Method and apparatus for bacterial monitoring

Abstract

A system for detecting target elements such as bacteria in a host analyte, comprising a substrate with an ordered array of wells having diameters to fit the size of the targets. The substrate may be a periodic macro-PSi array structure (MPSiAS) illuminated with a broadband source. The reflected light spectrum diffracted from the substrate is optically analyzed to provide the effective optical depth of the wells. Fast Fourier Transform analysis may be used for the optical analysis. Entry of target elements into wells is detected by the change in the effective optical depths of the wells. Micro-organisms as large as bacteria and viruses having dimensions comparable with the wavelength of the illumination can thus be detected. Wells with an inner section impenetrable by the target cells enables compensation for environmental changes. The detection may be performed in real time, such that production line bacterial monitoring may be achieved.

Claims

1. A system for detecting target elements in a host analyte, said system comprising: a substrate containing an ordered array comprising a predetermined pattern of wells formed in its surface, at least some of said wells having predetermined lateral dimensions such that said target elements can fit therein; a broadband illumination source configured to illuminate said surface of said substrate with broadband illumination over a range of wavelengths; an optical detector disposed such that it collects illumination diffracted from said substrate, and outputs a reflected spectrum signal; and a signal processing unit, adapted to analyze said reflected spectrum signal to provide a measure of the effective optical depth of said wells, wherein at least some of said wells have a lateral dimension at least as large as those wavelengths of said illumination source which are diffracted from said substrate and detected by said optical detector, to output said reflected spectrum signal which is analyzed by said signal processing unit, such that said system can detect said target elements.

2. A system according to claim 1 wherein said signal processing unit analyzes said reflected spectrum signal using the Fast Fourier Transform.

3. A system according to claim 1 wherein said optical detector is disposed normal to said substrate, such that it collects diffracted light of zero order from said substrate.

4. A system according to claim 1, wherein said effective optical depth of said wells provides an indication of the concentration of said target elements captured within said array of wells.

5. A system according to claim 2, wherein said effective optical depth is determined from the position of a peak obtained from the results of Fast Fourier transform analysis on said reflected spectrum.

6. A system according to claim 1, wherein said ordered array of wells comprises a lamellar photonic crystal grating.

7. A system according to claim 1, wherein said substrate is a silicon chip, and said ordered array of wells are constructed by microelectronic fabrication processes.

8. A system according to claim 1, wherein said target elements are bacterial cells having dimensions larger than the wavelengths of said illumination source detected by said optical detector element.

9. A system according to claim 1, wherein said wells comprise capture probes having a high affinity to the target elements intended to be measured by said system.

10. A system according to claim 9, wherein said target elements are micro-organisms, and said capture probes are any one of antibodies, aptamers or other peptides.

11. A system according to claim 10, wherein said microorganisms are bacterial cells, and said capture probes are specific antibodies.

12. A system according to claim 1, wherein said system is adapted to provide real time detection of microorganisms.

13. A system according to claim 9, further comprising a cell nutrient supply, such that the growth of a microorganism can be observed after application of said nutrient supply.

14. A system according to claim 1 wherein at least some of said wells include a recognition moiety adapted to said target elements to be detected.

15. A system according to claim 14, wherein said substrate comprises at least two different regions, the wells in each of said regions including a different recognition moiety, such that each of said different regions can detect different target elements concurrently.

16. A system according to claim 1, wherein at least some of said wells have at least two sequential sections having different lateral dimensions, and wherein a second section, further from said surface than a first section, has a lateral dimension less than that of said first section.

17. A system according to claim 16, wherein the dimensions of said second section are such that said target elements cannot penetrate said second section, while said host analyte can.

18. A system according to claim 16, wherein change of said measured effective optical depth of said second section of at least some wells is utilized as a marker to compensate for changes in environmental conditions that cause said effective optical depth of both of said first and second sections to change.

19. A system according to claim 16, wherein said second section is provided with sensitivity to a material which targets trapped in said first section may secrete.

20. A system according to claim 19 wherein said sensitivity provides further information regarding the level of targets trapped in said first sections of said wells.

21. A system according to claim 1 wherein at least some of said wells have a depth at least several times as large as the wavelengths of said illumination source detected by said optical detector.

22. A system according to claim 1 wherein said wells in said predetermined pattern are arranged in an ordered array of rows and columns.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) The present invention will be understood and appreciated more fully from the following detailed description, taken in conjunction with the drawings in which:

(2) FIGS. 1A and 1B show SEM micrograph views of a two-dimensional periodic macro-PSi array structure (MPSiAS) of the type described in this disclosure;

(3) FIG. 1C illustrates schematically an exemplary system for using such a structure for detecting micro-organisms, as disclosed in the present application;

(4) FIG. 1D is an exemplary graph showing the optical output used to determine the target concentration in the system of FIG. 1C;

(5) FIG. 1E illustrates schematically the steps for fabrication of a device of the type shown in FIG. 1A, using the electro-chemical etching method;

(6) FIG. 2 illustrates schematically a bacterial monitoring system constructed using an MPSiAS of the type shown in FIG. 1A or 1B;

(7) FIG. 3 shows an example of the output of a typical real time detection of the level of E. coli K12, using a system such as that described in FIG. 2;

(8) FIGS. 4A to 4C illustrate schematically a model by which the filling of the MPSiAS pores by the bacterial species is reflected by the change in effective optical depth measurements performed by the system;

(9) FIG. 5 shows some measured EOT values for different sensing experiments and the corresponding fill fractions calculated by the model of FIGS. 4A to 4C; and

(10) FIG. 6A illustrates schematically an alternative MPSiAS geometry, having pores with two different cross sections, and FIG. 6B shows the ensuing FTT output plot of the light reflected from such an array.

DETAILED DESCRIPTION

(11) Reference is now made to FIG. 1A, which illustrates a cross-sectional SEM micrograph view and, in the insert, a top view, of an exemplary two-dimensional (2D) periodic macro-PSi array structure (MPSiAS), having a pore diameter adapted to fit the size of the target bacteria cells. The structure shown in FIG. 1A is of an array fabricated by electrochemical etching after photolithography. For the purpose of describing the systems and methods of the present disclosure, the non-limiting example used in this detailed description is of a substrate structure having pores suitable in size for the detection of E. coli K-12 bacteriawhich have diameters of 0.8 to 2 m. The periodicity of the structure shown in FIG. 1A is about 3 m. FIG. 1B shows a plan and cross-sectional SEM micrograph view an example of an MPSiAS produced using dry-etching as an alternative fabrication technology. In this example, the periodicity of the structure is about 3 m.

(12) Reference is now made to FIG. 1C which illustrates schematically how the ordered array of wells 12 in the MPSiAS substrate 10 acts as a lamellar (or a phase) grating that reflects the incident light 11 falling on the structure into a set of diffraction orders at various angles according to the relationship between the periodicity of the grating and the optical wavelength. For light collected normal to the array surface, only the zero-order diffraction is measured, i.e. backscattered light having =0, where is the diffraction angle, yielding the following expression for the intensity I of the zero-order reflected light:

(13) $\begin{array}{cc}I\left(=0\right)=I_0{\cos }^2\left({}_0/2\right)\mathrm{where}& \left(1\right)\\ 2{}_0=2\mathrm{kL}=\frac{2}{}\left(2n_{0}L\right)& \left(2\right)\end{array}$
and .sub.0 is the phase delay between the incident and the reflected beams, is the free space optical wavelength, k is the wave number, L is the depth of the pores, n.sub.0 is the refractive index of the medium filling the pores, and the term 2n.sub.0L is thus the optical path difference between the top surface of the device and the bottom surface of the wells, referred to as the effective optical thickness (EOT) of the lamellar grating. By using phase gratings having fairly deep wells, typically of the order of several times to several tens of times the optical wavelength, an interference pattern between the light reflected from the top of the grating structure and from the bottom of the wells can be obtained, analogous to the spectral response obtained from Fabry Perot interference.

(14) In order to monitor the EOT of the device as the analyte solution is loaded, Fast Fourier transform (FFT) of the reflectivity spectrum from the MPSiAS layer is performed, resulting in a characteristic single peak, whose position is indicative of the EOT, as shown in the example graph of FIG. 1D of the reflected intensity I in arbitrary units, as a function of the optical path difference in nm. In the example shown in FIG. 1D, the optical path difference is seen to be of the order of 15 m. Sensing is accomplished once bacteria penetrate into the macro-pores, inducing measurable changes in the EOT that can be monitored and quantified in real-time via RIFTS analysis. The magnitude of the peak is a function of the bacterial concentration above the macro-pore array.

(15) Reference is now made to FIG. 1E, which shows the fabrication chain for one exemplary process for fabricating the MPSiAS samples using photolithography followed by electrochemical anodization. As per conventional semiconductor processing, a SiO.sub.2 layer 13 is formed on the top surface of the porous Si wafer 10 by thermal oxidation, followed by the photoresist layer 14 and a mask, typically formed by e-beam or holographic lithography, for defining the desired 2-D pore or well structure pattern. Next, the unexposed photoresist is removed to expose the etching pattern 15, followed by alkaline etching through the photolithographic oxide mask to generate a 2-dimensional pattern of inverted pyramid grooves 16 on top of the Si wafer. These inverted pyramids are used as the initial openings for forming the pore or well structures to the desired profile and depth. An electrochemical-etching process 17 can be advantageously used for this, performed under dark conditions, using a solution of hydrofluoric acid (HF) and dimethylformamide (DMF), at a constant current density, typically of the order of 30 mA/cm.sup.2 for this type of substrate and well size. For the current example of E. coli detection, structures with a periodicity of 2.5 m to 4 m, and depths in the range of 2.5-7.5 m can be used, to allow a facile entrapment of the bacteria cells within the array. Following anodization, which may typically be carried out under dark conditions, using an electrolyte solution of HF (49%) and DMF (1:7 v/v), at a constant current density of 30 mA/cm.sup.2 for 125 s, the resulting freshly treated PSi is chemically oxidized, optionally in a 5:1:1 solution of H.sub.2O:NH.sub.3:H.sub.2O.sub.2 to create a hydrophilic porous SiO.sub.2 matrix. Next, the oxidized MPSiAS samples may be modified in order to functionalize the porous surface specific or non-specific chemistries to target the microorganisms of interest; for example, by surface functionalization with positively charged groups; e.g., by using 2-aminopropyltriethoxy silane (2% APTES diluted in toluene). As most bacteria carry a net negative surface charge, adhesion of E. coli is promoted on the positively charged surfaces.

(16) Reference is now made to FIG. 2, which illustrates schematically an exemplary bacterial monitoring system constructed to implement the methods described hereinabove. The MPSiAS 20 is shown mounted in the sampling chamber 21 of the sensor unit, which has a transparent cover 22 such as plexiglass sealed to the housing by means of an O-ring 23 to maintain the analyte host liquid within the sampling chamber. The sample liquid to be tested is flowed through the sampling chamber, as indicated by the arrows 24 in FIG. 2. The broadband light source, shown as a tungsten lamp 25 in FIG. 2, may be projected into the sampling chamber by means of a fiber optical feed, preferably a bifurcated fiber optical probe 26, and the light reflected from the silicon chip may be collected by the same fiber optical probe. The spectrum of this reflected light is now analyzed, which, in the exemplary system shown in FIG. 2, is performed by input into a mini spectrometer 27, such as the Ocean Optics USB 4000, available from Ocean Optics Inc. of Dunedin, Fla., USA, and the interferometric reflectance spectrum 28 is then FFT analyzed, conveniently by means of commercially available software, such as the Igor Pro, v. 6.03, available from Wavemetrics Inc, of Portland Oreg., USA. In the monitoring system shown in FIG. 2, the output 29 is shown as a continuous plot of the EOT displayed as a function of time, the change in the level of the EOT indicating a change in the level of target bacterial cells detected by the system. However, the processing system can be configured so that it provides a direct reading of the bacterial concentration. This requires prior calibration runs in which the bacterial concentration measured by the system is checked against another direct observation method, such as using confocal laser scanning microscopy.

(17) Reference is now made to FIG. 3, which shows an example of the output of a typical real time detection of the level of E. coli K12, using a system such as that described in FIG. 2. The top graph of FIG. 3 shows the value of EOT measured as a function of elapsed time, while the bottom graph shows the reflectance intensity plotted as a function of time. The output data was recorded every 15 sec. In these plots, the changes in the FFT spectrum before and after the introduction of E. coli bacteria are depicted as a function of time. Sensing is accomplished by measuring the variation in the EOT and the intensity of the sensor once the bacteria cells are trapped inside the pores. In the upper plot, there is shown on the right hand ordinate, the measured EOT of the pores plotted as a function of time. The left hand ordinate shows the change, EOT measured. The graphs show an initial baseline plot with a constant flow of 0.85% w/v NaCl saline solution at 0.1 ml/min. At the point in time represented by minute 47, a bacterial suspension of 10.sup.6 cell/mL E. coli K12 in saline, was delivered to the detector system. A rapid increase in EOT of approximately 45 nm was observed. This EOT change is attributed to the entrapment of bacteria within the pores, leading to a refractive index increase. Simultaneously, in the lower graph of FIG. 3, the reflectance intensity is plotted as a function of time, and as is observed, a decrease in the reflectance intensity of about 3% occurs, probably due to light scattering induced by the bacteria cells. Optical studies demonstrate a detection limit of the order of 10.sup.5 cell mL.sup.1 for E. coli.

(18) The inset graph of FIG. 3 shows a time-zoomed view of the EOT shift itself, to show the speed with which the bacterial level can be detected. As is observed, for this particular MPSiAS, the complete 45 nm. shift in EOT generated by the bacterial filling of the pores takes place in less than 1 minute, implying that an indication of the presence of a bacterial contaminant can be provided in a fraction of this time, thus illustrating the usefulness of the device for on-line, real time monitoring. Thus, the zero-order diffraction of the reflected light presents a spectral interference pattern according to the phase accumulated inside the pores, allowing for real-time detection of bacteria capture.

(19) Reference is now made to the three parts of FIG. 4 which illustrate schematically a model showing the manner in which the filling of the MPSiAS pores by the bacterial or other large cell species is reflected by the change in effective optical depth measurements performed by the system.

(20) The model attempts to correlate between the optical readout of the sensor i.e., EOT shift, and the bacteria concentration as expressed by the fill factor of the pores determined by direct microscopic investigation, such as by the Confocal Laser-Scanning Microscopy (CLSM) technique. From image analysis of the CLSM data, the relative number of pores occupied by bacteria can be quantified, this value being referred to as the fill fraction of the MPSiAS. In the line (A) of FIG. 4, there is shown a schematic drawing of an MPSiAS, having some of its pores containing trapped bacteria 40. In the line (B) of FIG. 4, the partly filled and partly empty pores are replaced by an effective continuous layer of thickness, l, and refractive index, n, where l is less than the average length h of the bacteria, since only some of the pores are filled. The line (C) of FIG. 4 shows the schematics of the mathematical model used to simulate the sensing experiments. Thus, for a given EOT shift, the model predicts the corresponding fill fraction. The refractive indices of the bacteria (n) and of the host saline solution (n.sub.0) are taken to be 1.4 and 1.33 respectively, so that the only free parameter in the model is the effective thickness of the filled pores, or equivalently the effective fill factor (eff). This may be defined as
eff=(l/L).sup.2(3)
where, L is the pore depth, and the squared relation comes because the array is a 2-dimensional array. This quantity can be estimated from the model presented in line (B) of FIG. 4 to be:

(21) $\begin{array}{cc}\mathrm{eff}=\frac{\mathrm{EOT}/\mathrm{EOT}}{n/n_0}& \left(4\right)\end{array}$
where (EOT)/EOT is the relative change of the EOT as measured during the sensing experiment, and n=nn.sub.0, is the absolute change of the refractive index due to bacteria capture. The fill fraction of the MPSiAS can directly be related to the effective fill factor as follows:

(22) $\begin{array}{cc}\mathrm{fill}\mathrm{fraction}=\mathrm{eff}.Math.\left(\frac{{\mathrm{volume}}_{\mathrm{pores}}}{{\mathrm{volume}}_{E.\mathrm{coli}}}\right)& \left(5\right)\end{array}$

(23) Reference is now made to FIG. 5, which is a table showing some measured EOT values for different sensing experiments and the corresponding fill fractions calculated by the model of FIG. 4. FIG. 5 shows the fill fraction values for MPSiAS sensors, characterized by different periodicities of 2.5 and 4 m. The values are calculated by averaging at least 5 images taken at different locations for each sensor. As is observed from FIG. 5, structures with high periodicity i.e., larger pores, exhibit greater EOT shifts, corresponding to superior bacteria capture. The model results are in fairly good agreement with the fill fraction estimated by the CLSM technique indicating that, despite of the model's approximations, it provides a reasonable description to the sensing events. For example, for sample 1, the model predicts a fill fraction of 25%, while the CLSM data yields a fill fraction value of approximately 21%. The deviations between the fill fraction values are mainly attributed to the model assumptions. First, the model refers to the bacteria as a homogenous liquid filling the pores. This assumption oversimplifies the complex structure of the cell and its heterogeneity. Second, as the reflected light is collected normal to the pore's surface, only the intensity of the zero-order diffraction is measured. Therefore, due to typical pore morphology of the MPSiAS (see FIG. 1A) the model takes into consideration a diameter of about 25% of the opening of the pore for the effective coherent reflective surface, which is used to calculate the effective pore volume and hence the resulting fill fraction. Another cause for deviation may result from the difference in spot size from the optical data collected in the experimental sensing setup and in the CLSM. In the sensing setup, the EOT signal was measured from a single spot (with a typical diameter of 1 mm), while CLSM measurements were averaged over five different areas of the sensor's surface.

(24) It is known from US Patent Application Publication US 2007/0108465 to C. Pacholski et al for Porous Microstructure Multi-Layer Spectroscopy and Biosensing that a conventional porous silicon thin film structure can be constructed having pores with two different cross sections, an outer or upper section having a larger average diameter (for a cylindrical pore distribution) and an inner or lower section, having a narrower average diameter. In that application, the pore sizes of the upper section are limited to 20 to 50 nm, as expected from prior art random porous silicon devices, while that of the lower sections are of the order of <20 nm. Therefore, the pores are incapable of measuring larger micro-organisms, as described in the current disclosure. In the Pacholsky application the structure is composed of two layers of PSi, each layer having a distribution of pores having different and random diameters, with the lower layer having smaller average diameters than the upper layer.

(25) Reference is now made to FIG. 6A, which illustrates an alternative MPSiAS structure to those described hereinabove in FIGS. 1A to 5, in that it has a two-section pore structure. In contrast to what is shown in Pacholski, the implementation of FIG. 6A shows a single lamellar grating structure 60 having a single periodicity, but in which each period of the grating consists of two-sections. The upper section is wider and has a diameter d.sub.1 and a length l.sub.1, while the lower section is narrower having a diameter d.sub.2 and a length l.sub.2. If the diameter of the target cell is denoted by D, then, the pore diameters of both sections should follow the following condition:
d.sub.2<D<d.sub.1
In this case, the target cells 61 are capable of penetrating into the upper section of the pores and of being captured there, but are not capable of penetrating into the lower section of the pores. The host solution, however, can flow into both sections of the pores as schematically illustrated in FIG. 6A.

(26) When illuminated by a white light broadband source, reflection from the grating into a set of diffraction orders occurs, again, according to the relationship between the periodicity, p, and the optical wavelength. As before, only the zero-order reflection is collected by the optics but a more complex interference pattern is obtained from the two-section pores. Reference is now made to FIG. 6B, which is a plot obtained of the intensity of the reflected light in FTT space as a function of the optical path differences EOT. Three peaks are shown marked as:

(27) (i) 2.Math.n.sub.0.Math.l.sub.1 (EOT of section 1 of each pore),

(28) (ii) 2.Math.n.sub.0.Math.l.sub.2 (EOT of section 2 of each pore) and

(29) (iii) 2.Math.n.sub.0.Math.(l.sub.1+l.sub.2) (EOT of the total length including both sections of the pore),

(30) where n.sub.0 is the refractive index of the host analyte solution.

(31) However, these peaks behave differently from those described in Pacholski, since the physical mechanism is different, being directly based on the optical thicknesses and the refractive indices of the host solution (with a refractive index, n.sub.0) and the target cells (with a refractive index, n). On the other hand, in the Pacholski structure the effective optical thicknesses (EOTs) and the effective refractive indices of the layers are based on complex averaging over the refractive indices of the solution/target molecules with those of the porous media. Hence, the two-section MPSiAS of the present application allows direct sensing of the optical thicknesses of the host solution (in the lower section, subscripted 2, of the pores shown in FIG. 6A and the target cells (in the upper section, subscripted 1, of the pores shown in FIG. 6A, while in the Pacholski structure the EOTs measure more complex functions, that involve complicated averaging over the properties of the host/targets and the porous media in which they are embedded.

(32) In the two-section MPSiAS of the present application, if both of the sections of the pores are filled with the host analyte solution, then any change in environmental conditions causes all three peaks to move proportionally, maintaining, at least to first order, the same ratios between them.

(33) However, if now bacterial cells, having a refractive index n, enter the top section 1 of a pore, peaks (ii) and (iii), being dependent on l.sub.1, will, assuming that n>n.sub.0, shift to the right to a larger EOT value. Thus peak (ii) moves to position 2nl.sub.1, and peak (iii) moves to a position 2(n.Math.l.sub.1+n.sub.0.Math.l.sub.2). Denoting the value of (nn.sub.0) as n, each of peaks (ii) and (iii) move by a distance 2nl.sub.1.

(34) Peak (i) on the other hand, being dependent only on l.sub.2, does not shift, since, in the absence of a target element entry, no change has occurred to its optical path length l.sub.2. Hence, the distance between the first peak 2.Math.n.sub.0.Math.l.sub.2, which is not affected by the capture of the target cells, and the other two peaks is a measure to the number of bacteria (or target) cells which have been captured by the pores of the device, independent of environmental changes which affect the peaks in a similar manner. The greater the number of target elements captured, the larger the shift right of peaks (ii) and (iii). Furthermore, the distance between the second and third peaks can be used as a reference to eliminate random fluctuations in the signals that may cause false alarm events. Of course, if n<n.sub.0, then the peak shifts will be towards shorter EOT levels.

(35) This array configuration makes it possible to eliminate the effect of environmental change on the measurement of the bacterial concentration, by measuring the relative shift between the first peak (i) and the other two peaks, (ii) and (iii), whereby the shift between the peak due to the lower section and that due to the upper or both sections is used as an environmental marker.

(36) An additional and novel advantage of such a dual cross-section pore can be implemented if the lower, narrow section of the pore is provided with target selectivity to a material which the microorganisms trapped in the upper, wide section of the pore, may secrete. Such an implementation then provides a further level of validation as to the presence and even the quantity of microorganisms trapped in the device. Thus, for instance, the lower narrower sections of the pore may contain a trapping material for combining with proteins or toxins that bacteria trapped in the upper section may secrete. The selective trapping of these materials will result in a change in the refractive index of the narrow section of the pores, which will be reflected in a further change in the EOT characteristic of the narrow section of the pores, thereby providing a secondary measure of the presence of the specific bacteria in the upper, wider section of the pores. For example, Shiga toxin-producing E. coli organisms (STEC) are pathogens capable of producing sporadic and epidemic diarrhea, hemorrhagic colitis, and potentially life threatening hemolyticuremic syndrome. STEC possess a number of virulence factors, and the production of Shiga toxins (Stx1 and/or Stx2) is the most critical. Detection and identification of non-O157:H7 STEC serotypes in a timely fashion are more difficult even in a laboratory setting. Thus, by using the dual cross-section pore structure (in combination with appropriate surface modification of the narrow pores with receptors for Shiga toxins (antibodies of glycans) it would be possible to distinguish between STEC and non-STEC.

(37) It is appreciated by persons skilled in the art that the present invention is not limited by what has been particularly shown and described hereinabove. Rather the scope of the present invention includes both combinations and subcombinations of various features described hereinabove as well as variations and modifications thereto which would occur to a person of skill in the art upon reading the above description and which are not in the prior art.