4 LacI Biochem and Genetics2.docxPage 1 of 2

LacI Biochemistry

Read Lac Book 1.9 thru 1.12

•These sections discuss biochemistry and some very early molecular biology that was used to study LacI and its interaction with lacO. Also discussed are the methods used to purifiy a lot of LacI and what was done with it.

Lac Book 1.13--LacI is modular

•Lac region was thought to go: pI opZYA and Julian Davies was generating lacI deletions that deleted lacI and lacO. These should have been I- and Oc. (Note that he probably didn't find many of these because the real order is: pI poZYA, and deleting I and O would also have removed the promoter and given a Lac- phenotype).

During this screen he found some lacI- deletions that were trans-dominant: I-d. That is they could stop LacI+ from working.

•Ultimately, 187 I-d mutants were isolated and mapped using Jeff Miller’s deletion set (see next page).

-most bound IPTG, therefore their inducer binding domain was functional.

•I-d were thought to be dominant becaue LacI normally works as a tetramer. The "bad" LacI would poison the tetramer.

Question: If I-d LacI and w.t. LacI are present in equal amounts, what percentage of tetramers will be functional -- that is, not include a I-d protein monomer? What if the I-d to wt LacI ratio is 10 to 1?

The lacI-d mutants were mapped to the lacI gene using Jeffery Miller's lacI deletion set.

•Jeffery Miller's deletion set:

•He used a special strain of E. coli where the original lac operon was deleted and a copy inserted near tonB (Schmeissner_JMB_1977):

•He selected for phage and colicin resistance (tonB-) and Lac constitutive (lacI-) mutations. Most of these were deletions that took out tonB and at least part of lacI:

Using such deletions, it is possible to finely order nonsense or missense mutations. (SLIDES)

•All I-d mutations fell in the 1st 7 deletions of Miller's set (i.e. in the first ~50 amino acids of LacI). This operator binding is therefore about 50 a.a. long and at the N-term.

•Trypsin cuts LacI at residue 51 and/or 59 (lys or arg). The small piece can bind operator DNA.

•Do a SMART database search to introduce the database and see how it works

•Finally IPTG-binding mutations (i.e. lacIs) were shown to be downsteam of the lacI-d mutations. This meant that the LacI protein had two separate parts with different jobs: DNA binding and IPTG binding.

SLIDES of LacI structure:

Lac Book 1.14 Miller's Analysis.

•Miller in a monumental series of ~15 papers correlated the genetic and physical maps of the lacI gene.

What does this mean? He showed for each of >5000 nonsense mutations where, each mutation was, and what the phenotype was (I-, Is, I+) when the mutations were suppressed by 5 nonsense suppressors that put in different amino acids at the nonsense site.

•Amber: UAG, Ochre UAA, Opal (UGA).

•There are 90-100 sites in lacI where a single base change can generate a nonsense mutation.

•Amber and Ocher suppressors insert leu, typ, glu, ser or lys

•UGA suppressors insert trp

•To find the location of nonsense mutations, he needed to have well characterized mutagens

mutT: A=>C Aminopurnine G=>A, A=>G

Strategy:

1. Isolate many nonsense muations using mutagens of known specificity

2. Separate mutants into groups based on their map sites (deletion mapping) and suppression patters (are they Amber or Ochre etc). End up with something like "Amber near the end of 138, caused by mutT which means that it came from a A=>C mutation.

3. Correlate with an amino acid based on seq. of LacI protein and mutagen specificity. EG serine near the end of 138 went from UCG to UAG giving an amber mutation.

Mutations enhanced by mutagens

Go over slides

amberX/F' amberY---> how frequent wt recombinants occur is proportional to distance separating the ambers:

Top lacI- constitutives

Middle lacI(ts)

Bottom lacI superrepressors

Note that lacIS mutations are never in the domain that encodes the DNA-binding HTH

Lac book 1.5 cAMP and CRP(aka CAP)

•Makman and Sutherland measured cAMP levels in E. coli. It appears that they did this because cAMP was known to act as a second messenger in multicellular organisms, where it conveys information to cells about the presense of different hormones. (Sutherland had done some of this work in eukaryotes)

•They found that cAMP levels could vary ~1000 fold from 10-7 to 10-3 M depending on the conditions the cells were incubated. The presence of glucose caused cAMP levels to drop, and when cells ran out of glucose during growth, cAMP levels spiked.

•Agnes Ullman (a-complementation) working with Monod, and Ira Pastan recognized that cAMP and its relation to glucose may have had something to do with diauxic growth (and adaption) where glucose was used first and kept other carbon sources from being utilized.

From Ullman and Monod FEBS Letters 1968:

•Perlman and Pastan took a genetic approach and isolated mutants that were unable to make cAMP (Perlman_BBRC_1969)

•They screened on lactose+galatose +TTC for colonies (red) that failed to ferment either sugar. Idea was that a cMAP deficient strain would not be able to ferment either.

•They found several which could not ferment lactose or galactose unless cAMP was added to the plates. They studied on of these- strain 5336 further.

•Table 1 strain 5336 does not make cAMP

•Table 2. Strain 5336 fails to ferment a variety of C-sources unless cAMP is added. Note that MacConkey's is more sensitive that TTC for dectecting fermentation of glucose, frutose and galactose. (Note that if they had used MacConkey's to do the screen, they would never have found Lac- Gal- strains.

Table 3: Strain 5336 makes less b-gal than w.t., but cAMP added to the medium doesn't fix the deficiency completely.

•Mutants that were unable to ferment a variety of carbon sources, and that could not be rescued with cAMP were isolated. Some of these were defective in the protein that binds cAMP (called CAP or Crp). Rates of b-gal synthesis in the presence of inducer were reduced about 50-fold in these strains.

The Crp protein was thought to bind DNA and its binding site was mapped to the lac promoter region. The lac region was then know (thought) to be ordered as such: pI CAP-P-O-ZYA.

•CAP-binding site mutants are Lac-. From these, Lac+ suppressors were obtained and one of these had a lac promoter that no longer needed cAMP/CAP in order to work (it was cAMP independent). This ended up being a famous strain, and was in many papers/experiments: called lacUV5

Lac Book Chapter 1.16 Sequence of the lac operator

•Walter Gilbert and Allan Maxam (who would later go on to develop DNA sequencing) sequenced the operator region of the lac operon.

•The operator region was sized as follows:

Operator region was isolated by:

1. Isolating lamda DNA containing the lac region.

2. Shearing it into small bits by sonication.

3. Binding it with purified LacI

4. Adding the mixture to filters that bound protein, but not DNA. This bound the LacI/lacO complexes.

5. After washing, the filters were treated with IPTG which caused LacI to release the lac DNA.

6. The released fragments were transcribed and the resulting RNAs sequenced to give the following:

•Go over inverted repeat and sequence symmetry

•(Slides showing DNA-binding at inverted repeats)