Camelid single-domain antibody directed against phosphorylated tau proteins and methods for producing conjugates thereof

Abstract

The present invention relates to variable domains of a camelid heavy-chain antibodies directed against a phosphorylated tau protein and conjugates thereof. The present invention also relates to the use of these domains or conjugates for treating or diagnosing disorders mediated by neurofibrillary tangles, neuropil threads or dystrophic neurites, such as tauopathies.

Claims

1. An in vitro or ex vivo method, comprising: a) contacting in vitro a biological sample from a subject with a diagnostic agent comprising an isolated variant of the variable domain of a camelid heavy-chain antibody (VHH) of SEQ ID NO. 5 linked, directly or indirectly, covalently or non-covalently to a substance of interest, wherein said VHH variant is directed against the phosphorylated serine 422 of a phosphorylated tau protein, and wherein the amino acid sequence of said variant has at least 95% identity with the amino acid sequence SEQ ID NO: 5, and b) determining the presence or the absence of phosphorylated-tau protein in said biological sample.

2. The method of claim 1, wherein the amino acid sequence of said variant comprises, from the N-terminus to the C-terminus, the amino acid sequence SEQ ID NO. 1, the amino acid sequence SEQ ID NO. 2 and the amino acid sequence SEQ ID NO. 3.

3. The method of claim 1, wherein said variant has the amino acid sequence SEQ ID NO. 5 having the following mutations: A) the Glutamine residue (Gln, Q) at position 3 of the amino acid sequence SEQ ID NO. 5 is substituted with an amino acid residue selected from the group consisting of Aspartic acid (Asp, D) and Glutamic acid (Glu, E), B) the Isoleucine residue (Ile, I) at position 52 of the amino acid sequence SEQ ID NO. 5 is substituted with an amino acid residue selected from the group consisting of Alanine (Ala, A), Glycine (Gly, G), and C) the Valine residue (Val, V) at position 86 of the amino acid sequence SEQ ID NO. 5 is substituted with an amino acid residue selected from the group consisting of Alanine (Ala, A), Serine (Ser, S), Threonine (Thr, T), Asparagine (Asn, N), Glutamine (Gin, Q), Aspartic acid (Asp, D), Glutamic acid (Glu, E), Lysine (Lys, K), Arginine (Arg, R) and Glycine (Gly, G).

4. The method of claim 1, wherein said variant consists of the amino acid sequence of SEQ ID NO. 15 or of the amino acid sequence of SEQ ID NO. 16.

5. The method of claim 1, wherein the subject has a disorder selected from tauopathies, Alzheimer's disease (AD), Pick disease (PD), fronto-temporal dementia (FTD), corticobasal degeneration (CBD) and progressive supranuclear palsy (PSP).

6. The method of claim 1, wherein the presence of said phosphorylated-tau protein indicates that said subject has a disorder selected from tauopathies, Alzheimer's disease (AD), Pick disease (PD), fronto-temporal dementia (FTD), corticobasal degeneration (CBD) and progressive supranuclear palsy (PSP).

7. The method of claim 1, wherein the substance of interest is selected from the group consisting of an enzyme, a fluorophore, a Nuclear Magnetic Resonance (NMR) or Magnetic Resonance Imaging (MRI) contrast agent, a radioisotope, and a nanoparticle.

8. An in vitro or ex vivo method, comprising: a) providing a biological sample from a subject having a disorder selected from tauopathies, Alzheimer's disease (AD), Pick disease (PD), fronto-temporal dementia (FTD), corticobasal degeneration (CBD) and progressive supranuclear palsy (PSP), b) contacting in vitro the biological sample with a diagnostic agent comprising an isolated variant of the variable domain of a camelid heavy-chain antibody (VHH) of SEQ ID NO. 5 linked, directly or indirectly, covalently or non-covalently to a substance of interest, wherein said VHH variant is directed against the phosphorylated serine 422 of a phosphorylated tau protein, and wherein the amino acid sequence of said variant has at least 95% identity with the amino acid sequence SEQ ID NO: 5, and c) measuring the amount of phosphorylated-tau protein in said biological sample, and d) comparing the amount measured in step c) with an amount of phosphorylated-tau protein previously measured in said subject.

9. The method of claim 8, wherein the amino acid sequence of said variant comprises, from the N-terminus to the C-terminus, the amino acid sequence SEQ ID NO. 1, the amino acid sequence SEQ ID NO. 2 and the amino acid sequence SEQ ID NO. 3.

10. The method of claim 8, wherein said variant has the amino acid sequence SEQ ID NO. 5 having the following mutations: A) the Glutamine residue (Gln, Q) at position 3 of the amino acid sequence SEQ ID NO. 5 is substituted with an amino acid residue selected from the group consisting of Aspartic acid (Asp, D) and Glutamic acid (Glu, E), B) the Isoleucine residue (Ile, I) at position 52 of the amino acid sequence SEQ ID NO. 5 is substituted with an amino acid residue selected from the group consisting of Alanine (Ala, A), Glycine (Gly, G), and C) the Valine residue (Val, V) at position 86 of the amino acid sequence SEQ ID NO. 5 is substituted with an amino acid residue selected from the group consisting of Alanine (Ala, A), Serine (Ser, S), Threonine (Thr, T), Asparagine (Asn, N), Glutamine (Gin, Q), Aspartic acid (Asp, D), Glutamic acid (Glu, E), Lysine (Lys, K), Arginine (Arg, R) and Glycine (Gly, G).

11. The method of claim 8, wherein said variant consists of the amino acid sequence of SEQ ID NO. 15 or of the amino acid sequence of SEQ ID NO. 16.

12. The method of claim 8, wherein a significant increase in the amount of phosphorylated-tau protein is measured in step c) compared to the amount in a previous sample, and wherein the subject has progression of said disorder.

13. The method of claim 8, wherein a significant decrease in the amount of phosphorylated-tau protein is measured in step c) compared to the amount in a previous sample, and wherein the subject has regression of said disorder.

14. The method of claim 8, wherein the substance of interest is selected from the group consisting of an enzyme, a fluorophore, a Nuclear Magnetic Resonance (NMR) or Magnetic Resonance Imaging (MRI) contrast agent, a radioisotope, and a nanoparticle.

15. A method for in vivo imaging neurofibrillary tangles, neuropil threads or dystrophic neurites in a subject, comprising: a) administering to a subject a detectable quantity of a diagnostic agent comprising an isolated variant of the variable domain of a camelid heavy-chain antibody (VHH) of SEQ ID NO. 5 linked, directly or indirectly, covalently or non-covalently to a substance of interest, wherein said VHH variant is directed against the phosphorylated serine 422 of a phosphorylated tau protein, and wherein the amino acid sequence of said variant has at least 95% identity with the amino acid sequence SEQ ID NO: 5, and b) detecting the diagnostic agent in said subject by an imaging method.

16. The method of claim 15, wherein the amino acid sequence of said variant comprises, from the N-terminus to the C-terminus, the amino acid sequence SEQ ID NO. 1, the amino acid sequence SEQ ID NO. 2 and the amino acid sequence SEQ ID NO. 3.

17. The method of claim 15, wherein said variant has the amino acid sequence SEQ ID NO. 5 having the following mutations: A) the Glutamine residue (Gln, Q) at position 3 of the amino acid sequence SEQ ID NO. 5 is substituted with an amino acid residue selected from the group consisting of Aspartic acid (Asp, D) and Glutamic acid (Glu, E), B) the Isoleucine residue (Ile, I) at position 52 of the amino acid sequence SEQ ID NO. 5 is substituted with an amino acid residue selected from the group consisting of Alanine (Ala, A), Glycine (Gly, G), and C) the Valine residue (Val, V) at position 86 of the amino acid sequence SEQ ID NO. 5 is substituted with an amino acid residue selected from the group consisting of Alanine (Ala, A), Serine (Ser, S), Threonine (Thr, T), Asparagine (Asn, N), Glutamine (Gin, Q), Aspartic acid (Asp, D), Glutamic acid (Glu, E), Lysine (Lys, K), Arginine (Arg, R) and Glycine (Gly, G).

18. The method of claim 15, wherein said variant consists of the amino acid sequence of SEQ ID NO. 15 or of the amino acid sequence of SEQ ID NO. 16.

19. The method of claim 15, wherein the substance of interest is selected from the group consisting of an enzyme, a fluorophore, a Nuclear Magnetic Resonance (NMR) or Magnetic Resonance Imaging (MRI) contrast agent, a radioisotope, and a nanoparticle.

20. The method of claim 15, wherein the subject has a disorder selected from tauopathies, Alzheimer's disease (AD), Pick disease (PD), fronto-temporal dementia (FTD), corticobasal degeneration (CBD) and progressive supranuclear palsy (PSP).

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 shows the binding of Inti and Rascar sera against phospho-tau and tau proteins. The alpaca polyclonal antibodies were detected with rabbit anti-alpaca antibodies.

(2) FIG. 2 shows the binding of purified VHH Tau-A2 to phospho-tau, tau (unphosphorylated) and Ova-pS422 peptide.

(3) FIGS. 3A to 3B show the immunohistochemical staining of NFTs using the VHH Tau-A2 and mAb AT8 on human paraffin sections from cases with various tauopathies: AD (FIG. 3A), FTD (FIG. 3B), PSP (FIG. 3B) and PD (FIG. 3C).

(4) FIG. 4 shows the immunohistochemical staining of NFTs using the VHH Tau-A2 and mAb AT8 on Tg4510 mouse free-floating sections.

(5) FIG. 5 shows a western blot on AD brain extracts and phospho-tau (p-tau) revealed by VHH Tau-A2 and Tau-pS422 mAb. Lane 1 and 3: p-tau+Tau-A2 and Tau-pS422 mAb, respectively. Lane 2 and 4: AD extracts+Tau-A2 and Tau-pS422 mAb, respectively.

(6) FIG. 6 shows the solid-phase synthesis of maleimido-(DOTA/Gd).sub.3 (compound 1).

(7) FIGS. 7A to 7F show the biochemical and histological characterization of Tau-A2 and its derivatives. A: Protein profile analysis of VHH Tau-A2-SH (lane 1) and Tau-A2-S-AF488 (lane 2) by instantBlue stained SDS-PAGE gel. B: Determination of pI of Tau-A2 derivatives by NEPHGE on 3-10 IEF gel. C: In vitro imaging of NFTs by Tau-A2-S-AF488 on Tg4510 paraffin sections. The presence of VHH was revealed by the addition of a rabbit polyclonal antibodies directed against VHH. D: Negative control: IHC on TauPS2APP VHH free paraffin section using rabbit anti-VHH Tau-A2 only. E: Direct fluorescent staining of NFTs by Tau-A2-S-AF488 in the cortex on Tg4510 free floating sections. F: Direct fluorescent staining of NFTs by Tau-A2-S-AF488 in the hippocampus on Tg4510 free floating sections.

(8) FIG. 8 shows the in vivo imaging of NFTs after cerebral injection of Tau-A2-S-AF488 in a 8-month-old Tg4510 mouse. Arrows indicate labeling of NFTs. Scale bar=50 μm.

(9) FIGS. 9A to 9B show the in vivo two-photon imaging of Tau-A2-S-AF488 diffusing from blood vessels and labeling NFTs. A: Maximum intensity projection of fluorescence in a cortical area of a Tg4510 mouse 1 min, 120 min, 180 min after intravenous injection of Tau-A2-S-AF488 at 10 mg/kg. Whereas only faint non-specific signal could be detected before iv injection, a strong staining of arborescent vessels was observed few second after iv injection (T1). Specific labeling of NFTs (white arrows) was observed after 120 min, with a maximum intensity after 180 min (3D reconstruction (MIP)). Experiments were performed on 2 mice. B: Four hours after IV injection of Tau-A2-S-AF488, the brain was harvested and IHC was performed on 5 μm-paraffin sections. Rabbit polyclonal anti-VHH antibodies were used to reveal the presence and the labeling of NFTs by Tau-A2-S-AF488.

(10) FIG. 10 shows the optimization of the amino acid sequence of Tau-A2-SH for expression. Tau-A2-SH of SEQ ID NO. 14. Tau-A2var-SH of SEQ ID NO. 17.

(11) FIGS. 11A to 11D show the biochemical and histological characteristics of Tau-A2 variant. A: Protein profile analysis of VHH Tau-A2-SH (left) and Tau-A2var-SH (right) by instant Blue stained SDS-PAGE gel. B: Determination of pI of Tau-A2-SH (lane 1) and Tau-A2var-SH (lane 2) by NEPHGE on 3-10 IEF gel. C: ELISA Binding of VHH Tau-A2-SH and VHH Tau-A2var-SH on coated Phospho-Tau protein. A negative control was performed by using an irrelevant VHH. The VHH binding was revealed by the addition of a mouse anti-His tag mAb revealed by an anti-mouse antibody labelled with peroxydase D: Comparison of VHH Tau-A2-SH and VHH Tau-A2var-SH by IHC on Tg4510 mice. VHH were used at the concentration of 2 μg/ml and their presence was revealed by anti-His mouse mAb followed by a peroxydase labelled anti-mouse-Ig.

(12) FIG. 12 shows the in vivo imaging of NFTs after intravenous injection of Tau-A2var-S-(DOTA/Gd).sub.3 in a 8-month-old Tg4510 mouse. Four hours after IV injection of Tau-A2var-S-(DOTA/Gd).sub.3, the brain was harvested and IHC was performed on 5 μm-paraffin sections. Rabbit polyclonal anti-VHH antibodies were used to reveal the presence and the labeling of NFTs by Tau-A2var-S-(DOTA/Gd).sub.3.

EXAMPLE I

Generation of Anti-Phosphorylated Tau VHHS Coupled to Alexa Fluor®488 or Gadolinium Contrast Agent and Their Evaluation In Vitro/In Vivo

(13) Materials and Methods

(14) 1. Production, Selection and Purification of Anti-Tau Specific VHH (Tau-A2)

(15) 1.1 Antigen Preparation and Induction of a Humoral Immune Response in Alpaca

(16) Subjects

(17) Human cortical brain tissues from AD patients (Braak stage V and VI) were obtained from the NeuroCEB brain bank. This bank is associated to a brain donation program run by a consortium of patients associations (including France Alzheimer Association) and declared to the French Ministry of Research and Universities, as requested by French Law. An explicit written consent was obtained for the brain donation in accordance with the French Bioethical Laws.

(18) Tissue extraction was performed according to Mercken et al. (1992 Acta Neuropathol., 84, 265-272). Cortex from AD brain (0.2 g) was homogenized in 10 volumes of 10 mM Tris, 1 mM EGTA, 0.8 M NaCl pH7.4 containing 10% sucrose and was centrifuged at 27,000×g for 20 min at 4° C. The pellet was removed and the supernatant was adjusted to 1% N-laurylsarcosine and 1% beta-mercaptoethanol and incubated while rotating for 2.5 hours at 37° C. The supernatant mixture was centrifuged at 100,000 g for 35 min at 20° C. The PHF containing pellet was gently washed with PBS and finally resuspend in the same buffer.

(19) One alpaca (Inti) was immunized with tau pellet and an other alpaca (Rascar) was immunized with the single phospho-peptide derived from the C-terminus of a tau protein of sequence CSIDMVDS(PO.sub.3H.sub.2)PQLATLAD (SEQ ID NO. 6), coupled to KLH (Eurogentec).

(20) 250 μl (500 μg) of both antigens was mixed with 250 μl of Freund complete adjuvant for the first immunization, and with 250 μl of Freund incomplete adjuvant for the following immunizations. After three immunizations at day 0, 21 and 40, a serum sample was taken at day 52 and the immune response monitored by ELISA using recombinant phospho-tau protein or recombinant non-phosphorylated tau protein.

(21) 1.2 Library Construction and Panning

(22) 250 ml of blood of the immunized animals was collected at day 52 and the peripheral blood lymphocytes isolated by centrifugation on a Ficoll (Pharmacia) discontinuous gradient and stored at −80° C. until further use. Total RNA and cDNA was obtained as previously described in Lafaye P. et al. (1995 Res Immunol., 146, 373-382), and DNA fragments encoding VHH domains amplified by PCR using CH2FORTA4 and VHBACKA6 primers, which anneal to the 3′ and 5′ flanking region of the VH genes, respectively. The amplified product was used as template in a second round of PCR using either of the primers VHBACKA4 and VHFOR36. The primers were complementary to the 5′ and 3′ ends of the amplified product and incorporated SfiI and NotI restriction sites at the ends of the VHH genes. The PCR products were digested and ligated into phage expression vector pHEN1. The resulting library was composed of two sub-libraries, one derived from tau pellet and the other from phospho peptide coupled to KLH. Phages were produced and isolated using both sub-libraries, and subsequently pooled.

(23) The library (>6×10.sup.8 clones) was panned against full-length phospho-tau protein. Nunc Immunotubes (Maxisorp) tubes were coated overnight at 4° C. with the antigen (10 μg/ml) in PBS. Phages (10.sup.12 transducing unit) were panned by incubation with the coated tubes for 1 h at 37° C. with gentle agitation. A different blocking agent was used at each of the three rounds of panning: 2% skimmed milk, Licor blocking buffer (Biosciences) diluted 1:4, and 4% BSA were respectively used. Phage clones were screened by standard ELISA procedures using a HRP/anti-M13 monoclonal antibody conjugate (GE Healthcare) for detection (see below). The screening was performed in parallel with phospho-tau protein and tau protein.

(24) 1.3 Expression of VHHs

(25) The coding sequence of the selected nanobodies in vector pHEN1 was sub-cloned into a bacterial expression vector pET23d containing a 6-Histidine tag using NcoI and NotI restriction sites. Transformed E. coli BL21 (DE3) pLysS cells express VHH in the cytoplasm after overnight induction with IPTG (0.5 mM) at 16° C. Purified VHHs were isolated by IMAC from cytoplasmic extracts using a HiTrap crude column charged with Ni.sup.2+ (GE Healthcare), according to the manufacturer's instructions. The VHHs were eluted in 50 mM sodium phosphate buffer, 300 mM NaCl and 500 mM imidazole buffer and dialyzed in PBS buffer containing 300 mM NaCl (PBS/NaCl).

(26) The coding sequence of selected VHHs was also sub-cloned into a modified pASK IBA2 expression vector containing a Strep-tag at C-ter end (Skerra and Schmidt 1999 Biomol Eng., 16, 79-86). Same restriction sites were used. E. coli XL2-Blue Ultracompetent cells were transformed to express VHHs in the periplasm after overnight induction with anhydrotetracycline (AHT, 200 μg/l) at 16° C. VHHs were then purified from bacterial extract with Strep-Tactin column (IBA) according to manufacturer's instructions.

(27) 2. VHH Tau-A2 Coupling to MRI Contrast Agents and to Fluorophores

(28) 2.1 General Synthesis Methods

(29) Unless otherwise specified, the amino-acid derivatives and the reagents were purchased from Novabiochem and Sigma-Aldrich, respectively. The concentration of the peptide and VHH solutions (net protein content) was determined by quantitative amino acid analysis (AAA) using a Beckman 6300 analyzer after hydrolysis of the compounds with 6N HCl at 110° C. for 20 hours. The RP-HPLC/MS analyses were performed on an Alliance 2695 system coupled to a UV detector 2487 (220 nm) and to a Q-Tofmicro™ spectrometer (Micromass) with an electrospray ionisation (positive mode) source (Waters). The samples were cooled to 4° C. on the autosampler. The linear gradient was performed with acetonitrile+0.025% formic acid (A)/water+0.04% TFA+0.05% formic acid (B) over 10 or 20 min. The column used was a XBridge™ BEH300 C18 (3.5 μm, 2.1×100 mm) (Waters) (gradient 10-100% A). The source temperature was maintained at 120° C. and the desolvation temperature at 400° C. The cone voltage was 40 V. The samples were injected at 0.4-1 mg/ml concentration in their respective buffer added with B.

(30) A maleimido-(DOTA/Gd).sub.3 compound 1 was prepared by solid-phase peptide synthesis using 9-fluorenylmethoxycarbonyl (Fmoc) chemistry as shown in FIG. 6.

(31) 2.2 Production of Tau-A2-SH

(32) The coding sequence of a Cys-engineered VHH (Tau-A2-SH) was cloned into a bacterial expression vector pET23d using NcoI and XhoI restriction sites. To summarize, Tau-A2-SH (SEQ ID NO. 14) comprises from the N to the C terminus a 6-Histidine tag, a thrombin cleavage site, VHH Tau-A2 sequence followed by a G3 S spacer and three extra amino acids CSA. Transformed E. coli BL21 (DE3) pLysS cells express Tau-A2-SH in the cytoplasm after overnight induction with IPTG (0.5 mM) at 16° C. Purified VHHs were isolated by IMAC from cytoplasmic extracts using a HiTrap crude column charged with Ni2+ (GE Healthcare), according to manufacturer's instructions. The protein was eluted in PBS/NaCl containing 500 mM imidazole and then dialyzed in PBS/NaCl.

(33) AAA: Ala 14.3 (14), Arg 9.6 (10), Asp+Asn 7.3 (6), Glu+Gln 11.2 (10), Gly 18.4 (19), His 5.7 (6), Ile 4.2 (4), Leu 8.6 (8), Lys 5.3 (5), Phe 4 (4), Pro 2.9 (2), Ser 18.9 (23), Thr 10.5 (12), Tyr 6.1 (7), Val 11.5 (11).

(34) MS: 15499.286 (C669H1047N205O213S4 calculated 15498.190). The MS corresponds to the protein with N-ter deleted methionine.

(35) 2.3 Synthesis of Tau-A2-S-Alexa Fluor® 488 (Tau-A2-S-AF488)

(36) The cysteine present in the C-ter tripeptide of Tau-A2-SH was used to couple the Tau-A2-SH to the maleimido Alexa Fluor® 488 fluorophore (Invitrogen). pH of the solution containing Tau-A2-SH was adjusted between 6.8 and 7. The solution was then gently stirred with 10-fold molar excess of tris(2-carboxyethyl) phosphine (TCEP) at room temperature for 30 min, to allow complete reduction of any intermolecular disulfide bond. A 10-fold molar excess of maleimido Alexa Fluor® 488 fluorophore dissolved in dimethylformamide (DMF) was added. Notably, for high efficiency of conjugation, the volume percentage of DMF in the final solution was kept below 5%. The conjugation was performed for 2 hours at room temperature, under protection from light. The non-conjugated maleimido fluorophore was then removed by successive dialysis with PBS containing 300 mM NaCl (overall concentration).

(37) 2.4 Synthesis of Tau-A2-S-(DOTA/Gd).sub.3

(38) Prior to the conjugation, Tau-A2-SH (4.2 ml, 0.32 mg/ml in PBS/NaCl pH 6.8) was treated with TCEP (24.6 μg, 5 eq) for 30 min to prevent the dimerization of the VHH. Maleimido-(DOTA/Gd).sub.3 (0.78 mg, 3.8 eq relative to 1 thiol group per VHH) in aqueous solution (78 μl) was added to the protein and the solution was stirred at 4° C. for 3 h. The solution was then dialyzed in PBS/NaCl using Slide-A-Lyzer cassettes (Thermo Scientific) (3,500 MWCO). Aliquots (20 μl) of Tau-A2-SH and Tau-A2-S-(DOTA/Gd).sub.3 were diluted with buffer B (20 μl) for RP-HPLC/MS analyses. Further, aliquots (10 μl) of the same compounds were diluted in 20 mM Tris buffer pH 7.3 (90 μl) for ELISA analyses. 4.2 ml of Tau-A2-S-(DOTA/Gd).sub.3 (0.24 mg/ml) was obtained with a yield of 65%. It was calculated by dividing the actual amount of the final product Tau-A2-S-(DOTA/Gd).sub.3 by its expected amount (net protein contents). For further experiments, the solution of Tau-A2-S-(DOTA/Gd).sub.3 was concentrated four times using Vivaspin 2 centrifugal filter device (3,000 MWCO PES).

(39) AAA: Ala 13.4 (14), Arg 9.6 (10), Asp+Asn 8.0 (6), Glu+Gln 11.8 (10), Gly 24.4 (22), His* (6), Ile 4.2 (4), Leu 8.6 (8), Lys 18.9* (8), Phe 4 (4), Pro 2.9 (2), Ser 16.8 (23), Thr 10.7 (12), Tyr 6.1 (7), Val 11.4 (11). [*His cannot be determined due to co-elution with ammonium. Lys is overestimated due to co-elution with maleimido derivative in the conditions of the analysis.].

(40) MS: 17887.549 (C751H1174N227O244S4Gd3 calculated 17886.980).

(41) 3. In Vitro Characterization of Tau-A2, and Tau-A2 Conjugates by Immunohistochemistry and Biochemistry

(42) 3.1 Subjects

(43) Human brain tissue was obtained from the NeuroCEB brain bank. Preclinical experiments were performed on Tg4510 (Santacruz et al. 2005 Science, 309, 476-81) transgenic mice. Animal experimental procedures were performed in strict accordance with the ethical standards of French and European laws (European Communities Council Directive 2010/63/EU on the protection of animals used for scientific purposes) and after approval from local Animal Care and Use committee. The animals were sacrificed using a high dose of sodium pentobarbital (100 mg/kg) and then perfusion-fixed with 10% buffered formalin. Their brains were then removed, immersed in formalin for at least 24 hours and stored at 4° C.

(44) 3.2 Tissue Extracts

(45) Tissue extraction was performed according to Mercken et al. (1992 Acta Neuropathol., 84, 265-272).

(46) 3.3 Immunoblots

(47) Phospho-tau protein was resuspended in NuPAGE® LDS sample buffer (Invitrogen). Brain extracts were resuspended in NuPAGE® LDS sample buffer (Invitrogen) containing 8 M urea. Following separation by polyacrylamide gel electrophoresis (PAGE) using NuPAGE Novex 4-12% Bis-tris gel (Invitrogen), semi-dry transfer onto Hybond-C (Amersham) and western blotting were carried out using the Xcell II blot module (Invitrogen). Prior to the immunochemical reaction, membranes were blocked in a 4% skimmed milk solution. Immunoblotting of membranes was accomplished with VHH (with His or Strep tag) or anti p-tau 422 mAb (Grueninger et al. 2011 Mol Cell Biochem., 357, 199-207) and revealed by rabbit anti-His tag (eBioscience) polyclonal antibodies followed by peroxidase labeled goat anti-rabbit immunoglobulins (Abcam) or by an anti-Strep tag monoclonal antibody (such as the antibody C23-21 produced by the hybridoma filed with the CNCM under the number 1-4703) followed by peroxidase labeled rabbit anti-mouse immunoglobulins (Bio-rad). Finally, peroxidase activity was visualized using a chemiluminescent kit (GE Healthcare).

(48) 3.4 ELISA

(49) Microtiter plates (Nunc, Denmark) were coated by incubation overnight at 4° C. with 1 μg/ml of phospho-tau or tau protein or the phospho-peptide of SEQ ID NO. 6 coupled to ovalbumine protein diluted in PBS. Plates were washed with buffer 0.1% Tween 20 in PBS. Tau-A2 (with His or Strep tag) was diluted in buffer 0.5% gelatin 0.1% Tween 20 in PBS. After 2 h incubation at 37° C., plates were washed again before adding respectively a rabbit anti-His tag polyclonal antibody (eBiosciences), followed by peroxidase labeled goat anti-rabbit immunoglobulins (Abcam) or by an anti-Strep tag monoclonal antibody (such as the antibody C23-21) followed by peroxidase labeled rabbit anti-mouse immunoglobulins (Bio-rad), and finally revealed by OPD (o-phenylendiamine dihydrochloride, Dako) according to manufacturer's protocol.

(50) 3.5 Sequences Analysis

(51) VHH encoded DNAs were sequenced by GATC Biotech and sequences were treated with Serial Cloner.

(52) 3.6 Determination of pI

(53) The pI of VHHs was determined by isoelectric focusing using IEF 2-9 Gel (Invitrogen). NEPGHE (non equilibrium pH gradient gel electrophoresis) with sample application at the anode because it allows optimal protein analysis in the basic range of the gel including pH 8.5 to 10.5. The protocol was detailed in SERVAGel IEF 3-10 instruction manual.

(54) 3.7 Immunohistochemistry and Immunofluorescence

(55) Immunohistochemistry was performed on formalin-fixed tissues (paraffin-embedded or frozen sections or vibratome sections). Standard IHC protocols were applied and adapted for each tissue conditions. As most of immunostaining experiments were performed using paraffin sections, a detailed protocol for paraffin-embedded material is described herein. Immunostaining of brain tissue was performed on 4 μm thick paraffin sections. Both human and mouse tissues were used (Human patients with AD or other tauopathies and Tg4510 mice (Santacruz et al. 2005 Science, 309, 476-481). Sections were de-paraffinized in xylene, rehydrated through ethanol (100%, 90%, and 70%), 5 min for each solution and finally brought to running tap water for 10 min. They were then incubated in 98% formic acid for 5 min, washed again under running tap water, quenched for endogenous peroxidase with 3% hydrogen peroxide and 20% methanol, and finally washed in water. Non-specific binding was blocked by incubating the sections for 30 min in 2% bovine serum albumin in TBS+0.5% Tween. Appropriate dilutions of primary antibodies (1-10 μg/ml of VHH with either His or Strep tag) were then applied and slices incubated overnight in a humidified chamber at 4° C. Slides were washed with TBS-Tween and incubated with secondary antibodies rabbit anti-His Tag for 1/1000 or home-made biotinylated anti-strep mAb C23-21 in TBS-Tween at room temperature for 1 hour. Slides were then incubated with reagents of Dako REALTM Detection System, Peroxidase/DAB+ according to manufacturer's instructions. Chromogenic (DAB) revelation was developed until a good signal-to-noise ratio was obtained (about 5 min). After washing with TBS-Tween, slides were counter-stained with hematoxylin. For labeling of NFTs, biotinylated mAb AT8 (ThermoScientific) was used as a positive control in parallel.

(56) Immunofluorescent staining of NFTs using Tau-A2-S-AF488 was performed on 40 μm thick free floating sections obtained from Tg4510 mice using vibratome (Leica VT1000S). After 3×5 min washing in PBS, sections were blocked by incubating for 15 min with PBS-triton 0.2% containing 2% of BSA, which was then replaced by 1 μg/ml of Tau-A2-S-AF488 in PBS-triton 0.2% and incubated overnight at 4° C. Sections were finally washed with PBS and mounted with an aqueous mounting medium (Mowiol).

(57) 4. In Vivo Evaluation of Tau-A2 and Tau-A2 Conjugates by Two-Photon Imaging and Correlative Immunohistochemistry

(58) 4.1 Subjects

(59) In vivo evaluation of Tau-A2, and Tau-A2-S-(DOTA/Gd).sub.3 was performed on Tg4510 transgenic mice.

(60) 4.2 Stereotaxic Injection of VHH and IHC

(61) Stereotaxic injections were performed in Tg4510 female transgenic mice (n=2) anesthetized mice with 2 μl of VHH per injection at the rate of 0.5 μl/min. The mice were anesthetized with a mixture of isoflurane (1-2%) and air (1 l/min). They were placed on a stereotaxic frame and the skull was bilaterally perforated with a Dremel. Blunt Hamilton syringes were used to inject MRI contrast agent. Each mouse received 4 injections, in the frontal cortex and the hippocampus in each hemisphere. The stereotaxic coordinates in the frontal cortex were +0.86 mm anterior from bregma, ±1.5 mm lateral from the midline, −0.65 mm ventral from dura. The stereotaxic coordinates in the hippocampus were −2.18 mm posterior from bregma, ±1.5 mm lateral from the midline, −1.8 mm ventral from dura. Two or 24 hours after the injection, mice were euthanized and perfused intracardially with 4% paraformaldehyde in PBS (pH 7.6). Brains were removed and post-fixed in the same fixative overnight at 4° C. 4 μm thick paraffin sections were prepared. The presence of the VHH in cerebral tissue was detected using immunohistochemical procedures described above.

(62) 4.3. Two Photon Microscopy in Tg4510 Mice Using Tau-A2-S-AF488

(63) 1) Craniotomy

(64) Two 8-month-old Tg4510 mice were anesthetized by inhalation of isoflurane (1% vol/vol in pure O.sub.2) and placed onto a warming blanket (37° C.). A stereotaxic frame was used to identify the location of the motor cortex. 50 μl of 2% lidocaine was injected subcutaneously for local anesthesia at the incision site location where the skin was to be removed. A less than 2 mm craniotomy was performed using a scalpel. In the case of intracerebral injections of VHH Tau-A2-S-AF488, the dura mater was incised. In the case of intravenous injections, the dura mater was kept intact. To avoid movement artifacts, the skull opening was covered with 2% low melting point agarose and a coverglass. The optical window was secured and sealed to skull with dental cement, covering all the exposed skull, wound margins and coverglass edges.

(65) 2) Administration of Tau-A2-S-AF488

(66) Intracerebral injection: 1.2 μg (1.5 μL) of Tau-A2-S-AF488 was injected in the brain of one Tg4510 mouse at 1.5 mm depth from cortical surface. Two-photon imaging was then performed during 4 h after injection.

(67) Intravenous injection: 270 μg (150 μl) of Tau-A2-S-AF488 was slowly injected into the caudal vein of one Tg4510 mouse. Two-photon imaging was then performed in the following 3 hours.

(68) 3) Two-Photon Imaging

(69) Two-photon imaging was performed with a two-photon laser-scanning microscope system and PrairieView software (Prairie Technologies, Middelton, Wis., USA), using a 16×0.9 NA water immersion objective (Nikon, Tokyo, Japan) with the 2-photon laser tuned to 920 nm (MaiTai DeepSee, Spectra Physics, Mountain View, Calif., USA). The images were acquired at 512×512 with a pixel size of 0.5 μm. Care was taken to use less than 20 mW of laser power in the tissue.

(70) 5. Obtention of Rabbit Polyclonal Anti-VHH Antibodies

(71) Purified alpacas immunoglobulins were used to immunize one rabbit. Rabbit polyclonal antibodies against VHH were purified in 2 steps. First polyclonal antibodies were isolated by immunochromatography using protein A-sepharose 4B beads. Then another immunochromatography was realized by using sepharose 4B beads labelled with Tau-A2var-SH. The sepharose labelling was realized according to manufacturer's instructions (GE). The resulting purified antibodies represented less than 1% of total rabbit antibodies and are referred to as rabbit polyclonal anti-VHH antibodies. These antibodies bind to Tau-A2 VHH.

(72) Results

(73) 1. Polyclonal Response, Library Construction, and Selection of Specific Anti-Tau VHH

(74) After 3 immunizations, the sera of Inti (brain extract) and Rascar (tau-pS422) were collected and their binding to phospho-tau and non phosphorylated tau was analyzed by ELISA (FIG. 1). The sera of Inti and Rascar recognized both phosphorylated tau and non phosphorylated tau, although the immune response of Inti appeared to be better than that of Rascar. However, the former did not distinguish significantly phospho-tau and non phospho-tau and the latter showed a higher immune response for phospho-tau.

(75) In order to ensure that one or some of anti phospho-tau antibodies present in the sera could recognize tau lesions in fixed mouse tissues, IHC was performed on both frozen microtome floating sections and non-frozen vibratome floating sections from TauPS2APP mice (Grueninger F. et al., 2010, Neurobiol Dis., 37:294-306). The serum of Inti did not show any immune staining of tau lesions but otherwise the serum of Rascar demonstrated a specific tau staining of NFTs on both microtome and vibratome sections (data not shown).

(76) Despite the absence of immunodetection for NFTs in IHC with Inti polyclonal serum, it can not be excluded the existence of phospho-tau specific antibodies in its blood plasma. Indeed, the lack of IHC signal could be due to the low frequency of anti phospho-tau antibodies, which could recognize the epitopes of phospho-tau present in NFTs of TauPS2APP mice. The two libraries obtained from these two animals were pooled for panning experiments.

(77) Total RNA from peripheral blood lymphocytes was used as template for cDNA synthesis. Using this cDNA, the VHH encoding sequences were then amplified by PCR and cloned into vector pHEN1. Subsequent transformations yielded two libraries of about 3×10.sup.8 clones each. Both libraries were pooled and VHHs displaying the best affinity were selected by phage display through 3 panning cycles with phosphorylated tau. 96 individual clones were tested by ELISA on phosphorylated tau and on tau proteins. Only one clone was found to bind specifically on phosphorylated tau (VHH Tau-A2).

(78) This VHH was subcloned in vector pET23 or in vector pASK IBA2 to allow a high level of expression of VHH with, respectively, a His-tag or a Streptavidin-tag. Yields of <1 mg/l of bacterial culture were obtained. The single domain products were shown to be pure to homogeneity by SDS-PAGE and by RP-HPLC/MS (data not shown); its pI values was above 9.5. Dynamic light scattering experiments showed that Tau-A2 is monomeric (RH=4.5±0.3 nm) and is not aggregated after purification. Amino-acid sequence of VHH Tau-A2 is referred to as SEQ ID NO. 4.

(79) 2. Recognition of NFTs by VHH Tau-A2

(80) 2.1 VHH Tau A2 Recognizes the Phosphorylated Serine 422 in the C-Ter Tau Peptide

(81) Several serine and threonine are phosphorylated on the pathogenic form of tau. To evaluate the role of phosphorylation in the epitope recognition, ELISA was performed on phosphorylated tau and single phospho-peptide derived from the C-terminus of a tau protein (sequence CSIDMVDS(PO.sub.3H.sub.2)PQLATLAD; SEQ ID NO. 6), coupled to ovalbumin (Ova) protein. Phosphorylated tau was the full-length tau protein phosphorylated at multiple sites including S422. VHH Tau-A2 binds both compounds suggesting that VHH Tau-A2 recognized an epitope including the phospho serine 422 (pS422) (FIG. 2).

(82) 2.2 Immunoreactivity of VHH Tau A2 for NFTs

(83) The distribution of VHH-specific immunoreactivity was examined in human AD brains, in tissues from other tauopathies (fronto-temporal dementia [FTD], progressive supranuclear palsy [PSP] and Pick's disease [PD]) and in transgenic Tg4510 mice brains.

(84) VHH Tau-A2 showed good ability to immunodetect NFTs in paraffin sections from AD patients (FIG. 3A). Also tau-positive glial inclusions were observed in the FTD case (oligodendroglial coiled bodies) and in the PSP case (astrocytic tufts) (FIG. 3B). Interestingly the typical cellular inclusions (Pick bodies) identified by the reference anti-tau AT8 mAb were negative for Tau-A2 (FIG. 3C) underlining some specificity of the VHH. As opposed to other tauopathies Pick bodies are only constituted of 3R tau proteins. Hence it can be hypothesized that Tau-A2 mainly recognizes tau with 4 repeats. No labeling was observed with tissues from wild type mice.

(85) Paralleling result on paraffin-embedded tissues, it was showed that NFT immunodetection using VHH Tau-A2 can be readily obtained on mouse free-floating vibratome sections (FIG. 4).

(86) To confirm the immunoreactivity of VHH Tau-A2 on brain tissues, western-blot immunoassays were performed on brain extracts obtained from AD patients. Tau-pS422 mAb (Grueninger et al. 2011 Mol Cell Biochem., 357, 199-207) was used as a reference antibody. Recombinant phospho-tau (p-tau) was loaded in parallel on SDS-PAGE gel. Phospho-tau protein was revealed by both VHH Tau-A2 and Tau-pS422 mAb. One main band, corresponding to phosphorylated tau between 70 and 100 kDa, was immunodetected with Tau-A2 and Tau-pS422 in AD extracts. With Tau-pS422 mAb, two additional bands with a molecular weight between 130 and 150 kDa were observed indicated the presence of aggregated tau-pS422 in AD brain, which was very slightly revealed by Tau-A2 (FIG. 5).

(87) 3. Cys-Engineered Tau-A2 and Antibody Coupling to Alexa Fluor® 488 Maleimide

(88) Cys-engineered Tau-A2 containing from the N to the C terminus a 6-Histidine tag, a thrombin cleavage site, VHH Tau-A2 sequence followed by a G3 S spacer and three extra amino acids CSA was cloned in vector pET23d to allow a high level of expression (referred to as Tau-A2-SH or A2-SH; SEQ ID NO. 14). Tau-A2-SH was conjugated to maleimido Alexa Fluor® 488 by thioaddition. The resulting product Tau-A2-S-AF488 was analyzed by SDS-PAGE, IEF/NEPGHE (FIGS. 7A and 7B) and RP-HPLC/MS. It was shown to be pure to homogeneity by SDS-PAGE and by RP-HPLC/MS. All analyses showed that Tau-A2-SH was totally converted into the well-defined conjugate, referred to as Tau-A2-S-AF488, with a single AF488 on the VHH Tau-A2. Tau-A2-SH possesses a pI around 10 (between 9.5 and 10.5) and this value slightly decreases after AF488 conjugation, still remaining>9.5 (see FIG. 7B). The hydrodynamic radius (R.sub.H) of Tau-A2-SH and Tau-A2-S-AF488 was separately measured by DLS. The size distribution over time showed an average R.sub.H of 2.66±0.0788 nm for Tau-A2-SH and an average R.sub.H of 3.576±0.225 nm for Tau-A2-S-AF488, which suggested that both of them were in monomeric form in solution.

(89) 4. Detection of NFTs with VHH Tau-A2-S-AF488

(90) 4.1 In Vitro Labeling on Brain Slices of Tg4510 Mouse

(91) After direct incubation with mouse brain sections, Tau-A2-S-AF488 showed good ability to detect NFTs in Tg4510 mice by standard IHC (FIG. 7C-D), and direct fluorescent staining (FIG. 7E-F).

(92) 4.2 Two Photon Imaging After Intracerebral Injection

(93) In vivo two-photon imaging of Tau-A2-S-AF488 was performed after direct intracerebral injection of 1.2 μg (1.5 μL) in a Tg4510 mouse brain following a craniotomy and perforation of the dura mater. A large amount of CSF was observed and a high auto-fluorescence (even before the injection) was observed. Despite these limitations for appropriate bi-photon imaging, specific staining of NFTs was detected after topic cortical injection (FIG. 8).

(94) 4.3 Two-Photon Imaging After Intravenous Injection of VHH Tau-A2-S-AF488

(95) A 10 mg/kg dose of Tau-A2-S-AF488 was injected in the tail vein (270 μg, 150 μL) of two Tg4510 mice. BBB integrity of these mice was previously checked by MRI and absence of signal modification in the mice suggested no disruption of BBB (data not shown). The conjugate extravasation and staining in the brain was recorded for 4 hours post injection using two-photon microscopy on brain window (z=from the surface up to 360 μm deep). FIG. 9A displayed in vivo imaging (Maximum Intensity Projection—MIP) of Tau-A2-S-AF488 over time up to 180 min in the same region. Few seconds after iv injection, strong staining of arborescent vessels was observed and declined dramatically 20 min later with only few capillary vessels remaining stained. This suggested a short half-life of conjugated VHH in the circulation (10-20 min). 120 min after iv injection, NFTs began to be visualized. The absence of signal in the red channel demonstrated that the fluorescent signal was specific (data not shown) and not due to general autofluorescence. VHH Tau-A2-S-AF488 extravasation and staining in the brain was registered for 3 hours post injection using two-photon microscopy on brain window (from the immediate cortical surface up to 350 μm deep). In addition to strong auto-fluorescence background and large amount of CSF (see above), the Tg4510 mice exhibited marked cerebral atrophy. A few seconds after iv injection, strong staining of densely-packed vessels was observed. Specific tau staining of NFTs was observed 2 h after injection despite remaining autofluorescence signal. A persistent and specific labeling of NFTs was observed even 3 hours after injection (FIG. 9A), suggesting a brain half-life of Tau-A2-S-AF488 extending over several hours.

(96) Four hours after the intravenous injection of Tau-A2-S-AF488, the brain was harvested and 5 μm-thick paraffin sections were prepared. IHC was then performed with rabbit polyclonal anti-VHH antibodies to confirm the diffusion and labeling of NFTs by Tau-A2-S-AF488 (FIG. 9B).

(97) 6. Control Experiments

(98) Evaluation in NFTs-Free Mouse

(99) Tau-A2-S-AF488 was intravenously injected in a wild type, NFT-free, C57BL/6 mouse. No specific in vivo staining in the brain parenchyma was observed using two-photon microscopy assay (data not shown).

(100) Comparison with Conventional IgG Antibody

(101) Injection of AF488-conjugated anti-Tau mAb (Grueninger et al. 2010 Neurobiol. Dis., 37, 294-306) iv in a Tg4510 mouse did not allow detection of NFTs indicating no significant extravasation of this standard anti-Tau-pS422 immunoglobulin.

(102) 7. Conclusion

(103) Using two-photon microscopy after iv injection, VHH Tau-A2-S-AF488 showed its ability to cross the BBB and to penetrate into neurons after crossing a second (plasma membrane) barrier to reach its cytoplasmic target. Long-term detection (3 h) of NFTs labeling suggested also a long half-life of this VHH in the brain.

(104) 8. Antibody Coupling to MRI Contrast Agent

(105) VHH Tau-A2-SH was labeled with gadolinium (MRI contrast agent) using 1,4,7,10 tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA) as the chelating agent. A site specific coupling strategy was used which involves the thioaddition of Tau-A2-SH to the synthetic maleimido-(DOTA/Gd)3 compound. Tau-A2-SH was totally converted into the well-defined conjugate Tau-A2-S-(DOTA/Gd).sub.3, as shown by RP-HPLC/MS, with 65% yield. The pI of Tau-A2-S-(DOTA/Gd).sub.3 was slightly reduced compared to the one of the unlabeled Tau-A2-SH (data not shown).

EXAMPLE II

Synthesis of a Variant of Tau-A2-SH: Tau-A2Var-SH

(106) VHH Tau-A2 specifically detected phosphorylated Tau but its production level is rather low, in average 150 μg/L, and it tends to aggregate. It was therefore needed to provide a variant of VHH Tau-A2 with improved properties.

(107) Two hydrophobic amino acids located in 2 hydrophobic areas of VHH Tau-A2-SH of SEQ ID NO. 14 were mutated: isoleucine at position 76 replaced by a glycine (I76G) and valine at position 110 by a glycine (V110G) (FIG. 10). These mutations increased the hydrophilicity of the VHH with an increase of the GRAVY index from −0.280 for Tau-A2-SH to −0.365 for the variant (GRAVY or Grand average of hydropathicity index indicates the solubility of the proteins: positive GRAVY (hydrophobic), negative GRAVY (hydrophilic)). Glutamine at position 26 was also mutated by a glutamic acid (Q26E) and an aspartic acid (D) and anvaline (V) were added between position 22 and 23 to slightly decrease the pI (less basic) (FIG. 10). Three of these mutated amino acids (D23, E26, and G110) are located outside the CDRs. This mutant is referred to as Tau-A2var-SH (SEQ ID NO. 17).

(108) VHH Tau-A2var-SH (SEQ ID NO. 17) was cloned in pET23d vector and expressed in E. coli. It was purified on a Ni-NTA resin by elution in imidazole. Analyses of Tau-A2-SH in comparison with Tau-A2var-SH were performed by several methods: SDS-PAGE showed only one band at an apparent molecular weight around 15-16 KDa (FIG. 11A). A better production of Tau-A2var-SH was observed with an expression rate estimated to 0.8-1 mg/L, 5 times more than the parental counterpart. Moreover Tau-A2var-SH has a lower tendency to aggregate. The pI is 9.68 slightly less basic than Tau-A2-SH (FIG. 11B). the ability of both VHHs to recognize phospho-Tau was evaluated by ELISA (FIG. 11C). Comparison of the binding curves of Tau-A2var-SH and Tau-A2-SH did not show any difference for their recognition of phospho-Tau (FIG. 11C). to further assess the functional activity of Tau-A2var-SH, IHC was then carried out on tissue from mice model of NFTs (Tg4510 mice). After incubation of the brain slices with the different compounds and revelation by an anti-His secondary antibody, a detection of NFTs was observed with both compounds on mouse tissues (FIG. 11D). Immunostaining was even slightly enhanced using Tau-A2var as compared to Tau-A2.

(109) The VHH Tau-A2var-SH was labeled using a similar procedure on the non-dialized protein solution, directly after the affinity column elution.

(110) Then the ability of in vivo detection of NFTs with VHH Tau-A2var-S-(DOTA/Gd).sub.3 have been realized. A 10 mg/kg dose of Tau-A2var-S-(DOTA/Gd).sub.3 was injected in the tail vein (270 μg, 150 μL) of two Tg4510 mice. Four hours after the intravenous injection of Tau-A2var-S-(DOTA/Gd).sub.3, the brain was harvested and 5 μm-thick paraffin sections were prepared. IHC was then performed with rabbit polyclonal anti-VHH antibodies. NFTs were labeled in the cortex (FIG. 12). These data confirm the brain penetration, diffusion and labeling of NFTs by Tau-A2var-S-(DOTA/Gd).sub.3.

CONCLUSION

(111) These results show that Tau-A2var behaves in a similar way as Tau-A2 for the recognition of phospho tau and NFTs. Using IHC studies after iv injection, VHH Tau-A2var-S-(DOTA/Gd).sub.3 showed its ability to cross the BBB and to penetrate into neurons after crossing a second (plasma membrane) barrier to reach its cytoplasmic target.