Advisor(s)

William S. Hancock

Contributor(s)

Paul Vourus, Penny J. Beuning, Zoran Sosic

Date of Award

2011

Date Accepted

11-2011

Degree Grantor

Northeastern University

Degree Level

Ph.D.

Degree Name

Doctor of Philosophy

Department or Academic Unit

College of Science. Department of Chemistry and Chemical Biology.

Keywords

chemistry, analytical chemistry, MALDI-TOF MS method, glycomic

Subject Categories

Glycomics, Biochemical markers

Disciplines

Biochemistry

Abstract

Glycosylation has been determined to be involved in essentially all aspects of life, ranging from development to ageing to health and to disease. In the last 25 years with the advent of recombinant biotherapeutics, it has even become pivotal in drug discovery and development. On the heels of genomics and the completion of the human genome, came the emergence of other spin-off or daughter "omics" fields. Included in this category is glycomics; the study of the "entire set of glycans in an organism". With glycomic studies, came the observation that glycosylation changes are associated with various diseases and disease states. In response to these observations, methods were developed to characterize and quantitate these changes, which have led to advances in biomarker discovery. Meanwhile in the biopharmaceutical industry, numerous efforts were being made to control, characterize, monitor and quantitate product glycosylation. As often is the case, these efforts were seemingly separated by an imaginary line between academia and industry. For the most part, this is understandable as the needs of each "side" are inherently different. Academic pursuits are often lengthy and labor intensive resulting in highly detailed information, while industrial applications are often more focused on robust, rapid analysis that provides very specific types of information. The goal of this work was to erase this line by developing a method that is universally applicable to both academia and industry.

An innovative method utilizing well-established chemistries, commercially-available reagents and commonly-used instruments has been developed and applied to three different types of samples: (1) a purified biotherapeutic (2) CHO cells (3) human plasma. The method employs reductive amination of PNGase F-released glycans with carbon-isotopes of anthranilic acid and quantitative analysis by MALDI-TOF MS. The method displays high reproducibility (CV generally less than 10%) and data accuracy (generally 90% or greater) with a linear dynamic range over two orders of magnitude and sub picomolar sensitivity.

The method was applied to the analysis of a purified biotherapeutic over the course of a bioreactor campaign; six time points were sampled. The change in glycosylation over time (at each time point relative to the first time point) was quantitated, and it was observed that sialylation showed a marked decrease at the final time point. As sialylation often plays a role in the pharmacokinetic properties of biotherapeutics, this type of information may be of importance and could be used to modify cell culture parameters.

The method was also applied to the glycomic analysis of biotherapeutic-expressing CHO cells over the course of a bioreactor campaign; two different cell lines expressing two different monoclonal antibodies were sampled at four time points. The change in glycosylation over time was quantitated. Numerous glycans were observed; some increased over time, while others decreased. The impact of this information is unknown at this point and will require further investigation. The method was used in this application to demonstrate the feasibility of its use for quantitative glycomic investigations of CHO cells. In future experiments, it will be used to investigate whether CHO cell glycosylation has any impact on cell health and/or product quality.

The method was then applied to the glycomic analysis of normal vs. diseased human plasma. The method was combined with other preparative methods to establish a platform methodology for the purification and quantitative glycomic analysis of human plasma. Differences in glycosylation between normal and diseased plasma were observed and quantitated. Diseased plasma was observed be elevated in sialylation, by as much as 500% for a particular glycan. The observation of increased sialylation was expected as other cancers have been observed to show increases in sialylation. The method was utilized in this setting to demonstrate proof of concept for its use in biomarker discovery.

Document Type

Dissertation

Rights Holder

Samnang Tep


View Full Text

Click button above to open, or right-click to save.

Included in

Biochemistry Commons

Share

COinS