Vitamin B12 and Homocysteine Associations with Gait Speed in Older Adults: The Baltimore Longitudinal Study of Aging.

OBJECTIVES: This study aimed to assess the independent associations of serum levels of vitamin B12 and plasma concentrations of homocysteine with gait speed decline. DESIGN, SETTING, PARTICIPANTS: This study utilized longitudinal analysis of participants 50 years or older from The Baltimore Longitudinal Study of Aging, N=774. MEASUREMENTS: Gait speed (m/s) was assessed using the 6-meter usual pace test. Vitamin B12 and homocysteine concentrations were collected using standard clinical protocols. Linear mixed effects regression was stratified by baseline age category (50-69, 70-79, and ≥80 years old). RESULTS: Mean follow-up time for the total study sample was 5.4 ± 2.0 years. No association between vitamin B12 and gait speed decline over the follow-up time for any age group was found. Elevated homocysteine concentrations were associated with decline in gait speed after adjustment for covariates (50-69: ß= -0.005, p=.057; 70-79: ß= -0.013, p<.001, ≥80: ß= -0.007, p=.054). CONCLUSION: Homocysteine and vitamin B12 are inversely related, yet only homocysteine was associated with gait speed decline in this population of healthy older adults. Given these results, future research should be directed towards investigating the relationship in populations with greater variation in vitamin B12 concentrations and other mechanisms influencing homocysteine concentrations.

Longer duration of combination antiretroviral therapy reduces the risk of Hodgkin lymphoma: A cohort study of HIV-infected male veterans.

BACKGROUND: Hodgkin lymphoma (HL) incidence has increased since combined antiretroviral therapy (cART) introduction. It is unclear how different cART classes (e.g., protease inhibitors (PI), non-nucleoside reverse transcription inhibitors (NNRTI)) influence HL. This study aimed to determine the effects of cART duration on HL incidence among HIV-infected veterans. METHODS: We performed a retrospective cohort study utilizing the Veterans Affairs HIV Clinical Case Registry (1985-2010). HL cases were identified using ICD-9 codes (201.4-9). cART, PI, and NNRTI duration was the aggregate number of treatment days delivered. Incidence rates (IR) and rate ratios (IRR) were calculated from Poisson regression models to examine the effects of cART duration on HL. RESULTS: 31,576 cART users contributed 288,736 person-years (PY) and 211 HL cases (IR=7.3/10,000 person-years). HL incidence decreased from 25.1/10,000 PY (95%CI=18.9-33.4) within the first year of cART to 0.6/10,000 PY (95%CI=0.3-1.6) after ≥ 10 years. In multivariable models, each additional year of cART was associated with decreased HL incidence (IRR=0.80; 95%CI=0.75-0.86); similar effects were observed in models assessing HL incidence by PI and NNRTI. CONCLUSION: Our findings indicate long-term cART of any class is associated with decreased HL risk. High HL incidence directly following cART initiation supports a potential immune reconstitution mechanism in HIV-related HL. Further research is needed to evaluate the interaction between early cART, immune reconstitution, and HL.

Accuracy of self-reported weight, height and BMI in US firefighters.

BACKGROUND: Obesity is of increasing concern especially among firefighters. Bias in self-reported body weight, height and body mass index (BMI) has received a great deal of attention given its importance in epidemiological field research on obesity. AIMS: To determine the validity of self-reported weight, height and BMI and identify potential sources of bias in a national sample of US firefighters. METHODS: Self-reported and measured weight and height (and BMI derived from them) were assessed in a national sample of 1001 career male firefighters in the USA and errors in self-reported data were determined. RESULTS: There were 1001 participants. Self-reported weight, height and BMI were significantly correlated with their respective measured counterparts, i.e. measured weight (r = 0.990; P < 0.001), height (r = 0.961; P < 0.001) and BMI (r = 0.976; P < 0.001). The overall mean difference and standard deviation between self-reported weight, height and BMI were 1.3±2.0kg, 0.94±1.9cm and 0.09±0.9kg/m(2), respectively, for male firefighters. BMI-based weight status (P < 0.001) was the most consistent factor associated with bias in self-reported BMI, weight and height, with heavier firefighters more likely to underestimate their weight and overestimate their height, resulting in underestimated BMIs. Therefore, using self-reported BMI would have resulted in overestimating the prevalence of obesity (BMI ≥ 30.0) by 1.8%, but underestimating the prevalence of more serious levels of obesity (Class II and III) by 1.2%. CONCLUSIONS: Self-reported weight and height (and the resulting BMI) were highly correlated with measured values. A primary and consistent source of error in self-reported weight, height and BMI based on those indices was BMI-based weight status.

The analysis of relapse-free survival curves: implications for evaluating intensive systemic adjuvant treatment regimens for breast cancer.

Results of adjuvant dose intensification studies in patients with localised breast cancer have raised questions regarding the clinical usefulness of this treatment strategy. Here, we develop and fit a natural history model for the time to clinical tumour recurrence as a function of the number of involved lymph nodes, and derive plausible predictions of the effects of dose intensification under various conditions. The time to tumour recurrence is assumed to depend on the residual postoperative micrometastatic burden of tumour, the fractional reduction of residual tumour burden (RTB) by treatment, and the rate of regrowth of the RTB to a clinically detectable size. It is assumed that a proportion of micrometastatic tumours are unresponsive to adjuvant chemotherapy even at maximal dose intensity. Data fitted included the San Antonio Cancer Institute (SACI) database of untreated patients, and CALGB #9082, a study comparing a highly intensive and moderately intensity adjuvant regimen in patients with 10+ positive axillary nodes. The proportion of tumours unresponsive to maximally intensive adjuvant treatment is estimated to be 48% (29-67%). The estimated log kill for intermediate-dose therapy from CALGB #9082 was 6.5 logs, compared with 9 logs or greater for high-dose therapy. The model is consistent with a modest but nonnegligible advantage of dose intensification compared with standard therapies in patients with sensitive tumours who have 10+ positive axillary nodes, and suggests that much of this clinical benefit could be achieved using intermediate levels of treatment intensification. The model further suggests that, in patients with fewer than 10 involved axillary nodes, any advantage of treatment intensification over standard therapy would be much reduced, because in patients with smaller tumour burdens of sensitive tumour, a larger proportion of cures achievable with intensified therapy could be achieved as well with standard therapy.

Modeling cellular processes with variational Bayesian cooperative vector quantizer.

Gene expression of a cell is controlled by sophisticated cellular processes. The capability of inferring the states of these cellular processes would provide insight into the mechanism of gene expression control system. In this paper, we propose and investigate the cooperative vector quantizer (CVQ) model for analysis of microarray data. The CVQ model could be capable of decomposing observed microarray data into many different regulatory subprocesses. To make the CVQ analysis tractable we develop and apply variational approximations. Bayesian model selection is employed in the model, so that the optimal number processes is determined purely from observed micro-array data. We test the model and algorithms on two datasets: (1) simulated gene-expression data and (2) real-world yeast cell-cycle microarray data. The results illustrate the ability of the CVQ approach to recover and characterize regulatory gene expression subprocesses, indicating a potential for advanced gene expression data analysis.

Oncology thinking cap: scaffolded use of a simulation to learn clinical trial design.

BACKGROUND: Physicians often are called on to participate in and interpret clinical trials, but their training in this area may not provide them with the inquiry skills that are needed. Simulations have the potential to be a promising tool for helping medical students learn the skills involved in clinical trial design. However, simulations may be complex and require additional scaffolding to support learning. DESCRIPTION: The goal of this study was to teach aspects of cancer clinical trial design through the scaffolded use of a simulation, the Oncology Thinking Cap. The software-based scaffolding provided guidance in designing the trial. Subsequently, the simulation allowed students to run the designed trial, which produces detailed patient histories. This feedback then could be used to redesign the trial. EVALUATION: Twenty-four 4th-year medical students were asked to design a clinical trial in advance, on paper, to test a new anticancer drug. Student groups then designed and simulated running the clinical trial assisted by the software environment. Instructional effectiveness was measured using a pretest-posttest design that included having students (a) write a group research proposal and (b) individually critique a flawed proposal. At the group level (N = 6 groups), students demonstrated a 34% increase in the number of elements of a clinical trial that they included in their research proposals. At the individual level (N = 24), students improved by 48% in their critiques of flawed proposals. CONCLUSIONS: Scaffolding embedded in the simulator is a promising approach to helping students learn about clinical trial design.

Initiation of strand incision at G:T and O(6)-methylguanine:T base mismatches in DNA by human cell extracts.

Extracts of the human glioma cell line A1235 (lacking O(6)-methylguanine-DNA methyltransferase) are known to restore a G:T mismatch to a normal G:C pair in a G:T-containing model (45 bp) DNA substrate. Herein we demonstrate that substitution of G:T with O(6)-methylguanine:T (m6G:T) results in extract-induced intra-strand incision in the DNA at an efficiency comparable to that of complete repair of the G:T-containing substrate, although the m6G:T mispair serves as a poor substrate for later repair steps (e.g. gap filling, as judged by defective DNA repair synthesis). The A1235 extract, when supplemented with ATP and the four normal dNTPs, incises 5' to the mismatched T, as inferred by the generation of a single-stranded 20mer fragment. Unlike its parental (A1235) counterpart, an extract of the alkylation-tolerant derivative cell line A1235-MR4 produces no 20mer fragment, even when thymine-DNA glycosylase (TDG) is added to the reaction mixture. In contrast, the A1235 extract, when augmented with TDG, catalyzes enhanced incision at m6G:T in the 45 bp DNA, yielding 5-10-fold greater 20mer than that of either extract or TDG alone. Interestingly, the absence of m6G:T incision activity in the A1235-MR4 extract is similar to that seen for extracts of several known mismatch repair-deficient cell lines of colon tumor origin. Together these results suggest that derivative A1235-MR4 cells are defective in m6G:T incision activity and that the efficiency of this activity in the parental (A1235) cells may depend on the presence of several ill-defined mismatch repair recognition proteins along with TDG and ATP.

Pharmacokinetic and pharmacodynamic study of the combination of docetaxel and topotecan in patients with solid tumors.

PURPOSE: The sequence in which chemotherapeutic agents are administered can alter their pharmacokinetics, therapeutic effect, and toxicity. We evaluated the pharmacokinetics and pharmacodynamics of docetaxel and topotecan when coadministered on two different sequences of administration. PATIENTS AND METHODS: On cycle 1, docetaxel was administered as a 1-hour infusion at 60 mg/m(2) without filgrastim and at 60, 70, and 80 mg/m(2) with filgrastim on day 1, and topotecan was administered at 0.75 mg/m(2) as a 0.5-hour infusion on days 1 to 4. On cycle 2, topotecan was administered on days 1 to 4, and docetaxel was administered on day 4. Cycles were repeated every 21 days. Blood samples for high-performance liquid chromatography measurement of docetaxel (CL(DOC)) and topotecan (CL(TPT)) total clearance were obtained on day 1 of cycle 1 and day 4 of cycle 2. CL(DOC) and CL(TPT) were calculated using compartmental methods. RESULTS: Mean +/- SD CL(DOC) in cycles 1 and 2 were 75.9 +/- 79.6 L/h/m(2) and 29.2 +/- 17.3 L/h/m(2), respectively (P: <.046). Mean +/- SD CL(TPT) in cycles 1 and 2 were 8.5 +/- 4.4 L/h/m(2) and 9.3 +/- 3.4 L/h/m(2), respectively (P: >. 05). Mean +/- SD neutrophil nadir in cycles 1 and 2 were 4,857 +/- 6, 738/microL and 2,808 +/- 4,518/microL, respectively (P: =.02). CONCLUSION: Administration of topotecan on days 1 to 4 and docetaxel on day 4 resulted in an approximately 50% decrease in docetaxel clearance and was associated with increased neutropenia.

Detection of bladder cancer using a novel nuclear matrix protein, BLCA-4.

We have identified previously six nuclear matrix proteins (NMPs) that are bladder cancer specific. In this study, we analyzed the expression of one of these proteins, BLCA-4, in bladder tumors and normal bladder tissue. We also examined the appearance of BLCA-4 in the urine as a biomarker for bladder cancer. BLCA-4 was isolated from nuclear matrix preparations of bladder tumors, and its peptide sequence was determined. The antibodies generated against the resulting BLCA-4 peptides were then used to detect its presence in immunoblots and in urine samples by immunoassay. We analyzed tissue samples of bladder tumor and normal donor bladders and urine obtained from 51 normal individuals and 54 patients with pathologically confirmed bladder cancer. The BLCA-4 peptide sequences do not resemble any known human protein sequences. On immunoblot analysis, BLCA-4 expression was detectable in tumor and normal tissues from patients with bladder cancer but not in any of the normal bladder tissue obtained from organ donors. Using a prospectively determined cutoff level of 13 A (absorbance) units/microg protein, all 51 normal individuals tested were negative for BLCA-4 expression, whereas 53 of 55 samples from patients with bladder cancer were positive. These results suggest that BLCA-4 is present throughout the bladder in both the tumor and morphologically normal areas in bladder cancer patients. BLCA-4 is a very sensitive (96.4%) and specific (100%) marker for bladder cancer. BLCA-4 is a bladder cancer-specific marker that can be detected using a urine-based assay and can be used in the diagnosis of bladder cancer.

Analysis of therapeutic and immunologic effects of R(24) anti-GD3 monoclonal antibody in 37 patients with metastatic melanoma.

BACKGROUND: Antitumor effects of antibodies against ganglioside antigens of melanoma have been reported, but neither optimal doses nor mechanisms have been established. METHODS: This Phase IB trial of the murine immunoglobulin IgG(3) monoclonal antibody R(24) against disialoganglioside GD3 was conducted with 37 patients to define better the dose-response relation and mechanism of action of R(24) in patients with metastatic melanoma. RESULTS: Dose-limiting toxicity consisted of a pulmonary capillary leak syndrome in 3 of 5 patients in the 80 mg/M(2)/day dosage tier. Serial blood and tumor biopsy samples were obtained prior to therapy and on Days 5, 9, and 22 following R(24) infusion. Tumor biopsy-infiltrating lymphocytes were enumerated in peritumoral, endotumoral, and perivascular compartments: endotumoral CD4(+) and CD8(+) T cells and HLA-DR(+) T cells increased over time on R(24) antibody. Endotumoral CD4 lymphoid infiltrate activation (DR expression) and antibody-dependent cytotoxicity were the greatest in the one patient who achieved a complete response. CONCLUSIONS: Clinical response was associated with depression in natural killer (CD56(+) and CD56(+)DR(+)) blood cells (P = 0.03) and was associated with R(24) dosage (P = 0.01). A complete response that lasted 2 years and a partial response that lasted 2 months occurred at a dose of 1 mg/M(2)/day. The limited number of clinical responses observed in this trial hampered the correlation of antitumor and immune parameters but provided a rational foundation for the future evaluation of antiganglioside antibodies.

Brain tumor volume measurement: comparison of manual and semiautomated methods.

PURPOSE: To compare the reliability of two approaches to measuring enhancing brain tumor volumes--the conventional manual trace method and a threshold-based, semiautomated computer software method. MATERIALS AND METHODS: Two operators rated contrast material-enhanced, T1-weighted axial magnetic resonance (MR) image data sets from 16 patients aged 21-71 years with high-grade gliomas. Each MR data set was rated twice by using manual tracing and twice by using the semiautomated method. The semiautomated measurement method involved a thresholding algorithm based on mixture modeling. The data collection time for each method was recorded. Reliability was measured by using inter- and intraoperator agreement indexes. RESULTS: Mean intraoperator agreement indexes (+/- SD) were 0.90 +/- 0.09 (operator 1) and 0.83 +/- 0.15 (operator 2) for the manual trace method and 0.83 +/- 0.17 (operator 1) and 0.84 +/- 0.16 (operator 2) for the semiautomated measurement method. The mean interoperator agreement was 0.85 +/- 0.14 for the manual method and 0.82 +/- 0.18 for the semiautomated method. The semiautomated method was faster than the manual trace method by an average of 4.6 minutes per patient. CONCLUSION: The semiautomated computer method of measuring tumor volume was faster than the manual trace method. Semiautomated computer approaches offer an alternative to manual tracing for measuring serial tumor volumes in patients with high-grade brain neoplasms.

Decreased host-cell reactivation of UV-irradiated adenovirus in human colon tumor cell lines that have normal post-UV survival.

An ongoing study in our laboratories is to examine the relationship of DNA repair defects to human cancer. Our underlying hypothesis has been that human tumors may arise that lack interesting DNA repair pathways if these pathways are important in preventing cancer. In this study, we found that the UV-irradiated adenoviruses showed hypersensitivity when assayed on monolayers of certain human colon tumor cell lines, including three that are reported to have defects in long patch DNA mismatch repair genes and one with no reported defect in mismatch repair. The survival curves showed two components. The first sensitive component was characteristic of 77-95% of the infections depending upon the cell line and the experiment and had an average slope indicating 4.8-fold hypersensitivity to UV. The average of the second-component slopes indicated that the remainder of the infections was accompanied by near-normal repair. Although the value of the first component indicated that the colon tumor lines supported the growth of UV-damaged adenoviruses poorly, the cell lines themselves showed the same post-UV colony-forming ability as did normal human fibroblasts, and their ability to support the growth of N-methyl-N'-nitro-N-nitrosoguanidine-damaged adenoviruses was normal, i.e. it parallelled their ability to repair O6-methylguanine in vitro. We previously observed two-component survival curves when assaying UV-irradiated adenovirus on monolayers of all of seven strains of fibroblasts from Cockayne's syndrome patients. By contrast, single-component curves have been obtained using 21 strains of normal human fibroblasts and seven other tumor lines. We interpret the two-component survival curves in terms of the defective transcription-coupled repair of UV-induced DNA damage that is characteristic both of Cockayne's and certain colon tumor cell lines. In addition, four mismatch repair-deficient colon tumor lines were resistant to killing by elevated levels of dG.

Correlation of B-FABP and GFAP expression in malignant glioma.

The murine brain fatty acid binding protein (B-FABP) is encoded by a developmentally regulated gene that is expressed in radial glial cells and immature astrocytes. We have cloned the human B-FABP gene and have mapped it to chromosome 6q22-23. We show that B-FABP mRNA is expressed in human malignant glioma tumor biopsies and in a subset of malignant glioma cell lines, as well as in human fetal retina and brain. Malignant glioma tumors are characterized by cytoplasmic bundles of glial fibrillary acidic protein (GFAP), a protein normally expressed in mature astrocytes. Establishment of malignant glioma cell lines often results in loss of GFAP. The subset of malignant glioma cell lines that express GFAP mRNA also express B-FABP mRNA. Co-localization experiments in cell lines indicate that the same cells produce both GFAP and B-FABP. We suggest that some malignant gliomas may be derived from astrocytic precursor cells which can express proteins that are normally produced at different developmental stages in the astrocytic differentiation pathway.

The integration of a novice user interface into a professional modeling tool.

This paper describes a software tool, the Oncology Thinking Cap (OncoTCAP) and reports on our efforts to develop a novice user interface to simplify the task of describing biological models of cancer and its treatment. Oncology Thinking Cap includes a modeling tool for making relationships explicit and provide dynamic feedback about the interaction between cancer cell kinetics, treatments, and patient outcomes. OncoTCAP supports student learning by making normally invisible processes visible and providing a representational tool that can be used to conduct thought experiments. We also describe our novice interface and report the results of initial usability testing.

Intact G2-phase checkpoint in cells of a human cell line lacking DNA-dependent protein kinase activity.

Cells respond to radiation-induced DNA damage in a cell cycle phase-specific manner as shown by (1) variation in radiosensitivity across the cell cycle and (2) checkpoints in G1 and G2 phase at which arrest of progression of cells through the phases of the cell cycle occurs. We studied these processes in cells of human glioma cell lines which lack (M059J(PK-)) or express (M059K(PK+)) DNA-dependent protein kinase (DNA-PK) activity. Cell populations enriched with cells of a specific cell cycle phase were y-irradiated and analyzed for cell survival. Although both cell lines were relatively sensitive in G1 phase and resistant in S phase, the differential sensitivity was greater in M059J(PK-) cells. In the studies on checkpoints, unsynchronized cells were irradiated and examined for evidence of cell cycle arrest. Neither cell line showed a postirradiation G1-phase arrest, presumably because of mutant p53 status. For M059J(PK-) cells, all doses tested (2.5-10 Gy) resulted in a significant increase in the proportion of G2/M-phase cells; however, for M059K(PK+) cells, a significant increase in G2/M phase was observed only after 10 Gy. These results suggest that the ability to activate the G2-phase checkpoint remains intact in cells which lack DNA-PK activity.

Increased radiosensitivity of p16 gene-deleted human glioma cells after transfection with wild-type p16 gene.

The A1235 and T98 cell lines derived from human gliomas have homozygous deletions in their p16 genes and are radiosensitive and radioresistant, respectively, with respect to other established glioma cell lines. These differences in radiosensitivity may be due to variations to some extent among cell lines, rather than genetically defined resistance or sensitivity. We examined the effect on radiation sensitivity of introducing a wild-type p16 gene into both p16-deficient glioma cell lines. The plasmid pOPMTS containing human wild-type p16 cDNA and a neomycin resistance gene, or the control plasmid pOPRSV1, were transfected into these cells. Clones from both cell lines, which expressed wild-type p16 mRNA constitutively after transfection with pOPMTS, were more radiosensitive than the parental cells and clones obtained after transfection with the negative control plasmid.

Vitamin D inhibition of prostate adenocarcinoma growth and metastasis in the Dunning rat prostate model system.

OBJECTIVES: Risk factors for prostate cancer (PCa)-related mortality include old age, black race, and residence in northern latitudes. The objectives of this study are to examine the in vitro and in vivo effects of 1,25-dihydroxycholecalciferol (1,25-D3) and less-hypercalcemic analogues on the Dunning rat prostate adenocarcinoma model. METHODS: To evaluate the effect of 1,25-D3 on PCa in vitro, we used the highly metastatic Mat-lylu (MLL) and moderately metastatic R3327-AT-2 (AT-2) Dunning prostate cell lines, and examined effects on growth, clonogenicity, differentiation, and cell cycle. In vivo analysis included examination of the effects of these compounds on tumor growth and metastasis. RESULTS: Using both the 3-day MTT and 7-day clonogenic assay, 1,25-D3 demonstrated a growth inhibitory effect with a concentration for 50% inhibition (IC50) of approximately 20 microM for both MLL and AT-2. Cell cycle analysis of treated MLL cells (10 microM 1,25-D3 for 48 hours) had 25% more cells in the G0/G1 phase than did control cells. To examine the in vivo effect of 1,25-D3 and the less hypercalcemic vitamin D analogue, Ro25-6760 (or 6760), on MLL PCa growth and metastasis, tumors (5 x 10(5) cells) were implanted subcutaneously into the flank of Copenhagen rats on the same day that treatment was initiated with 1,25-D3 (1 microgram) or 6760 (1 or 5 micrograms); rats received treatment three times a week. After 3 weeks, 1,25-D3 and 6760 (5 micrograms dosing) resulted in an inhibition of tumor volume and a reduction in the number and size of lung metastases. CONCLUSIONS: These preclinical studies demonstrate the profound in vitro, or in vivo, or both antiproliferative and differentiating effects of 1,25-D3 and 6760 on PCa and suggest that these drugs may have potential beneficial effects in the treatment of advanced PCa.

Radiation-induced damage in two human glioma cell lines as measured by the nucleoid assay.

We measured radiation-induced damage and repair in MO59J and MO59K human glioma cell lines using the nucleoid halo assay. Although these two cell lines have different radiosensitivities when assayed for colony forming efficiency, our results indicated that there was no significant difference between the two in terms of the unwinding and rewinding of DNA supercoils, radiation-induced changes in nucleoid halo size or the kinetics of nucleoid halo lysis. The only differences noted were in the kinetics of recovery of radiation-induced changes in nucleoid halo size, with the more sensitive cell line (MO59J) showing a slightly faster recovery than the more resistant cell line (MO59K). However, this difference was not statistically different. Our data indicate that the different cellular radiosensitivities of MO59J and MO59K cells are probably not due to any differences in their supercoiled DNA structure as measured by the nucleoid halo assay.

Using multiple decrement models to estimate risk and morbidity from specific AIDS illnesses. Multicenter AIDS Cohort Study (MACS).

A simple non-parametric approach is developed to simultaneously estimate net incidence and morbidity time from specific AIDS illnesses in populations at high risk for death from these illnesses and other causes. The disease-death process has four-stages that can be recast as two sandwiching three-state multiple decrement processes. Non-parametric estimation of net incidence and morbidity time with error bounds are achieved from these sandwiching models through modification of methods from Aalen and Greenwood, and bootstrapping. An application to immunosuppressed HIV-1 infected homosexual men reveals that cytomegalovirus disease, Kaposi's sarcoma and Pneumocystis pneumonia are likely to occur and cause significant morbidity time.

Base analog and neighboring base effects on substrate specificity of recombinant human G:T mismatch-specific thymine DNA-glycosylase.

We studied the substrate specificity of the human G:T mismatch-specific thymine glycosylase that initiates the repair of G:T and G:U base mismatches to G:C base pairs. Such mismatches arise when 5-methylcytosine or cytosine deaminate spontaneously (and hydrolytically) in DNA. Substrates were 45-bp DNA heteroduplexes that bore single G:T, m6G:T, 2,6-diaminopurine:T, 2-amino-6-(methylamino)-purine:T, 2-aminopurine:T, and G:m4T mispairs. The bases 5' to the poorly matched G were altered in selected G:T substrates to yield mispairs in four different contexts, ApG, CpG, GpG, and TpG. The recombinant thymine glycosylase was incubated with the 45-bp DNA substrates, each labeled at the 5'-terminus of the strand containing the mismatched T. The DNAs were then treated with 0.1 N NaOH to catalyze phosphodiester bond breakage at the newly-generated AP sites, and the products were analyzed on DNA sequencing gels. As indicated by the amounts of the 20-nt incision product, the removal of the thymine base by the enzyme increased linearly between 0 and 40 min at which time the generation of product from all substrates ceased, probably because of enzyme inactivation. The rate of incision was greatest (0.7 fmol/min) with DNA containing the G:T mispair followed by the DNA containing the m6G:T mispair (0.38 fmol/min) and the DNA with the 2-amino-6-(methylamino)purine:T mispair (0.15 fmol/ min); the extent of reaction was 90%, 40%, and 20% respectively. By contrast to previous findings with cell-free extracts, DNA substrates containing 2,6-diaminopurine:T, 2-aminopurine:T, and G:m4T mispairs were not incised (< 2%). The amount of incision of the 45-bp DNA substrates containing G:T mispairs in the CpG context was 3-12-fold greater than in the TpG, GpG, and ApG contexts.