Structure and folding of the Tetrahymena telomerase RNA pseudoknot
Nucleic Acids Research
Structure and folding of the Tetrahymena telomerase RNA pseudoknot
Darian D. Cash 0
Juli Feigon 0
0 Department of Chemistry and Biochemistry, and Molecular Biology Institute, University of California , Los Angeles, CA 90095-1569 , USA
Telomerase maintains telomere length at the ends of linear chromosomes using an integral telomerase RNA (TER) and telomerase reverse transcriptase (TERT). An essential part of TER is the template/pseudoknot domain (t/PK) which includes the template, for adding telomeric repeats, template boundary element (TBE), and pseudoknot, enclosed in a circle by stem 1. The Tetrahymena telomerase holoenzyme catalytic core (p65-TER-TERT) was recently modeled in our 9 A˚ resolution cryo-electron microscopy map by fitting protein and TER domains, including a solution NMR structure of the Tetrahymena pseudoknot. Here, we describe in detail the structure and folding of the isolated pseudoknot, which forms a compact structure with major groove U•A-U and novel C•G-A+ base triples. Base substitutions that disrupt the base triples reduce telomerase activity in vitro. NMR studies also reveal that the pseudoknot does not form in the context of fulllength TER in the absence of TERT, due to formation of a competing structure that sequesters pseudoknot residues. The residues around the TBE remain unpaired, potentially providing access by TERT to this high affinity binding site during an early step in TERTTER assembly. A model for the assembly pathway of the catalytic core is proposed.
Telomerase is a ribonucleoprotein (RNP) complex
composed of the catalytic telomerase reverse transcriptase
(TERT), telomerase RNA (TER) and accessory proteins
that vary with species (1,2). Telomerase maintains the
telomeric DNA at the 3 ends of linear chromosomes by
adding de novo telomeric DNA repeats (3,4). Telomeres,
composed of telomeric DNA and bound proteins,
counteract DNA damage due to incomplete replication,
degradation, chromosomal fusion and illicit DNA damage
repair (5–7). TERT is generally well conserved ranging from
∼900 to 1100 amino acids (aa) among the commonly
studied vertebrates, yeast and ciliates. It contains four conserved
domains: the TERT essential N-terminus (TEN),
telomerase RNA binding domain (TRBD), reverse transcriptase
(RT) domain (palm and fingers) and C-terminal extension
(CTE) (thumb) (8). TER is more divergent, ranging in size
from ∼150 nucleotides (nt) in ciliates to over 3000 nt in
yeasts (9). TER has a number of conserved elements, which
are the template, pseudoknot (PK), template boundary
element (TBE) and stem terminus element (STE) (1,10). The
template/pseudoknot domain (t/PK) (also called core
domain) of TER includes the template that is used to copy
the telomeric sequence and an adjacent pseudoknot. The
pseudoknot is important for activity and has been proposed
to have a direct role in catalysis, in template positioning,
and/or in assembly (11–15). The TBE, within the t/PK, is
typically a stem or hairpin upstream of the template that
binds tightly to the TRBD and helps prevent copying of
nontemplate residues (16,17). Ciliates also have a template
recognition element (TRE) 3 of the template, which TERT
uses to direct efficient use of the template for reverse
transcription (18). The STE is distal from the t/PK and
stimulates telomerase activity typically through TERT TRBD
interactions (10). Along with TERT, the t/PK and STE of
TER are the minimal components necessary to reconstitute
telomerase activity in vitro (1,19).
The structures of individual TER domains have been
extensively studied in yeast, human and ciliates (15,20–29).
The NMR solution structures of the human and yeast K.
lactis TER pseudoknots revealed a number of conserved
features (15,21,29). Despite a divergent sequence and
secondary structure between the two pseudoknots, the 3D
structures are very similar, indicating conserved functions
or interactions. Both pseudoknots contain an extended
triple helix where the loop residues form base triples with
the Watson–Crick (WC) paired stem(s). Formation of the
base triples was shown to be important for catalytic activity
in vitro and in vivo (14,15,29–31). Early biochemical
studies have also linked the pseudoknot to telomerase assembly
and telomere repeat addition processivity (RAP) (32,33).
A smaller pseudoknot was predicted to form in
ciliate TERs (34). The telomere-rich ciliate Tetrahymena has
served as a model organism from which telomeres and
telomerase was discovered (35,36). Ciliate TERs are
generally composed of four base paired regions (numbered 1–4
and depicted in Figure 1) where Stems 3A and 3B form the
pseudoknot (34,37,38), Stem 2 and adjacent single-strand
nucleotides are the TBE, and Stem 1 closes the t/PK circle.
Distal stem–loop (SL) 4 is the ciliate STE, and is required
along with the t/PK for activity. NMR structures of SL2
and SL4 have been determined (22,25,26,39). Stem 4 was
crystallized in the presence of telomerase accessory protein
p65 C-terminal domain (40). p65 binds Stem 4, inducing a
105◦ bend at the conserved GA bulge that positions Loop
4 to interact with the TRBD (40,41). The conformational
change promotes hierarchical assembly of telomerase, with
the p65-TER complex facilitating assembly with TERT (40–
43). The ciliate pseudoknot has been particularly
challenging to characterize, due to its conformational dynamics,
which was revealed in early chemical probing studies (44).
It has a much shorter Stem A and Loop A than those found
in vertebrates (P2 and J2b/3) and yeasts. Interestingly, the
short length of Stem A is conserved amongst ciliates (3–4
bp) suggesting a possible function for pseudoknot
conformational flexibility (45). Based on phylogenetic analysis and
modeling this pseudoknot was predicted to have 2 U•A-U
triples in Stem B-Loop A (shown in Figure 1B) (45).
A number of studies have examined pseudoknot
formation in the Tetrahymena TER. Chemical probing revealed
that the isolated Tetrahymena pseudoknot (tetPK) can form
a pseudoknot, while tetPK does not form in the context of
full length (protein-free) TER (44). The authors concluded
that in full-length TER the pseudoknot region is in
equilibrium with alternate conformations. More recent
singlemolecule FRET experiments verified that tetPK forms in
isolation but not in full-length TER (46). Furthermore,
FRET values for free TER suggested that the pseudoknot
residues were involved in competing interactions with other
regions of TER (46). Based on SHAPE data, a model for
this alternative secondary structure was proposed, in which
residues from the pseudoknot region are base paired with
residues from the template and TRE (47). Importantly, both
SHAPE and FRET studies show that while tetPK does not
form in free TER, it does form in the presence of TERT.
Recently, a structural model of the Tetrahymena
telomerase holoenzyme was generated using a 9-A˚ resolution
cryo-electron microscopy (cryo-EM) map and fitting with
NMR and X-ray crystal domain structures and homology
models of proteins and RNA (12). This study revealed new
subunits and interactions, including two previously
unidentified proteins: telomeric repeat binding (Teb)2 and Teb3,
which interact with Teb1 to form TEB, a replication protein
A (RPA) paralog, that enhances processivity through DNA
binding. Furthermore, fitting of high-resolution structures
of domains of TER, TERT and p65 into the cryo-EM map
provided a detailed model of the RNP catalytic core. The
TERT TRBD-RT-CTE forms a ring which is encircled by
the TER t/PK. The template traverses the RT domain,
while the PK is distal from the active site, on the other side
of the TERT ring from the template near the CTE. p65
binds and bends Stem 4, and Loop 4 is inserted at the
interface between TRBD and CTE, on the opposite side of
TERT from the PK.
We determined the solution NMR structure of the
Tetrahymena telomerase pseudoknot and used it in
modeling the path of TER in our 9 A˚ cryo-EM map of the
telomerase holoenzyme (12). Here we used NMR to study the
structure and folding of the pseudoknot as an isolated
domain and in the context of the full length t/PK core. We
present details of the NMR data analysis and describe the
solution NMR structure of the isolated pseudoknot, which
we have further refined here. The isolated pseudoknot folds
into a structure that has stacked stems and loops interacting
in the respective major and minor grooves. The tetPK
tertiary structure is more compact than the human and yeast
pseudoknots, with only two stable major groove triples, an
expected U•A-U and a novel C•G-A+ triple with a
protonated A and syn G, and two minor groove triples.
Telomerase activity assays and NMR data confirm the importance
of the C•G-A+ triple for pseudoknot stability and
telomerase activity. NMR data shows that the pseudoknot does
not form in full length TER due to the competing
interactions between the template-TRE and pseudoknot residues
to form an alternative stem. Based on the determined
secondary structure of the t/PK and the cryo-EM structure
(12) a model for assembly of the catalytic core is proposed.
MATERIALS AND METHODS
RNA preparation for NMR studies
TetPK (and mutants tetPK-C75A, -C75U, -A80C, -G95U),
Stem 1 (nts 5–12:100–107), Stem 2 (nts 15–40),
Templatestem (nts 43–53:88–100), TRE-stem (nts 51–88),
tet5158:82–88, Stem 3alt (nts 42–100), t/PK (nts 5–107) and
TER (nts 1–159) RNAs (see Figures 1 and 4, and
Supplementary Figure S7 for sequences) were made as follows for
NMR studies. RNAs were synthesized by in vitro
transcription, using T7 RNA polymerase, from a synthetic DNA
template as previously described (48,49). The DNA
templates for t/PK and TER were cloned into the ApaI (5 )
and XhoI (3 ) sites of pRSF-1 vector. After linearization
with XhoI enzyme, 0.5–1.0 mg of DNA was used for in
vitro transcription with T7 RNA polymerase. The tetPK,
t/PK and TER templates contain a minimal consensus
hammerhead ribozyme sequence at the 3 end to allow
selfcleavage of each RNA molecule at a precise position (50,51).
Briefly, purified T7 RNA polymerase (P266L mutant) (52)
is added to a reaction containing 25–40 mM MgCl2, 4–6
mM of each NTP, and 1 M DNA template in 10 mM
Tris–HCl pH 8, 1 mM spermidine, 2.5 mM DTT and 0.01%
Triton. For tetPK and TRE-stem, uniformly 13C,15N
labeled rNTPs were appropriately substituted in the
reaction to make 13C,15N-(A,U) and 13C,15N-(G,C) double
labeled samples. For tetPK, a uniformly 13C,15N-labeled
sample was also made for RDC measurements. The reactions
were incubated at 37◦C for 3–4 h for transcription, and
an additional 3–4 h for cleavage of hammerhead ribozyme
constructs. The RNAs were subsequently purified by
electrophoresis on a 10–15% denaturing polyacrylamide gel
followed by electroelution of the excised RNA bands (53). The
RNAs were washed, using Amicon centrifugal filters, once
with high salt (1.5 M KCl), three times with water, and then
NMR buffer (10 mM NaPO4 pH 6.3, 50 mM KCl). The
RNA solution was heated under dilute conditions (10–50
M) at 95◦C for 4 min and then snap cooled on ice. The
RNAs were then concentrated to 0.5–1 mM for NMR
The solution NMR structure of tetPK was previously
reported (PDB 2N6Q) (12). Below, we describe the details for
NMR assignments, dihedral angle restraints, and RDCs for
structure calculation, and NMR assignments for tetPK and
other RNAs used in this study. NMR spectra were recorded
on Bruker DRX 500 MHz and 600 MHz and Avance
800MHz spectrometers equipped with HCN cryoprobes.
Exchangeable proton resonances for RNA constructs were
assigned at 283 K with samples in 90%H2O/10%D2O,
using 1D and 2D NOESY spectra in H2O, and
additionally 1H–15N HSQC and JNN-COSY spectra for tetPK.
Non-exchangeable proton resonances for tetPK and
TREstem were assigned with samples in 100% D2O, using
2D NOESY, 2D TOCSY, 1H–13C HSQC, 2D
HCCHCOSY, 3D HCCH-TOCSY, and a suite of filtered/edited
NOESY (F1fF2f, F2f, F1fF2e, F1eF2e) experiments at
283K (49,54–56) on 13C,15N-A,U and -G,C labeled samples.
Proton attached 13C and 15N resonances were assigned with
the HSQC experiments. The proton assignment strategy
used sequential NOE connectivity as previously described
(21,27,56). NOESY experiments were acquired with 100,
200 and 300 ms mixing times to determine which mixing
time best approximated a linear relationship between NOE
crosspeak volume and inter-proton distance. For use in
tetPK structure calculations, hydrogen bonds for
WatsonCrick (WC) and Hoogsteen base pairs were confirmed by
JNN-COSY (57,58). The imino proton resonances of the
terminal base pairs of tetPK (A69-U87, A76-U99) were not
detected due to rapid exchange with water, but these base
pairs were confirmed by NOEs (in D2O NOESY spectra).
Non WC base pairs (A80-G95, A91-C72, A90-C71) were
also confirmed by indicative NOEs (in D2O NOESY
RDCs for tetPK were measured for C–H (1DC1 H1 ,
1DC2H2, 1DC5H5, 1DC6H6, 1DC8H8) and N–H (1DN1H1,
1DN3H3) using the uniformly 13C,15N-labeled sample on 800
MHz spectrometer at 283 K (59). The spectra were
processed with Bruker TOPSPIN and analyzed with
NMRDraw. A total of 69 C–H and 10 N–H RDCs were
determined by measuring the difference in J-coupling in the
absence and presence of 15 mg/ml Pf1 phage.
New structure calculations using Xplor-NIH 2.42
For structure calculations, the NOE cross-peaks were
integrated to generate distance restraints using the
pyrimidine H5–H6 crosspeak as an internal standard (2.45 A˚ ),
using 200 ms mixing time NOESY spectra. NOEs were
categorized as very strong (2.5 A˚), strong (3.5 A˚),
moderate (4.5 A˚), weak (5.5 A˚ ) or very weak (6.5 A˚), with a
range of ±1 A˚ (except ‘very strong’ was limited to van der
Waals lower bound, 1.8 A˚). Loose A-form dihedral angle
restraints were used for the stem residues ( = −62◦ ± 30◦,
= −179◦ ± 30◦, = 47◦ ± 30◦, = −73◦ ± 30◦, =
37◦ ± 30◦). The ribose sugar pucker, correlated to angle,
was determined based on the H1 -H2 crosspeak intensity
in the DQF-COSY: strong (C2 endo, = 145◦ ±30◦),
intermediate (C2 endo, = 120◦ ± 30◦), or no cross peak
(C3 endo, = 82◦ ± 30◦). Syn/anti configuration is
correlated to , and determined based on characteristic NOE
pattern, where strong H8–H1 crosspeak intensity indicates
a syn conformation. Only G95 was syn ( = 25◦ ± 30◦),
and all other residues were set as anti ( = −160◦ ± 30◦).
Hydrogen bond restraints were used to restrain all
experimentally determined base pairs, with two distance restraints
per hydrogen bond.
The previously reported structure was refined with
Xplor-NIH version 2.9.8 (60). The structure was
recalculated here using Xplor-NIH 2.42, with a modified van
der Waal radius of 1.10 (up from 0.9) (61). An initial 100
structures were calculated starting from an extended single
strand, using NOE, hydrogen bond, weak planarity (weight
= 300.0 for individual base and weight = 6.0 for base pairs),
and dihedral angle restraints (21,60). This was followed by
a second round of NOE refinement with a lower starting
temperature (1000 K) and more cooling steps (40 000). The
next step of refinement incorporates the 79 RDCs, in which
a grid search produced optimal values for the axial (Da)
and rhombic (Dr) components of the alignment tensor: Da
= −39.0, Dr = 0.13 (62). In the last step, the database
potentials are applied. The structural statistics for the lowest
10 (out of 100) energy structures are given in
Supplementary Table S1. The structures were viewed and analyzed with
MOLMOL, PYMOL and MolProbity (63).
Telomerase activity assays
Telomerase activity was measured by primer extension
assay as previously described (64). TERT was in vitro
translated in rabbit reticulocyte lysate (RRL) using the TNT
transcription/translation kit (Promega). The 10 l
translation reaction contained 0.4 l of PCR enhancer (0.5 M KCl,
12.5 mM Mg(OAc)2), 0.2 l of 1 mM methionine, 8 l RRL
mix and 150 ng TERT DNA plasmid (pCITE-TERT) (gift
from Kathy Collins), and was incubated at 30◦C for 1 h. The
DNA templates of TER and variants TER-C75A, -C75U,
-A80C, -G95U, -U73C, -U73-bp (A93G/U82C), -U73-T
(U73C/A93G/U82C), -U74C, -U74-bp (A94G/U81C),
U74-T (U74C/A94G/U81C), -A91G and -U74C/C75U
were made and the RNAs were transcribed and purified as
described above, except they were exchanged into water
instead of buffer for use in activity assays. 0.5–1.0 M of the
purified RNA was added to the RRL translated TERT (1
l volume RNA per 10 l RRL reaction) and incubated at
30◦C for 30 min for RNP reconstitution. p65 was purified
as described (65), and was added to 1 M during RNP
reconstitution when necessary.
A 20 l telomerase primer extension reaction contains
50 mM Tris–HCl pH 7.0 or 8.3, 1.25 mM MgCl2, 1 mM
TCEP, 1 M primer ((GT2G3)3), 100 M TTP, 9 M
nonradioactive dGTP, 0.4 l of ( -32P)dGTP at 3000 Ci/mmol
and 6 l of the reconstituted RNP. The reaction was
incubated at 30◦C for 60 min and terminated by addition of
quench buffer (1 mM Tris–HCl, pH 8.0, 0.5 M EDTA)
containing a 15-mer ( -32P)5 -end labeled RNA recovery
control (RC). The nucleic acid products and control were
recovered by phenol/chloroform extraction and ethanol
precipitation, and loaded on a 10% polyacrylamide sequencing
gel (19:1 polyacrylamide, 7 M urea, 1× TBE buffer). The
gel was electrophoresed at 50 W for 1 h, dried and exposed
overnight to a phosphor imaging screen. The screen was
scanned and analyzed with QuantumOne Software. The
relative activity was determined by normalizing the integrated
density of each lane relative to the RC, and comparing to
the WT (as 100%).
Model of a step in catalytic core assembly
For help in visualizing the proposed assembly pathway
(Figure 7) a model of an early step in assembly of the catalytic
core was built as follows. Tetrahymena TERT (TEN, TRBD,
RT and CTE), SL4 and p65 were modeled as described in
(12), and positioned using Chimera. SL1 and Stem 3alt were
generated by the online server RNAComposer (66). The
structure and relative position of SL2 and TRBD were
obtained from a recent crystal structure (PDB 5C9H) (67). To
generate the full-length TER model, the remaining
singlestranded regions of TER connecting the above subdomains
(SL1, SL2, PK and SL4) were modeled by Pymol and Coot
(68), and standard bond angles and lengths of backbone of
these nucleotides were optimized by using the ‘Regularize
Zone’ tool in Coot.
Folding of the Tetrahymena telomerase RNA pseudoknot
For structural studies of the isolated pseudoknot, an RNA
construct was designed from the WT TER sequence (nt 69–
100) with an additional two G’s added at the 5 end to
enhance in vitro transcription (Figure 1B). The in vitro
transcribed RNA includes a 3 hammerhead ribozyme sequence
which self-cleaves the RNA product precisely at A100,
ensuring 3 end homogeneity. TetPK has low stability
compared to the TER pseudoknots of human and K. lactis,
likely due to its short Stem A and Loop A. 1D imino and
2D TOCSY (H5–H6) data show that the pseudoknot
begins to unfold (or pre-melt) at temperatures greater than
10◦C or if the pH is increased to 8.0 (Supplementary Figure
S1). The pseudoknot is in equilibrium with alternate
conformations, which can include the hairpins of each stem and
single-stranded RNA, as seen to a lesser extent for human
TER PK, which is in equilibrium with a small amount of a
Stem 1 (P2b; equivalent to Stem A in Tetrahymena) hairpin
at 25◦C (27). The alternate conformations of tetPK are
evidenced by the appearance of additional TOCSY (H5–H6)
crosspeaks of increasing intensity as the temperature is
increased >10◦C or the pH above 6.3 (with 10 mM sodium
phosphate and 50 mM KCl) (Supplementary Figure S1A).
The additional crosspeaks arise from separate unique
structure(s) that are in slow exchange with the folded
pseudoknot. As temperature and pH increase, the alternate
conformations increase in population. Additional Mg2+ or KCl
does not have a significant effect on this equilibrium.
Increasing salt (50–200 mM KCl and 0–5 mM Mg2+) also did
not affect the pseudoknot fold, as evidenced by
insignificant changes in the 1D imino spectra (Supplementary
Figure S1B). For these reasons, the tetPK structure was solved
at 10◦C and pH 6.3, with 10 mM sodium phosphate and 50
The base pairing of tetPK was determined by analyzing
the imino region of 2D NOESY spectra and JNN-COSY
spectra, which detects hydrogen bonds (57,58). The stems
of the pseudoknot form the predicted Watson-Crick base
pairs (shown in Figure 1B), with the expected NOE
connectivities, including an NOE crosspeak between Stem A
and Stem B residues (G84H1 to U92H3, respectively)
indicating stacking of the two stems. In Stem B, A80 and G95
form an unusual base pair. G95 is in the syn conformation
and its Hoogsteen edge faces the Watson-Crick edge of A80.
A80N1 is protonated and hydrogen bonds with G95N7 to
form a cWH base pair (Neocles and Westhof
nomenclature (69)). Although the A80 imino resonance was not
observed (even at pH 5.0), there is substantial evidence that the
A80N1 is protonated. N1 protonation causes the associated
amino protons to shift significantly downfield and split
further apart (70). This is observed with the A80 amino
protons of tetPK (Supplementary Figure S2B). Similar
chemical shift changes were observed for the aminos of the
protonated cytosine in the C+•G-C base triple formed in the
K. lactis telomerase RNA pseudoknot (29,71). In addition,
A80C2 is upfield shifted, another characteristic of N1
adenine protonation (Supplementary Figure S2A) (72). G95
was identified as syn by the strong NOE between its H8
and H1 , which is stronger than its H8 to H2 NOE. In
addition, there is an NOE between A80H2 and G95H8,
which is a strong indication of the synG-A+ base pair. This
type of synG-A+ pair was previously observed in the crystal
structure of an RNA 16-mer duplex with G-A mismatches
(73). Of the two predicted U•A-U base triples (U73•A93–
U82 and U74•A94–U81), only U74•A94–U81 (Figure 2D)
could be directly confirmed by JNN-COSY. Surprisingly,
the loop C75 interacts with the G-A base pair to form a
unique C•G-A+ base triple (C75•synG95-A+80) (Figure
2E) (cWHcWW nomenclature in (69)). The synG95 has its
Watson-Crick edge in the major groove and can thus form
a canonical Watson-Crick base pair with the loop C75.
Solution NMR structure of tetPK reveals a compact fold
The solution NMR structure of tetPK was previously
calculated by simulated annealing (60) using Xplor-NIH
version 2.9.8 (12). Here, we recalculated the structure using
the updated Xplor-NIH version 2.42, with the previously
reported restraints (414 NOE, 171 dihedral angle and 79
RDC restraints). The new calculation improved the RMSD
(between lowest 10 energy structures) from 0.83 A˚ to 0.42 A˚ ,
resulting in a better defined structure (Figure 2B and
Supplementary Table S1). The new structure is slightly more
extended than the previous one, with an RMSD between
lowest energy structures of the old and new calculations of
1.5 A˚ . There were improvements in steric contacts, confor
mation and minimal energy (better fit to NOE and dihedral
data, Supplementary Table S1).
In tetPK, the two stems form a quasi-continuous
Aform helix, stacked on top of each other without any
significant bend. Loops A and B bind in the major groove
(of Stem B) and minor groove (of Stem A), respectively.
The pseudoknot is compact, with a major groove width of
11.4 A˚ (defined as the phosphate distance between i and
i+6 cross-strand residues). Loop A and Stem B interact to
form a small triplex, composed of consecutive U•A-U and
C•G-A+ triples (discussed above). A second U•A-U triple
(U73•A93–U82) was predicted based on phylogenetic
analysis (45). The U73 is positioned in the major groove, but in
most of the structures it is not within hydrogen bond
distance for a Hoogsteen base pair to A83. The U73 imino
resonance was not observed in any NMR spectra. In
addition, the non-exchangeable resonances for U73 exhibited
broader linewidths, indicative of dynamics, and the NOE
pattern expected for this base triple, in which U73 would
have crosspeaks to U92 and A93 (i.e. U73-H1 to U92-H8,
-H1 , -H2 and U73-H1 to A93-H8), was not present (15).
We conclude that this base triple is conformationally
dynamic in the free tetPK.
The adenine-rich Loop B interacts with the minor groove
of Stem A of tetPK, to form A91•C72-G84 and
A90•C71G85 triples. An adenine-rich Loop B is common in
vertebrate TER pseudoknots, as well as in ribosomal
frameshifting viral pseudoknots (74,75). In these RNAs, the loop
adenines form minor groove triples involving sugar 2 -OH
and base protons (15,75). In tetPK, A90 and A91 are
stacked on top of each other and pair with with C71 and
C72, respectively, to form identical A•C-G triples. The
adenine amino protons form two hydrogen bonds with the
cytosine carbonyl and 2 -OH (shown in Figure 2F). This
interaction is consistent with in vivo footprinting data which
indicated that the Loop B––CAAA residues were protected
(from dimethyl sulfate modification) (76).The human TER
pseudoknot forms similar minor groove triples with its Stem
1 and Loop 2 adenine residues, further signifying the
importance of these types of minor groove interactions on
pseudoknot structure (15).
Pseudoknot folding is important for telomerase activity
The loop-stem tertiary interactions of TER pseudoknots
have previously been shown to be important for
telomerase activity in humans and yeast (in vitro and in vivo)
(14,15,29,31). By direct primer extension assay, we tested
the potential importance of tertiary interactions found in
the tetPK structure. First, we examined in detail the novel
C75•G95-A+80 triple. In full-length TER, we made C75A,
C75U, G95U and A80C mutations. TER variants were
then reconstituted with TERT, which had been in vitro
transcribed/translated in rabbit reticulocyte lysate (RRL),
and assayed for telomerase activity (Figure 3B,
Supplementary Figure S3). Since tetPK folding and stability is pH
dependent due to the protonated A in the C•G-A+ triple, these
and other activity assays were done at pH 7.0 in addition to
the standard pH 8.3 (64,65).
For the substitutions in the C75•G95-A+80 base triple,
TER-C75A has the biggest effect on telomerase activity,
decreasing activity by ∼50% compared to WT. This
substitution would abolish the tertiary interaction (loop-stem C75–
G95 bp) that stabilizes the unusual Stem B synG95-A+80
bp. TER-G95U and TER-80C, which would be expected
to replace the synG95-A+80 bp with a U–A and G–C base
pair, respectively, have 92% and 78% activity, respectively.
For both of these substitutions, an alternative triple
(C•UA and C+•G-C, respectively) could form with the loop
nucleotide. TER-C75U has 78% of WT activity; U75 could
pair with G95 similar to C75. Interestingly, decreasing pH
to 7.0 from the usual activity assay conditions of pH 8.3
increases activity for each TER variant (13–37%) This
increase is the smallest for the TER variants whose tertiary
interactions are not pH dependent (C75A, A80C and C95U)
and largest for WT and C75U (Figure 3B, Supplementary
To investigate the effect of these mutations on
pseudoknot structure and stability, the C•G-A+ base triple
substitutions were made in the context of the isolated tetPK,
and 1D and 2D imino NMR spectra were acquired to
provide secondary structure information (Figure 3D,
Supplementary Figure S4). Comparison of these spectra with that
of WT reveals that only tetPK-C75A does not form a folded
pseudoknot structure. The C75A substitution disrupts
formation of Stem A, as indicated by the near disappearance
of Stem A U86, G85 and G84 imino resonances (Figure
3D, Supplementary Figure S4). This explains why this TER
variant has the lowest activity. The tetPK substitutions
with higher activity, tetPK-C75U, tetPK-G95U and
tetPKA80C, each form a pseudoknot with Stem A and Stem B as
shown in the 1D and 2D spectra, although based on the
increase in imino proton linewidths all are somewhat less
stable than the WT tetPK (Figure 3D, Supplementary Figure
S4). As discussed above, tetPK-C75U and tetPK-A80C can
potentially form U•G-A+ and C+•G-C base triples
respectively. There is evidence for a U•G-A+ base triple in
tetPKC75U, which folds into a pseudoknot with similar imino
resonance chemical shifts and NOE crosspeak patterns as
WT tetPK, with the same protonated adenine amino
resonances (Figure 3D, Supplementary Figure S4). We were
unable to confirm whether a C+•G-C base triple forms in
tetPK-A80C. TetPK-G95U replaces the relatively unstable
synG95-A+80 base pair with a Watson–Crick U-A base
pair, where U95 could form a possible tertiary interaction
with C75. In summary, all of the nucleotide substitutions for
the C•G-A+ triple except for C75A had only modest effects
on activity, since an alternative stem base pair and/or loop
interaction could form. The C75A mutation strongly
destabilized the pseudoknot and showed a concomitant decrease
We next examined the effects of nucleotide substitutions
in the U74•A94–U81 and predicted U73•A93–U82 triples
on telomerase activity. For each set of triples, base pair (AU
to G–C, referred to as -bp), loop (U to C), and
compensatory base triple (U•A-U to C+•G-C, referred to as -T)
substitutions were made. The compensatory triple requires
protonation at CN3, so is also expected to be pH
dependent. For the U74•A94–U81 triple, TER-U74C and
TERU74-bp (A94G/U81C) substitutions which disrupt the base
triple have activity levels of 45% and 25% respectively, while
the compensatory mutant U74C/A94G/U81C
(TER-U74T) restores activity to 102%. Replacing U•A-U with
C+•GC base triples has previously been shown to (partially)
restore activity in the human and yeast pseudoknots (15,29).
For the predicted U73•A93–U82 triple, TER-U73C and
TER-U73-bp (A93G/U82C) have a smaller but significant
decrease in activity levels to 71% and 60% respectively, and
TER-U73-T (U73C/A93G/U82C) restores activity to 98%.
The activity assay results on substitutions in the predicted
U73•A93–U82 triple indicate that at least in the context
of assembly with TERT this triple is formed although
perhaps without optimal hydrogen bond geometry. Finally, we
tested the effect of disrupting the consecutive U•A–U and
C•G–A+ triples by changing the loop nucleotides
(TERU74C/C75U). This decreases activity to 33%, compared to
45% for TER-U74C and 78% for TER-C75U.
Last, we tested the importance of the minor groove
interaction between Loop B A91 and Stem A base pair G84–
C72. Substitution of A91G, which should abolish the
tertiary interaction of A91 with the relatively unstable Stem
A, decreases activity to 73%. As seen for the C•G–A+ base
triple substitutions, all TER variants showed an increase in
activity at pH 7 compared to pH 8.3, with most showing an
increase of more than 50%.
Taken together, the activity assays and NMR data
provide a direct correlation between stable pseudoknot folding
and telomerase activity levels. Disruption of the base triples
by changing the stem base pair or loop nucleotide results in
a decrease in activity. Not surprisingly, the largest decrease
in activity for substitutions in a single triple is seen for the
central U74C•A94–U81 triple.
Telomerase activity was also assayed in the presence of
holoenzyme assembly protein p65. p65 has been shown to
rescue a number of TER mutations that affect assembly,
including mutations within the pseudoknot region (65). For
all of the nucleotide substitutions tested here, p65 increased
the activity to near WT levels (90–98% relative to WT with
p65) (Figure 3B, Supplementary Figure S3). This is
consistent with a role for the pseudoknot in assembly rather than
catalysis, as proposed based on the location of the
pseudoknot in the pseudoatomic model of the catalytic core in the
cryo-EM map (12).
Secondary structure of TER t/PK
As discussed above, previous studies have shown that in the
absence of TERT, the Tetrahymena pseudoknot does not
form in the context of full length TER (46,47). Based on
SHAPE data, an alternate stem–loop structure with base
pairing between template-TRE and pseudoknot residues
was proposed for free TER (Figure 4A) (44,46,47). mFOLD
predicted a similar alternate structure, but with a
significant difference in the apical stem loop of the alternate stem–
loop (nt 51–88) (Figure 4A). We therefore investigated the
folding of free TER by NMR. The t/PK resonances in full
length TER (nt 1–159) and t/PK alone (nt 5–107) have
similar NOE crosspeak patterns, indicating that SL4 does
not affect the structure of the t/PK (Supplementary Figure
S5). Addition of MgCl2 also had no apparent effect on the
conformation of the t/PK (Supplementary Figure S5). We
therefore focused our NMR studies on the t/PK alone (in
10mM phosphate, pH 6.3, 50 mM KCl), which has the best
spectra quality. RNA constructs with sequences of Stem 1,
Stem 2, and the predicted template-stem (nt 43–53:88–100)
were also made for comparison of their NMR spectra to
those of the t/PK (Figure 4A and B). Comparison of the
chemical shifts and crosspeak patterns in 2D imino NOESY
spectra of these constructs with those of t/PK shows that
Stem 1, Stem 2, and the template-stem form in the t/PK
(Figure 4C). The lack of the tetPK NOE crosspeak pattern
(spectra in Figure 4C compared to Supplementary Figure
S7A) and the presence of the template-stem that sequesters
pseudoknot residues, confirms that the pseudoknot does
not form within the free t/PK. There are additional
unassigned imino resonances in t/PK, which include a distinct
G-U base pair pattern, which must arise from the
alternative structure. Both the SHAPE and mFOLD models
predict that the TRE is paired in an alternate stem–loop
(TREstem); the exact secondary structures differ but both have an
identical central stem (nt 53–58:82–87) that contains a G-U
basepair. We made a construct of this central stem sequence
capped with a UUCG tetraloop (tet51-58:82–88) for
comparison to t/PK. Tet51-58:82–88 imino resonances were
assigned, which confirmed that the predicted stem forms
(Figure 4D, orange). Furthermore, comparison of the
spectra of tet51-58:82–88 and t/PK confirmed that the central
TRE-paired stem is present in the t/PK (Figure 4C,
orange). Finally, we made an RNA containing residues 51–
88 (TRE-stem) and sequential assignments were obtained
from analysis of 2D imino NOESY (Figure 4D, purple) and
D2O NOESY spectra (Supplementary Figure S6). The AH2
proton is the only non-exchangeable resonance with
crossstrand NOEs: (weak NOE) to the H1 of its base-paired
residue and (strong NOE) to the H1 of its 3 (or i – 1)
cross-strand neighbor, depicted in Supplementary Figure
S6. These NOEs were used to validate assignments and
determine TRE-stem secondary structure, which is shown in
Figures 4B and 5A. This secondary structure is a long
helix, which contains a single bulge A(80) base, a U–U and C–
C base pair, capped by a hexaloop. While the central stem
(nt 53–58:82–87) is the same as the SHAPE and mFOLD
models, the rest of the helix lacks the extensive bulges
predicted by these two methods. Comparison of chemical shifts
and NOE crosspeak patterns in the imino NOESY of
TREstem with t/PK confirmed that the secondary structure of
TRE-stem also forms in t/PK. However, there are a few
additional broad imino resonances in t/PK that suggest that
other alternate conformations may be present.
For the NMR studies described above, the RNA was
heated and snap cooled to achieve a homogeneous well
folded sample. However, it is possible that TER could fold
into a different native structure during transcription. To
test for this possibility, we also purified t/PK under
nondenaturing conditions, using ion exchange and size
exclusion column chromatography. NMR spectra of t/PK
purified under denaturing and non-denaturing conditions were
essentially identical, indicating that co-transcriptional
folding does not trap a different structure than the one
obtained after heating and snap-cooling (Supplementary
Figure S5). We also investigated a construct that spans the
template, TRE and pseudoknot regions (nts 42–100; Stem
3alt), which comprises the residues that form the
alternative stem–loop, but opens the t/PK circle. Interestingly, the
NMR data shows that for this construct the pseudoknot
is present in ∼1:1 equilibrium with the alternate
conformation (Supplementary Figure S7). Apparently the
topological constraints of the closed TER t/PK circle also
influence the structural equilibrium, disfavoring the
pseudoknot. The identification of an alternate stem–loop
structure, in which the template and the TRE are paired with
residues from the pseudoknot sequence, explains why the
pseudoknot does not form in the context of the full length
TER in the absence of TERT. Phylogenetic analysis of the
t/PK from 12 Tetrahymena species shows that each species
can form comparable template- and TRE-stems (Figure 5).
Taken together, these results suggest that not only the folded
pseudoknot but also the Stem 3alt structure in the free TER
may be biologically relevant.
Comparison of ciliate, human, and yeast pseudoknots
The Tetrahymena TER pseudoknot has a number of
similarities and differences to those of human (hPK) and yeast
K. lactis (kPK), which were previously solved by NMR
(Figure 6) (15,21,29). All three pseudoknots have
continuous helical stacking interactions through the stems and
junction. Only hPK has an intervening loop-loop
interaction at the junction, while tetPK and kPK have stem-stem
stacking. All three pseudoknots have a stabilizing triple
helix that includes major groove tertiary interactions, and
both tetPK and hPK also have minor groove base triples.
kPK cannot form the typical minor groove interactions due
to the lack of adenines in its second loop, but has a much
longer major groove triplex. The base triples were shown
to be important for proper pseudoknot folding for all three
pseudoknots. In both tetPK and kPK, mutations in the loop
residues that disrupt major groove base triples abolished
formation of Stem A (Tetrahymena) and the equivalent stem
(Stem 1) in yeast (29). In hPK, mutations in loop residues
that disrupt major groove base triples destabilized the base
pairs in both stems near the junction (77). The stems of TER
pseudoknots have varying degrees of stability depending on
sequence, bulges, G-C content and length, but all require the
base triple interactions to ensure proper folding.
TetPK has a significantly different tertiary structure when
compared to hPK and kPK (Figure 6B). The backbone
of hPK and kPK overlay very well and the structures
appear remarkably similar despite differences in sequence and
tertiary interactions. TetPK is more compact. This may
be partly due to the different ratios in the number of
nucleotides in Loop A (human J2b/3) to Stem B (human P3).
hPK has a ratio of ∼1:1, kPK is 1:2, and tetPK is closer
to ∼1:3 (Loop A: Stem B nt ratio) (Figure 6A). TetPK has
eight base pairs in Stem B and only three Loop A nts, while
hPK has nine stem base pairs and eight loop nts. This means
that tetPK Loop A must span a greater distance per
nucleotide and is consistent with a more compact molecule.
These pseudoknots belong to different organisms with
significantly different TER size, sequence and structure, and to
a lesser extent different TERT sequence and structure, and
therefore their similarities may reflect a conserved function
while their differences reflect their diverse environments.
One study investigated a chimeric human TER, where hPK
was replaced with tetPK, which could potentially
compensate for function since the secondary structures of the two
pseudoknots are similar (Figure 6A) (78). However TERT
assembled with the chimeric TER was only ‘weakly
active’ when assayed in vitro, supporting the conclusion that
the pseudoknot differences reflect species-specific
Correlation between pseudoknot structure, p65, and
Previous studies of ciliate, yeast, and human telomerase
RNA have shown that formation of the pseudoknot and
its tertiary interactions are essential for activity in the
context of the minimal telomerase RNP (i.e. TERT + TER
only) (14,15,33,65,79–81). It has also been shown that for
Tetrahymena p65 can rescue mutations that affect
assembly, if they are not too severe, but not mutations that affect
catalysis (42,65,82). In the cryo-EM model of the
Tetrahymena telomerase holoenzyme, the pseudoknot is on the
opposite side of the TERT ring from the template (active site),
too far away to be directly involved in catalysis. Thus, it
has been proposed that the pseudoknot has a role in
assembly, with the tertiary interactions stabilizing the correct fold
on TERT (12). Consistent with this, all of the pseudoknot
substitutions reported here, which disrupt the U•A-U and
C•G-A+ base triples, were rescued by p65, in the context of
assembly of TER with TERT in RRL.
Some early studies seemed to indicate that the
pseudoknot was not essential for telomerase activity in
Tetrahymena, since mutations which were presumed to abolish
pseudoknot structure, such as deleting nt 86–89 or adding
a 4nt (CAAU) bulge at position 81–82 did not significantly
affect in vitro or in vivo activity (assayed in the presence of
p65) (83). However, the structural analysis presented here
and in yeast TER pseudoknots (29) suggest that the
pseudoknot might still form with these nucleotide changes albeit
less stably. TER- 86-89 would disrupt the terminal A-U
base pairs of Stem A, but would leave the G–C base pairs at
the junction and the tertiary interactions which stabilize the
pseudoknot intact (14,21). Adding a bulge in Stem B would
not necessarily abolish tetPK formation, since all secondary
and tertiary interactions remain intact. A similar bulge is
predicted in the S. cerevisiae pseudoknot (29,84).
Interestingly, when pseudoknot formation was indeed abolished, by
deleting Stem B, telomerase activity only decreased to 63%
(85), in assays where p65 was presumably present. This
deletion would prevent formation of Stem 3alt and would result
in a smaller t/PK circle. Activity in this mutant might be
explained by the assembly pathway postulated below, since
deletion of Stem B would decrease the size of the t/PK
circle, similar to that of the folded PK and allow assembly to
proceed up to the last step.
The Stem 3alt and pseudoknot structures explain FRET and
chemical probing data
The structure of Stem 3alt is completely consistent with the
identification of flexible residues at the apical loop and near
the bulge A80 in the free TER by SHAPE chemical probing
(Supplementary Figure S8A) (47). The SHAPE data for the
pseudoknot in the presence of TERT indicates that a
significant portion of Stem B is flexible (Supplementary Figure
S8A). The SHAPE data was acquired at pH 8, which we
have shown destabilizes the pseudoknot due to the pH
dependence of the tertiary C•G–A+ base triple, and in the
absence of p65, which assembles more active complexes. These
conditions could explain the observed ‘flexibility’ of Stem
B, since some proportion of the complexes might have
partially or misfolded pseudoknots. Telomerase activity assays
showed lower activity at pH 8 versus pH 7 and two-fold
lower activity in the absence (versus presence) of p65.
Previous FRET studies showed that the pseudoknot does
not form in the protein-free Tetrahymena TER and
suggested that instead of a Stem 3B hairpin, an alternative
structure forms (46). In the Stem 3alt structure the FRET
dyes would be farther apart than they would be if a Stem
B hairpin forms, consistent with the FRET results
(Supplementary Figure S8A). FRET studies of the much larger
human t/PK also indicated that the pseudoknot was not
fully folded in the free TER (77). Examination of human
TER sequence suggests that it can form a similar alternate
structure with P2 flipping out to mimic the TRE-stem and
the template base pairing with pseudoknot residues
(Supplementary Figure S8B). This alternate structure, in which
the FRET dye-labeled pseudoknot residues (C92 and U85)
would be far from each other (Supplementary Figure S8B),
would explain the FRET data on human TER and TERT
(Supplementary Figure S8B) (77).
A model for assembly of the catalytic core
Although the Tetrahymena pseudoknot does not form in
TERT-free TER, it is fully folded in the pseudoatomic
model of Tetrahymena telomerase from the cryo-EM map
(12). We note in this context that, in the absence of TERT, it
was only by studying the isolated tetPK that the pseudoknot
structure could be determined. The cryo-EM structure is of
the apo enzyme, without telomeric DNA. Single-molecule
FRET experiments where the conformation of the
pseudoknot was monitored during catalysis showed that only RNP
molecules containing a properly folded pseudoknot were
catalytically active and that the pseudoknot does not unfold
during catalysis (46).
Human cells contain pools of free hTER and hTERT
which assemble into the telomerase RNP as necessary
(86). Telomerase RNP assembly begins with the
cotranscriptional binding of species-specific accessory
proteins, i.e. H/ACA proteins in vertebrates and p65 in
ciliates (1,87), which bind TER regions outside the t/PK, and
would not be expected to affect the folding of the
(TERTfree) t/PK. The structure of free TER potentially plays
an important role in telomerase biogenesis. The Stem 3alt
structure in Tetrahymena t/PK has more base pairs and is
more stable than the ‘open circle’ structure with a
singlestranded template-TRE and PK. The double-helical RNA
would better protect TER from degradation since
singlestranded RNA is more likely to self-cleave due to the more
accessible 2 OH groups. Base-paired RNA is also more
protected from mutations, since single-stranded cytosines are
deaminated (converted to U) at a much higher rate (88).
This is especially critical in the template region, where
mutations would be propagated to the telomere repeat (47).
Aside from stabilizing free TER, the alternate structure
may play a role in assembly, given that it appears to be
conserved to base pair. In the assembled catalytic core, the t/PK
encircles the TERT ring approximately perpendicular to the
plane of the ring. The free t/PK circle, with the Stem 3alt,
is too small to allow entry of the TERT ring and also
sequesters the template. However, the TBE is still exposed.
Based on the secondary structure of the free t/PK, we can
speculate on a step-wise pathway for p65-TER-TERT
pathway (depicted in Figure 7): (Step 1) First, p65 binds to two
sites in SL4: the p65 La-RNA recognition motif (LaRRM,
or La module) binds the 3 poly U tail (40,42) and p65
Cterminal xRRM binds Stem 4, bending it so that Loop 4 is
closer to SL2 (40–43). (Steps 2, 3) Next, the TRBD binds the
high affinity binding site TBE, which is still accessible in the
Stem 3alt structure. Binding of the TBE to TERT also
positions Loop 4 (if bound to p65) close to the TRBD. Loop
4 interacts at the interface of the TRBD and CTE, where
it is proposed to stabilize the closed TERT ring (12,89,90).
Complete binding to the TRBD requires opening of the
unstable template-stem. (Step 4) The template and RT interact
due to their affinity and proper positioning. Further entry
of RT-TRBD into the t/PK circle facilitates the opening of
the rest of Stem 3alt (TRE-stem). (Step 5) The residues of
the open circle re-fold to form stem–loop 3B of the
pseudoknot. (Step 6) Stem 3A and the pseudoknot tertiary
interactions form (stabilized by TERT). The pseudoknot acts like
a ‘watchband ratchet clasp,’ decreasing the size of the TER
t/PK circle around the TERT ring, and locking the RNP
complex into place. We note that the TER circle passes
between the TERT ring and the TEN domain, which is
connected to the TRBD by a long linker. Telomerase activity
can be reconstituted with TEN added in trans to the TERT
ring and TER (91). It is therefore likely that the TEN
domain stacks over the CTE only after step 6, or even after
association of TERT with the holoenzyme protein p50 (39).
The TEN domain interacts directly with p50, where p50
serves as a central hub connecting the RNP catalytic core to
other accessory proteins to promote processivity (12,39,92).
In this speculative model of the catalytic core assembly,
TERT enters the TER circle from the ‘bottom’, i.e. on the
same side as SL4 (as viewed in Figure 7). TERT could
possibly also enter from the top, but entry from the bottom
would be facilitated by p65, supporting its role in assembly,
since by bringing SL4 closer to SL2, it would help bring the
TERT ring to the TER circle. Either way, the determined
secondary structure of the t/PK in the absence of TERT
and, from the cryo-EM model, the t/PK structure in the
presence of p65 and TERT, provides a working model for
thinking about assembly of the catalytic core. Furthermore,
this model could potentially explain why p65 can ‘rescue’
(or mask) pseudoknot destabilizing substitutions in vitro,
since both components work in tandem during TERT-TER
assembly. In the absence of p65, assembly of the TER
circle around TERT and insertion of Loop 4 at the
TRBDCTE interface would occur independently of each other,
since stem Loop 4 would point away from TERT without
bound p65. Thus, the chance of properly assembling TER
with TERT would be decreased. In the presence of p65, once
the TER circle assembles around the TERT ring, even a
destabilized pseudoknot might be able to fold on TERT.
In summary, this work highlights the important
conserved features of TER pseudoknots, in particular base
triples that stabilize the fold. Examination of the sequences
of Tetrahymena and human TER t/PK suggests that an
alternative structure, in which the pseudoknot nucleotides are
paired with other regions of the t/PK in the TERT-free
TER, may also be a common feature of TERs and play a
role in assembly.
Coordinates and restraints for the 10 lowest energy
structures of the recalculated tetPK have been deposited in the
Protein Data Bank with accession code 5KMZ (supercedes
Supplementary Data are available at NAR Online.
The authors thank Dr Yaqiang Wang for making the
assembly model structure.
National Institutes of Health [GM048123]; National
Science Foundation [MCB1022379 to J.F.]; Ruth L.
Kirschstein NRSA pre-doctoral training grant [GM007185
fellowship to D.D.C.]; NMR core facility by Department
of Energy [DE-FC0302ER63421]. Funding for open access
charge: National Institutes of Health [GM048123].
Conflict of interest statement. None declared.
1. Collins , K. ( 2006 ) The biogenesis and regulation of telomerase holoenzymes . Nat. Rev. Mol. Cell. Biol ., 7 , 484 - 494 .
2. Harrington , L. ( 2003 ) Biochemical aspects of telomerase function . Cancer Lett. , 194 , 139 - 154 .
3. Blackburn , E.H. ( 2000 ) The end of the (DNA) line . Nat. Struct. Biol ., 7 , 847 - 850 .
4. Nandakumar , J. and Cech , T.R. ( 2013 ) Finding the end: recruitment of telomerase to telomeres . Nat. Rev. Mol. Cell. Biol ., 14 , 69 - 82 .
5. Blackburn , E.H. ( 2001 ) Switching and signaling at the telomere . Cell , 106 , 661 - 673 .
6. Blackburn , E.H. and Collins , K. ( 2011 ) Telomerase: an RNP enzyme synthesizes DNA . Cold Spring Harb. Perspect. Biol ., 3 , a003558 .
7. O'Sullivan , R.J. and Karlseder , J. ( 1994 ) Telomeres: protecting chromosomes against genome instability . Nat. Rev. Mol. Cell. Biol ., 11 , 171 - 181 .
8. Mason , M. , Schuller , A. and Skordalakes , E. ( 2010 ) Telomerase structure function . Curr. Opin. Struct. Biol ., 21 , 92 - 100 .
9. Kachouri-Lafond , R. , Dujon , B. , Gilson , E. , Westhof , E. , Fairhead , C. and Teixeira , M.T. ( 2009 ) Large telomerase RNA, telomere length heterogeneity and escape from senescence in Candida glabrata . FEBS Lett. , 583 , 3605 - 3610 .
10. Egan , E.D. and Collins , K. ( 2012 ) Biogenesis of telomerase ribonucleoproteins . RNA , 18 , 1747 - 1759 .
11. Chen , J.L. and Greider , C.W. ( 2005 ) Functional analysis of the pseudoknot structure in human telomerase RNA . Proc. Natl. Acad. Sci. U.S.A. , 102 , 8080 - 8085 .
12. Jiang ,J., Chan ,H., Cash , D.D. , Miracco , E.J. , Ogorzalek Loo , R.R. , Upton , H.E. , Cascio , D. , O'Brien Johnson ,R., Collins , K. , Loo , J.A. et al. ( 2015 ) Structure of Tetrahymena telomerase reveals previously unknown subunits , functions, and interactions. Science , 350 , aab4070.
13. Ly , H. , Blackburn , E.H. and Parslow , T.G. ( 2003 ) Comprehensive structure-function analysis of the core domain of human telomerase RNA . Mol . Cell . Biol., 23 , 6849 - 6856 .
14. Qiao , F. and Cech , T.R. ( 2008 ) Triple-helix structure in telomerase RNA contributes to catalysis . Nat. Struct. Mol. Biol ., 15 , 634 - 640 .
15. Theimer ,C.A., Blois , C.A. and Feigon , J. ( 2005 ) Structure of the human telomerase RNA pseudoknot reveals conserved tertiary interactions essential for function . Mol. Cell , 17 , 671 - 682 .
16. Chen , J.L. and Greider , C.W. ( 2003 ) Template boundary definition in mammalian telomerase . Cancer Lett. , 17 , 2747 - 2752 .
17. Lai , C.K. , Miller , M.C. and Collins , K. ( 2002 ) Template boundary definition in Tetrahymena telomerase . Genes Dev. , 16 , 415 - 420 .
18. Miller , M.C. and Collins , K. ( 2002 ) Telomerase recognizes its template by using an adjacent RNA motif . Proc. Natl. Acad. Sci. U.S.A. , 99 , 6585 - 6590 .
19. Autexier , C. , Pruzan , R. , Funk , W.D. and Greider , C.W. ( 1996 ) Reconstitution of human telomerase activity and identification of a minimal functional region of the human telomerase RNA . EMBO J., 15 , 5928 - 5935 .
20. Kim , N.K. , Zhang , Q. and Feigon , J. ( 2014 ) Structure and sequence elements of the CR4/5 domain of medaka telomerase RNA important for telomerase function . Nucleic Acids Res ., 42 , 3395 - 3408 .
21. Kim , N.K. , Zhang , Q. , Zhou , J. , Theimer , C.A. , Peterson , R.D. and Feigon , J. ( 2008 ) Solution structure and dynamics of the wild-type pseudoknot of human telomerase RNA . J. Mol. Biol ., 384 , 1249 - 1261 .
22. Chen , Y. , Fender , J. , Legassie , J.D. , Jarstfer , M.B. , Bryan, T.M. and Varani , G. ( 2006 ) Structure of stem-loop IV of Tetrahymena telomerase RNA . EMBO J., 25 , 3156 - 3166 .
23. Leeper , T. , Leulliot , N. and Varani , G. ( 2003 ) The solution structure of an essential stem-loop of human telomerase RNA . Nucleic Acids Res ., 31 , 2614 - 2621 .
24. Leeper , T.C. and Varani , G. ( 2005 ) The structure of an enzyme-activating fragment of human telomerase RNA . RNA, 11 , 394 - 403 .
25. Richards , R.J. , Theimer , C.A. , Finger , L.D. and Feigon , J. ( 2006 ) Structure of the Tetrahymena thermophila telomerase RNA helix II template boundary element . Nucleic Acids Res . 34 , 816 - 825 .
26. Richards , R.J. , Wu , H. , Trantirek , L. , O'Connor , C.M. , Collins , K. and Feigon , J. ( 2006 ) Structural study of elements of Tetrahymena telomerase RNA stem-loop IV domain important for function . RNA , 12 , 1475 - 1485 .
27. Theimer ,C.A., Finger , L.D. , Trantirek , L. and Feigon , J. ( 2003 ) Mutations linked to dyskeratosis congenita cause changes in the structural equilibrium in telomerase RNA . Proc. Natl. Acad. Sci. U.S.A. , 100 , 449 - 454 .
28. Theimer ,C.A., Jady , B.E. , Chim , N. , Richard , P. , Breece , K.E. , Kiss , T. and Feigon , J. ( 2007 ) Structural and functional characterization of human telomerase RNA processing and Cajal body localization signals . Mol. Cell , 27 , 869 - 881 .
29. Cash , D.D. , Cohen-Zontag , O. , Kim , N.K. , Shefer , K. , Brown , Y. , Ulyanov , N.B. , Tzfati , Y. and Feigon , J. ( 2013 ) Pyrimidine motif triple helix in the Kluyveromyces lactis telomerase RNA pseudoknot is essential for function in vivo . Proc. Natl. Acad. Sci. U.S.A. , 110 , 10970 - 10975 .
30. Shefer , K. , Brown , Y. , Gorkovoy , V. , Nussbaum , T. , Ulyanov , N.B. and Tzfati , Y. ( 2007 ) A triple helix within a pseudoknot is a conserved and essential element of telomerase RNA . Mol . Cell . Biol., 27 , 2130 - 2143 .
31. Tzfati , Y. , Knight , Z. , Roy , J. and Blackburn , E.H. ( 2003 ) A novel pseudoknot element is essential for the action of a yeast telomerase . Genes Dev. , 17 , 1779 - 1788 .
32. Gilley , D. and Blackburn , E.H. ( 1999 ) The telomerase RNA pseudoknot is critical for the stable assembly of a catalytically active ribonucleoprotein . Proc. Natl. Acad. Sci. U.S.A. , 96 , 6621 - 6625 .
33. Lai , C.K. , Miller , M.C. and Collins , K. ( 2003 ) Roles for RNA in telomerase nucleotide and repeat addition processivity . Mol. Cell , 11 , 1673 - 1683 .
34. ten Dam , E. , van Belkum , A. and Pleij , K. ( 1991 ) A conserved pseudoknot in telomerase RNA . Nucleic Acids Res ., 19 , 6951 .
35. Blackburn , E.H. , Greider , C.W. and Szostak , J.W. ( 2006 ) Telomeres and telomerase: the path from maize, Tetrahymena and yeast to human cancer and aging . Nat. Med. , 12 , 1133 - 1138 .
36. Greider , C.W. and Blackburn , E.H. ( 1985 ) Identification of a specific telomere terminal transferase activity in Tetrahymena extracts . Cell , 43 , 405 - 413 .
37. Romero , D.P. and Blackburn , E.H. ( 1991 ) A conserved secondary structure for telomerase RNA . Cell , 67 , 343 - 353 .
38. Ye , A.J. and Romero , D.P. ( 2002 ) Phylogenetic relationships amongst tetrahymenine ciliates inferred by a comparison of telomerase RNAs . Int. J. Syst. Evol. Microbiol ., 52 , 2297 - 2302 .
39. Jiang , J. , Miracco , E.J. , Hong , K. , Eckert , B. , Chan,H., Cash , D.D. , Min , B. , Zhou , Z.H. , Collins , K. and Feigon , J. ( 2013 ) The architecture of Tetrahymena telomerase holoenzyme . Nature , 496 , 187 - 192 .
40. Singh , M. , Wang , Z. , Koo , B.K. , Patel , A. , Cascio , D. , Collins , K. and Feigon , J. ( 2012 ) Structural basis for telomerase RNA recognition and RNP assembly by the holoenzyme La family protein p65 . Mol. Cell , 47 , 16 - 26 .
41. Akiyama , B.M. , Loper , J. , Najarro , K. and Stone , M.D. ( 2012 ) The C-terminal domain of Tetrahymena thermophila telomerase holoenzyme protein p65 induces multiple structural changes in telomerase RNA . RNA, 18 , 653 - 660 .
42. O'Connor , C.M. and Collins , K. ( 2006 ) A novel RNA binding domain in tetrahymena telomerase p65 initiates hierarchical assembly of telomerase holoenzyme . Mol. Cell. Biol ., 26 , 2029 - 2036 .
43. Stone , M.D. , Mihalusova , M. , O'Connor , C.M. , Prathapam , R. , Collins , K. and Zhuang , X.W. ( 2007 ) Stepwise protein-mediated RNA folding directs assembly of telomerase ribonucleoprotein . Nature , 446 , 458 - 461 .
44. Bhattacharyya , A. and Blackburn , E.H. ( 1994 ) Architecture of telomerase RNA . EMBO J., 13 , 5721 - 5731 .
45. Ulyanov , N.B. , Shefer , K. , James , T.L. and Tzfati , Y. ( 2007 ) Pseudoknot structures with conserved base triples in telomerase RNAs of ciliates . Nucleic Acids Res ., 35 , 6150 - 6160 .
46. Mihalusova , M. , Wu , J.Y. and Zhuang , X. ( 2011 ) Functional importance of telomerase pseudoknot revealed by single-molecule analysis . Proc. Natl. Acad. Sci. U.S.A. , 108 , 20339 - 20344 .
47. Cole , D.I. , Legassie , J.D. , Bonifacio , L.N. , Sekaran , V.G. , Ding , F. , Dokholyan , N.V. and Jarstfer , M.B. ( 2012 ) New models of Tetrahymena telomerase RNA from experimentally derived constraints and modeling . J. Am. Chem. Soc. , 134 , 20070 - 20080 .
48. Batey ,R.T., Inada , M. , Kujawinski , E. , Puglisi , J.D. and Williamson , J.R. ( 1992 ) Preparation of isotopically labeled ribonucleotides for multidimensional NMR spectroscopy of RNA . Nucleic Acids Res ., 20 , 4515 - 4523 .
49. Dieckmann , T. and Feigon , J. ( 1997 ) Assignment methodology for larger RNA oligonucleotides: application to an ATP-binding RNA aptamer . J. Biomol. NMR , 9 , 259 - 272 .
50. Pley , H.W. , Flaherty , K.M. and McKay , D.B. ( 1994 ) Three-dimensional structure of a hammerhead ribozyme . Nature , 372 , 68 - 74 .
51. Scott , W.G. , Finch , J.T. and Klug , A. ( 1995 ) The crystal structure of an all-RNA hammerhead ribozyme . Nucleic Acids Symp. Ser. , 214 - 216 .
52. Gullerez , J. , Lopez , P.J. , Proux , F. , Launay , H. and Dreyfus , M. ( 2005 ) A mutation in T7 RNA polymerase that facilitates promoter clearance . Proc. Natl. Acad. Sci. U.S.A. , 102 , 5958 - 5963 .
53. Petrov , A. , Wu , T. , Puglisi , E.V. and Puglisi , J.D. ( 2013 ) RNA purification by preparative polyacrylamide gel electrophoresis . Methods Enzymol. , 530 , 315 - 330 .
54. Cromsigt , J.A. , Hilbers , C.W. and Wijmenga , S.S. ( 2001 ) Prediction of proton chemical shifts in RNA. Their use in structure refinement and validation . J. Biomol. NMR , 21 , 11 - 29 .
55. Peterson , R.D. , Theimer , C.A. , Wu , H. and Feigon , J. ( 2004 ) New applications of 2D filtered/edited NOESY for assignment and structure elucidation of RNA and RNA-protein complexes . J. Biomol. NMR , 28 , 59 - 67 .
56. Wu , H. , Yang , P.K. , Butcher , S.E. , Kang , S. , Chanfreau , G. and Feigon , J. ( 2001 ) A novel family of RNA tetraloop structure forms the recognition site for Saccharomyces cerevisiae RNase III . EMBO J., 20 , 7240 - 7249 .
57. Dingley , A.J. and Grzesiek , S. ( 1998 ) Direct observation of hydrogen bonds in nucleic acid base pairs by internucleotide (2)J(NN) couplings . J. Am. Chem. Soc. , 120 , 8293 - 8297 .
58. Dingley , A.J. , Nisius , L. , Cordier , F. and Grzesiek , S. ( 2008 ) Direct detection of N-H( ...) N hydrogen bonds in biomolecules by NMR spectroscopy . Nat. Protoc. , 3 , 242 - 248 .
59. Hansen , M.R. , Mueller , L. and Pardi , A. ( 1998 ) Tunable alignment of macromolecules by filamentous phage yields dipolar coupling interactions . Nat. Struct. Biol ., 5 , 1065 - 1074 .
60. Schwieters , C.D. , Kuszewski , J.J. , Tjandra , N. and Clore , G.M. ( 2003 ) The Xplor-NIH NMR molecular structure determination package . J. Magn. Reson. , 160 , 65 - 73 .
61. Bermejo , G.A. , Clore , G.M. and Schwieters , C.D. ( 2016 ) Improving NMR structures of RNA . Structure, 24 , 806 - 815 .
62. de Alba , E. and Tjandra , N. ( 2002 ) NMR dipolar couplings for the structure determination of biopolymers in solution . Progr. Nuclear Magn. Reson. Spectrosc. , 40 , 175 - 197 .
63. Davis , I.W. , Leaver-Fay , A. , Chen , V.B. , Block , J.N. , Kapral , G.J. , Wang , X. , Murray , L.W. , Arendall ,W.B. 3rd , Snoeyink, J. , Richardson , J.S. et al. ( 2007 ) MolProbity: all-atom contacts and structure validation for proteins and nucleic acids . Nucleic Acids Res ., 35 , W375 - W383 .
64. Bryan , T.M. , Goodrich , K.J. and Cech , T.R. ( 2000 ) A mutant of Tetrahymena telomerase reverse transcriptase with increased processivity . J. Biol. Chem. , 275 , 24199 - 24207 .
65. Berman , A.J. , Gooding , A.R . and Cech , T.R. ( 2010 ) Tetrahymena telomerase protein p65 induces conformational changes throughout telomerase RNA (TER) and rescues telomerase reverse transcriptase and TER assembly mutants . Mol. Cell. Biol ., 30 , 4965 - 4976 .
66. Popenda , M. , Szachniuk , M. , Antczak , M. , Purzycka , K.J. , Lukasiak , P. , Bartol , N. , Blazewicz , J. and Adamiak , R.W. ( 2012 ) Automated 3D structure composition for large RNAs . Nucleic Acids Res ., 40 , e112.
67. Jansson , L.I. , Akiyama , B.M. , Ooms , A. , Lu , C. , Rubin , S.M. and Stone , M.D. Structural basis of template-boundary definition in Tetrahymena telomerase . Nat. Struct. Mol. Biol ., 22 , 883 - 888 .
68. Emsley , P. and Cowtan , K. ( 2004 ) Coot: model-building tools for molecular graphics . Acta Crystallogr. D Biol . Crystallogr., 60 , 2126 - 2132 .
69. Abu Almakarem , A.S. , Petrov , A.I. , Stombaugh , J. , Zirbel , C.L. and Leontis , N.B. ( 2012 ) Comprehensive survey and geometric classification of base triples in RNA structures . Nucleic Acids Res ., 40 , 1407 - 1423 .
70. Kazantsev , A.V. and Pace , N.R. ( 1998 ) Identification by modification-interference of purine N-7 and ribose 2 -OH groups critical for catalysis by bacterial ribonuclease P . RNA, 4 , 937 - 947 .
71. Sklenar , V. and Feigon , J. ( 1990 ) Formation of a stable triplex from a single DNA strand . Nature , 345 , 836 - 838 .
72. Legault , P. and Pardi , A. ( 1994 ) 31P chemical shift as a probe of structural motifs in RNA . J. Magn. Reson., 103 , 82 - 86 .
73. Pan , B. , Mitra , S.N. and Sundaralingam , M. ( 1999 ) Crystal structure of an RNA 16-mer duplex R(GCAGAGUUAAAUCUGC)2 with nonadjacent G(syn) . A+(anti) mispairs. Biochemistry , 38 , 2826 - 2831 .
74. Chen , J.L. , Blasco , M.A. and Greider , C.W. ( 2000 ) Secondary structure of vertebrate telomerase RNA . Cell , 100 , 503 - 514 .
75. Su , L. , Chen , L. , Egli , M. , Berger , J.M. and Rich , A. ( 1999 ) Minor groove RNA triplex in the crystal structure of a ribosomal frameshifting viral pseudoknot . Nat. Struct. Biol ., 6 , 285 - 292 .
76. Zaug , A.J. and Cech , T.R. ( 1995 ) Analysis of the structure of Tetrahymena nuclear RNAs in vivo: telomerase RNA, the self-splicing rRNA intron, and U2 snRNA . RNA, 1 , 363 - 374 .
77. Hengesbach , M. , Kim , N.K. , Feigon , J. and Stone , M.D. ( 2012 ) Single-molecule FRET reveals the folding dynamics of the human telomerase RNA pseudoknot domain . Angew. Chem. Int. Ed. Engl., 51 , 5876 - 5879 .
78. Marie-Egyptienne , D.T. , Cerone , M.A. , Londono-Vallejo , J.A. and Autexier , C. ( 2005 ) A human-Tetrahymena pseudoknot chimeric telomerase RNA reconstitutes a nonprocessive enzyme in vitro that is defective in telomere elongation . Nucleic Acids Res ., 33 , 5446 - 5457 .
79. Licht , J.D. and Collins , K. ( 1999 ) Telomerase RNA function in recombinant Tetrahymena telomerase . Genes Dev. , 13 , 1116 - 1125 .
80. Sperger , J.M. and Cech , T.R. ( 2001 ) A stem-loop of Tetrahymena telomerase RNA distant from the template potentiates RNA folding and telomerase activity . Biochemistry , 40 , 7005 - 7016 .
81. Lin , J. , Ly , H. , Hussain , A. , Abraham , M. , Pearl , S. , Tzfati , Y. , Parslow, T.G. and Blackburn , E.H. ( 2004 ) A universal telomerase RNA core structure includes structured motifs required for binding the telomerase reverse transcriptase protein . Proc. Natl. Acad. Sci. U.S.A. , 101 , 14713 - 14718 .
82. Prathapam , R. , Witkin , K.L. , O'Connor , C.M. and Collins , K. ( 2005 ) A telomerase holoenzyme protein enhances telomerase RNA assembly with telomerase reverse transcriptase . Nat. Struct. Mol. Biol ., 12 , 252 - 257 .
83. Cunningham , D.D. and Collins , K. ( 2005 ) Biological and biochemical functions of RNA in the tetrahymena telomerase holoenzyme . Mol. Cell. Biol ., 25 , 4442 - 4454 .
84. Liu , F. , Kim , Y. , Cruickshank , C. and Theimer , C.A. ( 2012 ) Thermodynamic characterization of the Saccharomyces cerevisiae telomerase RNA pseudoknot domain in vitro . RNA , 18 , 973 - 991 .
85. Autexier , C. and Greider , C.W. ( 1998 ) Mutational analysis of the Tetrahymena telomerase RNA: identification of residues affecting telomerase activity in vitro . Nucleic Acids Res ., 26 , 787 - 795 .
86. Xi , L. and Cech , T.R. ( 2014 ) Inventory of telomerase components in human cells reveals multiple subpopulations of hTR and hTERT . Nucleic Acids Res ., 42 , 8565 - 8577 .
87. Fu , D. and Collins , K. ( 2003 ) Distinct biogenesis pathways for human telomerase RNA and H/ACA small nucleolar RNAs . Mol. Cell , 11 , 1361 - 1372 .
88. Frederico , L.A. , Kunkel , T.A. and Shaw , B.R. ( 1990 ) A sensitive genetic assay for the detection of cytosine deamination: determination of rate constants and the activation energy . Biochemistry , 29 , 2532 - 2537 .
89. Bley , C.J. , Qi , X. , Rand , D.P. , Borges , C.R. , Nelson , R.W. and Chen , J.J. ( 2011 ) RNA-protein binding interface in the telomerase ribonucleoprotein . Proc. Natl. Acad. Sci. U.S.A. , 108 , 20333 - 20338 .
90. Huang , J. , Brown , A.F. , Wu , J. , Xue , J. , Bley , C.J. , Rand , D.P. , Wu , L. , Zhang , R. , Chen , J.J. and Lei , M. ( 2014 ) Structural basis for protein-RNA recognition in telomerase . Nat. Struct. Mol. Biol ., 21 , 507 - 512 .
91. Eckert , B. and Collins , K. ( 2012 ) Roles of telomerase reverse transcriptase N-terminal domain in assembly and activity of Tetrahymena telomerase holoenzyme . J. Biol. Chem. , 287 , 12805 - 12814 .
92. Hong , K. , Upton , H. , Miracco , E.J. , Jiang , J. , Zhou , Z.H. , Feigon , J. and Collins , K. ( 2013 ) Tetrahymena telomerase holoenzyme assembly, activation, and inhibition by domains of the p50 central hub . Mol. Cell. Biol ., 33 , 3962 - 3971 .