Technical Comments

Comment on "Uracil DNA Glycosylase Activity Is Dispensable for Immunoglobulin Class Switch"

See allHide authors and affiliations

Science  17 Dec 2004:
Vol. 306, Issue 5704, pp. 2042
DOI: 10.1126/science.1104396

The uracil base-excision repair pathway has been strongly implicated in the process of immunoglobulin (Ig) heavy chain class switch recombination (CSR) (14). This process likely involves deamination of cytosine residues of the Ig switch region by activation-induced cytidine deaminase (AID), removal of the resulting uracil base by uracil DNA glycosylase (UNG), and cleavage of the DNA backbone at the product abasic site by an endonuclease (APE1) to generate double-strand DNA breaks (DSBs), the key recombinogenic intermediate in CSR (5). The findings in the recent report by Begum et al. (6) cast doubt on this mechanism by suggesting that the catalytic activity of the DNA repair enzyme UNG is not required for the formation of DSBs. Instead, the authors suggest that UNG plays a structural role in downstream events following the formation of DSBs. However, the experiments they reported fall short of eliminating the powerful UNG activity in the formation of DSBs.

Begum et al.(6) argue that UNG catalytic activity cannot be involved in the formation of DSBs because, when they ectopically expressed three catalytically crippled forms of UNG in UNG–/– B cells, DSBs and overall CSR were observed [UNG–/– B cells have previously been found deficient in the formation of DSBs and CSR (7).]. However, the three UNG single mutants—D145N, N204V, and H268L—are still very powerful catalysts and can excise uracil from duplex DNA with a half-life of about 1 min (8, 9). This residual activity may be sufficient to give rise to DSBs at a rate comparable to, or faster than, the subsequent steps in CSR.

Begum et al. also showed that two double mutants (D145N:N204V and H268L:D145N) that are at least an order of magnitude more catalytically deficient than the single mutants were unable to restore CSR. To reconcile this contradictory result with that of the single mutants, the authors attributed it to “a structural requirement for an unknown function of UNG.” However, structural studies of the H268Q and D145N single mutants provide no evidence that structural perturbations arise from these mutations, and it would be unexpected if the double mutations differed (10, 11). Structural studies clearly place the His268, Asn204, and Asp145 catalytic groups deep within the active site pocket of UNG, such that it is unlikely that they have any other role except in the catalytic process of uracil recognition and excision. A key experimental omission in (6) is to determine whether the double mutants retain the ability to form DSBs. If Begum et al. are correct that another enzymatic activity gives rise to the DSBs, then the double mutants should show DSBs, as should the parent UNG–/– B cells. In contrast, DSBs should not be observed in either case if UNG activity is required.

In an orthogonal approach to establish the mechanism, Begum et al. demonstrated that DSBs still occurred in B cells that express the potent uracil DNA glycosylase inhibitor protein (Ugi) (12) but that the overall process of CSR was abolished. This result was interpreted to provide additional evidence that another enzymatic activity was responsible for the DSBs and, once again, that UNG might play a structural role in CSR after DSB formation. However, before this conclusion can be solidified, it is important to determine how tightly Ugi interacts with the full-length human UNG, which is currently unknown (13). In lieu of this measurement, Ugi has been shown to have markedly weaker affinity for the catalytic domain of human UNG as compared with bacterial enzymes (12, 13), and it has no detectable affinity for a pox-virus UNG that is structurally indistinguishable from the human and bacterial UNGs (13, 14). In vitro studies indicate that when Ugi is present in a 10-fold molar excess over human UNG, about 2% residual activity remains. This is similar to the activity seen in UNG single mutants (13). Another concern is that human UNG is an abundant enzyme and is present at concentrations as high as 7 μM in some cell types (15, 16). Thus, a simple explanation for the Ugi effects follows along similar lines as described above for the single UNG mutations: Sufficient UNG activity remains to promote DSBs, but the bulky Ugi protein blocks weaker protein interactions that may be important for later steps in CSR. Until these alternative explanations can be dismissed by further experimentation, the catalytic role of UNG in CSR remains unclear.

References and Notes

Navigate This Article