Cis and trans-acting elements in somatic hypermutation

Somatic hypermutation in the Ig genes is a paradigm of a
site-specific, stage-specific, lineage-specific "mutator system"
that generates point mutations at high rates at the Ig locus in an
AID mediated and transcription dependent manner during affinity
maturation of activated B-cells. The questions in the field of
hypermutation today are (i) what targets the mutator factors to the
Ig genes? And (ii) what are these factors? To address some aspects
of these questions, I engineered and compared retroviral vectors to
monitor hypermutation. Retroviral vectors combine a high
transduction rate with integration at random sites within the host
cell genome, thus equalizing positional effects on the reporter
gene. The vectors contain a reporter gene with a premature TAG
termination codon; upon reversion, a full-length fluorescent
protein is expressed. The number of fluorescent cells can be easily
measured in flow cytometry, and thus mutation frequencies can be
determined with accuracy. I tested the green and yellow
fluorescence proteins (GFP and YFP); and various proteins with red
fluorescence (dsRed). Furthermore, I tested various selection
markers to select for transduced cells. I found that GFP as a
reporter, combined with a drug selection marker, gave the most
consistent and convenient mutation rate measurements. DsRed is a
good alternative to GFP, but variants with greater fluorescence
intensity are needed when combined with green fluorescence
measurements. To study the activity of enhancers on transcription
and hypermutation, I introduced various cis-acting enhancer
elements into the reporter construct. Using such constructs I found
that no immunoglobulin specific sequence is needed to target
hypermutation. Also, depending on their position in these
ectopically expressed constructs and on the transcription rate,
enhancers can have positive or negative effects on hypermutation. I
also applied the indicator system to investigate whether or not
mismatch repair influences AID mediated hypermutation in a non-B
lymphocyte line. To do so MLH 1 expression, which is essential in
mismatch repair, was regulated in a non-lymphocyte cell line that
had been transduced by an AID containing vector. Whether or not
MLH1 was expressed, we found no difference in mutation rates of an
indicator gene. This is in contrast to activated B cells, where the
absence of mismatch repair results in a reduction of hypermutation.
I, therefore, conclude that in order to contribute to
hypermutation, mismatch repair needs additional factors that are
present in activated B lymphocytes but absent in the cell line
investigated. During my cloning efforts to create DsRed containing
indicator vectors, I found that a single base substitution that
does not change the amino acid sequence in the gene encoding DsRed2
resulted in an increase in relative fluorescence intensity of the
protein. Surprisingly, the mutated codon is slightly less favored
than the original one and, therefore, would be expected to have a
negative effect on fluorescence intensity, if any at all. I suggest
that in highly expressed genes, the tRNA with the anticodon
corresponding to a commonly used codon might become depleted. As a
consequence, overall less protein is synthesized, but mRNAs using
less frequent codons at some positions may be translated more
efficiently.