A novel role of Rab11 in trafficking GPI-anchored trans-sialidase to the plasma membrane of Trypanosoma cruzi

Abstract

Trypanosoma cruzi, the causative agent of Chagas disease, is a unicellular parasite that possesses a contractile vacuole complex (CVC). This organelle is usually present in free-living protists and is mainly involved in osmoregulation. However, in some organisms, like for example Dictyostelium discoideum, other roles include calcium homeostasis and transference of proteins to the plasma membrane. T. cruzi plasma membrane is very rich in glycosylphosphatidylinositol anchored proteins (GPI-AP) and a very important group of GPI-AP is that of the trans-sialidases. These enzymes catalyze the transfer of sialic acid from host glycoconjugates to mucins present in the surface of the parasite and are important for host cell invasion among other functions. We recently reported that a pathway dependent on the Rab GTPase Rab11 is involved in the traffic of trans-sialidases to the plasma membrane through the CVC of the infective stages of the parasite and that preventing this traffic results in considerable reduction in the ability of T. cruzi to infect host cells. We also found that traffic of other GPI-anchored proteins is also through the CVC but uses a Rab11-independent pathway. These represent unconventional pathways of GPI-anchored protein traffic to the plasma membrane.

Abbreviations

CVC

contractile vacuole complex

GPI-AP

glycosylphosphatidylinositol-anchored protein

GFP

green fluorescent protein

DN

dominant negative

TS

trans-sialidase

Unicellular free-living organisms frequently live in hyposmotic environments and possess a contractile vacuole complex (CVC) to expel water and maintain their volume.¹ In contrast, T. cruzi is subjected to a variety of osmotic changes, mainly hyperosmotic, during its life cycle. Osmolarity in the rectum of the insect vector, where the parasite accumulates before its elimination, can reach values of 1,000 mOsm/kg² while circulating forms in the blood of the mammalian host can be in contact with osmolarities of up to 1,400 mOsm/kg when in the renal medulla circulation.³ Previous studies from our laboratory revealed the osmoregulatory role of the CVC of T. cruzi under both hyposmotic⁴ and hyperosmotic⁵ stress conditions.

When performing the proteomic analysis of the CVC of T. cruzi and the validation of the localization of the proteins identified by expression of green fluorescent protein (GFP)-tagged proteins and fluorescence assays we were surprised by the presence in the CVC of several proteins involved in traffic, fusion and tethering/docking of vesicles, such as SNAREs, VAMP7, AP180, Rab11, and Rab32.⁶ An interesting case is Rab11, a small GTPase that usually is located in recycling endosomes of eukaryotic cells. Rab11 is also localized in the CVC of D. discoideum and it was proposed that this could indicate that the CVC is an evolutionary precursor to the recycling endosomes of other eukaryotes.⁷

One of the functions of Rab11 is targeting proteins to organelles and cell compartments and in Trypanosoma brucei, which belongs to the group of parasites that produce human African trypanosomiasis and lacks a CVC, Rab11 is important for recycling Variant Surface Glycoprotein or VSG.⁸ VSG is a very abundant GPI-AP that covers the surface of these parasites and is responsible for antigenic variation,⁹ thereby helping the parasite to evade the adaptive immune response of the host. T. cruzi lacks VSG but has several surface GPI-APs, among them trans-sialidases (TSs), and mucins. TS genes (more than 1,400 in the T. cruzi genome) are included in several families only one of which is composed by genes encoding proteins with enzymatic activity.¹⁰ The TS enzymes catalyze the transfer of sialic acid from host glycoconjugates to surface mucins present in the parasites. This phenomenon is important to prevent killing by complement ¹¹ and lytic anti-α-Gal antibodies,¹² to attach and infect host cells ¹³ and to exit the parasitophorous vacuole once the parasites are within the host cells.¹⁴ When shed into the circulation TSs induce hematological abnormalities ¹⁵ and alter the immune system.¹⁶

Recently we reported that a number of GPI-AP, like TSs and mucins, accumulate in the CVC of culture and infective stages of T. cruzi before reaching the plasma membrane, and that the CVC is rich in lipid rafts-like structures.¹⁷ After accumulating in the CVC TSs are observed in vesicles in close contact to the plasma membrane and finally they appear in surface clusters, as detected by immunofluorescence assays and transmission electron microscopy¹⁷ (Fig. 1).

Figure 1.

Schematic representation of traffic of trans-sialidases to the plasma membrane of T. cruzi. Trans-sialidase (TS, orange oval) is synthesized in the endoplasmic reticulum (ER), where GPI anchor synthesis, protein synthesis and transamidation occurs. TS undergoes fatty acid remodeling in the Golgi complex (G) and then is transferred to the contractile vacuole complex (CVC) where Rab11 (purple oval) is located. Vesicular transport takes TS to the flagellar pocket (FP) from where it distributes to clusters in the plasma membrane. When dominant negative Rab11 is expressed, TS cannot reach the CVC and is probably degraded by the proteasome. F, flagellum, N, nucleus.

To investigate the role of Rab11 in the traffic of these proteins to the plasma membrane we designed Rab11 dominant negative mutants since T. cruzi lacks the RNA interference pathway and this approach cannot be used. Dominant-negative (DN) Rab mutants compete with endogenous Rabs for binding to Rab effector proteins (REPs), therefore preventing their downstream interaction with target proteins and forming ‘dead-end’ complexes.¹⁸ TcRab11DN-transfected epimastigotes have a punctate cytosolic localization that is maintained in the different life-cycle stages of the parasite. This indicates that coordinated regulation of TcRab11 by correct cycling of GTP and GDP bound forms and binding to effector proteins is instrumental for the correct localization of TcRab11, a phenomenon described for several other Rab proteins. We found that Rab11 dominant negative mutants were not able to transport TSs to the cell surface, while no effect was observed in the traffic of other GPI-AP (TSSA II mucins, gp35/50 mucins and TcMUC II mucins that react with lytic anti-α-Gal antibodies from Chagasic patients) or in the traffic of other membrane proteins like the P-type H⁺-ATPase.¹⁷ Rab11 dominant negative mutants were less infective than controls and infectivity could be partially restored by preincubating the cells with a sialic acid donor and recombinant active, but not inactive, TS suggesting that the reason for the deficient infectivity was the lack of sialylation of mucins by the surface TSs.¹⁷

Our results suggest that after GPI-anchor addition at the endoplasmic reticulum and transfer to the Golgi complex, TSs accumulate in the CVC from where they undergo vesicular transport to the flagellar pocket and/or plasma membrane of the cells. In Rab11 dominant negative mutants TSs cannot accumulate in the CVC and are probably degraded by the proteasome (Fig. 1).

In addition to its function in trafficking TSs to the plasma membrane, CVC-located Rab11 has also a function in osmoregulation. Rab11 dominant negative mutants were less able to undergo regulatory volume decrease after hyposmotic stress and were less able to shrink under hyperosmotic stress.¹⁷ There is evidence that acidocalcisomes, acidic calcium stores rich in polyphosphate, fuse to the CVC of T. cruzi resulting in translocation of an aquaporin (TcAQP1) during osmotic stress and this fusion is necessary for cell recovery.⁴ The results suggest that Rab11 is probably also involved in fusion of acidocalcisomes to the CVC.

In conclusion these new results reveal an unexpected function of the CVC of T. cruzi in trafficking of GPI-AP to the plasma membrane, the specific role of Rab11 in trafficking of TSs, and in osmoregulation, and the demonstration that sialylation of mucins by TSs is essential for host cell invasion by T. cruzi, a role that has been difficult to demonstrate before because of the impossibility of knocking out all the genes of this numerous gene family.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Acknowledgments

We thank Christina Moore for the drawing in Figure 1.

Funding

Work reported here was funded by the US. National Institutes of Health (grant A1-107663 to RD).