the wild-type promoter can deliver, which is still modest compared with what a phage promoter can do.

Why the lac promoter is weak even when fully induced

The fully induced lac promoter is not a strong promoter. Its -35 and -10 elements deviate from consensus, its spacer is not optimal, and its activator contact is real but limited. The measured output is sufficient for the cell to grow on lactose, but it is far below what a T7 promoter or a tac promoter can produce.

This is why expression vectors for high-level protein production rarely use the wild-type lac promoter as the main driver. The pET system uses a T7 promoter, which is recognized by T7 RNA polymerase and is far stronger than any sigma-70 promoter. The tac and trc promoters are hybrids that combine the -35 element of the trp promoter with the -10 element of the lac promoter, and they retain the lac operator so that LacI can still regulate them. These hybrid promoters are stronger than the parental lac promoter and are still IPTG-inducible.

The trade-off is not free. Stronger promoters are harder to keep fully off. The pET system addresses this with a double-repression strategy: the T7 promoter is regulated by a lac operator placed downstream, and the host supplies LacI, so the promoter is repressed until IPTG is added [9]. Without the lac operator, basal expression from a T7 promoter can be high enough to cause plasmid instability or cell toxicity, especially for proteins that are harmful to the host [9].

Medium composition can also undermine repression. Plant-derived peptones and other complex medium components can carry inducers that spontaneously derepress T7-lac promoters, which is a real problem for toxic proteins [10]. The same phenomenon explains why some auto-induction protocols work without added IPTG. In one study, cellobiose 2-epimerase was expressed from a T7-lac promoter in TB medium with no IPTG and no lactose, and the activity reached 5.88 U/mL, which was 3.70 fold higher than with 1.0 mM IPTG [11]. When the same strain was grown in M9 medium without yeast extract or tryptone, no expression occurred without inducer, which confirms that the complex medium was supplying an inducer [11]. The practical lesson is that "uninduced" does not always mean uninduced.

How the lac promoter is studied in practice

The standard readout is a transcriptional or translational reporter. A promoter fragment is cloned upstream of a fluorescent protein such as GFP, and fluorescence is measured in bulk or at single-cell resolution. Single-cell measurements are important because induction is often heterogeneous. Bimodal induction with IPTG has been documented at roughly ten-fold lower concentrations than with TMG [8].

A second approach is in vivo single-RNA imaging. By tagging the mRNA with a fluorescent aptamer and imaging individual molecules, researchers can measure the intervals between consecutive transcription events. In a lac promoter variant controlled by both IPTG and arabinose, the kinetics of closed and open complex formation were independently controlled by the two inducers, and transcript production was sub-Poissonian under all conditions tested [12]. That means the promoter fires more regularly than a simple Poisson process would predict, which has implications for how cells manage noise in gene expression.

A third approach is mutational scanning. A 75 base pair region covering the lac promoter has been mutagenized and each variant's induced transcriptional activity measured with a fluorescent reporter. Additive contributions of individual mutations explain about two-thirds of the explainable phenotype variance, pairwise epistasis explains about 7 percent of the variance across the full sequence and about 15 percent within the protein binding sites, and there is no evidence for third-order epistatic contributions [13]. The inferred fitness landscape is essentially single-peaked with a small amount of antagonistic epistasis, and the wild-type sequence appears to be multi-objective optimal for expression in different nutrient environments [13].

Comparative relevance

The lac promoter is not just a bacterial curiosity. The lac operator and LacI have been ported into mammalian cells, fission yeast, and other organisms to build inducible expression systems. In Schizosaccharomyces pombe, a chimeric nmt::lacO promoter can be regulated by LacI up to two orders of magnitude in response to IPTG, with near-full induction about 40 minutes after addition [14]. In mouse hematopoietic cells, a lacO-modified Vav promoter conferred limited repression and partial reversion after IPTG, which illustrates that the system does not always transplant cleanly [15]. In Corynebacterium glutamicum, photocaged IPTG has been used to control a tac-based expression module with light, giving tight and homogeneous induction upon short UV-A exposure [16].

Optogenetic control of the lac operon is an active area. The OptoLAC circuits use blue light to regulate IPTG-inducible promoters, and in metabolic engineering applications they improved mevalonate and isobutanol production by 24 percent and 27 percent respectively compared with IPTG induction, in light-controlled fermentations scalable to at least two-liter bioreactors [17]. These systems matter because light is tunable, reversible, and spatially controllable in ways that a chemical inducer is not.

Common Mistakes and Limitations

Treating the lac promoter as a strong promoter. It is weak by design. If you need high-level protein production, use a T7 or tac promoter. The wild-type lac promoter will not deliver the yield.

Assuming IPTG is the only inducer. Lactose, allolactose, TMG, and even components of complex media can induce. Auto-induction works because of this. It also means "uninduced" controls can be misleading.

Forgetting that IPTG is not metabolized. This is usually an advantage because the concentration stays constant. It also means IPTG cannot be cleared by the cell, so washout requires dilution or centrifugation.

Ignoring LacY and LacA. IPTG uptake at low concentrations depends on LacY [8], and LacA activity can lower the effective IPTG concentration over time [8]. Both affect induction kinetics in ways that a simple repressor model does not capture.

Assuming the operator is a single site. Three operators cooperate, and the pseudo-operators contribute substantially to repression. Removing O2 and O3 together reduces repression more than 50 fold [5].

Assuming CAP and LacI are independent. They are not. The ternary complex is 4 to 11 fold more stable than independent binding predicts [6]. This matters for quantitative models of the system.

Overlooking medium effects. Plant-derived peptones and yeast extracts can carry inducers that derepress T7-lac promoters, which is a serious problem for toxic proteins [10].

Expecting a simple on-off switch. The lac operon can show bistable, all-or-nothing behavior at low inducer concentrations, and the growth rate itself feeds back on the system dynamics [7].

Quick Review

1. The lac promoter is a weak sigma-70 promoter with a TTTACA -35 element and a TATGTT -10 element, both deviating from consensus.
2. CAP-cAMP binds an upstream site near -61.5 and activates transcription by contacting RNA polymerase directly, mainly by accelerating open complex formation.
3. The main operator O1 overlaps the promoter, and LacI binding blocks polymerase.
4. O2 and O3 cooperate with O1 through DNA looping, and their loss reduces repression more than 50 fold.
5. IPTG is a gratuitous inducer that binds LacI but is not metabolized, so its concentration stays constant.
6. Both inputs are required: low glucose raises cAMP and activates CAP, and inducer removes LacI.
7. The fully induced lac promoter is still weak, which is why T7 and tac promoters dominate overexpression work.

Frequently Asked Questions

What is the difference between the lac promoter and the lac operator?

The promoter is where RNA polymerase binds to start transcription. The operator is where LacI binds to block transcription. In the lac operon they overlap, which is why repressor binding physically prevents polymerase from working.

Why is the lac promoter considered weak?

Its -35 and -10 elements deviate from the sigma-70 consensus and its spacer is not optimal, so RNA polymerase binds it poorly without help. CAP-cAMP provides that help, but even the activated promoter is far weaker than a T7 or tac promoter.

What does IPTG do that lactose does not?

IPTG binds LacI and relieves repression but is not metabolized by LacZ, so its concentration stays constant. Lactose is consumed and its inducing species, allolactose, is produced only after LacZ is expressed.

Does glucose affect the lac promoter directly?

No. Glucose lowers cAMP, which reduces CAP-cAMP and therefore reduces activation of the promoter. The effect is indirect and is called catabolite repression.

Why do some cultures express protein without IPTG?

Complex media such as TB, yeast extract, and plant-derived peptones can contain inducer-like compounds that derepress lac-based promoters. In one study, expression from a T7-lac promoter in TB medium without IPTG was 3.70 fold higher than with 1.0 mM IPTG [11].

Can the lac promoter be used in organisms other than E. coli?

Yes. LacI and lacO have been ported into fission yeast, mammalian cells, and Corynebacterium glutamicum, with varying success. Induction strength and leakiness depend on the host and the promoter context [14][15][16].

Related Articles

• Lac Operon: Structure, Function, and Regulation in E. coli
• Operon Promoter: Structure, Function, and Regulation in Prokaryotes
• Operon Example: The Lac Operon and Gene Regulation
• Feline Diabetes Mellitus: Remission Induction and Monitoring
• Promoter bashing protocol
• Promoter Sequence: Definition, Function, and Examples

Sources

1. The -45 region of the Escherichia coli lac promoter: CAP-dependent and CAP-independent transcription.
2. Substitution of 2 base pairs (1 base pair per DNA half-site) within the Escherichia coli lac promoter DNA site for catabolite gene activator protein places the lac promoter in the FNR regulon.
3. Identification of the activating region of catabolite gene activator protein (CAP): isolation and characterization of mutants of CAP specifically defective in transcription activation.
4. Synergy between Escherichia coli CAP protein and RNA polymerase in the lac promoter open complex.
5. The three operators of the lac operon cooperate in repression.
6. Co-operative interactions between the catabolite gene activator protein and the lac repressor at the lactose promoter.
7. Bistable behavior in a model of the lac operon in Escherichia coli with variable growth rate.
8. lac operon induction in Escherichia coli: Systematic comparison of IPTG and TMG induction and influence of the transacetylase LacA.
9. High-level expression of Bacillus naganoensis pullulanase from recombinant Escherichia coli with auto-induction: effect of lac operator.
10. T7-lac promoter vectors spontaneous derepression caused by plant-derived growth media may lead to serious expression problems: a systematic evaluation.
11. Efficient heterologous expression of cellobiose 2-epimerase gene in Escherichia coli under the control of T7 lac promoter without addition of IPTG and lactose.
12. Regulation of mean and noise of the in vivo kinetics of transcription under the control of the lac/ara-1 promoter.
13. Genotype to phenotype mapping and the fitness landscape of the E. coli lac promoter.
14. An IPTG-inducible derivative of the fission yeast nmt promoter.
15. Generation and evaluation of an IPTG-regulated version of Vav-gene promoter for mouse transgenesis.
16. Light-Controlled Cell Factories: Employing Photocaged Isopropyl-β-d-Thiogalactopyranoside for Light-Mediated Optimization of lac Promoter-Based Gene Expression and (+)-Valencene Biosynthesis in Corynebacterium glutamicum.
17. Optogenetic control of the lac operon for bacterial chemical and protein production.