Slow stochastic transgene repression with properties of a timer

Slow stochastic transgene repression with properties of a timer
Clifford L Wang, Desiree C Yang and Matthias Wabl
Genome Biology 2006, 7:R47 (doi:10.1186/gb-2006-7-6-r47)

Thoughts/Implications/Summary
The ability to identify clonal populations exhibiting predictable repression kinetics is intriguing. This finding has implications in the context of retroviral-mediated gene therapy. In this paper the location of integration was associated with the decay kinetics observed. This phenomenon is most likely cell-type/individual dependent. It may be possible to screen cells during retroviral-mediated gene therapy for the kinetics of transgene repression and identify those cells bearing an integration event at a genomic position allowing for predictable/reproducible expression kinetics and selection of clones with a low probability of tumorigenicity. One could envision a scenario where cells are isolated from an individual, "corrected" by way of retroviral integration of the functional gene in question, clonal populations of cells are identified that exhibit predictable expression/repression kinetics with subsequent use of these cells to treat the individual. If in vitro expression/repression kinetics are exhibited in vivo then one could tailor a treatment plan based on these kinetics. (That is assuming that repression is an issue of concern in individuals undergoing gene therapeutic modalities to treat the disease in question. I am ignorant of the long-term prognoses of individuals undergoing these treatments...) As an aside, one could also monitor these cells for things such as aneuploidy.

Basic Experimental Steps

• Used pre-B cell line 18-81 and created transgenic cell lines that expressed GFP
• Isolated single, infected cells exhibiting fluorescence > 100 relative fluorescent units (RFUs)
• Isolated clones were expanded
• Clonal cultures were not initially selected with antibiotics
• Also, they waited 2 days before isolating fluorescent clones to avoid those clones where GFP inactivation was rapid
• Used Flow cytometry to measure GFP fluorescence to track gene expression of a population on a cell-by-cell basis
• Tracked 93 clones that differed in patterns of GFP expression
• Selected 3 of these clonal populations exhibiting different bi-modal patterns of transgene expression at the population level
• Characterized the dynamics of transgene repression in these three clonal populations.

They found evidence that even though repression of a transgene in an individual cell can appear sporadic or stochastic, when observing these events at a population level a probability distribution arises that can be utilized to predict the pattern of repression over time. For instance, they provide data for clonal populations giving one the ability to sample a subset of cells at a particular point in time allowing for the prediction of the time in culture since it was started by a single cell and, subsequently, the amount of time before the transgene was repressed in the entire population of cells. When they selected a high GFP expressing cell, they observed that transgene repression was independent of previous culture-based events (i.e, time in culture and variation in the number of cell generations per day). The attractor state in non-selecting conditions was that of transgene repression and the pattern of repression for one of the clones fit near perfectly that of first-order decay kinetics. The dynamics of repression were dependent on the location of integration of the viral-based vector within the genome and for the clone that exhibited near perfect first-order decay kinetics the site of integration was at a position in chromosome 9.