Human Epidermal Growth Factor Receptor Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Chapter 1

Breast cancer is a major health problem; it has been on the increase in the past two decades. In 2008, a global estimate of 12.7 million cancer cases and 7.6 million deaths were recorded. Breast cancer (occurs in both men and women) accounts for 23% of all cancer cases and 14% of all cancer deaths. It is the leading cause of cancer death in females worldwide, with high incidence in Western and Northern Europe among women between 35-55 years (Jemal et al., 2011).

The biology of breast cancer is a complex one, with several factors involved in its development and progression. Several gene products are implicated in the development of the disease; Estrogen receptors (ER), retinoic acid receptors β (RARβ), epidermal growth factor receptors (EGFR) family, Progesterone receptors (PR), Protein 53 (p53, a tumor suppressor), Breast cancer 1 and 2 (BRCA1 & BRCA2, tumor suppressors which normally regulate the growth of cells in the breast), steroid hormones, peptide growth factors, etc. These proteins may serve as markers for prognosis and therapeutic targets (Keen and Davidson, 2003). Breast cancer is associated with the over expression of specific proteins on the tumor cell surface. Amongst the expressed protein, the estrogen receptors(ER), progesterone receptors (PR) and Human epidermal growth factor 2 (HER2) are useful in prognosis (Cooke et al., September 12-16 1999).

1.1.2 HER2 and Breast Cancer: HER2 activation and Signaling

HER2 is a peptide growth factor receptor implicated in normal development of the mammary gland and also implicated in carcinogenesis (Olayioye et al., 2000). HER2 is one of the four known membrane tyrosine kinases (HER1, HER2,HER3, and HER4) (MENARD et al., 2000). This HER2 membrane protein is over expressed in 25-30% of human breast cancer cases. Its over expression (an indication of an aggressive tumor with high risk of recurrence) is usually linked to the amplification of the proto-oncogen gene which encodes for the HER2 protein (Caroline, 2003, Chang, 2010, Cooke et al., September 12-16 1999).

Figure 1: EGFR signal transduction cascade (Baselga J, 2002)

HER2 is a transmembrane glycoprotein of about 185KDa; it has cysteine rich external domains (I to IV), a hydrophobic transmembrane region and a cytoplasmic tyrosine kinase domain (Freudenberg et al., 2009, MENARD et al., 2000, Yarden, 2001, Weiner et al., 1989). The EGFRs exist as monomers in the absence of their ligands. Ligand binding stimulates dimerization of the receptors; homodimers and heterodimers comprising different combinations of EGFRs are formed. The EGFR are activated when a ligand such as epidermal growth factor (EGF), transforming growth factor-α (TGF-α) etc., binds to their extracellular domain (Baselga, 2002). In the dimerized state, activation of the intrinsic protein tyrosine kinase activity and phosphorylation of specific tyrosine residues within their cytoplasmic domains occurs. The phosphorylated tyrosine residues recruit and phosphorylate other proteins in the signal transduction pathway; the phosphorylated cytoplasmic domain serves as a docking site for intracellular proteins. In this way, the cells are able to convey environmental signals into the cell via their phosphorylated cytoplasmic domains: The activated tyrosine residue on the intracellular domain of HER2 activates lipid kinase phosphoinositide 3-kinase (PI3-K), which phosphorylates a phosphatidylinositol that in turn binds and phosphorylates the enzyme Ak transforming factor (Akt), driving cell survival. In parallel, a guanine nucleotide exchange factor, the mammalian homologue of the son of sevenless (SOS), activates the rat sarcoma (RAS) enzyme that, in turn, activates receptor activation factor (RAF) and then the mitogen-activated protein kinase (MAPK) and mitogen extracellular signal kinase (MEK). MEK phosphorylates, among others (Hudis, 2007a). The end result is a cascade of signal transduction that activates genes regulating cell growth, proliferation, hormonal dependence, apoptosis, angiogenesis, invasion and metastasis (MENARD et al., 2000, Olayioye et al., 1998, Yarden, 2001).

There are many known ligands for the HER family proteins. Unlike the other hEGFRs, HER2 has no known ligand and it preferentially binds to other ligand-bound hEGFRs proteins. It is therefore assumed to play the role of a coreceptor (Olayioye et al., 2000). Some of the ligands include; amphiregulin (AR), transforming growth factor-α (TGF- α) specific for HER1, Betacellulin (BTC) specific for HER1 and 4, neuregulins (NRG 1,2, 3 & 4). HER2 in breast cancer dimerizes in the absence of the ligand whereas HER1, HER3, and HER4 only form dimers in the presence of the ligand. The possible reason explaining this is that in the oncogene, a point mutation causes the substitution of Valine for Glutamic acid which confers a net negative charge on the transmembrane region of HER2. This charge promotes aggregation of the protein in the absence of any ligand, leading to a signal transduction pathway that causes uncontrolled cell proliferation and cell mobility. Also, worth noting is that HER3 has no tyrosine kinase activity and is the preferred partner for HER2 in dimer formation (Freudenberg et al., 2009, MENARD et al., 2000, Weiner et al., 1989, Yarden, 2001, Olayioye et al., 2000).

Figure2: Schematic representation of possible combinations of HER protein dimers (Olayioye et al., 2000).

1.1.3 HER2 Detection in Breast Cancer

Current methods that are used to determine the HER2 status in breast cancer are Immunohistochemistry (IHC) and Fluorescence in situ Hybridization (FISH).

IHC is more often used due to its simplicity, low cost and availability. This method employs antibodies (monoclonal or polyclonal) directed against the extracellular part of HER2 on a tissue biopsy and generates a color signal that is appraised by bright field microscopy. The intensity of the stain determines the level of HER2 expression. The major setbacks of the technique are the subjective appraisal of the stain score and it is low reproducibility, with an error rate of about 20% (Perez et al., 2006). In addition, possible loss of HER2 protein in the course of tissue fixation and storage cannot be under rated (PEREZ et al., 2002).

FISH is mainly used as a semi quantitative technique to quantify the amplification of the HER2 oncogene. The principle is that DNA probes complementary to the genomic sequence of interest are generated and labeled with a fluorophore. The probes are then applied to the tissue biopsy. Positive fluorescent signals are visualized and counted by fluorescent microscopy (Jimenez et al., 2000). HER2 loss on cell surface in the course of tissue fixation is not observed in FISH. However, the technique requires some training and it is time consuming. Fluorescent quenching may also reduce the signal.

1.1.4 Targeted Therapy of HER2 positive cancer

Available treatments for breast cancer targets the HER2 which is over expressed in 25-30% of breast cancer cases: One of such current treatments is with Trastuzumab (HerceptinTM, Genentech/Roche, South San Francisco, CA). Trastuzumab is a monoclonal humanized antibody direct against the HER2 protein. It binds to the extracellular domain of the HER2 and blocks signaling via the inhibition of HER2-receptor dimerization needed for its tyrosine kinase activity, and activation of antibody dependent cell-mediated cytotoxicity and endocytosis of the receptor (Hudis, 2007b).

New Picture New PictureFigure 3: Mechanisms of action of Trastuzumab : (A) Four members of the HER family and the signal transduction cascade driven by these receptors when activated by their respective ligands, (B) Trastuzumab, (C) Trastuzumab binds and blocks the shedding of HER2's extracellular domain, (D) Trastuzumab physically inhibits either homodimerization or heterodimerization of HER2, thereby reducing signaling by HER2, (E) Trastuzumab induces antibody-dependent cell-mediated cytotoxicity leading to tumour cell dead, and (F) receptor down regulation through endocytosis (Hudis, C.A., 2007).

Cases of Cardiotoxicity leading to heart failure were reported in the phase II clinical trial of Trastuzumab. Risk factors were age and prior exposure to anthracycline (which is thought to expose more HER2 on the cardiomyocytes (SENGUPTA et al., 2008). As a consequence, good diagnostic tools are needed to select patients who will benefit from the treatment with Trastuzumab.

In addition, cases of resistance to the drug have been reported. Recently, a derivative of the Trastuzumab, Trastuzumab-DM1, has been developed to overcome the resistance shown towards Trastuzumab by the HER2 tumor cells. DM1 (maytansine) is a fungal toxin that inhibits the formation of microtubules in cells, leading to cell death. Trastuzumab functions here as a carrier molecule to allow specific targeting of the HER2 tumor cells (Fanga et al., 2011). This derivative of Trastuzumab is independent of the HER2 mediated signaling pathway.

Another drug that targets the HER2 protein is Lapatinib (TykerbTM, GlaxoSmithKline, Research Triangle Park, NC), a tyrosine kinase inhibitor. Lapatinib binds to the catalytic domain of the HER2 (tyrosine kinase domain), blocking ATP binding and thus preventing autophosphorylation necessary for the initiation of the signal transduction cascade. Lapatinib also binds to HER 1. Lapatinib seems to be a better option to Trastuzumab for treatment of brain metastasis because it has a low molecular weight, thus can cross the blood brain barrier and target the brain metastases. Besides, cases of cardiodysfunction are rare (Fanga et al., 2011).

Pertuzumab (OmnitargTM; Genentech/Roche, South San Francisco, CA) is another humanized monoclonal antibody that targets the extracellular domain of HER2 (domain II) and blocks homo and heterodimerisation of HER2 with other EGFR family members and thus blocks signal transduction.

1.2.0 Proximity Ligation Assays (PLA)

The existing techniques used for the detection of HER2 status in breast cancer are limited by an error rate of 20% and the lack of correlation between FISH and IHC. Furthermore, the side effect associated with the existing drugs targeting HER2 in breast cancer (such as cardio dysfunction) and the expensive nature of the drugs, (Jimenez et al., 2000, PEREZ et al., 2002) requires a more stringent selection procedure for patients who will benefit from the drug. This is the reason why the search for a more and robust HER2 assays is appropriate. Proximity Ligation Assay (PLA) might be a good alternative to FISH and IHC.

After the completion of the human genome project in 2004, it was revealed that the human genome has about 20000-25000 genes coding for proteins (Consortium, 2004). With this information at hand, researchers have being exploiting the human proteome with the aims of studying the structure and functions of the available proteins that are expressed and to find possible markers for therapeutic and diagnostic purposes.

Several approaches have been presented by researchers for the detection of proteins or biomarkers for diagnostic and therapeutic purposes. In this section I will focus on the probe based targeting techniques.

1.2.1 Immunoassays

Several immunoassays are available in the proteomic tool box for the targeting or profiling of a proteome of an organism with the goal of finding possible diagnostic and therapeutic candidates. Amongst which is the radioimmunoassay (RIA), first described for the detection of plasma insulin in man (Yelow and Berson, 1960). This was a magnificent contribution in the medical field as the team won the 1977 Nobel Prize in medicine. The drawback of this great proteomic tool is the hazard of radioactivity to health and the need for special equipments to measure the γ-radiation.

A non-radioactive based analogue of RIA, Enzyme-linked immunosorbent assay (ELISA) was developed to eliminate the use of radioactive labeling: In this approach, the antibodies are coupled to an enzyme which converts its substrate to a colored product in the presence of the antigen (Engvall and Paulsson, 1971). In the same year, Enzyme immunoassay (EIA) was described, demonstrating that human chorionic gonadotropin concentrations in urine could be quantified (Van Weemen and Schuurs, 1971). ELISA and EIA though with different designs, both are based on the principle of coupling an enzyme to an antibody and not a radio isotope. The idea of coupling antibodies to enzymes as well as fixing them on solid phases were described in the 1960s (Avrameas, 1969, Nakane and Pierce, 1967, Wide and Porath, 1966), long before the development of RIA and ELISA. Western blot is another probe based method that applies the same principles of ELISA in its functioning, but is rather a qualitative than a quantitative test.

However, cross reactivity was the major drawback of RIA, ELISA and Western blot (Weibrecht et al., 2010). ELISA was later improved to involve two recognition events as in Sandwich ELISA, aiming at solving this problem of cross reactivity. Nonetheless, nonspecific binding of the detecting antibody still remains a bottleneck in this assay, giving background signals. This background signal makes it difficult to detect low abundant molecules present in a sample (Weibrecht et al., 2010).

In the early 1990s, immune-PCR was developed, aiming at improving the sensitivity above those of existing assays such as the classical ELISA: In this method, an oligonucleotide sequence is conjugated to an antibody and used as an affinity probe to target a protein. A single recognition event is greatly amplified by the DNA polymerase above background signal (Sano et al., 1992). This method still uses a single recognition event, liable to cross reactivity. Nonspecific binding will also make the coupled oligonucleotide available for amplification, giving background signal.

Proximity ligation assays (PLA) is a novel probe based techniques, recently added to the proteomic tool box. PLA is a method whose functioning is based on the simultaneous recognition of a target molecule by a pair of affinity probes, giving rise an amplifiable DNA sequence. As a requirement for PLA, the two affinity probes must be in close proximity to produce this sequence by ligation. Once the DNA sequence is present, it can be amplified like in quantitative PCR (qPCR) (Fredriksson et al., 2002).

1.2.2 Proximity Ligation Assay

PLA was first described in 2002 by Ulf Landegran research team for the detection and quantification of the platelet derived growth factor receptor, in a homogenous phase.

Principles behind PLA: Two antibodies are generated against the target protein. An oligonucleotide sequence is then coupled to each antibody to form PLA probes. Binding of the two antibodies to the same antigen brings them into close proximity. A connector oligonucleotide, complementary to the ligation ends of the conjugated oligos, is then added to function as a template for ligating the ends of the conjugated oligonucleotides, forming an amplifiable DNA sequence (readout) and is amplified by qPCR. This greatly amplifies the signal for a single recognition event. Only dual recognition of the target by the pair of affinity probes can give a signal.

The initial PLA was done in homogenous phase and was limited in just detecting the presence or absence of a protein in a biological sample. This has been exploited in detecting cytokines in femtomolar sensitivity compared to the picomolar sensitivity of ELISA (Gullburg et al., 2004).

PLA was later adapted to allow visualization of proteins in situ. In this modified version, binding of the two proximity probes to the antigen guide the formation of a circular ssDNA which serves as a template for rolling cycle amplification; producing concatemers of single stranded repeats of the circular ssDNA that remain attached to one of the probes. The amplified DNA is then detected by a fluorescent probe, allowing localized visualization of the target protein in situ (SÃ-derberg et al., 2006). This has been used to visualize the interaction between Myc and Max (oncogene transcriptional factors) in response to IFN-γ signaling, interaction of HER2 with HER3 on cell lines, and the phosphorylated state of the platelet derived growth factor receptor(PDGFR-β) after stimulation with platelet derived growth factor-BB (PDGF-BB) (Jarvius et al., 2007, Leuchowius et al., 2009).

Triple PLA (3PLA) was later developed to further increase the sensitivity of the assay to allow detection of very low level protein and isoforms of proteins resulting from splicing and post translational modifications (Schallmeiner et al., 2007). Furthermore, the development of solid phase PLA only further increased the sensitivity of the assay, allowing detection of very low protein in undiluted plasma, serum and whole blood (Darmanis et al., 2010).

The conjugation of an oligonucleotide sequence to antibody is quite a difficult step in developing the assay. To solve this problem, a more generalized secondary probes based model was developed. In this model, the oligonucleotides are conjugated to secondary antibodies instead of the primary antibodies. These pairs of secondary probes can be used with any pair of primary antibodies generated from the right animal species (Leuchowius et al., 2010).

PLA is a very sensitive and promising technique but it has been described with the conventional antibodies. Conventional antibodies are quite large and out of reach for certain conserved epitopes, limiting the application of the assay in targeting certain antigens such as the VSG of Trypanosomes which constantly shuffles their coat protein. Secondly, conventional (monoclonal) antibodies are quite expensive to produce. The polyclonal antibodies are more economic in production, but suffer from specificity and a limited availability in time. Small antigen binding fragments such Fab, Fv and scFv, derived from the conventional monoclonal antibodies are less stable, have lower solubility, easily aggregate and provide low expression yields in microbial expression vectors. Nature has offered a wonderful analogue of these small intact antigen binding fragments as seen below.

1.3.0 Immunoglobulin (Ig)

1.3.1 The Conventional immunoglobulins and the immune system

The immune system in simple terms could be defined as one which assists living animals in fighting against invading pathogens (foreign or non self) and also help eliminate altered self-cells such as tumor cells in a normal healthy condition. Specialized cells and molecules in the body perform these defense functions.

One of such important molecules are the immunoglobulins (Ig), expressed and secreted by B-cells on antigen exposure. Generally, all immunoglobulins have four polypeptide chains; two identical heavy chains (HC) and two identical Light chains (LC).

New Picture

Figure 4: The structural organization of the main human immunoglobulins; picture modified from Charlse et al, 2001.

Based on the LC, antibodies can be classified into two types: Kappa (κ) and the lambda (λ). Only one type of light chain can exist in a particular antibody at a time.

Based on the Heavy chain, there are five immunoglobulin classes (Isotypes) designated α (IgA), δ (IgD), ε (IgE), γ (IgG) and μ (IgM). IgM is a pentamer, giving it a high avidity. It is the first Ig to be produced by naive B-cells upon first antigen exposure. It has a broad recognition but low specificity. On the other hand, IgG is produced by activated B-cells on a second exposure to the same antigen. They are monomers with high specificity (Charles et al., 2001). IgA is a dimer on secretion in mucous secretions; saliva, tears, intestinal juice, colostrum, respiratory epithelium etc. and a small amount in blood. IgE is implicated in allergic reactions and parasitic worm infections.

1.3.2 Structure of an Antibody

Each antibody comprises two identical polypeptide chains (figure. 6a) (LC and HC), each of them having two distinct regions, a constant region (CV and CH) and a variable region (VL and VH). The variability in these variable regions are restricted to three regions called complementarity determining regions (CDRs: CDR1, CDR2 and CDR3) or hypervariable loops, responsible for the antigen specificity of a particular Ig. CDR3 has the highest variability in amino acid composition and length. The CDRs of VL and VH combine to form the antigen binding side (a total of 6 CDRs forms the antigen binding side). The CDRs are connected by framework regions with less variability in amino acid composition. The Light chain has just one constant region while the heavy chain has three to four constant regions depending on the Ig. In both cases, the constant regions are located at the C-terminus while the variable regions are located at the N-terminus (Charles et al., 2001).

Besides the conventional antibodies with two identical H-chains and L-chains, there exist in nature some species forms of IgG consisting of only two identical H-chains, and still having antigen specificities. Such antibodies have been revealed in camelids, Ilamas (heavy chain only antibodies,) and cartilaginous fishes (Immunoglobulin New Antigen Receptor, IgNAR).

1.3.3 IgNAR

The IgNAR (figure 5) are a unique immunoglobulin isotype present in the serum of Wobbegong (Oreclolobus maculats ) and Nurse sharks (Ginglymostoma cirratum) (Greenberg et al., 1995, Nuttall et al., 2001). Generally, the IgNAR comprises two polypepetide chains (H-chains) linked together by disulphide bridges with each chain having one variable domain and five constant domains (Nuttall et al., 2003). The variable domain has three complementarity determining regions CDRS, CDR1, CDR2 (very short) and CDR3 (Nuttall et al., 2003, Roux et al., 1998).

So far, three types of IgNAR have been revealed: The type1 bears two or four Cysteine residues in the CDR3 region which form intra-chain disulphide bridge(s). On the other hand, type2 have one Cysteine residue each in CDR1 and CDR3 which form inter-chain disulphide bridge between the two CDRs (Nuttall et al., 2003, Roux et al., 1998). The type3 are found in neonates and its population disappears as the shark grows older while the type2 and 3 are found predominantly in adult sharks. The type2 and 3 undergo affinity maturation after the immune system is stimulated by an antigen (Diaz et al., 2002, Nuttall et al., 2003, Roux et al., 1998).

1.3.4 Camelid Heavy-Chain only Antibodies

The Heavy-chain only antibodies are abundant in the serum of all camelids (figs. 5-6) (Camels and Ilamas). They belong to the IgG2 and IgG3 sub-isotypes. They co-exist with the IgG1 conventional antibodies. IgG1 binds to protein A and G whereas IgG2 binds to protein A and IgG3 to protein A and G. The IgG2 has a molecular weight of about 90kDa, IgG3 about 80KDa and IgG1 is about 150KDa. IgG2 and IgG3 lack the CH1 domain resulting in the absence of a light chain as well. As such they are called Heavy chain only antibodies (HcAb). IgG2 has a longer hinge region compared to its IgG3 counterpart (Hamers-Casterman et al., 1993). The absence of the light chain has facilitated the engineering of small size antibody fragments from these HcAbs since they bind their antigen with high affinity and specificity (Muyldermans, 2001).

1.3.5 Single Domain antibody Fragments

The Fab portion of the conventional antibodies is responsible for its antigen binding ability while the Fc portion mediates the cytotoxic effector function via complement activation and or binding to Fc receptors for gamma globulins. This Fc portion increases the serum half-life of the antibody, a property not good for imaging applications due to poor contrast associated to the increased serum half-life.

To eliminate this property of increased serum half-life, a hurdle in applying conventional antibodies in imaging, several antibodies fragments lacking the Fc region have been engineered e.g. Fab'2, Fab, VH, VL, etc. However, size and stability have limited the production and uses of some of these engineered fragments: Separation of Fab into VH and VL exposes the hydrophobic portions of these chains, creating a stability problem (Holliger and Hudson, 2005, Nuttall et al., 2001). The isolated VH and VL easily aggregate due to exposure of their hydrophobic residues, making their production a difficult task. In addition, the engineered antibodies fragments from conventional antibodies rarely retain the affinity of the original antibody (Holliger and Hudson, 2005).

The discovery of HcAb in Camelids (Hamers-Casterman et al., 1993) and in sharks (Nuttall et al., 2001) has greatly revived the engineering of single domain antibodies fragments. Soluble intact antigen binding fragments, VHH (Nanobodies) and V-NAR have been engineered from these new types of natural existing antibodies.

New Picture (3)

Figure 5: Picture of existing antibody formats, both natural occurring forms and varieties of engineered forms (Holliger and Hudson, 2005).

1.3.6 Nanobodies (VHH)

Nanobodies named by Ablynx company because of its nanometer range dimension of the VHH (diameter ~2,5nm and height ~4nm) (Muyldermans et al., 2009), are the smallest available intact antigen binding fragments, with a molecular weight ~15kDa, that still have their full antigen binding capacities as the original heavy chain only antibody, HcAbs, (Deffar et al., 2009).

Nanobodies are derived from a unique class of antibodies devoid of the light chain, existing in the serum of Camelidae and sharks (Hamers-Casterman et al., 1993, Nuttall et al., 2001). The existence of this class of heavy chain only antibodies was long discovered in the 1990s by Prof. Raymond Hamers (Deffar et al., 2009). In the analysis of the Camelid (Alpaca and Ilama) serum, three fractions of immunoglobulin of the IgG class were observed (van der Lindena et al., 2000, Hamers-Casterman et al., 1993): The IgG1 fraction of about 150kDa, which dissociates into heavy and light chain (50 and 25kDa) respectively, under reducing conditions; IgG2 and IgG3 fractions of about 90 and 80KDa under non reducing conditions and migrate as 45 and 40KDa under reducing conditions (Hamers-Casterman et al., 1993). The HcAbs have a unique structure that comprises a variable domain (VHH, so called to distinguish it from the VH of classical antibodies), a hinge region and two constant regions CH2 and CH3, which are homologous to the Fc domain of the classical antibodies (Muyldermans et al., 2008). The CH1 domain is absent and explains to an extend the absence of the light chain because the CH1 is the point of anchor for the constant light chain in conventional antibodies (Muyldermans, 2001).

Figure 6: (A) composition of a classical antibody and a Heavy-chain only antibody and a single domain antigen binding fragment, VHH or Nanobody, derived from the Heavy chain antibody. (B) Organization of VH and VHH sequence with the CDRs shown as well as the inter CDR disulfide bond shown. The folded structure of VH and VHH are also shown (Muyldermans et al., 2009).

VHH and VH have a similar organization. However there are some slight but importance differences between the two. Both VH and VHH comprises four conserved stretches; framework regions surrounding the three Complementarity Determining Regions (CDR1, CDR2 and CDR3). They both adopt an Ig fold of two β-sheets, one made of four strands and the other of five strands. The three CDRs are clustered at one end of the domain, in the loops connecting the β-strands (Muyldermans et al., 2009).

The difference between the VH and VHH lies in the CDRs: The CDR1 and CDR3 of VHH are more extended than those of VH and often contain a Cysteine residue. The Cys in CDR1 and CDR3 form and inter CDR disulphide bond that assist in shaping the loop structure (Muyldermans, 2001, Muyldermans et al., 2009).

Another major difference between the VH and VHH lies in framework 2 (FR2): the FR2 of VH has more hydrophobic amino acids that anchore the VL domain whereas the FR2 off VHH has more hydrophilic amino acids, explaining the absence of VL and the solubility of the VHH as a single entity (Muyldermans et al., 2009). The substitution of hydrophobic amino acids for hydrophilic ones in VHH mainly occur at positions 37, 44, 45 and 47 (Muyldermans, 2001).

Köhler and Milstein introduced a powerful research tool called monoclonal antibodies, which have been applied in the development of diagnostic tools as well as human therapeutics. Amongst the many drawbacks of this great research tool, ranging from expensive production, expression in mammalian cells, potential immunogenicity when mouse IgGs are administered to humans, and the large size of conventional antibodies limit their application (Muyldermans, 2001).

To overcome this hurdle of large size, several derivatives of the conventional antibodies have been engineered (fig. 5) (Cortez-Retamozo et al., 2008). These small derivatives of monoclonal antibodies are less stable (the Fab and the Fv need the combination of both the VL and VH to form the antigen binding site) due to dissociation of the VL and VH, have low expression yield, less soluble (thus form inclusion bodies in expression systems) (Huang et al., 2010), leading to nonspecific uptake in organs. This greatly limits their application in diagnostic and therapeutic uses. In addition, the intact classical antibodies have low systemic clearance due to interaction of their Fc part with body tissues (Cortez-Retamozo et al., 2008, Muyldermans, 2001, Muyldermans et al., 2008). All these properties limit their use in applications like imaging because low systemic clearance and nonspecific organ uptake will give high background signal (Harmsen and De Haard, 2007, Cortez-Retamozo et al., 2008, Huang et al., 2010).

Nanobodies have terrific advantages over the classical antibodies: They are small in size (about 15KDa); rapid systemic clearance, mainly through the kidney; good tumor uptake and retention, giving a high signal to background ration after 2 hours of application. This rapid systemic clearance together with their good tumor uptake properties makes them good candidates for imaging (Cortez-Retamozo et al., 2008, Lieven et al., 2010, Muyldermans, 2001, Muyldermans et al., 2008).

In addition, Nanobodies are cheap and easy to produce (a factor for large scale application), soluble (due the substitution of amino acids at the framework region 2, FR2, by hydrophilic amino acids), stable and functional even when exposed at temperatures as high as 90oC, low toxicity and immunogenicity, high expression yield in microbial systems, ease of purification and they can be humanized (Vincke et al., 2009), and can bind to cavities (unique conformational epitopes) out of reach for the conventional antibodies (Cortez-Retamozo et al., 2008, Ghahroudi et al., 1997, Muyldermans, 2001, Muyldermans et al., 2008, Van De Broek et al., 2011, Vaneycken et al., 2011a, Vaneycken et al., 2010).

1.4 Aim and Outline of this Thesis

HER2 is over expressed in 25-30% of breast cancer cases. The current techniques for diagnosis are IHC and FISH. IHC has an error rate of ~20%; appraisal of biopsy stains is subjective and non-reproducing, making it not efficient in the selection of patients who will benefit from HER2 target based therapy.

The Current HER2 based therapy is Trastuzumab, a monoclonal antibody that targets the extracellular domain of HER2, has as side effect cardiotoxicity followed by heart failure. Thus, there is need to improve the selection procedure for patients who will benefit from breast cancer treatment with Trastuzumab so that HER2 negative patients will not undertake this expensive treatment with more side effects than benefits.

Proximity ligation assays have high sensitivity (femtomolar) and specificity (dual recognition of the target), makes it a very promising assay. The fact that it allows for in situ visualization of protein (Gullburg et al., 2004) makes it even more interesting in cancer diagnosis. However, it has been described with conventional antibodies with some disabilities; large size (making them out of reach of some conserve epitopes), slow systemic clearance etc. Smaller fragments engineered from them are hampered with low stability, low production yields just to name a few, limit their application.

In this work we aim at exploring the small size of Nanobodies (~15KDa) to proof that proximity ligation assay can be done with Nanobodies, with HER2 as the target antigen. The Success of the work will also go a long way to further proof that Nanobodies can be labeled (despite their small nature) without loss in antigen recognition.