AFRICAN TRYPANOSOMIASIS THERAPY WITH A NANOBODY-CONJUGATED HUMAN TRYPANOLYTIC FACTOR
The present invention relates to the treatment of trypanosomiasis using a nanobody- conjugated trypanolytic factor. More specifically, it relates to a therapy with nanobody- conjugated human apo-L1 , whereby said apo-L1 is preferably truncated at the C-terminal end to avoid neutralization by the trypanosomes.
African trypanosomes include several species and subspecies that infect a wide range of game and domestic animals as well as humans. Several of there parasites are the agents of diseases of major veterinary and/or medical importance. Whereas Trypanosoma brucei gambiense and Trypanosoma brucei rhodesiense are the causative agents of sleeping sickness in humans, Trypanosoma brucei brucei, Trypanosoma congolense and Trypanosoma vivax cause nagana in cattle and Trypanosoma evansi and Trypanosoma equiperdum are, respectively, responsible for surra in camels and dura in horses. The subspecies T b. rhodesiense, T b. gambiense and T b. brucei are closely related, the latter being distinguished by its sensitivity to normal human serum. The two former subspecies differ both in their geographical distribution and the acuteness of the disease they cause Trypanosoma brucei rhodesiense is the causative agent of human African trypanosomiasis (HAT) in Eastern Africa (Hitchinson et al., 2003; Barrett et al., 2003). In endemic foci in Uganda, Angola and Sudan, large scale livestock infections cause a continuous threat for epidemic outbreaks amongst rural populations (Barrett et al., 2003; Welburn et al., 2001 a; Welburn et al., 2001 b). In order to evade an effective host immune response, trypanosomes have acquired a system of antigenic variation (Pays et al., 2004; Horn, 2004). By rapidly altering the expression of their variant-specific surface glycoproteins (VSG) that constitute a dense layer covering their entire outer membrane, efficient antibody-mediated clearance is avoided. As trypanosome genomes contain hundreds of genes encoding different VSGs, there is little prospect for a conventional vaccine. Thus the treatment of sleeping sickness relies mainly on therapy. The disease is fatal, if untreated, and the two drugs currently available for treatment of T .b. rhodesiense, suramin and melarsoprol (Legros et a/., 2002) have serious adverse events due to drug toxicity, relapses, and long duration of the treatment (Burri et a/., 2000; Pepin and Milford, 1994). Furthermore, drug resistance in trypanosomes is increasing in the field, hindering efforts to control HAT (Bacchi, 1993). Hence, development of highly specific new drugs, preferentially with low toxicity as compared to the current drugs, is mandatory. Human natural defense against African trypanosomes is mediated by normal human serum (NHS) that lyses most parasites, except those that cause HAT (Vanhamme and Pays, 2004). Recently, a human-specific serum protein, apolipoprotein L-I (apoL-l), was identified as the trypanolytic factor of NHS (Vanhamme et al., 2003). When isolated from livestock T. b. rhodesiense is in general sensitive to lysis by human serum (Welburn et a/., 2001 a), but this
parasite is capable of switching to a resistant phenotype becoming infectious to humans (Van Meirvenne et al., 1976). This happens through antigenic variation-associated transcriptional switching that enables the parasite to express the serum resistant associated (SRA) protein, a VSG homologue that was shown to confer resistance to NHS (Xong et al., 1998; De Greef et al, 1994). SRA interacts with the C-terminal domain of apoL-l, hereby neutralizing the trypanolytic activity of the latter. Consequently, deleting the SRA-interacting domain of apoL-l resulted in a truncated apoL-l (Tr-apoL-l) that could not be neutralized by SRA and thus was capable of lyzing both NHS sensitive and resistant T. b. rhodesiense (Vanhamme et al., 2003). Tr-apoL-l represents a natural trypanolytic agent to cure T. b. rhodesiense infections. However, the pharmaco-kinetic properties of Tr-apoL-l constitute a possible major obstacle for its therapeutic use. For instance, competition with endogenous natural apoL-l (10 μg/ml in NHS) (Duchateau et al., 1997; Duchateau et al., 2000) might prevent an efficient delivery of Tr- apoL-l. Therefore, the Tr-apoL-l might only be successful in eliminating trypanosomes if targeted specifically to the parasites, preferably accessing conserved surface epitopes. Nanobodies (Nbs), the single-domain antigen-binding fragments derived from camel heavy- chain antibody, may represent targeting tools because of their small size (13 kD) and strict monomeric behavior (Nguyen et al., 2001 ; Muyldermans, 2001 ; Conrath et al., 2001 ). In particular, we have identified a nanobody (Nb) able to reach a conserved VSG carbohydrate epitope present on different sub-species of T. b. brucei parasites (cAbAn33, here referred as NbAn33, accession no. AY263490) (Stijlemans et al., 2004). Alternatively, the we developed a nanobody targeted to a transferring receptor subunit situated on the surface of Trypanosoma brucei. Surprisingly, we found that a fusion protein between these antiboies, and the truncated apoL-l is very efficient in clearing a trypanosome infection. This effect is unexpected, as the binding efficiency of the nanobody to the conserved VSG epitopes, or to the transferring receptor subunit is due to its small size, and the person skilled in the art would forecast that the attachment to the nanobody of a lager entity such as Tr-ApoL-l would lead to a seriously reduced binding capacity of the antibody.
A first aspect of the invention is a fusion protein comprising a single-domain antigen binding fragment derived from a camel heavy-chain antibody (= nanoboy) against a trypanosome specific invariant surface antigen, and a functional fragment or variant of ApoL-l. Preferably, said trypanosome specific invariant surface antigen is VSG, or a transferring receptor subunit such as pESAGδ or pEASG7. Anti-VSG specific nanobodies have been described by Stijlemans et al. (2004). Preferably, said trypanosome specific anti-VSG nanobody is NbAn33. Preferably said anti-transferrin subunit antibody is directed against pESAGδ. Even more preferably, said anti-transferrin subunit antibody is selected from the group consisting of NbESδ, NbES31 or NbES48. Most preferably, it is NbESδ or NdES31. A functional fragment or variant of ApoL-l is any fragment, including total ApoL-l, that is capable of lysing a NHS
sensitive strain of trypanosomes, preferably a Trypanosoma brucei brucei. Preferably, said functional fragment is a fragment that is capable of lysing NHS resistant Trypanosoma brucei rhodesiense. Even more preferably, said functional fragment or variant comprises a C-terminal mutation, deletion or truncation whereby the C-terminal α-helix of ApoL-l (A368-L385; numbered according GenPept accession number 014791 , GL67462123) has been disrupted or removed. Suitable functional fragments are, as a non-limiting example, the fragment with a C-terminal F343-L398 deletion and the fragment with a S356-L398 deletion. A suitable variant is, as a non-limiting example, carrying mutations at position L345, L352 and Y354, preferably L345Y, L352A and Y354L. Most preferably, said functional fragment has a deletion of at least aa 345-355 of apoL-1. A preferred embodiment is a functional fragment as comprised in SEQ ID N°1 (aa147-457). The anti-VSG nanobody and the ApoL-l functional fragment may be connected by a hinge fragment.
Preferably, said fusion protein comprises SEQ ID N° 1 , even more preferably said fusion protein consists of SEQ ID N° 1. Another aspect of the invention is the use of a fusion protein according to the invention for the preparation of a medicament.
Still another aspect of the invention is the use of a fusion protein according to the invention for the preparation of a medicament to treat trypanosome infection in a mammal, including human. Preferably, said trypanosome infection is an African trypanosome. More preferably said trypanosome is selected from the group consisting of Trypanosoma brucei gambiense, Trypanosoma brucei rhodesiense, Trypanosoma brucei brucei, Trypanosoma congolense, Trypanosoma vivax, Trypanosoma evansi and Trypanosoma equiperdum. Even more preferably, said trypanosome is a Trypanosoma brucei, most preferably, it is a trypanosome causing HAT and belonging to the subspecies Trypanosoma brucei rhodesiense. Still another aspect of the invention is a pharmaceutical composition comprising a fusion protein according to the invention. The fusion protein of the invention may be mixed with a pharmaceutically acceptable carrier. The pharmaceutical composition may be any pharmaceutical composition known to the person skilled in the art. Preferably, said pharmaceutical composition is intended for injection into the patient.
BRIEF DESCRIPTION OF THE FIGURES
Figure 1 : Targeting modules for the trypanosome surface, a. Schematic representation of the different constructs (NbAn33-apoL-l, NbAn33-Tr-apoL-l, NbCEA5-Tr-apoL-l) that were used in this study. The N and C-end of the molecules are indicated and the numbers refer to the amino acid positions of apoL-l. The Signal peptide and SRA interacting domain are absent in the Tr- apoL-l form. Chimeras contain a six histidine (HiS6) tag at the C terminal end for purification purposes, b. Staining of T b. brucei AnT at 1.1 in the blood smear of an infected mouse
incubated with Alexa Fluoro 488-labelled NbAn33-Tr-apol_-l illuminated with visible light (Phase) or UV (Alexa).
Figure 2: In vitro trypanolysis. a. Concentration-dependent lytic activity of NbAn33-Tr-apoL-l on T. b. brucei AnTat 1.1. (after 4 h incubation) compared to lysis by 50% NHS or background lysis in presence of 10μg/ml NbAn33. b. Trypanolytic activity of different chimeras (10μg/ml, 4 h incubation) on the T. b. brucei AnTat 1.1 and T. b. rhodesiense ETat 1.2S (S) and ETat 1.2R (R) clones.
Figure 3: Therapeutic effects of NbAn33-Tr-apoL-l in acute infection. Parasitemia (a, c) and survival (b, d) of mice infected with T. b. rhodesiense ETat 1.2 R (a, b) or T. b. brucei (c, d) and treated with NbAn33-Tr-apoL-l 20μg( " ),10μg(»), NbCEA-Tr-apoL-l, 100μg(o) or NHS (in panel a. and b) or PBS (in panel c and d) (D) at day 3 post infection (arrow). Figure 4: Therapeutic effects of NbAn33-Tr-apoL-l treatment at the second peak of parasitemia. Parasitemia (a), survival (b) and anemia expressed in % of red blood cells (RBC) (c) of T. b. brucei infected mice treated with NbAn33-Tr-apoL-l ( " ),NbCEA-Tr-apoL-l (o) or PBS (π)at day 12 and 14 days post infection (arrows).
Figure 5: (a) Coomassie stained SDS-PAGE of purified recombinant pESAGδ. (b) The IgG immune response of dromedary immunized with recombinant pESAGδ, as measured by ELISA. The fractionated IgGI (filled square), lgG2 (open square) and lgG3 (open circle) bound to coated pESAGδ were detected with anti-dromedary IgG antiserum, (c) Three consecutive rounds of panning as described by Conrath et a/.(2001 b) were used to select pESAGδ-binding Nbs. After two pannings a significant enrichment of phages specifically binding to pESAGδ was observed, (d) Amino acid sequence alignment of three anti-pESAG 6 Nbs (NbES) with NbBCIH O according to IMGT (http://lmgt.cines.fr/) and Kabat numbering (Kabat et al. 1991 ). Figure 6: (a) Coomassie stained SDS-PAGE of different anti-pESAGδ Nbs (NbESs) after IMAC and gel filtration purification, (b) The kon, kof and kD of different NbESs determined by plasmon surface resonance in BIAcore. (c) Epitope mapping of NbESs by BIAcore analysis. First one Nb was added (indicated by the arrow at 200 sec) to the pESAGδ-coupled chip and after 100 sec another Nb was added (indicated by the arrow at 300 sec) to analyze the binding of second Nb. Figure 7: Analysis of NbESs binding to different trypanosomes. The fluorescently-labelled NbESs were incubated with different purified fixed parasites and were analyzed by FACS. Figure 8: (a) Uptake of FITC-labelled Tf by trypanosomes that were pre-incubated with different antipESAGδ Nbs (NbESδ, NbES31 and NbES48), or without any Nb (No Nb). The levels of internalized Tf were measured after different time points, (b) Growth of in vitro cultured trypanosomes in complete HMI-9 medium where different NbESs (NbESδ, NbES31 , and NbES48) at the concentration of 100 μg/ml, or no Nb (No Nb) was added, (c) In vitro trypanolysis of T. b. rhodesiense AnTat1.2R with different concentrations of NbES31-Tr-
APOL1 in 4 h time. The values are expressed as the percentage of the starting number of the parasites (106/ml). (d) In vitro trypanolysis of different trypanosomes with NbES31 -Tr-APOLI (20ug/ml) in 4 h.
Figure 9: Parasitemia (a and d) and mice survival (c and d) during infection by T. b. rhodensiense ETat 1.2R and treatment at day three post-infection by intravenous injection of NbES31 -Tr-APOLI (a and b) or NbES8-Tr-APOL1 (c and d) (10 μg or more per mouse (filled square), 5 μg per mouse (open circle), or control buffer (open square)).
Materials and methods to the examples
Cloning and Expression of Nb-apoL-l
The truncated apoL-l gene was amplified by polymerase chain reaction (PCR) from a plasmid containing the apoL-l gene1 as No1\-Xho\ fragment with the following primer set (the No1\-Xho\ restriction sites are underlined):
Apo42F: 5'-ATAAGAATGCGGCCGCAGAGGAAGCTGGAGCGAGGGT- 3'
The amplified TR-apoL-l fragment was cloned in Not\-Xho\ digested pET-21d (+) plasmid
(Novagen). The NbAn33-llama γ2c hinge was amplified as βamHI-Λ/ofl fragment with the following primer set (BamH-\-No1\ sites are underlined):
Nb F: 5'- ATCCGGGATCCCAGATGTGCAGCTGGTGGACTCT-S'
The reverse primer includes the nucleotide sequence of the 15-mer llama γ2c hinge, coding for amino acid sequence AHHSEDPSSKAPKAP. The amplified fragment was cloned in the pET- 21 d (+) plasmid, containing Tr-apoL-l described above. After sequence verification, the construct was transfected in BL 21 Escherichia coli cells. The anti-CEA nanobody (Nb-CEA5) conjugated to Tr-apoL-l was engineered likewise. As additional control, full apoL-l was amplified as Not\-Xho\ fragment with the following primers set: Apo42F (see above) and ApoR: δ'-ATCCGCTCGAGCAGTTCTTGGTCCGCCTGCA-S'. All recombinant proteins were expressed as cytosolic His-tagged products. After lysis of the transformed BL21 cells, the recombinant fusion proteins were purified using a Ni-NTA Superflow column (Qiagen). Further purification was performed by gel filtration in HPLC using a Superdex 75 (10/30) column (Pharmacia, Akta Explorer 10S) equilibrated with PBS. The proteins were checked in 10% SDS-PAGE.
Cloning and expression of ESAG 6
An ESAG6 gene fragment devoid of the sequences encoding the N- and C-terminal signal peptides (respectively 18 and 32 amino acids) was amplified as a Sph\-Sma\ fragment by polymerase chain reaction (PCR) from a plasmid containing ESAG6 from T. b. rhodesiense ETat 1.2R (Xong et al., 1998), using the following primer set : ESAG6 TR F: 5'- TTACATGCATGCGAAAATGAAAGGAATGCATTAAACG-3' ESAG6 TR R: 5'- TTTCCCCCGGGTCATCCACGGACAGGTTCTGC-3'. The PCR fragment was cloned in Sph\ /Sma\ digested pQE80 plasmid (Qiagen) which was transformed in Topl OF' Escherichia coli. Recombinant pESAGδ was expressed by inducing the bacterial culture with 1 mM isopropyl-β- D-thiogalactopyranoside (IPTG). Cells were pelleted after 3 h of induction and were lysed by brief sonication.
Purification of His6-tagged recombinant protein.
The Hisδ-tagged recombinant protein was purified by Immobilized Metal Affinity Chromatography (IMAC) using a Ni-NTA Superflow column (Qiagen). Further purification was performed by gel filtration in HPLC using a Superdex 75 (10/30) column (Pharmacia, Akta Explorer 10S) equilibrated with PBS. The samples were checked for purity by SDS-PAGE and their concentration was measured by BCA kit (PIERCE) where known concentration of bovine serum albumin (BSA) was used as standard.
Immunization for production of antibodies against pESAGβ
One dromedary (Camelus) at the CentralVeterinary Laboratory (Dubai, U.A.E) was immunized six times subcutaneously every week with 1 mg recombinant pESAGδ mixed in Gerbu adjuvant (GERBBU Biochemicals) (Lauwereys, 1998). Two weeks after the final immunization, 50 ml anticoagulated blood was collected from which plasma and peripheral blood lymphocytes were isolated (Lymphoprep).
Fractionation of IgG subclasses
Separation of the different plasmaimmunoglobulin G (IgG) subclasses was done byadsorption on Hitrap-protA and Hitrap-protGcolumns (Amersham Biosciences). lgG3 andlgGI subclasses were eluted from protG columnwith acetate buffer (pH 3.5) and glycine buffer (pH 2.7), respectively (Conrath et al., 2001 a). The flow through was subsequently loaded on a protA column to recover the lgG2 subclass by elution with the acetate buffer (pH 3.5).
The mRNA was extracted from peripheral blood lymphocytes by using QuickprepTM micro mRNA purification kit (Amersham Bioscience). cDNA was synthesized by using Ready-to-Go
You-prime-strand-beads (Amersham Bioscience). Primers CALL001 (5'- GTCCTGGCTCTCTTCTACAAGG-3') and CALL002 (5'- GGTACGTGCTGTTGAACTGTTCC- 3') were used in PCR to amplify the variable domains of both VHH (600 bp) and VH (900 bp). These two fragments were separated in agarose gel and the VHH gene fragments were purified. A nested PCR, using primers A6E (5'- GATGTGCAGCTGCAGGAGTCTGGAGGAG G-3', which anneals on framework -1 ) and FR4FOR (5'- GGACTAGTGCGGCCGCTGGAGACGGTGACCTGGGT- 3', which anneals on framework -4) was performed to amplify all VHH genes. The final PCR fragments were ligated into the phagemid vector pHEN4 (Arabi Gharoudi et ai, 1997) using the restriction sites Psfl and Λ/ofl. The ligated vector was used to transform electrocompetent E. coli cells (TG 1 ).
Selection of binders ofpESAGβ
The VHH repertoire was expressed on phage after infection with M13K07 helper phages.
Specific VHHs against pESAGδ were enriched by several consecutive rounds of in vitro selection on microtiter plates coated with antigen (10μg/ml). 100 mM triethylamine (pH 1 1.0) was used to elute the bound phage particles. The elutes were immediately neutralized with 1 M Tris-HCI (pH 7.4) and were used to infect exponentially growing TG1 cells. To assess the enrichment of phage particles carrying antigen-specific VHHs, a serial dilution of the phages eluted from antigen coated versus non-coated wells was used to transfect the exponentially growing TG1 cells. Individual colonies were picked and the cloned VHHs were expressed as soluble periplasmic proteins by induction with 1 mM IPTG. Recombinant VHHs extracted from the periplasm were tested for antigen recognition in an ELISA.
Expression of single-domain antibody against pESAGβ The VHH genes of the clones that scored positive in ELISA were re-cloned using the restriction enzymes Psfl and SsEII, into the expression vector pHENδ as described (Conrath et a/., 2001 a). The ligated plasmids were transformed into E. coli WK6 su-/- cells. Production of recombinant single-domain Nbs was performed by growing the bacteria in Terrific Broth supplemented with 0.1% glucose and ampicillin. When the optical density at 600 nm (OD600) was between 0.6 and 0.9, VHH expression was induced with 1 mM IPTG overnight at 28°C. After pelleting the cells, periplasmic proteins were extracted by applying osmotic shock (Skerra and Plukthun, 1988) and the Nbs were purified as His6-tagged proteins.
Solid-phase ELISA Maxisorb 96-well plates (Nunc) were coated with 1 μg/ml recombinant pESAGδ overnight at 4°C in phosphate-buffered saline (PBS). 1 % casein solution in PBS was used for 2 h at room temperature to block the residual protein-binding sites in the wells. Afterwards, serial diluted
fractionated IgGs or soluble recombinant VHHs were added to the wells. Detection of dromedary IgGs and VHHs was done using a rabbit anti-dromedary IgG antiserum and a mouse anti-hemagglutinin decapeptide tag (Covance/Babco) or a mouse anti-histidine tag (Serotec), respectively. Subsequent detection was done with horseradish peroxidase anti- rabbit or anti-mouse conjugate (Sigma). Finally, the substrate for peroxidase (2, 2'-azino-bis (3-ethylbenzathiazoline-6-sulfonic acid) was added and the absorption at 405 nm was measured after 15 min.
Affinity measurement of anit-pESAG6 nanobodies The affinities of the Nbs were measured with the BIAcore-3000 instrument (Biacore). First, pESAGδ was coupled to a CM chip according to the manufacturers' instructions (Biacore). Then different concentrations of Nbs (NbESδ, NbES31 , NbES48), ranging from 7.85 nM to 500 nM, were added. The measurements were performed with a flow rate of 30μl/ml in 150 mM EDTA, 0.005% Tween-20, 10 mM Hepes (pH 7.5). For affinity measurement of NbESδ, which did not bind to the CM-chip coupled pESAGδ; the Nb was biotinylated and subsequently coupled on streptavidin coated chip. The kinetic parameters kon, koff and kD were calculated by the BIA-evaluation software version 3.0 (Biacore). For the epitope mapping of the Nbs, one Nb was first added in saturating amounts on a chip containing pESAGδ, and then saturating amounts of the first Nband a second Nb were loaded, the resulting resonance units were compared with the signal obtained when adding saturating amounts of only the respective Nb on the chip.
Cloning and Expression of NbES-Tr-APOU
Cloning and expression of Nb-conjugated truncated APOL1 was performed as described above. The Nb-llama γ2c hinge was amplified with a reverse primer coding for the amino acid sequence AHHSEDPSSKAPKAP. This amplified fragment was cloned in the pET-21d (+) plasmid containing Tr-APOU. After sequence verification, the construct was transformed in E. coli BL21 cells. The control Nb conjugated to Tr-APOLI was engineered likewise. All recombinant proteins were expressed in the cytosol as Hisδ- tagged and were purified as described above.
Trypanosoma brucei brucei AnTat 1.1 , T b. rhodesiense ETat 1.2R, T. b. gambiense ABBA, T evansi KETRI 2480 and T congolense TC13 were used. Infected mice with one of these parasites were sacrificed at the peak of parasitemia. Parasites were purified from heparinized blood on DEAE-cellulose (DE52, Whatman) column (Lanham and Godfrey; 1970). Purified
parasites were fixed with a fixing solution (PBS containing 3% formaldehyde and 1 % glutaraldehyde).
Transferrin uptake assay 106 parasites (in 1 % fetal calf serum (FCS) in PBS) in a volume of 100 μl were, separately, pre-incubated with 20 μg of different NbESs at 37°C for 1 h. Afterwards, 3 μg of FITC- labeled human Tf (Sigma) was added and incubated at room temperature. After 10, 30 or 60 min, parasites were washed and the amount of FITC labelled Tf retained in the parasites was measured by fluorescence reading (Cytoflour II, Perceptive Biosystem).
In vitro parasite culture
T. b. rhodesiense ETati .2R parasites were purified, washed and resuspended at 2 x 105 ml-1 cells in complete HMI-9 medium (Hirumi and Hirumi, 1994) supplemented with 10 % FCS. Trypanosomes were counted and then diluted at 105 ml-1 cells in the same medium every 24 h. To check the effect of anti-pESAG6 Nbs on growth of trypanosomes in vitro, different NbESs were added at the concentration of 100 μg/ml which is equal in molarity to Tf in 10% FCS. Parasites were counted every 24 h and cumulative parasite numbers were calculated.
Binding of nanobodies to parasites. All Nbs were labeled with Alexa 488 (Molecular probes) and the labeling efficiency was determined spectrophotometrically according to manufacturers' instruction. The purified and fixed parasites were first washed with PBS and then incubated at room temperature with different Alexa 488- labelled Nbs. After 30 min of incubation, parasites were washed twice with PBS to remove unbound Nbs. The slides were visualized with visible as well as UV light for immuno-fluoroscence microscopy (Nikon ECLIPSE E600 with phase contrast) for the NbAn33- Tr-apoL-l construct, and analyzed by Fluorescence-Activated Cell Sorting (BD Bio-sciences), in case of the anti-pESAG6 antibodies.
In vitro trypanolysis assay. T. b. brucei (AnTat 1.1 ), T. b. rhodesiense ETat 1.2 S (sensitive to NHS) and T. b. rhodesiense ETat 1.2 R (resistant to NHS) clones were used. Purified parasites (106 parasites/ml phosphate saline-glucose buffer with 5 % fetal calf serum) were incubated with different chimeric proteins at 37°C. Live parasites were counted every hour under the microscope to calculate the trypanolytic capacity of chimeras.
In vivo therapy experiment.
C57/bl mice (4 animals/group) were infected intra peritoneally (i.p.) with 5,000 parasites. Once parasites were detected in blood by microscopy, mice received different i.p. doses (from 10 μg to 100 μg/mouse) of NbAn33-Tr-apoL-l or NbES-T r-apoLI. Control mice were either untreated or received NbCEA5-Tr-apoL-l or NbAn33-apoL-l. After treatment, the parasitemia was followed microscopically every alternate day and survival of the mice was recorded. To ensure the NHS resistance of T. b. rhodesiense ETat 1.2 R, mice i.p injected with 500μl of NHS were infected with 5000 parasites.
For the treatment after the second peak of parasitemia, mice infected with T. b. brucei (AnTat 1.1 ) were left untreated on the first peak of parasitemia. When the second wave of parasites was detected microscopically, mice received i.p. injections of NbAn33-Tr-apoL-l (20 μg/animal), twice on alternate days. Control mice were either untreated or received NbCEA5- Tr-apoL-l. Parasitemia and survival were recorded as described above.
Example 1 : construction of a nanobody-ApoL-l fusion
Fig. 1 a summarizes the design of chimerical constructs of Nbs and apoL-l that were generated and tested in this study. ApoL-l contains an N-terminal signal peptide (aa 1-27), a C-terminal SRA interacting domain (aa 343-398) and a central lytic domain (aa 28-342) (Vanhamme et al., 2003; Perez-Morga et al, 2005). In the truncated form of apoL-l (Tr-apoL-l) signal peptide was removed and the C-terminal domain was deleted to abrogate recognition and neutralization by SRA. A Nb encoding region was added, separated from Tr-apoL-l by a sequence encoding the natural llama γ2c antibody hinge, allowing independent folding of two protein subunits (Roux et al., 1998). Fig. 1 b shows that the NbAn33-Tr-apoL-l chimera specifically recognized trypanosomes and did not bind to mammalian host cells.
Example 2: NbAn33-Tr-apo-L-l can lyse both sensitive as well as resistant trypanosomes in vitro.
NbAn33-Tr-apoL-l exerted a dose-dependent trypanolytic activity when tested on parasite cultures with 10 μg/ml (180 nM) lysing 100% of the parasites within 4 hours of incubation (Fig.2a). In this experimental setting, a 20 molar excess of NbAn33 alone had no lytic activity, and around 70% of the parasites lysed in 50% NHS containing approximately 5 μg/ml apoL-l. In order to evaluate the contribution of the targeting module to the lytic activity, the NbAn33 sequence was substituted by NbCEA5 (encoding cAbCEAδ accession no. Ax800153) (Cortez- Retamozo et al., 2004), a Nb recognizing an antigen absent in both trypanosomes and healthy hosts. No lysis of trypanosomes was observed in presence of NbCEA5-Tr-apoL-l (Fig. 2b), demonstrating the crucial importance of trypanosome-surface targeting provided by NbAn33.
This was further corroborated by complete inhibition of the lytic activity upon pre-incubation of
NbAn33-Tr-apol_-l with soluble VSG (Fig. 2b). NbAn33-Tr-apol_-l was also highly trypanolytic for NHS resistant T. b. rhodesiense ETat 1.2R (Fig. 2b). Truncation of the SRA-interacting domain was absolutely required for this activity. Indeed, while NbAn33-Tr-apoL-l lysed both NHS resistant (ETat 1.2R) and sensitive (ETat 1.2S) clones with equal efficiency, NbAn33- apoL-l was only lytic for ETat 1.2S (Fig. 2b). Collectively, these data showed that in vitro, NbAn33-Tr-apoL-l lyses NHS sensitive T. b. brucei as well as NHS resistant and sensitive T. b. rhodesiense parasites in a highly specific and efficient manner.
Example 3: NbAn33-Tr-apoL-l can cure trypanosome infection in vivo NbAn33-Tr-apoL-l was tested as a potential curative treatment in murine models for HAT. First, mice were infected with the virulent NHS-resistant T. b. rhodesiense ETat 1.2R, and upon detection of parasites in the blood (day 3) animals were treated with a single inoculation of NbAn33-Tr-apoL-l or NbCEA5-Tr-apoL-l, ranging from 10 up to 100 μg/mouse. Treatment with one single dose of 20 μg (Fig. 3a) and higher of NbAn33-Tr-apoL-l resulted in complete parasite clearance and long term survival (60 days when experiments were terminated). Mice treated with a sub-optimal dose of 10 μg/mouse were partially protected as evidenced by a delayed parasitemia and longer median survival time of 9 days as compared to control NHS- treated mice (6 days). Using non-targeted NbCEA5-Tr-apoL-l at a high dose (100 μg/mouse) only a marginal protection effect was obtained (Fig.3b), which was similar to the effect obtained with recombinant apoL-l at the same dose, the amount of apoL-l required to abrogate parasitemia being around 1 mg/mouse.
Second, NbAn33-Tr-apoL-l treatment was assessed in a more chronic T. b. brucei pleomorphic AnTat 1.1 infection model. Also here, parasitemia was cleared completely with one single dose of 20μg NbAn33-Tr-apoL-l (Fig. 3c) and mice survived for up to 150 days, when the experiment was terminated (Fig. 3d). Mice that were treated with sub-curative dose of 10 μg showed reduced first peak of parasitemia and prolonged survival but did succumb to late stage parasitemia (median survival of 61 days). Control mice, treated with phosphate saline buffer or 100 μg of the non-targeting NbCEA5-Tr-apoL-l chimera, died due to parasitemia with a median survival of 30 and 36 days, respectively. Finally, curative potential of NbAn33-Tr-apoL-l treatment was evaluated during chronic phase of infection. To this end, mice were infected with T. b. brucei AnTat 1.1 and before the second wave of parasitemia (day 12), animals were injected twice with 20μg of NbAn33-Tr-apoL-l at a two days interval. Parasitemia was promptly cleared and treated mice remained parasite-free for up to 50 days (Fig. 4a). However, between days 55 and 65 of infection parasites reappeared in the blood, and finally, mice did succumb with a median survival of 70 days (Fig. 4b). Control mice that had been treated with 20 μg of the non-targeted NbCEA5-Tr-apoL-l
chimera showed unaltered parasitemia and similar survival time (median survival: 30 days) as PBS-treated mice.
During chronic trypanosome infection, anemia is associated with systemic inflammation (Chisi et a/., 2004; Naessens et al., 2005). Treatment with 20 μg of NbAn33-Tr-apoL-l at the onset of second peak of parasitemia rescued infection-associated anemia, and RBC levels remained normal until the parasites reappeared in the blood (Fig. 4c). Even though a 20 μg/mouse treatment was not sufficient to completely cure a chronic infection, it was beneficial in alleviating infection-associated pathology. Additional NbAn33-Tr-apoL-l treatment (20 μg/mouse) upon reappearance of parasites in the bloodstream (day 60) had no effect on both parasitemia level and survival (data not shown). Thus, in this experimental setting the infected host became gradually refractory to the therapeutical efficacy of NbAn33-Tr-apoL-l chimera, with full parasite control at the first peak, partial control at the second and no control at the final lethal peak. Selection of resistant parasites does not occur in this process since purified parasites from treated/infected animals were found to be equally susceptible to the lytic activity of the NbAn33-Tr-apoL-l chimera in vitro as well as in vivo (data not shown). Rather induction of host humoral responses should be considered here, in view of the observation that sera from mice treated twice with the chimera (50 μg/mouse intra-peritoneal at alternate day and boosted after 10 days) reacted strongly with apoL-l (ELISA signal OD405 > 0.65 at 1 :500 dilution) and neutralized apoL-l activity, but did not react against NbAn33. However in human treatment this type of humoral response should not occur, as apoL-l is a human self-antigen. In addition, the nanobody moiety is not expected to raise an immune response in humans because its sequence is almost identical to that of human variable immunoglobulin domains and, if necessary, the nanobodies could be further 'humanised' (Conrath et al., 2005).
Example 4: induction and selection of anti-pESAG6 antibodies pESAGδ, one of the two subunits of the heterodimeric Tf-R of T. brucei, was expressed as a recombinant antigen to immunize a camel. To this end, the ESAG6 gene was amplified by PCR and cloned in the pQE 80 vector to transform E. coli. The recombinant protein was expressed in the cytosol after induction with IPTG. His6-tagged recombinant pESAGδ was purified to near-homogeneity by chromatography on Ni-column and gel filtration (Fig. 5a). A dromedary was immunized with the recombinant pESAGδ, and the immunoglobulins from the plasma of the immunized dromedary were subfractionated to conventional (IgGI ) and Heavychain (lgG2 and lgG3) antibodies by differential affinity to protein A and protein G (Conrath et al. ,2001 a) Serial dilutions of these fractionated IgGs subclasses were used in a solid-phase ELISA to evaluate the immune response against pESAGδ. An antigenspecific
immune response composed of both Heavy-chain (lgG2 and lgG3) and conventional antibodies (IgGI ) was observed (Fig. 5b).
Lymphocytes were purified from the blood of the pESAGδ-immunized dromedary. Total mRNA was purified and cDNA was generated by reverse transcription using an oligodT primer. This cDNA was used to amplify the genes encoding variable domains of the heavy chain antibodies. The VHHs were cloned in the pHEN4 phagemid vector (Arabi Ghahroudi et a/., 1997) and transformed in E. coli. A VHH library of 107 transformants was obtained where more than 80% of the clones contained a VHH gene insert, as determined by colony PCR. This library was used to select pESAGδ-binding Nbs, using three consecutive rounds of panning as described by Conrath et a/.(2001 b). After two pannings a significant enrichment of phages specifically binding to pESAGδ was observed (Fig. 5c). After a third round of panning 48 clones were randomly chosen and used to express their VHH proteins (Nbs) as soluble proteins in the periplasm. These periplasmic extracts were tested on ELISA to identify the clones containing pESAGδ-specific VHH. 18 out of 48 clones scored positive. A restriction fragment length polymorphism (RFLP) analysis of these 18 VHHs was performed with a frequent cutter restriction enzyme Hinft to analyze whether these binders were different or similar. By this analysis three different VHHs fingerprints were identified which were further confirmed by DNA sequencing. Accordingly, three pESAGδ-specific VHHs were generated and isolated, i.e. NbESδ, NbES31 and NbES48 (Fig. 5d). These three Nbs harboured the VHH hallmark amino acids substitutions in framework 2 (Val37Phe, Gly44Glu, Leu45Arg and Trp47Xaa, in Kabat numbering). Furthermore, NbESδ and NbES48 contained two extra cysteines, Cys33 and Cysi OOb for NbESδ, and Cys50 and Cysi OOd for NbES48, which might be involved in formation of an extra disulfide bond. The CDR3 of all Nbs were of similar length (16-20 amino acids) and there was no difference in the length of CDR1 and CDR2.
Example 5: Characterization of the pESAG 6-specific nanobodies
The VHHs were recloned in the expression vector pHENδ (Conrath et al. 2001 a) and were transformed into E.coli. After sequence verification, the Nbs were expressed as Hisδ-tagged soluble proteins in the periplasm and purified by chromatography on Ni-column and gel filtration. Fig. 6 a shows that this two-steps purification resulted in more than 95% pure protein. The affinity of these Nbs on immobilized recombinant pESAGδ was determined by surface plasmon resonance measurements. NbESδ did not bind to EDC/NHS immobilized pESAGδ, and was therefore biotinylated before immobilization on streptavidin chips. The affinity for all these Nbs was in the nanomolar range (85 nM, 77 nM and 104 nM for NbESδ, NbES31 and NbES48 respectively) (Fig. 6b). Next, we analyzed whether the different Nbs recognize distinct epitopes on pESAGδ. To this end, pESAGδ was first immobilized on a chip.
Subsequently one of the three Nbs was allowed to bind at saturating concentration to the immobilized pESAGδ, followed by addition of saturating amounts of both the first and a second Nb. When NbES31 was first allowed to bind and then NbES48 was added, or vice versa, the binding of either of these Nbs was not influenced by the presence of the other Nb (Fig. 6c), suggesting that NbES31 and NbES48 recognize different epitopes. As mentioned above, NbES8 did not bind to immobilized pESAGδ, suggesting that this Nb binds to an epitope that is not accessible when pESAGδ is immobilized via EDC-chemical linkage on the chip and is therefore different from the epitope recognized by nbES31 and NbES48. Clones from different species and subspecies of trypanosomes, namely T. b. brucei AnTat 1.1 , T. b. rhodesiense ETat 1.2 R, T. b. gambiense ABBA, T. evansi KETRI 2480 and T. congolense TC13 were incubated with Alexalabelled NbESs after fixation. Cytofluorimetry measurements demonstrated the binding of the three NbESs to all trypanosomes except T. congolense (Fig. 7). Therefore, the three NbESs might bind conserved pESAGδ epitopes shared by different strains of closely related trypanosomes. T. b. rhodesiense ETat 1.2R parasites were pre-incubated with the different NbESs at 37°C. FITC-labelled Tf was added and the trypanosomes were further incubated at 37°C. At different time points, the parasites were washed and the amount of parasite-associated FITCIabelled Tf was measured. While the uptake of Tf was unaffected by NbES31 or NbES48, it was clearly inhibited by NbES8 (Fig. 8a). This observation suggests that NbES8 interferes with either Tf binding or cellular internalization of the trypanosome Tf-R. As NbES8 appeared to block Tf uptake, we analyzed its effect on trypanosome growth. T. b. rhodesiense ETat 1.2 R were cultured in vitro in the presence of different NbESs. While NbES31 and NbES48 were without effect, NbES8 significantly hampered trypanosome growth, in keeping with its effect on Tf uptake (Fig. 8b).
Example 6: Anti-pESAG6 nanobody conjugated with truncated APOL1 (NbES-Tr-APOLI ) exerts trypanolytic activity in vitro
To further corroborate the approach of targeting ApoLI to the trypanosomes by fusion with antibody directed to a trypanosome surface antigen, one of the NbESs (i.e.NbES31 ) was fused with truncated APOL1. N bES31 -Tr-APOLI exerted a dose-dependent trypanolytic activity, with 20 μg/ml lysing 100% of the parasites within 4 h of incubation (Fig. 8c). Substitution of NbES31 by NbCEA5, an irrelevant Nb, yielded only background trypanolysis at this concentration (Fig. 8c), demonstrating that targeting of the parasite surface by NbES31 is crucial for efficient lytic activity. Furthermore, the in vitro lytic capacity of NbES31 -Tr-APOLI was tested on other trypanosomes that are recognized by the NbES31. NbES31 -Tr-APOLI efficiently lysed most of the trypanosomes tested, including NHSresistant ETat1.2R, with the exception of T. b. gambiense for which there was only background lysis (Fig. 8d).
Example 7: In vivo treatment with nanobody-conjugated truncated APOL1 NbES31 -Tr-APOLI was tested as a potential curative treatment in mouse models for trypanosomiasis. Mice were infected with the virulent NHS-resistant T. b. rhodesiense ETat 1.2R, and upon detection of parasites in the blood (day 3) animals were treated intravenously with a single inoculation of NbES31 -Tr-APOLI , ranging from 5 up to 50 μg/mouse. Treatment with one single dose of 10 μg and higher of NbAn33-Tr-APOL1 resulted in complete parasite clearance (Fig. 9a) and long term survival (60 days when experiments were terminated, Fig. 9b) without showing any unwanted symptoms. Mice treated with a sub-optimal dose of 5 μg/mouse were partially protected as evidenced by a delayed parasitemia (Fig. 9a) and longer median survival time of 13 days as compared to control NHS-treated mice (6 days, Fig. 9c). Similar treatment was repeated with NbESδ, and the results were very similar regarding both parasite development (Fig. 9c) and survival time (Fig. 9d) of the treated mice.
Bacchi, CJ. Resistance to clinical drugs in African trypanosomes. Parasitol Today 9, 190-3
Barrett, M. P., Burchmore, R.J., Stich.A., Lazzari, J.O., Frasch, A.C., Cazzulo, J.J. and Krischna, S. The trypanosomiases. Lancet 362, 1469-80 (2003).
Burri, C, Nkunku, S., Merolle, A., Smith, T., Blum, J. and Brun, R. Efficacy of new, concise schedule for melarsoprol in treatment of sleeping sickness caused by Trypanosoma brucei gambiense: a randomised trial. Lancet 355, 1419-25 (2000).
Chisi, J. E., Misiri, H., Zverev, Y., Nkhoma, A. and Sternberg, J. M. Anaemia in human African trypanosomiasis caused by Trypanosoma brucei rhodesiense. East Afr Med J 81 , 505-8 (2004).
Conrath, K. E., Lauwereys, M., Wyns, L. and Muyldermans, S. Camel single-domain antibodies as modular building units in bispecific and bivalent antibody constructs. J Biol Chem 276, 7346-50 (2001 ). Conrath, K. , Vincke C, Stijlemans, B., Schymkowitz, J., Decanniere, K., Wyns, L., Muyldermans, S. and Loris, R. Antigen Binding and Solubility Effects upon the Veneering of a Camel VHH in Framework-2 to Mimic a VH. J MoI Biol 350, 1 12-125 (2005).
Cortez-Retamozo, V., Backmann, N ., Senter, P. D., Wernery, U., De Baetselier, P., Muyldermans, S and Revets, H. Efficient cancer therapy with a nanobody-based conjugate. Cancer Res 64, 2853-7 (2004).
De Greef, C. and Hamers, R. The serum resistance-associated (SRA) gene of Trypanosoma brucei rhodesiense encodes a variant surface glycoprotein-like protein. MoI Biochem Parasitol 68, 277-84 (1994).
Duchateau, P. N., Pullinger, C. R., Orellana, R. E., Kunitake, ST., Naya-Vigne, J., O'Connor, P.M., Malloy, M.J. and Kane, J. P. Apolipoprotein L, a new human high density lipoprotein apolipoprotein expressed by the pancreas. Identification, cloning, characterization, and plasma distribution of apolipoprotein L. J Biol Chem 272, 25576-82 (1997).
Duchateau, P.N. Movsesyan, I., Yamashita, S., Sakai, N., Hirano, K., Schoenhaus, S.A., O'Conner-Keams, P.M., Spencer, S. J., Jaffe, R. B., Redberg, R. F., Ishida, B.Y., Matsuzawa, Y., Kane, J. P. and Malloy, M.J. Plasma apolipoprotein L concentrations correlate with plasma triglycerides and cholesterol levels in normolipidemic, hyperlipidemic, and diabetic subjects. J Lipid Res 41 , 1231-6 (2000).
Horn, D. The molecular control of antigenic variation in Trypanosoma brucei. Curr MoI Med 4,
Hutchinson, O. C, Fevre, E. M., Carrington, M. and Welburn, S. C. Lessons learned from the emergence of a new Trypanosoma brucei rhodesiense sleeping sickness focus in Uganda. Lancet Infect Dis 3, 42-5 (2003).
Legros, D. Ollivier, G., Gastellu-Etchegorry, M., Paquet.C, Burri, C, Jannin, J. and Buscher.P. Treatment of human African trypanosomiasis-present situation and needs for research and development. Lancet Infect Dis 2, 437-40 (2002).
Muyldermans, S. Single domain camel antibodies: current status. J Biotechnol 74, 277-302 (2001 ).
Naessens, J. Kitani, H., Nakamura.Y., Sekikawa, K. and Iraqi, F. TNF-alpha mediates the development of anaemia in a murine Trypanosoma brucei rhodesiense infection, but not the anaemia associated with a murine Trypanosoma congolense infection. Clin Exp Immunol 139, 405-10 (2005).
Nguyen, V.K., Desmyter, A. and Muyldermans, S. Functional heavy-chain antibodies in Camelidae. Adv Immunol 79, 261-96 (2001 ). Pays, E., Vanhamme, L. and Perez-Morga, D. Antigenic variation in Trypanosoma brucei: facts, challenges and mysteries. Curr Opin Microbiol 7, 369-74 (2004).
Pepin, J. and Milord, F. The treatment of human African trypanosomiasis. Adv Parasitol 33, 1- 47 (1994).
Perez-Morga, D. Vanhollebeke, B., Paturiaux-Hanocq, F., Nolan, D. P., Lins, L., Horrible, F., Vanhamme, L., Tebabi, P., Pays, A., Poelvoorde, P., Jacquet.A., Brasseur, R. and Pays, E. Apolipoprotein L-I promotes trypanosome lysis by forming pores in lysosomal membranes. Science 309, 469-472 (2005)
Roux, K.H., Strelets, L., Brekke, O. H., Sandlie, I. and Michaelsen, T. E. Comparisons of the ability of human lgG3 hinge mutants, IgM, IgE, and lgA2, to form small immune complexes: a role for flexibility and geometry. J Immunol 161 , 4083-90 (1998).
Stijlemans, B., Conrath, K., Cortez-Retamozo, V., Xong H.V., Wyns, L., Senter, P., Revets, H., De Baetselier, P, Muyldermans, S and Magez, S. Efficient targeting of conserved cryptic epitopes of infectious agents by single domain antibodies. African trypanosomes as paradigm. J Biol Chem 279, 1256-61 (2004). Vanhamme, L., Paturiaux-Hanocq, F., Poelvoorde, P., Nolan, D. P., Lins, L., Van Den Abbeele, J., Pays, A., Tebabi, P., Xong, H.V., Jacquet, A., Moguilevsky, N., Dieu, M., Kane, J. P., De Baetselier, P, Brasseur, R and Pays, E. Apolipoprotein L-I is the trypanosome lytic factor of human serum. Nature 422, 83-7 (2003).
Vanhamme, L. and Pays, E. The trypanosome lytic factor of human serum and the molecular basis of sleeping sickness. I nt J Parasitol 34, 887-98 (2004). van Meirvenne, N., Maginus, E. and Janssens, P. G. The effect of normal human serum on trypanosomes of distinct antigenic type (ETat 1 to 12) isolated from a strain of Trypanosoma brucei rhodesiense. Ann Soc BeIg Med Trop 56, 55-63 (1976).
Welburn, S. C, Picozzi, K., Fevre, E. M., Coleman, P. G., Odiit, M., Carrington, M and Maudlin, I. Identification of human-infective trypanosomes in animal reservoir of sleeping sickness in Uganda by means of serum-resistance-associated (SRA) gene. Lancet 358, 2017-9 (2001 a).
Welburn, S. C, Fevre, E. M., Coleman, P. G., Odiit, M. and Maudlin, I. Sleeping sickness: a tale of two diseases. Trends Parasitol 17, 19-24 (2001 b).
Xong, H.V., Vanhamme, L., Chamekh, M., Chimfwembe, C. E., Van Den Abbeele, J., Pays, A., Van Meirvenne, N., Hamers, R., De Baetselier, P and Pays, E. A VSG expression site- associated gene confers resistance to human serum in Trypanosoma rhodesiense. Cell 95, 839-46 (1998).