T. brucei Lister strain 427 was used throughout this study. All BF cells were grown in HMI-9 media supplemented with 10% heat-inactivated Fetal Bovine Serum (FBS) at 37°C and 5% CO_2. T. brucei Lister 427 strain expressing the T7 polymerase and Tet repressor (single marker, AKA SM) (Wirtz et al., 1999) was grown in media containing 2 μg/mL of G418; TbTERT-FLAG-HA-HA (F2H) cells grown with 2 μg/mL of G418, 0.1 μg/mL Puromycin; TbTR ∆C/D box mutant cells were grown with 2 μg/mL of G418, 4 μg/mL of Hygromycin, 2.5 μg/mL of Phleomycin, 5 μg/mL of Blasticidin, 0.1 μg/mL of Puromycin and 0.1 μg/mL of Doxycycline to constitutively induce the TbTR mutations.

The TbTR WT gene without 3’ C/D box region (nt 841–943) together with 400 bp upstream and 380 bp of downstream TbTR flanking sequences were cloned into the pLew111 plasmid to generate pLew111-TbTR ΔC/D box plasmid.

To generate the TbTR ΔC/D box mutant strain, pLew111-TbTR ΔC/D box plasmid (nt 841–943 deleted) was digested with NotI and targeted to an rDNA spacer in the SM/TbTR^−/− cells under the phleomycin selection. Clones were confirmed by northern analysis. XhoI digested pSK-TbTERT-3C-FLAG-HA-HA-PUR plasmid (Dey et al., 2021) was subsequently transfected into the same cells under the puromycin selection to generate the SM/TbTR ΔC/D box/TbTERT ⁺/^F2H strain. Clones were confirmed by western and Southern blotting.

Immunoprecipitation of T. brucei telomerase was performed using a custom made anti-TbTERT antibody (Dey et al., 2021) to purify native telomerase complexes from BF Wild-type (WT) and BF TbTR ∆C/D box cells. Approximately, 5 × 10⁸ cells/300 mL were collected by centrifugation at 1900 RPM for 6 min. Following centrifugation, cells were lysed by homogenization in 500 µL of 1X immunopurified (IP) lysis buffer (25 mM Tris-HCl pH 7.5, 150 mM KCl, 1 mM EDTA, 10 mM MgCl₂, 0.5% IGEPAL CA630, 1X protease cocktail inhibitor, and 20 units of Ribolock RNase inhibitor). Lysate was then cleared of cell debris by centrifugation at 3000 RPM for 5 min at 4°C. The lysate was then pre-cleared by incubated with 50 µL of pre-washed Dynabeads protein G (10003D) for 1 h at 4°C on rotation. Pre-cleared lysates were then incubated overnight at 4°C on rotation with 5 µg of a custom anti-TbTERT antibody and an IgG antibody was added to the control (Dey et al., 2021). The next day, 50 µL of pre-washed Dynabeads protein G was added to the lysate antibody mixture and incubated at 4°C for 2 h on rotation. After incubation, the beads were collected in a magnetic stand and washed twice in ×1 IP lysis buffer. After washing, the bound protein was eluted off the beads by boiling in 100 µL of 1X SDS-PAGE dye for 5 min at 95°C. Eluted proteins were then stored in −80°C until further use. Each experiment was performed in biological triplicate (3 IPs and 3 IgG controls). Bound complexes were assayed for the presence of TbTERT by using an anti-FLAG antibody as the BF WT cells were TbTERT-FLAG-HA-HA tagged. Briefly, 4 µL of 100 µL of sample was loaded onto 4%–12% Novex Tris-glycine gel (Invitrogen, XP04120BOX). Western blotting was done using an anti-FLAG antibody, diluted 1:500, and the VeriBlot for IP Detection Reagent (HRP, ab131366) diluted to 1:10,000. Detection was then done using Pierce ECL Plus Chemiluminescence kit (Thermo Fisher Scientific, 32,106). Imaging was then done using Bio-Rad ChemiDoc MP system.

Immunopurification was also performed using Pierce Anti-DYKDDDDK magnetic beads (A36797). Approximately 6 × 10⁸ cells/300 mL were harvested and lysed in 300 μL of immunopurified (IP) lysis buffer (25 mM Tris-HCl pH 7.5, 150 mM KCl, 25 mM NaCl, 1 mM EDTA, 10 mM MgCl₂, 0.5% IGEPAL CA630, 1× protease cocktail inhibitor and 20 units of Ribolock RNase inhibitor). Lysate was cleared of debris by centrifugation at 3,000 rpm for 5 min at 4°C and incubated with pre-washed 50 μL of Pierce Anti-DYKDDDDK magnetic beads (A36797) at 4°C for 2 h with rotation. Following incubation, the beads were washed twice by ice cold IP buffer and once with ice cold DEPC water. The beads were then resuspended in 50 μL of RNAse free water.

Proteins were separated by SDS-PAGE and Gel segments were cut and subjected to in-gel digestion using trypsin. Peptides were desalted using C18 ZipTips (Millipore). Peptides were analyzed on a Q-Exactive HF hybrid quadrupole-Orbitrap mass spectrometer (Thermo Fisher) equipped with an Easy LC 1200 UPLC liquid chromatography system (Thermo Fisher). Peptides were first trapped using trapping column Acclaim PepMap 100 (75 uM x 2 cm, nanoViper 2Pk, C18, 3 μm, 100A), then separated using analytical column Acclaim PepMap RSLC (75 um × 25 cm, nanoViper, C18, 2 μm, 100A) (Thermo Fisher). The flow rate was 300 nL/min, and a 120-min gradient was used. Peptides were eluted by a gradient from 3% to 28% solvent B (80% (v/v) acetonitrile/0.1% (v/v) formic acid) over 100 min and from 28% to 44% solvent B over 20 min, followed by a short wash at 90% solvent B. For DDA acquisition, the precursor scan was from mass-to-charge ratio (m/z) 375 to 1,600 and the top 20 most intense multiply charged precursors were selected for fragmentation. Peptides were fragmented with higher-energy collision dissociation (HCD) with normalized collision energy (NCE) 27.

The resulting raw data files were searched against a concatenated library (TbruceiTREU927 release 32 databases with 11,202 entries) using MaxQuant (Tyanova et al., 2016). Carbamidomethyl Cysteine was set as a fixed modification. Oxidation of methionine and N-terminal acetylation were set as variable modifications. Tolerance for precursor ions was set to 4.5 ppm and 20 ppm for fragment ions. A maximum of two missed cleavages was allowed. MaxQuant was set to match in between runs and report LFQ. All other parameters were at the default setting. The proteinGroups.txt file generated by MaxQuant was further processed using Perseus. Reverse and possible contaminants were removed from the protein groups. Samples were separated into an experimental group consisting of the pull downs of isotype matched IgG (control) and a group consisting of the TbTERT Immuno-Precipitation pull downs (3 IPs and 3 IgG controls). Protein groups were filtered to contain at least three quantifications in one experimental group. The remaining missing quantifications were imputed with random numbers from a normal distribution (width 0.3, shift = 1.8). A two-sided Student’s t-test was performed across replicates between each experimental group.

The mass spectrometry proteomic data was analyzed by a range of approaches. Volcano plot was generated using GraphPad Prism software version 9.3.1. The STRING database was used for classifying proteins based on functional categories and gene ontology (GO) terms. Protein-protein interaction network analysis was done using STRING version 11.5 (https://string-db.org/) and visualized by using the Cytoscape software version 3.9.1.

Proteins identified by mass spectrometry contained several hits that have very little primary sequence identity with proteins from other phyla, so their functional orthologs were not entirely evident from simple sequence homology. To identify whether structural homology exists between the local folds of these proteins and those reported in other organisms, we obtained predicted structure models of these T. brucei proteins from AlphaFold (Jumper et al., 2021) using their Uniprot IDs and then queried these ‘PDB’ entries using programs PDBeFold (Krissinel and Henrick, 2004) and DaliLite (Holm and Rosenström, 2010). The top scoring and only relevant hit with a DaliLite Z score cut-off of >2 is considered as biologically informative structural neighbor of the protein of interest (Supplementary Table S1).

All TbTERT western blots were done using either an anti-FLAG antibody (1:500) or a custom anti-TbTERT C terminus antibody (1:500) unless otherwise indicated. Nucleolar protein 58 (NOP58) was detected using an anti-NOP58 antibody (Thermo Fisher Scientific, PA5-54321) diluted 1:500 and an anti-Rabbit HRP conjugated secondary antibody diluted to 1:10,000. To qualitatively check protein levels, 4 µL of IP eluate was separated on a 4%–12% Novex Tris-glycine gel and stained with Coomassie Brilliant Blue R-250 Dye (Thermo Scientific, 20,278) for 30 min. The gel was then destained until bands were visible in destain solution (40% MeOH, 10% acetic acid).

TbTERT IP was performed as described in the methods Section 2.4. To detect the presence of TbTR in the IPed complex, total RNA was isolated from the protein G magnetic beads using the TRIzol reagent (Thermo Fisher Scientific, 15596026) following the manufacturers protocol. 100 ng of isolated RNA was then used for cDNA synthesis utilizing the SuperScript II reverse transcriptase (Thermo Fisher Scientific, 18064022) following the manufacturers protocol. The generated cDNA was then used for qRT-PCR analysis using TbTR specific primers (Fwd: CTG​TGG​AAA​TTT​GTC​GTA​AGT​G, Rev: AGT​AGG​GTT​AGG​GAT​CGT​ATA​G).

To determine the activity of the Immunopurified T. brucei telomerase complex, a modified version of the exponential isothermal amplification of telomere repeat (EXPIATR) assay was performed (Tian and Weizmann, 2013; Dey et al., 2021)Briefly, A master mix was prepared on ice consisting of Nicking Telomerase Substrate (NTS, GTG​CGT​GAG​AGC​TCT​TCC​AAT​CCG​TCG​AGC​AGA​GTT), Nicking Probe (NP, AGC​AGG​AAG​CGC​TCT​TCC​TGC​TCC​CTA​ACC​CTA​ACC​C), 1X EXPIATR buffer (30 mM Tris-HCl, pH 8.3, 1.5 mM MgCl2, 100 mM KCl, 1 mM EGTA, 0.05% v/v Tween20), 200 µM dNTPs, Bst 2.0 Warm start DNA polymerase (0.96 units) and Nt. BspQ1 NEase (5 units). 17 μL of the master mix was aliquoted to PCR tubes containing, 3 µL of anti-FLAG bead-bound T. brucei telomerase, RNase A treated or heat-inactivated telomerase RNP bound beads, telomerase positive control (TPC8, GTG​CGT​GAG​AGC​TCT​TCC​AAT​CCG​TCG​AGC​AGA​GTT​AGG​GTT​AGG​GTT​AGG​GTT​AGG​GTT​AGG​GTT​AGG​GTT​AGG​GTT​AGG​G) (0.5 μM) and blank beads as a negative control. Telomerase activity was initiated by initial incubation of tubes at 28°C for 45 min for Nicking telomerase substrate (NTS) extension followed by amplification of resultant telomerase products at 55°C for 30 min. The amplified products were then analyzed on 12% Native PAGE gel by loading 10 μL of the reaction mixture.

Characterizing the global interactors of a protein of interest can be done through affinity-purification mass spectrometry (AP-MS). Identifying the interactome of a protein is key in understanding its function in the cell and how it is regulated. To identify the global protein interactors of T. brucei telomerase, we utilized AP-MS to identify the global interactome of TbTERT at the BF stage. We first immunopurified (IP) TbTERT using a custom anti-TbTERT antibody along with its associated proteins and performed LC-MS/MS (Figure 1A). For immuno-affinity purification, 500 ul of lysate containing 1 mg of total protein was used per IP sample. TbTERT was pulled down from the lysate using a custom anti-C terminus TbTERT antibody (Dey et al., 2021). In addition to verifying the specificity of the custom antibody for binding to TbTERT, the presence of some non-specific cross-reactive bands were also observed. The presence of the immunopurified TbTERT was confirmed using western blotting and SDS-PAGE analysis (Figures 1B–D). Immunoblot analysis (anti-FLAG antibody) of the IP fractions from TbTERT-F2H cells showed that TbTERT was enriched in the pulldown products from anti-TbTERT C terminus antibody IPs but not from control groups. The anti-TbTERT C-terminus antibody IP samples were then subjected to SDS-PAGE and in-gel protease digestion, followed by mass spectrometry as described in ‘Materials and Methods’. To further validate the presence of telomerase components in the IP, RNA was extracted from IP beads and subjected to qRT-PCR analysis to confirm the presence of the telomerase RNA in the complex (Figure 1E). To determine if the immunopurified telomerase complex was catalytically active, we performed a telomerase activity assay (EXPIATR) using the IPed complex (Figure 1F). The result confirmed that the immunopurified T. brucei telomerase complex was catalytically active. To independently validate our proteomic screen, another set of affinity enrichment of TbTERT protein using an anti-FLAG antibody was performed in duplicates in TbTERT-F2H cells (two biological replicates) and screened for known TbTERT interactors. Both the TbTERT IP-MS and TbTERT FLAG pulldown MS data were compared. Since TbTERT and several other proteins that were previously linked to telomerase were identified in both of the FLAG and C-terminus IP mass spec data sets (Figure 2; Supplementary Table S1), it was apparent that the ribonucleoprotein complex identified by this approach is biologically relevant to telomerase function.

We identified 1,056 proteins with 2 or more peptides. After stringent filtering of this dataset for those proteins highly enriched (Log₂(fold change) > 1.9) in the IP vs. control (Supplementary Table S1), 66 high-confidence proteins remained. To study the interactome of TbTERT, the relative abundance (log₂(fold change)) and statistical significance -log₁₀(P-value) of the proteins from the IP samples and controls were calculated (3 IP samples and 3 IgG controls samples) (Figure 2A). This resulted in the enrichment of 56 proteins for the anti-TbTERT IP samples (Supplementary Table S2). The protein samples that were significantly enriched in the IP samples versus the controls (Student’s t-test, -log₁₀(P-value) > 1.3) included both nuclear and mitochondrial proteins that have previously been shown to interact with TERT or play a role in telomere maintenance, such as T. brucei telomerase reverse transcriptase (TbTERT, Tb927.11.10190), yeast telomerase cell cycle turnover-related (anaphase promoting complex) proteins, CDC16 (Tb927.6.2150) and CDC27 (Tb927.10.10330) homologs (Sealey et al., 2011; Ferguson et al., 2013), mitochondrial stress-response protein, human HSP60 homolog (Tb927.11.15040) which is known to accumulate with hTERT in the same fractions of human mitochondria (Sharma et al., 2012), Splicing factor 3B subunit 1 (SF3B1) homolog (Tb927.11.11850) involved in various cellular functions including DNA damage response (Te Raa et al., 2015) and telomere maintenance (Wang et al., 2016), damage specific DNA binding protein 1 (DDB1) homolog (Tb927.6.5110), involved in ubiquitin-mediated TERT protein degradation (Jung et al., 2013), and several other proteins involved in telomerase and telomere metabolism (selected proteins shown in Figure 2A; Table 1).

To highlight the connectivity of candidate TbTERT interactors, we used the STRING database and Cytoscape to generate a functional protein-protein interaction network of TbTERT (Figure 2B; Supplementary Figure S1). To further determine the functions of proteins in the network, STRING GO analysis was done and proteins were grouped by biological process, and cellular component (Figures 3A, B). Terms for biological process that were enriched included, telomere maintenance by telomerase, cell cycle control, DNA repair, and response to stress. Enriched cellular component terms included, box C/D snoRNP complex, DNA replication factor C, anaphase-promoting complex, and cullin-RING ubiquitin ligase complex. Notably, proteins that are important for telomerase RNA biogenesis, processing and trafficking were significantly enriched in these GO terms.

An independent set of IP-MS using an anti-FLAG antibody to IP TbTERT also validated a majority of the proteins identified in the mass spec run described above from the anti-C terminus TbTERT IP because these proteins were identified with both IP approaches. A false discovery rate (FDR) of 1% was used as cut off for this data (Supplementary Table S3).

Homologs of telomerase-associated proteins which are known regulators of telomerase functions were found to be part of the IP telomerase complex in T. brucei. Typically, three types of known telomerase-associated proteins were reported to co-purify with TERT. The foremost of the three are the telomerase RNA-binding proteins that are involved in TR biogenesis, trafficking and TR-TERT assembly. These include, for example, dyskerin in vertebrates (Mitchell and Collins, 2000; Mochizuki et al., 2004), Sm proteins in yeasts (Tang et al., 2012), and La-motif proteins, such as p65, in ciliates (Singh et al., 2012). Interestingly, dyskerin binding H/ACA domain of human TR is replaced by a novel C/D box domain in T. brucei (Gupta et al., 2013). Several unique C/D box snoRNA binding proteins (snoRNPs) were identified in our TbTERT immunopurified complex which are described in the next section. In addition, a T. brucei La protein, Tb927.10.2370, which shows 24% amino acid sequence identity and a Z-score of 8.6 with Tetrahymena thermophila, TR binding protein p65 was also identified in the IP-MS. Telomerase RNP assembly also requires molecular chaperones, such as AAA+ family of ATPases, known as Pontin and Reptin, which can directly interact with TERT and play critical roles in telomerase RNP accumulation (Venteicher et al., 2008). Both the Pontin (Tb927.4.1270) and Reptin (Tb927.4.2000) homologs, annotated as RuvB-like DNA helicases in the T. brucei genome database, were also identified by this TbTERT AP-MS analysis. Additionally, another AAA+ ATPase protein, a yeast CDC48 homolog, Tb927.10.5770, was also identified. CDC48, which was previously identified in a molecular complex that recognized and bound ubiquitinated proteins (Schuberth et al., 2004), was also found to be associated with yeast telomerase as a novel regulator of telomere length homeostasis. Notably, TERT turnover is dependent on ubiquitin-proteasome mediated degradation process (Jung et al., 2013) and therefore proteins that are important for ubiquitination were previously identified in the telomerase complexes, such as several isoforms of E3 ubiquitin ligases (Lin et al., 2015). Interestingly, several ubiquitin-family proteins were also identified in the IP-MS data including ubiquitin ligases, although the roles of these proteins in T. brucei telomerase biology remains uncertain until the ubiquitination status of TbTERT is determined.

Mammalian studies have identified chaperone proteins p23 and HSP90 as two important proteins that are physically and functionally associated with telomerase activity (Holt et al., 1999). The proteomic mapping also identified a mammalian HSP90 homolog, Tb927.3.3580, with enrichment of several unique peptides in TbTERT IP samples identified by MS. Poly(A)-specific ribonuclease (PARN) is a 3′-exoribonuclease that is known to play important role in the maturation of telomerase RNA (Moon et al., 2015). A T. brucei homolog of PARN, Tb927.9.13510 was identified in this proteomic mapping data that may relate with the fact that T. brucei telomerase RNA is a Pol II transcript (Sandhu et al., 2013) that may require PARN processing for maturation. All these known telomerase homologs of T. brucei are listed in Supplementary Table S1. In terms of proteomic identification, it should be noted that several of the above proteins were identified in the range of low scoring functions or higher FDR%, however, these proteins are identified in all four of the biological replicates analyzed by AP-MS and therefore could be biologically relevant. Importantly, all the above proteins identified were part of the IP sample that was able to extend T. brucei telomeric repeats using synthetic TTAGGG as substrates in activity assays (Figure 1F), indicating that the proteins in this IP are potentially part of an active telomerase complex.

In contrast to hTR, the telomerase RNA in T. brucei contains a unique C/D snoRNA-like domain (Figure 4A bottom). Core C/D box RNPs, like NOP58, have previously been shown to interact with TbTR (Gupta et al., 2013). NOP58 is a 57 KDa protein that contains a coiled-coil (CC) domain and a NOP domain (Figure 4A top). NOP58 is highly conserved across eukaryotes and plays important roles in ribosomal RNA (rRNA) processing (Barth et al., 2008). In our AP-MS analysis of TbTERT, we identified three core C/D box binding proteins: NOP58, NOP56, and Fibrillarin (NOP1) (Table 2; Supplementary Table S3). To validate the interaction with NOP58, we performed a co-immunoprecipitation (Co-IP) assay of TbTERT and detected TbTERT and NOP58 through western blotting (Figure 4B top). For further confirmation of this interaction, we performed Co-IPs of TbTERT in both WT and ∆C/D box mutant cell lines. In the WT cells, NOP58 is present in the IP product, while in the ∆C/D mutant cells, the interaction with NOP58 is greatly diminished (Figure 4B bottom). This data supports the fact that the C/D box motif is important for the interaction between TbTR and NOP58.