WO2011020497A1

WO2011020497A1 - C-terminal mutant of apolipoprotein l-i and its therapeutical or prophylactic use

Info

Publication number: WO2011020497A1
Application number: PCT/EP2009/060687
Authority: WO
Inventors: Etienne Pays; Laurence Lecordier; Philippe Poelvoorde
Original assignee: Universite Libre De Bruxelles
Priority date: 2009-08-18
Filing date: 2009-08-18
Publication date: 2011-02-24
Also published as: AP2012006112A0

Abstract

The present invention is in the field of Molecular Biology and is related to new c-terminal mutants of Apolipoprotein L-I and their pharmaceutical use especially for a treatment and/or a prevention of diseases induced in mammals especially in cattle by Trypanosoma, especially African Trypanosoma, more particularly Trypanosoma brucei brucei, Trypanosoma brucei rhodesiense, Trypanosoma Evansi, Trypanosoma congolense and/or Trypanosoma vivax.

Description

C-TERMINAL MUTANT OF APOLIPOPROTEIN L-I AND ITS

THERAPEUTICAL OR PROPHYLACTIC USE Field of the invention

[0001] The present invention is in the field of

Molecular Biology and is related to new Apolipoprotein L-I sequence (c-terminal mutant of Apolipoprotein L-I (apoLl)) and its pharmaceutical (therapeutical or prophylactic) use, especially for a treatment and/or a prevention of diseases induced in mammals, especially in cattle, preferably infections induced by Trypanosoma, especially African Trypanosoma, more particularly Trypanosoma brucei brucei, Trypanosoma brucei rhodesiense, Trypanosoma congolense , trypanosome evansi and/or trypanosoma vivax and their related diseases ( AGA A) .

[0002] The present invention is also related to a transgenic cattle and a method for obtaining this transgenic cattle which is able to produce (express) the sequence of the invention, the obtained transgenic cattle being tolerant or resistant to Trypanosoma infection ( s ) , especially to African Trypanosoma infection ( s ) , more preferably infection (s) induced by Trypanosoma brucei brucei, Trypanosoma brucei rhodesiense, trypanosoma congolense, trypanosoma evansi and/or trypanosoma vivax and related diseases (NAGANA) . Background of the invention and state of the art

[0003] Apolipoprotein L-I (apoLl) is a human- specific serum protein that kills Trypanosoma brucei through ionic pore formation in endosomal membranes of the parasite. The T. brucei subspecies rhodesiense and gambiense resist this lytic activity and can infect humans, causing sleeping sickness. In the case of T. b. rhodesiense resistance to lysis involves interaction of the Serum Resistance-Associated (SRA) protein with the C-terminal helix of apoLl .

[0004] Normal human serum (NHS) is able to kill T. b. brucei, but not T. b. rhodesiense and T. b. gambiense . The lytic factor was identified as being apoLl . This protein is associated with HDL particles that are efficiently taken up by the parasite through specific binding to a haptoglobin-hemoglobin surface receptor, due to the simultaneous presence of haptoglobin-related protein (Hpr) acting as a ligand in these particles. Trypanosome lysis results from anionic pore formation by apoLl in the lysosomal membrane of the parasite. Resistance to lysis has only been studied in case of T. b. rhodesiense, where it was shown to depend on a parasite protein termed SRA. As synthesis of SRA only occurs after transcriptional activation of a given Variant Specific Glycoprotein (VSG) gene expression site from a repertoire of 10-20 candidates, T. b. rhodesiense clones can be either sensitive or resistant to NHS depending on which expression site is active. The mechanism by which SRA inhibits the activity of apoLl is unclear. Direct coil-coiling interaction between the C-terminal α-helix of apoLl and the N-terminal a-helix of SRA was demonstrated in vitro, but in vivo only evidence for tight co-localization between the two proteins was obtained. Total deletion of the C-terminal helix appeared to confer toxic activity to recombinant apoLl on T. b. rhodesiense, suggesting that in vivo SRA neutralizes apoLl through interaction with its C-terminal domain. However, the trypanolytic effect of the deleted apoLl was weak and incomplete. Moreover, data obtained following transgenic expression of a similarly truncated apoLl in mice suggested that its trypanolytic potential was lost in vivo. Therefore, additional work was required to evaluate if the interaction observed in vitro was relevant for the in vivo mechanism of T. b. rhodesiense resistance to human serum. While human serum is unable to kill T. b. rhodesiense, the serum of some African primates like Papio sp. was equally active on both T. b. rhodesiense and T. b. brucei. The phenotype of trypanolysis by Papio serum strikingly resembled that induced by human serum, as it was also dependent on HDL particles and was similarly inhibited by competing amounts of haptoglobin. Aims of the invention

[0005] The present invention aims to provide a new

Apolipoprotein L-I (apoL-1) (amino acid) sequence and its encoding nucleotide sequence, a vector comprising this nucleotide or amino-acid sequence, a cell transformed by this nucleotide or amino acid sequence or this vector that do not present the drawbacks of the state of the art and that could be used for improving the treatment and/or the prevention of infections and related diseases induced in mammals, especially in cattle, preferably in cattle infected by Trypanosoma, especially African Trypanosoma, more particularly Trypanosoma brucei brucei, Trypanosoma brucei rhodesiense, trypanosoma congolense, trypanosoma evansi and trypanosoma vivax and their related diseases ( AGA A) .

[0006] Another aim of the present invention is to propose a new pharmaceutical composition comprising one or more of these elements (Apolipoprotein amino acid or nucleotide sequence (s), vector or cell) that could be administrated to mammals, especially to cattle to render these mammals (cattle) tolerant or resistant to these Trypanosoma infections and related diseases (NAGANA) .

[0007] A last aim of the present invention is to obtain a transgenic ( i . e . genetically modified or recombinant) cattle which comprises a genetic modification that allows and renders this transgenic mammal, especially this transgenic cattle, tolerant or resistant to one or more of these Trypanosoma infections and related diseases (NAGANA) .

Summary of the invention [0008] The present invention is related to a

(recombinant and modified) human Apolipoprotein L-I or a monkey (recombinant and modified or wild-type) , preferably a Papio species Apolipoprotein L-I, more specifically a Papio anubis Apolipoprotein L-I, this primate, preferably human Apolipoprotein L-I being further characterised in that its C-terminal portion of its wild-type sequence (preferably SEQ.ID.NO:l, SEQ.ID.NO:8 or SEQ. ID. NO: 25)

either (naturally) comprises an amino acids sequence selected from the group consisting of TKIQ, NKIQ, NNIQ, NNNQ, TNNY, TKNY or TKIY;

or comprises a replacement of its last 13 C-terminal amino acids (NNNYKILQADQEL) by an addition of the sequence TKIQKILQADEL or by an addition of 4 amino acids having a sequence selected from the group consisting of TKIQ, NKIQ, NNIQ, NNNQ, TNNY, TKNY, TKIY or NNNY, preferably the addition of 4 amino acids having the sequence TKIQ.

[0009] Therefore, this (recombinant and modified sequence) human Apolipoprotein L-I sequence may also correspond to this wild type human Apolipoprotein sequence (SEQ. ID. O: 1, SEQ.ID.NO:8 or SEQ . ID. O : 25 ) modified by (which comprises) a deletion of 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19 or 20 (preferably 9) of its last C-terminal amino acids.

[0010] More preferably, the Apolipoprotein L-I sequence according to the invention is a monkey, preferably a Papio (more preferably Papio Anubis) species Apolipoprotein L-I sequence which may comprises one more fragments of the putative sequence (s) described in the figure 5.

[0011] Preferably, the human Apoliprotein according to the invention presents a sequence which is selected from the group consisting of SEQ.ID.NO:2, SEQ.ID.NO:3, SEQ. ID. NO.4, SEQ. ID. NO.5, SEQ.ID.NO:6, SEQ.ID.NO:7, SEQ. ID. NO: 22, SEQ.ID.NO:23 or SEQ . ID . NO : 24.

[0012] Another aspect of the present invention is related to a polynucleotide sequence encoding the Apolipoprotein L-I according to the invention and a vector comprising the (amino acid sequence of) Apolipoprotein L-I of the invention or its corresponding (coding) polynucleotide sequence.

[0013] The vector could be any chemical (lipidic) and/or nucleic acid composition which comprises element (s) for obtaining a transformation (or transduction) of a cell, preferably a vector selected from the group consisting of plasmids, viruses, lipids or cationic lipid vesicles.

[0014] Another aspect of the present invention is related to a cell transformed by this amino acid or polynucleotide sequence according to the invention and/or expressing the (recombinant and modified) Apolipoprotein L- I according to the invention; this cell is preferably a mammal cell, more preferably a cattle cell, more preferably a cow cell.

[0015] Another aspect of the present invention is related to a pharmaceutical composition (including a vaccine) comprising an adequate pharmaceutical carrier (or diluent and possibly one or more adequate adjuvant (s)) and a sufficient amount of an element selected from the group consisting of the Apolipoprotein L-I (amino acid sequence) according to the invention, the polynucleotide according to the invention, the vector according to the invention or the cell according to the invention; preferably, this pharmaceutical composition (vaccine) is used in (for) the treatment and/or the prevention of diseases induced in humans (by Trypanosoma rhodesiense infection) and/or in cattle, preferably in a cow, by Trypanosoma (NAGANA) , preferably the above identified Trypanosoma species; and wherein the Apolipoprotein L-I of the invention is preferably capable impeding its interaction with the Serum Associated protein (SRA) .

[0016] Cattle include any kind of (domesticated or not) bovidae (preferably a cow) , susceptible to be infected by Trypanosoma , especially African Trypanosoma, preferably by Trypanosoma brucei brucei, Trypanosoma brucei rhodesiense, trypanosoma congolense, trypanosoma evansi and/or trypanosoma vivax and suffering from the induced disease (NAGANA) .

[0017] Another aspect of the present invention is related to a non human genetically modified mammal, preferably a cattle, more preferably a cow, which is expressing the Apolipoprotein L-I according to the invention or which may comprise the polynucleotide, the vector or the cell according to the invention or which may express the synthesis of the amino acid sequence of the Apolipoprotein L-I of the invention.

[0018] Preferably, this non human genetically modified mammal is a genetically modified cattle, preferably genetically modified cow, which could be resistant or tolerant to infection (s) induced by Trypanosoma and non or slowly affected by the related diseases (NAGANA) , preferably infection (s) and disease (s) induced by Trypanosoma brucei brucei, Trypanosoma brucei rhodesiense, trypanosoma congolense, trypanosoma evansi and/or trypanosoma vivax.

[0019] The pharmaceutical composition according to the invention could be a therapeutic or a prophylactic vaccine, possibly comprising one or more adequate carrier (s), diluent (s) and/or adjuvant (s) for inducing the immune system against this trypanosome, this vaccine being dedicated ( admini strable ) to humans and/or to cattle and which comprises an adequate and pharmaceutical carrier (or diluent) well known by the person skilled in the art and possibly one or more adjuvant (s) to induce and stimulate an effective cellular and/or humoral immune response against infection (s) induced by these Trypanosoma.

The present invention is also related to a therapeutic and/or prophylactic (vaccinal) method for a treatment and/or prevention of a disease affecting mammals, especially humans (infected or infectable by Trypanosoma rhodesiense) and/or cattle (including cows) , more preferably cattle (including cows) contaminated (or susceptible to be contaminated) by trypanosoma species (especially the ones above described) and their related diseases (NAGANA) , this method comprises the step of administrating a sufficient amount of the pharmaceutical composition (vaccine) of the invention to this mammal to reduce and/or suppress the symptoms of this disease in this mammal .

[0020] The present invention will be described in more details in the following detailed description of the invention in reference to the enclosed figures presented as non limited illustrations of the present invention.

Short description of the figures [0021] Fig. 1 represents lytic activities of apoLl .

[0022] Fig. 2 represents SRA interaction with apoLl .

[0023] Fig. 3 represents effects of mutations in the conserved c-terminal leucine zipper of apoLl .

[0024] Fig. 4 represents the trypanolytic activity of apoLl-like sequences in Papio.

[0025] Fig. 5 represents ApoLl-like sequences in

Papio .

[0026] Fig.6 represents effects of mutations in the

C-terminus of apoLl-like sequences in Papio.

[0027] Fig. 7 represents the effect of transient transgenic expression of WT or mutant apoLl on trypanosome infection in mice.

Detailed description of the invention

[0028] The serum protein apolipoprotein L-I (apoLl) is responsible for human innate immunity against Trypanosoma brucei brucei, because this protein kills the parasite by generating ionic pores in the lysosomal membrane. Two T. brucei subspecies (T. b. rhodesiense and T. b. gambiense) can resist apoLl and therefore, infect humans and cause sleeping sickness. In T. b. rhodesiense resistance to human serum is linked to interaction of the Serum Resistance-Associated (SRA) protein with the C- terminal region of apoLl . Mutations targeted to this region reduced its interaction with SRA, while preserving the activity of the ionic pore-forming domain. While some mutants also lost their trypanolytic potential, C-terminal mutants inspired by apoLl-like sequences of Papio anubis conserved this activity, but also acquired the ability to efficiently kill T. b. rhodesiense, both in vitro and in mice. These findings demonstrate that interaction of SRA with the C-terminus of apoLl inactivate this protein and is responsible for the resistance of T. b. rhodesiense to human serum. Moreover, they suggest that apoLl-like proteins could be responsible for the trypanolytic potential of Papio species. Finally, Papio-like human apoLl mutants could be used to generate transgenic cattle that would resist both T. b. brucei and T. b. rhodesiense .

[0029] By a mutational and deletional analysis of the C-terminal helix of apoLl, the linkage between interaction with SRA and lytic potential for different T. brucei subspecies was investigated. Although in E. coli the SRA-interacting domain was dispensable for ionic pore- forming activity, its interaction with SRA resulted in inhibition of this activity. Different mutations affecting the C-terminal helix reduced the interaction of apoLl with SRA. However, mutants in the L370-L392 leucine zipper also lost in vitro trypanolytic activity. Mutants in the conserved G361-S364 motif still interacted with SRA, but lost trypanolytic potential in some cases. Truncating and/or mutating the C-terminal sequence of human apoLl like that of apoLl-like sequences of Papio anubis resulted in both loss of interaction with SRA and acquired ability to efficiently kill human serum-resistant T. b. rhodesiense parasites, in vitro as well as in transgenic mice. These findings demonstrate that SRA interaction with the C- terminal helix of apoLl inhibits its pore-forming activity and determines resistance of T. b. rhodesiense to human serum. In addition, they provide a possible explanation for the ability of Papio serum to kill T. b. rhodesiense, and lead to generation of transgenic cattle resistant to both T. b. brucei and T. b. rhodesiense .

[0030] The inventors analyzed the effects of various deletions and mutations in the C-terminal domain of apoLl on the trypanolytic potential of this protein against T. b. brucei and T. b. rhodesiense . The obtained results confirmed the interaction model presented in [1], as well as the role of this interaction in resistance to apoLl in T. b. rhodesiense. In accordance with these findings, Papio-like apoLl mutants able to efficiently kill both T. b. brucei and T. b. rhodesiense were identified.

Materials and Methods

Production of recombinant WT and mutant versions of apoLl . [0031] The apoLl coding sequence was amplified by

PCR from the pcDNA3. l-ApoLl-V5-His₆ construct [1] with primers creating a 5' Ncol site and ATG codon upstream from the E28 codon and a 3' Notl site downstream from the His₆ tag coding sequence. The PCR fragment was cloned in Ncol and Notl sites of the first polylinker of pCDF-Duetl expression vector (Novagen) . Mutant versions of apoLl were created by site-directed mutagenesis with Accuprime Pfx DNA polymerase (Invitrogen) and verified by sequencing. The complete SRA coding sequence was obtained by PCR from pTSA- Rib-SRA [3] with primers creating a 5' Ecll36II and 3' Sail sites. The DNA fragment was cloned in frame with the S-tag coding sequence in EcoRV and Xhol sites of the second polylinker of pCDF-Duetl vector and pCDF-Duet ApoLl-V5-His₆ construct . [0032] The plasmid constructs were transfected into

E. coli BL21(DE3) and expression of recombinant proteins was induced from an exponentially growing culture by 1 mM isopropyl β-D-l-thiogalactopyranoside (IPTG) overnight at 37°C. Bacteria were resuspended in 50mM Tris-HCl (pH 9.2) and lysed by two passages in French Press. Cell debris were removed by centrifugation at 5000g for 10 min, and inclusion bodies were recovered from supernatant by centrifugation at 16,000g for 15 min. Purification was performed from inclusion bodies dissolved in 6 M guanidium chloride, 150 mM NaCl, 50 mM Tris-HCl (pH 8.0). Solubilized proteins were incubated for 1 h with Ni-NTA beads (Qiagen) , and the bound proteins were extensively washed with 4 M guanidium chloride, 50 mM Tris-HCl (pH 7.5), 150 mM NaCl, 20 mM imidazole, before elution with 4 M guanidium chloride in 0.2 M acetic acid. After dialysis against 20 mM acetic acid, the proteins were concentrated (Vivaspin, Sartorius) . Purity and concentration were estimated by SDS-PAGE and Coomassie Blue staining.

Pore forming activity in bacteria.

[0033] Cultures of BL21 (DE3) strains transfected with pCDF-Duetl constructs were grown at 37°C in LB containing 1% glucose and 50 μg/ml streptomycin from freshly plated colonies until optical density at 600nm (OD600) reached 0.7 to 0.8. Serial dilutions were plated onto LB containing 50 μg/ml streptomycin and either 1% glucose (non-induced control) or 50 to 75 μΜ IPTG (induction of recombinant proteins) . Colonies were counted after overnight incubation at 37°C and results were expressed as the ratio of colonies with and without IPTG induction. These results reflect the pore-forming activity of apoLl .

Copurification of apoLl and SRA from bacteria.

[0034] Cultures of BL21(DE3) strains transfected with pCDF-Duetl constructs were grown at 37°C in LB containing 50 μg/ml streptomycin and 1% glucose from freshly plated colonies untill OD600 reached 0.7 to 0.8. Cultures were centrifuged and bacteria pellet was resuspended in fresh medium without glucose and distributed in 3 flasks to perform induction and copurification in triplicate. Expression of recombinant proteins was induced by addition of ImM IPTG overnight at 20°C and 80 rpm. Cell density was measured by OD600 and bacteria were centrifuged and resuspended in (10% of culture volume x OD600) in cold hypotonic buffer (50 mM MES pH 6.0 or pH 7.0) containing EDTA-free protease inhibitors (Roche) . Bacteria were lysed by Fast Prep 24 (MP Biomedicals) with 1/10 volume of glass beads (Lysing Matrix B, MP Biomedicals), for 3 x 30 s at 6 m.s^-1. Cell lysate was complemented by 0.6M NaCl, 1% Triton X100, 20 mM imidazole, and vortexed vigorously. Cell debris were pelleted by centrifugation for 15 min at 16,000 g, and the supernatant was applied onto Ni-NTA beads (Qiagen) equilibrated in the same buffer, for 2 h at 4°C. Beads were washed with 20 volumes of beads with cold binding buffer and bound proteins were eluted with 2 volumes of SDS-PAGE sample buffer. Supernatant and bound fractions were analyzed by western blotting. ApoLl was revealed with anti- recombinant apoLl rat serum and peroxydase conjugated anti- rat antibody. SRA-Stag was detected by anti-Stag monoclonal antibody (Novagen) and peroxydase conjugated anti-mouse antibody. Peroxydase activity was revealed by ECL (Western Lighting Chemiluminescence Reagent PLUS, Perkin Elmer) . Signals were quantified in supernatant and bound fractions using ImageQuant TL Software (GE Healthcare) . Since only a fraction of each supernatant and elution was loaded onto the gel, the values obtained were reported to the total volume of lysis or elution ("supernatant" and "bound" lanes respectively) . ApoLl and SRA binding values to nickel were calculated separately and the binding of SRA to apoLl was expressed as "SRA binding versus apoLl binding" ratio. The ratio obtained for SRA binding to WT apoLl was considered as 100%.

Parasites . [0035] Trypanosoma brucei brucei 328-114 [10] were grown in HMI9 supplemented with 10% foetal bovine serum, 10% Serum Plus [8] at 37°C in 5% C0₂ . T. b. rhodesiense ETat 1.2 NHS-resistant (R) and NHS-sensitive (S) clones and T. b. gambiense LiTat 1.2 were grown in IMDM medium supplemented with 20% foetal bovine serum [9] . For lysis experiments, parasites were diluted at a final concentration of 5.10⁵ cells/ml, and recombinant WT or mutant apoLl was added at 10 to 20 μg/ml. Living parasites were counted in duplicate under the microscope. When lysis was not complete after 8 h, parasites were counted in duplicate after 24 h incubation and cell survival was expressed as % of control. Lysis experiments with human and Papio serum, as well as imaging of living immobilised trypanosomes , were performed as described in [11] . Sequence analysis .

[0036] Alignments were performed by CLUSTALW software and edited with GeneDoc (http://wwvi.psc.edu/biomed/genedoc) . Hydrophobic Cluster Analysis (HCA) was used to compare the proteins at 2D level

[5,6] (http : //mobyle . rpbs . univ-paris-diderot . fr/cgi- bin /'portal , py? form=HCA) . Transient expression in mice.

[0037] The hydrodynamic transfection of mice was performed according to the method described in [12] . Briefly, 2 ml of saline solution containing 10 yg of pcDNA3-l or pcDNA3-l WT apoLI-V5His / del3-V5His / delTKIQ- V5His were injected in less than 8 seconds in the vein of the tail of 8 weeks-old BALB/c female mice. At day 1, an aliquot of blood was analyzed by Western blot for expression of apoLl and mice were injected intraperitoneally with 10^s trypanosomes .

Results

Direct interaction of SRA with apoLl inactivates the pore- forming activity of this protein.

[0038] As illustrated in Fig. 1A, apoLl contains three functional domains responsible for its ionic pore- forming capacity, addressing to biological membranes and interaction with SRA, from N- to C-terminus respectively [2]

[0039] Fig. 1A shows a schematic representation of apoLl and apoLl variants analyzed in this work (see ref. 4 for details about the three apoLl domains) . [0040] Fig. IB represents the colicin-like activity of WT and mutant apoLl (ApoLl-encoding pCDF-DUET plasmids were transfected into E. coli BL21 (DE3) , and the bacterial plating efficiency was scored comparing expression induction by IPTG addition or non-induction by glucose

(Glc) addition, after overnight incubation at 37°C, the ΔΗ9-Η10 apoLl mutant lacks the region encoding helices 9 and 10 of the pore-forming domain) .

[0041] Fig. 1C represents the trypanolytic potential of various apoLl variants, as determined on T. b. brucei after 24 h-incubation in vitro (ctrl=control without apoLl ) .

[0042] Fig. 2A represents the double expression of apoLl and SRA in E. coli (the scheme of the plasmid construct is shown above Western blot data illustrating the detection of the two tagged proteins from the lysis supernatant (sup=supernantant) and their recovery in the fractions bound to nickel beads. ApoLl and SRA were revealed by anti-apoLl and anti-S tag antibodies, respectively. The pH of the extraction buffer was between 6 and 7, as indicated) .

[0043] Fig. 2B represents the influence of apoLl and/or SRA expression on bacterial platting efficiency determined by the ratio of colonies counted following induction versus non -induced controls as illustrated in fig. IB.

[0044] Fig. 2C represents the evaluation of the level of protein association with the nickel beads.

SRA binding to apoLl was expressed as SRA binding percentage divided by the percentage of apoLl binding to the nickel beads. This ratio was considered as 100% for SRA binding to WT apoLl . In E. coli co-expressing WT apoLl and SRA, the typical yield of each protein was respectively 3 yg and lOOng/10 cells. The values resulted from 3 independent experiments, each performed in triplicate.

[0045] In E.coli, the pore-forming activity of apoLl can be measured by the reduction of cell viability following induction of apoLl expression upon IPTG addition as detailed in [2], in this system apoLl closely mimicked the toxicity of the genuine pore-forming domain of bacterial colicin A, including depolarization of the plasma membrane. As a control, deletion of the essential helix 9 of the pore-forming domain completely inhibited this activity (Fig. IB) . In addition to its effect on E. coll, the pore-forming activity of apoLl is also responsible for the ability of this protein to kill trypanosomes , although the trafficking and intracellular targeting of the toxin are obviously very different between the two systems [2] . Addition of recombinant apoLl to T. b. brucel resulted in efficient killing of the parasite, as measured after overnight incubation (Fig. 1C) . The trypanolytic activity of apoLl is known to be inhibited by the T. b. rhodeslense protein SRA [3] . In order to analyze the mechanism by which SRA neutralizes apoLl pore-forming activity, the inventors constructed a bicistronic plasmid co-expressing the two proteins in E. coll, under the dependence of two inducible T7 transcription promoters (pCDF-DUET; Fig. 2A) . In this system apoLl was tagged at the C-terminus with V5 and 6-His peptides, whereas SRA was provided with a C-term S tag. Following induction of expression, apoLl was purified by binding to nickel beads, and after elution it was revealed using anti-apoLl antibodies. As shown in Fig. 2A, apoLl was clearly detected in the nickel-bound fraction. When co- expressed with apoLl, SRA was also present in this fraction as revealed by anti-S tag antibodies. However, in the absence of apoLl, SRA was no longer found in the bound fraction (Fig. 2A) . These results indicated that in E. coli, co-expression of apoLl and SRA results in the formation of a complex associating these two proteins consistent with the lysosomal localization of this complex in T. brucei (1), the formation of the apoLl-SRA complex appeared to be favoured at acidic pH, as its amount was strongly reduced upon E. coli lysis at different pH values between pH 6 and pH 7 (Fig. 2A) . Significantly, the pore- forming activity of apoLl was strongly inhibited upon co- expression of SRA (Fig. 2B) . Therefore, in the E. coli expression system apoLl appeared to be inactivated following direct interaction with SRA.

The C-terminal domain of apoLl is entirely responsible for the interaction of this protein with SRA.

[0046] The inventors evaluated the level of interaction of SRA with different mutants of apoLl by measuring the relative amounts of either protein bound to nickel beads. More precise measurement of this interaction using plasmon resonance was impossible, due to the propensity of both proteins to stick to various matrices. Thereafter, the inventors generated apoLl variants deleted of either one of the three functional domains (del 1, del 2 and del 3 from N- to C-term, see Fig. 1A) . The presence in each case of an N-terminal bacterial signal peptide (pelB) allowed the determination of the pore-forming potential of the variants in E. coli irrespective of the deletion of the membrane-addressing domain [2] . Only deletion of the N- terminal domain (del 1) resulted in the loss of the pore- forming activity (Fig. 2B) . Co-expression of SRA with del 2 resulted in strong inhibition of the killing activity like it was observed with wild-type apoLl, whereas SRA only mildly affected the activity of del 3 (Fig. 2B) . In accordance with these data, SRA was found to bind to del 1 and del 2, but not to del 3 (Fig. 2C) . Altogether these data confirmed that SRA interacts with the apoLl C-terminal domain, and revealed that this interaction inhibits apoLl activity independently of the presence of the membrane- addressing domain.

Deletion or mutations of the C-terminal domain do not affect ionic pore-formation , but impede trypanolytic activi ty.

[0047] Recombinant forms of the three apoLl variants del 1, del 2 and del 3 were produced in E. coli , tested for their trypanolytic potential. As expected del 1 and del 2 were inactive (Fig. 1C) . Surprisingly, despite the full conservation of its intrinsic pore-forming activity (Fig. 2B) , del 3 was also inactive (Fig. 1C) . These findings confirmed a recent report [4], but contrasted with previous work where recombinant del 3 expressed in Chinese hamster ovary cells was found to kill NHS-sensitive trypanosomes [1].

[0048] The C-terminal domain of apoLl, as well as that of the other apoL family members, is characterized by the presence of a leucine zipper (Fig. 3A)

[0049] In Fig. 3A, the upper panel shows the sequence alignment of the leucine zipper within the human apoL family. The lower panels show hydrophobic cluster analysis of this region, in WT and various mutant apoLls.

[0050] Fig. 3B is the SRA binding to WT or mutants apoLls .

[0051] Fig. 3C is a quantification of SRA binding to

WT or mutant apoLls.

[0052] Fig. 3D is a bacterial plating efficiency of various mutant apoLls. [0053] Fig. 3E represents the trypanolytic potential of various apoLl variants (ctrl-control) .

[0054] Different mutants affecting this zipper were generated (Hel 1: L378 /382 /385S ; Hel 2: L378/382/385A; Hel 3: L378P) . As shown in Fig. 2A, in each case hydrophobic cluster analysis predicted a strong reduction of hydrophobic interaction potential. This was confirmed by the prediction of interaction energy using measurement of the mean force potential [7] according to the apoLl-SRA interaction model presented in [1, 7] . This energy was decreased in the mutant sequences (Table 1) .

Table 1

Predicted energy of interaction between SRA and the C- terminal domain of various apoLl mutants. The interaction model presented in [1] was used for energy calculation based on the mean force potential [7] .

Interaction ene

(kcal/mol)

WT -490

Hell -452

TKIQ -433

delTKIQ -409

delNNNY -456

As expected from this prediction, SRA interaction was reduced in the three mutants (Fig. 3B,C) . However, like in the case of the del 3 variant, these mutants also largely lost their trypanolytic potential although they conserved full pore-forming activity in E. coli (Fig. 3D,E) . Therefore, conservation of the C-terminal helix appeared to be necessary for the trypanolytic activity of recombinant apoLl . Identification of Papio apoLl-like proteins possibly involved in trypanolytic activity.

[0055] Fig. 4A represents the trypanolytic activity of Papio serum on NHS-resistant (R) or sensitive (S) clones of T. b. rhodesiense, and effect of DIDS on this activity.

[0056] Fig. 4B represents the trypanolytic potential of Papio and human serum, and effect of haptoglobin z (Hp) on this potential.

[0057] Fig. 4C is the phenotype of T. b. rhodesiense

NHS-resistant (R) or sensitive (S) clones incubated with human or Papio serum.

[0058] Fig. 4D is a western blot analysis with anti- apoLl, of human or Papio serum and of serum fractions bound to either anti-apoAl or SRA.

[0059] As shown in Fig. 4A, the serum of Papio cynocephalus was equally able to lyze NHS-resistant and - sensitive T. b. rhodesiense clones, although it did not affect T. b. gambiense . As NHS or recombinant apoLl cannot lyze T. b. rhodesiense [1], the Papio serum must contain a trypanolytic factor different from apoLl . However, several observations suggest that this factor actually resembles apoLl . First, like apoLl the Papio trypanolytic factor is bound to HDL particles, as Papio trypanolytic activity was present in a serum fraction binding to anti-apoAl, the main constituent of HDLs. Second, like apoLl [2] , it was sensitive to inhibition by the anionic channel inhibitor DIDS (Fig. 4A) . Third, like apoLl it was inhibited by competition with an excess of haptoglobin, suggesting its association with Hpr (Fig. 4B) . Fourth, the cellular phenotype of trypanosome lysis by Papio serum, involving considerable swelling of the lysosome, was indistinguishable from that induced by NHS (Fig. 4C) . Finally, proteins isolated from Papio serum through binding to anti-apoAl could be detected by anti-apoLl antibodies (Fig. 4D) . However, as expected considering the lack of resistance of T. b. rhodesiense to Papio serum, the apoLl- like proteins of Papio were unable to bind to SRA (Fig. 4D) . In order to evaluate the possible presence of apoLl- like proteins in Papio sp . , the inventors examined the currently available Papio anubis genomic sequence information .

[0060] Fig. 5 represents apoLl-like sequences in Papio. The upper panel represents the alignment of papio apoLl-like sequences reconstituted from information present in current Papio genome databases. The arrowhead identifies a frameshift predicted in the two apoLl-like genes and the arrow indicates the position and orientation of the 10 primers used for the RT-PCR analysis. In the lower panel, the details of the frameshift are represented: nucleotides of the human gene deleted in the papio sequence are boxed. As shown in Fig. 5, two apoLl-like sequences were identified, one of which was shorter due to C-terminal truncation. From the actual state of information the two genes appeared to be interrupted by frameshift mutations (Fig. 5), and the inventors confirmed this frameshift in reverse transcripts of these genes. From RNA of either blood cells of P. cynocephalus or endometrium of P. Anubis, using various combinations between 5 forwards and 5 reverse primers from different regions (Fig5) .

C-terminal apoLl mutants inspired by Papio sequences kill both T. b. brucei and T. b. rhodesiense .

[0061] The inventors hypothesized that converting the C-terminal sequence of human apoLl into those found in the two apoLl-like sequences of Papio anubis could impede the interaction of this protein with SRA and confer the capacity to kill T. b. rhodesiense . This involved the replacement of the 386-389 sequence NNNY into TKIQ (TKIQ mutant) , as well as the same replacement together with the removal of the 9 C-terminal amino acids (delTKIQ mutant) (Fig. 6A)

[0062] In Fig. 6A, the upper panel shows the sequence of the various mutants. The lower panels show hydrophobic cluster analysis of this region, in WT and two mutants of apoLl .

[0063] Fig. 6B represents SRA binding to WT or mutants apoLls.

[0064] Fig. 6C is the quantification of SRA binding to WT or mutant apoLls.

[0065] Fig. 6D is the bacterial plating efficiency of various mutant apoLls.

[0066] Fig. 6E represents the trypanolytic activity of various apoLl variants, as determined on NHS-resistant (R) clones of T. b. rhodesiense, T. b. brucei and T. b. gambiense (ctrl=control)

[0067] In addition to the latter mutant, the inventors also generated similar mutants, where the four C- terminal amino acids of the truncated version were intermediate between Papio and human apoLls (NKIQ, NNIQ, NNNQ, NNNY) (Fig. 6A) . These changes did not strongly affect the hydrophobic cluster pattern of the C-terminal helix (Fig. 6A) , but they significantly reduced the predicted energy of interaction of apoLl with SRA according to the model presented in [1] (Table 1) . Accordingly, these apoLl variants lost their capacity to interact with SRA (Fig. 6B,C) . All mutants conserved their pore-forming activity in E. coli (Fig. 6D) . As shown in Fig. 6E, the different Papio-like apoLl variants efficiently killed both T. b. brucei and NHS-resistant clones of T. b. rhodesiense . However, they were unable to kill T. b. gambiense . [0068] The inventors evaluated if the Papio-like apoLl variants could exhibit trypanolytic activity in mammals (mice) as they did in vitro.

[0069] Fig. 7A is a detection of WT apoLl at 1,4 and 8 days after hydrodynamic injection of the plasmid construct, monitored by incubation of Western blots of mouse serum proteins with rat anti-apoLl antibodies. Mice 1 and 2 were injected with control (empty) plasmid. The lane labelled NHS shows the result obtained with normal human serum.

[0070] Fig. 7B represents the detection of the different apoLl variants at day 1 post-injection of DNA (CTRL=no injection) (the mice whom sera was analyzed here were used for the trypanosome infection experiments reported in panel C (S=T. b. rhodesiense ETat 1.2S; R=T. b. rhodesiense ETat 1.2R; q=T. b. gambiense LiTat 1.2) and normal human serum (NHS) was used for comparison) . Loading control by Ponceau red staining of albumin is shown below each panel.

[0071] Fig. 7C shows 24 h post-injection of DNA, 10^s parasites of the indicated strains were inoculated intraperitoneally into mice (Parasitemia was measured 3 days after parasite inoculation (control : empty plasmid).

[0072] As shown in Fig. 7A, expression of apoLl can optimally be detected in mice one day after hydrodynamic injection of 10 yg of pCDNA3 plasmid encoding the protein. Similarly, apoLl variants could be detected one day post- injection of DNA, although the apoLl mutants appeared to be less expressed than WT apoLl (Fig. 7B) . Intraperitoneal inoculation of 10^s trypanosomes from different T. brucei subspecies was performed at that day post-DNA injection. Infection by NHS-sensitive T. b. rhodesiense ETat 1.2S parasites was inhibited following expression of either WT or delTKIQ apoLl, as determined by the measurement of the parasite number at the peak of parasitaemia (Fig. 7C) . As expected, transgenic expression of WT apoLl did not confer protection against the NHS-resistant T. b. rhodesiense clone ETat 1.2R (Fig. 7C) . However, mice expressing the delTKIQ apoLl variant could resist both trypanosome lines (Fig. 7C) . In addition, these mammals (mice) were also able to kill T. congolense (Fig. 7C) . Therefore, transgenic cattle expressing this variant would resist infection by T. b. brucei, T. b. rhodesiense and T. congolense . In contrast, neither WT nor mutant apoLl conferred protection against T. b. gambiense (Fig. 7C) .

[0073] These data demonstrate that in E. coli, SRA inhibits the pore-forming activity of apoLl through direct interaction with the C-terminal helix of this protein, and show that apoLl variants unable to bind SRA through deletion or mutations of this helix could overcome this inhibition. It is particularly interesting to note that the apoLl variant lacking the original membrane-addressing domain (but containing a bacterial signal peptide to compensate this activity) was still fully inhibited following co-expression of SRA in E. coli. This result indicates that the membrane-addressing domain is dispensable for the control of the pore-forming domain by SRA, suggesting that this control does not operate through refolding of the protein. A possible explanation would be that interaction with SRA prevents membrane targeting of apoLl . Similarly to the situation in E. coli, apoLl variants unable to bind SRA were found to kill NHS- resistant trypanosomes , confirming that the conclusions drawn in E. coli regarding the neutralization of the protein by SRA were also valid for trypanosomes. In particular, the fact that in E. coli SRA can directly inhibit the activity of apoLl suggests that in trypanosomes, its effect on apoLl would not necessarily operate through reorientation of apoLl trafficking in the cell. However, a clear difference was observed between E. coli and trypanosomes . In the latter case, conservation of most of the C-terminal helix was required to keep the lytic potential intact, even for trypanosomes devoid of SRA. Thus, it would appear that in contrast to what occurs in E. coli, in trypanosomes the C-terminal domain of apoLl is not completely dispensable for the killing activity of the protein .

[0074] Interestingly, it was through inspiration driven by sequence analysis of Papio apoLl-like genes that the inventors were able to generate apoLl variants unable to bind SRA, but still able to efficiently kill trypanosomes. As expected, these variants killed NHS- resistant T. b. rhodesiense clones as well as NHS-sensitive T. b. rhodesiense clones or T. b. brucei, like occurs with Papio serum. This finding strongly suggests that in Papio serum similar apoLl variants are responsible for the trypanolytic activity. The existence of baboon apoLl variants able to resist SRA would be consistent with the recent proposal that variations of apoL sequences are frequent at sites of interaction with pathogen proteins [13], although the apoLl mutations/deletion described here were not described in this report.

[0075] As is the case with Papio serum, the Papio- like human apoLl was unable to kill T. b. gambiense . This observation is in keeping with the fact that in this subspecies the mechanism of resistance to apoLl must be different from that of T. b. rhodesiense, as SRA is absent from T. b. gambiense . Therefore, it appears that resistance to NHS in T. b. gambiense is independent from the C- terminal domain of apoLl .

[0076] The results lead to the generation of transgenic cattle able to resist infection by African trypanosomes in Eastern Africa. Indeed, mammals (mice) transiently expressing Papio-like human apoLl variants resist not only T. b. brucei, but also NHS-resistant clones of T. b. rhodesiense and the cattle pathogen T. congolense . Moreover, given the sensitivity of T. evansi to apoLl, a transgenic cattle should resist T. evansi as well, and similar prediction could be proposed for T. vivax.

[0077] Finally, in view of these results it can be envisaged that understanding the mechanism of resistance of T. b. gambiense to NHS would allow to generate mutant versions of apoLl also able to kill this parasite.

References 1. Vanhamme L et al . (2003) Nature 422: 83-87.

2. Perez-Morga D et al . (2005) Science 309: 469-472.

3. Xong H et al . (1998) Cell 95: 839-846.

4. Molina-Portela MP et al (2008) J Exp Med 205: 1721-1728.

5. Gaboriaud C et al(1987) FEBS Lett 224: 149-155. 6. Callebaut I, et al . (1997) Cell Mol Life Sci 53: 621- 645.

7. Thomas A, et al . (2006) Proteins 65: 889-897.

8. Hirumi H et al . (1989) J Parasitol 75:985-989.

9. Baltz T et al . (1985) EMBO J. 4:1273-1277. 10. Wirtz E, et al (1999) Mol Biochem Parasitol 99: 89-101.

11. Vanhollebeke B et al . (2007) J Eukaryot Microbiol. 54:448-51.

12. Kobayashi N et al .

(2004) http : //www . cbi . nlm. nih . gov/pubmed/15133769?ordinalpo s=5&itool=EntrezSystem2. PEntrez . Pubmed. Pubmed ResultsPanel . Pubmed DefaultReportPanel . Pubmed RVDocSum J Gene Med 6:584- 592.

Claims

1. A recombinant human Apolipoprotein L-I wherein its C-terminal portion of its wild-type sequence (SEQ. ID. O: 1, SEQ.ID.NO:8 or SEQ. ID. O: 25) comprises a replacement of its last 13 C-terminal amino acids (NNNYKILQADQEL) by an addition of the sequence TKIQKILQADQEL or an addition of 4 amino acids having a sequence selected from the group consisting of TKIQ, NKIQ, NNIQ, NNNQ, TNNY, TKNY, TKIY or NNNY .

2. The Apolipoprotein L-I according to the claim 1, which is selected from the group consisting of SEQ.ID.NO:2, SEQ.ID.NO:3, SEQ.ID.NO:4, SEQ.ID.NO:5, SEQ.ID.NO:6, SEQ.ID.NO:7, SEQ . ID . NO : 22 , SEQ.ID.NO:23 or SEQ. ID. NO: 24.

3. A polynucleotide encoding the Apolipoprotein L-I according to the claim 1 or 2.

4. A vector comprising Apolipoprotein L-I according to the claim 1 or 2 or the polynucleotide according to the claim 3.

5. A cell transformed by the polynucleotide according to the claim 3 or by the vector according to the claim 4.

6. A pharmaceutical composition comprising an adequate pharmaceutical carrier or diluent and a sufficient amount of an element selected from the group consisting of the Apolipoprotein L-I, the polynucleotide, the vector or the cell according to any of the preceding claims 1 to 5.

7. The pharmaceutical composition according to the claim 6 for use in the treatment and/or the prevention of diseases induced in human by Trypanosoma rhodesiense .

8. The pharmaceutical composition according to the claim 7 for use in the treatment and/or the prevention of diseases induced in cattle by Trypanosoma ( AGA A) .

9. The pharmaceutical composition according to the claim 8, wherein Trypanosoma are selected from the group consisting of Trypanosoma brucei brucei, Trypanosoma brucei rhodesiense, trypanosoma congolense, trypanosoma evansi and trypanosoma vivax.

10. A method of treatment and/or prevention of a disease affecting a mammal, which comprises the step of administrating a sufficient amount of the pharmaceutical composition of claim 6 to 9 to this mammal to reduce and/or suppress the symptoms of this disease in the said mammal.

11. A non human genetically modified mammal which is expressing the Apolipoprotein L-I according to the claim 1 or 2, the polynucleotide according to the claim 3, the vector according to the claim 4 comprising the cell according to the claim 5.

12. The non human genetically modified mammal according to the claim 11, which is a genetically modified cattle .

13. The non human genetically modified mammal according to the claim 11 or 12, which is a genetically modified cow.

14. The non human genetically modified mammal according to the claims 11 to 13, which is resistant or tolerant to diseases induced by Trypanosoma and related diseases .

15. The non human genetically modified mammal according to the preceding claims 11 to 14, wherein Trypanosoma is selected from the group consisting of Trypanosoma brucei brucei, Trypanosoma brucei rhodesiense, tryponosoma congolense, trypanosoma evansi and trypanosoma vivax .