Understanding Bacterial Adhesion
Tailor-made Glycosylated Surfaces Reveal Molecular Details of Adhesion Processes
- Fig.1: Structure of a-mannosides which are attached to variable aglycon moieties in nature and are recognized by type 1-fimbriated E. coli.
- Fig. 2: Making things better: (A) A classical ELISA is the most time and material consuming way to detect bacterial adhesion and requires four steps. (B) Biotin-labeling of the bacteria shortens the detection of bacterial adhesion and (C) using GFP-tagged bacteria allows direct detection of adhered bacteria by fluorescence readout without any further additives. Abbreviations: ABTS: 2,2 a-azino-bis(3-ethylbenzothiazoline-6-sulfonic acid); ELISA: enzyme-linked immunosorbent assay; GFP: green fluorescent protein; HRP: horseradish peroxidase; OD: optical density.
- Fig. 3: Tailor-made carbohydrate surfaces are composed of three variable building blocks. (A) Mannoside ligands (blue), the aglycon moiety and the linkage chemistry are kept constant, (B) aglycon moieties of the immobilized mannosides are varied, (C) carbohydrate glycons, not aglycons are varied, and (D) complexity and nature of saccharides (multiplexing) are varied within the respective glycoarray.
- Mirja Hartmann, MSc, PhD student of the Lindhorst Group, Christiana Albertina University of Kiel
- Prof. Thisbe K. Lindhorst, Group Leader and Director, Christiana Albertina University of Kiel
All cells are covered by a glycocalyx, a ‘sweet coating', that serves different functions and is made up by complex glycoconjugates of variable architecture. In eukaryotic cells it plays a key role in cell communication and cell adhesion and expands to 100 µm width or even more. The glycocalyx and its molecular composition vary depending on the cell type and its developmental and health status. Thus, cancer cells often exhibit an especially thin or even disrupted glycocalyx. The sugar coating also serves as attachment site for microbial adhesion such as in bacterial colonization.
It depends on the tissue type on the one hand and the type of bacteria on the other whether or not bacterial adhesion causes a problem for the host. It can be a synergistic advantage like Escherichia coli colonization of the bowel or it might lead to inflammation, apoptosis, or even problems such as peptic ulcer . Thus, investigation of the mechanisms of bacterial adhesion is an important research field to promote our understanding of its consequences in health and disease.
Bacterial Adhesion to Host Cell Surfaces
Bacterial colonization is often accompanied by infectious diseases, constituting a major global health problem . In addition, biofilm formation on medical devices and implants frequently cause complications . Therefore, it is important to study the prerequisites for colonization of surfaces by bacteria and to investigate the details of the initial adhesion process. To facilitate adhesion, most bacteria express specialized adhesive organelles, called fimbriae (or pili). Fimbriae are hair-like protein structures on the bacterial cell surface. They possess protein subunits (lectin domains), that display carbohydrate-binding properties . The investigation of the molecular and supramolecular details of lectin-carbohydrate interactions is important to understand and treat bacterial colonization and as a consequence to develop means against infection and attachment to synthetic surfaces.
Addressing the Glycome
The molecular diversity of the carbohydrates that constitute the glycocalyx is extremely difficult to handle .
Thus, studying the "glycome" and its function in cell biology is a challenge of current research in addition to the investigation of the genome or the proteome, respectively. A feasible approach to tackle the problem of complexity is to narrow it down to simplified but highly specified systems. We have followed a two-step methodology: First, reduction of the supramolecular complexity of the glycocalyx to distinct saccharide moieties, such as certain monosaccharides. This approach allows to study the effect of detailed structural variations of the glycoside aglycon and glycon, respectively. Secondly, approximation of the surface scenario of the glycosylated cell by assembling the distinct glycoside constituents under investigation in the form of a glycoarray. This setup allows to study cellular adhesion to a glycosylated surface rather than just looking at receptor-ligand interactions in solution.
Assays to Test Bacterial Adhesion to Glycosylated Surfaces
In our own studies we have concentrated on mannose-specific adhesion of E. coli. This is mediated by the so-called type 1 fimbriae, which provide uropathogenic E. coli (UPEC) with the ability to attach to certain niches in the urinary tract . Type 1-fimbriated E. coli recognize a-mannosides with variable aglycon moiety (fig. 1). To study mannose-specific bacterial adhesion, we have regularly used the bacterial strain HB101pPKL4 , that exposes only type 1 fimbriae.
To allow the study of the molecular mechanisms of cellular adhesion to tailored glycosylated surfaces a powerful assay is an important prerequisite. A classical method is an ELISA, which has often been used in studies on bacterial adhesion with synthetic a-mannoside ligands. In this adhesion-inhibition assay mannose-specific bacteria, together with the serially diluted mannoside were incubated in mannan-coated microtiter plates. The polysaccharide mannan and the synthetic mannosides in solution compete for binding to the type 1 fimbrial lectins. After washing, adhered bacteria are visualized by addition of a monoclonal antibody against a protein on the type 1 fimbrial rod. Treatment with a horseradish peroxidase-conjugated secondary antibody and subsequent staining with ABTS facilitates the optical density readout (fig. 2 A). Higher OD-values thus parallel with better bacterial adhesion to the mannan surface and a low inhibitory potency of the added inhibitor.
The bottleneck of any ELISA is the primary monoclonal antibody, which has to be produced exclusively for one specific application. A second disadvantage is that the binding event is detected indirectly, as each single step is an individual error source. We have therefore set out to develop new methods for investigation of mannose-specific adhesion of E. coli, which are faster and more direct.
The next generation of bacterial adhesion assays was initiated by the idea to equip the bacteria directly with a detectable moiety. Therefore, prior to use, bacteria were biotinylated. For assaying bacterial adhesion, these biotin-labeled bacteria are again allowed to bind either the inhibitor or the mannosides on the surface. In a following incubation step, a streptavidin-HRP-conjugate can practically be anchored to the bacteria through the very strong biotin-streptavidin interaction. Subsequent staining with ABTS allows readout of the OD, analogous to the ELISA (fig. 2 B).
To shorten the assay time and boost the test's robustness we developed an assay using the fluorescing bacterial strain PKL1162. It contains the gene for GFP, the Nobel Prize-decorated green fluorescent protein  inserted on the chromosome. When these GFP-tagged bacteria were applied with mannan-coated microtiter plates, competitive binding between the well surface and a specific inhibitor in solution was tested as before. However, after incubation and washing, readout of bacterial adhesion could be performed directly with a fluorescence intensity reader at 485 nm (fig. 2 C). Moreover, the GFP-based assay is cheaper, when compared to other assays and highly reproducible. Its kit-like character makes the GFP-based assay a valuable tool for quick screening of potential new inhibitors of type 1 fimbriae-mediated bacterial adhesion and for testing adhesion to glycoarrays . In addition, GFP and its corresponding cDNA has been altered many times to give fluorescent proteins of different spectral characteristics and improved properties, all representing potential candidates for our assay.
Tailoring Carbohydrate Surfaces to Mimic Cellular Adhesion to the Glycocalyx
Instead of using mannan-coating, pre-functionalized 96-well microtiter plates can be modified according to the characteristics of a natural glycocalyx. In order to cope with the complexity of a glycosylated cell surface environment, specific structural elements of a glycoarray (fig. 3A) are varied accordant rational planning. For example, the aglycon moieties of the exposed glycosides can be modified (fig. 3B), altering parameters such as scaffolding, flexibility of the linker or additional affinity effects. However, the complexity of the glycoarray can be expanded by employing a mixture of carbohydrate ligands in a ‘multiplexing' approach (fig. 3C), or increase the saccharide complexity by going from mono- to oligosaccharide ligands (fig. 3D).
Reliable, reproducible, and accurate assays either serve as a tool to identify compounds with superior antiadhesive properties against one special bacterial strain or they can be utilized to screen the adhesion characteristics of different bacteria on tailor-made carbohydrate surfaces. This methodology has the potential, to investigate subtle structural changes within the carbohydrate ligands and maintain the supramolecular character of a glycosylated surface, which forms the basis of the adhesion processes operating at cell surfaces.
 Blaser M. J.: Sci. Am. 292, 38-45 (2005)
 Mulholland E. K., Adegbola R. A.: N. Engl. J. Med. 352, 75-77 (2005)
 Costerton J. W. et al.: Int. J. Artif. Organs., 11, 1062-1068 (2005)
 Imberty A. et al.: Curr. Opin. Struct. Biol., 15, 525-534 (2005)
 Werz D. B. et al.: ACS Chem. Biol. 2, 685-691 (2007)
 Ohlsen K. et al.: Top. Curr. Chem. 288, 109-120 (2009)
 Schembri M. A. et al.: Mol. Microbiol. 41, 1419-1430 (2001)
 Tsien R.: Annu. Rev. Biochem. 67, 509-544 (1998)
 Hartmann M. et al.: Chem. Commun. 46, 330-332 (2010)