1. The RNA Helicase Mtr4p Modulates Polyadenylation in the TRAMP Complex
Huijue Jia, Xuying Wang, Fei Liu, Ulf-Peter Guenther, Sukanya Srinivasan, James T. Anderson, Eckhard Jankowsky 
Cell 
Volume 145, Issue 6, Pages (June 2011)
DOI: /j.cell
Copyright © 2011 Elsevier Inc. Terms and Conditions

2. Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

3. Figure 1 Modulated Polyadenylation Activity by TRAMP
(A) Polyadenylation reaction with radiolabeled (asterisk) tRNAiMet (0.5 nM tRNAiMet, 150 nM TRAMP, 2 mM equimolar ATP-Mg2+). Aliquots were removed at 1 min intervals and resolved on denaturing PAGE. Added adenosines are marked on the right.
(B) Left: contourplot of the fraction of the adenylated intermediates (Ai) versus reaction time for the time course in (A). The color bar shows the color progression from Ai = 0 to 0.2 (contours: Ai = 0.04, 0.08, 0.12, 0.16, 0.2). Right: contourplot for a simulated reaction with equal rate constants for each adenylation step (k = 1.5 min−1).
(C) Quantitative analysis of individual adenylation steps. Kinetic scheme for the polyadenylation reaction. For corresponding equations and fitting of the dataset, see Experimental Procedures. Plots show representative time courses for selected species (A0, A1, A2, A10) from the reaction displayed in (A). Lines indicate the fit.
(D) Observed rate constants for individual adenylation steps. Points represent averages for multiple independent experiments as shown in (A). The error bars mark one standard deviation. The modulation of individual observed rate constants was independent of the order of addition of TRAMP, RNA, and ATP (Figures S1C–S1E).
(E) Rate constants at TRAMP and ATP saturation (kmax) for individual adenylation steps. Rate constants were determined from multiple reactions with increasing TRAMP and ATP concentrations. Error bars mark the deviation of values obtained at ATP and TRAMP saturation (Figures S1F–S1L).
(F) Apparent ATP affinity (K1/2ATP) for individual adenylation steps. Values were determined from multiple reactions with increasing ATP concentrations (Figures S1F–S1L). Error bars indicate the standard deviation.
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

4. Figure 2
Accumulation of Poly(A) Tails with Approximately Four Adenosines on Hypomethylated pre-tRNAiMet In Vivo
(A) Experimental scheme to measure poly(A) tail lengths of pre-tRNAiMet in vivo by Sanger sequencing. The heterogeneous 3′ termini of pre-tRNAiMet (IMT1∼4) are displayed.
(B) Left panel: In vitro-transcribed tRNAiMet with three 3′-terminal uridines, polyadenylated by TRAMP. The number of appended adenosines is marked. Right panel: Representative Sanger sequencing chromatogram for this RNA after the poly(A) tail length measurement procedure shown in (A). The dashed line at A8 indicates the start of the decrease in the A signal and the increase in G signal.
(C) Representative sequencing chromatogram for the cellular pre-tRNAiMet sample. The dashed line at A4 indicates the start of the decrease in A signal and the increase in G signal. Experiments were repeated multiple times and virtually identical chromatograms were obtained.
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

5. Figure 3
Modulated Polyadenylation Activity with Generic Model Substrates
(A) Polyadenylation of an RNA substrate consisting of a 16 bp duplex with 1 nt 3′-terminal overhang (100 nM TRAMP, 2 mM ATP-Mg2+, and 0.5 nM RNA). The asterisk marks the radiolabel. The 16 nt top strand contained a 3′-terminal 2′,3′-dideoxy residue to prevent adenylation. Plots show rate constants at TRAMP and ATP saturation (kmax) and the apparent ATP affinity (K1/2ATP) for individual adenylation steps. For apparent substrate affinities (K1/2TRAMP) of individual steps, see Figure S2A. Values were determined from multiple reactions with increasing TRAMP and ATP concentrations (Figures S1F–S1L). Error bars indicate the standard deviation.
(B) Polyadenylation of a 23 bp RNA duplex with 1 nt 3′ overhang (100 nM TRAMP, 2 mM ATP-Mg2+, and 0.5 nM RNA, top strand with 3′-terminal 2′,3′-dideoxy residue). Plots correspond to those in (A). Error bars indicate the standard deviation. For apparent substrate affinities, see Figure S2B.
(C) Polyadenylation of a 24 nt ssRNA substrate (100 nM TRAMP, 2 mM ATP-Mg2+, and 0.5 nM RNA). Plots correspond to those in (A). Error bars indicate the standard deviation. For apparent substrate affinities, see Figure S2C.
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

6. Figure 4
Removal or Mutation of Mtr4p Diminishes Modulation of Polyadenylation Activity
(A) Polyadenylation of the 24 nt ssRNA substrate (asterisk: radiolabel) with Trf4p/Air2p (100 nM Trf4p/Air2p, 2 mM ATP-Mg2+, and 0.5 nM RNA). Plots correspond to those in Figure 3. For apparent substrate affinities, see Figure S3C. Values were determined from multiple reactions with increasing TRAMP and ATP concentrations (Figures S1F–S1L). Error bars indicate the standard deviation. As a reference, the dashed line marks A4.
(B) Polyadenylation of the 23 bp RNA duplex with 1 nt 3′ overhang by Trf4p/Air2p. The y axis for the plot of apparent ATP affinities was broken to enable direct comparison of the identical reaction with WT TRAMP (Figure 3B). Error bars indicate the standard deviation. For apparent substrate affinities, see Figure S3D.
(C) Polyadenylation of the 24 nt ssRNA substrate by TRAMPMtr4-20p (100 nM TRAMPMtr4-20p, 2 mM ATP-Mg2+, and 0.5 nM RNA) (Figure S3E). Plots correspond to those in Figure 3. Error bars indicate the standard deviation. For apparent substrate affinities, see Figure S3F.
(D) Polyadenylation of the 23 bp RNA duplex (1 nt 3′ overhang) by TRAMPMtr4-20p. The inset shows the data with 10-fold magnification in the y axis, to enable direct comparison of the identical reaction with WT TRAMP (Figure 3C). Error bars indicate the standard deviation. For apparent substrate affinities, see Figure S3G.
(E) Polyadenylation of the 16 bp duplex (1 nt 3′-terminal overhang) by TRAMPMtr4-20p. Error bars indicate the standard deviation. For apparent substrate affinities see Figure S3H.
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

7. Figure 5
TRAMP Adjusts Polyadenylation Activity Based on the Number of 3′-Terminal Adenosines
(A) Polyadenylation of a 24 nt ssRNA substrate with four terminal adenosines (filled symbols) by TRAMP. For comparison, the identical substrate without the terminal adenosines is shown (open symbols, values identical to those in Figure 3C). Asterisks mark the radiolabel. Values were determined from multiple reactions with increasing TRAMP and ATP concentrations. Error bars indicate the standard deviation.
The dashed lines mark k1 and k5, A1 and A5, the arrows emphasize the shift of the peaks for adenylation rate constants and apparent ATP affinities by four nucleotides.
(B) Polyadenylation of a 24 nt ssRNA substrate with four consecutive adenosines 5 nt removed from the 3′ terminus (filled symbols). For comparison, values for the identical substrate without the terminal adenosines are shown (open symbols, panel A). Plots correspond to those in (A). Error bars indicate the standard deviation.
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

8. Figure 6
E947 in Mtr4p Is Critical for the Modulation of Polyadenylation
(A) Domain structure of Mtr4p (Weir et al., 2010). Domain names are shown. The blue bar represents E947.
(B) Crystal structure of Mtr4p in complex with ADP and 5 nt oligo(A). Molecule B from Weir et al. (2010) is shown. The domains are colored as in (A). E947 is shown in blue and RNA in orange. The dashed circle marks the area magnified in (C).
(C) Close up view of E947 and the 5 nt oligo(A). For clarity, only resides 945–1026 in the helical bundle domain are shown (gray). E947 is positioned to contact N6 of the 4th adenine from the 5′ end (Weir et al., 2010).
(D) Polyadenylation of the 24 nt ssRNA substrate by TRAMPMtr4p(E947A) (100 nM TRAMPMtr4p(E947A), 2 mM ATP-Mg2+, 0.5 nM RNA).
(E) Rate constants at TRAMPMtr4p(E947A) and ATP saturation (kmax, upper panel), and apparent ATP affinity (K1/2ATP, lower panel) for individual adenylation steps. For comparison, values for WT TRAMP are shown (open shapes). For apparent substrate affinities, see Figure S4A. Values were determined from multiple independent reactions. Error bars indicate the standard deviation.
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

9. Figure 7
TRAMP Processivity and Mtr4p Effects on Multiple Reaction Parameters
(A) Reaction scheme illustrating the principle of processivity (P1…n) for individual adenylation steps (T: TRAMP, A0…n: adenylated RNA species, TA0…n: TRAMP bound to the respective adenylated species, kf1…n: adenylation rate constant for individual step, kdiss1…n: dissociation rate constant for individual step). For more experimental details, see Experimental Procedures, Extended Experimental Procedures, and Figure S5.
(B) Processivity of TRAMP for individual adenylation steps with the 24 nt ssRNA substrate. The average number of steps (N), shown at the right, corresponds to the processivity according to: P = (N-1)/N (Ali and Lohman, 1997). The dotted line marks P = 0.5, N = 2. Processivity values are the average from multiple independent measurements; the error bars mark one standard deviation.
(C) Processivity of Trf4p/Air2p for individual adenylation steps of the 24 nt ssRNA substrate. Values are the average from multiple independent measurements; the error bars mark one standard deviation.
(D) Actual adenylation rate constants of TRAMP (filled circles) and Trf4p/Air2p (open circles) for individual adenylation steps with the 24 nt ssRNA substrate. Rate constants were calculated according to Equation 1 with kfn + kdissn = kmaxn (Ali and Lohman, 1997). Values shown were calculated from the data in Figure 3C and Figure 4A and panels B and C. Error bars mark one corresponding standard deviation.
(E) Dissociation rate constants of TRAMP (filled circles) and Trf4p/Air2p (open circles) for individual adenylation steps with the 24 nt ssRNA substrate. Rate constants were calculated with Equation S7, using the values for Pn and kfn determined in (B)–(D). Error bars mark one corresponding standard deviation.
(F and G) Free activation enthalpies (ΔG‡) for adenylation (upper panels) and dissociation (middle panels), and the free energy for ATP affinities (ΔG°, lower panels) for individual adenylation steps for TRAMP (F) and Trf4p/Air2p (G), measured for the 24 nt ssRNA substrate. Free activation enthalpies were calculated according to ΔG‡ = -RT·ln(hk/kbT) (R: gas constant, T: temperature, h: Planck constant, k: rate constants determined in panels D and E, kb: Boltzmann constant). Free energies for functional ATP affinities were calculated according to ΔG° = -RT·ln(1/K1/2ATP), using the ATP affinities (K1/2ATP) determined in Figure 3C (TRAMP) and Figure 4A (Trf4p/Air2p).
(H) Mtr4p effects on free activation enthalpies for adenylation (upper panel) and dissociation (middle panel), and on the free energies of functional ATP affinities (lower panels) for individual adenylation steps. The effect is expressed as difference in the respective free activation enthalpies and free energies shown in (F), e.g., ΔΔG‡ = ΔG‡(TRAMP) − ΔG‡(Trf4p/Air2p). The arrows on the right show how energy differences correspond to slower/faster rate constants and weaker/tighter ATP binding for each adenylation step.
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

10. Figure S1
Purification and Reconstitution of S. cerevisiae TRAMP and Quantitative Measurement of Polyadenylation Activity In Vitro, Related to Figure 1
(A) SDS-PAGE with recombinant Trf4p/Air2p, Mtr4p after the respective purification steps and the TRAMP complex after final purification. Coomassie staining colors Air2p to a lesser extent than Trf4p and Mtr4p.
(B) All TRAMP components co-fractionate after size exclusion chromatography. Distribution of Mtr4p (black circles), Trf4p (red circles), and Air2p (blue circles) after fractionation with a gel filtration column (Superose 12, GE Healthcare). Proteins were quantified by SDS-PAGE and Coomassie staining. The corresponding curves represent best fits to a Gaussian distribution. The dashed lines indicate peak positions for molecular weight standards under conditions used for analysis of TRAMP samples.
(C) Modulation of polyadenylation activity is seen regardless of the order of addition of reaction components. Polyadenylation reactions were performed with 0.5 nM tRNAiMet, 100 nM WT TRAMP, and 2 mM ATP-Mg2+, as described in Figure 1, and in the Experimental Procedures section. Reaction schemes in the upper panels correspond to the plot of individual adenylation rate constants in the lower panels. TRAMP and RNA were incubated in reaction buffer for 5 min, ATP was added to start the reaction. Points represent averages of multiple independent experiments. The error bars mark one standard deviation. The dashed line marks the peak adenylation rate constant at A3.
(D) TRAMP and ATP were incubated in reaction buffer for 5 min, RNA was added to start the reaction.
(E) RNA and ATP were incubated together in reaction buffer for 5 min, TRAMP was added to start the reaction.
(F) Determination of maximal adenylation rate constants and apparent substrate affinities from titrations of TRAMP and ATP. Polyadenylation reactions were performed with 0.5 nM tRNAiMetas described in Figure 1, and in the Experimental Procedures section. Representative plots of observed adenylation rate constants for individual steps at different TRAMP concentrations (25 nM TRAMP: open circles; 250 nM TRAMP: filled circles).
(G) Representative plots of adenylation rate constants for each step versus the TRAMP concentration (2 mM ATP-Mg2+). Examples shown are for step 1 (filled circles) and step 3 (open circles). Plots of the adenylation rate constants versus TRAMP concentration ([TRAMP]) were fit to a binding isotherm according to kobs = kmax[TRAMP] / (K1/2TRAMP + [TRAMP]). Rate constants at TRAMP saturation at each step (kmaxTRAMP) and the TRAMP affinity (K1/2TRAMP) were obtained from the fits.
(H) Representative plots of adenylation rate constants for each step versus ATP concentration (100 nM TRAMP). Examples shown are for step 1 (filled circles) and step 3 (open circles). Plots of the adenylation rate constants versus ATP concentration ([ATP]) were fit to a binding isotherm according to kobs = kmax[ATP] / (K1/2ATP + [ATP]). Rate constants at ATP saturation at each step (kmaxATP) and the ATP affinity (K1/2ATP) were obtained from the fits.
(I) Resulting plot of maximum rate constant at TRAMP saturation. Error bars indicate the standard deviation of the values obtained by fitting the curves in (G).
(K) Apparent substrate affinity(K1/2TRAMP) for individual steps. Error bars indicate the standard deviation of the values obtained by fitting the curves in (G).
(L) Observed rate constants determined from ATP and TRAMP titrations (kmaxATP and kmaxTRAMP respectively) extrapolated to TRAMP saturation (kmaxATP,TRAMP, open circles) and to ATP saturation (kmaxTRAMP,ATP, filled circles), respectively. The error bars mark one corresponding standard deviation. The average of these two values is shown as kmax in Figure 1E.
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

11. Figure S2
Apparent Substrate Affinities (K1/2TRAMP) for ds and ssRNA Model Substrates and Reactions with the 17 nt ssRNA, Related to Figure 3
Affinities were determined as described for the tRNAiMet (Figures S1F–S1L), but for the substrates used in Figure 3. The error bars mark one standard deviation. Cartoons indicate the substrates, the asterisks mark the radiolabel.
(A) Apparent substrate affinity for the RNA substrate consisting of a 16 bp duplex with a single nt 3′-terminal overhang.
(B) Apparent substrate affinity for the 23 bp RNA duplex with a single nt 3′-terminal overhang. Twelve instead of ten steps are shown for consistency with Figure 3B.
(C) Apparent substrate affinity for the 24 nt ssRNA substrate.
(D) Representative time course of a polyadenylation reaction with the 17 nt ssRNA (bottom strand of the 16 bp-1 nt overhang duplex). The reaction was performed with 0.5 nM RNA, 100 nM TRAMP, 2 mM ATP-Mg2+, as described.
(E) Observed adenylation rate constant for the first adenylation step plotted against WT TRAMP concentration, for the 17 nt ssRNA substrate (filled circles), its duplex counterpart (16 bp-1 nt overhang, open circles) and the 24 nt ssRNA (filled triangles).
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

12. Figure S3
Polyadenylation of the 16 bp Duplex Substrate by Trf4p/Air2p, Apparent Substrate Affinities of Trf4p/Air2p (K1/2TRAMP) for ds and ssRNA Model Substrates, Integrity of the TRAMPMtr4-20p Preparation, and Apparent Substrate Affinities of TRAMPMtr4-20p for ds and ssRNA Model Substrates, Related to Figure 4
(A) Polyadenylation of the 16 bp duplex substrate (cf. Figure 3A) by Trf4p/Air2p. The reaction was performed with 0.5 nM P32-labeled RNA, 100 nM Trf4p/Air2p, 2 mM ATP-Mg2+, conditions equivalent to the reactions with TRAMP.
(B) Observed adenylation rate constant for the first step plotted against Trf4p/Air2p concentration, for the 16 bp duplex substrate (filled circles), and the 24 nt single strand (open circles).
(C and D) Apparent substrate affinities of Trf4p/Air2p (K1/2Trf4p/Air2p) for the 24 nt ss RNA substrate, and the 23 bp duplex substrate. Affinities were determined as described for TRAMP andtRNAiMet in Figure S1. The error bars mark one standard deviation. The cartoons depict the substrates, the asterisks mark the radiolabel.
(E) SDS-PAGE for reconstituted and purified TRAMPMtr4-20p. The individual components are marked on the left.
(F–H) Apparent substrate affinities of TRAMPMtr4-20p (K1/2TRAMP-Mtr4-20p) for the ds and ssRNA model substrates. The error bars mark one standard deviation. Cartoons indicate the substrates, the asterisks mark the radiolabel.
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

13. Figure S4
E947 in Mtr4p Is Critical for Modulation of TRAMP Activity, Related to Figure 6
(A) Apparent substrate affinity of TRAMPMtr4(E947A) for the 24 nt ssRNA substrate. The error bars mark one standard deviation.
(B) Effects of the Mtr4-20p mutation and the Mtr4p E947A mutation on observed free activation enthalpies for observed adenylation rate constants adenylation (upper panels), on the free energies of functional ATP affinities (middle panels) and apparent substrate affinities (lower panels) for individual adenylation steps. The effect is expressed as difference in the respective free activation enthalpies and free energies, e.g., ΔΔG‡ = ΔG‡(WT TRAMP) - ΔG‡(mtTRAMP). Free activation enthalpies were calculated according to ΔG‡ = -RT·ln(hk/kbT) (R: gas constant, T: temperature, h: Planck constant, k: rate constants determined in Figure 3C (WT), Figure 4C (TRAMPMtr4-20p), or Figure 5E (TRAMPMtr4(E947A)), kb: Boltzmann constant). Free energies for functional ATP affinities were calculated according to ΔG° = -RT·ln(1/K1/2ATP), using the ATP affinities (K1/2ATP) determined in Figure 3C (WT), Figure 4C (TRAMPMtr4-20p), or Figure 5E (TRAMPMtr4(E947A)). Free energies for substrate affinities were calculated according to ΔG° = -RT·ln(1/K1/2TRAMP), using the substrate affinities (K1/2TRAMP) determined in Figure S2C (WT), Figure S3F (TRAMPMtr4-20p), or Figure S4B (TRAMPMtr4(E947A)). The arrows on the right indicate how these energy differences correspond to slower/faster rate constants and weaker/tighter ATP or substrate binding, thus quantitatively describing effects of the Mtr4p mutations on the reaction parameters for each adenylation step.
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions

14. Figure S5
Measurement of TRAMP Processivity for the Polyadenylation Reaction, Related to Figure 7
(A) Reaction diagram for processivity measurements of TRAMP-catalyzed polyadenylation. Polyadenylation reactions were allowed to proceed for a defined time (t1). Then, scavenger RNA was added to prevent re-binding of TRAMP (73 nt RNA, sequence unrelated to substrate RNA). The scavenger was verified to completely sequester TRAMP at the concentrations used in the reaction (data not shown). After scavenger addition, the reaction was allowed to proceed for a defined time (t2), until the distribution of polyadenylated species no longer changed (10 min following scavenger addition, data not shown). Samples were then applied to denaturing PAGE, as described (see Experimental Procedures and panel B for representative data).
(B) Representative PAGE for processivity measurements. Polyadenylation reactions were performed with 0.5 nM 24 nt ssRNA substrate, 150 nM WT TRAMP or Trf4p/Air2p, 2 mM ATP. Polyadenylation was allowed to proceed for t1 = 2 min and t1 = 8 min, respectively (lanes 1 and 4, before addition of scavenger RNA), after which times scavenger RNA was added. After scavenger addition (73 nt ssRNA, 10 μM final concentration), aliquots were removed from the reaction at t2 = 2 min (lanes 2, 5) and t2 = 4 min (lanes 3, 6). The fraction of each adenylated RNA was quantified, yielding distributions of polyadenylated species at a given reaction time before (t1) and after (t2) scavenger addition.
(C) Shift in the distribution of polyadenylated RNA species after scavenger addition. Relative abundance of individual species (Ai / Σ (A0…An), i = 1…n) was plotted versus poly(A) length before (upper panel) and after scavenger addition (lower panel). The shift in the peak position of the distribution before and after RNA scavenger addition is marked by the arrow. Processivity for each step is calculated by comparing the distributions before and after scavenger addition, as described in the Extended Experimental Procedures.
Cell  , DOI: ( /j.cell )
Copyright © 2011 Elsevier Inc. Terms and Conditions