Continuous endosomes form functional subdomains and orchestrate rapid membrane trafficking in trypanosomes

Reviewer #1 (Public Review):

Summary: Bloodstream stages of the parasitic protist, Trypanosoma brucei, exhibit very high rates of constitutive endocytosis, which is needed to recycle the surface coat of Variant Surface Glycoproteins (VSGs) and remove surface immune complexes. While many studies have shown that the endo-lysosomal systems of T. brucei BF stages contain canonical domains, as defined by classical Rab markers, it has remained unclear whether these protists have evolved additional adaptations/mechanisms for sustaining these very high rates of membrane transport and protein sorting. The authors have addressed this question by reconstructing the 3D ultrastructure and functional domains of the T. brucei BF endosome membrane system using advanced electron tomography and super-resolution microscopy approaches. Their studies reveal that, unusually, the BF endosome network comprises a continuous system of cisternae and tubules that contain overlapping functional subdomains. It is proposed that a continuous membrane system allows higher rates of protein cargo segregation, sorting and recycling than can otherwise occur when transport between compartments is mediated by membrane vesicles or other fusion events.

Strengths: The study is a technical tour-de-force using a combination of electron tomography, super-resolution/expansion microscopy, immune-EM of cryo-sections to define the 3D structures and connectivity of different endocytic compartments. The images are very clear and generally support the central conclusion that functionally distinct endocytic domains occur within a dynamic and continuous endosome network in BF stages.

Weaknesses: The authors suggest that this dynamic endocytic network may also fulfil many of the functions of the Golgi TGN and that the latter may be absent in these stages. Although plausible, this comment needs further experimental support. For example, have the authors attempted to localize canonical makers of the TGN (e.g. GRIP proteins) in T. brucei BF and/or shown that exocytic carriers bud directly from the endosomes?

Reviewer #2 (Public Review):

The authors suggest that the African trypanosome endomembrane system has unusual organisation, in that the entire system is a single reticulated structure. It is not clear if this is thought to extend to the lysosome or MVB. There is also a suggestion that this unusual morphology serves as a trans-(post)Golgi network rather than the more canonical arrangement.

The work is based around very high-quality light and electron microscopy, as well as utilising several marker proteins, Rab5A, 11 and 7. These are deemed as markers for early endosomes, recycling endosomes and late or pre-lysosomes. The images are mostly of high quality but some inconsistencies in the interpretation, appearance of structures and some rather sweeping assumptions make this less easy to accept. Two perhaps major issues are claims to label the entire endosomal apparatus with a single marker protein, which is hard to accept as certainly this reviewer does not really even know where the limits to the endosomal network reside and where these interface with other structures. There are several additional compartments that have been defined by Rob proteins as well, and which are not even mentioned. Overall I am unconvinced that the authors have demonstrated the main things they claim.

The approaches taken are state-of-the-art but not novel, and because of the difficulty in fully addressing the central tenet, I am not sure how much of an impact this will have beyond the trypanosome field. For certain this is limited to workers in the direct area and is not a generalisable finding.

Reviewer #3 (Public Review):

Summary:
As clearly highlighted by the authors, a key plank in the ability of trypanosomes to evade the mammalian host's immune system is its high rate of endocytosis. This rapid turnover of its surface enables the trypanosome to 'clean' its surface removing antibodies and other immune effectors that are subsequently degraded. The high rate of endocytosis is likely reflected in the organisation and layout of the endosomal system in these parasites. Here, Link et al., sought to address this question using a range of light and three-dimensional electron microscopy approaches to define the endosomal organisation in this parasite.

Before this study, the vast majority of our information about the make-up of the trypanosome endosomal system was from thin-section electron microscopy and immunofluorescence studies, which did not provide the necessary resolution and 3D information to address this issue. Therefore, it was not known how the different structures observed by EM were related. Link et al., have taken advantage of the advances in technology and used an impressive combination of approaches at the LM and EM level to study the endosomal system in these parasites. This innovative combination has now shown the interconnected-ness of this network and demonstrated that there are no 'classical' compartments within the endosomal system, with instead different regions of the network enriched in different protein markers (Rab5a, Rab7, Rab11).

Strengths:
This is a generally well-written and clear manuscript, with the data well-presented supporting the majority of the conclusions of the authors. The authors use an impressive range of approaches to address the organisation of the endosomal system and the development of these methods for use in trypanosomes will be of use to the wider parasitology community.

I appreciate their inclusion of how they used a range of different light microscopy approaches even though for instance the dSTORM approach did not turn out to be as effective as hoped. The authors have clearly demonstrated that trypanosomes have a large interconnected endosomal network, without defined compartments and instead show enrichment for specific Rabs within this network.

Weaknesses:
My concerns are:

i) there is no evidence for functional compartmentalisation. The classical markers of different endosomal compartments do not fully overlap but there is no evidence to show a region enriched in one or other of these proteins has that specific function. The authors should temper their conclusions about this point.

ii) the quality of the electron microscopy work is very high but there is a general lack of numbers. For example, how many tomograms were examined? How often were fenestrated sheets seen? Can the authors provide more information about how frequent these observations were?

iii) the EM work always focussed on cells which had been processed before fixing. Now, I understand this was important to enable tracers to be used. However, given the dynamic nature of the system these processing steps and feeding experiments may have affected the endosomal organisation. Given their knowledge of the system now, the authors should fix some cells directly in culture to observe whether the organisation of the endosome aligns with their conclusions here.

iv) the discussion needs to be revamped. At the moment it is just another run through of the results and does not take an overview of the results presenting an integrated view. Moreover, it contains reference to data that was not presented in the results.