Fig 1. Purification and in vitro characterization of the S. cerevisiae MutLγ complex.
A. Purification of affinity-tagged S. cerevisiae Mlh1-Mlh3 heterodimers from baculovirus-infected insect cells. A Coomassie-stained SDS-PAGE gel is shown; 2 μg of protein was loaded. B. Effects on meiotic recombination of N-terminal tagging of Mlh1 (HisFlagmlh1) and Mlh3 (HisFlagmlh3). Crossover frequencies in two test intervals (CEN8-ARG4 and ARG4-THR1) and MI nondisjunctions frequencies were measured by the spore-autonomous fluorescence assay (described in detail in Fig 4 below; values for mlh1Δ and mlh3Δ are reproduced from Fig 4 for comparison). Error bars represent standard errors for crossovers and 95% confidence interval of the proportion for MI nondisjunction. Statistically significant differences are indicated with asterisks (*, p<0.05). Selected non-significant differences are highlighted (n.s.). C. Effects on MMR of N-terminal tagging of Mlh1 and Mlh3. Graphs show Lys+ reversion frequencies normalized to the wild-type value. See text associated with Fig 5 for details on the assay. Error bars are standard deviation of the mean (n = 3–6). Values for mlh1Δ and mlh3Δ are reproduced from Fig 5 for comparison. D. DNA cleavage activity of MutLγ. Reactions containing 100 nM MutLγ and 5.7 nM pUC19 plasmid were incubated at 30°C for 1 hour, then stopped, separated on an agarose gel, and stained with ethidium bromide. The nuclease-dead mutant (MutLγ-nd) carries the active site mutation D523N on Mlh3. E. Michaelis-Menten data of MutLγ ATPase activity (Vmax = 0.225 ± 0.012 min-1; Km = 0.064 ± 0.013 mM; mean ± SE). Error bars show the range from two independent experiments. F. Effect of DNA on the ATPase activity of MutLγ. Substrates are identical to the ones used for binding assays in Fig 2C (see Materials and methods). Reactions contained 280 nM MutLγ with 4.8 pg DNA (2, 1 and 0.5 μM ssDNA, dsDNA and HJ, respectively). Error bars show the standard deviations from three independent determinations. Data were normalized to the mean of the ATPase activity in the absence of DNA (kcat = 0.112 ± 0.005 min-1). G. Partial proteolysis of MutLγ. A Coomassie-stained SDS-polyacrylamide gel is shown. Asterisks indicate some of the protein fragments that are different in the presence or absence of ATP (5 mM). H. AFM imaging of MutLγ. The image illustrates the heterogeneity of the size and configurations of MutLγ particles observed. I. Examples of the four configurations of dimeric MutLγ particles observed by AFM. The inference of the domains involved in dimerization is based on the characterization of other MutL proteins, which show that the C-terminal domains dimerize independently of ATP, while the N-terminal domains dimerize upon ATP binding [52, 53]. Which subunit is folded in the ‘one arm’ configuration is not known; the cartoon showing Mlh1 folded is intended only as an illustration. J. Volume analysis of MutLγ particles. The predicted volume of the heterodimer (175 kDa) is 257 nm3 according to the equation: Vc = (M0/N0)(V1+dV2), where M0 is the molecular weight, N0 is Avogadro’s number, V1 and V2 are specific volumes for protein (0.74 cm3 g-1) and water (1 cm3 g-1) and d is the extent of protein hydration (0.4 g H2O/g protein) [76]. K. Classification of MutLγ particles in the absence and presence of 1 mM ATP. A cutoff of 170 nm3 was chosen to classify particles as dimers (see panel J). The difference between distributions is statistically significant by G test, p = 0.0082.