NYMC Student Theses and Dissertations

Date of Award


Document Type

Doctoral Dissertation - Restricted (NYMC/Touro only) Access

Degree Name

Doctor of Philosophy


Biochemistry and Molecular Biology

First Advisor

Marietta Y.M.T. Lee

Second Advisor

Ernest Y.C. Lee


There are two forms of human Pol δ - Pol δ4, a heterotetramer, and Pol δ3, which lacks the p12 subunit. The Pol δ3 form is generated by regulated degradation in response to DNA damage, and also during the S-phase. A number of studies on their kinetics and proofreading capabilities suggest that Pol δ3 is intrinsically more accurate than Pol δ4. However, whether DNA synthesized by these two enzymes exhibit a difference in error frequency has not been shown. In this study, I use a forward mutation assay to determine error frequencies of Pol δ4 and Pol δ3. This assay uses a gapped plasmid that is extended by the polymerase of interest, followed by a genetic screen based on lacZα complementation to detect the plasmids containing a mutation. My results demonstrate that Pol δ4 exhibits a greater error frequency than Pol δ3. PDIP46 acts as an accessory protein that stimulates Pol δ4 processivity by an order of magnitude on singly-primed M13 template but has no effect on Pol δ3. I tested the hypothesis that PDIP46 modulates the fidelity of Pol δ4 by detailed analysis in forward mutation assays. Measurement of error rates in highly accurate polymerases is difficult as these are very close to baseline values. It is a standard protocol to use exo deficient mutants which removes the proofreading function and thereby increases the error rates. This tactic leads to increased error rates which allow statistical analysis and has added value that this also permits determination of the error rate of the polymerization reaction per se (in isolation from the combined actions of polymerase/exo). My results show that PDIP46 reduces the error rate of exo-deficient Pol δ4 by ca. 3x. Analysis of the mutation spectrum revealed that both transition and transversion substitutions are effectively suppressed. Furthermore, PDIP46 also suppresses the deletion errors. The effect of PDIP46 was confirmed by increasing the concentration of each one of the four dNTPs in turn to increase mis-incorporation events. In all cases, PDIP46 suppressed the error frequency. To verify that this is due to the action of PDIP46, I used the PDIP46-5A mutant which does not interact with either PCNA or Pol δ. My results reveal that PDIP46, in addition to its action on Pol δ4 activity, also increases its fidelity, and brings this closer to that of Pol δ3. The ability of PDIP46 to improve the fidelity of Pol δ4 could answer the the paradox that Pol δ4 and Pol δ3 have significantly different error rates.