Compound for the sequestration of undesirable antibodies in a patient

Abstract

The present invention provides a compound for the sequestration of undesirable antibodies (e.g. related to an autoimmune disease) in a patient. The compound comprises an inert biopolymer scaffold and at least a first peptide n-mer of the general formula P(SP).sub.(n-1) and a second peptide n-mer of the general formula P(SP).sub.(n-1); wherein, independently for each occurrence, P is a peptide with a sequence length of 2-13 amino acids and S is a non-peptide spacer, wherein, independently for each of the peptide n-mers, n is an integer of at least 1, wherein each of the peptide n-mers is bound to the biopolymer scaffold. Also provided are pharmaceutical compositions comprising the compound, as well as a method of sequestering one or more antibodies present in an individual and a method of inhibiting an immune reaction to a treatment with an active agent.

Claims

1. A method of rapid, selective depletion of an antibody when the antibody has become undesirable in an individual, the method comprising the following steps: (i) identifying at least one mimotope for which the antibody is specific; (ii) administering a first pharmaceutical composition comprising the antibody to a human individual in need thereof; and (iii) administering a second pharmaceutical composition to said individual, wherein the second pharmaceutical composition comprises at least one pharmaceutically acceptable excipient and a compound comprising a biopolymer scaffold selected from the group consisting of a human albumin, a human transferrin, a human alpha1-globulin, a human alpha2-globulin, a human beta globulin, and a human haptoglobin, and at least a first mimotope of said antibody, wherein the first mimotope has a sequence length of 6-13 amino acids and does not bind to any human leukocyte antigen (HLA) class I molecule, and a second mimotope of said antibody, wherein the second mimotope has a sequence length of 6-13 amino acids and does not bind to any HLA class I molecule; wherein each of the first and second mimotopes is covalently bound to the biopolymer scaffold; and wherein the second pharmaceutical composition of step (iii) is non-immunogenic in the individual; and wherein step (iii) is performed after step (ii), wherein step (iii) is performed when the antibody has become undesirable in the individual.

2. The method of claim 1, wherein the biopolymer scaffold is selected from the group consisting of human albumin, human alpha1-globulin, human alpha2-globulin and human beta-globulin.

3. The method of claim 2, wherein the biopolymer scaffold is a human transferrin.

4. The method of claim 2, wherein the biopolymer scaffold is a human albumin.

5. The method of claim 1, wherein the antibody is a monoclonal antibody.

6. The method of claim 1, wherein the individual has been administered an antibody-drug conjugate or a nanobody.

7. The method of claim 1, wherein the individual has been administered a diagnostic antibody.

8. The method of claim 1, wherein the individual has been administered a therapeutic antibody.

9. The method of claim 1, wherein the individual has a malignancy or a cancer.

10. The method of claim 8, wherein step (iii) is performed in case of an adverse event induced by the antibody.

11. The method of claim 8, wherein step (iii) is performed as an emergency intervention.

12. The method of claim 1, wherein the molar ratio of the first and second mimotopes together to biopolymer scaffold in the composition is from 7:1 to 50:1.

13. The method of claim 1, wherein the first mimotope and the second mimotope are different.

14. A method of rapid, selective depletion of an antibody when the antibody has become undesirable in an individual, the method comprising the following steps: (i) identifying at least one mimotope for which the antibody is specific; (ii) administering a first pharmaceutical composition comprising the antibody to a human individual in need thereof, wherein the individual has a malignancy or a cancer; and (iii) administering a second pharmaceutical composition to said individual, wherein the pharmaceutical composition comprises at least one pharmaceutically acceptable excipient and a compound comprising a biopolymer scaffold selected from the group consisting of a human transferrin, a human alpha1-globulin, a human alpha2-globulin, a human beta globulin, and a human haptoglobin, and at least a first mimotope P of said antibody and a second mimotope P of said antibody; wherein, independently for each occurrence, P is a peptide with a sequence length of 7-13 amino acids and does not bind to any HLA class I molecule, and wherein each of the first and second mimotopes is covalently bound to the biopolymer scaffold; wherein the molar ratio of the first and second mimotopes P together to biopolymer scaffold in the composition is from 7:1 to 50:1; and wherein the second pharmaceutical composition is non-immunogenic in the individual; and wherein step (iii) is performed after step (ii), wherein step (iii) is performed when the antibody has become undesirable in the individual.

15. A method of rapid, selective depletion of an antibody when the antibody has become undesirable in an individual, the method comprising the following steps: (i) identifying at least one mimotope for which the antibody is specific; (ii) administering a first pharmaceutic composition comprising the antibody to a human individual in need thereof wherein the individual has a malignancy or a cancer; and (iii) administering a second pharmaceutical composition to said individual, wherein the pharmaceutical composition comprises at least one pharmaceutically acceptable excipient and a compound comprising a biopolymer scaffold, wherein the biopolymer scaffold is human transferrin, and at least a first mimotope P of said antibody and a second mimotope P of said antibody; wherein, independently for each occurrence, P is a peptide with a sequence length of 7-13 amino acids and does not bind to any HLA class I molecule, wherein each of the first mimotope and the second mimotope is covalently bound to the biopolymer scaffold; wherein the molar ratio of the first and second mimotopes P together to biopolymer scaffold in the composition is from 7:1 to 50:1; and wherein the second pharmaceutical composition is non-immunogenic in the individual; and wherein step (iii) is performed after step (ii), wherein step (iii) is performed when the antibody has become undesirable in the individual.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1: The compound of the present invention successfully reduces the titre of undesired antibodies. Each compound of the invention was applied at time point 0 by i.p. injection into Balb/c mice pre-immunized by peptide immunization against a defined antigen. Each top panel shows anti-peptide titers (0.5? dilution steps; X-axis shows log(X) dilutions) against OD values (y-axis) according to a standard ELISA detecting the corresponding antibody. Each bottom panel shows titers Log IC50 (y-axis) before injection of each compound of the invention (i.e. titers at ?48 h and ?24 h) and after application of each compound of the invention (i.e. titers+24 h, +48 h and +72 h after injection; indicated on the x-axis). (A) Compound with albumin as the biopolymer scaffold that binds to antibodies directed against EBNA1 (associated with pre-eclampsia). The mice were pre-immunized with a peptide vaccine carrying the EBNA-1 model epitope. (B) Compound with albumin as the biopolymer scaffold that binds to antibodies directed against a peptide derived from the human AChR protein MIR (associated with myasthenia gravis). The mice were pre-immunized with a peptide vaccine carrying the AChR MIR model epitope. (C) Compound with immunoglobulin as the biopolymer scaffold that binds to antibodies directed against EBNA1 (associated with pre-eclampsia). The mice were pre-immunized with a peptide vaccine carrying the EBNA-1 model epitope. (D) Compound with haptoglobin as the biopolymer scaffold that binds to antibodies directed against EBNA1 (associated with pre-eclampsia). The mice were pre-immunized with a peptide vaccine carrying the EBNA-1 model epitope. (E) Demonstration of selectivity using the same immunoglobulin-based compound of the invention binding to antibodies directed against EBNA1 that was used in the experiment shown in panel C. The mice were pre-immunized with an unrelated amino acid sequence. No titre reduction occurred, demonstrating selectivity of the compound.

(2) FIG. 2: The compound of the invention is non-immunogenic and does not induce antibody formation after repeated injection into mice. Animals C1-C4 as well as animals C5-C8 were treated i.p. with two different compounds of the invention. Control animal C was vaccinated with a KLH-peptide derived from the human AChR protein MIR. Using BSA-conjugated peptide probes T3-1, T9-1 and E005 (grey bars, as indicated in the graph), respectively, for antibody titer detection by standard ELISA at a dilution of 1:100, it could be demonstrated that antibody induction was absent in animals treated with a compound of the invention, when compared to the vaccine-treated control animal C (y-axis, OD450 nm).

(3) FIG. 3: Successful in vitro depletion of antibodies using SADCs carrying multiple copies of monovalent or divalent peptides. SADCs with mono- or divalent peptides were very suitable to adsorb antibodies and thereby deplete them. Monovalent means that peptide monomers are bound to the biopolymer scaffold (i.e. n=1) whereas divalent means that peptide dimers are bound to the biopolymer scaffold (i.e. n=2). In the present case, the divalent peptides were homodivalent, i.e. the peptide n-mer of the SADC is E006spacerE006).

(4) FIG. 4: Rapid, selective antibody depletion in mice using various SADC biopolymer scaffolds. Treated groups exhibited rapid and pronounced antibody reduction already at 24 hrs (in particular SADC-TF) when compared to the mock treated control group SADC-CTL (containing an unrelated peptide). SADC with albumin scaffoldSADC-ALB, SADC with immunoglobulin scaffoldSADC-IG, SADC with haptoglobin scaffold-SADC-HP, and SADC with transferrin scaffold-SADC-TF.

(5) FIG. 5: Detection of SADCs in plasma via their peptide moieties 24 hrs after SADC injection. Both haptoglobin-scaffold-based SADCs (SADC-HP and SADC-CTL) exhibited a relatively shorter plasma half life which represents an advantage over SADCs with other biopolymer scaffolds such as SADC-ALB, SADC-IG oder SADC-TF. SADC with albumin scaffold-SADC-ALB, SADC with immunoglobulin scaffold-SADC-IG, SADC with haptoglobin scaffold-SADC-HP, and SADC with transferrin scaffold-SADC-TF.

(6) FIG. 6: Detection of SADC-IgG complexes in plasma 24 hrs after SADC injection. Haptoglobin based SADCs were subject to accelerated clearance when compared to SADCs with other biopolymer scaffolds. SADC with albumin scaffold-SADC-ALB, SADC with immunoglobulin scaffold-SADC-IG, SADC with haptoglobin scaffoldSADC-HP, and SADC with transferrin scaffoldSADC-TF.

(7) FIG. 7: In vitro analysis of SADC-IgG complex formation. Animals SADC-TF and -ALB showed pronounced immunocomplex formation and binding to Clq as reflected by the strong signals and by sharp signal lowering in case 1000 ng/ml SADC-TF due to the transition from antigen-antibody equilibrium to antigen excess. In contrast, in vitro immunocomplex formation with SADC-HP or SADC-IG were much less efficient when measured in the present assay. These findings corroborate the finding that haptoglobin scaffolds are advantageous over other SADC biopolymer scaffolds because of the reduced propensity to activate the complement system. SADC with albumin scaffoldSADC-ALB, SADC with immunoglobulin scaffoldSADC-IG, SADC with haptoglobin scaffoldSADC-HP, and SADC with transferrin scaffoldSADC-TF.

(8) FIG. 8: Determination of IgG capturing by SADCs in vitro. SADC-HP showed markedly less antibody binding capacity in vitro when compared to SADC-TF or SADC-ALB. SADC with albumin scaffoldSADC-ALB, SADC with immunoglobulin scaffoldSADC-IG, SADC with haptoglobin scaffoldSADC-HP, and SADC with transferrin scaffoldSADC-TF.

EXAMPLES

Example 1: The Compound of the Present Invention Effectively Reduces the Titre of Undesired Antibodies

(9) Animal models: In order to provide in vivo models with measurable titers of prototypic undesired antibodies in human indications, BALB/c mice were immunized using standard experimental vaccination with KLH-conjugated peptide vaccines derived from established human autoantigens or anti-drug antibodies. After titer evaluation by standard peptide ELISA, immunized animals were treated with the corresponding test SADCs to demonstrate selective antibody lowering by SADC treatment. All experiments were performed in compliance with the guidelines by the corresponding animal ethics authorities.

(10) Immunization of mice with model antigens: Female BALB/c mice (aged 8-10 weeks) were supplied by Janvier (France), maintained under a 12 h light/12 h dark cycle and given free access to food and water. Immunizations were performed by s.c. application of KLH carrier-conjugated peptide vaccines injected 3 times in biweekly intervals. KLH conjugates were generated with peptide T3-2 (SEQ ID NO. 14: CGRPQKRPSCIGCKG), which represents an example for molecular mimicry between a viral antigen (EBNA-1) and an endogenous human receptor antigen, namely the placental GPR50 protein, that was shown to be relevant to preeclampsia (Elliott et al.). In order to confirm the generality of this approach, a larger antigenic peptide derived from the autoimmune condition myasthenia gravis was used for immunization of mice with a human autoepitope. In analogy to peptide T3-2, animals were immunized with peptide T1-1 (SEQ ID NO. 15: LKWNPDDYGGVKKIHIPSEKGC), derived from the MIR (main immunogenic region) of the human AChR protein which plays a fundamental role in pathogenesis of the disease (Luo et al.). The T1-1 peptide was used for immunizing mice with a surrogate partial model epitope of the human AChR autoantigen. The peptide T8-1 (SEQ ID NO. 16: DHTLYTPYHTHPG) was used to immunize control mice to provide a control titer for proof of selectivity of the system. For vaccine conjugate preparation, KLH carrier (Sigma) was activated with sulfo-GMBS (Cat. Nr. 22324 Thermo), according to the manufacturer's instructions, followed by addition of either N- or C-terminally cysteinylated peptides T3-2 and T1-1 and final addition of Alhydrogel? before injection into the flank of the animals. The doses for vaccines T3-2 and T1-1 were 15 ?g of conjugate in a volume of 100 ul per injection containing Alhydrogel? (InvivoGen VAC-Alu-250) at a final concentration of 1% per dose.

(11) Generation of prototypic SADCs: For testing selective antibody lowering activity by SADCs of T3-2 and T1-1 immunized mice, SADCs were prepared with mouse serum albumin (MSA) or mouse immunoglobulin (mouse-Ig) as biopolymer scaffold in order to provide an autologous biopolymer scaffold, that will not induce any immune reaction in mice, or non-autologuous human haptoglobin as biopolymer scaffold (that did not induce an allogenic reaction after one-time injection within 72 hours). N-terminally cysteinylated SADC peptide E049 (SEQ ID NO. 13: GRPQKRPSCIG) and/or C-terminally cysteinylated SADC peptide E006 (SEQ ID NO. 4: VKKIHIPSEKG) were linked to the scaffold using sulfo-GMBS (Cat. Nr. 22324 Thermo)-activated MSA (Sigma; Cat. Nr. A3559) or -mouse-Ig (Sigma, 15381) or -human haptoglobin (Sigma H0138) according to the instructions of the manufacturer, thereby providing MSA-, Ig- and haptoglobin-based SADCs with the corresponding cysteinylated peptides, that were covalently attached to the lysines of the corresponding biopolymer scaffold. Beside conjugation of the cysteinylated peptides to the lysines via a bifunctional amine-to-sulfhydryl crosslinker, a portion of the added cysteinylated SADC peptides directly reacted with sulfhydryl groups of cysteins of the albumin scaffold protein, which can be detected by treating the conjugates with DTT followed by subsequent detection of free peptides using mass spectrometry or any other analytical method that detects free peptide. Finally, these SADC conjugates were dialysed against water using Pur-A-Lyzer? (Sigma) and subsequently lyophilized. The lyophilized material was resuspended in PBS before injection into animals.

(12) In vivo functional testing of SADCs: Prototypic SADCs, SADC-E049 and SADC-E006 were injected intraperitoneally (i.p.; as a surrogate for an intended intravenous application in humans and larger animals) into the mice that had previously been immunized with peptide vaccine T3-2 (carrying the EBNA-1 model epitope) and peptide vaccine T1-1 (carrying the AChR MIR model epitope). The applied dose was 30 ?g SADC conjugate in a volume of 50 ?l PBS. Blood takes were performed by submandibular vein puncture, before (?48 h, ?24 h) and after (+24 h, +48 h, +72 h, etc.) i.p. SADC injections, respectively, using capillary micro-hematocrit tubes. Using ELISA analysis (see below), it was found that both prototypic SADCs were able to clearly reduce the titers over a period of at least 72 hrs in the present animal model. It could therefore be concluded that SADCs can be used to effectively reduce titers in vivo.

(13) Titer analysis: Peptide ELISAs were performed according to standard procedures using 96-well plates (Nunc Medisorp plates; Thermofisher, Cat Nr 467320) coated for 1 h at RT with BSA-coupled peptides (30 nM, dissolved in PBS) and incubated with the appropriate buffers while shaking (blocking buffer, 1% BSA, lx PBS; washing buffer, 1?PBS/0.1% Tween; dilution buffer, 1?PBS/0.1% BSA/0.1% Tween). After serum incubation (dilutions starting at 1:50 in PBS; typically in 1:3 or 1:2 titration steps), bound antibodies were detected using Horseradish Peroxidase-conjugated goat anti-mouse IgG (Fc) from Jackson immunoresearch (115-035-008). After stopping the reaction, plates were measured at 450 nm for 20 min using TMB. EC50 were calculated from readout values using curve fitting with a 4-parameter logistic regression model (GraphPad Prism) according to the procedures recommended by the manufacturer. Constraining parameters for ceiling and floor values were set accordingly, providing curve fitting quality levels of R.sup.2>0.98.

(14) FIG. 1A shows an in vivo proof of concept in a mouse model for in vivo selective plasma-lowering activity of a prototypic albumin-based SADC candidate that binds to antibodies directed against EBNA1, as a model for autoantibodies and mimicry in preeclampsia (Elliott et al.). For these mouse experiments, mouse albumin was used, in order to avoid any reactivity against a protein from a foreign species. Antibody titers were induced in 6 months old Balb/c mice by standard peptide vaccination. The bottom panel demonstrates that titers Log IC50 (y-axis) before SADC injection (i.e. titers at ?48 h and ?24 h) were higher than titers Log IC50 after SADC application (i.e. titers+24 h, +48 h and +72 h after injection; indicated on the x-axis).

(15) A similar example is shown in FIG. 1B, using an alternative example of a peptidic antibody binding moiety for a different disease indication. Antibody lowering activity of an albumin-based SADC in a mouse model that was pre-immunized with a different peptide derived from the human AChR protein MIR region (Luo et al.) in order to mimic the situation in myasthenia gravis. The induced antibody titers against the AChR-MIR region were used as surrogate for anti-AChR-MIR autoantibodies known to play a causative role in myasthenia gravis (reviewed by Vincent et al.). A clear titer reduction was seen after SADC application.

(16) FIGS. 1C and 1D demonstrate the functionality of SADC variants comprising alternative biopolymer scaffolds. Specifically, FIG. 1C shows that an immunoglobulin scaffold can be successfully used whereas FIG. 1D demonstrates the use of a haptoglobin-scaffold for constructing an SADC. Both examples show an in vivo proof of concept for selective antibody lowering by an SADC, carrying covalently bound example peptide E049.

(17) The haptoglobin-based SADC was generated using human Haptoglobin as a surrogate although the autologuous scaffold protein would be preferred. In order to avoid formation of anti-human-haptoglobin antibodies, only one single SADC injection per mouse of the non-autologuous scaffold haptoglobin was used for the present experimental conditions. As expected, under the present experimental conditions (i.e. one-time application), no antibody reactivity was observed against the present surrogate haptoglobin homologue.

(18) FIG. 1E demonstrates the selectivity of the SADC system. The immunoglobulin-based SADC carrying the peptide E049 (i.e. the same as in FIG. 1C) cannot reduce the Ig-titer that was induced by a peptide vaccine with an unrelated, irrelevant aminoacid sequence, designated peptide T8-1 (SEQ ID NO. 16: DHTLYTPYHTHPG). The example shows an in vivo proof of concept for the selectivity of the system. The top panel shows anti-peptide T8-1 titers (0.5? dilution steps starting from 1:50 to 1:102400; X-axis shows log(X) dilutions) against OD values (y-axis) according to a standard ELISA. T8-1-titers are unaffected by administration of SADC-Ig-E049 after application. The bottom panel demonstrates that the initial titers Log IC50 (y-axis) before SADC injection (i.e. titers at ?48 h and ?24 h) are unaffected by administration of SADC-Ig-E049 (arrow) when compared to the titers Log IC50 after SADC application (i.e. titers+24 h, +48 h and +72 h; as indicated on the x-axis), thereby demonstrating the selectivity of the system.

Example 2: Immunogenicity of SADCs

(19) In order to exclude immunogenicity of SADCs, prototypic candidate SADCs were tested for their propensity to induce antibodies upon repeated injection. Peptides T3-1 and T9-1 were used for this test. T3-1 is a 10-amino acid peptide derived from a reference epitope of the Angiotensin receptor, against which agonistic autoantibodies are formed in a pre-eclampsia animal model (Zhou et al.); T9-1 is a 12-amino acid peptide derived from a reference anti-drug antibody epitope of human IFN gamma (Lin et al.). These control SADC conjugates were injected 8? every two weeks i.p. into na?ve, non-immunized female BALB/c mice starting at an age of 8-10 weeks.

(20) Animals C1-C4 were treated i.p. (as described in example 1) with SADC T3-1. Animals C5-C8 were treated i.p. with an SADC carrying the peptide T9-1. As a reference signal for ELISA analysis, plasma from a control animal that was vaccinated 3 times with KLH-peptide T1-1 (derived from the AChR-MIR, explained in Example 1) was used. Using BSA-conjugated peptide probes T3-1, T9-1 and E005 (SEQ ID NO. 17: GGVKKIHIPSEK), respectively, for antibody titer detection by standard ELISA at a dilution of 1:100, it could be demonstrated that antibody induction was absent in SADC-treated animals, when compared to the vaccine-treated control animal C (see FIG. 2). The plasmas were obtained by submandibular blood collection, 1 week after the 3rd vaccine injection (control animal C) and after the last of 8 consecutive SADC injections in 2-weeks intervals (animals C1-C8), respectively. Thus it was demonstrated that SADCs are non-immunogenic and do not induce antibody formation after repeated injection into mice.

Example 3: Successful In Vitro Depletion of Antibodies Using SADCs Carrying Multiple Copies of Monovalent or Divalent Peptides

(21) Plasma of E006-KLH (VKKIHIPSEKG (SEQ ID NO: 4) with C-terminal cysteine, conjugated to KLH) vaccinated mice was diluted 1:3200 in dilution buffer (PBS+0.1% w/v BSA+0.1% Tween20) and incubated (100 ?l, room temperature) sequentially (10 min/well) four times on single wells of a microtiter plate that was coated with 2.5 ?g/ml (250 ng/well) of SADC or 5 ?g/ml (500 ng/well) albumin as negative control.

(22) In order to determine the amount of free, unbound antibody present before and after incubation on SADC coated wells, 50 ?l of the diluted serum were taken before and after the depletion and quantified by standard ELISA using E006-BSA coated plates (10 nM peptide) and detection by goat anti mouse IgG bio (Southern Biotech, diluted 1:2000). Subsequently, the biotinylated antibody was detected with Streptavidin-HRP (Thermo Scientific, diluted 1:5000) using TMB as substrate. Development of the signal was stopped with 0.5 M sulfuric acid.

(23) ELISA was measured at OD450 nm (y-axis). As a result, the antibody was efficiently adsorbed by either coated mono- or divalent SADCs containing peptide E006 with C-terminal cysteine (sequence VKKIHIPSEKGC, SEQ ID NO: 4) (before=non-depleted starting material; mono-divalent corresponds to peptides displayed on the SADC surface; neg. control was albumin; indicated on the x-axis). See FIG. 3. (Monovalent means that peptide monomers are bound to the biopolymer scaffold (i.e. n=1) whereas divalent means that peptide dimers are bound to the biopolymer scaffold (i.e. n=2). In the present case, the divalent peptides were homodivalent, i.e. the peptide n-mer of the SADC is E006-S-E006.)

(24) This demonstrates that SADCs with mono- or divalent peptides are very suitable to adsorb antibodies and thereby deplete them.

Example 4: Generation of Mimotope-Based SADCs

(25) Linear and circular peptides derived from wild-type or modified peptide amino acid sequences can be used for the construction of specific SADCs for the selective removal of harmful, disease-causing or otherwise unwanted antibodies directed against a particular epitope. In case of a particular epitope, linear peptides or constrained peptides such as cyclopeptides containing portions of an epitope or variants thereof, where for example, one or several amino acids have been substituted or chemically modified in order to improve affinity to an antibody (mimotopes), can be used for constructing SADCs. A peptide screen can be performed with the aim of identifying peptides with optimized affinity to a disease-inducing autoantibody. The flexibility of structural or chemical peptide modification provided a solution to minimize the risk of immunogenicity, in particular of binding of the peptide to HLA and thus the risk of unwanted immune stimulation.

(26) Therefore, wild-type as well as modified linear and circular peptide sequences were derived from a known epitope associated with an autoimmune disease. Peptides of various length and positions were systematically permutated by amino acid substitutions and synthesized on the PEPperCHIP? peptide array Platform (PEPperPRINT GmbH, Germany). This allowed screening of 60000 circular and linear wild-type and mimotope peptides derived from these sequences. The peptide arrays were incubated with an autoantibody known to be involved in the autoimmune disease. This autoantibody was therefore used to screen the 60000 peptides and 100 circular and 100 linear peptide hits were selected based on their relative binding strength to the autoantibody. Of these 200 peptides, 51 sequences were identical between the circular and the linear peptide group. All of the best peptides identified had at least one amino acid substitution when aligned to the original sequences, respectively and are therefore regarded as mimotopes. It also turned out that higher binding strengths can be achieved with circularized peptides.

(27) These newly identified peptides, preferentially those with high relative binding values, are used to generate SADCs that are able to remove autoantibodies directed against this particular epitope or to develop further mimotopes and derivatives based on their sequences.

Example 5: Rapid, Selective Antibody Depletion in Mice Using Various SADC Biopolymer Scaffolds

(28) 10 ?g of model undesired antibody mAB anti V5 (Thermo Scientific) was injected i.p. into female Balb/c mice (5 animals per treatment group; aged 9-11 weeks) followed by intravenous injection of 50 ?g SADC (different biopolymer scaffolds with tagged V5 peptides bound, see below) 48 hrs after the initial antibody administration. Blood was collected at 24 hrs intervals from the submandibular vein. Blood samples for time point 0 hrs were taken just before SADC administration.

(29) Blood was collected every 24 hrs until time point 120 hrs after the SADC administration (x-axis). The decay and reduction of plasma anti-V5 IgG levels after SADC administration was determined by anti V5 titer readout using standard ELISA procedures in combination with coated V5-peptide-BSA (peptide sequence IPNPLLGLDC-SEQ ID NO: 21) and detection by goat anti mouse IgG bio (Southern Biotech, diluted 1:2000) as shown in FIG. 4. In addition, SADC levels (see Example 6) and immunocomplex formation (see Example 7) were analyzed.

(30) EC50[OD450] values were determined using 4 parameter logistic curve fitting and relative signal decay between the initial level (set to 1 at time point 0) and the following time points (x-axis) was calculated as ratio of the EC50 values (y-axis, fold signal reduction EC50). All SADC peptides contained tags for direct detection of SADC and immunocomplexes from plasma samples; peptide sequences used for SADCs were: IPNPLLGLDGGSGDYKDDDDKGK(SEQ ID NO: 22)-(BiotinAca)GC (SADC with albumin scaffoldSADC-ALB, SADC with immunoglobulin scaffoldSADC-IG, SADC with haptoglobin scaffoldSADC-HP, and SADC with transferrin scaffoldSADC-TF) and unrelated peptide VKKIHIPSEKGGSGDYKDDDDKGK(SEQ ID NO: 23)-(BiotinAca)GC as negative control SADC (SADC-CTR).

(31) The SADC scaffolds for the different treatment groups of 5 animals are displayed in black/grey shades (see inset of FIG. 4).

(32) Treated groups exhibited rapid and pronounced antibody reduction already at 24 hrs (in particular SADC-TF) when compared to the mock treated control group SADC-CTL. SADC-CTR was used as reference for a normal antibody decay since it has no antibody lowering activity because its peptide sequence is not recognized by the administered anti V5 antibody. The decay of SADC-CTR is thus marked with a trend line, emphasizing the antibody level differences between treated and mock treated animals.

(33) In order to determine the effectivity of selective antibody lowering under these experimental conditions, a two-way ANOVA test was performed using a Dunnett's multiple comparison test. 48 hrs after SADC administration, the antibody EC50 was highly significantly reduced in all SADC groups (p<0.0001) compared to the SADC-CTR reference group (trend line). At 120 hrs after SADC administration, antibody decrease was highly significant in the SADC-ALB and SADC-TF groups (both p<0.0001) and significant in the SADC-HP group (p=0.0292), whereas the SADC-IG group showed a trend towards an EC50 reduction(p=0.0722) 120 hrs after SADC administration. Of note, selective antibody reduction was highly significant (p<0.0001) in the SADC-ALB and SADC-TF groups at all tested time-points after SADC administration.

(34) It is concluded that all SADC biopolymer scaffolds were able to selectively reduce antibody levels. Titer reduction was most pronounced with SADC-ALB and SADC-TF and no rebound or recycling of antibody levels was detected towards the last time points suggesting that undesired antibodies are degraded as intended.

Example 6: Detection of SADCs in Plasma 24 Hrs after SADC Injection

(35) Plasma levels of different SADC variants at 24 hrs after i.v. injection into Balb/c mice. Determination of Plasma levels (y-axis) of SADC-ALB, -IG, -HP, -TF and the negative control SADC-CTR (x-axis), were detected in the plasmas from the animals already described in example 5. Injected plasma SADC levels were detected by standard ELISA whereby SADCs were captured via their biotin moieties of their peptides in combination with streptavidin coated plates (Thermo Scientific). Captured SADCs were detected by mouse anti Flag-HRP antibody (Thermo Scientific, 1:2,000 diluted) detecting the Flag-tagged peptides (see also example 7):

(36) Assuming a theoretical amount in the order of 25 ?g/ml in blood after injecting 50 ?g SADC i.v., the detectable amount of SADC ranged between 799 and 623 ng/ml for SADC-ALB or SADC-IG and up to approximately 5000 ng/ml for SADC-TF, 24 hrs after SADC injection. However surprisingly and in contrast, SADC-HP and control SADC-CTR (which is also a SADC-HP variant, however carrying the in this case unrelated negative control peptide E006, see previous examples), had completely disappeared from circulation 24 hrs after injection, and were not detectable anymore. See FIG. 5.

(37) This demonstrates that both Haptoglobin scaffold-based SADCs tested in the present example ((namely SADC-HP and SADC-CTR) exhibit a relatively shorter plasma half-life which represents an advantage over SADCs such as SADC-ALB, SADC-IG oder SADC-TF in regard of their potential role in complement-dependent vascular and renal damage due to the in vivo risk of immunocomplex formation. Another advantage of SADC-HP is the accelerated clearance rate of their unwanted target antibody from blood in cases where a rapid therapeutic effect is needed. The present results demonstrate that Haptoglobin-based SADC scaffolds (as represented by SADC-HP and SADC-CTR) are subject to rapid clearance from the blood, regardless of whether SADC-binding antibodies are present in the blood, thereby minimizing undesirable immunocomplex formation and showing rapid and efficient clearance. Haptoglobin-based SADCs such as SADC-HP in the present example thus provide a therapeutically relevant advantage over other SADC biopolymer scaffolds, such as demonstrated by SADC-TF or SADC-ALB, both of which are still detectable 24 hrs after injection under the described conditions, in contrast to SADC-HP or SADC-CTR which both are completely cleared 24 hrs after injection.

Example 7: Detection of SADC-IgG Complexes in Plasma 24 Hrs after SADC Injection

(38) In order to determine the amount IgG bound to SADCs in vivo, after i.v. injection of 10 ?g anti V5 IgG (Thermo Scientific) followed by injection of SADC-ALB, -HP, -TF and -CTR (50 ?g) administered i.v. 48 h after antibody injection, plasma was collected from the submandibular vein, 24 hrs after SADC injection, and incubated on streptavidin plates for capturing SADCs from plasma via their biotinylated SADC-V5-peptide [IPNPLLGLDGGSGDYKDDDDKGK(SEQ ID NO: 22) (BiotinAca)GC or in case of SADC-CTR the negative control peptide VKKIHIPSEKGGSGDYKDDDDKGK(SEQ ID NO: 23) (BiotinAca)GC]. IgG bound to the streptavidin-captured SADCs was detected by ELISA using a goat anti mouse IgG HRP antibody (Jackson Immuno Research, diluted 1:2,000) for detection of the SADC-antibody complexes present in plasma 24 hrs after SADC injection. OD450 nm values (y-axis) obtained for a negative control serum from untreated animals were subtracted from the OD450 nm values of the test groups (x-axis) for background correction.

(39) As shown in FIG. 6, pronounced anti-V5 antibody signals were seen in case of SADC-ALB and SADC-TF injected mice (black bars represent background corrected OD values at a dilution of 1:25, mean value of 5 mice; standard deviation error bars), whereas no antibody signal could be detected in plasmas from SADC-HP or control SADC-CTR injected animals (SADC-CTR is a negative control carrying the irrelevant peptide bio-FLG-E006 [VKKIHIPSEKGGSGDYKDDDDKGK(SEQ ID NO: 23) (BiotinAca)GC] that is not recognized by any anti V5 antibody). This demonstrates the absence of detectable amounts of SADC-HP/IgG complexes in the plasma 24 hrs after i.v. SADC application.

(40) SADC-HP is therefore subject to accelerated clearance in anti V5 pre-injected mice when compared to SADC-ALB or SADC-TF.

Example 8: In Vitro Analysis of SADC-Immunoglobulin Complex Formation

(41) SADC-antibody complex formation was analyzed by pre-incubating 1 ?g/ml of human anti V5 antibody (anti V5 epitope tag [SVSP-K], human IgG3, Absolute Antibody) with increasing concentrations of SADC-ALB, -IG, -HP, -TF and -CTR (displayed on the x-axis) in PBS+0.1% w/v BSA+0.1% v/v Tween20 for 2 hours at room temperature in order to allow for immunocomplex formation in vitro. After complex formation, samples were incubated on ELISA plates that had previously been coated with 10 ?g/ml of human Clq (CompTech) for 1 h at room temperature, in order to allow capturing of in vitro formed immunocomplexes. Complexes were subsequently detected by ELISA using anti human IgG (Fab specific)-Peroxidase (Sigma, diluted 1:1,000). Measured signals at OD450 nm (y-axis) reflect Antibody-SADC complex formation in vitro.

(42) As shown in FIG. 7, SADC-TF and -ALB showed pronounced immunocomplex formation and binding to Clq as reflected by the strong signals and by sharp signal lowering in case 1000 ng/ml SADC-TF due to the transition from antigen-antibody equilibrium to antigen excess. In contrast, in vitro immunocomplex formation with SADC-HP or SADC-IG were much less efficient when measured in the present assay.

(43) Together with the in vivo data (previous examples), these findings corroborate the finding that haptoglobin scaffolds are advantageous over other SADC biopolymer scaffolds because of the reduced propensity to activate the complement system. In contrast, SADC-TF or SADC-ALB show higher complexation, and thereby carry a certain risk of activating the C1 complex with initiation of the classical complement pathway (a risk which may be tolerable in some settings, however).

Example 9: Determination of IgG Capturing by SADCs In Vitro

(44) Immunocomplexes were allowed to form in vitro, similar to the previous example, using 1 ?g/ml mouse anti V5 antibody (Thermo Scientific) in combination with increasing amounts of SADCs (displayed on the x-axis). SADC-antibody complexes were captured on a streptavidin coated ELISA plate via the biotinylated SADC-peptides (see previous examples), followed by detection of bound anti-V5 using anti mouse IgG-HRP (Jackson Immuno Research, diluted 1:2,000).

(45) Under these assay conditions, SADC-HP showed markedly less antibody binding capacity in vitro when compared to SADC-TF or SADC-ALB (see FIG. 8, A). The calculated EC50 values for IgG detection on SADCs were 7.0 ng/ml, 27.9 ng/ml and 55.5 ng/ml for SADC-TF, -ALB and -HP, respectively (see FIG. 8, B).

(46) This in vitro finding is consistent with the observation (see previous examples) that SADC-HP has a lower immunocomplex formation capacity when compared to SADC-TF or SADC-ALB which is regarded as a safety advantage with respect to its therapeutic use for the depletion of unwanted antibodies.

Example 10: SADCs to Reduce Undesired Antibodies Against AAV-8

(47) Three SADCs are provided to reduce AAV-8-neutralizing antibodies which hamper gene therapy (see Gurda et al. for the epitopes used; see also AAV-8 capsid protein sequence UniProt Q8JQF8, sequence version 1): (a) SADC-a with haptoglobin as biopolymer scaffold and at least two peptides with the sequence LQQQNT (SEQ ID NO: 18) covalently bound to the scaffold, (b) SADC-b with transferrin as biopolymer scaffold and at least two peptides with the sequence TTTGQNNNS (SEQ ID NO: 19) covalently bound to the scaffold, and (c) SADC-c with albumin as biopolymer scaffold and at least two peptides with the sequence GTANTQ (SEQ ID NO: 20) covalently bound to the scaffold.

(48) These SADCs are administered to an individual who will undergo gene therapy with AAV-8 as vector in order to increase efficiency of the gene therapy.

NON-PATENT REFERENCES

(49) Carter, John Mark, and Larry Loomis-Price. B cell epitope mapping using synthetic peptides. Current protocols in immunology 60.1 (2004): 9-4. Elliott, Serra E., et al. A pre-eclampsia-associated Epstein-Barr virus antibody cross-reacts with placental GPR50. Clinical Immunology 168 (2016): 64-71. Erlandsson, Ann, et al. In vivo clearing of idiotypic antibodies with antiidiotypic antibodies and their derivatives. Molecular immunology 43.6 (2006): 599-606. Garces, Jorge Carlos, et al. Antibody-mediated rejection: a review. The Ochsner Journal 17.1 (2017): 46. Gazarian, Karlen, et al. Mimotope peptides selected from phage display combinatorial library by serum antibodies of pigs experimentally infected with Taenia solium as leads to developing diagnostic antigens for human neurocysticercosis. Peptides 38.2 (2012): 381-388. Gfeller, David, et al. Current tools for predicting cancer-specific T cell immunity. Oncoimmunology 5.7 (2016): e1177691. Gurda, Brittney L., et al. Mapping a neutralizing epitope onto the capsid of adeno-associated virus serotype 8. Journal of virology 86.15 (2012): 7739-7751. Jansson, Liselotte, et al. Immunotherapy With Apitopes Blocks the Immune Response to TSH Receptor in HLA-DR Transgenic Mice. Endocrinology 159.9 (2018): 3446-3457. Jensen, Kamilla Kjaergaard, et al. Improved methods for predicting peptide binding affinity to MHC class II molecules. Immunology 154.3 (2018): 394-406. Jurtz, Vanessa, et al. NetMHCpan-4.0: improved peptide-MHC class I interaction predictions integrating eluted ligand and peptide binding affinity data. The Journal of Immunology 199.9 (2017): 3360-3368. Ko?alo?lu-Yal?in, et al. Predicting T cell recognition of MHC class I restricted neoepitopes. Oncoimmunology 7.11 (2018): e1492508. Hansen, Lajla Bruntse, Soren Buus, and Claus Schafer-Nielsen. Identification and mapping of linear antibody epitopes in human serum albumin using high-density peptide arrays. PLoS One 8.7 (2013): e68902. Homma, Masayuki, et al. A Novel Fusion Protein, AChR-Fc, Ameliorates Myasthenia Gravis by Neutralizing Antiacetylcholine Receptor Antibodies and Suppressing Acetylcholine Receptor-Reactive B Cells. Neurotherapeutics 14.1 (2017): 191-198. Howard Jr, James F. Myasthenia gravis: the role of complement at the neuromuscular junction. Annals of the New York Academy of Sciences 1412.1 (2018): 113-128. Howarth, M., & Brune, K. D. (2018). New routes and opportunities for modular construction of particulate vaccines: stick, click and glue. Frontiers in immunology, 9, 1432. Lazaridis, Konstantinos, et al. Specific removal of autoantibodies by extracorporeal immunoadsorption ameliorates experimental autoimmune myasthenia gravis. Journal of neuroimmunology 312 (2017): 24-30. Leung, Nicki Y H, et al. Screening and identification of mimotopes of the major shrimp allergen tropomyosin using one-bead-one-compound peptide libraries. Cellular & molecular immunology 14.3 (2017): 308-318. Lim, Sung In, and Inchan Kwon. Bioconjugation of therapeutic proteins and enzymes using the expanded set of genetically encoded amino acids. Critical reviews in biotechnology 36.5 (2016): 803-815. Lin, Chia-Hao, et al. Identification of a major epitope by anti-interferon-? autoantibodies in patients with mycobacterial disease. Nature medicine 22.9 (2016): 994. Lorentz, Kristen M., et al. Engineered binding to erythrocytes induces immunological tolerance to E. coli asparaginase. Science advances 1.6 (2015): e1500112. Luo, Jie, et al. Main immunogenic region structure promotes binding of conformation-dependent myasthenia gravis autoantibodies, nicotinic acetylcholine receptor conformation maturation, and agonist sensitivity. Journal of Neuroscience 29.44 (2009): 13898-13908. Luo, Jie, and Jon Lindstrom. AChR-specific immunosuppressive therapy of myasthenia gravis. Biochemical pharmacology 97.4 (2015): 609-619. Majowicz, Anna, et al. Seroprevalence of pre-existing NABs against AAV1, 2, 5, 6 and 8 in the South African Hemophilia B patient population. (2019): 3353-3353. Mazor, Ronit, et al. Tolerogenic nanoparticles restore the antitumor activity of recombinant immunotoxins by mitigating immunogenicity. Proceedings of the National Academy of Sciences 115.4 (2018): E733-E742. Meister, Daniel, S. Maryamdokht Taimoory, and John F. Trant. Unnatural amino acids improve affinity and modulate immunogenicity: Developing peptides to treat MHC type II autoimmune disorders. Peptide Science 111.1 (2019): e24058. Mingozzi, Federico, et al. Overcoming preexisting humoral immunity to AAV using capsid decoys. Science translational medicine 5.194 (2013): 194ra92-194ra92. Mingozzi, Federico, and Katherine A. High. Overcoming the host immune response to adeno-associated virus gene delivery vectors: the race between clearance, tolerance, neutralization, and escape. Annual review of virology 4 (2017): 511-534. Morimoto et. al., Bioconjugate Chemistry 25 (8) (2014): 1479-1491 Moussa, Ehab M., et al. Immunogenicity of therapeutic protein aggregates. Journal of pharmaceutical sciences 105.2 (2016): 417-430. M?ller, Manuel M. Post-translational modifications of protein backbones: unique functions, mechanisms, and challenges. Biochemistry 57.2 (2017): 177-185. Siang Ong, Yong, et al. Recent advances in synthesis and identification of cyclic peptides for bioapplications. Current topics in medicinal chemistry 17.20 (2017): 2302-2318. Peters, Bjoern, et al. A community resource benchmarking predictions of peptide binding to MHC-I molecules. PLoS computational biology 2.6 (2006): e65. Pishesha, Novalia, et al. Engineered erythrocytes covalently linked to antigenic peptides can protect against autoimmune disease. Proceedings of the National Academy of Sciences (2017): 201701746. Rey et al., Clinical Immunology 96 (3) (2000): 269-279 Ruff, Robert L., and Robert P. Lisak. Nature and action of antibodies in myasthenia gravis. Neurologic clinics 36.2 (2018): 275-291. Rummler, Silke, et al. Current techniques for AB0-incompatible living donor liver transplantation. World journal of transplantation 6.3 (2016): 548. Runcie, Karie, et al. Bi-specific and tri-specific antibodiesthe next big thing in solid tumor therapeutics. Molecular Medicine 24.1 (2018): 50. Ryan, Brent J., Ahuva Nissim, and Paul G. Winyard. Oxidative post-translational modifications and their involvement in the pathogenesis of autoimmune diseases. Redox biology 2 (2014): 715-724. Shanmugam, Arulkumaran, et al. Identification of PSA peptide mimotopes using phage display peptide library. Peptides 32.6 (2011): 1097-1102. Sorensen, Karen Kristine, et al. Liver sinusoidal endothelial cells. Comprehensive Physiology 5.4 (2011): 1751-1774. Spiess, Christoph, Qianting Zhai, and Paul J. Carter. Alternative molecular formats and therapeutic applications for bispecific antibodies. Molecular immunology 67.2 (2015): 95-106. Taddeo, Adriano, et al. Selection and depletion of plasma cells based on the specificity of the secreted antibody. European journal of immunology 45.1 (2015): 317-319. Teschner, Sven, et al. AB0-incompatible kidney transplantation using regenerative selective immunoglobulin adsorption. Journal of clinical apheresis 27.2 (2012): 51-60. Tetala, Kishore K R, et al. Selective depletion of neuropathy-related antibodies from human serum by monolithic affinity columns containing ganglioside mimics. Journal of medicinal chemistry 54.10 (2011): 3500-3505. Vincent, Angela, et al. Serological and experimental studies in different forms of myasthenia gravis. Annals of the New York Academy of Sciences 1413.1 (2018): 143-153. Wallukat, Gerd, et al. Patients with preeclampsia develop agonistic autoantibodies against the angiotensin AT 1 receptor. The Journal of clinical investigation 103.7 (1999): 945-952. Zhou, Cissy C., et al. Angiotensin receptor agonistic autoantibodies induce pre-eclampsia in pregnant mice. Nature medicine 14.8 (2008): 855.