Review article
Irreversibly binding anti-metal chelate antibodies: Artificial receptors for pretargeting

https://doi.org/10.1016/j.jinorgbio.2006.01.004Get rights and content

Abstract

Antibodies against metal chelates may potentially be used in biomedical applications such as targeted imaging and therapy of cancer. Highly specific monoclonal antibodies can be developed, but their binding strength needs to be maximized for them to be of practical use. In general, the half-life for dissociation of an antibody–ligand complex is more than an order of magnitude lower than the half-lifetimes for decay of medically useful radiometal ions. Practically speaking, the metal chelate-based ligand will not be bound to its receptor long enough for all of the bound radiometal to decay. A novel approach to this problem is a combination of synthetic chemistry and site-directed mutagenesis, to position a mildly reactive group on the metal chelate adjacent to a complementary reactive group on the antibody when the complex is formed. The partners are chosen to be sufficiently unreactive so that they coexist with other molecules in living systems without undergoing reaction. When the antibody–chelate complex is formed the effective local concentrations of the two groups can be non-physically large, so that a permanent link is formed in the complex even though no reaction occurs when the partners are free in solution.

Introduction

Cancer therapies that discriminate between malignant and normal tissues are preferred for the treatment of cancer over less selective conventional strategies [1], [2], [3], [4], [5], [6], [7]. Many targeted therapies rely on the use of monoclonal antibodies (mAbs1) designed to bind specifically to extracellular receptors that are overexpressed or found only on malignant cells. Normally, engineered mAbs are not able to effectively recruit the immune system to destroy cancer cells. Rituximab (Rituxan), a chimeric monoclonal antibody that binds the CD-20 antigen expressed on the surface of certain B-cell lymphomas, is a rare exception that effectively destroys non-Hodgkin’s lymphoma (NHL) positive cells by antibody-dependent cell-mediated cytotoxicity (ADCC) [8], [9]. More commonly a cytotoxic agent such as a radionuclide or immunotoxin is attached to the antibody to aid in the destruction of the cancer tissue (Fig. 1).

Radioiodination of targeting antibodies with the β-emitting isotope 131I was a popular option initially; Bexxar®, a 131I labeled anti-CD-20 binding antibody, was recently approved by the US Food and Drug Administration for the treatment of NHL [10], [11], [12], [13], [14], [15]. Radiometal ions became an attractive alternative to iodine for radiolabeling antibodies because of their diverse properties (range of half-lives, particle emissions and decay energies), which can be selected to suit a particular application. Metal ions are attached to targeting antibodies via bifunctional chelating agents (BCAs) such as (S)-2-(4-(2-bromo-acetamido)-benzyl)-DOTA (BAD, Fig. 1), which incorporate an electrophilic arm for attachment to proteins and a chelating moiety that binds β- or α-emitting metal ions (therapy) or γ or positron emitters (imaging) [10], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25]. Even the therapeutic efficacy of Rituximab is significantly improved when the related murine IgG is conjugated with the therapeutic radionuclide, yttrium-90 (Ibritumomab, also called Zevalin®) [26], [27].

Because monoclonal antibodies bind their ligands reversibly, an antibody retaining high affinity for its ligand is desirable to maximize the bound lifetime. Antibodies of the IgG type possess two identical binding sites in a single molecule (Fig. 1). The bivalent nature of IgGs increases the effective affinity (“avidity”) of these molecules when bound to a cell, virus or other entity containing many copies of a particular ligand [28].

In practice, first generation radioimmunotherapeutic (RIT) strategies have fallen short of significantly improving the therapeutic ratio (target:non-target) of conventional therapies. Antibodies have circulating half-lives in the bloodstream that can be on the order of several days, attributable to their large size, which largely exempts them from removal by glomerular filtration in the kidneys (Fig. 2) [29]. Consequently, radionuclide carrying antibody molecules that have not localized to tumors continue to circulate in the bloodstream, delivering radiation to non-target tissues. An additional complication of using whole IgG antibodies for radioimmunotherapy is that hepatocytes in the liver express Fc binding receptors on their surfaces [30], which leads to accumulation of radioactive antibodies and possible radiation toxicity to the liver.

One alternative to the use of whole IgG antibodies is to use IgG fragments such as Fab, Fab′, scFv, or other engineered molecules, which maintain the antigen binding capabilities of whole antibodies but eliminate all or part of the Fc region, avoiding undesirable binding to Fc receptors in the liver (Fig. 3) [31], [32], [33], [34], [35]. These lower molecular weight antibody fragments (approximately 50 kDa for Fab fragments, slightly larger for Fab′, and ∼28 kDa for scFv compared to 150 kDa for whole IgG) have decreased circulating lifetimes in the bloodstream, thereby decreasing the radiation dose to non-target tissues. Unfortunately, the shorter circulating lifetime also translates to decreased antibody uptake in the tumor, decreasing the amount of radiation delivered [36]. Another negative consequence to using antibody fragments is their monovalent nature; the bivalency of whole IgGs significantly enhances the bound lifetime of an antibody to a multivalent target [37], [38]. Unique constructs that fuse Fab, Fab′ or scFv domains can improve the in vivo pharmacokinetics and bound tumor lifetimes (Fig. 3) [28], [36], [39], [40], [41], [42], [43], [44].

Section snippets

Pretargeting – a shift in strategy

Manipulation of antibody binding domains has led to a variety of unique structures possessing pharmacokinetics properties that differ from those observed for whole IgGs. Although decreases in toxicity have been achieved using engineered antibody fragments, the total amount of radiation delivered to targeted tumors has often been less than that achieved with bivalent whole IgGs. Although much was learned from manipulating the size and valency of antibodies, a shift in strategy was required to

Anti-metal chelate antibodies: second generation receptors for pretargeting

Second-generation pretargeting strategies incorporating anti-metal chelate antibody fragments that bind their haptens with reasonably high affinity were designed to replace streptavidin-based receptors [44], [45], [61], [62], [63], [64]. In contrast to streptavidin, antibodies can be humanized to reduce their immunogenicity, and synthetic metal chelates are small molecules that display practically no immunogenicity. Complexes of Gd3+ have been widely used in vivo for years at high

Engineering covalent ligand–receptor interactions

Historically, the inhibition of protein-based enzyme targets has been a high priority of pharmaceutical research. Promising drug candidates often bind their targets with high affinity and are effective at low concentrations. Ideally, the best of the high affinity drug candidates bind their targets permanently, eliminating the activity of the bound target by deactivating the catalytic center. Enzymes possessing catalytic cysteine residues are well-suited to this type of inhibition [72], [73],

Irreversibly binding anti-metal chelate antibodies: artificial receptors for pretargeted imaging and therapy

Chmura et al. introduced the concept of “infinite affinity” to characterize the highly specific permanent bond formed between an engineered single-cysteine mutant of antibody CHA255 and the ligand In-AABE1 upon binding. AABE is an EDTA1 derivative incorporating a weakly electrophilic acrylamide substituent. The crystal structure of CHA255 bound to a reversible ligand, In-EOTUBE1 [89], was used to identify serine 95 (light chain, CDR3) as a promising site for mutagenesis due to its proximal

Conclusion

Much has been expected of antibody-based therapies for some time now. After many years of research only a handful of antibody-based drugs have been approved for therapeutic use. Conventional antibody targeting efforts were historically patterned after better understood small molecule-based approaches, partly because much was yet to be discovered regarding the in vivo pharmacokinetic behavior of protein-based drugs. While protein-based treatments present new challenges, there are advantages.

Abbreviations

 

    AABD

    (S)-2-(4-acrylamidobenzyl)-DOTA

    AABD-tBu

    (S)-2-(4-acrylamidobenzyl)-DOTA tetra-tertbutyl ester

    AABE

    (S)-2-(4-acrylamidobenzyl)-EDTA

    BAD

    (S)-2-(4-bromoacetamidobenzyl)-DOTA

    CDR

    complementarity determining region

    CH1 and CL

    regions of the heavy chain and light chain comprising the constant domains of the Fab

    DOTA

    1,4,7,10-tetraazacyclododecane-N,N′,N″,N‴-tetraacetic acid

    EDTA

    ethylenediaminetetraacetic acid

    EOTUBE

Acknowledgements

Supported by Research Grant CA016861 from the National Cancer Institute, National Institutes of Health, and by NIH Shared Instrumentation Grant RR014701.

References (96)

  • D.A. Goodwin et al.

    Biotechnol. Adv.

    (2001)
  • R.V. Chari

    Adv. Drug Deliv. Rev.

    (1998)
  • M.J. Glennie et al.

    Immunol. Today

    (2000)
  • M.E. Reff et al.

    Blood

    (1994)
  • P. McLaughlin

    Crit. Rev. Oncol. Hematol.

    (2001)
  • S.V. Govindan et al.

    Pharm. Sci. Technol. Today

    (2000)
  • T.M. Illidge et al.

    Blood

    (1999)
  • S.J. DeNardo et al.

    Curr. Opin. Immunol.

    (1999)
  • G.A. Wiseman et al.

    Crit. Rev. Oncol. Hematol.

    (2001)
  • A. Pluckthun et al.

    Immunotechnology

    (1997)
  • P.J. Hudson

    Curr. Opin. Immunol.

    (1999)
  • P. Holliger et al.

    Curr. Opin. Biotechnol.

    (1993)
  • R.J. Poljak

    Structure

    (1994)
  • A.A. Kortt et al.

    Biomol. Eng.

    (2001)
  • O.W. Press et al.

    Blood

    (2001)
  • M. Rusckowski et al.

    Nucl. Med. Biol.

    (1997)
  • D.A. Goodwin et al.

    Int. J. Rad. Appl. Instrum. B

    (1986)
  • G.C. Adam et al.

    Chem. Biol.

    (2001)
  • G.C. Adam et al.

    Mol. Cell. Proteomics

    (2002)
  • A.J. Chmura et al.

    J. Control. Release

    (2002)
  • P. Carter

    Nat. Rev.

    (2001)
  • C.F. Meares et al.

    J. Mol. Recognit.

    (2003)
  • I.I. Pastan et al.

    Adv. Drug Deliv. Rev.

    (1998)
  • G.R. Thrush et al.

    Annu. Rev. Immunol.

    (1996)
  • J.M. Burke et al.

    Cancer Control

    (2002)
  • G.L. DeNardo et al.

    J. Nucl. Med.

    (1999)
  • G.L. DeNardo et al.

    Clin. Cancer Res.

    (1999)
  • M.S. Kaminski et al.

    N. Engl. J. Med.

    (1993)
  • F. Buchegger et al.

    J. Nucl. Med.

    (1990)
  • M.S. Kaminski et al.

    J. Clin. Oncol.

    (2001)
  • S.J. DeNardo et al.

    J. Nucl. Med.

    (1997)
  • S.J. Knox

    Cancer Res.

    (1995)
  • S.J. Knox et al.

    Clin. Cancer Res.

    (1996)
  • S.J. DeNardo et al.

    Proc. Natl. Acad. Sci. USA

    (1997)
  • R.T. O’Donnell et al.

    Clin. Cancer Res.

    (2001)
  • M.R. McDevitt et al.

    Science

    (2001)
  • G.L. DeNardo et al.

    Cancer

    (1997)
  • G.A. Wiseman et al.

    Eur. J. Nucl. Med.

    (2000)
  • D.A. Goodwin et al.

    Cancer

    (1997)
  • U. Hopf et al.

    J. Immunol.

    (1976)
  • R.E. Bird et al.

    Science

    (1988)
  • J.S. Huston et al.

    Proc. Natl. Acad. Sci. USA

    (1988)
  • R. Fischer et al.

    Biotechnol. Appl. Biochem.

    (1999)
  • D. Colcher et al.

    Natl. Cancer Inst.

    (1990)
  • D. Colcher et al.

    J. Nucl. Med.

    (1998)
  • D.W. Mason et al.

    Biochem. J.

    (1980)
  • M. Rheinnecker et al.

    Immunology

    (1996)
  • A.M. Wu et al.

    Q. J. Nucl. Med.

    (2000)
  • Cited by (0)

    View full text