Intrabody induced cell death by targeting the T. brucei cytoskeletal protein TbBILBO1

Trypanosoma brucei belongs to a genus of protists that cause life-threatening and economically important diseases of human and animal populations in Sub-Saharan Africa. T. brucei cells are covered in surface glycoproteins some of which are used to escape the host immune system. Exo-/endocytotic trafficking of these and other molecules occurs via a single copy organelle called the flagellar pocket (FP). The FP is maintained and enclosed around the flagellum by the flagellar pocket collar (FPC). To date, the most important cytoskeletal component of the FPC is an essential, calcium-binding, polymer-forming protein called TbBILBO1. In searching for novel immune-tools to study this protein, we raised nanobodies against TbBILBO1. Nanobodies (Nb) that were selected according to their binding properties to TbBILBO1, were tested as immunofluorescence tools, and expressed as intrabodies (INb). One of them, Nb48, proved to be the most robust nanobody and intrabody. We further demonstrate that inducible, cytoplasmic expression of INb48 was lethal to these parasites, producing abnormal phenotypes resembling those of TbBILBO1 RNAi knockdown. Our results validate the feasibility of generating functional single-domain antibody derived intrabodies to target trypanosome cytoskeleton proteins.


42
Trypanosomes are flagellated protists, comprising of pathogenic species capable of infecting a wide 43 range of vertebrate hosts including humans, domestic and wild animals. Commonly transmitted by 44 insect vectors, trypanosomes cause lethal and economically important human and animal diseases expressing cancer specific antigens (47). Studies have shown that a single chain antibody fragment 93 (scFv) can have knockdown effects without having a direct neutralizing effect by trapping the target 94 membrane or secretory protein within the endoplasmic reticulum (48). By using targeting motifs, 95 intrabodies can also be directed to specific organelles or structures.

208
The expression of anti-TbBILBO1 intra-nanobodies is cytotoxic in T. brucei 209 Because anti-TbBILBO1 intrabodies co-localize with TbBILBO1 on isolated flagella, we hypothe-  However, concentrations from 5ng/mL to 1μg/mL induced a rapid growth defect from 24 hours post tetracycline indicates that the lethality of INb48::3cMyc expression in PCF is dose dependent. Taking in 217 account these results, we chose a tetracycline concentration of 1μg/mL as the standard concentration     Because FP biogenesis is hindered in INb48:: 3cMyc induced cells, we decided to investigate the organi-261 zation of the hook complex (HC), a structure intimately associated with the FPC, by immunolabelling of TbMORN1 (28) in the background of INb48::3cMyc induced mutant cells. The HC is an essential struc-263 ture positioned distal to the FPC, forming a distinctive hook shape, and is involved in the regulation 264 of entry of molecules into the flagellar pocket (20, 27, 28, 30). Figure 5C shows the normal location 265 and hook-like shape of anti-TbMORN1 labelling in non-induced cells. However, at 24hpi of 266 INb48::3cMyc expression, TbMORN1 labelling is elongated distally along the flagellum, and the hook 267 shape is lost ( Figure 5D). Importantly, little or weak TbMORN1 labelling is observed at the base of 268 the new flagellum, suggesting that HC biogenesis is also strongly affected in cells expressing an antibody marker anti-TbBLD10 was used, which was raised against TbBLD10, an essential protein 279 in pro-basal body biogenesis (75). Figure 5G shows a non-induced T. brucei cell with the typical 280 labelling of the mature and pro-basal bodies. At 24hpi of INb48:: 3cMyc, the labelling of the both the pro-281 and mature basal bodies of the mother and daughter flagella were unaffected ( Figure 5H).  Figure 6E shows TbMORN1 labelling and shows in more detail that, in the 303 presence of INb48::3cMyc (arrowheads), it is observed on the MtQ (arrows) extending in a linear fashion 304 proximally along the flagellum but not distally. These data demonstrate that the FPC and HC struc-     of helical polymers (23). We therefore used Mut2 EF-hand to test for co-labelling with Nb48. Excel-360 lent co-labelling was observed when the Mut2 EF-hand was expressed: the helical polymers of the mutated BILBO1 co-label with Nb48::HA::6His, ( Figure 8C), suggesting that mutating the calcium-binding 362 site, and preventing calcium binding, does not affect Nb48 binding. We next expressed the 363 TbBILBO1 truncations T1, T2, T3, T4 (refer to Figure 2F) in U-2 OS cells and used the previously   Figure 9C and D). Importantly, however, the typical polymer structures that TbBILBO1 normally forms in U-2 OS cells (refer to Figure 8B) were absent. Instead, dense compacted structures, attached 386 to thin minor polymers were formed. Similar dense structures were observed when Mut2-EFh was 387 co-expressed with INb48::3HA, indicating that the EF hand domain 2 mutation (which prevents calcium 388 binding) did not influence binding of INb48::3HA to TbBILBO1 ( Figure 9E). This data indicate that 389 binding of INb48:: 3HA to TbBILBO1 modifies and reduces linear polymerization in this environment.

410
Nanobodies are rapidly being developed, not only as revolutionary therapeutics but also as biological 411 tools. In mammalian cells they have been used to assess protein function intracellularly (intrabodies), 412 explore protein-protein interactions and as protein inhibitors (42)   Our research data support the hypothesis that targeting minor, yet essential, cytoskeletal proteins is 477 of considerable merit in the search to understand parasite biology. They also suggest that the use of anti-cytoskeletal intrabody in T. brucei, that is able to precisely target and bind to its target protein 482 epitope and induce disruption of a cytoskeletal structure leading to rapid cell death.

718
Grids were blocked and incubated in secondary antibody for 1 hour at RT (anti-mouse goat GMTA 719 5nm gold, and/or anti-rabbit goat 15nm gold GAR15, BBInternational) 1:10 in 0.2% fish skin gelatin 720 in PBS. Grids were blocked, and fixed in 2.5% glutaraldehyde in 0.2% fish skin gelatin in PBS.

722
Micrographs were taken on a Phillips Technai 12 transmission electron microscope at 100 kV.

1056
For the sake of convenience the protein being probed is named as anti-Nb, plus the relevant nanobody 1057 name rather than its tag. Scale bar = 5μm, inset = 1μm.     The third panel is a negative control of a WT cytoskeleton probed with anti-HA rabbit followed by 1204 anti-mouse FITC, which demonstrates that there is no cross-reaction between these antibodies on the

Anti-INb48
Anti-TbBLD10   TbBILBO1 (in this case anti-TbBILBO1 mouse monoclonal, IgM) and Nb48::HA::6His (anti-HA rabbit, IgG). The IgM was used here because it binds to the C-terminus whereas anti-TbBIL1BO (1-110) was raised to, and only binds to, the N-terminus of TbBILBO1. Scale bar = 5μm. For the sake of convenience the protein being probed is named as anti-Nb, plus the relevant nanobody name rather than its tag.  As with the experiments of Figure 9, TbBILBO1 proteins and INb48::3HA were expressed in U-2 OS cells for 24h; in this case INb48::3HA, was co-expressed T1-T4 truncations. Cells were then processed for IFA using either the anti-TbBILBO1 (1-110) rabbit polyclonal, raised to and binding only, the N-terminus of TbBILBO1, for detecting T1 and T2, or an in-house made anti-TbBILBO1 IgM mouse monoclonal, which recognizes the C' terminus of TbBILBO1; anti-HA was used to detect INb48::3HA followed by their respective secondary antibodies.  Figure 8G, are observed to a lesser extent but numerous smaller polymers are present in the cytoplasm and these are never present in cells expressing T4 alone. This suggests that the binding of INb48 to T4 modifies the polymer forming capacity of this truncation. Scale bar = 5μm. For the sake of convenience the protein being probed is named as anti-Nb, plus the relevant nanobody name rather than its tag.