top of page



2020 - PRESENT

Cell-free biosynthesis of conjugate vaccines


2015 - 2020

Engineering microorganisms for the efficient synthesis of capsular polysaccharides and related enzymes


2014 - 2015

Protein stabilization by insertional fusion


Cornell University, Smith School of Chemical & Biomolecular Engineering

Advisor: Prof. Matthew DeLisa

Cell-Free Biosynthesis of Conjugate Vaccines

Conjugate vaccines are proven to protect against infection by bacterial pathogens, representing a powerful approach for expanding the number of available vaccines and tackling currently unmet medical needs, especially in the developing world. However, conventional technologies for making conjugate vaccines are technically complex and rely on living cells, which necessitates highly centralized manufacturing, specialized equipment, lengthy development timelines, and cold-chain distribution.

With the support of a Cornell Presidential Postdoctoral Fellowship, I am working on developing an unconventional cell-free conjugate vaccine synthesis technology, which combines cell-free protein synthesis of licensed vaccine carrier proteins with N-linked glycosylation of bacterial polysaccharide antigens in freeze-dried lysates derived from detoxified, nonpathogenic Escherichia coli.

The outcome of this work will help to ameliorate the burden of enteric infections in children by enabling small-batch vaccine manufacturing at the point of care. Ultimately, this project aims to revolutionize vaccine manufacturing through the development of a modular platform that can be applied to accelerate the development of conjugates for diarrhea and extended to other vaccine-preventable diseases, for communities that are impoverished and/or isolated.


Acute diarrheal diseases rank 2nd amongst all infectious diseases as a killer in children below 5
years of age worldwide.

Source: Black RE, Morris SS, Bryce J. Where and why are 10 million children dying every year?


Conventional technologies are technically complex and present economic and logistical challenges that limit the reach of vaccination campaigns, especially in remote or low-resource settings. This can slow response to pathogen outbreaks.


  • easy to use

  • deployable in field conditions that lack a refrigerated supply chain

  • capable of delivering vaccine products that are safe, affordable and effective.

Research: postdoc



Rensselaer Polytechnic Institute, Howard P. Isermann Department of Chemical & Biological Engineering​

Research Advisors: Profs. Mattheos Koffas & Robert Linhardt

Engineering Microorganisms for the Efficient Synthesis of Capsular Polysaccharides and Related Enzymes

Heparin (HP) and chondroitin sulfate (CS) are glycosaminoglycans (GAGs) with many biological and physiological functions that contribute to their widespread use as anticoagulants and anti-inflammatory drugs. These sulfated polysaccharides are primarily extracted and purified from animal tissues, where factors like source material, manufacturing processes, and the presence of contaminants impact overall safety and biological and pharmacological efficacy. 

As part of ongoing efforts to separate the food chain from the drug chain, my graduate work explored engineering microbial metabolism in order to produce CS in Escherichia coli, and heparosan, a valuable precursor to heparin, in Bacillus megaterium. The projects summarized below tackle challenges associated with bioengineering approaches for HP and CS production and analysis.

Can engineered cells be used as drug factories to replace animal-sourced drugs?

Our low-cost bioengineering approach helps to move production past its currently limited supply chain – providing superior, safer, and more readily available drugs to meet worldwide demand in a more sustainable manner.

(i) Post-Polymerization Modification: Expression of chondroitin sulfotransferases in  E. coli.

The engineered production of non-sulfated chondroitin in moderate quantities in various microorganisms has been reported, but no industrial scale biotechnological process using E. coli to produce CS has been established. We engineering E. coli for the production of CS through a biotransformation scheme, where the microbial non-sulfated chondroitin backbone is sulfated in vitro using multiple E. coli strains to generate the required components.

The utility of this system involves the expression of several active biosynthetic pathway enzymes and requires an examination of their specificity and stability. To this end, we recombinantly expressed chondroitin 6-sulfotransferase in E. coli in active form and made activity improvements to E. coli-expressed chondroitin-4-sulfotransferase through protein engineering strategies adopted to develop more stable mutants. This was used to achieve for the first time, in vivo chondroitin sulfate production without the need for animal products. 

(ii) Precursor Synthesis: Heparosan Production in Bacillus megaterium

Although heparosan production in engineered Bacillus subtilis has been previously reported, the larger GRAS organism, Bacillus megaterium, is a more attractive alternative for industrial scale production since it possesses the intrinsic favorable properties of low protease activity and high secretion capability. We have engineered B. megaterium for heparosan production to exploit these attributes. The T7 RNA polymerase (T7 RNAP) expression system for B. megaterium, which allows tightly regulated and efficient induction of genes of interest, has been co-opted for control of heparosan synthase (PmHS2). Specifically, we showed for the first time that engineered B. megaterium MS941 can produce heparosan upon induction with xylose, providing an alternative, safe source of heparosan as a precursor for heparin production.

(iii) Structural Analysis: Cloning, expression, and characterization of chondroitinase AC II.

Chondroitinase ACII, a polysaccharide lyase isolated from Arthrobacter aurescens, has been widely used to aid the analysis and study of chondroitin and CS structure, but it is no longer commercially available. Two versions of recombinant chondroitinase ACII were prepared in E. coli, based on a nucleotide sequence derived from the crystal structure of the original ACII from Seikagaku. Their activity, stability, specificity and action pattern were examined, to fill this great need within the glycobiology research community. They were compared to a natively secreted A. aurescens enzyme, which displayed some subtle specificity differences towards uncommon CS substrates and can serve as a suitable replacement for the original, discontinued enzyme. Furthermore, the recombinant E. coli expressed versions of chondroitinase ACII that we developed and characterized, are suitable for general application in the structural determination of most standard chondroitin sulfates.

The sole supplier of this important enzyme, Seikagaku Corporation in Japan, discontinued its production after loss of the bacterial strain in the 2011 earthquake and tsunami. This resulted in a great need within the academic and industrial glycobiology research communities.

grad research
Research: undergrad


New York University, Department of Chemical & Biomolecular Engineering

Research Advisor: Prof. Jin Ryoun Kim 

Protein Stabilization by Insertional Fusion

As an undergraduate research assistant, I helped to develop a new and potentially general method of protein stabilization via insertional fusion of a relatively unstable “guest” enzyme into a highly stable “host” thermophilic protein; preventing the restriction of available sequence spaces or compromising other intrinsic properties.

Insertional fusion to Pyrococcus furiosus maltodextrin‐binding protein (PfMBP) has been shown to increase the  thermodynamic and kinetic stability of β‐lactamase (BLA) homologues and suppress their aggregation. This observed improvement in stability is derived from enthalpic as well as entropic mechanisms. Through our work, we demonstrated that insertional fusion to PfMBP might be a generally applicable means to improve protein stability and benefit protein evolution.

This novel method of protein stabilization does not restrict available sequence spaces or compromise other intrinsic  properties.

bottom of page