Antibody architecture

Five major classes of antibodies (immunoglobulins, Igs) have been characterized: IgM, IgG, IgA, IgD and IgE). Immunoglobulins of all classes display a similar basic four-chain structure consisting of two identical light (L) chains and two identical heavy (H) chains (Figure 13.B2). The overall structure is held together by disulfide linkages and non-covalent interactions. Different H chain types are present in immunoglobulins of different classes. In addition, some classes can be further subdivided into subclasses (isotypes) based upon more subtle differences. Thus, human IgG can be subdivided into IgG1, IgG2, IgG3 and IgG4. Murine IgG can be subdivided into IgG1, IgG2a, IgG2b and IgG3.

In their native conformations, each immunoglobulin chain is seen to be composed of discrete domain structures, stabilized by intrachain disulfide linkages (not shown below). Each domain contains approximately 110 amino acid residues. H chains and L chains contain both variable (V) and constant (C) domains. Variable regions house the actual antigen-binding site of the antibody. Variable regions of antibodies displaying different (antigen-binding) specificities differ in amino acid sequence. Constant regions (within any one antibody class/subclass) do not. L chains contain one variable (VL) and one constant (CL) domain. H chains contain one variable (VH) and three constant (CH1, CH2 and CH3) domains. In addition, H chains display a single short sequence joining CH1 and Ch2. This is the flexible hinge (H) region, which contains several proline residues.

Treatment with certain proteolytic enzymes (e.g. papain) results in cleavage of the immu-noglobulin at the hinge region, yielding two separate antigen-binding fragments (2 X F(ab)), and a constant fragment (Fc). The Fc region mediates the various antibody effector functions. Fab fragments, although retaining their antigen-binding properties, are no longer capable of precipitating antigen in vitro. However, immunoglobulin incubation with other proteases (e.g. pepsin) results in antibody fragmentation immediately below the hinge region. This leaves intact two interchain disulfide linkages towards the C-terminus of the hinge region. This holds the two antigen-binding fragments together. The products of this fragmentation are donated F(ab)2 and Fc. Because of its bivalent nature, F(ab)2 retains the ability to precipitate antigen in vitro.

FV fragments consist of VH and VL domains, and can easily be produced by recombinant DNA technology (as can other antibody fragments). Two FV domains can be stabilized by the introduction of an interchain covalent linkage (e.g. a disulfide linkage or via direct chemical coupling). 'Single-chain' FV fragments may also be generated by the introduction of a short peptide linker sequence between the two FV domains.

Selected regions within the antibody's variable domain display greater variability in amino acid sequence (from one antibody to another) than do other variable regions. These so-called 'hypervariable' regions (complementarity-determining regions (CDRs)) are brought into close proximity upon antibody folding into its native conformation, and represent the antigen binding sites. The remaining areas of the variable domain are termed framework regions.

Immunoglobulins are glycoproteins. The carbohydrate moiety is attached to the heavy chain (usually the CH2 domain) via an N-linked glycosidic bond. Removal of the carbohydrate group has no effect upon antigen binding, but it does affect various antibody effector functions and alters its serum half-life.

H

H

CH2

CH2

ch3

ch3

Figure 13.B2 IgG structure

= Carbohydrate

Figure 13.B2 IgG structure

13.3.3.1 Antibody-based strategies for tumour detection/destruction

Clear identification of tumour-associated antigens would facilitate the production of monoclonal antibodies capable of selectively binding to tumour tissue. Such antibodies could be employed to detect and/or destroy the tumour cells.

The antibody preparations could be administered unaltered or (more commonly) after their conjugation to radioisotopes or toxins. Binding of unaltered monoclonal antibodies to a tumour surface alone should facilitate increased destruction of tumour cells (Figure 13.4). This approach, however, has yielded disappointing results, as the monoclonal antibody preparations used to date have been murine in origin. The Fc region of such mouse antibodies is a very poor activator of human immune function. Technical advances, allowing the production of human/humanized monoclonals (see later) may render this therapeutic approach more attractive in the future.

Several clinical trials have evaluated (or continue to evaluate) monoclonal antibodies to which a radioactive tag has been conjugated. These are usually employed as potential anti-cancer agents. The rationale is selective delivery of the radioactivity directly to the tumour site. Most of the radioisotopes being evaluated are P-emitters. These include istopes of iodine (125I, 131I), rhenium (186Re, 188Re) and yttrium (90Y). The medium-energy radioactivity these emit is capable of penetrating a thickness of several cells. Congregation of radioactivity at the tumour surface could thus promote irradiation of several layers of tumour cells, as well as nearby healthy cells. Higher energy a-emitters are also being evaluated. Although their effective path length is only about one cell deep, each emission has a greater likelihood of killing all cells in its path.

Figure 13.4 Binding of appropriate antibody to tumour-associated antigens marks the tumour cell for destruction. This is largely due to the presence of a domain on the antibody Fc region (see also Box 13.2), which is recognized and bound by macrophages and NK cells. Therefore, congregation of such cells on the surface of the tumour is encouraged. This greatly facilitates their cytocidal activity towards the transformed cells

Figure 13.4 Binding of appropriate antibody to tumour-associated antigens marks the tumour cell for destruction. This is largely due to the presence of a domain on the antibody Fc region (see also Box 13.2), which is recognized and bound by macrophages and NK cells. Therefore, congregation of such cells on the surface of the tumour is encouraged. This greatly facilitates their cytocidal activity towards the transformed cells

An allied application of radiolabeled anti-tumour monoclonal antibodies is that of diagnostic imaging (immunoscintigraphy). In this case, the radioisotope employed must be a y-emitter (such that the radioactivity can penetrate outward through the body for detection purposes). Although various radioisotopes of iodine have been evaluated, technetium (99mTc) is the one most commonly employed. It has a y-ray emission energy that is sufficient, but a relatively short half life of 6 h (this minimizes long-term exposure of patient to high-energy y-rays). It can also be generated at nuclear installations relatively easily and is inexpensive. Equally important, chemical methodologies exist which facilitate its (stable) coupling to antibody molecules. Direct labelling with 99mTc generally entails initial reduction of antibody disulfide residues, forming free sulfahydryl (—SH) groups. This is achieved by incubation with a suitable reducing agent, such as ascorbic acid or sodium dithionite. A source of 99mTc (e.g. Na99mTcO4) is then reduced separately. Subsequent mixing under nitrogen gas (to maintain reducing conditions) results in direct linkage of the radioisotope to the antibody.

Upon administration, the anti-tumour 99mTc conjugate will congregate at the tumour site. The tumour can then be visualized using suitable y-ray detection equipment, such as a planar gamma camera.

A number of radiolabelled monoclonal antibodies have been approved as tumour diagnostic imaging agents (Table 13.2). Carcinoembryonic antigen (CEA)-SCAN, for example, is an antigen-binding fragment (Fab) of a specific murine monoclonal raised against human CEA. As discussed in detail in Section 13.3.4, CEA is expressed at high levels by some tumours. This is particularly true of tumours of the gastrointestinal tract, such as carcinomas of the colon or rectum. CEA-SCAN (non-proprietary name: Arcitumomab) is used to detect these carcinomas. However, CEA is expressed naturally (all be it at much lower levels) by some non-transformed cells. Therefore, this antibody fragment is used mostly to image recurrence and/or metastases of histologically demonstrated carcinoma of the colon or rectum. It is used as an adjunct to standard imaging techniques, such as a computed tomography scan or ultrasonography. Its industrial method of production is overviewed in Figure 13.5. The product is administered by i.v. injection, and only relatively mild side effects are usually noted. These can include nausea, fever, rash and headaches.

Anti-tumour monoclonal antibodies can also be used to deliver toxins to tumour sites. Toxins conjugated to therapeutic antibodies include ricin, pokeweed toxin, Pseudomonas toxin and

Figure 13.5 Outline of the production strategy of CEA-SCAN. The antibody-producing hybridoma cell line was originally obtained by standard methods of hybridoma generation. Spleen-derived murine B-lymphocytes were fused with murine myeloma calls. The resulting stable hybridomas were screened for the production of anti-CEA monoclonals. The clone chosen produces an IgG anti-CEA antibody. Note that the finished product outlined above is not radiolabelled. The freeze-dried antibody preparation (which has a shelf life of 2 years at 2-8 °C) is reconstituted immediately prior to its medical use. The reconstituting solution contains 99mTc, and is formulated to facilitate direct conjugation of the radiolabel to the antibody fragment

Figure 13.5 Outline of the production strategy of CEA-SCAN. The antibody-producing hybridoma cell line was originally obtained by standard methods of hybridoma generation. Spleen-derived murine B-lymphocytes were fused with murine myeloma calls. The resulting stable hybridomas were screened for the production of anti-CEA monoclonals. The clone chosen produces an IgG anti-CEA antibody. Note that the finished product outlined above is not radiolabelled. The freeze-dried antibody preparation (which has a shelf life of 2 years at 2-8 °C) is reconstituted immediately prior to its medical use. The reconstituting solution contains 99mTc, and is formulated to facilitate direct conjugation of the radiolabel to the antibody fragment diphtheria toxin. After binding to the cell surface, the antibody-toxin conjugate is often internalized via endocytosis. It is presumed that, rather than being destroyed, the toxin is subsequently made available inside the cell, such that it can induce its toxic effects. One such antibody-based product now approved for general medical use is Mylotarg (Table 13.2). The product consists of an engineered antibody (a 'humanized' antibody, as described later), conjugated to a cytotoxic anti-tumour antibiotic, calicheamicin (Figure 13.6). The antibody binds specifically to a cell surface antigen, CD33. This is a sialic-acid-dependent adhesion protein found on the surface of leukaemic cells in more than 80 per cent of patients suffering from acute myeloid leukaemia. The product production process entails initial culture of the antibody-producing mammalian cell line with subsequent purification of the antibody by a series of chromatographic steps. Downstream processing ch3

OCH3 OH

CH2CH3 J

och3

Figure 13.6 Schematic diagram of the antibody based product Mylotarg, with emphasis upon the toxin's chemical structure. In reality three to five molecules of toxin are attached to each antibody molecule

OCH3 OH

CH2CH3 J

och3

Figure 13.6 Schematic diagram of the antibody based product Mylotarg, with emphasis upon the toxin's chemical structure. In reality three to five molecules of toxin are attached to each antibody molecule

also incorporates ultrafiltration and low-pH incubation steps designed to remove/inactivate any virus potentially present. The cytotoxic antibiotic is obtained separately via fermentation of its producer microrganism, Micromonospora echinospora sp. calichensis. Direct chemical linkage of antibiotic to antibody is achieved using a bifunctional linker.

Mylotarg administration results in congregation of the antibody-toxin conjugate on the surface of (CD33 positive) leukaemic cells. Binding triggers internalization of the conjugate. Lysosomal degradation ensues, but a significant proportion of the intact antibiotic escapes and induces its cytotoxic affect by binding DNA in its minor groove. This, in turn, induces double-strand breakage.

Mylotarg (like most other drugs) does induce some side effects, the most significant of which is immunosuppression. This is induced because certain additional (non-cancerous) white blood cell precursors also display the CD33 antigen on their surface. The immunosuppressive effect is reversed upon termination of treatment, as pluripotent haematopoietic stem cells (Chapter 10) are unaffected by the product.

13.3.3.2 Drug-based tumour immunotherapy

In addition to tumour-selective delivery of toxins and radioisotopes, antibodies may also be used to mediate tumour-targeted drug delivery. At its simplest, this involves conjugation of a chemo-therapeutic drug to a tumour-specific antibody. Therapeutic drugs used include adriamycin, ami-nopterin, methotrexate and vinca alkaloids. This direct approach to tumour drug delivery has met with some success, mainly in animal studies. However, a limited number of drug molecules can be conjugated to each antibody molecule, thus somewhat limiting the drug delivery load.

An alternative approach is the use of a tumour-specific antibody to which a prodrug activating enzyme has been attached. Therapeutically inactive prodrugs could be administered by,

Prodrug U (inactive) p

Prodrug U (inactive) p

Key:

^-^^ = (Prodrug activating) enzyme-antibody conjugate

^ I = Tumor-associated antigen

Figure 13.7 Outline of antibody-directed enzyme prodrug therapy (ADEPT). Subsequent to its enzymatic activation, the active drug is taken up by the cell, upon which it exhibits a cytocidal effect. Refer to text for specific detail for example, i.v. injection. This would subsequently be activated only at the tumour surface (Figure 13.7). This approach has been termed ADEPT or antibody-directed catalysis.

Because of its catalytic nature, a single antibody-enzyme conjugate would activate many molecules of the prodrug in question. Much of the active cytocidal agent released at the tumour surface would be taken up by the tumour cells via simple diffusion or carrier-mediated active transport.

Administration of etoposide in prodrug form exemplifies this approach (Figure 13.8). Etoposide (C29H32O13; molecular mass 588.6) is a semi-synthetic derivative of podophyllotoxin, produced naturally by the North American plant Podophyllum peltatum. It is used as an anti-cancer agent. Its cellular uptake is diffusion dependent, and once inside the cell it exerts its cytocidal effect. Phosphorylated etoposide is non-diffusable and, hence, represents an inactive prodrug form of etoposide. (Attachment of a charged group to most diffusion-dependent drugs prevents their cellular uptake.) Alkaline phosphatase, however, can cleave the phosphate group, releasing free cytocidal agent. Administration of a tumour-detecting antibody-alkaline phosphatase conjugate thus effectively targets the enzyme to the tumour surface. Subsequent administration of phos-phorylated etoposide results in etoposide liberation at the tumour surface, which can then enter tumour cells by diffusion (Figure 13.8). Various other prodrug-enzyme combinations have now been developed, including phenoxyacetamide derivatives of doxorubicin (activated by penicillin amidase) and 5-fluorocytosine (activated by cytosine deaminase).

The prodrugs used should be inexpensive, readily available and should be stable to chemical/enzymatic degradation in vivo. Enzymes used should also be stable under physiological conditions, display a reasonable turn over number in vivo and not be dependent upon a co-factor for activity. Mammalian enzymes would be likely less immunogenic than microbial enzymes. However, the use of a prodrug capable of being activated by a mammalian enzyme can lead to complications if that enzyme's (human) endogenous counterpart is capable of activating the drug at sites distant from the tumour.

= Tumor-asociated antigen (TAA) Figure 13.8 The etoposide-alkaline phosphatase ADEPT system. Refer to text for specific details 13.3.3.3 First-generation anti-tumour antibodies: clinical disappointment

Despite the scientific elegance of the antibody-mediated approach to tumour detection/destruction, initial clinical trials proved disappointing. A number of factors contributed to their poor therapeutic performance, particularly against solid tumours. Most such factors relate directly/ indirectly to the fact that the first generation of such drugs utilized whole monoclonal antibody preparations of murine origin. These factors include:

• insufficient information exists regarding tumour antigens;

• murine monoclonals prompt an immune response when administered to humans;

• poor penetration of tumour mass by antibody;

• murine monoclonals display a relatively short half-life when administered to humans;

• poor recognition of murine antibody Fc domain by human effector mechanisms.

Supplements For Diabetics

Supplements For Diabetics

All you need is a proper diet of fresh fruits and vegetables and get plenty of exercise and you'll be fine. Ever heard those words from your doctor? If that's all heshe recommends then you're missing out an important ingredient for health that he's not telling you. Fact is that you can adhere to the strictest diet, watch everything you eat and get the exercise of amarathon runner and still come down with diabetic complications. Diet, exercise and standard drug treatments simply aren't enough to help keep your diabetes under control.

Get My Free Ebook


Post a comment