Academic journal article Genetics

High-Resolution Specificity from DNA Sequencing Highlights Alternative Modes of Lac Repressor Binding

Academic journal article Genetics

High-Resolution Specificity from DNA Sequencing Highlights Alternative Modes of Lac Repressor Binding

Article excerpt

(ProQuest: ... denotes formulae omitted.)

THE lactose regulatory system established the paradigm of a trans-acting factor binding to a cis-acting element to regulate the expression of the adjacent gene in response to an environmental signal (Jacob and Monod 1961). Many aspects of the lac repressor protein have been studied extensively (reviewed in Lewis 2005). Our primary interest is in the DNA binding specificity of the lac repressor. Measurements of affinity changes due to operator sequence variation, by base replacement, by the use of base analogs, or by changing the length of the operator, have been performed almost since the operator sequence was first determined (Goeddel et al. 1978; Sadler et al. 1983; Betz et al. 1986; Sartorius et al. 1989; Lehming et al. 1990; Sasmor and Betz 1990; Frank et al. 1997; Spronk et al. 1999; Falcon and Matthews 2001; Kalodimos et al. 2002, 2004b; Daber and Lewis 2009). But those analyses all measured binding affinity to only a few sequences.

The lac repressor has not, to our knowledge, been analyzed by current high-throughput methods that can determine specificity over thousands, or even millions, of sequences in parallel (Stormo and Zhao 2010), such as protein-binding microarrays (PBM) (Berger et al. 2006; Gordan et al. 2013), SELEX-seq [or HT-SELEX (Zhao et al. 2009; Zykovich et al. 2009; Jolma et al. 2010; Wong et al. 2011)], bacterial one-hybrid (BlH) (Meng et al. 2005; Noyes et al. 2008; Christensen et al. 2011), and mechanically induced trapping of molecular interactions (MITOMI) (Maerkl and Quake 2007). While those methods offer an expansive overview of binding specificity, the accuracies of the resulting models are highly variable. This can be due to low-quality data, but even with highly reproducible experiments specificity determination requires complex computational modeling (Stormo 2013; Weirauch et al. 2013; Orenstein and Shamir 2014) and the results can vary considerably, depending on the method used and the protein being analyzed, including their ability to predict in vivo binding site locations (see Weirauch et al. 2013 for a comparison of many different approaches to the analysis of PBM data for many different transcription factors).

Here we introduce a new high-throughput method, Specseq, that directly measures specificity by sequencing. It has several advantages over existing methods for quantifying large collections (thousands) of binding site energies in one experiment. Using lac repressor as an example, we show that this method has excellent reproducibility, giving energy measurements generally consistent within ~0.1 kT (k is the Boltzmann constant and T is temperature in degrees kelvin; at 0 degrees C, 1 kT = 0.54 kcal/mol). While not measuring in parallel as many different sequences as some of the higher-throughput methods, we obtain high accuracies because we measure exactly what is necessary, the relative affinity to a large collection of sequences, without any mathematical fitting procedures or approximations required. It is similar to MITOMI in the number of different sites that can be analyzed in parallel, but it is much simpler to perform, requiring only a means to separate bound and unbound sequences, which are then sequenced using high-throughput, short-read sequencing machines.

When applied to the lac repressor we obtain, in a single experiment, data covering a large fraction of all the previous studies, plus thousands of additional variants, allowing us to compile a much more comprehensive profile of its specificity. We confirm that the lac repressor can bind to sites of different lengths but that the preferred sequence, and the mode of binding, depends on the length. We also applied Spec-seq to two other members of the LacI/GalR protein family, PurR and YcjW, to obtain extensive models of their specificity and test the generality of the lac repressor's ability to bind to operators of variable length. We find that the lac repressor is apparently unique in its ability to bind with high affinity to sites of different lengths and with different modes of binding. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.