Dissecting microbiological systems using materials science

Dissecting microbiological systemsusing materials scienceAbishek Muralimohan, Ye-Jin Eun, Basudeb Bhattacharyya and Douglas B. Weibel

Department of Biochemistry, University of Wisconsin-Madison, 433 Babcock Drive, Madison, WI 53706, USA

Review

Glossary

Chemical etching: a process for patterning surfaces in which layers are

removed from the surface of a substrate via solvents (e.g. etchants).

Elastomer: a compliant rubber-like polymer.

Electron-beam lithography: a process in which a focused beam of electrons is

scanned across a surface and ‘writes’ patterns with dimensions below the

diffraction limit of visible light.

Hydrogels: this class of polymers is frequently cross-linked and is characterized

by a large amount of bound water.

Laminar flow: a regime of fluid flow characterized by motion that is dominated

by viscosity. In the laminar regime, adjacent layers of fluid slip past each other

without mixing. The exchange of material between laminar streams of fluid

occurs by molecular diffusion.

Microchannels: channels with micron-scale dimensions are typically embossed

or etched in the surface of a material.

Microfluidics: networks of channels with micron-scale dimensions that are

used to transport and manipulate small volumes of fluids.

Photolithography: a process used to transfer a pattern from a mask into a thin

film of photosensitive polymer coated on the surface of a substrate using UV

light.

Quantum dots: fluorescent semiconductor nanocrystals with optical properties

that can be tuned during their synthesis.

Reynolds number: the Reynolds number (Re) characterizes the different

regimes of fluid flow. Re is a dimensionless parameter that depends on the

properties of a fluid (viscosity and density), its rate of flow and the geometry in

which the fluid is flowing. Laminar flow occurs when Re < 2100. A typical range

of flow in microchannels produces values of Re ranging from 0.1 to 100.

Soft lithography: this set of techniques makes it possible to create

microstructures by printing, molding and embossing using a patterned,

elastomeric stamp or mold and/or a polymeric substrate.

Surface plasmon resonance: surface electromagnetic waves propagate in a

direction parallel to the surface of metal NPs. The wavelength of these

oscillations is sensitive to any changes at this boundary, such as the adsorption

of molecules on the metal surface.

Young’s modulus: a measure of stiffness that relates the deformation of a

material as a force is applied. A large value of Young’s modulus indicates a stiff

Materials science offers microbiologists a wide varietyof organic and inorganic materials with chemical andphysical properties that can be precisely controlled.These materials present new capabilities for isolating,manipulating and studying bacteria and other microor-ganisms and are poised to transform microbiology.This review summarizes three classes of materials thatspan a range of length scales (nano, micro and meso)and describes a variety of fundamental questions inmicrobiology that can be studied by leveraging theirproperties.

New materials for a new microbiologyThis review focuses on the application of materials scienceto the study ofmicrobiological systems.Materials science isa field that is positioned at the intersection of the physicalsciences and engineering; in it, organic, inorganic andpolymeric materials are routinely synthesized and fabri-cated with an extraordinary spectrum of dimensions andproperties. Biocompatible structures that match theintrinsic length scale of a wide variety of structures foundin microbiology (e.g. proteins, chromosomes, organelles,individual cells and multicellular structures) can be tai-lored to study a range of processes that might be difficult toexplore using the techniques of classical physiology-drivenmicrobiology, genetics, genomics and biochemistry(Figure 1). In contrast to microbiology and materialsscience, which are just beginning to converge, eukaryoticcell biology has been implementing new materials for atleast a decade. Eukaryotic cell biology and materialsscience have become intertwined to the point where cellbiologists routinely synthesize materials for their studiesandmaterials scientists are experienced in cell culture andgenetics. What does materials science and engineeringbring to eukaryotic cell biology that it cannot also bringto microbiology? We believe that the implementation ofnew materials will benefit the study of microbes just as ithas transformed eukaryotic cell biology.

Although the most comprehensive reviews at the inter-face of these fields are focused on applications in mamma-lian cell biology [1–3], a recent review summarizestechniques for creating structures relevant to the studyof microbes [4]. Because many microbiology laboratoriescurrently lack the equipment and expertise necessary todesign and fabricate materials, we believe a more usefulintroduction to this area for microbiologists will be asummary of several interesting classes of biocompatiblestructures and a discussion of where they might find

Corresponding author: Weibel, D.B. ([email protected]).

100 0966-842X/$ – see front matter � 2008 Elsevier

applications. This review draws connections betweenrecent advances in materials science and unansweredquestions in microbiology by focusing on three aims thatwill stimulate research at the interface of these two fields:(i) a discussion of several ‘classes’ of structures organizedaccording to physical length scale (nano to meso) that weanticipate will be particularly useful for microbiologicalstudies, (ii) a focus on specific areas of microbiology inwhich they will have an impact and (iii) a review ofexamples from the literature that demonstrates orindicates their potential application in these areas. Thefocal point of this review is on forging new connectionsbetween the characteristics and capabilities of materialsand fundamental, unanswered microbiological questions.Although we briefly discuss how specific examples ofmaterials are made and where they can be acquired (seeTable 1 in the Supplementary Data online), the emphasisis on their functions and characteristics. For a generaloverview of techniques for the synthesis and fabrication

material.

Ltd. All rights reserved. doi:10.1016/j.tim.2008.11.007 Available online 23 February 2009

mailto:[email protected]

http://dx.doi.org/10.1016/j.tim.2008.11.007

Figure 1. Length scales in materials science and biology. Nano-, micro- and

mesostructured materials are represented by a quantum dot [69], a microfluidic

device and patterns of star-shaped bacterial colonies of E. coli (left column) [53].

Common biological structures are shown for size comparison (right column).

Images of protein structures were generated from PDB entries 1EMA and 1CDR

using Pymol (http://www.pymol.org/). The structure of the quantum dot was

reproduced, with permission, from Ref. [69].

Review Trends in Microbiology Vol.17 No.3

of biocompatible materials, we direct the reader to severalreviews [3–6].

Nanostructures: nanoparticles‘Nano’ is typically used to define structures with a lengthscale of 1–100 nm. To put the dimensions of nanoscaleobjects in perspective: a structure that is 1 nm wide con-sists of tens of atoms. Nanoscale materials are not justanother step in miniaturization – they are a unique class ofmaterials with characteristics that are not found in theirmacroscopic counterparts, including a large surface-to-volume ratio in which themajority of the atoms are presenton the surface of a particle or structure; tunable opticalproperties (e.g. excitation and emission wavelengths andplasmon resonance); and extraordinary mechanical, ther-mal and conductive properties [7]. The length scale andproperties of nanoscale materials make them particularly

applicable to exploring phenomena occurring within bac-teria at the subcellular level. Below, we introduce this classof materials and elaborate on its application to the study ofintracellular structures in bacterial cells.

A wide variety of nanostructured materials (e.g. sur-faces, thin films, wires, spheres, rods, prisms and otherparticles) have applications in biology. Nanoparticles(NPs) are arguably the class of nanostructured materialsthat have been most frequently applied to the study ofbiological systems [7]. NPs can be synthesized in mono-disperse form – that is, the particles have a small coeffi-cient of variance in their mean diameter – which makes itpossible to control and predict their properties. Thediameter of NPs (>1 nm) is equivalent to or smaller thanthe cross-section (hydrodynamic radius) of many proteinsand other biological structures. NPs have been synthesizedin a variety of organic and inorganic materials (Table 1).Finally, several classes of NPs are photostable and haveoptical parameters that can be tuned by controlling theirdimensions during their synthesis. We believe that semi-conductor nanocrystals (quantum dots, or ‘QDs’; see Glos-sary) and gold NPs are two classes of nanostructures thatwill illuminate the structure of microbes. A summary of themost salient features of these materials is discussed below.

Quantum dotsQDs are photostable, bright and resistant to metabolicdegradation [8]. The emission spectrum of QDs is narrowand can be controlled by varying the diameter of particles(Figure 2b). QDs are particularly useful as donors forForster resonance energy transfer (FRET) [9]. In vivostudies of mammalian cells using QDs as tracers andFRET donors have taken advantage of their photostabilityand fluorescence intensity. The ability to coat the surface ofparticles with ligands that target their localization withinthe cell (Figure 2a) has made them particularly useful forstudying the subcellular structure ofmammalian cells [10].QDs are typically microinjected into mammalian cells, andtheir temporal and spatial location is measured usingepifluorescence microscopy over a range of time scales[7,8]. NPs can also be introduced in mammalian cells byattaching transport sequences to their surface [9].

In contrast to their application to the study of mamma-lian cell biology, QDs are still emerging as probes forstudying microbes [11]. One of the major hurdles of usingQDs to study bacteria is the difficulty of transporting theminto the cytoplasm. The cell dimensions of most strains ofbacteria make microinjection unfeasible. The peptidogly-can (PG) layer of the cell wall limits the passive transportof QDs into cells; this material can range from 10 to 80 nmthick in Gram-negative and Gram-positive bacteria [12].The average pore size of the mesh-like PG is believed to beconstant in bacteria and is �2 nm in cells of Bacillussubtilis and Escherichia coli [13]. The porosity of the PGis smaller than the diameter of most commercially avail-able QDs (�3–6 nm) and makes it unlikely that QDs willpassively diffuse through the PG into the cytoplasm.

Although the PG presents a physical barrier for theefficient uptake of QDs by bacteria, several groups havestarted exploring the transport of QDs into bacterial cells.Li and coworkers [14] transformed Ca2+-induced compe-

101

http://www.pymol.org/

Table 1. Properties of nanoparticles

Material Composition Preparation Important properties Biological applications Refs

Dendrimers Polymer Chemical synthesis � Hyper-branched polymer network � Encapsulation, delivery of

biomolecules into cells

[60]

� Porous

Nanocages DNA Self-assembly � Biological starting material � Encapsulation, delivery of

molecules or particles into cells

[61]

� Molecular recognition by hybridization

to complementary sequences

Metal

nanoparticles

Gold, silver Nucleation � Surface plasmon resonance

� Can be functionalized

� Antimicrobial agents [6,19,20]


� TEM imaging

� Metabolic sensors

� Contrast agents

Quantum dots CdSe, CdTe,

PbSe, InAs

Nucleation,

crystallization

� Fluorescent � Optical tracer for transformations [5,62]

� Photostable � In vivo FRET donor

� Narrow emission spectra � Cellular label for long-term imaging


Abbreviations: CdSe, cadmium selenide; CdTe, cadmium telluride; PbSe, lead selenide; InAs, indium arsenide.


tent cells ofE. coliwith 3–4 nm diameter QDs as tracers forstudying chemical competence. An interesting approachfor transforming cells with QDsmight be to take advantageof the native transport machinery in the bacterial cell wall.Kloepfer et al. [15] have found that adenine- and adenosinemonophosphate (AMP)-coated QDs are internalized byadenine auxotrophs of B. subtilis and E. coli(Figure 2c,d). They suggest that the uptake of QDs occursvia purine-dependent transport mechanisms. Hirscheyet al. [16] described the transport of QDs coated withorganic di- and tricarboxylic acids (e.g. citrate, isocitrate,succinate and malate) into the cytoplasm of cells of E. coli.It is unclear how the QDs cross the PG en route to thecytoplasm, particularly if the PG is a continuous polymernetwork with 2-nm wide pores. The systematic study ofQDs decorated with ligands that target the transportsystems of bacterial cells might introduce new materialsand mechanisms for studying transport processes in bac-teria and techniques for improving their transformation.

Gold nanoparticlesGold NPs are frequently used as contrast agents for ima-ging bacterial cells using transmission electronmicroscopy(TEM), in which the particles bind to proteins with exposedcysteine residues [17]. The application of gold NPs to thestudy of bacteria is not limited to TEM and might also beuseful for imaging the subcellular localization of proteinsand nucleic acid within cells using optical microscopy. GoldNPs exhibit a phenomenon referred to as localized surfaceplasmon resonance (LSPR), in which gold NPs scattervisible light at a specific wavelength [18]. When the surfacechemistry of a gold NP changes, the wavelength of lightscattered by the particle is red-shifted and provides a color-based biological sensor. The application of a related tech-nique to study the localization of proteins in live cellsmightbe possible using gold NPs coated with small moleculesthat are substrates for enzymes. The basis for thisapproach is a recent study by Liu et al. [19], in whichthe authors used an LSPR-based technique to monitor b-lactamase activity. In this assay, gold NPs were coatedwith cephalosporin. Cleavage of cephalosporin by b-lacta-mase caused the NPs to cluster, which produced a measur-able optical signal. A similar concept has been applied tothe development of an optical sensor of bacterial growththat depends on the crosslinking of dextran-coated goldNPs in the presence of concanavalin A and carbohydrates

102

in the growth medium. During periods of growth, theconcentration of free carbohydrates in solution decreasesand the size of gold NP clusters is reduced [20].

Nanoparticles: future directionsApplications of nanostructured materials in microbiologydo not yet take advantage of the full range of character-istics that have made them an important tool in studyingmammalian cells. NPs have certain limitations, such astheir tendency to ‘rust’ and aggregate, which are still beingsorted out. Despite these limitations, NPs will make itpossible to study a range of interesting structural issues inbacteria, including the molecular mechanisms of transportacross the cell wall, the physical structure of the PG andthe subcellular organization of proteins and nucleic acid inbacterial cells.

Studying the molecular mechanisms of transport across

the cell wall

Surface-modified QDs with controlled dimensions will bean intrinsically useful tool for studying transport acrossthe bacterial cell membrane and the cell wall. Thisapproach might answer fundamental questions aboutthe mechanisms of transport of biomolecules (includingDNA and othermolecules with a hydrodynamic radius thatis larger than the pore size of the PG) in and out of cells.

Studying the physical structure of the PG

Demchick andKoch [13] used fluorescently labeled dextranmolecules to determine the pore size and diffusion ofbiomolecules across the PG in E. coli and B. subtilis.QDs are ‘rigid’ and make it possible to determine theporosity of the PG in intact cells at a level of detail thatis not possible using flexible dextran molecules and iso-lated fragments of PG.

Studying the subcellular organization of proteins and

nucleic acid in bacterial cells

NPs might be useful in exploring the spatial and temporalorganization of bacterial cells. For example, the diffusion offluorescent NPs within cells might be useful for studying‘crowding’ in the cytoplasm [21].

Microstructures: elastomeric polymersMicrostructures have a size scale similar to that of indi-vidual or small groups of microorganisms and typically

Figure 2. Properties of quantum dots (QDs). (a) A cartoon depicting a QD [69]. The

inner CdSe core particle (diameter = 60 A) is encapsulated by a ZnS shell. The zinc

atoms are shown in pink, and the sulfur atoms are shown in teal. The QD surface is

functionalized with dihydrolipoic acid (red). A maltose-binding protein is shown to

the right of the QD for size comparison. (b) A plot of the fluorescence emission of

QDs versus the radius of the core particle [62]. (c,d) QDs labeled with adenine or

AMP are internalized by E. coli adenine auxotrophs (Eco aux Ad, right image); no

internalization is observed in wild-type E. coli (Eco Ad, left image) [15]. Figures

were reproduced, with permission, from Refs [15,62,69].


range from 100 nm to 100 microns. The overlap in theselength scales makes it possible to design and implementmicrostructured materials to isolate, manipulate andstudy individual bacteria or small multicellular structures.Microstructures can be used to control the interface be-

tween cells and their microenvironment – the region that issensed by a cell and is typically defined by molecularcontact, mass transport and diffusion – and, ultimately,the internal organization and physiology of the cell. In thenext section, we describe a class of microstructuredmaterials for controlling small volumes of fluids that pro-vide access to a variety of experimental conditions that arenot available using traditional techniques of microbiology.

Microfluidic systems have channels with micron-scaledimensions and are used to manipulate small volumes offluids (from fL to mL). Microchannels have a large surface-to-volume ratio that facilitates rapid mass and heat trans-fer. These systems offer predictable and reproducible con-trol over conditions for experiments with bacteria byminimizing the effects of environmental fluctuations(e.g. temperature and aeration) [22]. The concentrationof molecules in microchannels can also be controlled witha high degree of accuracy in space and time [23]. Micro-fluidic systems also have several practical characteristics:they consume small quantities of samples and reagents,they can be multiplexed to perform several assays simul-taneously and they are inexpensive. Most bacteria live inlow-Reynolds-number environments, and microfluidic sys-tems, in which fluids flow in the laminar regime, canreproduce such environments precisely.

Microchannels have been created in a variety ofmaterials, including glass, silicon, poly(methyl methacry-late) (PMMA) and other polymers (Table 2). The simplicityof embossing the elastomeric polymer, poly(dimethylsilox-ane) (PDMS), with channels and the unique properties ofthis polymer make it widely used for microfluidics [24].Below, we briefly discuss the areas ofmicrobiology inwhichmicrostructured materials might be particularly relevant.

Study of single bacterial cellsMicrostructured surfaces are useful for the isolation andstudy of single bacterial cells and might be particularlyrelevant to bacterial communities in environmentalsamples [25]. One of the challenges facing microbialecology is the difficulty of culturing a heterogeneous popu-lation of microorganisms extracted from their environ-ment, which makes it difficult to quantitativelydetermine their composition, genotype and function.Recent advances in PDMS microfluidic systems make itpossible to analyze the genotype and phenotype of a singlebacterium, which obviates the necessity for culturing cellsand bypasses the limitations of their isolation and growthin the laboratory. The Quake group have developed severaltechniques based on microfluidics for isolating single bac-teria from sparse environmental samples, including thehind gut of termites [26] and the human oral cavity [27].They have integrated multiplex PCR into these microflui-dic systems to amplify and sequence entire genomes fromsingle bacteria and have used these devices to genotypeand identify several new species of bacteria.

The greatest impact of microstructured materials inmicrobiology might be in the area of single-cell analysis[28,29]. Data from single-cell experiments are inherentlynoisy because of variations in the age of cells, the stochasticnature of metabolism and transcription, and fluctuationsin the microenvironment of cells. Microstructured

103

Table 2. Properties of common microstructured materials

Material Optical

properties

Young’s

modulus

Solvent compatibility Techniques for patterning Limitations Notes

Silicon Opaque;

reflective

<500 nm

129–186 GPa � Resistant to most

solvents

� Electron-beam lithography � Expensive � Scratch resistant

� Chemical etching � Serial patterning

� Photolithography � Brittle

Glass Transparent

>195 nm

(fused silica)

�100 GPa � Resistant to most

solvents

� Electron-beam lithography � Expensive � Scratch resistant

� Chemical etching � Serial patterning

PMMA Transparent

>350 nm

2–3 GPa � Acid sensitive � Hot embossing � Can be scratched � Used as a replacement

for glass� Degrades in many

organic solvents

� Electron-beam lithography

� Photolithography

PDMS Transparent

>280 nm

�2 MPa � Resistant to alcohols � Drop casting � Hydrophobic � Used in soft lithography

� Swells reversibly in

organic solvents

� Rapid prototyping � Surface chemistry is

difficult to modify

� Flexible

� Transparent to gas and

vapor


materials make it possible to average data collected fromthousands of isolated cells in parallel under identicalconditions. Single-cell analysis overcomes biases insampling that arise from ensemble averaging of data frombulk measurements [30], which makes it a powerful tool toilluminate minor phenotypic differences among clonalpopulations of cells. For example, Balaban and coworkers[31,32] used microfluidic channels to monitor single bac-terial cells over several life cycles after treatment withantibiotics and derived a model of bacterial persistence(Figure 3b).

Cell shapeThe molecular details of the coordination of cell growth,division, asymmetry and the cytoskeleton in bacteria arejust beginning to emerge [33–35]. Genetic and biochemicaltechniques have been the tools that have largely driven ourunderstanding of these processes. Physical constraints formanipulating bacterial cell shape are complementary tothese methods and provide a unique capability for thestudy of cellular physiology. Takeuchi et al. [36] embossedagarose with microchannels and used these structures toexert mechanical forces on growing bacterial filaments. Asthe cells grow, the deposition of the cell wall during fila-mentation is controlled and the filaments become perma-nently deformed in the shape of the channels (Figure 3a).Cabeen et al. recently used this technique to study themechanism of crescentin, a homolog of eukaryotic inter-mediate filaments that polymerizes into filaments in cellsof Caulobacter crescentus and influences cell curvature(Cabeen, M.T. et al., unpublished).

Bacterial motilityThe study of bacterial motility is another area of micro-biology in which microchannels are playing an importantpart. The dimensions and properties of channels can betailored to create new conditions or mimic the nativeenvironment of bacterial cells. DiLuzio et al. [37] studiedthe motility of cells of motile strains of E. coli in micro-channels in which the porosity of the channel surfaces wasvaried. They observed that bacteria preferably swim inhydrodynamic contact with the surface of hydrogels. Thisphenotype is particularly evident when cells are in chan-nels with a hydrogel ‘floor’ – the cells move along thechannel in contact with the right-hand wall because of

104

the clockwise rotation of the cell body (viewed from behindthe cell) during their translation through fluids. Hulmeet al. [38] have taken advantage of this phenotype todevelop a microfluidic system of ratchets that sorts motilestrains of bacteria based on cell length.

Microfluidics might also provide insight into bacterialchemotaxis. Traditional capillary-based chemotaxis assays[39] are incapable of producing gradients that aretemporally stable, complex and non-linear. Microfluidicsystems overcome these limitations and make it possibleto study concentration gradients of chemoattractants andchemorepellants at the nM level [40].

Laminar flow and fluid shearThe laminar flow regime of fluids and the effects of surfacetension and viscous forces that are present in microstruc-tured channels make it possible to study bacterial physi-ology from a new angle [22]. Parallel stream of fluidsflowing in the laminar regime do not mix convectively.This property enables users to pattern surfaces with chem-istry, complex topographic features and gradients thatmimic natural environments by usingmicrofluidic systems[41–43].

The shear created by the flow of fluids has been used toidentify the ‘catch-bond’ mechanism of pili in bacterialadhesion that is responsible for uropathogenic bacteriaadhering to tissues in high-shear environments [44].Microfluidics arguably offers the most precisely controlledplatform for studying the effects of shear on single cells[45]. The shear created by flowing fluids can also be used toprobe the mechanical properties of structural componentsof cells – e.g. the cytoskeleton and peptidoglycan – bymeasuring the mechanical deformation of cells in high-shear conditions.

Mesostructures: hydrogelsMesostructured materials typically have features withlength scales from 500 microns to several millimeters,which we refer to as ‘mesoscopic’. These structures providemechanisms of controlling populations of microbial cellsand confining them in geometries to study populationdynamics and collective behavior. Mesostructuredmaterials might provide a unique platform for culturingbacteria, in which the user can control the spatial organ-ization of colonies and manipulate cell–cell and cell–

Figure 3. Examples of micro- and mesostructured materials. (a) A micropatterned

agarose surface used to engineer bacterial cell shape. The image inset shows

helical, filamentous cells of E. coli that were released from the microchambers and

retain the imposed shape [36]. (b) Microchannels used for single-cell analysis of

bacterial persistence [31]. Cells form linear microcolonies in the microchannels

that were used to link persistence to heterogeneity in bacterial populations. (c) The

spatial self-organization of bacteria is observed when grown in confined

geometries [54]. Cells in region I were distributed almost exclusively

perpendicular to the long axis of the mesostructure, whereas cells in region II

were randomly distributed. (d) Bacteria-shaped structures of photoluminescent

Vibrio fischeri patterned on the surface of agar to form the word ‘ink’ [53]. Figures

(a), (b) and (c) were reproduced, with permission, from Refs [31,36,54].


environment interactions [46]. In this section, we introducemesostructured polymers with properties that provideseveral advantages over traditional materials used forbacterial cell culture.

Hydrogels are polymers that absorb and retain a largevolume fraction of bound water. Agar is an example of thisclass of materials that has been used in microbiology as asubstrate for bacterial growth. It is widely available, inex-pensive, and easy to prepare and use. Agar has manyuseful properties: it provides a ‘wet’ environment for bac-terial growth; it is relatively ‘transparent’ to the diffusionof nutrients, ions and metabolic waste; it is biocompatible;and it is not degraded by bacteria. Several characteristicsof agar limit its applicability to addressing certain micro-biological questions, including chemical and physical prop-erties that are typically not well defined. Agar offerslimited control over parameters such as surface chemistry,porosity, wetness, wettability and stiffness.

Agarose is chemically defined and provides better con-trol over some parameters relevant to bacterial culture.Poly(ethylene glycol) (PEG) and polyacrylamide (PAA) arehydrogels that might be excellent alternatives to agar andagarose for bacterial culture because their physical proper-ties can be tuned precisely during their synthesis [2](Table 3). The surface of these polymers can be topogra-phically patterned and functionalized to present differentmolecules [47], including chemistry that makes themresponsive to light, pH, temperature and moleculessecreted by cells. These modifications change the pore size,water content and diffusion of molecules through the poly-mer and might affect the growth rate of microbes on thesurfaces. Below, we describe applications of hydrogels inbacterial culture.

New techniques for microbial culturingCell culture is arguably one of the most fundamentaltechniques in microbiology and can benefit considerablyfrom the implementation of new materials. Most organ-isms in the biosphere have not been cultured by conven-tional microbiology to date, and many of these seem to berefractory to routine culture. Polymers might be useful forculturing and isolating these organismswhich, in turn, willprovide insights into the mechanisms of sensory trans-duction between the cell and its environment [48]. Mesos-tructured polymeric structures have recently been used tocontrol the length scale of interactions between differentmicroorganisms, making it possible to engineer syntrophicinteractions [49]. Kaehr and Shear [50] have developedprotein-based hydrogels that respond to chemical stimuliby undergoing changes in their hydration state, making itpossible to dynamically control the volume of the polymer.Structures fabricated in these hydrogels enable the user totrap, incubate and subsequently release cells of bacteria,which could provide the basis for a high-throughput, auto-mated cell-culture system.

Bacterial microarrays, like their DNA and proteincounterparts, have applications in high-throughput phe-notypic [51] and genotypic screens [52]. Reproducibilityand the ability to catalog (store) and replicate thousands ofparallel microcolonies while maintaining a form factor nolarger than a postage stamp make bacterial microarrays

105

Table 3. Properties of hydrogels

Material Synthesis Physical properties Surface

chemistry

Stimulus response Notes Refs

Agar Gels upon

cooling

Pore size: undefined; permits diffusion

of macromolecules

� Hydroxyl

groups

� None � Widely used as substrate for

bacterial growth

–

� Poorly defined parameters

Agarose Gels upon

cooling

Pore size: varied from 200–500 nm

(5%–1% gel)

� Hydroxyl

groups

� None � Surface structures are

created using soft lithography

[63,64]

Stiffness: 400 kPa (2.5% gel)

Alginate Crosslink with

Ca2+

Stiffness: 100 kPa (2% gel); can be

varied over a large range

� Guluronic

acid groups

� None � Used to encapsulate bacteria

for biotechnology applications

[65]

PEG Photo-

crosslinking

Pore size: 1–10 nm; stiffness: depends

on ratio of crosslinker. Typically mixed

with polyurethane to increase stiffness

� Ether

linkages

� Volume expands

upon water absorption

� Biocompatible [66]

� Inert

PAA Free-radical

polymerization

Pore size: 19 nm–142 nm; stiffness:

�10 GPa; pore size and stiffness

depends on % of Bis crosslinker

� Various � Responsive to

temperature (NIPAM)

and pH (chitosan-PAA)

� Acrylamide analogs provide

a wide range of properties for

use in microbial cell culture

[67,68]

Protein-

based

Photo-

crosslinking

Unknown � Amino acid

residues

� Responsive to

temperature, pH, ionic

strength, molecules

� Reversible unfolding of

proteins causes a change in

hydration properties

[50]

Abbreviation: NIPAM, poly-N-isopropylacrylamide.


an excellent alternative to conventional techniques ofstreaking and plating colonies. Weibel et al. [53] havedeveloped a technique for directly printing patterns ofbacterial colonies on surfaces using hydrogel stamps(Figure 3d). Eun and Weibel recently developed a tech-nique using polymer stencils to create and study arrays ofbiofilms (Eun and Weibel, unpublished).

Population dynamics and evolutionMesostructured materials can also be used to addressquestions of population dynamics and microbial evolution.Cho et al. [54] recently used mesostructures to identify therole of chemotaxis, quorum sensing and spatial confine-ment in the formation of biofilms (Figure 3c). The systema-tic variation of parameters that affect biofilm growth [55](including surface chemistry, microenvironment andmechanical stresses due to flow and geometry) will makeit possible to study the interaction of cells, the establish-ment of spatial heterogeneity and the differentiation ofcells within biofilms [56]. Mathematical models that

Box 1. Questions for future research

Materials with structures spanning a range of length scales from nano

to meso have precisely controlled physical and chemical properties

and have the potential to impact several areas of microbiology. We

pose several questions below that we believe will stimulate further

research at the interface of microbiology and materials science; other

examples are discussed in the text.

What is the fate of individual bacteria in multicellular communities?

Fluorescence microscopy has been instrumental in advancing our

understanding of microorganisms. Commonly used fluorescent

probes are limited by their low photostability and occasional

cytotoxicity. Quantum dots (QDs) are fluorescent semiconductor

nanoparticles that can be designed to overcome these problems. QDs

can be used to label subpopulations of bacteria and can be tracked

over several generations [10].

What is the role of physical interactions between cells of bacteria in

multicellular structures?

Most bacteria are too small and intractable to be used with micro-

manipulation techniques commonly used with larger eukaryotic cells.

Microfluidics offers a platform for precisely controlling the position of

single bacterial cells and engineering reproducible interactions

between cells, thereby making it possible to study cell–cell interactions.

106

describe complex multicellular behavior (such as coopera-tion [57], the one-third law of evolution [58] and the pred-ator–preymodel ofE. coli behavior [59]) can now be verifiedexperimentally using this class of materials.

Concluding remarks and future perspectivesMany of the materials that are currently used to studymicrobes have evolved little since their introduction, evenafter a century or more of routine use. Have thesematerials stuck around because they are convenient, ordo they offer unique capabilities or an ideal solution? Arelevant example is the agar plate. Agar surfaces are stillthe most frequently used platform for culturing and iso-lating the vast majority of bacterial strains grown in thelaboratory. Slants, stabs, plates and other forms of solidmedia that incorporate agar are used for the routinepropagation and storage of bacterial colonies. However,based on the diverse environments that microbes inhabit,it seems unlikely that the surface of an agar plate is afaithful reproduction of their microhabitat [48]. Because

How does asymmetry arise in bacteria?

The manipulation of single microbial cells might be also an

important capability for studying the intracellular organization of

bacteria. Rather than a simplistic view of the cytosol as a collection

of freely diffusing molecules, it is now recognized that there is a

remarkable amount of subcellular organization and asymmetry in

bacterial cells. Microfluidic techniques such as PARTCELL [70]

make it possible to ‘paint’ domains of a cell with reagents by

taking advantage of laminar fluid flow. This method, and others,

might be particularly useful for studying the origins of asymmetry

and its role in bacterial physiology.

How do interactions arise between different bacterial species?

Bacteria in their natural habitats exhibit a rich variety of inter-

species interactions including symbiosis, competition, parasitism

and commensalism. The systematic study of these interactions is

complicated by the challenge of accurately mimicking the condi-

tions for their growth. The physical and chemical properties of

certain biocompatible hydrogels can be tuned to present a spatially

heterogeneous substrate that resembles natural substrates more

closely than the surface of an agar plate does.


there are hundreds of other polymer gels with chemicaland physical properties that can be precisely tuned – incontrast to agar – and used for this application, it seemslikely that the popularity of agar might be a matter ofconvenience.

Newmaterials and techniques for studyingmicrobeswillmake it possible to control their microenvironment andstudy cell physiology and behavior at a new level of detailby interfacing these capabilities with the techniques ofgenetics, genomics and biochemistry. Materials scientistsknow how to control the physical and chemical properties ofmaterials that will have key roles in this area; they mightnot fully appreciate the capabilities that microbiologistsneed. Physical scientists and engineers that design andfabricate materials, and end users, who might be microbiol-ogists seeking new capabilities for studying microbes, havehistorically not worked closely together. Over the pastseveral years, however, the gap between these fields hasnarrowed as materials scientists have ventured into micro-biology in search of new applications for materials andmicrobiologists have been introduced to materials withnew capabilities. Materials science is poised to have animportant impact on microbiology (Box 1). The merger ofthese fields will almost certainly drive the emergence ofexciting new questions and directions in biology, chemistry,and materials science and engineering.

AcknowledgementsThe United States Department of Agriculture (WISO1192), the NationalScience Foundation under Grant No. DMR-0520527, a 3M Non-tenuredFaculty Award and a Searle Scholar Award support research in ourlaboratory at the interface of materials science and microbiology.

Appendix A. Supplementary dataSupplementary data associated with this article can befound at doi:10.1016/j.tim.2008.11.007.

References1 Jiang, X. and Whitesides, G.M. (2003) Engineering microtools in

polymers to study cell biology. Eng. Life Sci. 3, 475–4802 Mrksich, M. (2002) What can surface chemistry do for cell biology?

Curr. Opin. Chem. Biol. 6, 794–7973 Sniadecki, N.J. et al. (2006) Nanotechnology for cell–substrate

interactions. Ann. Biomed. Eng. 34, 59–744 Weibel, D.B. et al. (2007) Microfabrication meets microbiology. Nat.

Rev. Microbiol. 5, 209–2185 Clapp, A.R. et al. (2006) Capping of CdSe-ZnS quantum dots with

DHLA and subsequent conjugation with proteins.Nat. Protoc. 1, 1258–

12666 Daniel, M.C. and Astruc, D. (2004) Gold nanoparticles: assembly,

supramolecular chemistry, quantum-size-related properties, andapplications toward biology, catalysis, and nanotechnology. Chem.Rev. 104, 293–346

7 Grainger, D.W. and Castner, D.G. (2008) Nanobiomaterials andnanoanalysis: opportunities for improving the science to benefitbiomedical technologies. Adv. Mater. 20, 867–877

8 Jaiswal, J.K. and Simon, S.M. (2004) Potentials and pitfalls offluorescent quantum dots for biological imaging. Trends Cell Biol.14, 497–504

9 Alivisatos, A.P. et al. (2005) Quantum dots as cellular probes. Annu.Rev. Biomed. Eng. 7, 55–76

10 Michalet, X. et al. (2005) Quantum dots for live cells, in vivo imaging,and diagnostics. Science 307, 538–544

11 Helmick, L. et al. (2008) Spatiotemporal response of living cellstructures in dictyostelium discoideum with semiconductor quantumdots. Nano Lett. 8, 1303–1308

12 Vollmer, W. et al. (2008) Peptidoglycan structure and architecture.FEMS Microbiol. Rev. 32, 149–167

13 Demchick, P. and Koch, A.L. (1996) The permeability of the wall fabricof Escherichia coli and Bacillus subtilis. J. Bacteriol. 178, 768–773

14 Li, W. et al. (2004) Exploring the mechanism of competencedevelopment in Escherichia coli using quantum dots as fluorescentprobes. J. Biochem. Biophys. Methods 58, 59–66

15 Kloepfer, J.A. et al. (2005) Uptake of CdSe and CdSe/ZnS quantum dotsinto bacteria via purine-dependent mechanisms. Appl. Environ.Microbiol. 71, 2548–2557

16 Hirschey, M.D. et al. (2006) Imaging Escherichia coli usingfunctionalized core/shell CdSe/CdS quantum dots. J. Biol. Inorg.Chem. 11, 663–669

17 de Pedro, M.A. et al. (1997) Murein segregation in Escherichia coli. J.Bacteriol. 179, 2823–2834

18 Aslan, K. et al. (2005) Plasmon light scattering in biology andmedicine:new sensing approaches, visions and perspectives. Curr. Opin. Chem.Biol. 9, 538–544

19 Liu, R. et al. (2007) A Simple and specific assay for real-timecolorimetric visualization of b-lactamase activity by using goldnanoparticles. Angew. Chem. Int. Ed. 46, 8799–8803

20 Nath, S. et al. (2008) Dextran-coated gold nanoparticles for theassessment of antimicrobial susceptibility. Anal. Chem. 80, 1033–1038

21 Konopka, M.C. et al. (2006) Crowding and confinement effects onprotein diffusion in vivo. J. Bacteriol. 188, 6115–6123

22 Atencia, J. and Beebe, D.J. (2005) Controlled microfluidic interfaces.Nature 437, 648–655

23 Whitesides, G.M. (2006) The origins and the future of microfluidics.Nature 442, 368–373

24 Xia, Y. and Whitesides, G.M. (1998) Soft lithography. Annu. Rev.Mater. Sci. 28, 153–184

25 Costerton, B. (2004) Microbial ecology comes of age and joins thegeneral ecology community. Proc. Natl. Acad. Sci. U. S. A. 101,16983–16984

26 Ottesen, E.A. et al. (2006) Microfluidic digital PCR enables multigeneanalysis of individual environmental bacteria. Science 314, 1464–1467

27 Marcy, Y. et al. (2007) Dissecting biological darkmatter with single-cellgenetic analysis of rare and uncultivated TM7 microbes from thehuman mouth. Proc. Natl. Acad. Sci. U. S. A. 104, 11889–11894

28 Rosenfeld, N. et al. (2005) Gene regulation at the single-cell level.Science 307, 1962–1965

29 Mihalcescu, I. et al. (2004) Resilient circadian oscillator revealed inindividual cyanobacteria. Nature 430, 81–85

30 Brehm-Stecher, B.F. and Johnson, E.A. (2004) Single-cell microbiology:tools, technologies, and applications.Microbiol. Mol. Biol. Rev. 68, 538–

55931 Balaban, N.Q. et al. (2004) bacterial persistence as a phenotypic switch.

Science 305, 1622–162532 Gefen, O. et al. (2008) Single-cell protein induction dynamics reveals a

period of vulnerability to antibiotics in persister bacteria. Proc. Natl.Acad. Sci. U. S. A. 105, 6145–6149

33 Berlatzky, I.A. et al. (2008) Spatial organization of a replicatingbacterial chromosome. Proc. Natl. Acad. Sci. U. S. A. 105, 14136–14140

34 Cabeen, M.T. and Jacobs-Wagner, C. (2005) Bacterial cell shape. Nat.Rev. Microbiol. 3, 601–610

35 Shapiro, L. et al. (2002) Generating and exploiting polarity in bacteria.Science 298, 1942–1946

36 Takeuchi, S. et al. (2005) Controlling the shape of filamentous cells ofEscherichia coli. Nano Lett. 5, 1819–1823

37 DiLuzio,W.R. et al. (2005)Escherichia coli swim on the right-hand side.Nature 435, 1271–1274

38 Hulme, S.E. et al. (2008) Using ratchets and sorters to fractionatemotile cells of Escherichia coli by length. Lab Chip 8, 1888–1895

39 Adler, J. (1966) Chemotaxis in bacteria. Science 153, 708–71640 Mao, H. et al. (2003) A sensitive, versatile microfluidic assay for

bacterial chemotaxis. Proc. Natl. Acad. Sci. U. S. A. 100, 5449–545441 Regenberg, B. et al. (2004) Use of laminar flow patterning for

miniaturised biochemical assays. Lab Chip 4, 654–65742 Rhee, S.W. et al. (2005) Patterned cell culture inside microfluidic

devices. Lab Chip 5, 102–10743 Lin, F. et al. (2004) Generation of dynamic temporal and spatial

concentration gradients using microfluidic devices. Lab Chip 4, 164–

167

107

http://dx.doi.org/10.1016/j.tim.2008.11.007


44 Thomas, W.E. et al. (2002) Bacterial adhesion to target cells enhancedby shear force. Cell 109, 913–923

45 Gutierrez, E. et al. (2008) Microfluidic devices for studies of shear-dependent platelet adhesion. Lab Chip 8, 1486–1495

46 Ingham, C.J. et al. (2007) The micro-petri dish, a million-well growthchip for the culture and high-throughput screening of microorganisms.Proc. Natl. Acad. Sci. U. S. A. 104, 18217–18222

47 Kim, D. and Beebe, D.J. (2007) Hydrogel-based reconfigurablecomponents for microfluidic devices. Lab Chip 7, 193–198

48 Kaeberlein, T. et al. (2002) Isolating ‘uncultivable’ microorganisms inpure culture in a simulated natural environment. Science 296, 1127–

112949 Kim, H.J. et al. (2008) Defined spatial structure stabilizes a synthetic

multispecies bacterial community. Proc. Natl. Acad. Sci. 105, 18188–

1819350 Kaehr, B. and Shear, J.B. (2008) Multiphoton fabrication of chemically

responsive protein hydrogels for microactuation. Proc. Natl. Acad. Sci.U. S. A. 105, 8850–8854

51 Xu, C.W. (2002) High-density cell microarrays for parallel functionaldeterminations. Genome Res. 12, 482–486

52 Narayanaswamy, R. et al. (2006) Systematic profiling of cellularphenotypes with spotted cell microarrays reveals mating-pheromoneresponse genes. Genome Biol. 7, R6

53 Weibel, D.B. et al. (2005) Bacterial printing press that regenerates itsink: contact-printing bacteria using hydrogel stamps. Langmuir 21,6436–6442

54 Cho, H. et al. (2007) Self-organization in high-density bacterialcolonies: efficient crowd control. PLoS Biol. 5, e302

55 Lee, J.H. et al. (2008) Microfluidic devices for studying growth anddetachment of Staphylococcus epidermidis biofilms. Biomed.Microdevices 10, 489–498

56 Little, A. et al. (2008)Rules of engagement: interspecies interactions thatregulate microbial communities. Annu. Rev. Microbiol. 62, 375–401

Elsevier.com – linking scientists

Designed for scientists’ information needs, Elsevie

customer-focused navigation and an intuitive arch

greater prod

The easy-to-use navigational tools and structure

from one entry point. Users can perform rapid an

functionality, using the FAST technology of Scirus.

define their searches by any number of criteria to p

specific author or editor, book publication date,

physical sciences and social sciences – or by produ

1800 Elsevier journals, 2200 new books every year

In addition, tailored content for authors, editors an

on new products

Elsevier is proud to be a partner with the scientific

our mission and values at Elsevier.com. Discover

medical communities worldwide through partnershi

awards from The Els

As a world-leading publisher of scientific, technical

linking researchers and professionals to the best t

deepest coverage in a range of media types to

breakthroughs in research and discovery, and t

Elsevier. Building insightswww.elsev

108

57 Nowak, M.A. (2006) Five rules for the evolution of cooperation. Science314, 1560–1563

58 Ohtsuki, H. et al. (2007) The one-third law of evolutionary dynamics. J.Theor. Biol. 249, 289–295

59 Balagadde, F.K. et al. (2008) A syntheticEscherichia coli predator–preyecosystem. Mol. Syst. Biol. 4, 187

60 Klajnert, B. and Bryszewska, M. (2001) Dendrimers: properties andapplications. Acta Biochim. Pol. 48, 199–208

61 Erben, C.M. et al. (2006) Single-molecule protein encapsulation in arigid DNA cage13. Angew. Chem. Int. Ed. 45, 7414–7417

62 Medintz, I.L. et al. (2005) Quantum dot bioconjugates for imaging,labelling and sensing. Nat. Mater. 4, 435–446

63 Maaloum, M. et al. (1998) Agarose gel structure using atomic forcemicroscopy: gel concentration and ionic strength effects.Electrophoresis19, 1606–1610

64 Normand, V. et al. (2000) New insight into agarose gel mechanicalproperties. Biomacromolecules 1, 730–738

65 Augst, A.D. et al. (2006) Alginate hydrogels as biomaterials.Macromol.Biosci. 6, 623–633

66 Kim, D.H. et al. (2006) Guided three-dimensional growth of functionalcardiomyocytes on polyethylene glycol nanostructures. Langmuir 22,5419–5426

67 Bonina, P. et al. (2004) pH-sensitive hydrogels composed of chitosanand polyacrylamide – preparation and properties. J. Bioact. Compat.Polym. 19, 101–116

68 Saunders, B.R. and Vincent, B. (1996) Thermal and osmotic deswellingof poly(NIPAM) microgel particles. J. Chem. Soc. Faraday Trans. 92

69 Medintz, I.L. et al. (2004) A fluorescence resonance energy transfer-derived structure of a quantum dot-protein bioconjugatenanoassembly. Proc. Natl. Acad. Sci. U. S. A. 101, 9612–9617

70 Takayama, S. et al. (2003) Selective chemical treatment of cellularmicrodomains using multiple laminar streams. Chem. Biol. 10, 123–

130

to new research and thinking

r.com is powered by the latest technology with

itecture for an improved user experience and

uctivity.

connect scientists with vital information – all

d precise searches with our advanced search

com, the free science search engine. Users can

inpoint information and resources. Search by a

subject area – life sciences, health sciences,

ct type. Elsevier’s portfolio includes more than

and a range of innovative electronic products.

d librarians provides timely news and updates

and services.

and medical community. Find out more about

how we support the scientific, technical and

ps with libraries and other publishers, and grant

evier Foundation.

and health information, Elsevier is dedicated to

hinking in their fields. We offer the widest and

enhance cross-pollination of information,

he sharing and preservation of knowledge.

. Breaking boundaries.ier.com

Documents

Dissecting microbiological systems using materials science