dc.description.abstract
Cancer is the second leading cause of death for humans around the world. Aberrant cell surface glycosylation is a hallmark for cancer and can be used for targeted cancer therapy. Two such aberrant tumour-associated carbohydrate antigens (TACAs) are Globo-H and sialyl Lewis A (sLe A). Globo-H is expressed in various epithelial cancers, such as breast cancer, the highest incidence cancer in women. Its expression leads to cancer progression through angiogenesis, immunosuppression, and tumour survival. On the other hand, sLe A is so far the only FDA approved biomarker for pancreatic cancer that is often only detected at a later stage and, therefore, has a high mortality rate. Sialyl Le A has also been implicated in metastasis, tumour-associated inflammation, and immune evasion. Therefore, both TACAs are important targets, but the number of tools available to study them are limited.
Synthetic, pure, and well-defined glycans were used to obtain antibodies (Abs) and nanobodies (Nbs) against sLe A and Globo-H, respectively. I demonstrated through various methods that Abs and Nbs generated through this technique are highly specific for their targets both in vitro with synthetic glycans as well as in vivo with native glycans.
For the first part of my thesis, I studied sLe A-targeting Abs, GB11 and HA8. I compared their amino acid sequences to that of a commercial Ab, 1116-NS-19-9, clinically used to target sLe A and diagnose pancreatic cancer. I compared the binding of the three Abs to sLe A and obtained a crystal structure for GB11 to unravel the molecular origin of its specificity and higher affinity, compared to the commercial counterpart, 1116-NS19-9.
For the second part of my thesis, I showed that a Nb, GH46, is the best binder for Globo-H in a previously generated Nb library. I deduced its binding epitope using different methods and generated a multivalent construct of GH46 by successfully cloning and expressing GH46-trimer. The trimer has better binding to native Globo-H on cells than GH46-monomer. Furthermore, I used a site-directed mutagenesis strategy to improve upon solubility, thermostability, and affinity of GH46. I demonstrated that the framework region 3 of the Nb plays an important role in Nb stability. Through this, I generated a GH46 mutant which retains specificity for Globo-
24
H but gives 100-fold higher yield, 7-fold better binding affinity in vitro, and 2-fold better binding in vivo.
In summary, I show with this thesis how the immunisation with synthetic glycans can generate highly specific Abs and Nbs, which could be used in biomedical and biotechnological applications in the future particularly in the framework of cancer therapy. As proof of concept, two new Abs targeting sLe A were introduced and characterised that could potentially be used for early diagnosis of pancreatic cancer. Additionally, the first in class Nb binding to a TACA was characterised and functionalised to a multimer. The Nb structure was studied in depth and may add to the current knowledge available on structural stability of Nbs.
en
