Transcription-Coupled DNA Repair in Prokaryotes - Progress in Molecular Biology - page 33

Signature I Signature II

100 200 300 400 500 600 700 800 900

UvrB binding DNA binding

ATP-binding I ATP-binding I ATP-binding II

F IG . 3. Schematic linear representation of the UvrA protein (adapted from Ref. 52 ). Its
domains are colored: ATP-binding I, red; signature I, pink; ATP-binding II, dark blue; signature
II, green; UvrB-binding, yellow; insertion domain (labeled DNA binding), purple; and linker, black.
Inactivation of either one of the two ATP-binding domains causes the loss of ATPase activity in vitro
and UV sensitivity in vivo . Numbers below the bars refer to amino acids.

in eukaryotes, including human cells, mutants lacking TCR but not GGR or
GGR but not TCR have been identified. For example, cells from XPC patients
show TCR but not GGR, while cells from patients with Cockayne syndrome
(CSA or CSB) lack TCR but retain the capacity for GGR. Studies of cells
deficient in the CSB (ERCC6) protein have led to the idea that this protein and
Mfd play similar roles in TCR; however, the interactions of CSB with the other
components of TCR in mammalian cells are more complex and less clearly
understood than those of Mfd .
The ability of UvrA to interact with UvrB is required for TCR as well as for
GGR, and the interaction has been analyzed in detail. Specific mutations in
UvrA (e.g., R176, R206, and D219 but not E185) or in UvrB (e.g., E215 and
R223 but not E222) prevent the interaction and thus inhibit repair. 47,52,54

VI. The Role of UvrB in TCR

Like UvrA , UvrB ( Fig. 4 ) is required for both GGR and TCR. Because the
principal difference between TCR and GGR is in the initial steps of damage
recognition, once UvrA has been recruited and Mfd released from the vicinity
of a lesion, the transfer of UvrB to DNA and the subsequent incision and
resynthesis are thought to involve the same biochemical steps as GGR. The
UvrAB* complex recognizes damage in GGR, and when the UvrA protein
dissociates, UvrB must be positioned to allow the UvrC endonuclease to bind
and incise the damaged strand, not the undamaged strand. Assuming the same
is true for TCR, one might expect that, if there were a particular kind of
damage that only arrested the RNA polymerase when the lesion was in the
nontranscribed strand, the repair of the nontranscribed strand would be more
rapid than the repair of the transcribed strand. Although Mfd would recruit