We analysed a widely used barcode, the V9 region of the 18S rRNA gene, to study the effect of environmental conditions on the distribution of two related heterotrophic protistan lineages in marine plankton, kinetoplastids and diplonemids. We relied on a major published dataset (Tara Oceans) where samples from the mesopelagic zone were available from just 32 of 123 locations, and both groups are most abundant in this zone. To close sampling gaps and obtain more information from the deeper ocean, we collected 57 new samples targeting especially the mesopelagic zone. We sampled in three geographic regions: the Arctic, two depth transects in the Adriatic Sea, and the anoxic Cariaco Basin. In agreement with previous studies, both protist groups are most abundant and diverse in the mesopelagic zone. In addition to that, we found that their abundance, richness, and community structure also depend on geography, oxygen concentration, salinity, temperature, and other environmental variables reflecting the abundance of algae and nutrients. Both groups studied here demonstrated similar patterns, although some differences were also observed. Kinetoplastids and diplonemids prefer tropical regions and nutrient-rich conditions and avoid high oxygen concentration, high salinity, and high density of algae.

Background ApaH like phosphatases (ALPHs) originate from the bacterial ApaH protein and have been identified in all eukaryotic super-groups. Only two of these proteins have been functionally characterised. We have shown that the ApaH like phosphatase ALPH1 from the Kinetoplastid Trypanosoma brucei is the mRNA decapping enzyme of the parasite. In eukaryotes, Dcp2 is the major mRNA decapping enzyme and mRNA decapping by ALPHs is unprecedented, but the bacterial ApaH protein was recently found decapping non-conventional caps of bacterial mRNAs. These findings prompted us to explore whether mRNA decapping by ALPHs is restricted to Kinetoplastida or could be more widespread among eukaryotes. Results We screened 827 eukaryotic proteomes with a newly developed Python-based algorithm for the presence of ALPHs and used the data to characterize the phylogenetic distribution, conserved features, additional domains and predicted intracellular localisation of this protein family. For most organisms, we found ALPH proteins to be either absent (495/827 organisms) or to have non-cytoplasmic localisation predictions (73% of all ALPHs), excluding a function in mRNA decapping. Although, non-cytoplasmic ALPH proteins had in vitro mRNA decapping activity. Only 71 non-Kinetoplastida have ALPH proteins with predicted cytoplasmic localisations. However, in contrast to Kinetoplastida, these organisms also possess a homologue of Dcp2 and in contrast to ALPH1 of Kinetoplastida, these ALPH proteins are very short and consist of the catalytic domain only. Conclusions ALPH was present in the last common ancestor of eukaryotes, but most eukaryotes have either lost the enzyme, or use it exclusively outside the cytoplasm. The acceptance of mRNA as a substrate indicates that ALPHs, like bacterial ApaH, have a wide substrate range: the need to protect mRNAs from unregulated degradation is one possible explanation for the selection against the presence of cytoplasmic ALPH proteins in most eukaryotes. Kinetoplastida succeeded to exploit ALPH as their only or major mRNA decapping enzyme. 71 eukaryotic organisms outside the Kinetoplastid lineage have short ALPH proteins with cytoplasmic localisation predictions: whether these proteins are used as decapping enzymes in addition to Dcp2 or else have adapted to not accept mRNAs as a substrate, remains to be explored.

Background: ApaH like phosphatases (ALPHs) originate from the bacterial ApaH protein and are present in eukaryotes of all eukaryotic super-groups; still, only two proteins have been functionally characterised. One is ALPH1 from the Kinetoplastid Trypanosoma brucei that we recently found to be the mRNA decapping enzyme of the parasite . mRNA decapping by ALPHs is unprecedented in eukaryotes, which usually use nudix hydrolases, but the bacterial ancestor protein ApaH was recently found to decap non-conventional caps of bacterial mRNAs. These findings prompted us to explore whether mRNA decapping by ALPHs is restricted to Kinetoplastida or more widespread among eukaryotes. Results: We screened 824 eukaryotic proteomes with a newly developed Python-based algorithm for the presence of ALPHs and used the data to refine phylogenetic distribution, conserved features, additional domains and predicted intracellular localisation of ALPHs. We found that most eukaryotes have either no ALPH (500/824) or very short ALPHs, consisting almost exclusively of the catalytic domain. These ALPHs had mostly predicted non-cytoplasmic localisations, often supported by the presence of transmembrane helices and signal peptides and in two cases (one in this study) by experimental data. The only exceptions were ALPH1 homologues from Kinetoplastida, that all have unique C-terminal and mostly unique N-terminal extension, and at least the T. brucei enzyme localises to the cytoplasm. Surprisingly, despite of these non-cytoplasmic localisations, ALPHs from all eukaryotic super-groups had in vitro mRNA decapping activity. Conclusions: ALPH was present in the last common ancestor of eukaryotes, but most eukaryotes have either lost the enzyme since, or use it exclusively outside the cytoplasm in organelles in a version consisting of the catalytic domain only. While our data provide no evidence for the presence of further mRNA decapping enzymes among eukaryotic ALPHs, the broad substrate range of ALPHs that includes mRNA caps provides an explanation for the selection against the presence of a cytoplasmic ALPH protein as a mean to protect mRNAs from unregulated degradation. Kinetoplastida succeeded to exploit ALPH as their mRNA decapping enzyme, likely using the Kinetoplastida-unique N- and C-terminal extensions for regulation.