Biofunctionalized nanoporous aluminum oxide culture chips : for capture and growth of bacteria

Porous aluminum oxide (PAO) is a nanostructured material used for various biotechnological applications, including the culturing microorganisms and other types of cells. The ability to chemically modify the PAO surface and tailor its surface properties is a promising way to expand and refine its applications. The immobilization of biomolecules on PAO that specifically interact with and bind to target bacteria would enable the capture and subsequent growth of bacteria on the same surface, and this was the ultimate goal of the research presented in this thesis. After a general introduction to the overall subject of this thesis, presented in Chapter 1, the most commonly used and recent methods to prepare glycosurfaces are reviewed and compared on their merits and drawbacks in Chapter 2. Although there are a great number of techniques, the main challenge that still remains is to develop an accessible, reproducible and inexpensive approach that produces well-defined and stable glycosurfaces using as few steps as possible. The most used analytical techniques for the characterization of glycosurfaces and several applications of these surfaces in the binding, capture, and sensing of bacteria and bacterial toxins were also discussed in Chapter 2. Biofunctionalization of surfaces in general requires a stepwise approach, in which it is very important to have a stable monolayer as the first step. At the beginning of this research it was known that various functional groups were able to react with (porous) aluminum oxide, but there was no comprehensive study comparing the stability of these modified surfaces under the conditions that are important for microbiological applications. In Chapter 3, the PAO surface was modified with various functional groups known to react with PAO (carboxylic acid, α-hydroxycarboxylic acid, alkyne, alkene, phosphonic acid, and silane), and the stability of these modified surfaces was assessed over a range of pH and temperatures that are relevant for microbial growth. Silane and phosphonate-modified PAO surfaces with a hydrophobic monolayer proved to be the most stable ones, but the phosphonate modification was both more easily applied and reproducible. This modification was stable for at least two weeks in buffer solutions with pH values between 4 and 8, and at temperatures up to 40 °C. Only at elevated temperatures of 60 °C and 80 °C under hydrolytic conditions it was observed that the stability of the same monolayer on PAO decreased gradually. As a proof-of-principle for the biofunctionalization and bacterial capture on this PAO phosphonate monolayer, an alkyne-terminated monolayer was biofunctionalized via a CuAAC click reaction with an azido-mannoside and the binding and growth of Lactobacillus plantarum was successfully demonstrated. In Chapter 4 various approaches to install reactive groups onto the phosphonate-modified PAO surface were developed, creating a (bio)functionalization “tool-box”. PAO surfaces presenting different terminal reactive groups were prepared, such as azide, alkyne, alkene, thiol, isothiocyanate, and N-hydroxysuccinimide (NHS), starting from a single, straightforward and stable initial modification with a bromo-terminated phosphonic acid. These reactive surfaces were then used to immobilize (bio)molecules, including carbohydrates and proteins. Fluorescently labeled bovine serum albumin (BSA) was covalently immobilized on the PAO surface as a proof-of-principle, and it was shown that a range of bacteria could still grow on the BSA-functionalized PAO surface. With a PAO (bio)functionalization tool-box in hand, the successful proof-of-principle mannoside-dependent binding and growth of L. plantarum on PAO (Chapter 3) was further investigated and expanded upon (Chapter 5). The parameters involved in the preparation of these surfaces and in the binding with L. plantarum were investigated in more detail in Chapter 5, such as the nature of the spacer connected to the mannoside derivative and the presence of soluble carbohydrates and bovine serum albumin (BSA) in the medium. The surfaces with the azido-mannoside with the long hydrophobic spacer showed the best binding of L. plantarum when compared to a long PEG-based hydrophilic spacer and a short hydrophobic one. The presence of a soluble a-glucoside did not prevent the binding of the bacteria to the mannose-presenting PAO, and similar results were obtained when BSA was present. Additionally, a mutant strain of L. plantarum that does not have the mannose-specific adhesion was not able to bind to the mannose-presenting PAO. When taken together, this proves that the mannoside–adhesin interaction is the main mechanism of binding the bacteria to the mannose-biofunctionalized PAO in this system. In Chapter 6, the NHS-terminated PAO developed in Chapter 4 was used for the immobilization of antibodies against Escherichia coli. After an extensive optimization of the modification chemistry of the surfaces and the incubation conditions, commercially available anti-E. coli antibodies were immobilized on the PAO surface. Binding and washing experiments indeed demonstrated increased binding of E. coli on the antibody-presenting PAO surfaces, providing avenues for testing other bacteria such as Lactobacillus rhamnosus GG widely used in probiotic formulations worldwide. In Chapter 7, the most important achievements of this project are discussed, together with additional ideas and recommendations for further research. Most notably some preliminary results are presented on the immobilization of two antibodies against L. rhamnosus GG: anti-L. rhamnosus GG, against the whole bacterial cell, and anti-SpaC, against only the SpaC part of the pili present on the cell surface of L. rhamnosus GG. Anti-L. rhamnosus GG antibody showed promising but not yet optimal increased binding of L. rhamnosus GG. Finally, some reflections on PAO and its (bio)functionalization are provided in the context of a risk analysis and technology assessment.