ANTIGEN-BINDING CONSTRUCTS THAT BIND TO AND INHIBIT CATALASE AND/OR SUPEROXIDE DISMUTASE AS WELL AS PHARMACEUTICAL COMPOSITIONS CONTAINING THEM FOR TUMOR THERAPY

Abstract

The invention relates to single domain VHH fragments which specifically bind to and inhibit superoxide dismutase and/or bind to and inhibit catalase and/or bind to and inhibit superoxide dismutase and catalase, in particular for the use in the therapy of tumor diseases.

Claims

1. An antigen-binding construct that specifically binds to superoxide dismutase or catalase or both superoxide dismutase and catalase, characterized in that said construct is a single domain VHH fragment for use in the treatment of tumor diseases.

2. The antigen-binding construct according to claim 1, wherein said construct (i) specifically binds to and inhibits superoxide dismutase; (ii) specifically binds to and inhibits catalase; and/or (iii) specifically binds to and inhibits both superoxide dismutase and catalase.

3. The antigen-binding construct according to claim 1, wherein said construct (i) specifically binds to superoxide dismutase but does not inhibit it; (ii) specifically binds to catalase but does not inhibit it; and/or (iii) specifically binds to and does not inhibit both superoxide dismutase and catalase.

4. The antigen-binding construct according to claim 1, wherein said construct further includes a marker.

5. The antigen-binding construct according to claim 1, wherein said construct can be connected to a cytotoxic agent.

6. The antigen-binding construct according to claim 1, wherein said construct is connected to a cytotoxic agent.

7. The antigen-binding construct according to claim 1, wherein said construct is connected to a component that changes the retention time of said construct in a target organism.

8. The antigen-binding construct according to claim 1, wherein said construct contains at least one CDR sequence selected from the group consisting of SEQ ID NOs: 15-51.

9. The antigen-binding construct according to claim 1, wherein said construct contains at least two CDR sequences selected from the group consisting of SEQ ID NOs: 15-51.

10. The antigen-binding construct according to claim 1, wherein said construct contains at least three CDR sequences selected from group consisting of SEQ ID NOs: 19-51.

11. A pharmaceutical composition containing at least one antigen-binding construct according to claim 1.

12. The pharmaceutical composition according to claim 11, wherein said antigen-binding construct specifically binds to and inhibits catalase.

13. The pharmaceutical composition according to claim 11, wherein said antigen-binding construct specifically binds to and inhibits superoxide dismutase.

14. The pharmaceutical composition according to claim 11, wherein said composition contains at least one antigen-binding construct that binds to and inhibits catalase and at least one construct that specifically binds to and inhibits superoxide dismutase.

15. The pharmaceutical composition according to claim 11, wherein said composition further includes a tumor therapeutic agent.

16. The pharmaceutical composition according to claim 11, wherein said composition is formulated for the treatment of tumor diseases.

17. The pharmaceutical composition according to claim 15, wherein tumor therapeutic agent is taxol.

Description

BRIEF DESCRIPTION OF THE FIGURES

[0030] The following diagrammatic representations first unite the apoptosis-inducing ROS signal paths in transformed cells (scheme 1: HOCI path; scheme 2: NO/peroxynitrite path) and then, show the tumor cell-specific effect of membranous catalase and SOD on these signal paths (Scheme 3 and 4).

[0031] FIG. 1: Scheme 1 (HOCI signal path) shows the membrane of a malignant transformed cell characterized by expression of NOX1 and DUOX (consisting of an NOX1-related NOX domain and a peroxidase domain). Here, the expression of NOX1 represents a specific feature of malignant cells, whereas DUOX can also be detected in normal cells. By the effect of a matrix metalloprotease (MMP) the peroxidase domain is released {circle around (1)}. The superoxide anions {circle around (2)} generated by NOX1 dismutate to H.sub.2O.sub.2 {circle around (3)} that is used by peroxidase as a substrate {circle around (4)}. Here, from the native peroxidase (PODFe.sup.III) there is formed the reactive intermediate “Compound I” (PODFe.sup.IV═O.sup.+) that is able to oxidize chloride ions to HOCI {circle around (5)}. The HOCI present in the micromolar range of concentrations only develops a toxic effect if it reacts with superoxide anions to form apoptosis-inducing hydroxyl radicals {circle around (6)}. Here, the decisive step is the lipid peroxidation by hydroxyl radicals {circle around (7)} that via the formation of ceramides results in the induction of the mitochondrial path of the apoptosis (not shown). If there is relative excess of H.sub.2O.sub.2 over peroxidase the reactions shown in {circle around (8)} and {circle around (9)} by consuming HOCI {circle around (8)} or preventing its synthesis a may lead to the termination of the HOCI signal path.

[0032] FIG. 2: Scheme 2 shows the NO/peroxynitrite signal path of transformed cells. NO Synthase (NOS) generates nitric oxide (NO) {circle around (1)}. A quite considerable part of the NO is converted by NO dioxygenase (NOD) to nitrate {circle around (2)}. Here, NOD is controlled by cytochrome P450-dependent oxidoreductase (POR). NO exhibits a high membrane permeability {circle around (3)} and can react with superoxide anions generated on the outside of the transformed cells to peroxynitrite (ONOO.sup.−) {circle around (4)}. The peroxynitrite acid (ONOOH) {circle around (5)} formed by protonation of peroxynitrite extremely rapidly decomposes to NO.sub.2 and apoptosis-inducing hydroxyl radicals {circle around (6)}. The reaction sequence {circle around (7)}-{circle around (12)} shows an alternative reaction option for NO that altogether represents a consumption reaction and can weaken or terminate the NO/peroxynitrite path. However, this consumption reaction represents an option to modulate H.sub.2O.sub.2-dependent processes by increasing the NO concentration.

[0033] FIG. 3: Scheme 3 shows that tumor cells effectively suppress the HOCI signal path {circle around (1)}, {circle around (2)}, {circle around (4)}-{circle around (5)} by expression of membranous catalase (CAT) by preventing the HOCI synthesis by destroying H.sub.2O.sub.2 {circle around (3)}. Certainly, membranous SOD promotes the dismutation of superoxide anions to H.sub.2O.sub.2, what could influence the HOCI path, but this effect does not bring results by the effect of catalase that degrades said H.sub.2O.sub.2. However, the SOD-mediated decrease in the concentration of superoxide anions inhibits the interaction between HOCI and superoxide anions {circle around (5)} that is essential for the signal path and in this way enhances the protection of the tumor cell from ROS signaling. The analysis of the inhibiting effects shows that the catalase on the membrane of tumor cells is sufficient for a complete and dominant inhibition of the HOCI synthesis, whereas certainly the SOD effect alone offers clearly measureable, but only partial protection.

[0034] FIG. 4: Scheme 4 summarizes the complex interaction of membranous catalase and SOD in the protection of tumor cells from the NO/peroxynitrite signal path. Catalase prevents the formation of peroxynitrite by oxidation of NO to NO.sub.2 {circle around (5)} and destroys possibly forming peroxynitrite by degradation {circle around (6)}. SOD prevents the formation of peroxynitrite by scavenging superoxide anions {circle around (4)} and also is able to destroy peroxynitrite {circle around (11)} to a limited extent. It may also be speculated that the H.sub.2O.sub.2 formed by SOD in reaction {circle around (10)} by the consumption reaction shown in scheme 2 contributes to an additional decrease in the NO concentration. The scheme impressively shows that tumor cells in a coordinated manner are capable to repeatedly control the NO/peroxynitrite path.

[0035] In the following diagrammatic representations 5 (FIG. 5) and 6 (FIG. 6) there are summarized the enzymatic details that are required for the understanding of the multiple effect of the protective enzymes catalase and SOD.

[0036] FIG. 5: Scheme 5 shows that the inhibition of the HOCI signal path {circle around (1)} by catalase is effected by a two-stage mechanism {circle around (2)}, {circle around (3)} in which the catalase intermediate “Compound I” (CATFe.sup.IV═O.sup.+) is of central importance and is re-formed to native enzyme. On the one hand, the inhibition of the NO/peroxynitrite path {circle around (4)} takes place by degradation of peroxynitrite via a two-stage mechanism with the participation of Compound I, on the other hand by oxidation of NO {circle around (8)}. Oxidation of NO is in an equilibrium with an NO-dependent inhibition of catalase, however that only takes effect at relatively high NO concentrations.

[0037] FIG. 6: Scheme 6 shows that the SOD-mediated dismutation of superoxide anions to H.sub.2O.sub.2 {circle around (1)}, {circle around (2)} is effected by a two-stage mechanism in which the reduction of the enzyme-bound Cu.sup.++ and oxidation of the Cu.sup.+ play a key role. The reaction of superoxide anions counteracts both the interaction between HOCI and superoxide anions {circle around (3)} that is essential for the HOCI path and the formation of peroxynitrite {circle around (4)}. Also, SOD has the potential to destroy peroxynitrite via reaction steps {circle around (7)}-{circle around (9)}.

[0038] Schemes 5 and 6 show that catalase and SOD, contrary to the textbook knowledge, are not characterized by highly selective reactions, but rather can execute multiple, partially overlapping functions. Altogether, this results in an outstanding plastic and complex biological effect.

[0039] FIG. 7: Scheme 7 represents the overall context between the enzymes acting on the outside of tumor cells. Membranous NADPH oxidase (NOX) generates the superoxide anions that are essential for both signal paths. Membranous SOD converts a substantial part of the superoxide anions to H.sub.2O.sub.2 {circle around (1)}, {circle around (2)}. This results in a partial inhibition of the formation of peroxynitrite a and to the partial inhibition of the interaction between superoxide anions and HOCI {circle around (4)}, wherein HOCI was formed by peroxidase (POD) {circle around (5)}. Additionally, SOD also decomposes peroxynitrite (sequence of reactions {circle around (6)}-{circle around (8)}). Catalase prevents HOCI synthesis by degradation of H.sub.2O.sub.2 {circle around (9)}, degrades peroxynitrite {circle around (10)} and prevents the formation of peroxynitrite by oxidation of NO {circle around (11)}.

[0040] FIG. 8: Scheme 8 is a diagrammatic representation of the conclusion that results from the facts summarized in Scheme 7, wherein a sole inhibition of membranous catalase {circle around (9)}, {circle around (10)} can result in a reactivation of the HOCI signal path and the NO/peroxynitrite path, since SOD only achieves a partial protection of steps {circle around (3)} and {circle around (4)}.

[0041] FIG. 9: Scheme 9 demonstrates that the assumption that an inhibition of SOD would not be sufficient for a reactivation of the apoptosis-inducing signal paths, since the dominant inhibiting effect of catalase should be opposed thereto, is incorrect. The key for the understanding of this initially unexpected finding is the inhibiting effect of superoxide anions on catalase. Since after the inhibition of SOD a local increase of the concentration of superoxide anions is observed (that is only counteracted by the spontaneous dismutation reaction) an indirect inhibiting effect on the catalase is achieved if SOD was only inhibited directly.

[0042] FIG. 10: Scheme 10 explains the enzymatic basis for the inhibition of catalase mediated by superoxide anions. On the one hand, superoxide anions are able to convert native catalase into the inactive Compound III (CATFe.sup.IIIO.sub.2) {circle around (1)}, on the other hand, to convert Compound I (CATFeIV=O..sup.+) by an one-electron transition into the inactive Compound II (CATFe.sup.IV═O) and in this way substantially inhibit the enzyme effect.

[0043] FIG. 11: Scheme 11 demonstrates that an SOD-mediated decrease in the concentration of superoxide anions below the concentration required for the inhibiting effect contributes to the establishment of the catalase activity.

[0044] FIG. 12: Scheme 12 demonstrates that the direct inhibition of SOD necessarily involves an indirect inhibition of catalase.

[0045] FIG. 13A shows the specific apoptosis induction in MKN-45 gastric carcinoma cells by single domain VHH fragments against catalase. FIG. 13B shows the induction of apoptosis that arises from the application under same conditions of recombinant Fab fragments (consisting of a light and a heavy chain) that are directed against and neutralize human catalase (Abd aCAT15562).

[0046] FIGS. 14A and 14B show that the catalase-neutralizing single domain VHH fragment aCATcb0972 in gastric carcinoma cells of the MKN-45 line induces specific ROS signaling via the NO/peroxynitrite and HOCI path.

[0047] FIG. 15A shows the specific apoptosis induction in MKN-45 gastric carcinoma cells by single domain VHH fragments against SOD. FIG. 15B shows the induction of apoptosis that arises from the application under same conditions of recombinant Fab fragments (consisting of a light and a heavy chain) that are directed against and neutralize human catalase (Abd aSOD15660).

[0048] FIGS. 16A and 16B shows that the SOD-neutralizing single domain VHH fragment aSODcb0989 in gastric carcinoma cells of the MKN-45 line induces specific ROS signaling exclusively via the NO/peroxynitrite path. FIG. 16A depicts the induction of apoptosis arising from single domain VHH fragment aSODcb0989 alone or in the presence of 100 μM of the NOX1 inhibitor AEBSF or 25 μM of the peroxynitrite scavenger FeTPPS. FIG. 16B depicts the induction of apoptosis arising from single domain VHH fragment aSODcb0989 alone or in the presence of 50 mM of the HOCI scavenger taurine (TAU) or 2 mM of the singlet oxygen scavenger histidine (HIS).

[0049] FIG. 17A shows that apoptosis induction by the single domain VHH fragments directed against catalase only with higher density runs in an optimum manner. FIG. 17B shows that when the application of the single domain VHH fragments is directed against SOD, there is much lesser attenuation of the effect when the target cells are present with a lower density.

[0050] FIGS. 18A and 18B show that the concentration-dependent apoptosis-inducing effects already measured in the preliminary experiments can be confirmed both by aCATcb0972 (FIG. 18A) and aSODcb0989 (FIG. 18B).

[0051] FIG. 19 shows that single domain VHH fragments neutralizing catalase or SOD, respectively, cause a strong synergistic effect with the established chemotherapeutic agent taxol.

[0052] FIG. 20 shows that the synergistic effect between single domain VHH fragments aCAT and aSOD can be concentrated in one hybrid molecule.

[0053] FIGS. 21A and 21 B show that the hybrid molecule from the single domain VHH fragments aCATcb0972 and aSODcb0989 in gastric carcinoma cells of the MKN-45 line induces specific ROS signaling exclusively via the NO/peroxynitrite path. FIG. 21A depicts the induction of apoptosis arising the hybrid molecule in the presence of 100 μM of the NOX1 inhibitor AEBSF or 25 μM of the peroxynitrite scavenger FeTPPS. FIG. 21B depicts the induction of apoptosis arising from the hybrid molecule in the presence of 50 mM of the HOCI scavenger taurine (TAU) or 2 mM of the singlet oxygen scavenger histidine (HIS).

[0054] FIG. 22 presents the results of siRNA-based analysis confirming the specificity of the ROS signaling that is induced by the hybrid molecule and results in apoptosis.

[0055] FIGS. 23A and 23B shows that catalase- and SOD-neutralizing single domain VHH fragments (FIG. 23A) as well as the hybrid molecule aCATaSOD (FIG. 23B) can induce concentration-dependent apoptosis in the neuroblastoma line SHEP.

[0056] FIG. 24 shows that the additional administration of the NO donor (that alone is not able to induce apoptosis) both results in a sensitization concerning the reactivation of the apoptosis induction and effectively counteracts the supra-optimum right-side drop of the optimum curve.

[0057] FIG. 25A shows that the tumor cell line MKN-45 is much better protected from H.sub.2O.sub.2 than the normal cells and the transformed cell line. FIG. 25B shows that, in the presence of catalase-neutralizing single domain VHH fragments, the tumor cells are very clearly sensitized for the effect of H.sub.2O.sub.2, whereas single domain VHH fragments that only bind to catalase do not result in a sensitization.

[0058] FIGS. 26A and 26B show that the neutralizing single domain VHH fragments caused a sensitizing effect on the transformed line that however could not influence normal cells.

[0059] FIG. 27A-27F shows that the siRNA-mediated knockdown of the intracellular catalase of normal cells (208F) (as compared to transformed cells (208Fsrc3) and tumor cells (MKN45)) increases their sensitivity against H.sub.2O.sub.2.

[0060] FIGS. 28A and 28B show that the growth of a human colon carcinoma xenotransplant on immunocompromised mice is inhibited by the repeated administration of the single domain VHH fragment aSODcb0989. FIG. 28A shows that, when applying 0.03 mg/kg of aSODcb0989, a recognizable difference to the controls does not result. FIG. 28B, however, shows that, when administering 0.3 mg/kg of aSODcb0989, the group treated with the single domain VHH fragment very clearly differs from the control group, even though both groups underlie a very strong spread (FIG. 28B).

[0061] FIG. 29 shows that the growth of a human colon carcinoma xenotransplant on immunocompromised mice is inhibited by the repeated administration of the single domain VHH fragment aSODcb0989.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0062] Within the scope of the present invention there were prepared and sequenced various preferred antigen-binding fragments by genetic engineering. The sequences are disclosed in the present application. Particularly important for the antigen-binding sequences are the CDR regions of the constructs. The so-called “complementary determining regions” (in the following briefly CDR) are very specific parts of the variable chains of the immunoglobulins. Said CDR regions are embedded within the framework sequence of immunoglobulins, determine their specificity, and establish contact with the specific antigen to which the immunoglobulins bind. The CDR regions are the most variable parts of the immunoglobulins and substantially contribute to the variety of these molecules. In immunoglobulins having a heavy and a light chain there are six CDR regions. However, if the immunoglobulin only consists of one chain, such as in case of the single domain VHH fragments that are preferred according to the invention, there are three CDR regions.

[0063] Generally, it is important that the CDR regions are almost present unchanged if the bonding specificity is to be maintained. However, it is possible that minor mutations do not adversely affect the functionality of the antibody-binding constructs. This is especially true if the structure of the CDR is not adversely affected by the exchange of an amino acid. Such amino acid exchanges are possible if the newly inserted amino acid is very similar to the replaced amino acid. Thus, in a preferred embodiment the antigen-binding fragments have the CDR regions that were disclosed within the scope of the present application or they differ at most in a smaller number of amino acids from the respectively disclosed CDR sequences that do not substantially change or reduced the bondability and bonding specificity.

[0064] Within the scope of the present invention CDR regions of antigen-binding single domain VHH fragments are disclosed that descent from such constructs that either inhibit or not catalase and/or superoxide dismutase, but do bind thereto. In a preferred embodiment the single domain VHH fragments (nanobodies) according to the invention contain at least one, preferably at least two and most preferably at least three CDR regions, wherein those CDR regions descending from constructs that inhibit catalase and/or superoxide dismutase are especially preferred.

[0065] The constructs according to the invention can preferably be humanized when they are intended for therapeutic application. Here, the framework sequence is replaced by a human framework substance or the non-human sequence is changed into a human sequence by mutations, but the bonding properties are to be maintained.

[0066] For the development of therapeutically usable biological molecules (here nanobodies) often modifications of the amino acid sequence are unavoidable. Since the molecules do not descend from humans, but originally from camelidae it is possible or likely that antibodies against exogenous epitopes are generated. Such antibody reactions would neutralize the effect of the antibody fragment to be used in therapy. To avoid these difficulties the therapeutically used molecules are humanized. The humanization of antibodies or antibody fragments is a technology that is well known in this special field. Typically, it is looked for humane framework sequences (backbone) that have the highest possible similarity to the original molecule. Then, the CDR regions are excised from the original nanobody and transplanted into the human sequence. Here, it is not inevitable that certain adaptions of amino acid sequences have to be made.

[0067] A substantial aspect is that the site on the antigen to which the binding part of the antibody binds is defined by the CDR sequences. During the humanization it might be required to slightly modify one or two of the three CDR sequences in order that the advantageous properties of the antigen-binding part are maintained. Thus, the single domain VHH fragments according to the invention are characterized in that they contain at least one of the CDR sequences, preferably two and especially preferably three CDR sequences, as disclosed in the present application.

[0068] Thus, the object of the present invention are single domain VHH fragments that have at least one, preferably at least two and especially preferably at least three of the following CDR regions characterized by the SEQ ID numbers. These are the sequences with SEQ ID numbers 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, as well as 30. Further preferred CDR sequences derived from clones binding to SOD are CDRs with SEQ ID numbers 31, 32, 33, 34, 35, 36, 37, 38, and 39. Also preferred are CDR sequences with SEQ ID numbers 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, and 51.

[0069] In a further preferred embodiment the constructs according to the invention can further be provided with a marker. For this employment, it is not necessary that the constructs also inhibit the target enzyme in its function. Rather, a specific bond with sufficient affinity is sufficient. Such a bond can be measured in a classical ELISA system in which either human catalase or SOD is bound to a suitable carrier and the complex of enzyme and single domain VHH fragment is detected by a common detection method. On the one hand, a marker can be used to mark the desired tumor cells in order to arrive at diagnostic statements here. On the other hand, the marker may also serve to label the tumor cells for other effectors. For example, as markers such constructs can be employed that connect themselves to effectors that induce apoptosis of the tumor cells. One example of such a marker is an anti CD3 part reacting with the CD3 receptor of cytotoxic T cells. Such an anti CD3 part may be for example an antigen-binding part of an antibody directed against the CD3 receptor.

[0070] In further embodiments of the present invention an antigen-binding construct of the present invention may also be configured such that it can be connected to a cytotoxic agent. As the cytotoxic agent, there are known for example various toxins that descend from various sources, for example bacteria, fungi, or plants. Preferably, this may be the cholera toxin, the botulinus toxin, or streptolysin, to name just a few.

[0071] In a further embodiment, the antigen-binding constructs according to the invention can be connected to a cytotoxic agent via different types and possibilities of bonding. These could also be solid chemical bonds, such as covalent bonds as well as ionic interactions or van der Waals forces.

[0072] In a further embodiment, it is also possible that an antigen-binding construct according to the invention can be or is tightly connected to a radioactive isotope. The radioactive isotope either may serve for the diagnostic detection of the tumor cells or enhance the cytotoxic effect of the antigen-binding construct according to the invention by bringing a radioactive isotope into close proximity to a tumor cell. Preferably, in the therapeutic application such isotopes are employed that only radiate relatively short in order to keep the side effects as low as possible. Moreover, the half-life of the isotope should be relatively short in order to keep the burden of the body and the environment in an acceptable range. Preferably employed are yttrium-90, rhenium-186 or erbium-169.

[0073] In a further embodiment, the antigen-binding constructs according to the invention can be connected to a colorant via various types and possibilities of bonding. These can be solid chemical bonds such as covalent bonds as well as ionic interactions or van der Waals forces. The colorant can be detectable by different common methods. Such constructs should be usable for diagnostic purposes.

[0074] In a further embodiment, pharmaceutical compositions are disclosed that contain at least one antigen-binding construct according to the invention. These pharmaceutical preparations are preferably used to treat tumor diseases, in particular to treat gastric carcinoma.

[0075] The pharmaceutical compositions according to the invention contain at least one antigen-binding construct that specifically binds to and inhibits catalase. In another embodiment, the pharmaceutical compositions according to the invention contain at least one antigen-binding construct that specifically binds to and inhibits superoxide dismutase. In a further embodiment, the pharmaceutical composition contains a hybrid molecule that binds both to catalase and superoxide dismutase and inhibits both target enzymes.

[0076] It is surprising and was not predictable that the synergistic effect of the hybrid molecules exclusively reactivates the NO/peroxynitrite path, whereas anti-SOD alone also reactivates this path, but anti-catalase induces the HOCI path. The exclusive reactivation of the NO/peroxynitrite path by the hybrid molecules represents an advantage for the future applications, because the less complex NO/peroxynitrite path should be further optimized by an additional modulation of the NO metabolism, especially in view of a desired width of the plateau phase of the dose-effect relationship.

[0077] A further advantage of the exclusive reactivation of the NO/peroxynitrite path by anti-SOD or hybrid molecules from anti-SOD and anti-CAT is that due to the signal chemistry (ill. 1-4) in the course of the NO/peroxynitrite path no free H.sub.2O.sub.2 should be present. This is desired because H.sub.2O.sub.2 has a proliferation-stimulating effect on surviving tumor cells that is detrimental to the therapeutic effect. The principle of said advantage of the therapeutic use of anti-SOD or of the hybrid molecule from anti-catalase and anti-SOD disclosed here was not disclosed in EP 11170076.1 and has a surprising effect. This effect indicates the presence of an inventive step.

[0078] In a further preferred embodiment of the present invention there is used at least one antigen-binding construct of the present invention together with an active ingredient having antitumor activity. Various active ingredients having antitumor effects are known. Especially, as chemotherapeutics that can also be present in the present composition there can be mentioned substances such as taxol, cisplatin, endostatin, oxaliplatin, etopside, or colchicine, to name only a few active ingredients as examples. Preferably, taxol is employed.

[0079] The antigen-binding constructs according to the invention are single domain VHH fragments that can be prepared by genetic engineering.

[0080] A precise molecular arrangement of the individual fragments in the antigen-binding construct is of subordinate importance, as long as the desired function, namely binding to and inhibiting catalase and/or superoxide dismutase is preserved.

[0081] The antigen-binding constructs according to the invention are prepared with molecular-biological means. These are not naturally occurring antibodies or Fab fragments prepared therefrom by simple (enzymatic) cleavage. For better understanding, the structure of antibodies might be briefly recapitulated. Antibodies, for example of the IgG type, consist of two Fab fragments and one Fc fragment. Each Fab fragment consists of a light and a heavy chain, wherein the heavy chain can be divided into a variable part (V.sub.H) and a constant part (C.sub.H1) and the light chain can be divided into a variable part (V.sub.L) and a constant part (C.sub.L). Of particular interest are the variable parts V.sub.H and V.sub.L that in turn contain the CDRs (complementarity determining regions) 1-6 that are relevant for antigen-bonding. The bonding properties of the Fab fragments are determined by the CDR regions that are embedded in a framework structure that spatially arranges the individual CDR regions.

[0082] The antigen-binding constructs are single domain VHH fragments (nanobodies). Nanobodies contain only the parts of the heavy chain of the antibody relevant for binding and can be very good expressed in bacterial cells (Muyldermans S. Nanobodies: Natural single-domain antibodies. Ann. Rev. Biochem. 82: 775-797, 2013).

[0083] With the help of various methods of genetic engineering a number of antigen-binding constructs can be prepared. The methods used for that are quite diverse and well known to the person of average skill in the art. Typically, here it is proceeded such that laboratory animals (mice, rats, rabbits, chickens or camels, alpacas etc.) are immunized with the desired antigen. Since camels and alpacas in addition to the conventional antibodies naturally also possess IgG that is exclusively constructed of heavy chains the use of these animals in combination with established selective screening methods results in obtaining single domain antibody encoding nucleic acid sequences. Then, from suitable immunocytes (for example B cells) nucleic acid sequences can be isolated that are further optimized with suitable methods, for example with the so-called phage display. Then, with these methods antigen-binding construct molecules are obtained that specifically bind to the desired antigen. In this context, specifically means that the constructs preferably only bind to the molecule sought, more particularly only to an epitope of said molecule (SOD or catalase). Non-specific cross-reactions are generally undesired.

[0084] Another important property of said antigen-binding constructs is that they sensitively bind to the desired antigen. Sensitively means that already at a very low concentration of the antigen-binding construct a specific binding to the desired antigen or the desired epitope takes place. Expressed in simplified terms, the better an antigen-binding construct binds to the target antigen the more sensitive it is. Since the single domain VHH fragments do not enclose a certain epitope like classical Fab fragments do, but due to their molecular structure bind to spatial recesses of the antigen there result substantial differences with respect to the detectability of certain epitopes by these two types of single domain VHH fragments.

[0085] The antigen-binding constructs according to the invention specifically bind to the superoxide dismutase and inhibit this enzyme. The inhibition of the target molecule superoxide dismutase is effected by the fact that the antigen-binding construct either binds to the catalytically active center of the superoxide dismutase (SOD) or in the proximity of this catalytic center, whereby a steric inhibition of the enzyme is effected. Then, superoxide anions (the typical and specific substrate of SOD) can no longer bind to the enzyme and can not catalytically be converted to H.sub.2O.sub.2 by it.

[0086] The same applies to catalase. The antigen-binding constructs according to the invention specifically and sensitively bind to catalase and inhibit it so that the enzymatic conversion of H.sub.2O.sub.2 into H.sub.2O+1/2 O.sub.2, or peroxynitrite into NO.sub.2.sup.− and 1/2 O.sub.2 is inhibited and the oxidation of NO by the active intermediate “Compound I” of the catalase is prevented.

[0087] In the present application the abbreviations given in the list below were used:

AEBSF 4-(2-aminoethyl)-benzenesulfonyl fluoride [0088] (inhibitor of the NADPH oxidase)
3-AT 3-aminotriazole [0089] (catalase inhibitor)
anti-CAT antibody against catalase
anti-SOD antibody against SOD
(For Reasons of Space in the Pictures the Designations Anti-CAT and Anti-SOD are Replaced by aCAT and aSOD)
CAT catalase
Compound I activated intermediate stage both of catalase of formula [0090] CAT Fe.sup.IV═O.sup.+. and peroxidase of formula POD Fe.sup.IV═O..sup.+. Compound I is formed in the reaction of catalase or peroxidase with one molecule hydrogen peroxide. Catalase can also form Compound I with one molecule peroxynitrite.
Duox dual oxidase [0091] (membranous enzyme consisting of a NADPH oxidase and a peroxidase domain. The peroxidase domain is cleaved off with the help of proteases.)
FBS fetal bovine serum
FeTPPS 5-, 10-, 15-, 20-tetrakis(4-sulfonatophenyl)porphyrinato iron(III) chloride [0092] (peroxynitrite decomposition catalyst)
NO nitric oxide
NOD nitric oxide dioxygenase (oxidizes NO to nitrate)
NOS NO synthase
NOX NADPH oxidase (here, in particular the membranous NOX-1)
POD peroxidase [0093] (in this context in particular the ability of certain peroxidases takes effect that in the presence of hydrogen peroxide they are able to oxidize chloride to HOCI)
PON peroxynitrite
POR cytochrome P 450 oxidoreductase
RAS, RAC oncogenes
ROS reactive oxygen and nitrogen species [0094] (radical and non-radical species such as superoxide anions, hydroxyl radicals, nitric oxide, hydrogen peroxide, HOCI, peroxynitrite, etc.)
siRNA small interfering RNA [0095] (reagent to specifically down-regulate the synthesis of defined gene products)
SOD superoxide dismutase [0096] (here, in particular SOD-1 (Cu.sup.++ in the active center of the tumor cells and MnSOD from bacteria for analytical purposes)
TGF-beta transforming growth factor type beta

EXAMPLES

[0097] The present invention is explained in detail by the following examples. The results of the experiments according to the invention are often illustrated in the figures. There are disclosed sequences of particularly preferred embodiments.

Example 1: Provision of Constructs and Materials

[0098] The following examples were carried out with the following antibodies, Fab fragments, or single domain VHH fragments:

1) Monoclonal antibody (mouse, IgG1) against human SOD-1 (clone SD-G6) (charge number 035K4823). Manufacturer Sigma Aldrich, Schnelldorf, Germany (as a control).
2) Recombinant human Fab fragment against human catalase, format Fab-V5Sx2, prepared by AbDSerotec from a HuCAL® Library (described in detail in EP 859 841 and U.S. Pat. No. 6,300,064). There was employed the construct #AbD15562 with catalase-inhibiting effect (comparison).
3) Recombinant human Fab fragment against human SOD, format Fab-V5Sx2, prepared by AbDSerotec from a HuCAL® Library. There was employed construct #AbD15660 with SOD-inhibiting effect (comparison).
4) Recombinant single domain VHH fragments against human catalase (according to the invention), prepared in cooperation with a commercial supplier.

[0099] The preparation was by immunizing alpacas with human catalase (catalase [EC 1.11.16] purified from human erythrocytes, obtained from Sigma (Schnelldorf), catalogue number C 3556) under the supervision of a veterinary, obtaining RNA from the B cells of the animals, reverse transcription, cloning in E. coli and isolation via phage display technology. Clones encoding for single domain VHH fragments that bind to human catalase were selected by testing supernatants in a suitable ELISA. In a second run, by employing the cell culture system described by Heinzelmann and Bauer (Heinzelmann S. and Bauer G. Multiple protective functions of catalase against intercellular apoptosis-inducing ROS signaling of human tumor cells, Biol. Chem. 391, 675-693, 2010) it was checked which one of the single domain VHH fragments binding to catalase actually results in an inhibition of the catalase, what is expressed as ROS-dependent apoptosis induction in the cells. There were used the single domain VHH fragments aCATcb0972, aCATcb0974, that both bind to human catalase and neutralize it, and aCATcb0973 and aCATcb0975, that bind to human catalase, but do not neutralize it. The clones underlying the single domain VHH fragments were sequenced by standard methods and the amino acid sequence was determined therefrom.

[0100] Preferred embodiments of the invention have the following sequences.

[0101] The sequences of the antigen-binding fragments were analyzed both for DNA and protein level and the antigen-binding regions (CDR) were determined. In the following only the amino acid sequences are described. These antigen-binding regions are substantial for the specificity of the antigen-binding fragments.

[0102] In a preferred embodiment of the present invention the antigen-binding constructs, especially the single domain VHH fragments or nanobodies, contain at least one CDR sequence, preferably at least two and most preferably three CDR sequences.

[0103] In the following, there are described the CDR sequences at the protein level. In the complete sequence the respective positions are given by underlining.

TABLE-US-00001 cb 0972 (binding to and neutralizing catalase): protein sequence (SEQ ID NO: 10): MAQVQLVESGGGLVQAGGSLRLSCAASERTFNTYGMGWFRQAPGKEREFV ATISWSGDSTYYADSVKGRFTISRDNAKNTMYLQMNSLKPEDTAVYYCNA NSEYGDSYWGQGTQVTVSSKKKHHHHHH CDR sequence at protein level: CDR1: RTFNTYGMG (SEQ ID NO: 19) CDR2: TISWSGDSTYYADSVKG (SEQ ID NO: 20) CDR3: NSEYGDSY (SEQ ID NO: 21) cb 0973 (binding to catalase, but not neutralizing): protein sequence (SEQ ID NO: 11): MAEVQLVESGGGLVQPGGSLRLSCAVSGFIFNTYSMRWGRQAPGKGLEWV SSISTGGYSTYADSVKGRFTISRDNAKNLVYLQMNSLKPEDTAVYYCGWG AFVRGERPQGQGTQVTVSSKKKHHHHHH CDR sequence at protein level: CDR1: FIFNTYSMR (SEQ ID NO: 22) CDR2: SISTGGYSTYADSVKG (SEQ ID NO: 23) CDR3: GAFVRGERP (SEQ ID NO: 24) cb 0974 (binding to and neutralizing catalase): protein sequence (SEQ ID NO: 12): MAQVQLVESGGGLVQPGGSLRLSCAASGSIFSIASMGWYRQAPGKQRDLV ATITSDGSTKYADSVKGRFTISRDNAKNTMYLQMNSVKPEDAAVYYCNAD ADDLEPGSYDYDYWGQGTQVTVSSKKKHHHHHH CDR sequence at protein level: CDR1: SIFSIASMG (SEQ ID NO: 25) CDR2: TITSDGSTKYADSVKG (SEQ ID NO: 26) CDR3: DADDLEPGSYDYDY (SEQ ID NO: 27) cb 0975 (binding to catalase, but not neutralizing): protein sequence (SEQ ID NO: 13): MAQVQLVESGGGLVQPGGSLRLSCAASASIFSIYVMAWYRQAPGKQRELV ATVTSGGATNYANSVKGRFTISRDNAKNTMDLQMNSLKPEDTAVYYCNAE DYYDYGLSRSKIYWGQGTQVTVSSKKKHHHHHH CDR sequence at protein level: CDR1: SIFSIYVMA (SEQ ID NO: 28) CDR2: TVTSGGATNYANSVKG (SEQ ID NO: 29) CDR3: EDYYDYGLSRSKIY (SEQ ID NO: 30)
5) Recombinant single domain VHH fragments against human SOD1, prepared in cooperation with a commercial supplier.

[0104] The preparation was by immunizing alpacas with human SOD1 (SOD1=Cu/ZnSOD [EC 1.15.1.1] purified from human erythrocytes, obtained from Sigma (Schnelldorf), catalogue number S 9636) under supervision of a veterinary, obtaining RNA from the B cells of the animals, reverse transcription, cloning in E. coli and isolation via the phage display technology. Clones encoding for single domain VHH fragments that bind to human SOD1 were selected by testing supernatants in a suitable ELISA. In a second run, by employing the cell culture system described by Heinzelmann and Bauer (Multiple protective functions of catalase against intercellular apoptosis-inducing ROS signaling of human tumor cells, Biol. Chem. 391, 675-693, 2010) it was checked which of the single domain VHH fragments binding to SOD actually results in an inhibition of SOD, what is expressed in the tumor system used as ROS-dependent apoptosis induction in the cells, since by the inhibition of SOD the concentration of free superoxide anions dramatically increases due to the absence of the enzymatic dismutation and results in a parallel indirect inhibition of catalase. In the following, this is expressed as ROS-dependent apoptosis induction. In a further control trial the specific inhibition of SOD by recombinant single domain VHH fragments was verified by the fact that the increasing effect of these fragments on apoptosis induction by exogenously added HOCI was examined, as described in Bauer 2013 (HOCI-dependent singlet oxygen and hydroxyl radical generation modulate and induce apoptosis of malignant cells. Anticancer Res 33: 3589-3602, 2013). There was used the single domain VHH fragment aSODcb0989 that binds to and neutralizes human SOD1, and the fragments aSODcb0987 and aSODcb0991 that bind to human SOD1, but do not neutralize it.

[0105] The clones underlying the single domain VHH fragments were sequenced by standard methods and the amino acid sequence was determined therefrom. Preferred embodiments of the invention have the following sequences:

TABLE-US-00002 anti-SOD VHHs: cb 0987 (binding to SOD, but not neutralizing): protein sequence (SEQ ID NO: 14): MAQVQLVESGGGIVQPGGSLRLSCVASESISEIDAMYWHRQAPGKERELV AGITNDGTRYYADSVKGRFTISRDNAKSTLYLQMNSLKFEDTAMYYCAAL PNPPPGYWGQGTQVTVSSKKKHHHHHH CDR sequence at protein level: CDR1: SISEIDAMY (SEQ ID NO: 31) CDR2: GITNDGTRYYADSVKG (SEQ ID NO: 32) CDR3: LPNPPPGY (SEQ ID NO: 33) cb 0989 (binding to and neutralizing SOD): protein sequence (SEQ ID NO: 15): MAQVQLVESGGGLVQSGGSLTLSCTASGFTISNYPMTWVRQAPGKGLEWV SRINSGGDRTLYADSVKGRFTVSRDNARNTMYLQMNNLKPEDTGLYFCAD SGAGWRYWGQGTQVTVSSKKKHHHHHH CDR sequence at protein level: CDR1: FTISNYPMT (SEQ ID NO: 34) CDR2: RINSGGDRTLYADSVKG (SEQ ID NO: 35) CDR3: SGAGWRY (SEQ ID NO: 36) cb 0991 (binding to SOD, but not neutralizing): protein sequence (SEQ ID NO: 16): MAQVQLVESGGGIVQPGGSLRLSCVASESISDIDAMYWHRQAPGKRRELV AGITNDGTEYFADSVKGRFAISRDNTKSSLYLQMNSLKLEDTAMYYCATL PNPPPGYWGQGTQVTVSSKKKHHHHHH CDR sequence at protein level: CDR1: SISDIDAMY (SEQ ID NO: 37) CDR2: GITNDGTEYFADSVKG (SEQ ID NO: 38) CDR3: LPNPPPGY (SEQ ID NO: 39)

[0106] By connecting clones cb 0972 (neutralizing catalase) and cb 0989 (neutralizing SOD) via a linker the bispecific hybrid single domain VHH fragments cb 1081 (anti-CATanti-SOD) and cb 1082 (anti-SODanti-CAT) were prepared by genetic engineering.

TABLE-US-00003 Bispecific anti-Catalase-SOD VHH cb 1081: protein sequence (SEQ ID NO: 17): MAQVQLVESGGGLVQAGGSLRLSCAASERTFNTYGMGWFRQAPGKEREFV ATISWSGDSTYYADSVKGRFTISRDNAKNTMYLQMNSLKPEDTAVYYCNA NSEYGDSYWGQGTQVTVSSGGGGSGGGGSGGGGSAQVQLVESGGGLVQSG GSLTLSCTASGFTISNYPMTWVRQAPGKGLEWVSRINSGGDRTLYADSVK GRFTVSRDNARNTMYLQMNNLKPEDTGLYFCADSGAGWRYWGQGTQVTVS SKKKHHHHHH CDR sequence at protein level: CAT CDR1: RTFNTYGMG (SEQ ID NO: 40) CAT CDR2: TISWSGDSTYYADSVKG (SEQ ID NO: 41) CAT CDR3: NSEYGDSY (SEQ ID NO: 42) SOD CDR1: FTISNYPMT (SEQ ID NO: 43) SOD CDR2: RINSGGDRTLYADSVKG (SEQ ID NO: 44) SOD CDR3: SGAGWRY (SEQ ID NO: 45) Bispecific anti-SOD-Catalase VHH cb 1082 protein sequence (SEQ ID NO: 18): MAQVQLVESGGGLVQSGGSLTLSCTASGFTISNYPMTWVRQAPGKGLEWV SRINSGGDRTLYADSVKGRFTVSRDNARNTMYLQMNNLKPEDTGLYFCAD SGAGWRYWGQGTQVTVSSGGGGSGGGGSGGGGSAQVQLVESGGGLVQAGG SLRLSCAASERTFNTYGMGWFRQAPGKEREFVATISWSGDSTYYADSVKG RFTISRDNAKNTMYLQMNSLKPEDTAVYYCNANSEYGDSYWGQGTQVTVS SKKKHHHHHH CDR sequence at protein level: SOD CDR1: FTISNYPMT (SEQ ID NO: 46) SOD CDR2: RINSGGDRTLYADSVKG (SEQ ID NO: 47) SOD CDR3: SGAGWRY (SEQ ID NO: 48) CAT CDR1: RTFNTYGMG (SEQ ID NO: 49) CAT CDR2: TISWSGDSTYYADSVKG (SEQ ID NO: 50) CAT CDR3: NSEYGDSY (SEQ ID NO: 51)

[0107] The NADPH oxidase inhibitor 4-(2-aminoethyl-benzenesulfonyl fluoride (AEBSF), the catalase inhibitor 3-aminotriazole (3-AT), the HOCI scavenger taurine, the singlet oxygen scavenger histidine, glucose oxidase (GOX) were obtained from Sigma (Schnelldorf, Germany). Peroxynitrite and the “peroxynitrite decomposition catalyst” (functional peroxynitrite scavenger) 5-, 10-, 15-, 20-tetrakis(4-sulfonatophenyl)porphyrinato iron(III) chloride (FeTPPS) were obtained from Calbiochem (Merck Biosciences GmbH, Schwalbach/Ts, Germany).

[0108] A precise description of these active ingredients is found in the publications Heinzelmann and Bauer (2010, Multiple protective functions of catalase against intercellular apoptosis-inducing ROS signaling of human tumor cells, Biol. Chem. 391, 675-693), and Bechtel and Bauer (2009, Catalase protects tumor cells against apoptosis induction by intercellular ROS signaling, Anticancer Res 29: 4541-4557).

Example 2: Gene Knockout with siRNAs

[0109] For the analysis described in FIG. 22 by employing the specific knockout of genes by means of the siRNA technique the following siRNAs (obtained from Qiagen, Hilden, Germany) were applied: [0110] A. control siRNA (“siCo”), (catalogue no. 1022076; sequence:

TABLE-US-00004 (SEQ ID NO: 1) r(UUCUCCGAACGUGUCACGU)dTdT (sense) (SEQ ID NO: 2) ACGUGACACGUUCGGAGAA)dTdT (antisense). [0111] The manufacturer has found that siCo does not influence the expression of any known gene. [0112] B. “High-performance validated siRNAs” for the knockdown of: FAS receptor (“siRNA FAS R.”)

TABLE-US-00005 (SEQ ID NO: 3) (Hs_FAS_7_HP validated siRNA, catalogue No. SI02654463; target sequence: AAG GAG TAC ACA GAC AAA GCC);
caspase-8 (“siRNA CASP8”)

TABLE-US-00006 (SEQ ID NO: 4) (Hs_CASP8_11_HP validated siRNA; catalogue No. SI02661946, target sequence: AAG AGT CTG TGC CCA AAT CAA);
caspase-9 (“siRNA CASP-9”)

TABLE-US-00007 (SEQ ID NO: 5) (Hs_CASP9_7_HP validated siRNA, catalogue No. SI02654610, target sequence: CAG TGA CAT CTT TGT GTC CTA);
C: HP siRNA against human NOX1 (“siRNA NOX1”);
target sequence:

TABLE-US-00008 (SEQ ID NO: 6) CCG ACA AAT ACT ACT ACA CAA
D: siRNA against human iNOS2 (siiNOS) [0113] target sequence:

TABLE-US-00009 (SEQ ID NO: 7) CTG GGC CGT GCA AAC CTT CAA

[0114] For the control examinations represented in FIG. 27 in addition to control siRNA there were also used siRNA against human catalase (for MKN-45 cells) [0115] Hs_CAT_4_HP siRNA, catalogue No. SI00027713 [0116] target sequence:

TABLE-US-00010 (SEQ ID NO: 8) CCG GAT CTC ACT TGG CGG CAA
and siRNA against murine catalase (for 208F and 208Fsrc3 cells) [0117] HP Mm_Cat_4_HP siRNA, catalogue No. SI00941976 [0118] target sequence:

TABLE-US-00011 (SEQ ID NO: 9) CCC AAT AGG AGA TAA ACT TAA.

[0119] The transfection technique by means of the siRNAs is described in detail in Heinzelmann and Bauer, 2010 (loc.cit.). The transfection efficiency was more than 95%. 24 hours after the transfection control examinations for the respective gene functions were carried out that made sure that the “functional knockdown” was complete. This means that the specific siRNAs had effectively suppressed the De-Novo synthesis of the analyzed gene products and that the natural degradation of the concentration of the gene products before the administration of siRNA took place up to below the detection limit.

[0120] The human gastric carcinoma line MKN-45 was kept in RPMI 1640 medium, supplemented with 10% inactivated fetal bovine serum and 40 U/ml penicillin, 50 μg/ml streptomycin, g/ml neomycin, 10 U/ml Moronal (antimycotic antibiotic agent) and 280 μg/ml glutamine. The human neuroblastoma line SHEP as well as normal rat fibroblasts (208F) and their offspring transformed by the src oncogene (208Fsrc3) were kept in eagle's minimum essential medium (EMEM) supplemented with 5% inactivated fetal bovine serum and 40 U/ml penicillin, 50 μg/ml streptomycin, 10 μg/ml neomycin, 10 U/ml Moronal and 280 μg/ml glutamine. Details on the cell lines and their culture are found in the works of Heinzelmann and Bauer, 2010 and Bechtel and Bauer, 2009.

Example 3: Standard Preparation for the Analysis of the Autocrine ROS-Controlled Apoptosis Induction in Tumor Cells

[0121] The experiments shown in FIG. 13-24 were carried out with a standard preparation for the apoptosis induction that is based on the method performed in Heinzelmann and Bauer 2010. The tumor cells used were taken from an optimally growing semi-sealed culture, centrifuged and placed in fresh medium. The test was performed in 96-well cell culture plates with 12 500 MKN-45 cells/100 μl medium or 10 000 SHEP cells/100 μl medium. MKN-45 cells grow in suspension, SHEP cells are adherent. The tests with SHEP cells were started as soon as the cells had grown on. By adding increasing concentrations of single domain VHH fragments that either inhibit catalase or SOD the protection of the tumor cells from their own ROS signaling was annulled so that based on the cell-owned extracellular superoxide anion production the NO/peroxynitrite signal path and the HOCI path (with MKN-45 cells) or in case of the SHEP cells only the NO/peroxynitrite path could start when the cells were incubated at 37° C. The analysis of the signal paths was carried out by the application of inhibitors of the NADPH oxidase (AEBSF), the HOCI scavenger taurine, the peroxynitrite scavenger FeTPPS and the singlet oxygen scavenger histidine. In some experiments for comparison instead of the single domain VHH fragments either recombinant Fab fragments or monoclonal antibodies were applied. This is indicated in the respective illustrations.

[0122] Double preparations were examined at the times shown in the text by means of phase contrast reverse microscopy for the percentage of apoptotic cells. Here, the classical apoptosis criteria described and documented in Heinzelmann and Bauer 2010 such as condensation of the nucleus, fragmentation of the nucleus, and membrane blebbing were used. Per single preparation at least 250 randomly selected cells were examined for the presence of apoptosis features.

[0123] Parallel control examinations, as e.g. documented in the work of Bauer et al. (Bauer G, Bereswill S, Aichele P and Glocker E. Helicobacter pylori protects oncogenically transformed cells from reactive oxygen species-mediated intercellular induction of apoptosis, Carcinogenesis 35: 1582-1591, 2014, Supplement) made sure that the applied morphological criteria correlated with apoptosis criteria such as DNA fragmentation (measured by the TUNEL reaction) or positivity for annexin V binding.

Example 4: Specific Sensitization of Tumor Cells for Apoptosis-Inducing ROS Signaling by Single Domain VHH Fragments Against Catalase

[0124] In the scope of the present invention many trials have been carried out with the results of the trials having been summarized in FIGS. 13 to 29, summarized above.

[0125] FIG. 13A shows the specific apoptosis induction in MKN-45 gastric carcinoma cells by single domain VHH fragments against catalase.

[0126] MKN-45 cells under standard conditions for autocrine apoptosis induction were mixed with the given concentrations of single domain VHH fragments that bind to but do not inhibit human catalase (aCATcb0973, aCATcb0975) and single domain VHH fragments that bind to and inhibit human catalase (aCATcb0972; aCATcb9074) and were further incubated for 3.5 hours at 37° C., 5% CO.sub.2. In parallel under same conditions recombinant Fab fragments (consisting of a light and a heavy chain) were applied that are directed against and neutralize human catalase (Abd aCAT15562; FIG. 13B). Thereafter, the percentage of apoptotic cells (double preparations) was determined in accordance with the above-mentioned classical apoptosis criteria.

[0127] FIGS. 13A and 13B show that only single domain VHH fragments inhibiting catalase, but not such that bind to the enzyme without inhibiting it induce apoptosis in the tumor cells. Here, the effect is represented in the form of an optimum curve, as has also been described in the work of Heinzelmann and Bauer, 2010 for the catalase inhibitor 3-aminotriazole. After the optimum curve has dropped there is a second increase in apoptosis induction. The addition of the recombinant Fab fragment Abd aCAT 15562 consisting of light and heavy chains also results in the apoptosis induction in the form of an optimum curve with a subsequent second increase, but to achieve the action optimum there is required a molar concentration of Abd aCAT 15562 that is about 250-fold higher in comparison to the catalase-inhibiting single domain VHH fragments. This is made evident by the superior efficacy of the constructs according to the invention.

[0128] The addition of single domain VHH fragments and classical Fab fragments inhibiting human catalase in the gastric carcinoma cell line MKN-45 results in the induction of apoptosis in the form of an optimum curve with respect to the concentration of the antibodies. The specificity of the induced process is made evident by the fact that single domain VHH fragments that bind to catalase but do not neutralize it do not result in an apoptosis induction (FIG. 13A, 13B). That is, for the sensitization it is not sufficient that antibodies only bind to the catalase. The specific inhibition of the function of catalase seems to be essential for the sensitization. Noteworthy and unexpected is the finding that the single domain VHH fragments have a much stronger apoptosis-inducing effect than classical Fab fragments. Due to the different structure of these two groups of fragments it would have been to be expected that the classical Fab fragments would have been to be applied with twice the concentration in comparison with single domain VHH fragments in order to achieve a similar effect. However, the difference makes up a factor of about 500 (on a molar basis of 250).

Example 5: Effect of a Single Domain VHH Fragment Binding Catalase and Inhibition in Gastric Carcinoma Cells

[0129] FIG. 14A shows that the catalase-neutralizing single domain VHH fragment aCATcb0972 in gastric carcinoma cells of the MKN-45 line induces specific ROS signaling via the NO/peroxynitrite and HOCI path.

[0130] To standard preparations for the induction of apoptosis the given concentrations of the single domain VHH fragments aCATcb0972 were added in the presence of 100 μM of the NOX1 inhibitor AEBSF, 50 mM of the HOCI scavenger taurine (TAU), 25 μM of the peroxynitrite scavenger FeTPPS or 2 mM of the singlet oxygen scavenger histidine (HIS). Control preparations were carried out parallel without inhibitors. After 3.5 hours at 37° C., 5% CO.sub.2 the percentages of apoptotic cells were determined.

[0131] FIGS. 14A and 14B show that the single domain VHH fragment aCATcb0972 (binds to and inhibits catalase) induces apoptosis in the form of an optimum curve and a subsequent second increase. Here, apoptosis induction is inhibited by the NOX1 inhibitor AEBSF in the entire range of concentrations of the single domain VHH fragment. The HOCI scavenger taurine does not inhibit in the left range of concentrations of the optimum curve, but then results in a strong inhibition of the apoptosis in the entire further range of the optimum curve (FIG. 14A). The second increase of the apoptosis in the concentration range from 6.2 pg/ml of the single domain VHH fragment is not inhibited by taurine. The peroxynitrite scavenger FeTPPS only in the left part of the optimum curve results in a strong inhibition and then, changes the optimum curve of apoptosis induction into a plateau curve. The singlet oxygen scavenger histidine does not result in an inhibition of apoptosis, but changes the optimum curve into a plateau curve.

[0132] That is, FIGS. 14A and 14B proves that the apoptosis-inducing effect of the single domain VHH fragment aCATcb0972 actually is caused by the reactivation of the ROS signaling, since at all applied concentrations a complete inhibition of the apoptosis induction takes place when the superoxide anion production is inhibited by AEBSF. Moreover, the result proven in FIGS. 14A and B show that the optimum range of the apoptosis induction (0-6.2 pg/ml) is characterized by a succession of NO/peroxynitrite path and HOCI path. In the range of 0.09 and 0.18 pg/ml of aCATcb0972 there is an inhibition by the peroxynitrite scavenger FeTPPS, but not inhibition by the HOCI scavenger taurine, what indicates the course of the NO/peroxynitrite path. From 0.75 pg/ml of the Fab fragment there is no substantial inhibition by FeTPPS, but a very strong inhibition by taurine, what proves the course of the HOCI path. As expected, at 0.37 pg/ml both signal paths overlap. The new increase of the apoptosis induction at higher concentrations of the single domain VHH fragment depends [0133] i) on the production of superoxide anions, [0134] ii) on the degree of the catalase inhibition and neither is affected by the HOCI scavenger nor the peroxynitrite scavenger.

[0135] Accordingly, it is a sole effect of H.sub.2O.sub.2 and not that of the specific ROS signal path.

[0136] Changing the optimum curve of the apoptosis induction into a plateau curve by the singlet oxygen scavenger histidine and the peroxynitrite scavenger FeTPPS indicates that at concentrations of the single domain VHH fragment aCATcb0872>0.75 pg/ml also singlet oxygen seems to play a role. Singlet oxygen can result from the reaction of H.sub.2O.sub.2 with peroxynitrite. Singlet oxygen is also known to be able to inactivate catalase. Then, the increased availability of H.sub.2O.sub.2 caused thereby can cause the side reactions {circle around (8)} and {circle around (9)} shown in FIG. 1, whereby HOCI is consumed or less HOCI is synthesized. In this way, a drop of apoptosis induction on the right side of the optimum curve is caused.

Example 6 Specific Sensitization of Tumor Cells for Apoptosis-Inducing ROS Signaling by Single Domain VHH Fragments Against SOD

[0137] The specific apoptosis induction in MKN-45 gastric carcinoma cells by single domain VHH fragments against SOD is shown in FIG. 15A. The percentage of apoptotic cells is plotted on the X axis.

[0138] MKN-45 cells under standard conditions for autocrine apoptosis induction were mixed with the given concentrations of single domain VHH fragments that bind to but do not inhibit human SOD1 (aSODcb0987, aSODcb0991) and single domain VHH fragments that bind to and inhibit human SOD (aSODcb0989) and were further incubated for 3.5 hours at 37° C., 5% CO.sub.2. In parallel under the same conditions recombinant Fab fragments (consisting of a light and a heavy chain) were applied that are directed against and neutralize human catalase (Abd aSOD15660; FIG. 15B). Thereafter, the percentage of apoptotic cells (double preparations) was determined in accordance with the above-mentioned classical apoptosis criteria.

[0139] FIGS. 15A and 15B show that single domain VHH fragments that inhibit SOD, but not such that only bind to the enzyme, induce apoptosis in the tumor cells. Here, the effect is represented in the form of a broad optimum curve. An optimum curve of the apoptosis induction is also achieved by the classical Fab fragment AbD aSOD 15660 (FIG. 15B), however, for that, in order to achieve the same effect, there is required a concentration that is 250-fold higher in comparison to that of the single domain VHH fragment.

[0140] That is, FIGS. 15A and 15B shows that it is not sufficient for the sensitization that antibodies bind to SOD. The specific inhibition of the function of SOD seems to be essential for the sensitization. It is noteworthy that the specific inhibition of the membranous SOD is sufficient to reactivate the ROS signaling that results in the apoptosis. Noteworthy and unexpected is the finding that neutralizing single domain VHH fragments have a much stronger apoptosis-inducing effect than classical neutralizing Fab fragments. Due to the different structure of these two groups of fragments it would have been to be expected that the classical Fabs would have been to be applied with twice the concentration in comparison with single domain VHH fragments in order to achieve a similar effect. However, the difference is a factor of about 250.

[0141] FIGS. 16A and 16B show that the SOD-neutralizing single domain VHH fragment aSODcb0989 in gastric carcinoma cells of the MKN-45 line induces specific ROS signaling exclusively via the NO/peroxynitrite path.

[0142] To the standard preparations for the induction of apoptosis the given concentrations of the single domain VHH fragment aSODcb0989 were added in the presence of 100 μM of the NOX1 inhibitor AEBSF, 25 μM of the peroxynitrite scavenger FeTPPS, 50 mM of the HOCI scavenger taurine (TAU) or 2 mM of the singlet oxygen scavenger histidine (HIS). Control preparations were carried out parallel without inhibitors. After 3.5 hours at 37° C., 5% CO.sub.2 the percentages of apoptotic cells were determined.

[0143] FIGS. 16A and 16B show that SOD-neutralizing single domain VHH fragments in gastric carcinoma cells of the MKN-45 line induce apoptosis in the form of a broad optimum curve which is followed by an indicated second increase. The apoptosis induction is inhibited by AEBSF in the entire concentration range. The entire optimum range is inhibited by FeTPPS, but not by taurine. Histidine results in a perceptible partial elimination of the drop of the right flank of the optimum curve.

[0144] That is, FIGS. 16A and 16B proves that apoptosis induced by aSODcb0989 in the tumor cells certainly is continuously induced via a superoxide anion-dependent process, but this (in contrast to aCATcb0972) exclusively is the NO/peroxynitrite path and HOCI signaling has no perceptible influence. This is also documented by the complete inhibition by means of FeTPPS and the disappearance of an inhibition by taurine. This result indicates that, given the absence of the SOD effect, obviously no sufficient H.sub.2O.sub.2 is available for the HOCI path. Since the NO/peroxynitrite path is very efficiently inhibited by membranous catalase it can be concluded from the result shown in FIGS. 16A and 16B that the inhibition of SOD by the single domain VHH fragment also must have resulted in an indirect inhibition of catalase. The different quality of the signal paths reactivated by aSODcb0989 and aCATcb0972 allows to exclude the theoretic assumption that the effect of aSODcb0989 could be a cross-reaction with catalase and thus also confirms the specificity of the effect of aSODcb0989.

Example 7: Relationship Between Single Domain VHH Fragments and Target Cells

[0145] FIGS. 17A and 17B show that single domain VHH fragments against catalase or SOD make different demands on the density of the target cells.

[0146] Preparations for the induction of apoptosis in addition to the standard cell density (12500 cells/100 μl) were also prepared with a lower cell density (4000 cells/100 μl) and increasing concentrations of the single domain VHH fragments aCATcb0972 or aSODcb0989 and incubated for four hours at 37° C., 5% CO.sub.2. Thereafter, the percentages of apoptotic cells were determined.

[0147] FIG. 17A shows that apoptosis induction by the single domain VHH fragments directed against catalase only with higher density runs in an optimum manner, whereas in the application of the single domain VHH fragments directed against SOD there is much lesser attenuation of the effect when the target cells are present with a lower density (FIG. 17B). This difference gives evidence for the different quality of the reactivated signal paths and especially is consistent with the fact that an inhibition of SOD preferably reactivates the NO/peroxynitrite path, since this does not depend on a high cell density to the same extent as the HOCI-signal path.

Example 8: Synergistic Effect of Single Domain VHH Fragments Against Catalase and SOD

[0148] FIGS. 18A and 18B show a noteworthy synergistic effect with the simultaneous application of single domain VHH fragments against catalase and SOD.

[0149] Standard preparations for the apoptosis induction with MKN-45 cells were mixed with increasing concentrations of the catalase-neutralizing single domain VHH fragment aCATcb0972 alone and in combination with 0.005 pg/ml of the SOD-neutralizing single domain VHH fragment aSODcb0989 or the single domain VHH fragment aSODcb0987 that binds to SOD but does not neutralize it (A). In the complementary experiment (B) mixing was done with increasing concentrations of aSODcb0989 alone or in combination with 0.005 pg/ml of the catalase-neutralizing single domain VHH fragment aCATcb0972. All preparations were incubated for 3.5 hours at 37° C., 5% CO.sub.2. Thereafter, the percentages of apoptotic cells were determined.

[0150] FIGS. 18A and 18B show that the concentration-dependent apoptosis-inducing effects already measured in the preliminary experiments can be confirmed both by aCATcb0972 (FIG. 18A) and aSODcb0989 (FIG. 18B). The combination with a minor concentration of the respectively complementary single domain VHH fragment that alone is not sufficient for the apoptosis induction (i.e. aSOD in increasing concentrations of aCAT and aCAT in increasing concentrations of aSOD) resulted in a noteworthy synergistic effect in the apoptosis induction. The specificity of said synergistic effect is proved by the fact that it has not appeared with the administration of a single domain VHH fragment that binds to but does not neutralize SOD (FIG. 18 A).

[0151] That is, FIG. 18 shows that the combination of single domain VHH fragments against catalase or SOD results in a very strong synergistic effect. When optimally using this effect the required concentrations of single domain VHH fragments could be drastically decreased.

Example 9: Synergistic Effect with Chemotherapeutic Agent

[0152] FIG. 19 shows that single domain VHH fragments neutralizing catalase or SOD, respectively, cause a strong synergistic effect with the established chemotherapeutic agent taxol.

[0153] Standard preparations for the apoptosis induction with MKN-45 cells were mixed with increasing concentrations of taxol alone or in combination with catalase-neutralizing aCATcb0972, SOD-neutralizing aSODcb0989 and SOD-binding but not neutralizing aSODcb0987 and incubated for 4 hours at 37° C., 5% CO.sub.2. Thereafter, the percentages of apoptotic cells were determined.

[0154] FIG. 19 shows that taxol induces apoptosis in the tumor cells in a concentration-dependent manner.

[0155] Here, the optimum effect was found in the concentration range of 1100 ng/ml. This effect was to be expected. In combination with single domain VHH fragments that either could inhibit catalase or SOD the taxol-dependent optimum curve surprisingly was drastically displaced to a lower concentration range. Now, the optimum of the effect was between 0.17 ng/ml and 0.5 ng/ml. This impressive displacement of the required concentration to less than a thousandth could only be achieved by single domain VHH fragments that also neutralize the respective target structure (catalase or SOD), while a mere bond did not cause any enhancing effect.

[0156] The effect shown in FIG. 19 at first is of high interest because it proves a previously unknown functional connection between the effect of taxol on tumor cells and ROS signaling. The appearance of a synergistic effect with single domain VHH fragments against SOD or catalase offers the opportunity to conserve resources in all by suitably combining the active ingredients, to thereby obtain a reduction in costs of the therapy and also opens the chance to reduce or even avoid the side effects caused by taxol and to avoid possible side effects of the single domain VHH fragments.

Example 10: Specific Sensitization of Tumor Cells for Apoptosis-Inducing ROS Signaling by Hybrid Molecules from Catalase- or SOD-Neutralizing Single Domain VHH Fragments

[0157] FIG. 20 shows that the synergistic effect between single domain VHH fragments aCAT and aSOD can be concentrated in one hybrid molecule.

[0158] Standard preparation for the apoptosis induction with MKN-45 cells were mixed with increasing concentrations of catalase-neutralizing aCATcb0972, SOD-neutralizing aSODcb0989, the hybrid molecules from aCATcb0972 and aSODcb0989 in the two possible arrangements as well as a neutralizing conventional monoclonal antibody against catalase (Sigma) as a control and incubated for 3.5 hours at 37° C., 5% CO.sub.2. Thereafter, the percentages of apoptotic cells were determined.

[0159] FIG. 20 shows that the synergistic effect observed with the combination of the individually added single domain VHH fragments aCATcb0972 and aSODcb0989 can actually be concentrated in hybrid molecules. Here, the arrangement aCATaSOD seemed to be gradually superior over the reverse arrangement. While the optimum effect of the individually applied single domain VHH fragments was in the concentration range of about 170 fg/ml, the optimum effect by the hybrid molecules was already achieved at 0.24 fg/ml. These noteworthy activities are conflicted with a dramatically lower efficacy of the conventional monoclonal antibody that develops its optimum effect at 111 ng/ml. In the illustration the data for classical recombinant Fab fragments that neutralize SOD or catalase have not been included for sake of clarity. Their optimum efficacy is in the range of 0.1-0.3 ng/ml.

[0160] That is, FIG. 20 confirms that [0161] i) the efficacy of single domain VHH fragments unexpectedly is much larger than that of conventional monoclonal antibodies or conventional recombinant Fab fragments and [0162] ii) a further very clear increase in the efficacy can be achieved when hybrid molecules are generated from single domain VHH fragments.

Example 11

[0163] FIGS. 21A and 21B show that the hybrid molecule from the single domain VHH fragments aCATcb0972 and aSODcb0989 in gastric carcinoma cells of the MKN-45 line induces specific ROS signaling exclusively via the NO/peroxynitrite path.

[0164] To standard preparations for the induction of the apoptosis there were added the given concentrations of the hybrid molecule aCATaSOD in the presence of 100 μM of the NOX1 inhibitor AEBSF, 25 μM of the peroxynitrite scavenger FeTPPS, 50 mM of the HOCI scavenger taurine (TAU) or 2 mM of the singlet oxygen scavenger histidine (HIS). Control preparations were carried out parallel without inhibitors. After 3 hours at 37° C., 5% CO.sub.2 the percentages of apoptotic cells were determined.

[0165] FIG. 21 confirms that the hybrid molecule aCATaSOD in the tumor cells induces apoptosis via the reactivation of the ROS signaling, because the process is completely prevented by the inhibition of the synthesis of superoxide anions by means of AEBSF. Further it is apparent that the apoptosis induction reactivated by the hybrid molecule is exclusively the NO/peroxynitrite path, since there is a complete inhibition by FeTPPS, but no inhibition by taurine. The singlet oxygen scavenger in turn has an attenuating effect on the right-side drop of the optimum curve. The dominance of the NO/peroxynitrite path was unexpected and unpredictable from the previous results.

Example 12

[0166] FIG. 22 by means of siRNA-based analysis confirms the specificity of the ROS signaling that is induced by the hybrid molecule and results in apoptosis.

[0167] MKN-45 cells were transfected with 24 nM siRNA that was directed against NOX1, iNOS2, caspase-9, FAS receptor or caspase-8. Control preparations were transfected with irrelevant control siRNA. After 24 hours at 37° C. the cells were taken up in fresh medium and mixed with the given concentrations of the hybrid molecule aCATaSOD. The percentage of apoptotic cells is determined after four hours.

[0168] FIG. 22 shows that the apoptosis induction in the form of an optimum curve caused by the hybrid molecule aCATaSOD can only take place when intact NADPH oxidase (NOX1), NO synthase (iNOS2) and caspase-9 that is essential for the course of apoptosis via the mitochondrial path are available. The siRNA-mediated knockdown of these enzymes each completely prevents the apoptosis induction induced by the hybrid molecule and thus, proves the underlying ROS signaling via the NO/peroxynitrite path with subsequent mitochondrial apoptosis signaling. On the other hand, apoptosis induction by the hybrid molecule does not require the FAS receptor and its downstream caspase-8, what emphasizes the exclusive effect of the mitochondrial apoptosis path and excludes that the apoptosis path via the death receptor APO/FAS for the apoptosis induction under the selected conditions plays a role. However, the knockdown of the FAS receptor and caspase-8 prevents the supra-optimum drop of the apoptosis induction curve. This can be explained by the participation of FAS receptor and caspase-8 and singlet oxygen-related processes, as stated in Bauer, 2012 and coincides with the finding regarding histidine established in FIG. 21B.

[0169] From FIGS. 13A, 13B, 15A, 15B and 20, there were determined the concentrations of the antibodies and single domain VHH fragments that were necessary for an optimum apoptosis induction in the tumor cells (table 1). This table explains the superior efficacy of the single domain VHH fragments in comparison to classical recombinant Fab fragments and monoclonal antibodies. Table 2 emphasizes the impressive synergistic effect that can be achieved by the use of the hybrid molecules. The superior effect of the single domain VHH fragments over classical recombinant Fab fragments was not predictable and thus, unexpected. Rather, due to the current state of knowledge it was to be expected that neutralizing classical Fab fragments and neutralizing single domain VHH fragments when considering the molar concentration should achieve the same effect on the apoptosis induction, when considering the concentrations (pg/ml) thereby only a difference by factor 2 should have to be achieved. This non-predictable dramatic difference in the efficacy is surprising. This effect is best suited for the therapeutic use.

TABLE-US-00012 TABLE 1 relation to opt. optimum conc. single domain relation Antibody conc. Fab (molar basis) aCAT cb0972 0.4 pg/ml (single domain VHH) AbD aCAT 15562 0.2 ng/ml    500   250 (recomb. Fab classical) monoclonal aCAT 111 ng/ml  277 500 46 250 (Sigma) aSOD cb0989 (single 1.2 pg/ml domain VHH) AbD aSOD 15660 0.3 ng/ml    250   125

TABLE-US-00013 TABLE 2 relation to opt. relation Antibody optimum conc. conc. hybrid Fab (molar basis) aCATaSOD 0.24 fg/ml (hybrid single dom. VHH) aCAT cb0972 0.17 pg/ml  708 1416 (single domain VHH) monoclonal aCAT  111 ng/ml 4.6 × 10.sup.8 1.54 × 10.sup.8 (Sigma) aSOD cb0989 0.35 pg/ml 1458 2916 (single domain VHH)

[0170] The values were taken from FIGS. 13A, 13B, 15A, 15B, and 20. First it was determined by which amount the concentration of conventional recombinant Fab fragments and monoclonal antibodies has to be higher to achieve the same effect as with recombinant single domain VHH fragments. By including the molar masses of the various Fab fragments and antibodies then the relation was determined on the basis of molarities. In this correction there was employed the valid assumption that classical Fab fragments make up a third of the molar mass of a complete IgG molecule and that the molar mass of a single domain VHH fragment makes up approximately half the molar mass of a classical Fab fragment.

Example 13: Effect of Single Domain VHH Fragments with Catalase- or SOD-neutralizing Effect on Human Tumor Cells that are Only Capable to Establish the NO/Peroxynitrite Path

[0171] While the human gastric carcinoma line MKN-45 used in the previous examples is characterized in that it can express the whole spectrum of the known intracellular ROS signaling (HOCI and NO/peroxynitrite path as main paths, nitryl chloride path as secondary path) when its membranous catalase is inhibited, in a series of other human tumor cell lines there is shown a limitation to the NO/peroxynitrite path (Heinzelmann and Bauer, 2010; Bauer, 2012). So, in the previous examinations it was found that a certain type of tumor each shows a uniform ROS signal system. A restriction to the NO/peroxynitrite signaling we so far only observed with neuroblastoma, Ewing's sarcoma, mammary carcinoma, ovarian carcinoma and small-cell lung carcinoma.

[0172] FIG. 23 shows that single domain VHH fragments can also reactivate apoptosis in tumor cells that can only form NO/peroxynitrite signaling.

[0173] 10 000 cells of the human neuroblastoma line SHEP per 100 μl medium were mixed with the given concentrations of the catalase-neutralizing single domain VHH fragment aCATcb0972, the SOD-neutralizing single domain VHH fragment aSOD0989, the SOD-binding but not neutralizing single domain VHH fragment aSODcb991 and the hybrid molecule aCATaSOD and incubated for 5 hours at 37° C., 5% CO.sub.2, before percentages of apoptotic cells were determined.

[0174] FIG. 23 shows that the catalase- and SOD-neutralizing single domain VHH fragments as well as the hybrid molecule aCATaSOD can induce concentration-dependent apoptosis in the neuroblastoma line SHEP. As expected, here the hybrid molecule proved to be much more effective than the single domain VHH fragments that each are only directed against one target structure. The SOD-binding but not neutralizing single domain VHH fragment aSODcb0991 at high concentrations also achieves an apoptosis-inducing effect, but for that there is required a 10.sup.6-fold higher concentration than for the neutralizing single domain VHH fragment.

[0175] That is, FIG. 23 confirms that also cell lines that after catalase inhibition are only capable for NO/peroxynitrite signaling can be brought into apoptosis by single domain VHH fragments directed against catalase or SOD. Also, here the synergistic effect between aCAT and aSOD appears. Noteworthy is the concentration-effect curve observed with this cell line that in comparison with the MKN-45 cell line only has a very flat long-lasting supra-optimum drop on the right flank.

[0176] The weak but significant apoptosis induction that is achieved by the single domain VHH fragment aSODcb0991 that can only bind but not neutralize can be explained best by the fact that after bonding of the single domain VHH fragment there is an internalization of the SOD and thus, its concentration on the surface is reduced, what should result in an effect analogous to the inhibition.

[0177] FIG. 24 shows that the effect of aSODcb0989 on SHEP cells can be further optimized when there is additionally employed the NO donor SNP.

[0178] 10 000 cells of the human neuroblastoma line SHEP per 100 μl medium were mixed with the given concentrations of the SOD-neutralizing single domain VHH fragment aSOD0989. Parallel preparations received 20 μM or 100 μM of the NO donor sodium nitroprusside or were incubated without a further additive. After 5 hours at 37° C., 5% CO.sub.2 the percentages of apoptotic cells were determined.

[0179] FIG. 24 shows that the additional administration of the NO donor (that alone is not able to induce apoptosis) both results in a sensitization concerning the reactivation of the apoptosis induction and effectively counteracts the supra-optimum right-side drop of the optimum curve. From this, modulation methods should be established that transform the optimum curve of the apoptosis induction into a plateau curve, what should result in a greater certainty of the therapeutic use.

Example 14: Single Domain VHH Fragments Exclusively Inhibit Membranous Catalase (that is Characteristic and Essential for Tumor Cells) and do not Achieve Intracellular Catalase (that Also Plays a Role in Normal Cells)

[0180] FIGS. 25A, 25B, 26A and 26B show that catalase-neutralizing single domain VHH fragments only neutralize the membranous catalase of tumor cells and cannot influence the intracellular catalase of normal cells.

[0181] 6000 non-transformed 208F cells, transformed 208Fsrc3 cells, and MKN-45 tumor cells, respectively, each were seeded in 100 μl medium and mixed with 0.1 or 1 pg/ml catalase-neutralizing aCATcb0972 or only catalase-binding aCATcb0973. Control preparations were left without single domain VHH fragments. Subsequently, the indicated concentrations of glucose oxidase were added and apoptosis induction was measured after 1.5 hours. Glucose oxidase (GOX) generates H.sub.2O.sub.2 which is cell-permeable and thus, can be degraded both by intracellular and membranous catalase. At a sufficient concentration H.sub.2O.sub.2 induces apoptosis without selectivity with respect to the malignant status of cells (Ivanovas et al., Selective and nonselective apoptosis induction in transformed and nontransformed fibroblasts by exogenous reactive oxygen and nitrogen species. Anticancer Research, Anticancer Res. 22:841-856, 2002).

[0182] FIG. 25A shows that the tumor cell line MKN-45 is much better protected from H.sub.2O.sub.2 than the normal cells and the transformed cell line. In the presence of catalase-neutralizing single domain VHH fragments the tumor cells are very clearly sensitized for the effect of H.sub.2O.sub.2, whereas single domain VHH fragments that only bind to catalase do not result in a sensitization (FIG. 25B). FIGS. 26A and 26B show that the neutralizing single domain VHH fragments caused a sensitizing effect on the transformed line that however could not influence normal cells.

[0183] FIGS. 25A, 25B, 26A and 26B confirm the specific effect of the single domain VHH fragments on the malignant cells, whereas the normal cells are not influenced. Here, the greatest effect was observed with the tumor cells, since these, above all are protected by membranous catalase. Transformed cells are known to bear detectable amounts of catalase on the surface, that however is present in a lower local concentration than in tumor cells and thus, is not sufficient for the protection from ROS signaling. The lack of reaction in the normal cells proves that the single domain VHH fragments do not penetrate the cell and cannot influence the catalase that is present there. That is, they act specifically on the membranous catalase, as is especially characteristic for tumor cells. However, this very important statement about the reaction site of the single domain VHH fragments is only significant if at the same time it can be proved that an inhibition of the intracellular catalase of normal cells actually would have had influence on their sensitivity against H.sub.2O.sub.2.

[0184] With this control aspect is dealt in FIG. 27.

[0185] FIG. 27A-27F shows that the siRNA-mediated knockdown of the intracellular catalase of normal cells increases their sensitivity against H.sub.2O.sub.2.

[0186] Normal cells (208F), transformed cells (208Fsrc3) and tumor cells (MKN-45) were transfected with control siRNA (siCo) and siRNA against catalase (siCAT) and kept for 24 hours at 37° C. and 5% CO.sub.2. Thereafter, the cells were taken up into fresh medium and taken up in a cell density of 6000 cells/100 μl medium. Subsequently, the preparations were treated either with increasing concentrations of GOX (27A-27C) or peroxynitrite (PON) (27D-27F). After 1.5 hours of incubation and 37° C. and 5% CO.sub.2 the percentages of apoptotic cells were determined. To assess this experiment, it has to be recapitulated that GOX generates H.sub.2O.sub.2 that has a very good cell-permeability and thus, can be degraded both by membranous and intracellular catalase. On the other hand, exogenously added peroxynitrite reacts with the cell membrane when it contacts the cell. Thus, protection from the effect of peroxynitrite can only be achieved by catalase sitting on the outside of the membrane.

[0187] At first, FIG. 27A-27F confirm that tumor cells are much better protected from H.sub.2O.sub.2 and peroxynitrite than normal or transformed cells. Further, the figure shows that a siRNA-mediated degradation of catalase results in a very strong sensitization of normal cells, transformed cells, and tumor cells against the effect of H.sub.2O.sub.2. After treatment with peroxynitrite there appeared a completely different picture: now, the siRNA-mediated degradation of catalase in the normal cells does not result in a sensitization, whereas in the transformed cells and tumor cells there is a strong sensitization. This shows that only on the surface of the malignant cells, i.e. the transformed and tumor cells, there is catalase that protects from exogenous peroxynitrite. A protective membranous catalase effect cannot be detected in the normal cells. The finding made by means of H.sub.2O.sub.2-generating GOX shows that normal cells possess functional protective catalase the effect of which can be detected e.g. by a siRNA-mediated knockdown. Since in the experiment shown in FIGS. 26A and 26 B however a sensitization of normal cells against H.sub.2O.sub.2 cannot be detected by means of catalase-neutralizing single domain VHH fragments it is proved that they could not act inside the cell.

Example 15: The In Vitro Effect Shown for Single Domain VHH Fragments Correlates with an Inhibition of the Tumor Growth In Vivo

[0188] FIGS. 28A, 28B and 29 show that the growth of a human colon carcinoma xenotransplant on immunocompromised mice is inhibited by the repeated administration of the single domain VHH fragment aSODcb0989.

[0189] The experiments carried out in FIGS. 28A and 28B and FIG. 29 were carried out by a commercial certified supplier (Oncotest GmbH, Freiburg). Here, xenotransplants of a human colon carcinoma were implanted in suitable mice. After the tumors had grown on and a certain minimum size had been achieved, the therapeutic test was started. aSODcb0989 in the indicated doses (mg/kg body weight) or buffer were intravenously applied twice a week. Also, the tumor sizes were measured twice a week by means of a caliber.

[0190] At first, FIGS. 28A and 28B show that the tumor growth in the control animals is characterized by a very clear spread. When applying 0.03 mg/kg of aSODcb0989 there does not result a recognizable difference to the controls (FIG. 28A). When administering 0.3 mg/kg of aSODcb0989 the group treated with the single domain VHH fragment very clearly differs from the control group, even though both groups underlie a very strong spread. The illustrated difference in the tumor volume between the control group and treated group is the expression of a clear proliferation inhibition by aSODcb0989. When further increasing the dose of aSODcb0989 to 0.9 mg/kg (FIG. 29) in some animals the growth inhibition is enhanced, whereas in other animals there is caused the contrary. This shows that also in vivo a dose-effect curve in the form of an optimum is present and the conditions in the experiment shown in FIG. 29 have reached the limit of the concentration for the supra-optimum inhibition.