CA2588106A1 - Single-domain antibodies and heavy chain antibody against egfr and uses thereof - Google Patents

Abstract

The present invention relates to single-domain antibodies (sdAbs), as well as fusion proteins containing the same, directed towards epidermal growth factor receptor (EGFR). The present invention is also involved in methods of diagnosing cancer and of targeting tumors.

Description

SINGLE-DOMAIN ANTIBODIES AND HEAVY CHAIN ANTIBODY AGAINST
EGFR AND USES THEREOF

FIELD OF THE INVENTION:

The present invention relates to the field of antibodies directed towards epidermal growth factor receptor (EGFR). More particularly, the present invention relates to nucleic acid sequences and amino acid sequences which are encoded thereby, directed towards single-domain antibodies (sdAb) and clones thereof, which target EGFR. The invention also concerns an sdAb which is fused with a crystallizable fragment (Fc) of an immunoglobulin protein in order to generate a fusion protein. This fusion protein can thus be used in the targeting of tumors presenting EGFR on their surface, as well as for diagnosing certain types of cancer.
BACKGROUND OF THE INVENTION:

Epidermal growth factor receptors (EGFRs) are over-expressed and/or dysregulated in many tumor types including head and neck, breast, non-small-cell lung and pancreatic cancer to name but a few (Sebastian, S., Settleman, J., Reshkin, S. J., Azzariti, A., Bellizzi, A., & Paradiso, A. (2006) Biochimica et biophysica acta 1766, 120-139).

The EGFR family contains four members: EGFR1 (ErbB1), HER2 (ErbB2), HER3 (ErbB3) and HER4 (ErbB4) (Carpenter, G. (1987) Annual review of biochemistry 56, 881-914). Targeting EGFR in cancer cells was initially proposed by Sato et al. (Sato, J. D., Kawamoto, T., Le, A. D., Mendelsohn, J., Polikoff, J., & Sato, G. H. (1983) Molecular biology & medicine 1, 511-529).

The first anti-EGFR antibody drug, Cetuximab (ErbituxR), was approved by the FDA in 2004 for the treatment of metastatic colon cancer either in combination with CamptosarT"", a chemotherapeutic, or as a single agent for patients who cannot

2 tolerate chemotherapy (Adams, G. P., Schier, R., McCall, A. M., Simmons, H.
H., Horak, E. M., Alpaugh, R. K., Marks, J. D., & Weiner, L. M. (2001) Cancer research 61, 4750-4755).

Despite the success of Cetuximab and other antibody drugs, their large size (-150 kDa) is considered a major limiting factor in tumor penetration (Jain, M., Chauhan, S. C., Singh, A. P., Venkatraman, G., Colcher, D., & Batra, S. K.
(2005) Cancer research 65, 7840-7846) and hence their inability to reach a higher therapeutic index. Furthermore, the high production cost of antibodies contributes largely to the high retail cost of antibody drugs (Nolke, G., Fischer, R., &
Schillberg, S. (2003) Expert opinion on biological therapy 3, 1153-1162).

Alternatively, most engineered small antibody fragments, mainly single chain variable fragments (scFvs), failed to show enhanced tumor targeting. scFvs are often cleared rapidly from circulation partly due to their low molecular weight (MW;
<60 kDa, the threshold of glomerular filtration) (Trejtnar, F. & Laznicek, M.
(2002) Q J
Nucl Med 46, 181-194). As a result, scFvs usually have a serum half-life of less than 10 minutes and a peak tumor uptake of about 5 percent injected dose per gram tissue (% ID/g) (Jain, M., Chauhan, S. C., Singh, A. P., Venkatraman, G., Colcher, D., & Batra, S. K. (2005) Cancer research 65, 7840-7846). Both parameters increase with increasing size of the antibody fragments, making divalent scFvs (Goel, A., Colcher, D., Baranowska-Kortylewicz, J., Augustine, S., Booth, B. J., Pavlinkova, G., & Batra, S. K. (2000) Cancer research 60, 6964-6971), tetravalent scFvs (Goel, A., Coicher, D., Baranowska-Kortylewicz, J., Augustine, S., Booth, B. J., Pavlinkova, G., & Batra, S. K. (2000) Cancer research 60, 6964-6971) and minibodies (Hu, S., Shively, L., Raubitschek, A., Sherman, M., Williams, L. E., Wong, J. Y., Shively, J. E., & Wu, A. M. (1996) Cancer research 56, 3055-3061) more attractive targeting molecules.

In addition, most of the antibody fragments detailed above, lack the Fc region and are, therefore, unable to activate antibody-dependent cellular cytotoxicity (ADCC) and cell-dependent cytotoxicity (CDC), the two major mechanisms involved in the eradication of tumor tissue upon antibody binding (Adams, G. P., Schier, R.,

3 McCall, A. M., Simmons, H. H., Horak, E. M., Alpaugh, R. K., Marks, J. D., &
Weiner, L. M. (2001) Cancer research 61, 4750-4755).

Indeed, Fc engineering has become a major focus of antibody engineering in recent years, resulting in either extended serum half-life (Hinton, P. R., Xiong, J. M., Johlfs, M. G., Tang, M. T., Keller, S., & Tsurushita, N. (2006) J Immunol 176, 356) or shortened serum half-life (Kenanova, V., Olafsen, T., Crow, D. M., Sundaresan, G., Subbarayan, M., Carter, N. H., lkle, D. N., Yazaki, P. J., Chatziioannou, A. F., Gambhir, S. S., et al. (2005) Cancer research 65, 622-631), and in certain cases enhanced ADCC (Natsume, A., Wakitani, M., Yamane-Ohnuki, N., Shoji-Hosaka, E., Niwa, R., Uchida, K., Satoh, M., & Shitara, K. (2006) Journal of biochemistry 140, 359-368).

By fusing scFv to Fc, a novel antibody molecule scFv-Fc was generated, which self assembled into a dimer with a molecular weight of about 105 kDa (Wu, A.
M., Tan, G. J., Sherman, M. A., Clarke, P., Olafsen, T., Forman, S. J., &
Raubitschek, A. A. (2001) Protein engineering 14, 1025-1033). When fused to Fcs with varying serum half-lives, excellent tumor-targeting antibodies with tumor uptake as high as 44% ID/g were generated, making this type of molecule a promising candidate for radioimmunotherapy (Kenanova, V., Olafsen, T., Williams, L. E., Ruel, N. H., Longmate, J., Yazaki, P. J., Shively, J. E., Colcher, D., Raubitschek, A. A., & Wu, A.
M. (2007) Cancer research 67, 718-726).

Single-domain antibodies (sdAbs), also known as domain antibodies (dAbs) or nanobodies, are the smallest antibody fragments with a size of 12-15 kDa. They are usually derived from the variable regions of heavy chain antibodies (HCAbs) of either camelid (Hamers-Casterman, C., Atarhouch, T., Muyldermans, S., Robinson, G., Hamers, C., Songa, E. B., Bendahman, N., & Hamers, R. (1993) Nature 363, 446-448) or nurse shark (Greenberg, A. S., Avila, D., Hughes, M., Hughes, A., McKinney, E. C., & Flajnik, M. F. (1995) Nature 374, 168-173). Recently, non-aggregating sdAbs have also been isolated from either heavy chain or light chain variable regions of human antibodies (To, R., Hirama, T., Arbabi-Ghahroudi, M., MacKenzie, R., Wang, P., Xu, P., Ni, F., & Tanha, J. (2005) The Journal of biological chemistry 280, 41395-

4 41403 and Jespers, L., Schon, 0., Famm, K., & Winter, G. (2004) Nature biotechnology 22, 1161-1165).

Camelids such as camel, llama and alpaca have HCAbs naturally devoid of light chains and consist only of VH, CH2 and CH3 domains (Hamers-Casterman, C., Atarhouch, T., Muyldermans, S., Robinson, G., Hamers, C., Songa, E. B., Bendahman, N., & Hamers, R. (1993) Nature 363, 446-448). sdAbs derived from camelid HCAbs are excellent building blocks for novel antibody molecules (Revets, H., De Baetselier, P., & Muyldermans, S. (2005) Expert opinion on biological therapy

5, 111-124) due to their high thermostability, high detergent resistance, relatively high proteolytic resistance (Dumoulin, M., Conrath, K., Van Meirhaeghe, A., Meersman, F., Heremans, K., Frenken, L. G., Muyldermans, S., Wyns, L., & Matagne, A.
(2002) Protein Sci 11, 500-515) and high production yield (Arbabi Ghahroudi, M., Desmyter, A., Wyns, L., Hamers, R., & Muyldermans, S. (1997) FEBS letters 414, 521-526).
They can be engineered to have very high affinity by isolation from an immune library (Arbabi Ghahroudi, M., Desmyter, A., Wyns, L., Hamers, R., & Muyldermans, S.
(1997) FEBS letters 414, 521-526) or by in vitro affinity maturation (Davies, J. &
Riechmann, L. (1996) Immunotechnology 2, 169-179 and De Genst, E., Handelberg, F., Van Meirhaeghe, A., Vynck, S., Loris, R., Wyns, L., & Muyldermans, S.
(2004) The Journal of biological chemistry 279, 53593-53601). Furthermore, since a single protein domain is responsible for antigen binding, sdAbs presumably do not confer a large conformational change or lose affinity when transferred to immunoglobulin (Ig) molecules.

Despite the immense potential of sdAbs, tumor targeting with sdAbs remains largely unexplored. It is known that monomeric (15 kDa) and bivalent (33 kDa) sdAbs targeting lysozyme, artificially expressed on the surface of a tumor cell line, have been isolated, constructed and tested. However these molecules failed to show sufficient tumor accumulation due to rapid blood clearance (Cortez-Retamozo, V., Lauwereys, M., Hassanzadeh Gh, G., Gobert, M., Conrath, K., Muyldermans, S., De Baetselier, P., & Revets, H. (2002) International journal of cancer 98, 456-462). Anti-CEA sdAbs were isolated and fused to the R-lactamase of Enterobacter cloacae.
The fusion protein was shown to efficiently activate prodrug in an in vitro study and induce tumor regression in an established tumor xenograft model (Cortez-Retamozo, V., Backmann, N., Senter, P. D., Wernery, U., De Baetselier, P., Muyldermans, S., &
Revets, H. (2004) Cancer research 64, 2853-2857). A similar approach was used to 5 link an sdAb against Type IV collagenase with an anti-tumor drug, lidamycin.
The fusion protein also demonstrated tumor growth inhibition (Miao, Q. F., Liu, X.
Y., Shang, B. Y., Ouyang, Z. G., & Zhen, Y. S. (2007) Anti-cancer drugs 18, 127-137).
sdAbs against EGFR (Roovers, R. C., Laeremans, T., Huang, L., De Taeye, S., Verkleij, A. J., Revets, H., de Haard, H. J., & van Bergen en Henegouwen, P.
M.
(2007) Cancer Immunol Immunother 56, 303-317) and its Type III variant (Omidfar, K., Rasaee, M. J., Modjtahedi, H., Forouzandeh, M., Taghikhani, M., &
Golmakani, N.
(2004) Tumour Biol 25, 296-305) have been isolated as well, but no in vivo data are available.

As such, there is a need for effective antibodies which are small in size and not very costly to produce on a large scale. There is also a need for sdAbs that have an Fc region to activate ADCC and CDC. Furthermore, there is a need for antibodies that show enhanced tumor targeting.

Hence, in light of the aforementioned, there is a need for an antibody, which, by virtue of its design and its components, would be able to overcome some of the above-discussed problems.

SUMMARY OF THE INVENTION:

A first object of the present invention is to provide sdAbs, as well as fusion proteins containing the same, directed towards EGFR.

That object is specifically achieved by providing a polypeptide sequence comprising a first region having an amino acid sequence substantially identical to SEQ ID NO:1, SEQ ID NO:2, SEQ ID NO:3 or SEQ ID NO:4, a second region having an amino acid sequence substantially identical to SEQ ID NO:5, SEQ ID NO:6, SEQ
ID NO:7, SEQ ID NO:8 or SEQ ID NO:9 and a third region having an amino acid

6 sequence substantially identical to SEQ ID NO:10, SEQ ID NO:11, SEQ ID NO:12 or SEQ ID NO:13.

Another object of the invention consists of a polynucleotide which encode by the polypeptide as defined above. A further object consists of a protein that comprises two to ten copies of a polypeptide as defined above, wherein the copies are identical or different.

An additional object of the present invention consists of a fusion protein with a binding specificity for EGFR that contains a heavy chain peptide comprising an amino acid sequence encoded by a Fc portion of an immunoglobulin gene, and an sdAb of the invention.

The invention is also directed towards chimeric polypeptide comprising a sdAb encoded by the polynucleotide as defined hereinabove and a another polypeptide such as but not limited to a toxin, a cytokine and an enzyme through protein fusion or conjugation.

Another object of the invention is to provide an immunoliposome comprising a polypeptide as defined above, wherein the polypeptide is the targeting moiety.
Another object of the present invention is to provide a method of diagnosing a cancer in a subject, wherein the method comprises the steps of:
- labelling the above described polypeptide with a contrast agent; and - administering an effective dose of the labelled polypeptide to said subject; and - detecting the signal generated by the contrast agent.
BRIEF DESCRIPTION OF THE DRAWINGS:

Figure 1 is a flow chart of the steps followed for the generation of the sdAbs.
Figure 2 represents the amino acid sequences of 11 sdAbs specific for EGFR-extracellular domain (ECD) with complementarity determining regions CDR1, CDR2 and CDR3 underlined. Based on the sequence identity of their CDRs, the sdAbs can

7 be divided into four (4) groups, which are separated by horizontal lines in the column in which the sdAb clones are listed. Group 1 consists of EG2, EG5 and EG28;
Group 2 consists of EG6 and EG10; Group 3 consists of EG7, EG16, EG29, EG30 and EG43; and Group 4 includes EG31. The frequency of the sequences for each clone is indicated in parentheses.

Figures 3A to 3C illustrate the characteristics of the sdAbs constructed.
Figure 3A is a schematic representation of the primary structures of an sdAb (EG2), a pentabody (V2C-EG2) and a cHCAb (EG2-hFc). Figure 3B is an SDS-PAGE of I pg purified EG2 (lane 1), V2C-EG2 (lane 2) and EG2-hFc (lane 3). Figure 3C
represents size exclusion chromatography of EG2, V2C-EG2 and EG2-hFc following EG2 and V2C-EG2 expression in E. co/i, and EG2-hFc expression in HEK293 cells.
Proteins were separated on an 8-25% gradient PhastGel (GE Healthcare) and Coomassie stained to visualize the proteins. Gel filtration chromatography was performed on purified EG2, V2C-EG2 and EG2-hFc using a Superdex 200TM column (GE
Healthcare). Superdex separations were carried out in PBS. The elution positions of molecular mass markers (GE Healthcare) are indicated. Data are normalized to a maximum 100 milliabsorbance unit (mAU).

Figure 4A to 4D show the interactions between EGFR-ECD and sdAbs as monitored by surface plasmon resonance. Figure 4A shows sensorgrams of the binding of 0.5 pM EG2, EG10, EG31 and EG43 to EGFR-ECD. The antigen was immobilized at a density of 500 RU on a CM5 sensor chip. For calculation of the affinities of the sdAbs, at least three independent experiments were performed using sdAb concentrations ranging from 0.4 nM to 1 pM. In Figures 4B and 4C, the binding of EG2, V2C-EG2 and EG2-hFc to surfaces with different antigen densities is shown.
On the same sensor chip in different flow cells, EGFR-ECD was immobilized at a density of 400 RU in Figure 4B, and at a density of 1500 RU in Figure 4C.
Binding of EG2, V2C-EG2 and EG2-hFc to antigen at different concentrations was analyzed;
only that of 0.5 pM is shown for each antibody. The data in Figures 4B and 4C
were normalized to a maximum RU of 100 in order to compensate for the different molecular weights of the binding proteins and allow comparison of resulting

8 sensorgrams. Figure 4D shows the interaction of EGFR-ECD with immobilized antibodies. EG2, V2C-EG2 and EG2-hFc were immobilized at a density of 300 RU.
Multiple concentrations of EGFR-ECD were used in the experiment, and only data at 0.5 pM is shown.

Figure 5A to 5C show fused microPET/CT images of a human pancreatic carcinoma model MIA PaCa-2. Mice bearing the established tumor were i.v.
injected with 64Cu-DOTA-EG2 at a dose of 396 pCi (Figure 5A), 64Cu-DOTA-V2C-EG2 at a dose of 393 pCi (Figure 5B) and 64Cu-DOTA-EG2-hFc at a dose of 438 pCi (Figure 5C). For EG2 and V2C-EG2, the mice were imaged at 1 hr, 4 hr and 20 hr post-injection (20 hr data not shown). For EG2-hFc, the mouse was imaged at 1 hr, 4 hr, hr and 44 hr post-injection. The top row in each sub-figure contains surface rendering images performed using AmiraTM (Mercury Computer System Inc.) to show relative tumor location (arrows). The bottom row in sub-figures 5A and 5B and top row in sub-figure 5C contains fused microPET/CT images. Images were acquired by 15 FLEX Trimodality micro CT/PET/SPECT system (Gamma Medica-Ideas Inc.).

DETAILED DESCRIPTION OF THE INVENTION:

As one skilled in the art already knows, antibody molecules are plasma proteins that bind specifically to particular molecules known as antigens, such as an 20 EGFR molecule, and are produced in response to immunization with such an antigen.

As used herein, the term "bind" or "binding" or any synonym thereof refers to the ability of a ligand, such as an antibody or an antibody fragment including single domain antibody (sdAb) to specifically recognize and detectably bind, as assayed by standard in vitro assays, to a EGFR molecule. For example, binding, as used herein, is measured by the capacity of an antibody, antibody fragment, or an antibody fragment to recognize an EGFR molecule on the surface of a cell using well described ligand-receptor binding assays, chemotaxis assays, histopathologic

9 analyses, flow cytometry and confocal microscopic analyses, and other assays known to those of skill in the art and/or exemplified herein.

Each antibody molecule has a unique structure that allows it to bind its specific antigen, but all conventional IgGs have the same overall structure consisting of two identical heavy chains and two identical light chains. The isotype of the heavy chain will specify the distinctive functional activity in the antibody molecule. The light chain consists of one V and one C domain and is disulfide bonded to the heavy chain.

The "light chains" of antibodies (immunoglobulins) from any vertebrate species can be assigned to one of two clearly distinct types, called kappa (k) and lambda (X), based on the amino acid sequences of their constant domains. Depending on the amino acid sequence of the constant domain of their heavy chains, immunoglobulins can be assigned to different classes. There are five major classes of immunoglobulins: IgA, IgD, IgE, IgG, and IgM, and several of these can be further divided into subclasses (isotypes), e.g., IgG1, IgG2, IgG3, IgG4, IgA1, and IgA2. The heavy chain constant domains that correspond to the different classes of immunoglobulins are called alpha, delta, epsilon, gamma, and mu, respectively.
The subunit structures and three-dimensional configurations of the different classes of immunoglobulins are well known.

Since variable regions of HCAbs, also called single-domain antibodies (sdAbs) or nanobodies, are considered excellent building blocks for antibody engineering due to their small size, enhanced thermostability and high production yield, the inventors of the present invention have isolated and engineered specific sdAbs directed towards EGFR.

As used in the present description, the term "isolation", "isolated" or "purified"
means altered "by the hand of man" from its natural state, i.e., if it occurs in nature, it has been changed or removed from its original environment, or both. For example, a polynucleotide naturally present in a living organism is not "isolated", the same polynucleotide separated from the coexisting materials of its natural state, obtained by cloning, amplification and/or chemical synthesis is "isolated" as the term is employed herein. Moreover, a polynucleotide that is introduced into an organism by transformation, genetic manipulation or by any other recombinant method is "isolated" even if it is still present in said organism.

As used in the present description, it is also understood that the term "sdAb", 5 "sdAbs", "nanobody" and "nanobodies" are all equivalent terms identifying the variable region of the HCAbs of the present invention. Other terms may also be used to identify the variable region of the HCAbs, and should not modify the scope of the present invention.

In this connection, it is therefore an embodiment of the invention to provide for

10 a sdAb and a fusion protein which comprises such an sdAb. The sdAb and fusion protein contemplated by the present invention have been developed by the present inventors to be used to target EGFRs on cell surfaces. As mentioned hereinabove, it is known that EGFR type receptors are mainly over-expressed on the surface of cancer cells.

1. Polynucleotides and polypeptides of the invention An sdAb of the present invention imay be consists',of a polypeptide comprising:
- a first region having an amino acid sequence substantially identical to SEQ
ID N0:1, SEQ ID NO:2, SEQ ID NO:3 or SEQ ID NO:4;

-a second region having an amino acid sequence substantially identical to SEQ ID NO:5, SEQ ID NO:6, SEQ ID NO:7, SEQ ID NO:8 or SEQ ID N0:9; and -a third region having an amino acid sequence substantially identical to SEQ
ID NO:10, SEQ ID NO:11, SEQ ID NO:12 or SEQ ID NO:13.

As one skilled in the art may appreciate, the first, second and third regions of the polypeptide of the inveniton correspond respectively to complementarity determining region (CDR) 1, CDR 2, and CDR 3.

As one skilled in the art may further appreciate, the polypeptide of the

11 invention may be encoded by a nucleic acid sequence substantially identical to SEQ
ID NO:14, SEQ ID NO:16, SEQ ID NO:18, SEQ ID NO:20, SEQ ID NO:22, SEQ ID
NO:24, SEQ ID NO:26, SEQ ID NO:28, SEQ ID NO:30, SEQ ID NO:32 or SEQ ID
NO:34.

In this connection and according to another embodiment, the present invention provides a polynucleotide encoding a polypeptide of the invention.

As used herein, the term "substantially identical", when referring to a nucleic acid sequence, is to be understood that the sequence of interest has a nucleic acid sequence which is at least 70% identical, or at least 80% identical, or at least 95%
identical to the nucleotide sequences contemplated by the present invention.

A sequence which "encodes" a selected polypeptide, is a nucleic acid molecule which is transcribed (in the case of DNA) and translated (in the case of mRNA) into a polypeptide, for example, in vivo when placed under the control of appropriate regulatory sequences (or "control elements"). The boundaries of the coding sequence are typically determined by a start codon at the 5' (amino) terminus and a translation stop codon at the 3' (carboxy) terminus. A coding sequence can include, but is not limited to, cDNA from viral, prokaryotic or eukaryotic mRNA, genomic DNA sequences from viral or prokaryotic DNA, and even synthetic DNA
sequences. A transcription termination sequence may be located 3' to the coding sequence. Other "control elements" may also be associated with a coding sequence.
A DNA sequence encoding a polypeptide can be optimized for expression in a selected cell by using the codons preferred by the selected cell to represent the DNA
copy of the desired polypeptide coding sequence.

The term "nucleic acid sequence" or "polynucleotide" or "nucleotide sequence"
as used interchangeably herein refers to any natural and synthetic linear and sequential arrays of nucleotides and nucleosides, for example cDNA, genomic DNA, mRNA, tRNA, oligonucleotides, oligonucleosides and derivatives thereof. For ease of discussion, such nucleic acids may be collectively referred to herein as "constructs", "plasmids" or "vectors." Representative examples of the nucleic acids of the present

12 invention include bacterial plasmid vectors including expression, cloning, cosmid and transformation vectors such as, but not limited to, pBR322, pSJF2, animal viral vectors such as, but not limited to, modified adenovirus, influenza virus, polio virus, pox virus, retrovirus, and the like, vectors derived from bacteriophage nucleic acid, and synthetic oligonucleotides like chemically synthesized DNA or RNA. The term "nucleic acid" further includes modified or derived nucleotides and nucleosides.

As used herein, "protein", "peptide" and "polypeptide" are used interchangeably to denote an amino acid polymer/residues or a set of two or more interacting or bound amino acid polymers/residues.

It will be understood that the polypeptide of the invention which refers to an sdAb may be encoded by, for instance, a nucleic acid sequence of SEQ ID NO:14, SEQ ID NO:16, SEQ ID NO:18, SEQ ID NO:20, SEQ ID NO:22, SEQ ID NO:24, SEQ
ID NO:26, SEQ ID NO:28, SEQ ID NO:30, SEQ ID NO:32 or SEQ ID NO:34.

As it may be further appreciated, the polypeptide of the invention may comprise an amino acid sequence substantially identical, for instance, to SEQ
ID
NO:15, SEQ ID NO:17, SEQ ID NO:19, SEQ ID NO:21, SEQ ID NO:23, SEQ ID
NO:25, SEQ ID NO:27, SEQ ID NO:29, SEQ ID NO:31, SEQ ID NO:33 and SEQ ID
NO:35.

As used herein, the term "substantially identical", when referring to an amino acid sequence, it is understood that the sequence of interest has an amino acid sequence which is at least 75% identical, or at least 85% identical, or at least 95%
identical to the amino acid sequences contemplated by the present invention.

As mentioned herein above, as shown in Figure 2, and according to the present invention, the sdAbs contemplated by the present invention can be divided into four (4) groups based on CDR sequence identity. Group 1 consists of EG2 (corresponding to SEQ ID NO: 14 and SEQ ID NO:15), EG5 (corresponding to SEQ
ID NO: 16 and SEQ ID NO:17) and EG28 (corresponding to SEQ ID NO: 18 and SEQ ID NO:19); Group 2 consists of EG6 (corresponding to SEQ ID NO:20 and SEQ

13 ID NO:11) and EG10 (corresponding to SEQ ID NO:22 and SEQ ID NO:23); Group 3 consists of EG7 (corresponding to SEQ ID NO:24 and SEQ ID NO:25), EG16 (corresponding to SEQ ID NO: 26 and SEQ ID NO:27), EG29 (corresponding to SEQ
ID NO:28 and SEQ ID NO:29), EG30 (corresponding to SEQ ID NO:30 and SEQ ID
NO:31) and EG43 (corresponding to SEQ ID NO:32 and SEQ ID NO:33); and Group 4 includes EG31 (corresponding to SEQ ID NO: 34 and SEQ ID NO:35). The groupings identified hereinabove are given so as to facilitate the understanding of the subject matter described in the section "EXAMPLE" provided herein below.

Yet another embodiment of the invention is to provide a protein which comprises two to ten copies of a polypeptide of the invention. It will be understood that the copies may be identical or different.

2. Fusion proteins, chimeric polypeptides and organisms of the invention According to another embodiment of the present invention, the invention provides for a fusion protein having a binding specificity to EGFR wherein said fusion protein comprises :
- a heavy chain peptide comprising an amino acid sequence encoded by a Fd portion of an immunoglobulin gene; and - a single variable domain peptidel.
As one skilled in the art may appreciate, the fusion protein of the invention may further comprise a linker so as to link the heavy chain peptide to the single variable domain peptide.
Typically, a linker will be a peptide linker moiety. The linker should be long enough and flexible enough to allow the single variable domain peptide to freely interact with a EFGR molecule. The linker, if required, may be for instance a peptide of at least two amino acid residues, or at least 5 amino acid residues, or at least 10 amino acids residues, or at least 15 amino acid residues. In other words, the linker contemplated by the present invention should link the carboxy-terminal of the sdAb

14 with the amino-terminal of the Fc region.

For example, a contemplated linker may have an amino acid sequence substantially identical to AEPKSCDKTHTCPPCP (or Ala-Glu-Pro-Lys-Ser-Cys-Asp-Lys-Thr-His-Thr-Cys-Pro-Pro-Cys-Pro).

A person skilled in the art will understand that the Fc region of an immunoglobulin molecule IgG is the crystallizable fragment produced when IgG
antibodies are cleaved with a papain enzyme. It is also understood that the Fc comprises the carboxy-terminal halves of the two heavy chains disulfide-bonded to each other. Furthermore, the Fc region is useful for recruiting the help of other cells and molecules to destroy and dispose of pathogens and confers functionally distinct properties to each of the various isotypes.

For instance, the Fc region may be derived from a human IgG1 immunoglobulin molecule and may consist of a nucleic acid sequence substantially identical to SEQ ID NO:36 which encodes an amino acid sequence of SEQ ID
NO:37. A person skilled in the field will also understand that the Fc region may also be derived from another IgG subclass such as IgG1, IgG2a, IgG2b, IgG3 or IgG4 or other immunoglobulin isotypes such as IgA, IgD, IgE and IgM. A person skilled in the field will also understand that the sources should be mammals including human or another source that need not be human.

According to another embodiment of the invention, the invention is concerned with a chimeric polypeptide comprising an sdAb encoded by the polynucleotide of the invention as defined herein above, fused or conjugated to any other polypeptide, which includes but is not limited to, a toxin, a cytokine, or an enzyme.

According to another embodiment of the invention, the invention is concerned with an immunoliposome of which an sdAb encoded by the polynucleotide of the invention serves as the targeting moiety.

A further embodiment of the invention concerns an organism that either expresses and secretes a polypeptide of the invention. As one skilled in the art will appreciate, the organism of the invention, such as a virus, expresses the polypeptide of the invention intracellularly, periplasmically or extracellularly.

3. Method of the invention 5 As mentioned previously, EGFRs are mainly over-expressed on the surface of cancer cells. As such, according to yet another embodiment of the invention, the present invention provides a method of diagnosing cancer in a subject, wherein the method comprises the step of administering to said subject (such as a human) an effective amount of the fusion protein described herein above, and the step of 10 detecting the fusion protein on cancerous cells of the subject.

As used herein, an "effective amount" is an amount sufficient to effect beneficial or desired results, including clinical results. An effective amount can be administered in one or more administrations. For purposes of this invention, an effective amount of the fusion protein is an amount that is sufficient to palliate,

15 ameliorate, stabilize, reverse, slow or delay the progression of the disease state.

According to another embodiment of the present invention, the invention is directed to a method of diagnosing cancer in a subject, wherein the method comprises the steps of:
- labelling a polypeptide of the invention with a contrast agent;
- injecting the labelled polypeptide in a subject, such as a human; and - detecting a signal generated by the contrast agent.

As one skilled in the art may appreciate, the contrast agent may be, but not limited to, a radionuclide, a fluorescent dye, a fluorescent nanoparticle, a magnetic contrast agent or a supermagnetic contrast agent.
It will be also be understood that the subject may further be for instance an experimental animal, such as a mouse, a rat, and a rabbit.
A person skilled in the art will understand that the fusion protein as described herein above is labelled with a marker or a contrast agent that will enable the imaging

16 of the fusion protein when it binds to a cell. For instance, the marker may be of the fluorescent, chemical or radioactive type. In this connection and as one skilled in the art will appreciate, the step of detecting the signal generated by a contrast agent, such as a radionuclide, is achieved for instance by a positron emission tomography (PET) scanner or micro-PET scanner.
In the case where the contrast agent is a fluorescent dye, a fluorescent nanoparticle, the step of detecting the signal generated by such a contrast agent is achieved for instance by an optical imaging scanner.
In the case where the contrast agent is a magnetic contrast agent or a supermagnetic contrast agent, the step of detecting the signal generated by such a contrast agent is achieved for instance by a magnetic resonounce tomography scanner.
Also, for instance, the fusion protein may be labeled with 64Cu and other radionuclides suitable for in vivo tumor imaging and used for imaging of a human pancreatic carcinoma established in mammals, such as nude mice.

Also, for instance, the fusion protein may be labeled with 64Cu and other radionuclides suitable for in vivo tumor imaging and used for imaging of tumor of a patient with pancreatic carcinoma.

Also, for instance, the fusion protein may be labeled with 64Cu and other radionuclides suitable for in vivo tumor imaging and used for imaging of tumor of a patient with other types of tumors that have EGFR expression.

EXAMPLE: GENERATION OF EGFR HCAb SINGLE-DOMAIN ANTIBODIES

The following example details the techniques employed in order to isolate eleven sdAbs from a phage display library constructed from the sdAb repertoire of a llama immunized with epidermal growth factor receptor vlll extracellular domain (EGFRvIII-ECD). More specifically, it details the steps identified in Figure 1 and provides the results and a discussion related thereto.

17 Briefly, pentameric sdAb, or pentabody, V2C-EG2 was constructed by fusing EG2, the sdAb with the highest affinity, to the D17E/W34A mutant of E. coli shiga toxin B subunit (StxB). cHCAb, EG2-hFc, was constructed by fusing EG2 to the Fc of human IgG1. E. coli expressed EG2 and V2C-EG2, and mammalian expressed EG2-hFc were tested for their tumor-targeting ability in a xenograft tumor model of human pancreatic carcinoma (MIA PaCa-2) in mice. Whereas EG2 and V2C-EG2 were shown to localize mainly in the kidneys after i.v. injection, EG2-hFc exhibited excellent tumor accumulation. As such, cHCAb is demonstrated to be one of the best antibody platforms for in vivo diagnostics and/or therapeutics for cancer.

Materials and Methods A.Construction and purification of the extracellular domains of EGFR and EGFRvIll.
Subcloning, production and purification of the extracellular domains of EGFR
(EGFR-ECD) and EGFRvIIi (EGFRvIII-ECD) was performed as previously described (29). Recombinant baculovirus containing the coding sequence for 6xHistidine (His)-tagged extracellular domains of EGFR or EGFRvIII (30) was used to infect Sf9 (Invitrogen, Burlington, ON) cells growing in suspension at 5-10 x 106 cells/ml.
Purification of the secreted proteins was performed by immobilized metal affinity chromatography (IMAC) using Ni-NTA-agarose (Qiagen, Mississauga, ON) following the manufacturer's instructions. Purified EGFR-ECD and EGFRvIII-ECD were confirmed by SDS-PAGE.

B. Isolation of EGFR-specific sdAbs from a llama immune phage display library.

A male llama (Lama glama) was injected subcutaneously with 100, 75, 75, 50 and 50 g EGFRvIII-ECD on days 1, 21, 36, 50 and 64, respectively (21).
Complete Freund's Adjuvant was used for primary immunization (Sigma, St. Louis, MO), incomplete Freund's Adjuvant for immunizations 2 - 4 (Sigma, St. Louis, MO), and no adjuvant for the final immunization. The llama was bled one week following each immunization and heparinized blood was collected for immediate isolation of the peripheral blood leukocytes, which were stored at -80 C until further use.

Total RNA was isolated from the leukocytes using QlAamp RNA Blood Mini Kit

18 (Qiagen, Mississauga, ON). cDNA was synthesized using pd(N)6 as primer and 566 ng total RNA as the template. Three different sense primers (called J' corresponding to the 5'-end of IgG) including MJ1 (GCCCAGCCGGCCATGGCCSMKGTGCAGCTGGTGGAKTCTGGGGGA), MJ2 (CAGCCGGCCATGGCCCAGGTAAAGCTGGAGGAGTCTGGGGGA) and MJ3 (GCCCAGCCGGCCATGGCCCAGGCTCAGGTACAGCTGGTGGAGTCT) and two anti-sense primers, corresponding to the CH2 domain DNA sequence, CH2 (CGCCATCAAGGTACCAGTTGA) and CH2b3 (GGGGTACCTGTCATCCACGGACCAGCTGA) were used to amplify the VH-CH1-Hinge-CH2 region of conventional IgG or VHH-Hinge-CH2. Amplified products of approximately 600 bp from the primer combination J'-CH2 were extracted from a 1%
agarose gel and purified with a QlAquick Gel Extraction Kit (Qiagen) and the amplified products from primers J'-CH2b3 were PCR purified. In a second PCR
reaction, the two primers of MJ7BACK
(CATGTGTAGACTCGCGGCCCAGCCGGCCATGGCC) and MJ8FOR
(CATGTGTAGATTCCTGGCCGGCCTGGCCTGAGGAGACGGTGACCTGG) were used to introduce Sfil restriction sites and to amplify the final sdAb fragments from the combined J'-CH2 and J'-CH2b3 amplified products (Arbabi Ghahroudi, M., Desmyter, A., Wyns, L., Hamers, R., & Muyldermans, S. (1997) FEBS letters 414, 521-526). The final PCR product was digested with Sfil and ligated into pMED1, a derivative of pHEN4, and transformed into E. coli TG1 (NEB, Ipswich, MA) by electroporation (Arbabi Ghahroudi, M., Desmyter, A., Wyns, L., Hamers, R., &
Muyldermans, S. (1997) FEBS letters 414, 521-526). Phages were rescued with helper phage M13K07 (NEB, Ipswich, MA).

The llama immune phage display library was panned against 1 mg/mI
EGFRvlll-ECD that was preadsorbed to a Reacti-BindTM maleic anhydride activated microtiter plate well. About 1011 phages were added to the well and incubated at 37 C for 2 hr for antigen binding. After disposal of unattached phages, the wells were washed six times with phosphate buffered saline supplemented with 0.05%
TweenTM
20 (PBST) for round one and washes were increased by one for each additional round. Phages were eluted by 10 min incubation with 100 pI 100 mM
triethylamine

19 and the eluate was subsequently neutralized with 200 NI 1M Tris-HCI (pH 7.5).
Phages were amplified as described above but on a smaller scale. After four rounds of panning, eluted phages were used to infect exponentially growing E. coli TG1.
Individual colonies were used in phage ELISA.

For phage ELISA, wells of a 96-well plate were coated overnight with 5 pg/ml EGFRvIII-ECD or EGFR-ECD and then blocked with 1% casein for 2 hr at 37 C.
Phages of individual clones were pre-blocked with casein overnight, added to the pre-blocked wells and incubated for 1 hr. Positive phage clones detected by standard ELISA procedure were sent for sequencing.

C. Construction of EG2 sdAbs and a pentabody.

DNA encoding four representative clones from each of the four groups, EG2, EG10, EG31 and EG43, was cloned into the Bbsl and BamHl sites of pSJF2 (Tanha, J., Muruganandam, A., & Stanimirovic, D. (2003) Methods in molecular medicine 89, 435-449).

EG2 was subcloned into the BspEl and BamHl sites of pVT2, generating an expression vector for pentabody V2C-EG2 (Stone, E., Hirama, T., Tanha, J., Tong-Sevinc, H., Li, S., MacKenzie, C. R., & Zhang, J. (2007) Journal of immunological methods 318, 88-94). EG2 sdAbs and V2C-EG2 pentabody were expressed periplasmically and purified by IMAC (Zhang, J., Li, Q., Nguyen, T. D., Tremblay, T.
L., Stone, E., To, R., Kelly, J., & Roger MacKenzie, C. (2004) Journal of molecular biology 341, 161-169).

Briefly, clones were inoculated in 25 ml LB-Ampicillin (Amp) and incubated at 37 C with 200 rpm shaking overnight. The next day, 20 ml of the culture was used to inoculate 1 I of M9 (0.2% glucose, 0.6% Na2HPO4, 0.3% KH2PO4, 0.1% NH4CI, 0.05% NaCI, 1 mM MgCI2, 0.1 mM CaCI2) supplemented with 0.4% casamino acids, 5 mg/I of vitamin B1 and 200 g/ml of Amp, and cultured for 24 hr. Next, 100 ml of 10 x TB nutrients (12% Tryptone, 24% yeast extract and 4% glycerol), 2ml of 100 mg/mI
Amp and 1 ml of 1 M isopropyl-beta-D-thiogalactopyranoside (IPTG) were added to the culture and incubation was continued for another 65-70 hr at 28 C with 200 rpm shaking. E. coli cells were harvested by centrifugation and lysed with lysozyme. Cell lysates were centrifuged, and clear supernatant was loaded onto High-TrapTM
chelating affinity columns (GE Healthcare, Uppsala, Sweden) and His-tagged proteins were purified.

5 D. Construction of cHCAb EG2-hFc.

Human Fc (hFc) gene comprising the nucleic acid sequence of SEQ ID NO:36 was inserted into a mammalian expression vector pTT5, a derivative of the pTT
vector (Durocher, Y., Perret, S. and Kamen, A. (2002) High Level and High-Throughput Recombinant Protein Production by Transient Transfection of the 10 Suspension-Growing Human 293-EBNA 1 Cell Line. Nucleic Acids Research, 30:e9) (34), to generate hFc fusion vector pTT5-hFc (J.Z., unpublished data). EG2 was amplified and inserted into pTT5-hFc so that the C-terminus of the sdAb was linked to the hinge region of human IgGl and then to Fc of human IgG1 with no extra residues added to the entire construct. The generated EG2-hFc was used in the 15 transient transfection of human embryonic kidney cells (HEK293).

Clone 6E of 293-EBNA1 (Y.D., unpublished data) was maintained as suspension culture in shaker flasks in serum-free F17 medium (Invitrogen, Burlington, ON). Cells were inoculated at a density of 0.25 x 106 cells/mI in a 2.5 I
shake flask (500 ml working volume) two days prior to transfection. Cells (usually at a

20 concentration of around 1.0-1.5 x 106 cells/mI) were transfected with 1 pg/mI plasmid DNA and 2 pg/mI linear 25 kDa polyethyleneimine, as previously described (Pham, P. L., Perret, S., Doan, H. C., Cass, B., St-Laurent, G., Kamen, A., &
Durocher, Y.
(2003) Biotechnology and bioengineering 84, 332-342). A feed with TN1 peptone (0.5%) was performed 24 hr post-transfection (Pham, P. L., Perret, S., Cass, B., Carpentier, E., St-Laurent, G., Bisson, L., Kamen, A., & Durocher, Y. (2005) Biotechnology and bioengineering 90, 332-344). Culture medium was harvested hr post-transfection.

EG2-hFc secreted into the medium was purified by affinity chromatography on a Protein A column (MabSelect SuRe, GE Healthcare, Uppsala, Sweden). Purified

21 material was desalted on a HiPrepTM 26/10 desaiting column (GE Healthcare, Uppsala, Sweden) equilibrated with phosphate buffered saline (PBS). Protein concentration was determined by absorbance at 280 using a molar extinction coefficient of 58830 calculated from EG2-hFc amino acid sequence (Gill, S.C.
and von Hippel, P.H. (1989) Calculation of protein extinction coefficients from amino acid sequence data. Anal. Biochem. 182:319-326(1989). [PubMed: 2610349].
E. Surface Plasmon Resonance Analysis.

Experiments were performed using a BIACORE 3000T"' optical sensor platform and research grade CM5 sensor chips (GE Healthcare, Uppsala, Sweden).
EGFR-ECD, sdAbs and multivalent sdAb constructs were immobilized on the sensor chip surface by standard amine coupling. All experiments were carried out in HEPES
buffer [10 mM HEPES (pH 7.4), 150 mM NaCI, 3.4 mM EDTA, 0.005% Tween 20] at 25 C. Antibodies were injected at different concentrations (0.004 - 1 pM) at a flow rate of 30 NI/min unless otherwise indicated. The amount of analyte bound after subtraction from the blank control surface is shown as relative resonance units (RU).
The double referenced sensorgrams from each injection series were analyzed for binding kinetics using BlAevaluation software (GE Healthcare, Uppsala, Sweden).
Dissociation constants (KD) were calculated from the on- and off-rates (k n and k ff, respectively), as determined by global fitting of the experimental data to a 1:1 Langmuir binding model (Chi2<1). The final reported KD was from at least three independent experiments.

F. Size Exclusion Chromatography.

Gel filtration chromatography of EG2, V2C-EG2 and EG2-hFc was performed on Superdex 200TM (GE Healthcare, Uppsala, Sweden). Superdex separations were carried out in PBS. Low molecular weight markers ribonuclease A (13.7 kDa), chymotrypsin A (25 kDa) and ovalbumin (43 kDa) were used to calculate the molecular weight of EG2. High MW markers catalase (232 kDa), ferritin (440 kDa), thyroglobulin (669 kDa) and blue dextran (2000 kDa) were used to calculate the molecular weight of V2C-EG2 and EG2-hFc.

22 G. 64Cu-labeling of antibodies.

1,4,7,10-tetraazacyclododecane-N,N',N",N"'-tetraacetic acid (DOTA) was activated by N-hydroxysulfosuccinimide (sulfo-NHS) and 1-ethyl-3-[3-(dimethylamino)propyl] carbodiimide (EDC) in a mixture solution (pH 5.5) at 4 C for 30 min. Purified antibody was reacted with a 1,000:1,000:100:1 molar ratio of DOTA:sulfo-NHS:EDC:antibody in 0.1 M Na2HPO4 (pH 7.5) at 4 C for 12-16 hr.
After conjugation, the reaction mixture was centrifuged repeatedly through a YM-30 centricon with 30 mM ammonium citrate buffer (pH 6.5) to remove unconjugated small molecules. The purified conjugate was concentrated to 1 mg/ml in 30 mM
ammonium citrate buffer and stored at -20 C for further use. 64Cu (64CuC12 in 0.1 M
HCI; radionuclide purity > 99%) was purchased from Washington University (St.
Louis, MO).

Typically, 150 pg of DOTA-conjugated antibody and 1 mCi of 64Cu were incubated in 30 mM ammonium citrate (pH 6.5) at 43 C for 45 min. The reaction was terminated by addition of 5 NI 10 mM diethylenetriaminepentaacetic acid solution.
Labeled antibody was separated by a size exclusion Bio-SpinTM 6 column (Biorad, Mississauga, ON). Radiolabeling efficiency was determined by integrating peak areas on Fast Protein Liquid Chromatography (FPLC) chromatograms and determining the percentage of radioactivity associated with the antibody peaks.

H. MicroPET/CT imaging techniques.

The human pancreatic carcinoma cell line MIA PaCa-2 was maintained in DMEM (Gibco, Gaithersburg, MD) supplemented with 10% fetal bovine serum (FBS;
Gibco, Gaithersburg, MD). Six-week old female nude mice were obtained from Harlan Laboratories. MIA PaCa-2 pancreatic cancer cells (3 x 106 in sterile saline) were injected subcutaneously into the right flank of the animals. The animal models were imaged when tumors reached the size of 300-500 mm3. About 400 pCi/120 pg of 64Cu-DOTA-antibody was administered via tail vain injection to mice under Metofane anesthesia. The animals were allowed free access to food and water.
The mice were re-anesthetized and imaged using microPET/CT scanner at the time points indicated. MicroPET/CT imaging of mice was performed using a tri-modality

23 microPET/CT/SPECT imager (Gamma Medica FLEX Inc., CA) for functional and anatomical imaging. MicroCT has an X-ray tube of 80 kVp, 0.5 mA fixed anode with tungsten target to provide anatomical imaging with spatial resolution of -100 pm. X-ray CT has a 4.72" bore suitable for imaging small animals. Images were acquired at a fast scan time of 1 min and reconstructed using cone beam filtered back-projection (modified Feldkamp) reconstruction algorithm with streak artifact reduction.
Live animal images were acquired at low radiation doses (1.2 cGy) for 1 min fly mode scan. The microPET scanner has a solid ring design of bismuth germanate (BGO) detector blocks and continuous automatic photomultiplier gain stabilization technology. The 16.5 cm diameter ring is located in the same gantry. The scanner provides a 10 cm transaxial and 11.6 cm axial field of view. The scanner is capable of an axial and transaxial resolution of 2 mm. Images were reconstructed using filtered back-projection (2D OSEM) and 3D filtered back-projection (3D OSEM).

1. Quantification of microPET data.

The calibration factor to convert PET image units of counts/sec/voxel to pCi/cc was calculated from a mouse-sized cylinder with a known concentration of '$F
in water assuming a tissue density of 1 g/cc. No additional attenuation correction was applied. The conversion of positron activity of18F to that of 64Cu was carried out by the ratio of the branching ratios of the positron decay of the isotopes. The calculated concentrations of radioactivity were multiplied by the volume of each region of interest (ROI) to determine total radioactivity present within regions. ROI
was analyzed using Analyzer AVW 3.0 software (Biomedical Imaging Resource, Mayo Foundation, Rochester, MN).

Results Isolation and characterization of sdAbs. Isolation of EGFR-specific sdAbs was achieved by llama immunization with EGFRvIII-ECD, construction of immune phage display library and subsequent panning. Following the 5th immunization, peripheral leukocytes were isolated from llama blood, total RNA was isolated and cDNA
synthesized. DNA encoding the variable regions of HCAbs was amplified and flanked with Sfil restriction sites using nested PCR. The amplified DNA was digested with Sfil

24 restriction enzyme and ligated into pMED1 (M. Arbabi, unpublished results).
The ligation products were transformed into E. coli TG1 cells, generating an immune sdAb phagemid library with a size of 5.5x107, which is rescued by helper phage M 13K07.

Four rounds of phage display panning were performed on immobilized EGFRvIII-ECD, and phage enrichment was observed with the pannings (data not shown). Phage ELISA demonstrated 44 of the 45 clones were positive for EGFRvIII-ECD binding (data not shown). Phage ELISA on EGFR-ECD indicated that these phages bound to wild type EGFR as well. Analysis of encoding sequences of the sdAbs displayed on the phage clones revealed 11 different sdAb genes. These sdAbs can be divided into four groups based on CDR sequence identity (see Figure 2).

One sdAb gene from each of the four groups was chosen and subcloned into an E. coli periplasmic expression vector, pSJF2, (Tanha, J., Muruganandam, A., &
Stanimirovic, D. (2003) Methods in molecular medicine 89, 435-449), generating four clones pEG2, pEG10, pEG31 and pEG43 (data not shown). The four sdAbs, each tagged with a 6xHistidine (His) at their C-termini (Fig. 4A, represented by EG2), were produced in E. coli and purified by IMAC. The yields of EG2, EG10, EG31 and were 11, 19.4, 7.8 and 43 mg per liter of TG1 culture, respectively.

The four anti-EGFR sdAbs were analyzed for their binding to EGFR-ECD by surface plasmon resonance. The on-rates of the sdAbs were quite similar, but their off-rates have significant differences. The dissociation constants (Kps) of the sdAbs range from 55 nM (EG2) to 440 nM (EG31) (Fig. 4A and Table 1).

Construction and characterization of EG2 pentabody and EG2 cHCAb.
EG2, the sdAb with the highest affinity for EGFR-ECD, was used to construct pentabody and cHCAb. DNA encoding EG2 was amplified by PCR and flanked with restriction sites BspEl and BamHl. The amplified DNA was digested and ligated into the pentamerization vector pVT2 (Stone, E., Hirama, T., Tanha, J., Tong-Sevinc, H., Li, S., MacKenzie, C. R., & Zhang, J. (2007) Journal of immunological methods 318, 88-94) digested with the same enzymes. The generated clone expresses pentameric EG2, V2C-EG2 (Fig. 3A), which was purified by IMAC. The yield of V2C-EG2 was 43.9 mg per liter of culture.

To generate EG2 cHCAb, sdAb was amplified and cloned into HCAb vector 5 pTT5-hFc (J.Z. unpublished results), which is designed to fuse a protein to the Fc of human IgG1. The generated clone EG2-hFc (Fig. 3A) was used to transiently transfect HEK293 cells. EG2-hFc produced by the cells was purified by Protein A
affinity chromatography. The yield of EG2-hFc was 21 mg per liter of HEK293 culture. Sequence analysis of EG2-hFc indicated that a Glu to Val mutation at 10 position 5 of EG2 occurred during PCR amplification of EG2 but this did not affect the binding of EG2-hFc to EGFR (Fig. 4).

EG2, V2C-EG2 and EG2-hFc were subjected by SDS-PAGE and size exclusion chromatography to analyze their subunits and molecular masses.
Denatured EG2 sdAb, V2C-EG2 pentabody and EG2-hFc cHCAb migrated at 15 13 kDa, 21 kDa and 37 kDa, respectively (Fig. 3B). Size exclusion chromatography results indicate that EG2, V2C-EG2 and EG2-hFc have molecular weights of 14 kDa, 108 kDa and 90 kDa, respectively (Fig. 3C). These results indicate that EG2 exists as a monomer, V2C-EG2 as a pentamer and EG2-hFc as a dimer. The measured size of V2C-EG2 (108 kDa) is slightly smaller than the predicted size (126 kDa).
20 Nevertheless, it is still considered to be a pentamer based on the approximation of the two data and our previous results of other pentabodies.

To evaluate the impact of multivalency on the functional affinities of V2C-EG2 and EG2-hFc, the binding profiles of these molecules were analyzed by SPR on the same EGFR-ECD surface. Oligomerization of EG2 sdAb, in either a dimeric or

25 pentameric format, resulted in higher apparent affinity (Fig. 4B and 4C).
Although both proteins showed a slightly slower kon compared to EG2, the main difference in apparent affinity is due to their koffs. The binding avidity for the multivalent constructs appears to increase with higher surface density (comparing Fig. 4B and 4C). In contrast, the KD for EG2, the monomeric sdAb, was not affected by antigen density (Fig. 4B and 4C).

26 To confirm that the higher apparent affinities of V2C-EG2 and EG2-hFc were due to the multivalency, EGFR-ECD binding to immobilized EG2, V2C-EG2 and EG2-hFc was analyzed. Sensorgrams of the interactions showed that EGFR-ECD
binds to all three proteins, either monomer, dimer or pentamer, with nearly identical affinity (Fig. 4D). This result confirms that the apparent affinity improvements are due solely to avidity effects.

MicroPET/CT imaging of human pancreatic carcinoma model in nude mice using the constructed antibodies. EG2, V2C-EG2 and EG2-hFc were labeled with 64Cu and used for imaging of a human pancreatic carcinoma model MIA
PaCa-2 established in nude mice. MicroPET/CT fused images indicated that the majority of EG2 and V2C-EG2 localizes in the kidneys within 1 hr post-injection (Fig 5A and 5B). Both proteins are barely detectable in the tumor at 1 hr, 4 hr and 20 hr.
In contrast, microPET/CT images of mice administered with EG2-hFc reveal gradual accumulation of EG2-hFc in the tumor even at 44 hr post-injection (Fig. 5C).
In addition, quite good tumor to muscle contrast is observed at 20 hr post-injection, and the contour of the tumor in the PET image matches the true tumor shape quite well.
Discussion The isolation of eleven sdAbs targeting EGFR and the construction of a pentabody (V2C-EG2) and a cHCAb (EG2-hFc) based on one of the sdAbs, EG2, is described herein. The three types of antibodies were radiolabeled with 64Cu and microPET/CT
imaging was used to analyze their in vivo distribution in a MIA PaCa-2 human pancreatic carcinoma xenograft mouse model. As expected, the sdAb was cleared from the circulation rapidly after injection. The pentabody, despite its large size (126 kDa), behaved like the sdAb. In contrast, the cHCAb accumulated in tumor over time and showed excellent tumor-targeting ability. This indicates that cHCAb, not sdAb and pentabody, is an appropriate sdAb-based tumor-targeting molecule.

The tumor targeting ability of an antibody relies a great deal on two factors:
serum clearance rate and tumor penetration rate (Graff, C. P. & Wittrup, K. D.
(2003) Cancer research 63, 1288-1296), which are related to affinity, size and the antibody

27 Fc. Intact Ig molecules are most frequently used in therapy due to their prolonged serum half-lives, which are usually longer than 100 hours. Truncated antibodies with a complete Fc, such as scFv fused to CH2-CH3 (Fc), are cleared at a rate slightly faster than the original mAb (Xu, X., Clarke, P., Szalai, G., Shively, J. E., Williams, L.
E., Shyr, Y., Shi, E., & Primus, F. J. (2000) Cancer research 60, 4475-4484 and Slavin-Chiorini, D. C., Kashmiri, S. V., Schlom, J., Calvo, B., Shu, L. M., Schott, M.
E., Milenic, D. E., Snoy, P., Carrasquillo, J., Anderson, K., et al. (1995) Cancer research 55, 5957s-5967s). In contrast, antibody fragments lacking Fc, such as scFv and Fab, are rapidly cleared from circulation by glomerular filtration and have a much shorter serum half-life than molecules with an Fc (Khawli, L. A., Biela, B., Hu, P., &
Epstein, A. L. (2003) Hybridoma and hybridomics 22, 1-9). Even when the Fc was only partially removed, as in the case of CH2-deleted antibody, the serum half-life is drastically reduced (Slavin-Chiorini, D. C., Horan Hand, P. H., Kashmiri, S.
V., Calvo, B., Zaremba, S., & Schlom, J. (1993) International journal of cancer 53, 97-103).

In contrast to serum half-life, tumor penetration is a more difficult parameter to measure. Smaller antibody fragments have been shown to penetrate into deeper areas of tumor tissue (Buchegger, F., Haskell, C. M., Schreyer, M., Scazziga, B. R., Randin, S., Carrel, S., & Mach, J. P. (1983) The Journal of experimental medicine 158, 413-427). However, the faster tumor penetration rate didn't result in improved tumor targeting because of accelerated clearance. Furthermore, removal of Fc would abrogate the induction of ADCC and CDC, which are generally critical for antibody therapy.

It is therefore difficult to retain the Fc while satisfying the moderate size requirement for good tumor penetration by antibodies. The relatively small size (-14 kDa) of sdAbs makes it possible to fulfill both requirements. EG2-hFc reported here has a complete human Fc and yet is only approximately 80 kDa in size. We refer to this type of molecule as chimeric HCAb because of its human Fc and llama sdAb.
Fully human HCAb (hHCAb) can be constructed if human sdAbs (17, 43) are used.
It is noteworthy that the advantage of the potentially low immunogenicity of fully human HCAb should not be overemphasized. Human sdAbs, even those from a stable sdAb

28 framework, often have less satisfactory biophysical properties as compared to llama VHHs (J. Z., unpublished results, and J. Tanha, personal communization). Given that high thermostability of antibodies is essential for in vivo tumor targeting, a balance must be obtained between biophysical properties and immunogenicity (Willuda, J., Honegger, A., Waibel, R., Schubiger, P. A., Stahel, R., Zangemeister-Wittke, U., &
Pluckthun, A. (1999) Cancer research 59, 5758-5767).

HCAbs are likely to be very similar to chimeric IgGs with regard to glycosylation pattern and immunogenicity. The eukaryotic expression system, HEK293 cell line, is expected to provide a glycosylation pattern similar to that of human IgGs. EG2 has 68% sequence identity to a human VH (AAA53000.1), very similar to that between chimeric VH and human myeloma VH (68-75%) and humanized VH and human myeloma VH (69-74%) (Clark, M. (2000) Immunology today 21, 397-402). This suggests that cHCAb is likely to have immunogenicity similar to that of chimeric or humanized antibodies.

The yield of the cHCAb, EG2-hFc, is 21 mg per liter of culture after transient transfection and purification. A yield of over 100 mg per liter of culture was achieved from a very similar construct, indicating the potential of reaching very high expression for cHCAb.

HCAbs, either cHCAb or hHCAb, have the potential to serve as better therapeutic antibody formats in comparison to conventional IgGs because of the following advantages: 1) potentially, better tumor penetration; 2) potentially, higher production yield due to simpler molecular structure; 3) potentially, lower dose requirement due to molecular weight that is approximately half that of IgG and 4) easier fusion to other entities such as cytokines. And the last, but perhaps the most important advantage of sdAb-based antibodies is their ability to target the so-called hidden epitopes that are inaccessible to conventional IgGs (Lauwereys, M., Arbabi Ghahroudi, M., Desmyter, A., Kinne, J., Holzer, W., De Genst, E., Wyns, L., &
Muyldermans, S. (1998) The EMBO joumal 17, 3512-3520 and Stijlemans, B., Conrath, K., Cortez-Retamozo, V., Van Xong, H., Wyns, L., Senter, P., Revets, H., De Baetselier, P., Muyldermans, S., & Magez, S. (2004) The Joumal of biological

29 chemistry 279, 1256-1261).

Much can be done to improve the tumor targeting ability of EG2-hFc. First, the affinity of EG2 can be improved. It has been shown that antibodies with a Kp of 10-9 M to 10"10 M obtained tumor accumulations that are not only higher than those with moderate affinity (10"' M to 10-8 M), but also higher than an antibody with very high affinity (10-" M) (Adams, G. P., Schier, R., McCall, A. M., Simmons, H. H., Horak, E.
M., Alpaugh, R. K., Marks, J. D., & Weiner, L. M. (2001) Cancer research 61, 4755)). It is therefore assumed that better tumor targeting can be achieved when EG2 is replaced by an sdAb with a higher affinity. Second, the wild type Fc can be replaced by mutants with prolonged (Hinton, P. R., Xiong, J. M., Johlfs, M.
G., Tang, M. T., Keller, S., & Tsurushita, N. (2006) J Immunol 176, 346-356) or shortened (Kenanova, V., Olafsen, T., Crow, D. M., Sundaresan, G., Subbarayan, M., Carter, N.
H., Ikle, D. N., Yazaki, P. J., Chatziioannou, A. F., Gambhir, S. S., et al.
(2005) Cancer research 65, 622-631) serum half-life, depending on the ultimate purpose of therapy or in vivo diagnosis. Third, cytokines can be fused to EG2-hFc, rendering the antibody capable of cytokine-mediated cytotoxicity yet maintaining a relatively low molecular weight. In the search for the best tumor targeting antibodies for either in vivo diagnostics or therapeutics, all these strategies can of course be combined.

Another tested antibody platform, the pentabody, failed to show successful tumor targeting. The pentabody was initially generated by fusing an sdAb to the B
subunit of E. coli shiga toxin (StxB) (Zhang, J., Tanha, J., Hirama, T., Khieu, N. H., To, R., Tong-Sevinc, H., Stone, E., Brisson, J. R., & MacKenzie, C. R. (2004) Journal of molecular biology 335, 49-56). This simple fusion resulted in a 1,000 to 10,000-fold increase in functional affinity of the sdAb to corresponding antigen densely immobilized on a solid surface. It was shown that pentabodies can be excellent antibody molecules for biomedical research (Zhang, J., Li, Q., Nguyen, T. D., Tremblay, T. L., Stone, E., To, R., Kelly, J., & Roger MacKenzie, C. (2004) Journal of molecular biology 341, 161-169), tumor diagnosis (Mai, K. T., Perkins, D. G., Zhang, J., & Mackenzie, C. R. (2006) Histopathology 49, 515-522) and in certain cases therapeutic candidates (Stone, E., Hirama, T., Chen, W., Soltyk, A. L., Brunton, J., MacKenzie, C. R., & Zhang, J. (2007) Molecular immunology 44, 2487-2491).

From in vivo distribution data of scFvs and their multivalent counterparts, diabody, tetrabody and (Fab)2, we predicted pentabody would demonstrate better tumor targeting ability due to its larger size (126 kDa) and five antigen binding sites.
5 However, our results indicate that V2C-EG2 performance is not improved over that of EG2 in this regard.

Development of sdAb-based therapeutics has lagged behind conventional IgGs. The cHCAb antibody format presented here can become an effective antibody drug platform that has the potential to exceed the efficacy of conventional IgGs.

10 Although a preferred embodiment of the present invention has been described in detail herein and illustrated in the accompanying drawings, it is to be understood that the invention is not limited to this embodiment and that various changes and modifications could be made without departing from the scope and spirit of the present invention.

Table 1. Kinetic Rate Constants and Equilibrium Rate Constants of anti-EGFR
sdAbs interacting with EGFR-ECD.

kon (M-'S 1) (3.7 0.25) X105 (2.5 0.07) ) (3.2 0.5) x105 (1.17 0.09) x105 , k ff (S ) (2.1 0.2) x10-2 (3.2 0.1) x10-2 (1.4 0.02) x10"1 (3.7 0.5) x10"2 Ko(nM) 55 10 126 7 440 100 316 25 ANNEX I

SEQ ID NO: 1 DYVMG

Asp Tyr Val Met Gly SEQ ID NO:2 SYAMG

1 o Ser Tyr Ala Met Gly SEQ ID NO:3 FDAWG
IDAWG

LDAWG

(Phe or Ile or Leu) Asp Ala Trp Gly SEQ ID NO:4 INAIG
I le Asn Ala I le Gly SEQ ID NO:5 AISRN GLTTR YADSV KG

Ala Ile Ser Arg Asn Gly Leu Thr Thr Arg Tyr Ala Asp Ser Val Lys Gly SEQ ID N0:6 AISGR SSIRN YDDSV KG

Ala Ile Ser Gly Arg Ser Ser Ile Arg Asn Tyr Asp Asp Ser Val Lys Gly SEQ ID NO:7 WGST GSTSY ADFVK G

Val Val Gly Ser Thr Gly Ser Thr Ser Tyr Ala Asp Phe Val Lys Gly SEQ ID N0:8 LVGST GSTSY ADSVK G
LVGSD GSTSY ADSVK G

Leu Val Gly Ser (Thr or Asp) Gly Ser Thr Ser Tyr Ala Asp Ser Val Lys Gly SEQ ID NO:9 RITSD GRTIL EDSVK G

Arg Ile Thr Ser Asp Gly Arg Thr Ile Leu Glu Asp Ser Val Lys Gly SEQ ID N0:10 NSAGT YVSPR SREYD Y

NSAGT YVSPR SRDYD G

Asn Ser Ala Gly Thr Tyr Val Ser Pro Arg Ser Arg (Glu or Asp) Tyr Asp (Tyr or Gly) SEQ ID N0:11 DTVFRSFWGNVKE

Asp Thr Val Phe Arg Ser Phe Val Val Gly Asn Val Lys Glu SEQ ID NO:12 RFQSLYNS
RFDSLYNS

Arg Phe (GIn or Asp) Ser Leu Tyr Asn Ser SEQ ID NO:13 EKGGSPLY
Glu Lys Gly Gly Ser Pro Leu Tyr EG2 - SEQ ID NO:14 CAGGTAAAGCTGGAGGAGTCTGGGGGAGGATTGGTGCAGGCTGGGGACTCTC
TGAGAGTCTCCTGTGCAGCCTCTGGACGCGACTTCAGTGATTATGTCATGGGCT

AATGGTCTTACGACTCGCTATGCAGACTCCGTGAAGGGCCGATTTACCATCTCC
AGAGACAATGACAAAAACATGGTGTACCTGCAAATGAACAGCCTGAAACCTGAG
GACACGGCCGTTTATTACTGTGCAGTAAATTCGGCCGGGACATACGTTAGTCCC
CGCTCGAGAGAGTATGACTACTGGGGCCAGGGGACCCAGGTCACCGTCTCCTC
A

EG2 - SEQ ID NO:15 QVKLEESGGGLVQAGDSLRVSCAASGRDFSDYVMGWFRQAPGKEREFVAAISRN
GLTTRYADSVKGRFTISRDNDKNMVYLQMNSLKPEDTAVYYCAVNSAGTYVSPRS
REYDYWGQGTQVTVSS

EG5 - SEQ ID NO:16 CAGGTAAAGCTGGAGGAGTCTGGGGGAGGATTGGTGCAGGCTGGGGACTCTC
TGAGACTCTCCTGTGTAGACTCTGGACGCGACTTCAGTGATTATGTCATGGGCT
GGTTCCGCCAGGCTCCAGGGAAGGAGCGTGAGTTTGTAGCAGCTATTAGCAGG

AGAGACAATGACAAAAACACGGTGTACCTGCAAATGAACAGCCTGAGACCTGAG
GACACGGCCGTTTATTACTGTGCAACAAATTCGGCCGGGACATACGTCAGTCCC
CGCTCGAGAGACTATGACGGCTGGGGCCAGGGGACCCAGGTCACCGTCTCCT
CA

EG5 - SEQ ID NO:17 QVKLEESGGGLVQAGDSLRLSCVDSGRDFSDYVMGWFRQAPGKEREFVAAISRN
GITTRYADSVKGRFTISRDNDKNTVYLQMNSLRPEDTAVYYCATNSAGTYVSPRSR
DYDGWGQGTQVTVSS

EG28 - SEQ ID NO:18 CAGGTACAGCTGGTGGAGTCTGGGGGAGGATTGGTGCAGGCTGGGGACTCTC
TGAGACTCTCCTGTGTAGACTCTGGACGCGACTTCAGTGATTATGTCATGGGCT
GGTTCCGCCAGGCTCCAGGGAAGGAGCGTGAGTTTGTAGCAGCTATTAGCAGG
AATGGTATTACGACACGCTATGCAGACTCCGTGAAGGGCCGATTTACCATCTCC
AGAGACAATGACAAAAACACGGTGTACCTGCAAATGAACAGCCTGAAACCTGAG
GACACGGCCGTTTATTACTGTGCAACAAATTCGGCCGGGACATACGTCAGTCCC
CGCTCGAGAGACTATGACGGCTGGGGCCAGGGGACCCAGGTCACCGTCTCCT

CA

EG28 - SEQ ID NO:19 QVQLVESGGGLVQAGDSLRLSCVDSGRDFSDYVMGWFRQAPGKEREFVAAISRN
GITTRYADSVKGRFTISRDNDKNTVYLQMNSLKPEDTAVYYCATNSAGTYVSPRSR
DYDGWGQGTQVTVSS

EG6- SEQ ID NO:20 CAGGTAAAGCTGGAGGAGTCTGGGGGAGGATTGGTGCAGGCTGGGGGCTCTC
1o TGACCCTCTCCTGTGCAGCCTCTGGAGGCACCTTCAGTAGCTATGCCATGGGCT
GGTTCCGCCAGGCTCCAGGGAAGGAGCGTGAGTTTGTAGCAGCTATTAGCGGG
CGTAGTTCTATAAGAAACTATGATGACTCCGTGAAGGGCCGATTCGCCATCTCC
AGAGACAACGCCAAGAACACGGTGTATCTGCAAATGAACAGCCTGAAACCTGAG
GACACGGCCGTTTATTATTGTGCAGCAGATACGGTATTCCGGTCGTTTGTTGTT
GGCAACGTTAAAGAATGGGGTCAGGGGACCCAGGTCACCGTCTCCTCA

EG6 - SEQ ID NO:21 QVKLEESGGGLVQAGGSLTLSCAASGGTFSSYAMGWFRQAPGKEREFVAAISGRS
SIRNYDDSVKGRFAISRDNAKNTVYLQMNSLKPEDTAVYYCAADTVFRSFWGNVK
EWGQGTQVTVSS

EG10 - SEQ ID NO:22 CAGGTGCAGCTGGTGGAGTCTGGGGGAGGATTGGTGCAGGCTGGGGGCTCTC
TGACCCTCTCCTGTGCAGCCTCTGGAGGCACCTTCAGTAGCTATGCCATGGGCT
GGTTCCGCCAGGCTCCAGGGAAGGAGCGTGAGTTTGTAGCAGCTATTAGCGGG
CGTAGTTCTATAAGAAACTATGATGACTCCGTGAAGGGCCGATTCGCCATCTCC
AGAGACAGCGCCAAGAACACGGTGTATCTGCAAATGAACAGCCTGAAACCTGA
GGACACGGCCGTTTATTATTGTGCAGCAGATACGGTATTCCGGTCGTTTGTTGT

TGGCAACGTTAAAGAATGGGGTCAGGGGACCCAGGTCACCGTCTCCTCA
EG10 - SEQ ID NO:23 QVQLVESGGGLVQAGGSLTLSCAASGGTFSSYAMGWFRQAPGKEREFVAAISGRS
SIRNYDDSVKGRFAISRDSAKNTVYLQMNSLKPEDTAVYYCAADTVFRSFWGNVK
EWGQGTQVTVSS

EG7 - SEQ ID NO:24 CAGGTACAGCTGGTGGAGTCTGGGGGAGGCTTGGTGCAGCCTGGGGGGTCTC
TGAGACTCTCCTGTGCAGCCTCTGAAAGCTTCTTCAATTTCGATGCCTGGGGCT
GGTACCGCCAGGCTCCAGGGAAGCAGCGCGAAATGGTCGCCGTAGTTGGTAGT
ACTGGTAGCACAAGTTATGCAGACTTTGTGAAGGGCCGATTCACCATCTCCAGA
GACAACGCCAACAACACGGTGTATCTGCAAATGAACACCCTGAGACCTGAGGA
CACGGCCGTCTATTACTGTTATGCGAGGTTTCAGAGCTTGTATAACTCCTGGGG
CCAGGGGACCCAGGTCACCGTCTCCTCA

EG7 - SEQ ID NO:25 QVQLVESGGGLVQPGGSLRLSCAASESFFNFDAWGWYRQAPGKQREMVAWGS
TGSTSYADFVKGRFTISRDNANNTVYLQMNTLRPEDTAVYYCYARFQSLYNSWGQ
GTQVTVSS

EG16 - SEQ ID NO:26 GCCGTGCAGCTGGTGGAGTCTGGGGGAGGCTTGGTGCAGCCTGGGGGGTCTC
TGAGACTCTCCTGTGCAGCCTCTGTTAGCATCTTCGATATCGATGCCTGGGGCT
GGTACCGCCAGGCTCCAGGGAAGCAGCGCGAAATGGTCGCGTTAGTTGGTAGT
ACTGGTAGCACAAGTTATGCAGACTCCGTGAAGGGCCGATTCACCCTCTCCAGA
GACAACGTCAACAACACGATGTATCTGCAAATGAACAGCCTGAGACCTGAGGAC
ACGGCCGTCTATTACTGTTATGCGAGGTTTGATAGCTTGTATAACTCTTGGGGC

CAGGGGACCCAGGTCACCGTCTCCTCA
EG16 - SEQ ID NO:27 AVQLVESGGGLVQPGGSLRLSCAASVSIFDIDAWGWYRQAPGKQREMVALVGSTG
STSYADSVKGRFTLSRDNVNNTMYLQMNSLRPEDTAVYYCYARFDSLYNSWGQGT
QVTVSS

EG29 - SEQ ID NO:28 CAGGTAAAGCTGGAGGAGTCTGGGGGAGGCTTGGTGCAGCCTGGGGGGTCTC
TGAGACTCTCCTGTGCAGCCTCTGAAAGCTTCTTCAATTTCGATGCCTGGGGCT
GGTACCGCCAGGCTCCAGGGAAGCAGCGCGAAATGGTCGCCGTAGTTGGTAGT
ACTGGTAGCACAAGTTATGCAGACTTTGTGAAGGGCCGATTCACCATCTCCAGA
GACAACGCCAACAACACGGTGTATCTGCAAATGAACACCCTGAGACCTGAGGA
CACGGCCGTCTATTACTGTTATGCGAGGTTTCAGAGCTTGTATAACTCCTGGGG
CCAGGGGACCCAGGTCACCGTCTCCTCA

EG29 - SEQ ID NO:29 QVKLEESGGGLVQPGGSLRLSCAASESFFNFDAWGWYRQAPGKQREMVAWGST
GSTSYADFVKGRFTISRDNANNTVYLQMNTLRPEDTAVYYCYARFQSLYNSWGQG
TQVTVSS

EG30 - SEQ ID NO:30 CAGGTAAAGCTGGAGGAGTCTGGGGGAGGCTTGGTGCAGCCTGGGGGGTCTC
TGAGACTCTCCTGTGCAGCCTCTGTTAGCATCTTCGATATCGATGCCTGGGGCT
GGTACCGCCAGGCTCCAGGGAAGCAGCGCGAAATGGTCGCGTTAGTTGGTAGT
ACTGGTAGCACAAGTTATGCAGACTCCGTGAAGGGCCGATTCACCCTCTCCAGA
GACAACGTCAACAACACGATGTATCTGCAAATGAACAGCCTGAGACCTGAGGAC
ACGGCCGTCTATTACTGTTATGCGAGGTTTGATAGCTTGTATAACTCTTGGGGC

CAGGGGACCCAGGTCACCGTCTCCTCA
EG30 - SEQ ID NO:31 QVKLEESGGGLVQPGGSLRLSCAASVSIFDIDAWGWYRQAPGKQREMVALVGSTG
STSYADSVKGRFTLSRDNVNNTMYLQMNSLRPEDTAVYYCYARFDSLYNSWGQGT
QVTVSS

EG43 - SEQ ID NO:32 CAGGTGCAGCTGGTGGAGTCTGGGGGAGGCTTGGTGCAGCCTGGGGGGTCTC
TGAGACTCCCCTGTGCAGCCTCTGGAAGCATCTTCAGTCTCGATGCCTGGGGC
TGGTACCGCCAGGCTCCAGGGAAGCAGCGCGAAATGGTCGCGTTAGTTGGTAG
TGACGGTAGCACAAGTTATGCAGACTCCGTGAAGGGCCGATTCACCATCTCCAG
AGACAACGCCAACAACACATTTTATCTGCAAATGAACAGCCTGAAACCTGAGGA
CACGGCCGTCTATTACTGTTATGCGAGGTTTCAAAGCTTGTATAACTCCTGGGG
CCAGGGGACCCAGGTCACCGTCTCCTCA

EG43 - SEQ ID NO:33 QVQLVESGGGLVQPGGSLRLPCAASGSIFSLDAWGWYRQAPGKQREMVALVGSD
GSTSYADSVKGRFTISRDNANNTFYLQMNSLKPEDTAVYYCYARFQSLYNSWGQG

EG31 - SEQ ID NO:34 CAGGTAAAGCTGGAGGAGTCTGGGGGAGGCTTGGTGCAGGTTGGGGGGTCTC
TGAGACTCTCCTGTGCACACTCTGGGCTGCCCTTCGGTATCAATGCCATCGGCT
GGTACCGCCAGGGTCCTGGGAATCAGCGCGACTTGGTCGCACGTATTACTAGT
GATGGTCGCACGATATTGGAAGACTCCGTGAAGGGCCGATTCACCATCTCCAG
AGACAACGCCAAGAAGACGGTATATGTGCAAATGAACAACCTGAAACCTGAGGA
CACGGCCGTGTATTACTGTGCTGCAGAGAAGGGGGGTAGTCCGCTCTACTGGG

GCCAGGGGACCCAGGTCACCGTCTCCTCA
EG31 - SEQ ID NO:35 QVKLEESGGGLVQVGGSLRLSCAHSGLPFGINAIGWYRQGPGNQRDLVARITSDG

VTVSS

hFc - SEQ ID NO:36 GCACCTGAACTCCTGGGGGGACCGTCAGTCTTCCTCTTCCCCCCAAAACCCAA
10 GGACACCCTCATGATCTCCCGGACCCCTGAGGTCacatgcgtggtggtggacgtgagccac gaagaccctgaggtcaagttcaactggtacgtggacggcgtggaggtgcataatgccaagacaaagccgcgggag gagcagtacaacagcacgtaccgtgtggtcagcgtcctcaccgtcctgcaccaggactggctgaatggcaaggagt acaagtgcaaggtctccaacaaagccctcccagcccccatcgagaaaaccatctccaaagccaaagggcagccc cgagaaccacaggtgtacaccctgcccccatcccgggatgagctgaccaagaaccaggtcagcctgacctgcctg 15 gtcaaaggcttctatcccagcgacatcgccgtggagtgggagagcaatgggcagccggagaacaactacaagac cacgcctcccgtgctggactccgacggctccttcttcctctacagcaagctcaccgtggacaagagcaggtggcagc aggggaacgtcttctcatgctccgtgatgcatgagggtctgcacaaccactacacgcagaagagcctctccctgtctcc gggtaaa 20 hFc - SEQ ID NO:37 APELLGGPSVFLFPPKPKDTLMISRTPEVTCVWDVSHEDPEVKFNWYVDGVEVHN
AKTKPREEQYNSTYRWSVLTVLHQDWLNGKEYKCKVSNKALPAPIEKTISKAKGQ
PREPQVYTLPPSRDELTKNQVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPPVL
DSDGSFFLYSKLTVDKSRWQQGNVFSCSVMHEGLHNHYTQKSLSLSPGK

EG2-hFc - SEQ ID NO:38 EG2-hFc amino acid sequence QVKLVESGGGLVQAGDSLRVSCAASGRDFSDYVMGWFRQAPGKEREFVAAISRN

GLTTRYADSVKGRFTISRDNDKNMVYLQMNSLKPEDTAVYYCAVNSAGTYVSPRS
REYDYWGQGTQVTVSSAEPKSCDKTHTCPPCPAPELLGGPSVFLFPPKPKDTLMIS
RTPEVTCVWDVSHEDPEVKFNWYVDGVEVHNAKTKPREEQYNSTYRWSVLTVL
HQDWLNGKEYKCKVSNKALPAPIEKTISKAKGQPREPQVYTLPPSRDELTKNQVSL

TCLVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYSKLTVDKSRWQQG
NVFSCSVMHEGLHNHYTQKSLSLSPGK
EG2-hFc - SEQ ID NO:39 CAGGTTAAGCTGGTGGAGTCTGGGGGAGGATTGGTGCAGGCTGGGGACTCTCT
GAGAGTCTCCTGTGCAGCCTCTGGACGCGACTTCAGTGATTATGTCATGGGCTG
GTTCCGCCAGGCTCCAGGGAAGGAGCGTGAGTTTGTAGCAGCTATTAGCAGGA
ATGGTCTTACGACTCGCTATGCAGACTCCGTGAAGGGCCGATTTACCATCTCCA
GAGACAATGACAAAAACATGGTGTACCTGCAAATGAACAGCCTGAAACCTGAGG
ACACGGCCGTTTATTACTGTGCAGTAAATTCGGCCGGGACATACGTTAGTCCCC

GCTCGAGAGAGTATGACTACTGGGGCCAGGGGACCCAGGTCACCGTCTCCTCA
GCTGAGCCCAAATCTTGTGACAAAACTCACACATGCCCACCGTGCCCAGCACCT
GAACTCCTGGGGGGACCGTCAGTCTTCCTCTTCCCCCCAAAACCCAAGGACAC
CCTCATGATCTCCCGGACCCCTGAGGTCacatgcgtggtggtggacgtgagccacgaagaccct gaggtcaagttcaactggtacgtggacggcgtggaggtgcataatgccaagacaaagccgcgggaggagcagta caacagcacgtaccgtgtggtcagcgtcctcaccgtcctgcaccaggactggctgaatggcaaggagtacaagtgc aaggtctccaacaaagccctcccagcccccatcgagaaaaccatctccaaagccaaagggcagccccgagaac cacaggtgtacaccctgcccccatcccgggatgagctgaccaagaaccaggtcagcctgacctgcctggtcaaagg cttctatcccagcgacatcgccgtggagtgggagagcaatgggcagccggagaacaactacaagaccacgcctcc cgtgctggactccgacggctccttcttcctctacagcaagctcaccgtggacaagagcaggtggcagcaggggaac gtcttctcatgctccgtgatgcatgagggtctgcacaaccactacacgcagaagagcctctccctgtctccgggtaaa

Claims

WHAT IS CLAIMED IS:

1. All novel and inventive subject matter disclosed herein.