Two lectin-binding methods were investigated as possible ways of monitoring the glycosylation of human monoclonal antibodies during their development and production. Carbohydrate composition was assessed in various preparations that were produced in different host cell types, cell sublines or batches of the same cells. The lectin binding was measured with ELISA and surface plasmon resonance (SPR). For comparative purposes, the monosaccharide content of many of the preparations was also measured by high-pressure anion exchange chromatography (HPAEC)/pulsed amperometric detection (PAD). Both lectin methods detected modifications in glycosylation when antibodies were produced in different ways; SPR was more sensitive than ELISA for some lectins and vice versa. Generally, the lectin results agreed with those obtained by the monosaccharide analysis; however, the former were much better for assessing N -acetylneuraminic acid changes. The latter were impossible to assess by HPAEC/PAD because of their low levels. The lectin-based methods also had the advantages that they were quicker to perform and required less expertise and could quickly identify structures that monosaccharide analysis might miss. It is suggested that, in the development of therapeutic proteins, monosaccharide analysis and/or oligosaccharide profiling is initially performed but later routine batches of the glycoprotein are screened with a lectin method. Of the two lectin methods used, SPR is much quicker when performing a screen, whereas ELISA is particularly useful for comparing a particular carbohydrate feature on different samples of the same glycoprotein.

A recent genome-wide association study (GWAS) for urinary bladder cancer (UBC) has identified multiple novel genetic risk factors (Rothman et. al, Nat Gen, 2010). One of these factors was an intronic SNP rs11892031 (p=1.0×10-7) located within the UGT1A locus on chromosome 2q37. The human cellular detoxification uridine 5’ diphosphate (UDP)-glucuronosyltransferases (UGTs) belong to a superfamily of proteins that conjugate diverse endo and exotoxins, increase their solubility and facilitate their removal via bile and urine. The UGT1A locus includes nine protein-coding genes and four pseudogenes. Each first exon of these genes is regulated by its own promoter and is spliced to four constant exons producing nine UGT1A proteins. Due to the complexity of the region and ∼90-95% similarity between the substrate-binding exon 1 sequences of UGT1A genes, this region is poorly represented in public databases (HapMap, 1000 Genomes, dbSNP). To ensure specificity of detection, we generated long-range amplicons and sequenced all UGT1A exons in HapMap individuals (CEU) and in 44 bladder cancer patients and identified 25 non-synonymous coding variations. All these variations have been genotyped in 1000 bladder cancer cases and 1000 controls from the Spanish Bladder Cancer Study (SBCS). We found one exonic SNP that showed association stronger than the original GWAS variant, and multiple variants that showed some evidence of gene and cigarette smoking interactions. Having ∼20 coding and potentially functional variants within the UGT1A region, we are exploring methods of simultaneous analysis of individual variants and their haplotypes. In conclusion, association of genetic variants within the UGT1A region confirms that altered cellular detoxification of environmental substrates is an important factor for development of bladder cancer. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 4680. doi:10.1158/1538-7445.AM2011-4680