Queuosine Formation in Eukaryotic tRNA Occurs via a Mitochondria-localized Heteromeric Transglycosylase

tRNA guanine transglycosylase (TGT) enzymes are responsible for the formation of queuosine in the anticodon loop (position 34) of tRNA^Asp, tRNA^Asn, tRNA^His, and tRNA^Tyr; an almost universal event in eubacterial and eukaryotic species. Despite extensive characterization of the eubacterial TGT the eukaryotic activity has remained undefined. Our search of mouse EST and cDNA data bases identified a homologue of the Escherichia coli TGT and three spliced variants of the queuine tRNA guanine transglycosylase domain containing 1 (QTRTD1) gene. QTRTD1 variant_1 (Qv1) was found to be the predominant adult form. Functional cooperativity of TGT and Qv1 was suggested by their coordinate mRNA expression in Northern blots and from their association in vivo by immunoprecipitation. Neither TGT nor Qv1 alone could complement a tgt mutation in E. coli. However, transglycosylase activity could be obtained when the proteins were combined in vitro. Confocal and immunoblot analysis suggest that TGT weakly interacts with the outer mitochondrial membrane possibly through association with Qv1, which was found to be stably associated with the organelle.Queuosine (Q³; (7-{[(4,5-cis-dihydroxy-2-cyclo-penten-1-yl)-amino]methyl}-7-deazaguanosine) is a modified 7-deazaguanosine molecule found at the wobble position of transfer RNA that contains a GUN anticodon sequence: tRNA^Tyr, tRNA^Asn, tRNA^His, and tRNA^Asp (1). The Q-modification is widely distributed in nature in the tRNA of eubacteria, plants, and animals; a notable exception being yeast and plant leaf cells (2, 3). Interestingly, Q-modification has also been detected in aspartyl tRNA from mitochondria of rat (4) and opossum (5). In most eukaryotes, the Q molecule can be further modified by the addition of a mannosyl group to Q-tRNA^Asp and a galactosyl group to Q-tRNA^Tyr (1).Eubacteria are unique in their ability to synthesize Q. As part of this biosynthetic process, the eubacterial tRNA guanine transglycosylase (TGT) enzyme inserts the Q precursor molecule, 7-aminomethyl-7-deazaguanine (preQ₁) into tRNA, which is then converted to Q by two further enzymatic steps at the tRNA level (6). Eukaryotes by contrast salvage queuosine from food and enteric bacteria either as the free base (referred to as queuine) or as queuosine 5′-phosphate subsequent to normal tRNA turnover (7). A Q-related molecule, archaeosine, is found at position 15 of the D loop of most archaeal tRNA, where it functions to stabilize the tRNA structure (8). The enzyme involved in archaeosine biosynthesis is structurally and mechanistically related to the eubacterial TGT but with adaptations necessitated by the differences imposed by its unique substrate and tRNA specificity (9, 10).The crystal structure of the Zymononas mobilis (Z. mobilis) TGT has been determined and revealed the enzyme to be an irregular (β/α)₈ TIM barrel with a C-terminal zinc-binding subdomain (11). Insight into the residues involved in catalysis came from mutational and kinetic analysis of the recombinant Escherichia coli enzyme (12) and from the Z. mobilis TGT structure as an RNA-bound intermediate complexed to the final preQ₁-modified RNA product (13). This work showed the essential role of Asp-280 (Z. mobilis numbering) as the active site nucleophile. Asp-102, which was originally ascribed the role of active site nucleophile, functions as a general acid/base during catalysis (12, 10). Although, the E. coli and Z. mobilis TGT enzymes are monomeric in solution (14), at high protein concentrations the E. coli enzyme can oligomerize (15), and structural data from the Z. mobilis TGT has shown the formation of a 2:1 complex with tRNA; a possible functional requirement for catalysis (10).In contrast to the eubacterial enzyme, which is a single protein species, purification of the eukaryotic TGT suggested that the catalytically active enzyme is a heterodimeric molecule: subunits of 60 and 43 kDa in rabbit erythrocytes (16), 66 and 32 kDa in bovine liver (17), 60 and 34.5 kDa in rat liver (18), and a homodimer of two 68-kDa proteins in wheat germ (16, 19). A partial amino acid sequence was recovered from two of these active enzyme preparations. The identity of the proteins from bovine liver (17) could not be assigned at the time of publication. However, our searches show that the peptides from the larger 65-kDa subunit are identical to asparaginyl tRNA synthetase, and those of the smaller 32-kDa subunit correspond to 2,4-dienoyl CoA reductase. A highly pure preparation from rabbit reticulocytes (20) gave peptides with homology to the immunophilin p59, human elongation factor 2 (EF2), and a deubiquitinating enzyme, USP14. It is noteworthy that none of the peptide sequences obtained showed similarity to the eubacterial TGT. The results do suggest, however, that in eukaryotes the TGT activity could be embedded in a multisubunit complex.Most recently, Deshpande and Katze (21) identified a cDNA clone encoding a putative TGT catalytic subunit. Cloning the cDNA into a mammalian expression plasmid reconstituted TGT activity in GC₃/c1 cells, which are known to be naturally deficient in Q-containing tRNA (22). In this study, we identify for the first time the composition of the eukaryotic tRNA guanine transglycosylase, reconstitute the catalytic activity in vitro, and examine the intracellular distribution of the active subunits.     

Dual Targeting of a tRNAAsp Requires Two Different Aspartyl-tRNA Synthetases in Trypanosoma brucei

The mitochondrion of the parasitic protozoon Trypanosoma brucei does not encode any tRNAs. This deficiency is compensated for by partial import of nearly all of its cytosolic tRNAs. Most trypanosomal aminoacyl-tRNA synthetases are encoded by single copy genes, suggesting the use of the same enzyme in the cytosol and in the mitochondrion. However, the T. brucei genome encodes two distinct genes for eukaryotic aspartyl-tRNA synthetase (AspRS), although the cell has a single tRNA^Asp isoacceptor only. Phylogenetic analysis showed that the two T. brucei AspRSs evolved from a duplication early in kinetoplastid evolution and also revealed that eight other major duplications of AspRS occurred in the eukaryotic domain. RNA interference analysis established that both Tb-AspRS1 and Tb-AspRS2 are essential for growth and required for cytosolic and mitochondrial Asp-tRNA^Asp formation, respectively. In vitro charging assays demonstrated that the mitochondrial Tb-AspRS2 aminoacylates both cytosolic and mitochondrial tRNA^Asp, whereas the cytosolic Tb-AspRS1 selectively recognizes cytosolic but not mitochondrial tRNA^Asp. This indicates that cytosolic and mitochondrial tRNA^Asp, although derived from the same nuclear gene, are physically different, most likely due to a mitochondria-specific nucleotide modification. Mitochondrial Tb-AspRS2 defines a novel group of eukaryotic AspRSs with an expanded substrate specificity that are restricted to trypanosomatids and therefore may be exploited as a novel drug target.In most animal and fungal mitochondria, the total set of tRNAs required for translation is encoded on the mitochondrial genome and thus of bacterial evolutionary origin. The aminoacyl-tRNA synthetases (aaRSs)² responsible for charging of mitochondrial tRNAs are always nuclear encoded and need to be imported into mitochondria. We therefore expect to find two sets of aaRSs, one for cytosolic aminoacyl-tRNA synthesis and a second one, of bacterial evolutionary origin, for aminoacylation of mitochondrial tRNAs (1, 2).In most cells, however, some aaRSs are targeted to both the cytosol as well as to mitochondria (3). In Saccharomyces cerevisiae, for example, four aaRSs are double-targeted to both compartments, indicating that they are able to aminoacylate tRNAs of both eukaryotic and bacterial evolutionary origin (4–6). In plants, the situation is more complex, since protein synthesis occurs in three compartments: the cytosol, the mitochondria, and the plastids. A recent analysis in Arabidopsis has shown that, rather than having three unique sets of aaRSs specific for the three translation systems, more than 15 aaRSs were dually targeted to the mitochondria and the plastid (7). Moreover, there is at least one aaRS that is shared between all three compartments. In summary, these examples indicate that the overlap between the different sets of aaRSs used in the various translation systems is variable and can be extensive.Most eukaryotes, except many animals and fungi, lack a variable number of mitochondrial tRNA genes. Mitochondrial translation in these organisms depends on import of a small fraction of the corresponding nucleus-encoded cytosolic tRNAs (8–10). As a consequence, imported tRNAs are always of eukaryotic evolutionary origin. An intriguing situation is found in trypanosomatids (such as Trypanosoma brucei and Leishmania spp.), where all mitochondrial tRNA genes have apparently been lost and all mitochondrial tRNAs are imported from the cytosol. In these organisms, all mitochondrial tRNAs derive from cytosolic tRNAs (11). It is therefore reasonable to assume that trypanosomal aaRSs are dually targeted to the cytosol and the mitochondrion. For the T. brucei glutaminyl-tRNA synthetase (GlnRS) and the glutamyl-tRNA synthetase, the dual localization has been shown experimentally (12). Moreover, dual targeting of essentially all aaRSs is suggested by the fact that the genome of T. brucei and other trypanosomatids encodes only 23 distinct aaRSs, fewer than any other eukaryote that has a mitochondrial translation system (13). Unexpectedly, two distinct genes were found for the tryptophanyl-tRNA synthetase (TrpRS), the lysyl-tRNA synthetase and the aspartyl-tRNA synthetase (AspRS). A recent study has shown that the two trypanosomal TrpRSs are required for cytosolic and mitochondrial tryptophanyl-tRNA formation (14). Trypanosomal tRNA^Trp is imported to the mitochondria, where it undergoes C to U editing at the wobble nucleotide and is thiolated at position 33. The RNA editing is required to decode the reassigned mitochondrial tryptophan codon UGA (14–16). Both nucleotide modifications are antideterminants for the cytosolic TrpRS (14). As we concluded previously (14), the presence of a second TrpRS with expanded substrate specificity is required to efficiently aminoacylate imported, mature tRNA^Trp in trypanosomal mitochondria.The present study focuses on the characterization and functional analysis of another pair of duplicated trypanosomal aaRSs, the AspRSs. We show that the two enzymes are individually essential for normal growth of insect stage T. brucei. We also demonstrate that the two trypanosomal AspRSs are of eukaryotic evolutionary origin and that the aminoacylation of the cytosolic and mitochondrial tRNA^Asp species requires these two distinct AspRSs.     

Defects in Transient tRNA Translocation Bypass tRNA Synthetase Quality Control Mechanisms

Quality control mechanisms during protein synthesis are essential to fidelity and cell survival. Leucyl-tRNA synthetase (LeuRS) misactivates non-leucine amino acids including isoleucine, methionine, and norvaline. To prevent translational errors, mischarged tRNA products are translocated 30Å from the canonical aminoacylation core to a hydrolytic editing-active site within a completely separate domain. Because it is transient, the tRNA translocation mechanism has been difficult to isolate. We have identified a “translocation peptide” within Escherichia coli LeuRS. Mutations in the translocation peptide cause tRNA to selectively bypass the editing-active site, resulting in mischarging that is lethal to the cell. This bypass mechanism also rescues aminoacylation of an editing site mutation that hydrolyzes correctly charged Leu-tRNA^Leu. Thus, these LeuRS mutants charge tRNA^Leu but fail to translocate these products to the hydrolytic site, where they are cleared to guard against genetic code ambiguities.Quality control during translation depends on the family of aminoacyl-tRNA synthetases (aaRSs),² which is responsible for the first step of protein synthesis. Each aaRS selectively aminoacylates just one of the 20 standard amino acids to its cognate tRNA (1). About half of this family of enzymes ensures fidelity by employing a “double sieve model” that relies on two active sites (2, 3). One sieve is synthetic and produces charged tRNA. The other is a hydrolytic editing-active site that clears mistakes. Defects in the editing mechanism cause cell death (4, 5) and also neurological disease in mammals (6).The aminoacylation site in the ancient canonical core of the aaRS activates its cognate amino acid but can also misactivate structurally similar amino acids (1). The editing-active site blocks the correctly charged amino acid (7, 8) and hydrolyzes mischarged amino acids from the tRNA. Amino acid editing destroys mistakes before they can be incorporated by the ribosome, which would result in the production of statistical proteins (1).Amino acid proofreading requires that the charged tRNA transiently migrates between two enzyme domains that are responsible for aminoacylation and editing. For leucyl-tRNA synthetase (LeuRS) and the homologous isoleucyl-(IleRS) and valyl-tRNA synthetases (ValRS), the editing domain resides in a structural insertion called CP1 (9) that splits the Rossmann ATP binding fold. The insert folds independent of the canonical core (10–12). The isolated CP1 domains from LeuRS, ValRS, and IleRS can independently and specifically hydrolyze mischarged amino acid from its cognate tRNA (13–15).The aminoacylation and editing-active sites of LeuRS are separated by about 30 Å. Thus, the charged 3′ end of the tRNA must be faithfully translocated a significant distance for proofreading and then hydrolysis if it is mischarged (16). It has also been suggested that the tRNA 3′ end binds initially near the editing-active site and requires translocation to the aminoacylation site (17).We hypothesized that flexible molecular hinges might facilitate conformational changes between the aminoacylation and the editing complexes (18). Two putative hinge sites were predicted by computational analysis of Thermus thermophilus LeuRS. One hinge at Ser-227 was located in the N-terminal β-strand that links the aminoacylation and CP1 editing domains (18). Mutations at the predicted hinge site in the β-strand linker of Escherichia coli LeuRS abolished aminoacylation activity and significantly decreased amino acid editing activity (18).A second hinge site at Glu-393 was identified in a flexible peptide within the CP1 domain of T. thermophilus LeuRS (18). Here, we describe results at a homologous Asp-391 site in E. coli LeuRS that demonstrate that this hinge comprises a portion of a translocation peptide. Unlike the predicted β-strand hinge mutation, the aminoacylation and editing activities of the CP1 domain-based hinge mutants in LeuRS were intact. Surprisingly however, mutations within the translocation peptide yield mischarged tRNA despite a robust deacylation activity. We hypothesize that impairing the LeuRS translocation peptide causes the charged tRNA 3′ end to bypass the editing sieve prior to product release. Defects in the translocation peptide and its mechanism result in amino acid toxicities that are lethal to the cell.     

Catalytic Properties of RNase BN/RNase Z from Escherichia coli: RNase BN IS BOTH AN EXO- AND ENDORIBONUCLEASE*

Processing of the 3′ terminus of tRNA in many organisms is carried out by an endoribonuclease termed RNase Z or 3′-tRNase, which cleaves after the discriminator nucleotide to allow addition of the universal -CCA sequence. In some eubacteria, such as Escherichia coli, the -CCA sequence is encoded in all known tRNA genes. Nevertheless, an RNase Z homologue (RNase BN) is still present, even though its action is not needed for tRNA maturation. To help identify which RNA molecules might be potential substrates for RNase BN, we carried out a detailed examination of its specificity and catalytic potential using a variety of synthetic substrates. We show here that RNase BN is active on both double- and single-stranded RNA but that duplex RNA is preferred. The enzyme displays a profound base specificity, showing no activity on runs of C residues. RNase BN is strongly inhibited by the presence of a 3′-CCA sequence or a 3′-phosphoryl group. Digestion by RNase BN leads to 3-mers as the limit products, but the rate slows on molecules shorter than 10 nucleotides in length. Most interestingly, RNase BN acts as a distributive exoribonuclease on some substrates, releasing mononucleotides and a ladder of digestion products. However, RNase BN also cleaves endonucleolytically, releasing 3′ fragments as short as 4 nucleotides. Although the presence of a 3′-phosphoryl group abolishes exoribonuclease action, it has no effect on the endoribonucleolytic cleavages. These data suggest that RNase BN may differ from other members of the RNase Z family, and they provide important information to be considered in identifying a physiological role for this enzyme.Maturation of tRNA precursors requires the removal of 5′ and 3′ precursor-specific sequences to generate the mature, functional tRNA (1). In eukaryotes, archaea, and certain eubacteria, the 3′-processing step is carried out by an endoribonuclease termed RNase Z or 3′-tRNase (2–6). However, in some bacteria, such as Escherichia coli, removal of 3′ extra residues is catalyzed by any of a number of exoribonucleases (7, 8). The major determinant for which mode of 3′-processing is utilized appears to be whether or not the universal 3′-terminal CCA sequence is encoded (2, 9). Thus, for those tRNA precursors in which the CCA sequence is absent, endonucleolytic cleavage by RNase Z right after the discriminator nucleotide generates a substrate for subsequent CCA addition by tRNA nucleotidyltransferase (1–3, 10). In view of this role for RNase Z in 3′-tRNA maturation, it is surprising that E. coli, an organism in which the CCA sequence is encoded in all tRNA genes (2), nevertheless contains an RNase Z homologue (11), because its action would appear not to be necessary. In fact, the physiological function of this enzyme in E. coli remains unclear, because mutants lacking this protein have no obvious growth phenotype (12). Hence, there is considerable interest in understanding the enzymatic capabilities of this enzyme.The E. coli RNase Z homologue initially was identified as a zinc phosphodiesterase (11) encoded by the elaC gene (now called rbn) (13). Subsequent work showed that the protein also displayed endoribonuclease activity on certain tRNA precursors in vitro (6, 14). However, more recent studies revealed that this protein actually is RNase BN, an enzyme originally discovered in 1983 and shown to be essential for maturation of those bacteriophage T4 tRNA precursors that lack a CCA sequence (15, 16). Using synthetic mimics of these T4 tRNA precursors, RNase BN was found to remove their 3′-terminal residue as a mononucleotide to generate a substrate for tRNA nucleotidyltransferase. Based on these reactions RNase BN was originally thought to be an exoribonuclease (13, 15, 17). However, subsequent work by us and others showed that it can act as an endoribonuclease on tRNA precursors (13, 18). RNase BN is required for maturation of tRNA precursors in E. coli mutant strains devoid of all other 3′-tRNA maturation exoribonucleases, although it is the least efficient RNase in this regard (7, 19). Thus, under normal circumstances, it is unlikely that RNase BN functions in maturation of tRNA in vivo except in phage T4-infected cells (15, 16).To obtain additional information on what types of RNA molecules might be substrates for RNase BN and to clarify whether it is an exo- or endoribonuclease, we have carried out a detailed examination of its catalytic properties and substrate specificity. We show here that RNase BN has both exo- and endoribonuclease activity and that it can act on a wide variety of RNA substrates. These findings suggest that E. coli RNase BN may differ from other members of the RNase Z family of enzymes.     

The N Terminus of Phosphodiesterase TbrPDEB1 of Trypanosoma brucei Contains the Signal for Integration into the Flagellar Skeleton

The precise subcellular localization of the components of the cyclic AMP (cAMP) signaling pathways is a crucial aspect of eukaryotic intracellular signaling. In the human pathogen Trypanosoma brucei, the strict control of cAMP levels by cAMP-specific phosphodiesterases is essential for parasite survival, both in cell culture and in the infected host. Among the five cyclic nucleotide phosphodiesterases identified in this organism, two closely related isoenzymes, T. brucei PDEB1 (TbrPDEB1) (PDEB1) and TbrPDEB2 (PDEB2) are predominantly responsible for the maintenance of cAMP levels. Despite their close sequence similarity, they are distinctly localized in the cell. PDEB1 is mostly located in the flagellum, where it forms an integral part of the flagellar skeleton. PDEB2 is mainly located in the cell body, and only a minor part of the protein localizes to the flagellum. The current study, using transfection of procyclic trypanosomes with green fluorescent protein (GFP) reporters, demonstrates that the N termini of the two enzymes are essential for determining their final subcellular localization. The first 70 amino acids of PDEB1 are sufficient to specifically direct a GFP reporter to the flagellum and to lead to its detergent-resistant integration into the flagellar skeleton. In contrast, the analogous region of PDEB2 causes the GFP reporter to reside predominantly in the cell body. Mutagenesis of selected residues in the N-terminal region of PDEB2 demonstrated that single amino acid changes are sufficient to redirect the reporter from a cell body location to stable integration into the flagellar skeleton.In eukaryotes, the ubiquitous second messenger cyclic AMP (cAMP) is generated from ATP by membrane-integral or by cytoplasmic, CO₂-regulated cyclases (35, 44). The cAMP signal is processed by a small group of receiver proteins, including the regulatory subunit of protein kinase A (28), cAMP-gated ion channels (4), and the guanine-nucleotide-exchange proteins EPAC1 and EPAC2 (39). The cAMP signal is terminated by the action of a family of cyclic nucleotide-specific phosphodiesterases (PDEs) (9). This paradigm is rather straightforward, involves a limited number of players, and is generally well understood, at least in mammalian cells. However, much less is known about how individual cAMP signals are temporally and spatially controlled. Since most eukaryotic adenylyl cyclases are integral membrane proteins, often restricted to specific membrane subdomains (10), cAMP signaling is usually initiated at the cell membrane (40). However, diffusion of cAMP away from its site of generation is rapid, with diffusion coefficients being about 400 μm²/s (8, 15, 29), translating into diffusion velocities of 30 to 40 μm/s. As a consequence, the signal would reach the center of the cell with a diameter of 3 μm within less than 50 ms and would rapidly saturate the entire cell. While regulation through fluctuating cellular levels of cAMP represents a valid paradigm of cAMP signaling, it has become clear that other, more localized modes of cAMP signaling must also exist. Several groups have shown that the cAMP response of a given cell can differ depending on what set of receptors activates the cyclase response (14, 30, 41, 42). Similarly, the cAMP response of endothelial cells depends on the subcellular site where the cAMP is produced. They tighten their barrier function when cAMP is produced by membrane-bound adenylyl cyclases but become more permeable when cAMP is produced in the cytoplasm (17, 45). The distinct subcellular localization of cAMP signals was experimentally demonstrated using an array of techniques (29, 40, 55, 56).Physically tethered PDEs might serve to confine newly synthesized cAMP to defined microdomains. Only cAMP-binding proteins that are localized within or extend into such microdomains would be able to receive the cAMP signal (17, 49). cAMP concentrations within such domains might rise and fall rapidly, reaching peak concentrations much more rapidly and locally far beyond the steady-state cAMP levels measured in whole-cell extracts. Such spatially organized, tethered PDEs can generate local sinks into which cAMP disappears (1, 23). This paradigm would allow the simultaneous presence of numerous local cAMP concentration gradients within a single cell, allowing great flexibility in signal generation and intracellular signal transmission. This concept is based on the distinct subcellular localization and physical association of PDEs with subcellular structures and on the existence of localized subcellular cAMP pools, for which there is extensive experimental support (3, 5, 13, 50, 52). Interestingly, PDEs localized in different subcellular regions may still be able to compensate for each other. Ablation of the cilium-specific PDE1C from the olfactory neurons in the mouse did not prolong response termination, as long as the cytoplasmic PDE4 in the cell body was still present (11).The unicellular eukaryote Trypanosoma brucei is the causative agent of human sleeping sickness in sub-Saharan Africa. It belongs to the large order of the kinetoplastida, which includes many medically and economically important pathogens of humans, their livestock, and their crops worldwide (27). Trypanosomes are very small cells (about 15 by 3 μm in diameter) that carry a single flagellum (10 by 0.5 μm). The volume of a procyclic trypanosome of strain 427 is (9.6 ± 0.8) × 10⁻¹⁴ liter (Markus Engstler, personal communication), with the flagellum representing about 15% of this. A signaling threshold concentration of 1 μM cAMP corresponds to just about 30,000 molecules of cAMP per cell. Given a diffusion coefficient of 400 μm²/s (29), unrestricted diffusion of cAMP would swamp the cell within 50 ms. Obviously, temporal and spatial control of cAMP signaling is crucial for T. brucei. Strategically located, physically tethered PDEs might thus play an important role in the architecture of the cAMP signaling pathways in T. brucei.The genomes of T. brucei and of other kinetoplastids, such as T. vivax, T. cruzi, Leishmania major, L. infantum, and L. braziliensis, all code for the same set of five cyclic nucleotide-specific PDEs (25, 53). In T. brucei, the genes for T. brucei PDEB1 (TbrPDEB1; subsequently termed PDEB1) and TbrPDEB2 (PDEB2) are tandemly arranged on chromosome 9 and code for two very similar cAMP-specific PDEs, each with two GAF (mammalian cyclic GMP-dependent PDEs, Anabaena adenylyl cyclases, Escherichia coli FhlA) domains (21) in their N-terminal regions (38, 57). These two PDEs were also studied experimentally in T. cruzi (12) and L. major (24, 52), and orthologues are present in all kinetoplastid genomes available so far. Despite their high overall sequence similarity, PDEB1 and PDEB2 exhibit distinct subcellular localizations (31). PDEB1 is predominantly found in the flagellum, where it is stably associated with cytoskeletal components that are resistant to detergent extraction. In contrast, PDEB2 is mostly localized in the cell body, from where it is fully extractable by nonionic detergents. However, a minor fraction of PDEB2 also associates with the flagellar skeleton in a Triton-resistant manner, most likely through interaction with PDEB1. Earlier work has shown that both PDEB1and PDEB2 are essential enzymes in bloodstream-form T. brucei (31), while TbPDEA, TbPDEC, and TbPDED play minor roles (20; S. Kunz, unpublished data).