Library Systems

Retroviral cDNA and Peptide Libraries

The power of complementation cloning, long appreciated in bacterial and yeast genetic systems, may now be more fully accessed for mammalian cells by applying retroviral delivery systems.

The basic tenet of the approach is that using retroviruses large libraries of encoded objects (cDNAs, peptides, RNAs)-- among them desired objects capable of exerting a dominant phenotype on target cells-- can be transferred to target mammalian cells. By applying appropriate genetic selections one can isolate those cells expressing the object of interest from among a background of unwanted objects.

We have applied this in two main areas:

cDNA libraries. In this case a cDNA library is made in a retroviral backbone from a source cell type. The cDNA library (generally around 5 x 10^6 independent full-length inserts) is converted into packaged retrovirus and used to infect a target cell. After infection of the target cell a suitable genetic selection is set that allows for one to distinguish those cDNAs capable of conferring a new phenotype on the target cell. We have used this approach to isolate cDNAs that block a variety of apoptotic stimuli as well as several surface markers. Published papers on this work can be found in Kitamura et al and Hitoshi et al, among others in the Publications section.
Peptide libraries. A library of short peptides, termed DOMINANT EFFECTORS, are encoded into the retroviral backbone. The peptides can be considered "shapes" in the pharmaceutical sense of the word (see the mini-tutorial link at the right for more info). An expectation is that rare peptides will be capable of interfering with intracellular signaling systems in interesting and novel ways. A further prediction is that the selected peptides will be useful for identifying the signaling molecules upon which they act (target identification). This can lead to reagents that can be used to understand function of biological pathways and discern new drug candidates. A publication of our work in this area can be found in the following Nature Genetics article Xu et al "Dominant Effector Genetics in Mammalian Cells". A News and Niews on the work can be found at this link. We have also used this approach to create novel inhibitors of T cell pathways and HIV-1 replication (Shigemi Kinoshita and GPN, unpublished). Background articles can be also found in Lorens JB, et al and basic library principles in the cDNA publication of Kitamura T, et al Efficient screening of retroviral cDNA expression libraries (PNAS).

Other applications of libraries include:

Mutant cDNA libraries in which PCR or other mutagenesis is used to mutate a known cDNA and a library selection system is set to find those mutant cDNAs that survive the selection. This can be used to find, in a non-biased and interesting manner, new sites on proteins that are not readily found using standard mutagenic approaches.

cDNA fragment libraries. Pioneered by Igor Roninson at the University of Chicago this approach uses randomly fragmented cDNA expressed in retroviral backbones. The approach allows for "mini-domain" interference in signaling systems with the notion that the fragments' ability to interfere, and its constitution, allows for biological inferences to be made.

Ribozyme Libraries. A variety of ribozymes have been encoded in retroviruses. Some individuals have suggested that by randomizing the targeting sequence, and setting appropriate selection criteria, one can isolate ribozymes with specific properties that are capable of cleaving "unknown" mRNAs. Of course, the inference is that once the ribozyme is sequenced one will be able to soon discover the nature of the target RNA and then determine how the cleavage event determines the selected event.

Other applications of libraries include:

Related links: