Distinct domains of a nucleolar protein mediate protein kinase binding, interaction with nucleic acids and nucleolar localization

An essential Trypanosoma brucei protein kinase: a functional analysis of regulation and the identification of inhibitors

Introduction

The protein serine/threonine kinase AEK1 is essential in the pathogenic stage of Trypanosoma brucei, the causative agent of African trypanosomiasis. AEK1 is a member of the AGC protein kinase family, although it is not closely related to a specific human AGC kinase. Our previous chemical genetic studies showed that targeted inhibition of AEK1 in parasites expressing analog-sensitive AEK1 blocked parasite growth and enhanced survival of infected mice.

Methods

To further validate AEK1 as a drug target, we used the chemical genetic system to determine the effect of a 24 hour loss of AEK1 activity on cell viability at the clonal level. A panel of 429 protein kinase inhibitors were screened against the wild-type protein for binding, using time-resolved fluorescence energy transfer (TR-FRET). The role of phosphorylation sites and motifs was probed by determining whether expression of proteins harboring mutations in these sequences could rescue AEK1 conditional knockout parasites. To determine the effect that mutations in the phosphosites have on the kinase activity of cellular AEK1 we compared the in vitro kinase activity of mutant and wild-type proteins immunoprecipitated from parasite lysates using the exogenous substrate MBP. Finally, the tagged AEK1 protein was localized by deconvolution microscopy.

Results

After a 24 hour exposure to an AEK1 inhibitory analog in the chemical genetic system, less than five percent of the remaining live cells can clonally expand, further validating AEK1 as a drug target. In the AEK1 inhibitor screening assay, we identified 17 hit compounds. Complementation studies showed that of the two known phosphorylation sites in the activation loop; mutation of one abolished function while mutation of the other had no discernable effect. Mutation of the other two AEK1 phosphosites gave intermediate phenotypes. Mutations in either the hydrophobic motif at the C-terminus of the protein or in the region of AEK1 predicted to bind the hydrophobic motif were also required for function. All parasites with defective AEK1 showed reduced proliferation and defects in cytokinesis, although the tested mutations differed in terms of the extent of cell death. Kinase activity of immunoprecipitated AEK1 phosphosite mutants largely paralleled the effects seen in complementation studies, although the mutation of the phosphosite adjacent to the hydrophobic motif had a greater impact on activity than predicted by the complementation studies. AEK1 was localized to cytoplasmic puncta distinct from glycosomes and acidocalcisomes.

Discussion

The rapid loss of viability of cells inhibited for AEK1 supports the idea that a short course of treatment that target AEK1 may be sufficient for treatment of people or animals infected with T. brucei. Key regulatory elements between AEK1 and its closest mammalian homolog appear to be largely conserved despite the vast evolutionary distance between mammals and T. brucei. The presence of AEK1 in cytoplasmic puncta raises the possibility that its localization may also play a role in functional activity.

Unusual features and localization of the membrane kinome of Trypanosoma brucei

In many eukaryotes, multiple protein kinases are situated in the plasma membrane where they respond to extracellular ligands. Ligand binding elicits a signal that is transmitted across the membrane, leading to activation of the cytosolic kinase domain. Humans have over 100 receptor protein kinases. In contrast, our search of the Trypanosoma brucei kinome showed that there were only ten protein kinases with predicted transmembrane domains, and unlike other eukaryotic transmembrane kinases, seven are predicted to bear multiple transmembrane domains. Most of the ten kinases, including their transmembrane domains, are conserved in both Trypanosoma cruzi and Leishmania species. Several possess accessory domains, such as Kelch, nucleotide cyclase, and forkhead-associated domains. Surprisingly, two contain multiple regions with predicted structural similarity to domains in bacterial signaling proteins. A few of the protein kinases have previously been localized to subcellular structures such as endosomes or lipid bodies. We examined the localization of epitope-tagged versions of seven of the predicted transmembrane kinases in T. brucei bloodstream forms and show that five localized to the endoplasmic reticulum. The last two kinases are enzymatically active, integral membrane proteins associated with the flagellum, flagellar pocket, or adjacent structures as shown by both fluorescence and immunoelectron microscopy. Thus, these kinases are positioned in structures suggesting participation in signal transduction from the external environment.

Trypanosoma brucei: Two mitogen activated protein kinase kinases are dispensable for growth and virulence of the bloodstream form

Mitogen activated protein kinase cascades function in eukaryotic responses to the environment and stress. Trypanosomatid parasites possess protein kinases with sequences characteristic of kinases in such cascades. In this report we use gene knockouts to demonstrate that two mitogen activated kinase kinase genes, MKK1 (Tb927.3.4860) and MKK5 (Tb927.10.5270), are not essential in the pathogenic bloodstream stage of Trypanosoma brucei, either in vitro or in vivo. Bloodstream forms lacking MKK1 showed decreased growth at 39°C as compared to the parental line. However, unlike its Leishmania orthologue, T. brucei MKK1 does not appear to play a significant role in flagellar biogenesis.

A novel protein kinase localized to lipid droplets is required for droplet biogenesis in trypanosomes

Ubiquitous among eukaryotes, lipid droplets are organelles that function to coordinate intracellular lipid homeostasis. Their morphology and abundance is affected by numerous genes, many of which are involved in lipid metabolism. In this report we identify a Trypanosoma brucei protein kinase, LDK, and demonstrate its localization to the periphery of lipid droplets. Association with lipid droplets was abrogated when the hydrophobic domain of LDK was deleted, supporting a model in which the hydrophobic domain is associated with or inserted into the membrane monolayer of the organelle. RNA interference knockdown of LDK modestly affected the growth of mammalian bloodstream-stage parasites but did not affect the growth of insect (procyclic)-stage parasites. However, the abundance of lipid droplets dramatically decreased in both cases. This loss was dominant over treatment with myriocin or growth in delipidated serum, both of which induce lipid body biogenesis. Growth in delipidated serum also increased LDK autophosphorylation activity. Thus, LDK is required for the biogenesis or maintenance of lipid droplets and is one of the few protein kinases specifically and predominantly associated with an intracellular organelle.

The Trypanosoma brucei life cycle switch TbPTP1 is structurally conserved and dephosphorylates the nucleolar protein NOPP44/46

Trypanosoma brucei adapts to changing environments as it cycles through arrested and proliferating stages in the human and tsetse fly hosts. Changes in protein tyrosine phosphorylation of several proteins, including NOPP44/46, accompany T. brucei development. Moreover, inactivation of T. brucei protein-tyrosine phosphatase 1 (TbPTP1) triggers differentiation of bloodstream stumpy forms into tsetse procyclic forms through unknown downstream effects. Here, we link these events by showing that NOPP44/46 is a major substrate of TbPTP1. TbPTP1 substrate-trapping mutants selectively enrich NOPP44/46 from procyclic stage cell lysates, and TbPTP1 efficiently and selectively dephosphorylates NOPP44/46 in vitro. To provide insights into the mechanism of NOPP44/46 recognition, we determined the crystal structure of TbPTP1. The TbPTP1 structure, the first of a kinetoplastid protein-tyrosine phosphatase (PTP), emphasizes the conservation of the PTP fold, extending to one of the most diverged eukaryotes. The structure reveals surfaces that may mediate substrate specificity and affords a template for the design of selective inhibitors to interfere with T. brucei transmission.

Arginine methylation regulates mitochondrial gene expression in Trypanosoma brucei through multiple effector proteins

Arginine methylation is a post-translational modification that impacts gene expression in both the cytoplasm and nucleus. Here, we demonstrate that arginine methylation also affects mitochondrial gene expression in the protozoan parasite, Trypanosoma brucei. Down-regulation of the major trypanosome type I protein arginine methyltransferase, TbPRMT1, leads to destabilization of specific mitochondrial mRNAs. We provide evidence that some of these effects are mediated by the mitochondrial RNA-binding protein, RBP16, which we previously demonstrated affects both RNA editing and stability. TbPRMT1 catalyzes methylation of RBP16 in vitro. Further, MALDI-TOF-MS analysis of RBP16 isolated from TbPRMT1-depleted cells indicates that, in vivo, TbPRMT1 modifies two of the three known methylated arginine residues in RBP16. Expression of mutated, nonmethylatable RBP16 in T. brucei has a dominant negative effect, leading to destabilization of a subset of those mRNAs affected by TbPRMT1 depletion. Our results suggest that the specificity and multifunctional nature of RBP16 are due, at least in part, to the presence of differentially methylated forms of the protein. However, some effects of TbPRMT1 depletion on mitochondrial gene expression cannot be accounted for by RBP16 action. Thus, these data implicate additional, unknown methylproteins in mitochondrial gene regulation.

Species specificity in ribosome biogenesis: a nonconserved phosphoprotein is required for formation of the large ribosomal subunit in Trypanosoma brucei

In the protozoan parasite Trypanosoma brucei, the large rRNA, which is a single 3.4- to 5-kb species in most organisms, is further processed to form six distinct RNAs, two larger than 1 kb (LSU1 and LSU2) and four smaller than 220 bp. The small rRNA SR1 separates the two large RNAs, while the remaining small RNAs are clustered at the 3' end of the precursor rRNA. One would predict that T. brucei possesses specific components to carry out these added processing events. We show here that the trypanosomatid-specific nucleolar phosphoprotein NOPP44/46 is involved in this further processing. Cells depleted of NOPP44/46 by RNA interference had a severe growth defect and demonstrated a defect in large-ribosomal-subunit biogenesis. Concurrent with this defect, a significant decrease in processing intermediates, particularly for SR1, was seen. In addition, we saw an accumulation of aberrant processing intermediates caused by cleavage within either LSU1 or LSU2. Though it is required for large-subunit biogenesis, we show that NOPP44/46 is not incorporated into the nascent particle. Thus, NOPP44/46 is an unusual protein in that it is both nonconserved and required for ribosome biogenesis.

A variant histone H3 is enriched at telomeres in Trypanosoma brucei

Variant histones play critical roles in transcriptional activation and repression, DNA repair and chromosome segregation. We have identified HTV, a single-copy gene in Trypanosoma brucei encoding a variant form of histone H3 (H3V). H3V is present at discrete nuclear foci that shift over the course of the cell cycle and associate with the mitotic spindle, a pattern of localization reminiscent of that described previously for both mini-chromosomes and telomeres. By combining fluorescence in situ hybridization with indirect immunofluorescence, we confirmed that the H3V foci overlap with a 177-bp repetitive sequence element found predominantly in mini-chromosomes, as well as with the TTAGGG repeats that compose telomeres. Chromatin immunoprecipitation studies, however, reveal that only the telomeric repeat DNA is substantially enriched with H3V. HTV is not essential for viability, mini-chromosome segregation, telomere maintenance or transcriptional silencing at the telomere-proximal expression sites from which bloodstream-form T. brucei controls antigenic variation. We propose that H3V represents a novel class of histone H3 variant, a finding that has evolutionary implications.

Nuclear DNA helicase II (RNA helicase A) binds to an F-actin containing shell that surrounds the nucleolus

Nuclear DNA helicase II (NDH II), alternatively named RNA helicase A (RHA), is an F-actin binding protein that is particularly enriched in the nucleolus of mouse cells. Here, we show that the nucleolar localization of NDH II of murine 3T3 cells depended on an ongoing rRNA synthesis. NDH II migrated out of the nucleolus after administration of 0.05 microg/ml actinomycin D, while nucleolin and the upstream binding factor (UBF) remained there. In S phase-arrested mouse cells, NDH II was frequently found at the nucleolar periphery, where it was accompanied by newly synthesized nucleolar RNA. Human NDH II was mainly distributed through the whole nucleoplasm and not enriched in the nucleoli. However, in the human breast carcinoma cell line MCF-7, NDH II was also found at the nucleolar periphery, together with the tumor suppressor protein p53. Both NDH II and p53 were apparently attached to the F-actin-based filamentous network that surrounded the nucleoli. Accordingly, this subnuclear structure was sensitive to F-actin depolymerizing agents. Depolymerization with gelsolin led to a striking accumulation of NDH II in the nucleoli of MCF-7 cells. This effect was abolished by RNase, which extensively released nucleolus-bound NDH II when added together with gelsolin. Taken together, these results support the idea that an actin-based filamentous network may anchor NDH II at the nucleolar periphery for pre-ribosomal RNA processing, ribosome assembly, and/or transport.

The NOG1 GTP-binding protein is required for biogenesis of the 60 S ribosomal subunit

NOG1 is a nucleolar GTP-binding protein present in eukaryotes ranging from trypanosomes to humans. In this report we demonstrate that NOG1 is functionally linked to ribosome biogenesis. In sucrose density gradients Trypanosoma brucei NOG1 co-sediments with 60 S ribosomal subunits but not with monosomes. 60 S precursor RNAs are co-precipitated with NOG1. Together with the nucleolar localization of NOG1, these data indicate that NOG1 is associated with a precursor particle to the 60 S subunit. Disruption of NOG1 function through RNA interference led to a dramatic decrease in the levels of free 60 S particles and the appearance of an atypical rRNA intermediate in which ITS2 was not cleaved. Overexpression of mutant nog1 with a defect in its GTP binding motif on a wild type background caused a modest defect in 60 S biogenesis and a relative decrease in processing of the large subunit rRNAs. In contrast to the mutant protein, neither the N-terminal half of NOG1, which contains the GTP binding motifs, nor the C-terminal half of NOG1 associated with pre-ribosomal particles, although both localized to the nucleolus.

Two families of RNA binding proteins from Trypanosoma brucei associate in a direct protein-protein interaction

We have previously reported the identification of two closely related RNA binding proteins from Trypanosoma brucei, termed p34 and p37. The predicted primary structures of the two proteins are highly homologous with one major difference, an 18 amino acid insertion in the N-terminal region of p37. These two proteins are localized to the nucleus based on immunofluorescence microscopy. Recently, we have shown that p34 and p37 interact with T. brucei 5S rRNA. In order to gain further insight into their function, we have utilized protein affinity chromatography and immune capture approaches to identify T. brucei proteins which associate with p34 and p37. We demonstrate here an interaction of both p34 and p37 with the NOPP44/46 proteins, identified in T. brucei as a family of tyrosine-phosphorylated RNA binding proteins primarily localized to the nucleolus. This interaction was mapped to the RNA-binding region of p34/p37 and an acidic region of NOPP44/46 by protein affinity chromatography using recombinant deletion constructs of p34 and p37 and yeast two-hybrid analysis. These data may suggest a role for p34 and p37 and NOPP44/46 in the import and/or assembly pathway of T. brucei 5S rRNA in ribosome biogenesis.

Molecular cloning of Trypanosoma brucei CK2 catalytic subunits: the alpha isoform is nucleolar and phosphorylates the nucleolar protein Nopp44/46

We have demonstrated previously that Nopp44/46, an abundant nucleolar phosphoprotein of Trypanosoma brucei, is associated with a protein kinase. In many organisms multiple nucleolar proteins are phosphorylated by the protein kinase CK2, formerly known as casein kinase II. Here we report the identification of two T. brucei genes, CK2a1and CK2a2, which encode protein kinases bearing signature motifs common to CK2 catalytic subunits. The protein specified by CK2a1, designated CK2alpha, was capable of associating with Nopp44/46 as assessed by yeast two-hybrid analysis. An epitope-tagged version of CK2alpha expressed in T. brucei colocalized with Nopp44/46, with a largely nucleolar localization. This localization contrasts with the predominantly nuclear localization of mammalian CK2. When expressed in Escherichia coli, TbCK2alpha was catalytically active and phosphorylated Nopp44/46. Together these data demonstrate that TbCK2alpha is a Nopp44/46-associated kinase. Competition assays revealed that, unlike most CK2s, TbCK2alpha discriminates highly between ATP and GTP. This distinction may be associated with the substitution of glutamic acid and alanine for the di-asparagine motif thought to participate in purine interaction.

Two splice variants of Nopp140 in Drosophila melanogaster

The Nopp140 gene of Drosophila maps within 79A5 of chromosome 3. Alternative splicing yields two variants. DmNopp140 (654 residues) is the sequence homolog of vertebrate Nopp140. Its carboxy terminus is 64% identical to that of the prototypical rat Nopp140. DmNopp140-RGG (688 residues) is identical to DmNopp140 throughout its first 551 residues, but its carboxy terminus contains a glycine/arginine-rich domain that is often found in RNA-binding proteins such as vertebrate nucleolin. Both Drosophila variants localize to nucleoli in Drosophila Schneider II cells and Xenopus oocytes, specifically within the dense fibrillar components. In HeLa cells, DmNopp140-RGG localizes to intact nucleoli, whereas DmNopp140 partitions HeLa nucleoli into phase-light and phase-dark regions. The phase-light regions contain DmNopp140 and endogenous fibrillarin, whereas the phase-dark regions contain endogenous nucleolin. When coexpressed, both Drosophila variants colocalize to HeLa cell nucleoli. Both variants fail to localize to endogenous Cajal bodies in Xenopus oocyte nuclei and in HeLa cell nuclei. Endogenous HeLa coilin, however, accumulates around the periphery of phase-light regions in cells expressing DmNopp140. The carboxy truncation (DmNopp140DeltaRGG) also fails to localize to Cajal bodies, but it forms similar phase-light regions that peripherally accumulate endogenous coilin. Conversely, we see no unusual accumulation of coilin in cells expressing DmNopp140-RGG.

A novel nucleolar G-protein conserved in eukaryotes

We describe here a novel, evolutionarily conserved set of predicted G-proteins. The founding member of this family, TbNOG1, was identified in a two-hybrid screen as a protein that interacts with NOPP44/46, a nucleolar phosphoprotein of Trypanosoma brucei. The biological relevance of the interaction was verified by co-localization and co-immunoprecipitation. TbNOG1 localized to the trypanosome nucleolus and interacted with domains of NOPP44/46 that are found in several other nucleolar proteins. Genes encoding proteins highly related to TbNOG1 are present in yeast and metazoa, and related G domains are found in bacteria. We show that NOG1 proteins in humans and Saccharomyces cerevisae are also nucleolar. The S. cerevisae NOG1 gene is essential for cell viability, and mutations in the predicted G motifs abrogate function. Together these data suggest that NOG1 may play an important role in nucleolar functions. The GTP-binding region of TbNOG1 is similar to those of Obg and DRG proteins, which, together with NOG, form a newly recognized family of G-proteins, herein named ODN. The ODN family differs significantly from other G-protein families, and shows several diagnostic sequence characteristics. All organisms appear to possess an ODN gene, pointing to the biological significance of this family of G-proteins.

Expression-site-associated gene 8 (ESAG8) of Trypanosoma brucei is apparently essential and accumulates in the nucleolus

Trypanosoma brucei variant surface glycoprotein expression sites are interesting examples of genomic loci under complex epigenetic control. In the infectious bloodstream stage, only one of about 20 expression sites is actively transcribed. In the Tsetse midgut (procyclic) stage, chromatin remodeling silences all expression sites. We have begun to explore the function of one of the expression-site-associated genes, ESAG8. Gene knockout experiments implied that ESAG8 is essential. ESAG8 is present at a very low level and apparently accumulates in the nucleolus. A 32-amino-acid domain, which contains a putative bipartite nuclear localization signal (NLS), is both necessary and sufficient to target fusions of ESAG8, with Aequorea victoria green fluorescent protein, to the trypanosome nucleolus. This same sequence functioned only as an NLS in mammalian cells, supporting the idea that nucleolar accumulation requires specific interactions. These results have implications for models of ESAG8 function.

Analysis of targeting sequences demonstrates that trafficking to the Toxoplasma gondii plastid branches off the secretory system

Apicomplexan parasites possess a plastid-like organelle called the apicoplast. Most proteins in the Toxoplasma gondii apicoplast are encoded in the nucleus and imported post-translationally. T. gondii apicoplast proteins often have a long N-terminal extension that directs the protein to the apicoplast. It can be modeled as a bipartite targeting sequence that contains a signal sequence and a plastid transit peptide. We identified two nuclearly encoded predicted plastid proteins and made fusions with green fluorescent protein to study protein domains required for apicoplast targeting. The N-terminal 42 amino acids of the apicoplast ribosomal protein S9 directs secretion of green fluorescent protein, indicating that targeting to the apicoplast proceeds through the secretory system. Large sections of the S9 predicted transit sequence can be deleted with no apparent impact on the ability to direct green fluorescent protein to the apicoplast. The predicted transit peptide domain of the S9 targeting sequence directs protein to the mitochondrion in vivo. The transit peptide can also direct import of green fluorescent protein into chloroplasts in vitro. These data substantiate the model that protein targeting to the apicoplast involves two distinct mechanisms: the first involving the secretory system and the second sharing features with typical chloroplast protein import.