Date Published: December 21, 2017
Publisher: Public Library of Science
Author(s): Alena Zíková, Zdeněk Verner, Anna Nenarokova, Paul A. M. Michels, Julius Lukeš, Laura J. Knoll.
Trypanosoma brucei is a parasitic protist that causes significant health burden in sub-Saharan countries endemic for the tsetse fly (Glossina spp.). During the bloodmeal of this insect vector, the flagellate is transmitted to a variety of mammals, including humans, in which T. brucei subs. gambiense and T. brucei subs. rhodesiense cause human African trypanosomiasis. During its life cycle, T. brucei encounters and adapts to very diverse environments that differ in available nutrients. In the mammalian host, it exists in two major forms: the replicating long-slender bloodstream form (LS-BSF) and the nondividing short-stumpy bloodstream form (SS-BSF), the latter being pre-adapted to infect the insect vector . While the BSF flagellates primarily colonize the mammalian bloodstream and utilize the plentiful glucose for their energy needs, they can also be found in the cerebrospinal fluid and in extracellular spaces of several tissues, including the brain, adipose tissue, and skin [2,3]. In the insect vector, trypanosomes occur in three major forms occupying different locations within the fly: the procyclic form (PCF) resides in the midgut and proventriculus, while epimastigotes and metacyclic trypomastigotes are found in the salivary glands. During the fly’s bloodmeal, the latter form infects the mammalian host. All three forms experience the glucose-poor and amino acid–rich environment within the insect host. These drastic environmental changes encountered by T. brucei during its development require significant morphological and metabolic changes and adaptations [4,5].
To map the BSF mitochondrial proteome (mitoproteome), we first used the available mass spectrometry data of purified PCF mitochondria [29–37] in order to assemble a comprehensive list of mitochondrial proteins. Next, we asked how many of these proteins were identified in any mass spectrometry data obtained from BSF cells [38–43]. To our surprise, out of 1,195 constituents of the PCF mitoproteome, 956 were also identified in at least one study of the BSF, suggesting that, when qualitatively measured, the corresponding mitoproteome is reduced by only approximately 20% (Fig 1; S1 Table). The surprisingly high, approximately 80% overlap with the PCF mitoproteome might also be a consequence of the heterogeneity of the examined BSF populations. The heterogeneity may be related to the experimental protocols, the environmental variations (cells grown in vivo versus in vitro), or variations within the cell cycle (e.g., ATP requirements vary between different cell cycle stages) as well as to the form type (monomorphic versus pleiomorphic). Indeed, some authors analyzed monomorphic strains grown in vitro [40,41], and others examined the pleiomorphic AnTat 1.1 strain grown either in immunosuppressed rats  or in vitro (S1 Table) . Therefore, some LS-BSF cells may have a mitochondrion that is close to the “classical” version, while a subset of these flagellates may express an extended mitoproteome. However, no apparent differences were detected between the mitoproteomes from the pleiomorphic and monomorphic BSF cells, suggesting that, regardless of their status, a surprisingly large repertoire of mitochondrial proteins is expressed in the BSF stage.
The current metabolic model for BSF excludes a role of the mitochondrion in the ATP production by either oxidative or substrate-level phosphorylation . In contrast to this premise, succinyl-CoA synthase (SCoAS), an enzyme responsible for substrate-level phosphorylation of ADP to ATP, has been detected in BSF cells, and more importantly, its RNA interference (RNAi)–mediated silencing produced a severe growth phenotype . This enzyme can be involved in two ATP-producing pathways. The first one includes activity of 2-oxoglutarate dehydrogenase (2-OGDH) producing succinyl-CoA from 2-oxoglutarate that originates from amino acids such as proline and glutamine or can result from transamination reactions by mitochondrial alanine and aspartate transaminases (Fig 2). While all the enzymes involved in these reactions were detected in the LS-BSF mitoproteome (Fig 2 and S1 Table), the activity of 2-OGDH remains contradictory because some authors failed to detect it in the pleiomorphic cells , while others recorded its low activity in culture-adapted monomorphic LS-BSF cells . Puzzlingly, the 2-OGDH subunits E1 and E2 were shown to be essential in BSF not because of their role in carbon metabolism but rather due to their moonlighting roles in glycosomes and mitochondrial DNA maintenance [46,48]. However, in an untargeted metabolomics study using isotope-labeled glucose, up to 30% of excreted succinate remained unlabeled, supporting its nonglucose origin  and making the occurrence of this substrate-level phosphorylation reaction even more plausible (Fig 2).
Combined, the available data reveal that the metabolic flexibility and adaptability of the BSF mitochondrion are much larger than appreciated so far. Mitochondrial metabolism appears to be controlled at various levels; a developmental program seems to be a major contributor, but recent advances in the field suggest that other cues may also play a role through fine-tuning mechanisms. However, the triggers and signaling pathways of these mechanisms remain to be identified. Furthermore, it should be realized that almost all metabolic studies have been performed with strains well adapted to laboratory conditions. While the proteomic data do not show any significant differences between the monomorphic and pleiomorphic strains, future work combining proteomics and metabolomics with functional genomics should be extended to the mitochondrion of trypanosomes isolated not only from blood but also from other tissues to determine whether their metabolism is tissue specific and, if so, what is/are the mechanism(s) that control(s) the changes. Therefore, the virtually unexplored array of pathways and enzymes begs for attention because it may have important implications for drug target identification and future novel chemotherapeutics. Moreover, a decreased morphological complexity, which is apparently not reflected in metabolic complexity, is an interesting and novel phenomenon that can now be efficiently addressed with emerging, increasingly sensitive methods.