Bacteria Pattern Spontaneously on Periodic Nanostructure Arrays you. Your

Surface-associated bacteria typically form self-organizing communities called biofilms. Spatial segregation is important for various bacterial processes associated with cellular and community development. Here, we demonstrate bacterial ordering and oriented attachment on the single-cell level induced by nanometer-scale periodic surface features. These surfaces cause spontaneous and distinct patterning phases, depending on their periodicity, which is observed for several strains, both gram positive and negative. This patterning is a general phenomenon that can control natural biofilm organization.

2 polymeric organic matrix. The formation of these composite structures is of concern to industry because biofilms can grow thick enough to obstruct pipelines in a variety of environments such as oil refineries, paper mills, and heat exchangers. In medical settings, biofilms cause extensive damage by triggering the human immune response, releasing harmful endotoxins and exotoxins, and clogging indwelling catheters. Hospital-acquired, or nosocomial, infections affect roughly 10% of patients in the United States -about 2 million annually, causing nearly 100,000 deaths 1 . Specifically, Pseudomonas aeruginosa is a human opportunistic pathogen and one of the most common nosocomial infections in the lining of catheters and the lungs of cystic fibrosis patients 2,3 . These infections are difficult to treat because the biofilm protects its constituent cells from antibiotic attack. Developing biomedical materials that are resistant to biofilm formation would significantly reduce the rate of nosocomial infections and the costs associated with treating them. Many negative effects of bacterial colonization stem from the formation of biofilms as protective structures and the associated cooperative behavior of bacterial cells mediated by quorum sensing and other virulence factors. As such, a comprehensive understanding of the interactions of bacterial cells at interfaces may lead to more effective treatments or surfaces that resist biofilm growth.
Persistently bacteria-resistant materials are difficult to achieve by surface chemistry alone. Even if bacteria are unable to attach directly to a substrate, non-specific adsorption of proteins or secreted surfactants to the surface eventually masks the underlying chemical functionality 4-6 . On the other hand, the effects of topographical features on bacterial adhesion and subsequent biofilm formation are poorly understood. Surface structures may provide a more persistent form of interaction between bacteria and surfaces. Nature provides some clues to preventing microbial colonization of surfaces. For example, ship hulls constantly amass layers of algae and crustaceans, whereas materials with topographical features mimicking the skin of sharks, for instance, have exhibited a remarkable resistance to marine biofouling 7 .
Recent studies have demonstrated that the behavior of mammalian cells can be manipulated using 3 only spatial and mechanical cues [8][9][10] . Various cellular processes, from apoptosis to proliferation and differentiation, are dependent on the spatial confinement of cells 11 , and even stem cell fate can be determined by the stiffness of the underlying growth substrate 12 . Bacteria also respond to mechanical cues from the environment. Indeed, surface attachment is an integral step in biofilm formation and precipitates chemical signaling pathways within and between bacterial cells 13 . Substrate stiffness, for example, has been suggested to affect the density of surface colonization 14 . The role that surface structures play in modifying bacterial attachment and subsequent behavior, however, is unclear 15 .
Biofilms contain a diversity of microbial phenotypes and form spatial patterns through cooperative organization at the macroscopic 16 and microscopic level. The formation of biofilms at interfaces (liquidsolid, liquid-air) is directed by gradients of nutrients, oxygen, and signaling molecules. In response to surrounding environmental factors, biofilms develop anisotropically and differentiated phenotypesdistinct from those of the planktonic state -segregate accordingly [17][18][19][20][21] . Topographical features can influence the arrangement and the resulting behavior of cells on surfaces and may affect biofilm development. Cellular interactions generally comprise diffusive compounds, such as quorum sensing molecules, from surrounding cells or environmental stresses, but can also include spatial signals. Some bacteria rely on physical interactions between neighboring cells for communication 22 , and several critical cellular processes, including division and external signal processing such as chemotaxis sensors, localize at the polar ends of the bacterium 23 . Therefore, disrupting the natural packing arrangement of cells within biofilms may influence some of the cooperative functions of these microbial populations.
Spatial confinement has been used to modify the configuration of surface-associated cells to a limited extent 15,24,25 , for example, and can alter the threshold for biochemical reactions 26,27 . Here we report on periodic arrays of high-aspect-ratio nanostructures, which direct the large-scale spontaneous patterning behavior of bacteria. The configuration of cells on these surfaces is sensitive to the spacing between neighboring features and the bacterial patterns register precisely with the symmetry of the underlying array. Moreover, the effect is general and is observed in biofilm mutants and both gram-positive and gram-negative species.
To study the effects of substrate topography on bacterial ordering and biofilm development, nanostructured substrates were fabricated with dimensions on the order of bacterial cells. Arrays of high aspect-ratio nanometer-scale polymer posts were generated using a fast replication molding technique described previously 28 . Using this method, many identical substrates with varying dimensional parameters, such as nanopost diameter, height, pitch, and array symmetry, were made to conduct systematic investigations of bacterial growth on structured surfaces. These substrates were sputtercoated with thin layers of PtPd or AuPd to reduce the autofluorescence background and to provide a compatible surface for thiol chemical functionalization. As a result, the surface chemistry, substrate stiffness, post dimensions and symmetry could be tuned to explore a wide range of interaction parameters.
Pseudomonas aeruginosa (strain PA14), a rod-like gram-negative bacterium, was grown on submerged polymer replicas with a gradient of post pitch, from 4 down to 0.9 µm. The posts had a diameter of about 300 nm, were 2 µm tall, and arranged in an array with square symmetry. As opposed to the random packing and three-dimensional growth of biofilms on flat substrates, bacteria grown on these post substrates spontaneously assemble into patterns dictated by the underlying array symmetry ( Figure 1). The fluorescence image in Figure 1a shows the interface between a flat region (upper) and one of patterned posts (lower) on the same substrate, upon which PA14 was cultured for 22 hr. in a rocking LB culture (see supporting information). The dissimilar ordering at the microcolony stage of biofilm formation is apparent, and the abrupt change at the interface suggests a localized response to topographical features rather than an induced cooperative behavior. Through-focusing fluorescence the cells could optimize their surface contact by adopting a diagonal orientation along the [11] directions of the square array. Indeed, this state is observed, where the cells are still perpendicular to each other but diagonal within the post array, as seen in the FFT by the increase in intensity of the [11] positional ordering peaks and slight extension of the central spot towards these peaks (Figure 3b, c). Bacteria adhere to surfaces by specific mechanisms, which can vary between species 13 . P. aeruginosa uses cellular appendages such as pili and flagella to find, attach to, and move about on surfaces. To investigate the relevant mechanisms by which PA14 assembled within the post arrays, mutants lacking 7 the necessary genes to synthesize either pili or flagella were grown on identical post substrates to the wild type strain. The assembly of these strains at different post pitches is shown in Figure 4a and b, respectively. The overall assembly is unchanged and characteristically similar to the wild type assembly. The lack of effect of the appendage knockouts on the assembly phenomenon indicates that a different mechanism is involved in bacterial affinity for the surface than for normal biofilm development. In addition, the flagellum knockouts tended to adhere and proliferate at a slower rate than either the wild type or pili mutants, which is consistent with the known adsorption mechanism of P. aeruginosa 13 . Control experiments on the same substrates also confirmed that the patterning is not an effect of shear flow or sedimentation. Identical spontaneous assembly was observed on rocking and static cultures, and on inverted submerged substrates with the same pitch gradient as those shown in the figures. These results suggest that a biological, rather than physical, mechanism is responsible for the patterning behavior of the bacteria.
Other bacterial strains also use different genetic pathways and cellular appendages to adhere to surfaces. Wild type strains of both Bacillus subtilis (strain 3610) and Escherichia coli (strain W3110) were grown on the substrates with a gradient of post pitches, and similar cellular patterning was observed (Figure 4c and d). These disparate species assemble into patterns dictated by the post array and exhibit the similar ordering phases to PA14. One significant difference between the assemblies of these species on the same post pitch gradient substrates is the inter-post spacing at which the patterns form.
Specifically, the spacing between neighboring posts at which bacteria began to order on the substrate correlates to the size of the cells. B. subtilis and E. coli cells ordered in subsequently larger gaps than P. aeruginosa, suggesting that the assembly phenomenon is related to interactions with the cell surfaces or biofilm components closely associated with the cell wall. Moreover, the patterned assembly is a general     [11] peaks and diagonal smearing of the center spot (note that neither the (11) peaks nor the diagonal smearing is observed in other packing phases on the substrate, see