We are committed to constantly improve our detection methods. While the current method uses the quorum sensing system, it is thus limited to bacteria that secrete autoinducers. Hence, we present an alternative approach for the detection of pathogens. In our alternative detection system, biomolecules are tagged with a fluorescent reporter that bind to the surface of the cell and reveal its presence.

Aachen 14-10-13 Galectin-3 iNB.png

Natural Functions of Galectin-3

Galectins are proteins of the lectin family, which possess carbohydrate recognition domains binding specifically to β-galactoside sugar residues. In humans, 10 different galantines have been identified, among which is galectin-3. Galectin-3 has a size of about 31 kDA and is encoded by a single gene, LGALS3. It is found to have many physiological functions, such as cell adhesion, cell growth and differentiation, and contributes to the development of cancer, inflammation, fibrosis and others.

Human galectin-3 is a protein of the lectin-family that was shown to bind the LPS of multiple human pathogens. Some of them, including P. aeruginosa protect themselves against the human immune system by mimicking the lipopolysaccharides (LPS) present on human erythrocytes. By making fusion proteins of galectin-3 with fluorescent reporter proteins, pathogens can be labelled and made visible by fluorescence-microscopy.

Aachen 14-10-13 Galectin-3-YFP iNB.png

An Alternative Sensing Molecule

Aachen 14-10-09 Pseudomonas LPS iNB.png
Cell wall composition of P. aeruginosa
Gram-negative bacteria have two cell membranes. The LPS are embedded in the outer membrane and are composed of a lipid and an O polysaccharide.

Characteristic parts of the lipopolysaccharide structure (LPS) of P. aeruginosa can be bound by galectin-3. Specifically, the O polysaccharide (see figure on the left) of the LPS is recognized by galectin-3. Therefore, this specific binding of galectin-3 enables the construction of a fluorescent based detection system. A fusion protein of galectin-3 and a reporter protein, such as a fluorescent protein, can be built and applied in the detection of P. aeruginosa.

In our approach, a galectin-3-YFP fusion protein is built and expressed in E. coli, including a his-tag and a snap-tag for purification. The fusion protein can then be incorporated into a cell-free biosensor system. These biosensors have many advantages over systems that use living cells such as an uncomplicated storage. Furthermore, from a biosafety and social acceptance point of view, it is advantageous if the sensor system does not contain alive genetically modified organisms.

Aachen 14-10-09 Cell Free Biosensor iNB.png

To detect P. aeruginosa cells, an agar chip could be used to sample a solid surface. However, other materials but agar can be considered to collect pathogens. The cells stick to the sampling chip which is then immersed in a detection buffer containing the galectin-3-YFP fusion protein. Excess protein is removed during washing in a suitable buffer. The galectin-3 remains bound to the pathogen and illumination with 514 nm, the excitation wavelength of YFP, in a modified version of our measurement device reveals the location of the cells. The picture taken by the measurement device can then be analyzed by our software Measurarty.

Aachen 14-10-15 Medal Cellocks iNB.png


Due to the generous support of Sophia Böcker and Prof. Dr. Elling of the Helmholtz Institute for Biomedical Engineering in Aachen, we got access to a pET17-derived expression plasmid for a His- and SNAP-tagged YFP-galectin-3 fusion protein. We transformed the fusion protein into E. coli Rosetta cells and conducted a batch fermentation to obtain large amounts of protein.

With the help of David Schönauer and Alan Mertens from the RWTH Aachen Institute of Biotechnolgy we then purified the fusion protein using FPLC. Subsequently, we aimed to test the binding of the Gal-3 fusion protein to the LPS of P. aeruginosa as shown previously (Kupper, Böcker, Liu et al., 2013). Apparently, because of insufficient sensitivity of the used fluorescence microscope, this could not be confirmed and would require further experiments, idealy using other detection methods.

After we received the collection of pSBX-expression vectors from Team Heidelberg, we used Gibson assembly to make K1319020 from K1319003 and pSBX1A3, which is the translational unit for an mRFP-Gal3 fusion protein with a C-terminal 6xHis tag:

Aachen K1319020.png
This BioBrick is a construction intermediate of K1319003 (gal3), E1010 (mRFP), K1319007 (6xHis tag) to K1319020 (translational unit of the fusion protein).

In addition, we cloned our BioBrick K1319003 into the pET17 expression vector and expressed all combinations of fusion proteins in E. coli BL21(DE3). An SDS-PAGE showed that all fusion proteins were fully translated:

Aachen 14-10-04 Expression Pellets iMO.png Aachen Gal3 Expression.png
Pellets of different fusion protein expressions
Expression in the pET17 vector was much more leaky than the expression in the pSBX vectors.
SDS-PAGE of K1319020 expression
The fusion protein was fully translated to the correct molecular mass of 74 kDa.


Kupper, C. E., Böcker, S., Elling, L., Lui, H., Adamzyk, C., de Kamp, J. v., et al. (2013). Fluorescent SNAP-tag galectin fusion proteins as novel tools in glycobiology. Current Pharmaceutical Design, 19(30), 5457-67. Available at: