Keywords

4.1 Introduction

Porous Si (PSi) or SiO2 matrices have emerged as an attractive and versatile material for the construction of complex functional nanostructures (Li et al. 2003, 2005; Yoon et al. 2003). PSi is typically synthesized by anodic electrochemical etching of a single-crystal Si wafer in a hydrofluoric acid (HF) based electrolyte solution. At the end of this simple and cost-effective process, a nanostructure with complex properties is obtained (Sailor and Link 2005). One of the favorable properties of this nanomaterial is its large surface area (up to 500 m2/cm3), which enables large amounts and a variety of biomolecular interactions, including enzymes (DeLouise et al. 2005b), DNA fragments (Zhang and Alocilja 2008) and antibodies (Bonanno and DeLouise 2007), occurring over a small working area. The resulting activated surface can be used for several biosensing applications, including the detection of DNA (Zhang and Alocilja 2008), proteins (Dancil et al. 1999; Pacholski et al. 2005, 2006; Schwartz et al. 2007), enzyme activity (DeLouise et al. 2005b) and bacteria (Alvarez et al. 2007; Chan et al. 2001; Massad-Ivanir et al. 2010; Massad-Ivanir et al. 2011; Radke and Alocilja 2005). PSi optical sensors are based on changes of photoluminescence (Chan et al. 2001; de Leon et al. 2004) or reflectivity (Archer et al. 2004; Stewart and Buriak 2000) upon exposure to the target analyte (D’Auria et al. 2006), which replace the media in the pores. A change in the refractive index of the film is observed as a modulation in the photoluminescence spectrum or as a wavelength shift in the reflectivity spectrum, respectively.

A key challenge in PSi biosensors is to effectively stabilize the nanostructure during experiments in biological solutions, as PSi oxidation and dissolution in aqueous environments lead to significant signal baseline drifts, signal loss, and ultimately to structural collapse of the PSi thin film (Jane et al. 2009; Janshoff et al. 1998; Kilian et al. 2009). Another difficulty presented by PSi transducers is the susceptibility of proteins to undergo undesired conformation changes during deposition and patterning onto the Si (Burnham et al. 2006). Thus, it was demonstrated that the use of PSi as template or as a host matrix may eliminate these issues, while providing the means for construction of complex optical structures from flexible materials, such as polymers (Bonanno and DeLouise 2009a, b; DeLouise et al. 2005a; Li et al. 2003; Massad-Ivanir et al. 2010; Segal et al. 2007; Yoon et al. 2003). Specifically, the incorporation of hydrogels offers significant advantages due to their high optical transparency, good mechanical properties, ability to store and immobilize reactive functional groups and biological compatibility (Bonanno and DeLouise 2009a, b; Massad-Ivanir et al. 2010). Recent work on (oxidized) PSi/hydrogel hybrids (Bonanno and DeLouise 2009a, b; Massad-Ivanir et al. 2010; Perelman et al. 2010; Sciacca et al. 2011; Segal et al. 2007; Wu and Sailor 2009) demonstrated the potential of these nanomaterials for application in drug delivery, sensing and biosensing.

In the present work we describe the basic considerations in designing PSi-based optical transducers for label-free detection of microorganisms e.g., bacteria. We compare two strategies for the preparation of the biosensor surface. In the first approach, neat oxidized PSi (PSiO2) Fabry–Pérot thin films are used as the transducer and biofunctionalized with a monoclonal antibody (as the capture probe) using well established coupling chemistry. In the second approach, an antibody-modified PSiO2/hydrogel hybrid is designed and synthesized. The hydrogel is synthesized in situ within the PSiO2 host and conjugated with a monoclonal antibody through a biotin–streptavidin (SA) system. We demonstrate rapid detection of E. coli K-12 bacteria (as a model microorganism) via a “direct cell capture” approach onto these two types of PSi-based biosensors.

4.2 Materials and Methods

4.2.1 Materials

Highly-doped p-type Si wafers (0.0009 Ω-cm resistivity, <100  >  oriented, boron-doped) were purchased from Siltronix Corp. Aqueous HF (48%) and ethanol absolute were supplied by Merck. Acrylamide, N,N′-methylenebis(acrylamide) (BIS), bis(acryloyl)cystamine (BIS-CA), 2,2-dimethoxy-2-phenyl-acetophenone (DMAP), tris(2-carboxyethyl) phosphine (TCEP), Bis(N-succinimidyl)carbonate (SC), (3-aminopropyl)triethoxysilane (APTES), diisopropylethylamine (DIEA) and ­PEO-iodoacetyl biotin were obtained from Sigma Aldrich Chemicals. Streptavidin (SA), E. coli antibody and biotinylated E. coli antibody were purchased from Jackson ImmunoResearch Labs Inc. E. coli (K-12) was generously supplied by Prof. Sima Yaron (Technion).

4.2.2 Preparation of PSiO2 Substrates

PSi Fabry-Pérot thin films are prepared by anodic etch of highly doped p-type single-crystal Si wafers in a solution of 3:1 v/v 48% aqueous hydrofluoric acid:ethanol. A constant current density of 385 mA/cm2 is applied for 30 s. The resulting freshly-etched samples are thermally oxidized at 800°C, to create a SiO2 matrix.

4.2.3 Preparation of PSiO2  /Hydrogel Hybrids

The detailed synthesis scheme of these hybrids was previously described (Massad-Ivanir et al. 2010). Briefly, an aqueous pre-gel solution contains acrylamide monomers, cross-linking agents (BIS and BIS-CA) and photoinitiator (DMAP). The pre-gel solution is cast onto the PSiO2 film and allowed to infiltrate into the nanostructure. Photo-polymerization is initiated by UV irradiation (254 nm, 10 min).

4.2.4 Immobilization of Recognition Motives

4.2.4.1 Biofunctionalization of Neat PSiO2 Films

Preparation of APTES-modified surfaces: A PSiO2 sample is incubated with an aqueous solution of 42 mM APTES and 56 mM DIEA for 30 min. After the solution is removed, the surface is rinsed with purified water and ethanol for 10 min each and dried under a nitrogen stream.

Preparation of NHS-modified surfaces: The APTES-modified surface is immersed in a 10 mM SC solution in acetonitrile for 7 min. After the solution is removed, the surface is washed extensively with acetonitrile three times for 10 min each and dried under a nitrogen stream.

Preparation of IgG-modified surfaces: The NHS-modified surface is incubated in 100 μg/mL E. coli antibody solution at room temperature for 60 min. After the solution is removed, the surface is rinsed with PBS, 1 M NaCl, and PBS for 10 min each.

4.2.4.2 Biofunctionalization of PSiO2  /Polyacrylamide Hybrids

The hybrids are exposed to a reducing agent, TCEP (10 mM), to generate reactive thiol groups throughout the hydrogel. The activated hydrogel is then treated with PEO-iodoacetyl biotin (100 μM), a thiol-reactive biotin linker molecule, and reacted with SA (100 μg/mL). The resulting SA-modified hybrids are incubated with biotinylated E. coli antibody (100 μg/mL).

4.2.5 Measurement of Interferometric Reflectance Spectra

Interferometric reflectance spectra of the samples are collected using an Ocean Optics charge-coupled device (CCD) USB 4000 spectrometer fitted with a microscope objective lens coupled to a bifurcated fiber-optic cable. A tungsten light source is focused onto the center of the sample surface with a spot size of approximately 1–2 mm2. Reflectivity data are recorded in the wavelength range of 400–1,000 nm, with a spectral acquisition time of 100 ms. Both illumination of the surface and detection of the reflected light are performed along an axis coincident with the surface normal. All the optical experiments are conducted in a fixed cell to ensure that the sample reflectivity is measured at the same spot during all the measurements. Spectra are collected using a CCD spectrometer and analyzed by applying fast Fourier transform (FFT).

4.2.6 Bacteria Culture

E. coli K-12 is cultivated in a 10 mL tube with LB medium (5 mL) (medium composition in deionized water (1 L): NaCl (5 g), yeast extract (5 g), and tryptone (10 g)). The bacteria are incubated overnight at 37°C with shaking.

4.2.7 Bacteria Sensing

IgG-modified PSiO2, IgG-modified hybrid, neat PSiO2 and unmodified hybrid (as controls) samples are incubated with E. coli K-12 suspensions (at concentrations ranging from 10  3 to 10  5 cell/mL) for 30 min in a fixed cell. After the bacteria suspension is removed, the cell is flushed for 30 min with a buffer solution. Optical measurements are recorded throughout the experiment. The FFT intensity changes are expressed as percentages and are calculated using the following equation:

$$ \text{Intensity change}(\%)=\frac{{\text{A}}_{1}-{\text{A}}_{2}}{{\text{A}}_{2}}\times 100$$
(4.1)

where A1 is the intensity before modification and A2 is the intensity after modification with the bacteria suspensions.

4.3 Results and Discussion

4.3.1 Preparation and Characterization of PSiO2 Films

The PSiO2 film is prepared from a highly doped p-type single-crystal Si wafer, polished on the <100  >  face using an anodic electrochemical etch, at a constant current density of 385 mA/cm2 for 30 s. The resulting freshly etched PSi template is then thermally oxidized at 800°C to create a hydrophilic PSiO2 matrix. The resulting porous layer is approximately 7,880 nm thick with interconnecting cylindrical pores ranging in diameter from 60 to 100 nm and the calculated porosity is approximately 80%. The structural properties, i.e., thickness and porosity, of the PSiO2 layer are thoroughly ­characterized by scanning electron microscopy, gravimetry (for porosity), and a spectroscopic liquid infiltration method, as previously described by Massad-Ivanir et al. (2010).

4.3.2 Biofunctionalization of Neat PSiO2 and PSiO2  /Hydrogel Hybrids with Antibodies

4.3.2.1 Neat PSiO2

The synthetic approach for grafting the monoclonal antibodies (IgG) onto neat PSiO2 surfaces is based on a well established silanization technology (Xia et al. 2006), as previously described (Massad-Ivanir et al. 2011).

4.3.2.2 PSiO2  /Hydrogel Hybrids

Biofunctionalization of the hybrids is achieved by exposure of the PSiO2/di-sulfide linked hydrogel to a reducing (TCEP) to generate reactive thiol groups throughout the hydrogel phase. Next, the activated hydrogel is incubated with a SH-reactive biotin linker molecule, followed by attachment of SA to the hydrogel via the SH-linked biotin. Functionalization is accomplished through the bioconjugation of biotinylated E. coli monoclonal antibodies to the SA linker molecule.

4.3.3 Stability of PSiO2–Based Transducers in Aqueous Media

The optical readout stability of the two different transducers (neat PSiO2 and hybrid) under flow conditions is characterized by long buffer (PBS) flow experiments of several hours. The samples are placed in a plexiglas custom-made flow cell, in which a buffer solution is delivered at a constant buffer flow of 0.5 mL/min for 5 h and the reflectivity spectra is collected every 15 s.

The reflectivity spectrum of the thin PSi layer consists of a series of interference fringes that result from a Fabry–Pérot interference. This fringe pattern arises from reflections at the top and at the bottom of the film, so that the measurement is made over the entire volume of the system. The maxima of these fringes are governed by the following relationship (Fabry–Pérot equation) (Sailor and Link 2005):

$$ \text{m}\lambda =2\text{nL}$$
(4.2)

The effective optical thickness (EOT) of the sample refers to the 2 nL term in the Fabry–Pérot formula (where m is an integer, n is the average refractive index, L is the thickness of the film, and λ is the wavelength of the incident light). A change in the average refractive index (n) leads to a shift in the observed reflectivity spectrum that correlates with EOT changes.

Figure 4.1 shows the results of the buffer flow experiments for both silanized-PSiO2 and hybrid transducers. Changes in the EOT values and the reflected light intensity with time are depicted in Fig. 4.1a and b, respectively. This study demonstrates that the optical readout of the hybrid transducer is highly stable throughout the entire experiment. Changes of less than 0.5% in the intensity and 10 nm in the EOT values are recorded. Thus, a robust optical readout with no evident baseline drift is obtained. On the contrary, the silanized-PSiO2 transducer exhibits significant baseline drifts of both the EOT and the intensity signals with time. The EOT value of the film decrease by approximately 90 nm within 5 h and the reflected intensity drops by 13%. These significant changes in the optical readout are ascribed to time-dependent degradation of the porous scaffold (Janshoff et al. 1998; Massad-Ivanir et al. 2010; Pacholski et al. 2006). Thus, these experiments clearly show that the ­presence of the hydrogel within and delicate and highly porous SiO2 scaffold stabilize the nanostructure under aqueous conditions. Similar results were recently reported by Cunin et al. (Sciacca et al. 2011) for a PSi/chitosan hybrid used as an optical transducer for the detection of carboxylic acid-containing drugs in water.

Fig. 4.1
figure 1_4

Buffer flow experiments onto silanized-PSiO2 and PSiO2/polyacrylamide hydrogel hybrid transducers. (a) EOT changes with time and (b) Intensity changes with time. The reflectivity spectra of the samples is collected every 15 s under constant PBS buffer flow of 0.5 mL/min for 5 h

Full size image

4.3.4 Optical Detection of E. coli Bacteria

Our biosensing approach is based on monitoring changes in the light reflected from the biofunctionalized PSiO2 or hybrid nanostructures. Changes in the amplitude (intensity) of the FFT peak of the biosensor are correlated to specific immobilization of bacteria cells, “direct cell capture” onto the transducer surface via antibody-antigen interactions. Since E. coli bacteria are large biological ­species, with typical dimensions of 0.8–2 μm (Sundararaj et al. 2004), binding of bacteria will occur only on the biosensor surface and not inside the porous nanostructure. Moreover, we expect that their presence on the surface will scatter the light, resulting in a significant decrease in the reflected light intensity (Alvarez et al. 2007; Massad-Ivanir et al. 2010, 2011). An experimental setup for continuous monitoring of the reflectivity spectrum of the hybrids upon incubation with bacteria suspensions is designed and constructed. The biosensors are exposed to E. coli K-12 ­suspensions with different ­concentrations, ranging from 103 to 105 cell/mL. The incubation time was set to 30 min, after which the samples were washed with a buffer solution for 30 min. Figure 4.2 displays the optical response of the different biosensors upon introduction to different concentrations of E. coli suspensions. Indeed, both biosensors exhibit a decrease in intensity upon exposure to E. coli, while insignificant changes in the FFT spectrum are recorded for the unmodified surfaces. In order to confirm that this intensity change results from bacteria capture onto IgG-modified surfaces, the biosensors are studied under a light microscope immediately after the biosensing experiment. Indeed, immobilized bacteria cells are observed onto the biosensor surface, while no cells are observed onto the unmodified surfaces (data not shown). Moreover, the intensity signals of the two biosensors are proportional to the bacteria suspension concentration (see Fig. 4.2). As expected, exposure of the biosensors to a lower bacteria concentration (103 cells/mL) results in a signal decrease; i.e., a smaller change in intensity is observed. When comparing the response of the two biosensors (PSiO2 vs. hybrid) to E. coli exposure, the hybrid-based biosensors exhibit significantly higher intensity changes (∼four fold higher). These results indicate that the hybrid system outperforms the PSiO2 biosensors in terms of their sensitivity. It is well established that the immobilization step is crucial in maintaining the conformation and the immunoactivity of the IgG molecules (Somasundaran 2006). We assume that the presence of the hydrogel (a soft organic component) maintains the proteins desired conformation during the immobilization and patterning onto the biosensor surface.

Fig. 4.2
figure 2_4

E. coli K-12 biosensing experiments. IgG-modified PSiO2 and hybrid are incubated with different concentrations of E. coli suspensions and the maximal intensity decrease value is presented. Control experiments, incubation of unmodified PSiO2 and hybrid surfaces with bacteria suspensions

Full size image

In terms of the sensitivity of the different biosensors, these preliminary experiments show relatively low detection limits of 103 and 104 cell/mL for the modified hybrids and PSiO2 biosensing platforms, respectively. For comparison, the detection limit of current state-of-the-art surface plasmon resonance (SPR) biosensors is in the range of 102–106 cells/mL (Dudak and Boyaci 2007, 2009; Skottrup et al. 2008; Taylor et al. 2008). Moreover, the response time of these biosensors to bacteria exposure is comparable to that of SPR techniques.

4.4 Conclusions

Two classes of label-free optical biosensors for bacteria detection are synthesized and characterized. The first platform is based on a PSiO2 nanostructure (Fabry–Pérot thin film) and the second consists of a PSiO2/polyacrylamide hydrogel hybrid. The different nanostructures’ surfaces are biofunctionalized with IgG, as a capture probe, and their potential applicability as a biosensor for bacteria detection is demonstrated. The hybrids show improved optical readout stability under aqueous conditions. Moreover, the presence of the hydrogel also enables desired conformation of the antibodies onto the biosensor surface during deposition and patterning, resulting in a higher sensitivity. This proof-of-concept work demonstrates a simple and sensitive detection scheme of bacteria via a “direct cell capture” approach. Our preliminary biosensing experiments demonstrate a detection limit of 103–104 cells/mL for E. coli and a response time of several minutes. We are currently exploring several approaches to enhance the sensitivity of these biosensors, including improving optimization of the antibody concentration and orientation, enhancement of the coupling chemistry and use antibody fragments.