Upload
others
View
5
Download
0
Embed Size (px)
Citation preview
Nature Structural & Molecular Biology: doi:10.1038/nsmb.3224
Supplementary Figure 1
Domain construction, sequence, overall structure and active site of R.i. and E.r. maturases.
a. Organization of the maturase RT construct. All maturases contain an N-terminal RT domain (blue) and a C-terminal thumb/X domain (orange). b. Sequence alignment of the full-length maturase, comparing E.r. (Eu.re.I2) and R.i. (Ro.in.I1). The black boxes indicate identical residues, the grey boxes indicate similar residues, and the white box indicates residues with substantial differences. The red bar indicates the boundary for the RT domain used for crystallization. The blue arrow indicates the N-terminal RT domain and the orange arrow indicates the C-terminal X or thumb domain. c. R.i. and E.r. RT domain structures in cartoon representation. The R.i. RT is shown in green and the E.r. RT is shown in gray. d. Comparison of the E.r. RT active site and primer grip with an active telomerase (PDBID: 3KYL)
49. The E.r. RT is shown in grey and telomerase is shown in cyan. For the residues shown as sticks, red color indicates
oxygen and blue color indicates nitrogen atoms. The potassium in the R.i. RT active site is shown in purple, and the Mg2+
in the telomerase active site is shown in yellow. Metal ion coordination is indicated by orange dashed lines.
Nature Structural & Molecular Biology: doi:10.1038/nsmb.3224
Nature Structural & Molecular Biology: doi:10.1038/nsmb.3224
Supplementary Figure 2
Maturase RT forms a dimer.
a. Each panel shows the content of an ASU in different crystals, and the organism names and the space groups are indicated at the bottom of each panel. In the dimer, one monomer is colored in grey and the other is colored in green. The ASU of R.i. RT crystallized in P21 space group from condition A (see online methods) contains 2 molecules. The ASU of R.i. RT crystallized in P212121 space group from condition B (see online methods) contains 4 molecules. The ASU of R.i. RT crystallized in P1 space group from condition C (see online methods) contains 4 molecules. The ASU of E.r. RT crystallized in P21 space group from condition D (see online methods) contains 8 molecules. b. The active site within the R.i. RT domain is partially obstructed by the dimerization interface. On the left panel, monomer A is represented as a cartoon diagram in grey, and monomer B is represented as a space-filling model in green. For monomer A, the active site loop is colored in red, the primer grip is colored in magenta and the insertion loop containing a short α-helix a10 is colored in yellow. The right panel is a close-up view of the interactions that stabilize the insertion loop shown in sticks. Oxygen atoms are colored in red, nitrogen atoms are colored in blue and hydrogen bonds are shown as orange dashed lines.
Nature Structural & Molecular Biology: doi:10.1038/nsmb.3224
Supplementary Figure 3
A possible model for association of the maturase dimer with a group II intron RNA.
A cartoon diagram of oceanobacillus iheyensis group II intron (PDBID: 4FAQ)29
is shown on the left, where domain 1 (D1) is colored yellow and domain 4 (D4) is colored in cyan. All other domains are grey. The electrostatic surface of the E.r. RT dimer is shown at right, where blue indicates positive charge and red indicates negative charge. The sizes of the group II intron model and the maturase RT dimer model are not to scale. Based on the electrostatic surface potentials of the RT dimer, and previous results showing that maturase interacts with the intron as a dimer
15-16 through a high-affinity binding site in intron D4
14 and additional auxiliary sites in D1
48, the
potential surfaces involved in RNA-protein interactions are marked by arcs, and the potential interactions are indicated by dashed lines.
Nature Structural & Molecular Biology: doi:10.1038/nsmb.3224
Nature Structural & Molecular Biology: doi:10.1038/nsmb.3224
Supplementary Figure 4
E.r. RT binds RNA with high affinity and specificity.
a. Comparison of the secondary structures for D4 in the L1.LtrB14
and E.r. introns. The residues colored in red belong to the D4A region. The start codon is highlighted in green. The rest of the open reading frame (ORF) encoding the maturase is indicated by a circle. b. Electrophoretic mobility shift assay (EMSA) for E.r. RT binding to D4A or D2 (Fig. 3b and Fig. 3c). The protein concentrations (nM) for each lane are indicated at the top. c. Phosphate binding site on the R.i. RT domain. The R.i. RT is shown as a surface rendering in grey. The phosphorus atom is orange and the oxygens are red. The simulated annealing omit map for the phosphate ion is shown in blue at the 3σ level. The interactions stabilizing the phosphate are shown on the right panel. Carbon atoms are grey, oxygens are red and nitrogen are blue. Hydrogen bonds are shown as orange dashed lines. d. Molecular weight analysis of E.r. D4A RNA using SEC-MALS. The experiments for E.r. D4A RNA (shades of grey) were performed over two concentrations (in mg/mL) as indicated, and the MW at each concentration was plotted as triangles and circles respectively (upper right legend) on the right y-axis. For each concentration, the UV trace (curve) was plotted on the left y-axis with the elution volume indicated on the x-axis. The MW at elution peak is 23 kDa at 0.03 mg/mL and 24 kDa at 0.07 mg/mL.
Nature Structural & Molecular Biology: doi:10.1038/nsmb.3224
Nature Structural & Molecular Biology: doi:10.1038/nsmb.3224
Supplementary Figure 5
Primer extension assay for the maturase.
Lanes 1-3 are DNA sequencing ladders (corresponding to nucleotides A, G and T respectively) showing the sequence and the resolution of the primer extension products on the gel. Lane 4 is the control lane, which only contains the primer. Lane 5 is the primer extension product from the E.r. RT construct, and lane 6 is the primer extension product from the E.r. full-length maturase. In order to resolve DNA products of vastly different length on the same gel (enabling us to compare nonprocessive and processive activity by the RT domain and full-length maturase constructs, respectively), we used a long sequencing gel. The gel is slightly twisted, so each corresponding lane is also labeled at the bottom of the gel to avoid confusion.
Nature Structural & Molecular Biology: doi:10.1038/nsmb.3224
Reference model: R.i. RT
Prp8 HCV NS5B TERT HIV RT
Dali de novo search Z-score 15.4 14.9 8.9 7.4
TM-score 0.698 0.682 0.534 0.505
Dali pairwise alignment Z-score 15.4 14.6 11.2 10.5
Reference model: R.i. RT-ΔRT0
TM-score 0.700 0.671 0.646 0.586
Dali pairwise alignment Z-score 12.3 12.0 12.2 10.5
Supplementary Table 1: TM-align scores and Dali pairwise alignment Z-scores for the
alignment between R.i. RT domain and various RT or RT-like domains. The de novo Dali
search was only performed using complete R.i. RT monomer (R.i. RT, top row) as the query
structure because R.i. RT monomer without the RT0 motif (R.i. RT-ΔRT0) failed to identity
additional significant hits besides Prp8 and HCV NS5B. In structural alignments, either the
complete R.i. RT monomer (R.i. RT, top rows) or the R.i. RT monomer without the RT0 motif (R.i.
RT-ΔRT0, bottom rows) were used as references as indicated in the table heading. The target
structures for comparison are the finger and palm subdomains from Prp8 (PDBID: 4I43, residues
882–1303)32, HCV RNA polymerase (PDBID: 1C2P, residues 1–385)42, TERT (PDBID: 3DU6,
residues 151–406)40 and HIV RT p66 subunit (PDBID: 2HMI, residues 1–246)41. The TM-scores
were produced by TM-align44 with a value between 0–1. Higher TM-score indicates higher
structural similarity. The Dali pairwise alignment Z-scores were produced by the Dali pairwise
alignment server43 and similarly higher Z-score suggests higher structural similarity.
Nature Structural & Molecular Biology: doi:10.1038/nsmb.3224