December 2008 technical highlight
Speedy cloning and mutagenesis
A ligase-independent approach to cloning rapidly produces large numbers of constructs for expression testing and crystallization trials.
Schematic of the PIPE cloning method. The thin, black lines represent template DNA. The thick, black lines with dashed or dotted ends represent the primers with 5′ complementary extensions. The black square dashes represent sequences complementary to each other as do the black dots. The full dark gray lines represent complete strand synthesis. The dashed dark gray lines represent heterogeneous primer extension resulting from PIPE (PIPE occurs all across a PCR template and up to the very end of each primer. This is not shown in its entirety here).
For some proteins, obtaining crystals suitable for structural studies requires the investigator to generate multiple mutants and truncations in the hope that one of them will produce something useful. Techniques such as partial proteolysis or deuterium-exchange mass spectrometry can help to define useful constructs, but without known structural homology, many rounds of cloning and expression may be needed.
To speed up this step, the PSI JCSG uses a fast and efficient method for creating clones and for screening the protein products to predict crystallization success. This approach uses polymerase incomplete primer extension (PIPE), which does not rely on any particular sequence or strain, and has helped the center tackle proteins from its PSI target list.
In a nutshell, it uses unpurified PCR products plus a vector and directly transforms them into bacteria to create clones without any extra manipulations. Normal PCRs generate a range of partially single-stranded DNA fragments. Through appropriate primer design, both insert and vector amplifications result in compatible overhanging ends. The insert and vector amplifiers are simply mixed and transformed to yield the clone of interest.
Using this technique, the team generated full-length expression clones for 448 PSI targets in about a week. Of these, 114 full-length targets produced soluble protein, resulting in nine structures from this initial target set. Of the other 105 soluble proteins, 96 were selected for construct optimization via N- and C-terminal deletions using primers specifying the truncation point. Within 2 weeks using the PIPE method, 2143 deletions were generated. These nested truncations were evaluated for expression of soluble proteins and characterization of their aggregation properties. Over 60% of the targets showed improved behavior after the terminal truncation series.
Another step towards improving crystallization rates that JCSG has taken is to use microscreening to predict whether a target will produce crystals. Expression constructs are screened in 96 deep-well microtiter blocks for soluble expression and undergo limited biophysical testing for yield, purity, integrity and aggregation. Many soluble expression constructs produce aggregated protein, which is not usually suitable for structural studies, and so they use analytical size-exclusion chromatography, at the microscreening stage, to ascertain the distribution of the protein in solution. Previous studies indicated that monodispersed proteins are most likely to crystallize.
Together these methods contribute to the high throughput needed for structural genomics.