【Science】

Structure of the Tribolium castaneum telomerase catalytic subunit TERT 

 

Andrew J. Gillis¹, Anthony P. Schuller¹ & Emmanuel Skordalakes¹ 

 

A common hallmark标志 of human cancers is the overexpression of telomerase端粒酶, a ribonucleoprotein complex that is responsible for负责 maintaining the length and integrity完整性 of chromosome ends. Telomere length deregulation失控 and telomerase activation激活 is an early, and perhaps necessary, step in cancer cell evolution发展. Here we present呈现 the high-resolution structure of the Tribolium castaneum赤拟谷盗 catalytic subunit of telomerase, TERT. The protein consists of three highly conserved domains, organized into ring-like structure that shares common features with retroviral reverse transcriptases逆转录酶, viral RNA polymerases and B-family DNA polymerases. Domain organization places放置 motifs implicated牵连 in substrate binding and catalysis in the interior内部 of the ring, which can accommodate容纳 seven to eight bases of double-stranded nucleic acid. Modelling of an RNA–DNA heteroduplex异源双链 in the interior of this ring demonstrates perfect fit完全合适 between the protein and the nucleic acid substrate, and positions安置 the 39-end of the DNA primer at the active site of the enzyme, providing evidence for the formation of an active telomerase elongation延伸 complex. 

 

Telomerase is active in the early stages of life to maintain telomere length and therefore the chromosomal integrity of frequently dividing cells, and it becomes dormant静止的 in most somatic cells during adulthood1,2. The ability of telomeres to provide genomic stability is diminished削减 over time owing to由于 both the natural loss of telomeric structure with every cell division, and the loss of telomerase activity—a process which leads to ageing衰老3,4. In cancer cells, however, telomerase becomes reactivated and works tirelessly无休止地 to maintain the short length of telomeres of rapidly dividing cells, leading to their immortality永生性5,6. The essential role of telomerase in cancer and ageing makes it an important target for the development of therapies to treat cancer and other age-associated disorders. 

Telomerase functions as both a monomer and a dimer7–10,and consists of a protein subunit (TERT) and an integral RNA component (TER) which contains the template模板 that TERT uses to add several DNA repeats to the 39-end of linear chromosomes11,12 . TERT, the catalytic subunit催化亚单位 of telomerase, is highly conserved among phylogenetic系统发生的 groups and shares common motifs with conventional常见的 reverse transcriptases, suggesting an overall总的 conservation of the basic catalytic mechanism between these two classes of enzymes13,14. Although TER varies considerably很 in size, sequence and structure between species, core structural elements are conserved, suggesting that there is a common mechanism of telomere replication among organisms15,16 . 

A functional telomerase holoenzyme全酶 requires the stable association of the ribonucleoprotein complex, a process mostly carried out by the RNA-binding domain (TRBD)17,18. Weak interactions have been reported between TER and both the far amino-terminal domain (a low conservation region of TERT) and the polymerase domain (reverse transcriptase)18,19. Current evidence suggests that TRBD binds to the template boundary边界 element of TER, usually a stem loop or a pseudoknot假结 flanked两侧有 by regions of single-stranded RNA20–23. The TRBD–TER association also promotes repeat addition processivity持续合成能力, which is a unique独特的 feature of telomerase19–22,24. Telomerase repeat addition processivity is also attributed to归于 the IFD (insertion in fingers domain) motif of reverse transcriptase and the carboxy-terminal extension (CTE) proposed to constitute the putative推定的 ‘thumb’ domain of telomerase25–27. 

Initiation of telomere synthesis requires the loading of telomerase onto the end of the chromosomes and the pairing of the 39-end of the linear DNA substrate with the templating region (usually one and a half repeats of the telomeric repeat)28–30. Pairing of the DNA with the RNA template places the 39-end of the DNA substrate at the active site of the enzyme for nucleotide addition, whereas the RNA template provides the platform for the successive rounds连续的循环 of nucleotide addition and selectivity. RNA–DNA pairing alone is not sufficient for a stable and active telomerase elongation complex and requires extensive广泛的 contacts of the DNA substrate with both the reverse transcriptase and the putative thumb domain of TERT25,31. In some organisms, contacts between the far N-terminal domain and a DNA site upstream of the RNA–DNA hybridization region allow the enzyme to remain attached to the end of the chromosomes during translocation易位32,33. 

Here we present, to our knowledge, the first high-resolution structure of the catalytic subunit of telomerase. This structure, together with previous biochemical data, provides insights into TERT–TER– DNA assembly and elongation complex formation. 

 

Architecture of the TERT structure 

We have solved the structure of the full-length catalytic subunit of the T. castaneum active telomerase34,35, TERT, to 2.71 A° resolution. There is a dimer in the asymmetric unit; however, the protein alone is clearly monomeric in solution as indicated by gel filtration凝胶过滤 and dynamic light scattering散射 (results not shown) suggesting that the dimer we observe in the crystal is the result of crystal packing. This notion is further supported by the fact that a different crystal form (Supplementary Table 1) of the same protein also contains a dimer in the asymmetric unit of a different configuration than the one presented here. It is worth noting that the TERT from this organism does not contain an N-terminal domain, a low conservation region of telomerase (Fig. 1a, b).  

The TERT structure is composed of three distinct domains: an RNA-binding domain (TRBD), the reverse transcriptase domain and the CTE thought to represent the putative thumb domain of TERT (Fig. 1a, c). The TRBD is mostly helical and contains an indentation凹陷 on its surface formed by two conserved motifs (CP and T) known to bind the double-and single-stranded RNA regions of the template boundary element, respectively24 (Fig. 2a). Structural comparison of the TRBD from T. castaneum with that of the previously determined structure from Tetrahymena thermophila四膜虫24 shows similarity between the two structures (root mean squared deviation均方根偏差 (r.m.s.d.) 2.7 A° ), suggesting that a high degree程度 of structural conservation occurs between these domains across organisms of diverse phylogenetic groups.

 

 

 

The reverse transcriptase domain is a mixture of α-helices and β-strands organized into two subdomains that are most similar to the ‘fingers’ and ‘palm’ subdomains of retroviral reverse transcriptases36, viral RNA polymerases37 and B-family DNA polymerases38 (Supplementary Fig. 1a–d), and contains important signature motifs that are hallmarks of these families of proteins14 (Fig. 2b). Structural comparisons of TERT with the HIV reverse transcriptases show that the fingers subdomain of TERT is arranged in the open configuration with respect to有关 the palm subdomain, which is in good agreement with the conformation adopted by HIV reverse transcriptases in the absence of bound nucleotide and nucleic acid substrates39. One notable difference between the putative palm domain of TERT and the HIV reverse transcriptases is a long insertion between motifs A and B’ of TERT; this is referred to as the IFD motif and is required for telomerase processivity27. In the TERT structure, the IFD insertion consists of two antiparallel α-helices (α13 and α14) located on the outside periphery外周 of the ring and at the interface of the fingers and the palm subdomains (Fig. 2b). These two helices are almost in a parallel position with the central axis of the plane of the ring, make extensive contacts with helices α10 and α15, and have an important role in the structural organization of this part of the reverse transcriptase domain. A similar structural arrangement is also present in viral polymerases, and the equivalent等价物 of helix α10 in these structures is involved in direct contacts with the nucleic acid substrate40 (Supplementary Fig. 1c). 

In contrast to和……相比 the reverse transcriptase domain, the CTE is an elongated helical bundle束 that contains several surface-exposed long loops (Fig. 2c). A search in the protein structure database using the secondary-structure matching software (http://www.ebi.ac.uk/msd-srv/ssm)41 produced no structural homologues同源物, suggesting that the CTE domain of telomerase adopts采取 a new fold. Structural comparison of TERT with the HIV reverse transcriptase, with the viral RNA polymerases and with the B-family DNA polymerases places the thumb domain of these enzymes and the CTE domain of TERT in the same spatial空间的 position with respect to the fingers and palm subdomains. This suggests that the CTE domain of telomerase is the thumb domain of the enzyme, a finding that is in good agreement with previous biochemical studies25 (Supplementary Fig. 2). 

TERT domain organization brings the TRBD and thumb domain—which constitute the terminal domains of the molecule—together, an arrangement that leads to the formation of a ring-like structure that is reminiscent回忆起 of the shape of a doughnut油炸圈饼 (Fig. 1a, b). Several lines of evidence suggest that the domain organization of the TERT structure presented here is biologically relevant. First, the domains of four TERT monomers observed in two different crystal forms (two in each asymmetric unit) all have the same organization (average r.m.s.d. = 0.76 A° between all four monomers). Second, contacts between the N-and the C-terminal domains of TERT are extensive (1,677 A° ²) and largely hydrophobic疏水的 in nature, an observation that is consistent with previous biochemical studies42 (Supplementary Fig. 3). Third, TERT domain organization is similar to that of the polymerase domain (p66 minus the RNase H domain) of its closest homologue, HIV reverse transcriptase36. It is also similar to the domain organization of the viral RNA polymerases37 and that of the B-family DNA polymerases, particularly RB69 (ref. 38; Supplementary Fig. 1a–d). The arrangement of the TERT domains creates a hole in the interior of the particle that is ~26 A° wide and ~21 A° deep, sufficient足够的 to accommodate double-stranded nucleic acid approximately seven to eight bases long and in good agreement with existing biochemical data43,44. 

 

The TERT ring binds double-stranded nucleic acid 

To understand better how the TERT ring associates with RNA–DNA to form a functional elongation complex, we modelled double-stranded nucleic acid into its interior using the complex of HIV reverse transcriptase with DNA36, the closest structural homologue of TERT (Fig. 3a). The TERT–RNA–DNA model immediately shows some notable features that support our model of TERT–nucleic-acid associations. The hole of the TERT ring, and where the nucleic acid heteroduplex is projected突出 to bind, is lined with several key signature motifs that are hallmarks of this family of polymerases and have been implicated in nucleic acid association, nucleotide binding and DNA synthesis (Fig. 3a). Moreover, the organization of these motifs results in the formation of a spiral螺旋 in the interior of the ring that resembles the geometry几何学 of the backbone骨架 of double-stranded nucleic acid (Fig. 3b). Several of the motifs, identified as contact points with the DNA substrate, are formed partly by positively charged residues, the side chains of which extend towards the centre of the ring and are poised摆好姿态 for direct contact with the backbone of the DNA substrate. For example, the side chain of the highly conserved K210 (Supplementary Fig. 4) that forms part of helix α10, is within coordinating同等的 distance of the backbone of the modelled DNA, thus providing the stability required for a functional telomerase enzyme. Helix α10 lies in the upper segment of the reverse transcriptase domain and faces the interior of the ring. The location and stabilization of this helix is heavily influenced by its extensive contacts with the IFD motif implicated in telomerase processivity27. Disruption of the IFD contacts with helix α10—by deletion or mutation of this motif—would lead to displacement转移 of helix α10 from its current location, which would in turn effect DNA-binding and telomerase function. 

    

      

 

Structural elements of the thumb domain that localize to the interior of the ring also make several contacts with the modelled DNA substrate (Fig. 3a). In particular, the loop (thumb loop) that connects the palm to the thumb domain and constitutes an extension of motif E, also known as the ‘primer grip引物夹’ region of telomerase, preserves维护 the geometry of the backbone of double-stranded nucleic acid to a notable degree (Fig. 3b). The side chains of several lysines and asparagines that form part of this loop extend towards the centre of the TERT molecule and are in coordinating distance of the backbone of modelled double-stranded nucleic acid. Of particular interest is K406, located in proximity接近 of motif E. The side chain of this lysine extends towards the nucleic acid heteroduplex and it is poised for direct contacts with the backbone of the nucleotides located at the 39-end of the incoming DNA primer. It is therefore possible that the side chain of this lysine, together with motif E, help facilitate placement of the 39-end of the incoming DNA substrate at the active site of the enzyme during telomere elongation. Sequence alignments of the thumb domain of TERTs from a wide spectrum of phylogenetic groups show that the residues predicted to contact the DNA substrate are always polar (Supplementary Fig. 4). Another interesting feature of the thumb domain, which supports double-stranded nucleic acid binding, is helix α19 (Fig. 2c). Thisisa 310 helix (thumb 310 helix) that extends into the interior of the ring and seems to dock停靠 itself into the minor groove of the modelled double-stranded nucleic acid, thus facilitating RNA–DNA hybrid binding and stabilization (Fig. 3b). Deletion or mutation of the corresponding residues in both yeast and human TERT results in severe loss of TERT processivity, clearly indicating the important role of this motif in TERT function25,26,45 . 

 

 

The active site of TERT and nucleotide binding 

The TERT structure presented here was crystallized in the absence of nucleotide substrates and magnesium; however, the location and organization of TERT’s active site and nucleotide-binding pocket can be predicted on the basis of existing biochemical data14 and structural comparison with the polymerase domain of its closest homologue, the HIV reverse transcriptase46. The TERT active site consists of three invariant aspartic acids (D251, D343 and D344) that form part of motifs A and C, which are two short loops located on the palm subdomain and adjacent to the fingers of TERT (Fig. 4a). Structural comparisons of TERT with HIV reverse transcriptases, as well as with RNA and DNA polymerases, show a high degree of similarity between the active sites of these families of proteins (Fig. 4b), suggesting that telomerase also uses a two-metal mechanism for catalysis. Alanine mutants of these TERT aspartic acids resulted in complete loss of TERT activity, indicating that the role of these residues in telomerase function is essential14. 

The telomerase nucleotide-binding pocket is located at the interface of the fingers and palm subdomains of TERT (Fig. 4a) and consists of conserved residues that form the motifs 1, 2, A, C, B’ and D which are implicated in template and nucleotide binding47,48 (Supplementary Fig. 5). Structural comparisons of TERT with viral HIV reverse transcriptases bound to ATP46 support the presence of a nucleotide substrate in this location. Two highly conserved surface-exposed residues Y256 and V342 of motifs A and C, respectively, form a hydrophobic pocket adjacent to临近 and above the three catalytic aspartates天冬氨酸 and this could accommodate the base of the nucleotide substrate. Binding of the nucleotide in this oily油的 pocket places the triphosphate moiety一半，部分 in proximity of the enzyme’s active site for coordination with one of the Mg²⁺ ions. In contrast, it positions the ribose核糖 group within coordinating distance of an invariant glutamine (Q308) that forms part of motif B’, which is thought to be an important determinant决定因素 of substrate specificity49. Protein contacts with the triphosphate moiety of the nucleotide are mediated by motif D, a long loop located beneath在下方 the active site of the enzyme. In particular, the side chain of the invariant K372 is within coordinating distance of the γ-phosphate of the nucleotide, an interaction that probably helps position and stabilize the triphosphate group during catalysis. The side chains of the highly conserved K189 and R192 of motifs 1 and 2, which together form a long β-hairpin that forms part of the fingers subdomain, are also within coordinating distance of both the sugar and triphosphate moieties of the modelled nucleotide. Contacts with either or both the sugar moiety and the triphosphate moiety of the nucleotide substrate would facilitate nucleotide binding and positioning for coordination to the 39-end of the incoming DNA primer. 

 

TRBD facilitates template positioning at the active site 

As with most DNA and RNA polymerases, nucleic acid synthesis by telomerase requires pairing of the templating region (usually seven to eight bases or more) of TER with the incoming DNA primer28. TRBD–reverse-transcriptase domain organization forms a deep cavity腔 on the surface of the protein that spans横跨 the entire width of the wall壁 of the molecule, forming a gap缝隙 that allows entry into the hole of the ring from its side (Fig. 3a). The arrangement of this cavity with respect to the central hole of the ring provides an elegant mechanism upon TERT–TER assembly for the placement of the RNA template in the interior of the ring and where the enzyme’s active site is located. Of particular significance is the arrangement of the β-hairpin that forms part of the T motif. This hairpin extends from the RNA-binding pocket and makes extensive contacts with the thumb loop and motifs 1 and 2 (Fig. 3a). Contacts between this hairpin and both the fingers and the thumb domains place the opening of the TRBD pocket that faces the interior of the ring in proximity to the active site of the enzyme (Fig. 5). It is therefore possible that this β-hairpin acts as an allosteric变构的 effector switch that couples RNA binding in the interior of the ring and placement of the RNA template at the active site of the enzyme. Placement of the template into the interior of the molecule would facilitate its pairing with the incoming DNA substrate, which together would form the RNA–DNA hybrid required for telomere elongation. RNA–DNA pairing is a prerequisite of telomere synthesis in that it brings the 39-end of the incoming DNA primer in proximity to the active site of the enzyme for nucleotide addition of identical repeats of DNA at the ends of chromosomes. Notably, modelling of the RNA–DNA heteroduplex in the interior of the TERT ring places the 59-end of the RNA substrate at the entry of the RNA-binding pocket and where TERT is expected to associate with TER, whereas it places the 39-end of the incoming DNA primer at the active site of TERT providing a snapshot快照 of the organization of a functional telomerase elongation complex (Fig. 5). 

 

Conclusions 

The structure presented here provides a view of the full-length catalytic subunit of telomerase. The structure shows that TERT is organized into an unexpected ring configuration that resembles—both structurally and functionally—the HIV reverse transcriptases, the viral RNA polymerases and the B-family DNA polymerases, suggesting that there is an evolutionary link between these families of enzymes. It also provides insights into the mechanism of TERT and RNA–DNA association, which in turn explains how TERT may assemble with RNA–DNA and offers a snapshot of a functional telomerase elongation complex required for telomere synthesis. Moreover, because telomerase has a critical role in both cancer and ageing, these findings could potentially assist our efforts to identify and develop inhibitors and/or activators of this enzyme for the treatment of cancer and ageing, respectively. 

 

METHODS SUMMARY 

The full-length TERT of T. castaneum was overexpressed in bacteria and purified by nickel镍, ion-exchange and gel-filtration chromatography. Co-crystallization of the protein–telomeric-DNA ((TCAGG)3) produced two crystal forms (orthorhombic and hexagonal), which were grown by the vapour diffusion, sitting-drop method. Data were collected at the National Synchrotron Light Source国家同步光源 (NSLS) at beamline X6A and were processed with MOSFILM (Supplementary Table 1). Phases for the orthorhombic正交晶的或斜方晶的 crystal were obtained by the method of single isomorphous同形的 replacement with anomalous不规则的 signal using a mercury水银 derivative (CH3HgCl; Supplementary Table 1). The model from the orthorhombic crystal was subsequently used to solve the hexagonal六角形 crystal form by molecular replacement. Both models were refined to good stereochemistry立体化学 (Supplementary Table 1).  

转载本文请联系原作者获取授权，同时请注明本文来自徐文科学网博客。
链接地址：https://blog.sciencenet.cn/blog-200233-201805.html

上一篇：我的第一篇英文演讲稿（RNAi）
下一篇：艾滋病疫苗-5年内搞定！