Field of the invention The present invention relates to a sensing platform and method for detecting an analyte.

Background of the art In order to achieve accurate disease diagnosis, it is preferable to entail studies into the expression patterns of bio-molecules secreted in human samples. As such, it is desirable to develop a rapid, reliable, quantitative, and low-cost platform for the detection of biomarkers such as genes and proteins. Disease diagnosis based on visual inspection alone is generally inadequate as early symptoms may appear innocuous and therefore overlooked during checkup (Micheal et al, Clin. Cane. Res. 2001, 7, (3), 607) ¹ . Studies of protein expression patterns on the other hand can provide a much more accurate diagnosis on the basis that the onset of a disease development is normally preceded with abnormal alterations in the expression patterns of the circulating proteins. For instance, of the 300 prostate-specific genes identified, approximately 60 possess signal peptides that are thought to be secreted, and detecting a plurality of these proteins could greatly enhance diagnosis accuracy, as opposed to recognising only a single peptide, which can detect only 50% of the early and late prostate cancers. ² In addition, monitoring the expression patterns of a variety of proteins at various stages of a disease could not only assist prognosis, but also allow one to follow disease progression. It is therefore desirable to have a platform that could offer parallel detection of multiple analytes, which in certain cases may not necessarily be peptide-based molecules, in a rapid, sensitive, and quantitative manner (Hood et al, Sci. 2004, 306, (640-3)) ² . Preferably, the platform should also be amendable to miniaturisation, and be affordable.

RAMAN SPECTROSCOPY

Two optical processes can occur when a monochromatic light interacts with a molecule. In the first, and the dominant process, a large portion of the incident light is elastically scattered with no photon energy being absorbed - this is known as the Rayleigh scattering. A secondary optical phenomenon relates to a process in which a small amount of the incident photon energy is absorbed and results in a transition between vibrational states within the probed molecule. This process unfortunately is relatively weak and would generally lead to a subsequent "re-emission" of a photon quantum whose frequency is "shifted" from that of the incidence iight. This optical effect is conventionally known as Raman Scattering or Raman Spectroscopy (RS), and was first observed experimentally by Raman and Krishnan in 1928 (Raman et al, Nat. 1928, 121 , 501 ) ⁴ . Both the Raman scattering as well as the Rayleigh scattering are depicted in Fig. 1 .

SURFACE ENHANCED RAMAN SPECTROSCOPY

While RS is able to provide high chemical specificities, such a feature is unfortunately offset by the weak Raman scatterings in most molecules. It is estimated that only about one-millionth of the incident photons is "Raman- scattered" by a given molecule. Such a shortcoming can be overcome by field enhancements occurring on optically-excited nano-structured metallic surfaces in a phenomenon conventionally known as Surface Enhanced Raman Spectroscopy or SERS.

SERS was first discovered by Fleischman et al in 1974 when they observed remarkably strong Raman signals for pyridine adsorbed on an electrochemically roughened silver electrode. (Reischman et al, Chem. Phys. Lett. 1974, 26, 123) ⁷ However, the mechanism responsible for such an observation was not understood, and was initially attributed to the increased surface area, until 1977, where it was recognised, by two separate groups, Jeanmarie and Van Duyne, Albercht and Creighton, that the observed enhancement could be attributed both to surface plasmon resonance and to chemical effects on the metallic surfaces. (Jeanmaire et al, J. Electroanal. Chem. 1977, 84, (1 ), 1 -20) ^5, (Albrecht et al, J. Am. Chem. Soc. 1977, 99, 5215) ⁶ Ever since, enhanced Raman scattering has been reported for other compounds that have been brought into close proximity with a metallic surface (typically Silver, Gold or Copper) with nano-scale features.

Generally, it is widely accepted that two mechanisms are responsible for enhancements in Raman scatterings (which can be as high as 10 ¹⁴ times the unenhanced signal): (Kneipp et al, Chem. Rev. 1999, 99, (10), 2957-76) ⁸ Electromagnetic enhancement (EM) and Chemical enhancement (CM).

Electromagnetic enhancement

The majority of the SERS effects can be accounted for by electromagnetic enhancements arising from interactions between the adsorbate analyte and the surface plasmon (SP) fields produced on the metallic nano-structured surface by the excitation laser beam. (Moskovits, J. Raman Spectro. 2005, 36, (6-7), 485-96) ⁹ SP is the collective oscillation of charges bound to the metal-dielectric interface brought about as a result of coupling between the light-fields and the surface charges. On a roughened metallic surface, such a coupling become extremely efficient, and the resultant oscillating SP fields generated are thus concomitantly amplified. As a consequence, this leads to the adsorbed analyte molecule experiencing a strong laser excitation, and in turn results in a large Raman scattering, i.e. the SERS effect. Chemical enhancement

Chemical enhancement contributes only an order of 10- 0 ² to the overall enhancement, (Liang et al, J. Raman Spectro. 1996, 27, (12), 879-85) ⁰ and is currently not fully understood. Nonetheless, the widely accepted mechanisms are that charge-transfers between the analyte and metallic surface form an analyte-surface species capable of coupling, resonantly, with the excitation light, thereby leading to amplified Raman scatterings. The strength of chemical enhancement is generally affected by the surface potential. PREVIOUS ART IN BIO-SENSING

Immunoassay of protein expressions has long been the method of choice for bio-markers detection in a clinical as well as research setting. (Lemarchand et al, Experientia. 1965, 21 , (6), 353-6. ) ^{1 1} Immunoassay normally involves a solid platform, made of either polymer or glass, bearing several immobilised antibodies spotted on well-defined locations. Samples of interest are then applied onto these antibody-functionalised spots to initiate specific bindings between the targeted antigens and their respective capturing antibodies. Upon completion of the antigen-antibody binding process, the sample is thoroughly washed off the immunoassay platform before secondary antibodies are added to bind to the already-captured protein antigens. A tertiary antibody carrying a label is subsequently applied to facilitate read-out. Two forms of labels are generally used: An enzyme that reacts with a specific chemical substrate and converts it into a coloured substance to provide for a colourmetric read-out and a fluorophore, in which case, a fluoro-spectrometer is required to detect the fluorescent signals.

Disadvantages in previous protein sensing techniques

The time-consuming analysis procedures (i.e. multiple washing steps and long incubation time) required in a conventional immunoassay have rendered its application in screening an extensive panel of proteins/biomarkers virtually impossible. Additionally, high-operating costs and significant material wastages owing to the need for multiple antibodies (i.e. primary, secondary and tertiary antibodies) for detecting each targeted protein antigen also prohibit its use for such purposes. Furthermore, colourmetric read-out used in some immunoassay protocol is susceptible to interference by background signals, rendering it impossible for the immunoassay to reach its maximum possible sensitivity with this read-out approach. (Waggoner et al, Lab. Chip 2007, 7, 238-55) ¹²

While it may be possible to improve detection multiplexicity in a conventional fluorescent immunoassay by incorporating more than one fluorescent labellers, (Nishihira et al, Nucleic. Acids Symp. Ser. 2004, 48, 135-6) ¹³ the broad fluorescence bandwidths (60 - 90 nm) unfortunately limits the maximum number of detectable fluorophores per sensing area to about 3. In other words, the maximum number of proteins detectable for each sensing area in a conventional immunoassay cannot exceed 3.

Summary of the invention

The present invention addresses the problems above, and in particular provides a novel and useful sensing platform and method for detecting and/or measuring an analyte in a sample.

According to a first aspect of the present invention, there is provided a sensing platform for detecting and/or measuring an analyte in a sample, the sensing platform comprising: a SERS-active member; a recognition element configured to bind with the analyte, the binding causing a change in an environment around the recognition element; and a molecular transducer arranged with the recognition element and the SERS-active member such that a SERS spectrum of the molecular transducer changes in response to the change in the environment around the recognition element; wherein the change in the SERS spectrum of the molecular transducer is used for detecting and/or measuring the analyte.

According to another aspect of the present invention, there is provided a method for detecting an analyte in a sample, the method comprising: presenting the sample to a sensing platform according to any aspect of the present invention; measuring the SERS spectrum of the molecular transducer; and determining if the measured SERS spectrum of the molecular transducer is different from a SERS spectrum of the molecular transducer prior to presenting the sample to the sensing platform and if so, determining that the analyte is detected.

According to another aspect of the invention, there is provided a method for measuring an analyte in a sample, the method comprising: measuring the SERS spectrum of the molecular transducer; presenting the sample to a sensing platform according to the invention; and analyzing the measured SERS spectrum of the molecular transducer to determine an analyte concentration. More in particular, the analyzing step may comprises: calculating a ratio of intensities between a first peak and a second peak in the measured SERS spectrum of the molecular transducer; comparing the ratio of intensities to a ratio of intensities between the first peak and the second peak in a SERS spectrum of the molecular transducer prior to presenting the sample to the sensing platform; and determining the analyte concentration. The analyzing step may also comprises : using a pattern recognition algorithm capable of quantitative whole-spectrum analysis to analyze differences between the measured SERS spectrum of the molecular transducer and a SERS spectrum of the molecular transducer prior to presenting the sample to the sensing platform; and determining the analyte concentration. The pattern recognition algorithm may be used in combination with at least one statistical learning technique.

According to a further aspect of the present invention, there is provided a sensing kit for detecting and/or measuring at least one analyte in a sample, the sensing kit comprising a plurality of sensing platforms according to any aspect of the present invention.

According to another aspect, the present invention provides a sensing platform for detecting and/or measuring an analyte in a sample, wherein the sensing platform is substantially as described according to the whole content of the present invention.

According to another aspect, the present invention provides a method for detecting and/or measuring an analyte in a sample, wherein the method is substantially as described according to the whole content of the present invention.

According to a further aspect, the present invention provides a sensing kit for detecting and/or measuring at least one analyte in a sample, wherein the sensing kit is substantially as described according to the whole content of the present invention.

Detailed description of the invention

The embodiments of the current invention are primarily concerned with a detection and/or sensing platform that can offer parallel detection and/or measurement of multiple analytes, which in certain cases may not necessarily be peptide-based molecules, in a rapid, sensitive, and quantitative manner, and that is amendable to miniaturisation and affordable. Embodiments of the current invention are related to a novel sensing platform conceptualised based on unique SERS-active molecular transducers that are responsive, in some specific manners, towards bindings of capturing agents to their target analytes. More importantly, the current technology used in the embodiments of the current invention exploits the narrow peak-widths and the photo-stability of SERS spectra in order to minimise signal cross-talks between SERS-active molecular transducers so as to allow for simultaneous measurements of multiple binding events within a single laser spot. The sensing scheme in the embodiments of the current invention is thus capable of a highly- multiplexed label-free detection of biologically and chemically related agents.

The embodiments of the current invention deal with detecting and/or measuring binding events between a capturing-agent and its targeted analyte partner via SERS-active molecular transducers. In particular, the embodiments of the current invention are concerned about the indirect measurements of changes (for example, chemical and/or electric-field changes) in the environment (for example, local micro-environment) of the capturing-agents occurring during a binding event by means of SERS measurements.

An objective of the embodiments of the current invention is to eliminate the need for multiple antibodies per targeted antigen, hence minimising material wastage. The embodiments of the current invention also seek to reduce the number of washing steps, and to allow for a label-free read-out scheme, hence rapid detection. The embodiments of the current invention also aim to provide a solution to the limited detection multiplexity in a conventional immunoassay.

Raman frequency shifts of a molecule are closely related to the vibrational modes - here, vibrational modes refer to the "manner" in which the molecule vibrates, which in turn is usually dependent upon both the molecular structure of the molecule as well as the chemical nature of the molecule's immediate surroundings. (Jeanmaire et al, J. Electroanal. Chem. 1977, 84, (1 ), 1-20) ⁵ , (Albrecht et al, J. Am. Chem. Soc. 1977, 99, 5215) ⁶ The embodiments of the current invention involve the use of such a sensing property of a Raman-active molecule for detecting binding events between a capturing-agent and its targeted analyte partner. A Raman active molecule is a molecule that is capable of giving a Raman signal.

SERS-based Molecular Transducers for Protein Detection

Fig. 2 depicts a sensing platform for detecting and/or sensing an analyte in a sample according to one possible embodiment of the current invention. The configuration includes a recognition element (RE) in the form of a primary antibody 202 covalently attached to a molecular transducer in the form of a SERS-active molecular transducer 204 (SERS-MT), which itself is covalently immobilised to a SERS-active member in the form of a SERS-active metallic nano-structured surface 206. Although single bonds are shown as chemical joints between elements, it is also possible to use multiple covalent bonds to link two neighbouring elements. It should be understood that the primary antibody 202 depicted in the figure, and hereafter, is to be taken as an example. Other types of REs that exhibit binding specificities such as aptamer, protein, DNA, RNA and a synthetic chemical are equally applicable.

As shown in Fig. 3, an analyte in the form of an antigen 302 is presented to the sensing platform. If this antigen 302 is the target analyte partner of the antibody 202, it binds with the antibody 202. This causes changes in the environment (usually the immediate local micro-environment) around the antibody 202.

Sensing is accomplished through the SERS-MT 204, which is placed in close proximity to the antibody 202. The SERS-MT 204 is arranged with the antibody 202 such that the environment around it changes together with the environment around the antibody 202. Furthermore, the SERS-MT 204 is configured such that its SERS spectrum changes in response to changes in the environment around it. Therefore, when the antigen 302 binds with the antibody 202, the SERS spectrum of the SERS-MT 204 changes and the antigen 302 can hence be detected. Note that the change in the SERS spectrum of the SERS-MT 204 can be detected using any conventional SERS technology as will be known to a person skilled in the art (for example, see Figs. 1 (a) and (b) which respectively illustrates the Stokes Raman Scattering and the Anti-Stokes Raman Scattering). In one example, the SERS-MT 204 acts as an optical beacon that senses changes (for example, chemical and/or electric-field changes) in the environment (for example, immediate local micro-environment) of the antibody 202 upon binding to the targeted antigen 302 as illustrated in Fig. 3. The chemical changes include, but are not limited to, pH changes, and/or changes in the ionic-strengths. In one example, a potentiometric molecule (i.e. a molecule which changes its spectral properties in response to electric- field/voltage changes) is used as the SERS-MT 204 and changes in the local electric-field strengths surrounding the antibody 202 upon binding to the targeted antigen 302 are measured. Sensing is then achieved by means of optically detecting changes in the SERS spectrum of the SERS-MT 204 as a result of changes in its local micro-environment. Such changes could, for example, be a consequence of environmentally-induced charge re-distributions within the SERS-MT 204, and will be reflected as peak-shifts (i.e. changes in locations of peaks in the SERS spectrum) and/or changes in peak-to-peak ratios in the SERS spectrum.

Please note that the orientation and position of the bound antigen 302 in Fig. 3 is for illustration purposes only. The actual orientation and position of the bound antigen 302 can vary. Furthermore, the analyte need not be in the form of an antigen. Instead, the analyte may also be in the form of any one of the following: chemical agent (e.g. chloride, magnesium), micro-organism (e.g. bacterium, virus), protein, gene.

In general, a "molecular transducer" is a molecular sensor that is capable of detecting a parameter or information in one form and reporting it in another. In other words, it is capable of transforming one type of information or data, to another (for example, transforming pH changes/ion strength changes to readable data such as SERS spectral changes). This differs from a typical "reporter" which usually merely reports changes (such as reporters used in ELISA which merely release signals in the form of colours or fluorescence to indicate the presence of a particular reaction or compound).

The molecular transducer may be in the form of a molecule whose vibrational mode (i.e. manner in which the molecular transducer vibrates) changes upon detecting changes in the environment around the recognition element. In particular, the molecular transducer may comprise one or more of the following: a highly Raman active molecule comprising aromatic ring(s), a molecule comprising a carboxyl group, a molecule with a thiol group, a molecule with a thiol group without an amino group, a molecule with both a thiol group and an amino group. More specifically, the molecular transducer may comprise one or more of the following: Cysteamine hydrochloride, Crystal violet, Malachite Green, Basic fuchsin (BF), para-mercaptobenzoic acid (pMBA) (Michota et al, J. Raman Spectrosc. 2003, 34, 21 - 25.) ¹⁴ · (Lee et al, J. Raman Spectrosc. 1991 , 22, 81 1 - 817) ¹⁵ , sulfanilic acid sodium salt (Yohannan Panicker et al, J. Raman Spectrosc. 2010, 41 , 944 - 951 ) ¹⁶ , metoclopramide (Leopold et al, J. Raman Spectrosc. 2010, 4 , 248 - 255) ¹⁷ . These studies ^{14 " 7} have shown that pMBA, sulfanilic acid sodium salt and metoclopramide are pH sensitive (i.e. they respond to pH changes). More details regarding these molecules can be found in references 14 - 17:

(Michota et al, J. Raman Spectrosc. 2003, 34, 21 - 25. ) ¹⁴ ' (Lee et a!, J. Raman Spectrosc. 1991 , 22, 81 1 - 817) ¹⁵ ,

(Yohannan Panicker et al, J. Raman Spectrosc. 2010, 41 , 944 - 951) ¹⁶ , and (Leopold et al, J. Raman Spectrosc. 2010, 4 , 248 - 255) ¹⁷

, the contents of which are herein incorporated by reference.

In the configuration shown in Fig. 2, the SERS-MT 204 is understood to be a highly Raman active molecule, whose chemical structure possesses aromatic ring(s). Of course, molecules containing no aromatic ring but are highly Raman active are equally applicable.

The embodiments of the current invention exploit the vibrational characteristics of a SERS-MT 204 as a means to optically probe antigen-antibody induced chemical changes in the local micro-environment. More specifically, it is expected that such changes could result in alterations in the vibrational states of the SERS-MT 204 which can, in turn, be reflected as changes in its SERS spectra. An example can be given of the carbonyl stretching vibration mode (C=0) in a carboxyl group, whose vibrational frequency is known to be pH- dependent. A placement of such a chemical group in the vicinity of the antibody 202's binding sites could thus provide a means for indirect binding readout based on the local pH changes.

To quantify the response of the SERS-MT 204 to antigen-antibody induced chemical changes in the local micro-environment, the measured SERS spectra of the SERS-MT 204 after presenting the sample to the sensing platform may be analyzed and compared to the SERS spectra of the SERS-MT 204 prior to presenting the sample to the sensing platform.

For example, a ratio of intensities between a first peak and a second peak in the measured SERS spectrum of the SERS-MT 204 may be calculated based on the following formula, ! _α - 0α· ^{+ Ι} α"> ^/2

R =

wherein l _x denotes the Raman intensity at a spectral position, x cm ^"1 and the first and second peaks are at spectral positions a cm ^"1 and b cm ^"1 respectively. In this example, l _a is adjusted by subtracting the background level of Raman radiation represented by the terms (Ι ₃ · + l _a "V2, wherein the Raman intensity at positions a' cm ^"1 and a" cm ^"1 are close to the background intensity, and the SERS spectrum between these positions contains one peak at a cm ^"1 . Similarly, l in this example is adjusted by subtracting the background level of Raman radiation represented by the term (l _b - + l _b ")/2. The ratio of SERS spectrum intensities between the first and second peaks may be obtained for different concentrations of antigen, and this data set may be used to derive a quantitative response of the SERS-MT 204 to antigen-antibody induced chemical changes in the local micro-environment.

In another example, a pattern recognition algorithm capable of quantitative whole-spectrum analysis may be used to analyze differences between the measured SERS spectrum of the molecular transducer and a SERS spectrum of the molecular transducer prior to presenting the sample to the sensing platform. In a similar fashion to the previous example, this analysis may be repeated for different antigen concentrations and the resultant data set used to derive a quantitative response of the SERS-MT 204 to antigen-antibody induced chemical changes in the local micro-environment. Further, the pattern recognition algorithm may be used in combination with at least one statistical learning technique. It should be understood that the upright orientation of the primary antibody 202 (i.e. the recognition element) assumed in Fig. 2 must only be taken as an example. Other possible orientations can also be used. Fig. 4 illustrates three additional possible positioning of the SERS-MTs 204A, 204B, 204C around the antibody 202A, 202B, 202C. In Fig. 4(A), SERS-MTs 204A are depicted to be placed in the vicinity of, but not attached to, the antibody 202A. In Fig. 4(B), sthe antibody 202B is shown to be covalently immobilised to a SERS-active nano-structured surface 206B via a weakly Raman active linker molecule. SERS-MT 204B, in this particular case, is attached to the linker molecule. In Fig. 4(C), the SERS-MT 204C is covalently attached to the immobilised antibody 202C.

An alternative embodiment of the current invention is shown in Fig. 5, in which a SERS active member in the form of a SERS-active metallic nano-particle (NP) 502 functionalised with SERS-MT/antibody complexes 504 is depicted. Such a SERS-MT/antibody-NP construct would be suitable for intracellular studies.

Multiplexed Protein Detection based on SERS-active Molecular Transducers

Another possible embodiment is illustrated in Fig. 6, whereby the sensing area is partitioned into sub-regions (sub-regions 1 , 2, 3), each of which bears a unique SERS-MT/antibody pair 602, 604, 606. If the sensing area is made to be comparable in size to the laser-spot, simultaneous detection of multiple target antigens becomes possible as depicted in Fig. 6(B). This is made possible by the fact that SERS peaks generated from different SERS-MT molecules are generally narrow and are not significantly overlapped spectrally, thus enabling post-computational separations. This particular embodiment therefore offers the possibility of a highly-multiplexed protein-sensing platform. The embodiment in Fig. 6 may serve as a sensing kit comprising a plurality of sensing platforms whereby each sensing platform comprises a SERS-MT/antibody pair 602, 604, 606. Each SERS-MT/antibody pair 602, 604, 606 may comprise a different molecular transducer arranged with a respective different recognition element, the same molecular transducer arranged with a respective different recognition element or the same recognition element arranged with a respective different molecular transducer. The SERS-MT/antibody pairs 602, 604, 606 may also comprise the same molecular transducer arranged with the same recognition element for increased sensitivity. As mentioned above, the sensing platforms may be arranged such that a single laser spot is sufficient for detecting the binding between all the recognition elements and their respective analytes.

Aside from a simple ratio of peak intensities, changes in the spectral profile or pattern at different antigen concentrations can be used to improve the sensor's resolution and sensitivity. One can view each of the measured spectra as a vector in an N-dimensional space (called the manifold), with each axis in this abstract space represents Raman frequency. A pattern recognition (or classification) algorithm may be used to group the "closest" vectors, in a training-data set, corresponding to the same antigen concentration into one representative vector. The resultant set of representative vectors is then modelled with a piecewise linear curve that minimizes the mean-squared error over this collection of vectors. This antigen-Raman manifold model will then be used to produce the estimated antigen concentration from any given SERS spectrum. Example

The following experimental data serves to illustrate the feasibility of the invention, It should be noted that the molecular transducer system used in the current experimental study simply serves as an example. Any other molecular transducer system possessing the necessary Raman effects for bio-sensing as described above may also be used. Fabrication of SERS-active bio-sensor based on pH-sensitive 4- mercaptobenzoic acid (4-MBA)

The current experimental study demonstrates the use of 4-mercaptobenzoic acid (4-MBA) as a pH-sensitive SERS-active transducer for bio-sensing. The SERS-active bio-sensor system described here consists of a SERS-active Au/Ag bi-metallic nano-structure, 4-MBA as the optical transducer, Anti-PSA (Prostate Specific Antigen) antibody, and Cysteamine (Cyst) as the anchor point for the antibody attachment. This is schematically shown in Fig. 7.

Functionaiization of SERS-active Au/Ag bi-metallic nano-structure

The Au/Ag bi-metallic nano-structure was cleaned thoroughly with copious amount of ethanol, and dried with Argon gas. The cleaned substrate was then submerged in a solution containing 5 mM 4-MBA and 20 mM Cyst, and let to react for 1 hr. Upon completion of the reaction, the substrate was washed first with copious amount of ethanol to remove unbound 4-MBA and Cyst, followed by washing with PBS. For antibody conjugation, Anti-PSA solution was prepared by diluting the stock Anti-PSA in PBS. EDC/NHS (171 mM: 427.5 mM) solution is then prepared in PBS. 5 μΙ of the EDC/NHS solution was then added to the Anti-PSA solution and let to react for 5 min. The final concentrations of the EDC, NHS and Anti-PSA in this solution are, respectively, 34.2 mM, 85.5 mM, and 276.8 nM. This activated the carboxyl-terminal at the crystal-segment of the Anti-PSA antibody. The 4-MBA/Cyst coated substrate was then incubated in the EDC/NHS/Anti-PSA solution for 2 hrs at room temperature. This resulted in the activated carboxyl-terminal of the Anti-PSA to react with the amino-group of the Cysteamine, causing the antibody to covalently attach to the Cysteamine - in an "up-right" orientation, which is crucial in maximizing the overall binding capacity of the sensor. Upon completion of the reaction, the substrate was washed thoroughly with PBS for 4 min, to remove excess of EDC, NHS, and unbound antibody. This concludes the fabrication procedure. pH response of 4-MBA

To verify the sensitivity of 4-MBA to pH changes, we subjected 4-MBA/Cyst coated substrates to phosphate buffers of different pH values, and measure their corresponding SERS spectra. Typical spectra obtained at pH = 3.18, 7.0, and 1 1.0, are shown in Fig. 8.

To quantify the response of the 4-MBA/Cyst layer to pH changes, we calculated the ratio value between the peak intensity at 1080 cm ^"1 to that at 1 180 cm ^"1 based on the following formula,

where l _x denotes the Raman intensity at the spectral position, x cm ^" .

The value of l _x as a function of the pH is shown in Fig. 9.

In the subsequent experiment, SERS response of our sensor will be quantified the above peak-intensity ratio, R. Of course, other methods including neural- network based pattern recognition algorithm can also be used to quantify SERS response.

Response of the 4-MBA-based sensor with and without Antibody- conjugation

A subsequent study verified that the observed SERS responses from the Anti- PSA/4-MBA sensor were indeed due to the binding of antigen (PSA) to the immobilized Anti-PSA antibody. To achieve this, we measured the SERS responses of the 4-MBA/Cyst layer upon exposure to PSA (prepared in PBS), and compare with those of the Anti-PSA conjugated 4-MBA/Cyst layer (hereby denoted as Anti-PSA 4-MBA). Typical SERS spectra derived from these experiments are shown in Figs. 10a and 10b. Fig. 10a shows changes in the SERS spectra of the unconjugated 4-MBA/Cyst layer before and after the application of PSA, and Fig. 10b shows SERS responses of Anti-PSA conjugated 4-MBA/Cyst layer. Fig. 1 1 is a bar chart showing the corresponding responses of the 4-MBA/Cyst layer and Anti-PSA/4-MBA in terms of the intensity ratio (lio80cm-i ln80cm-i )- As can be seen, maximum response occurs only when 4-MBA/Cyst is conjugated with Anti-PSA, indicating that antigen- antibody binding events could be sensed via a pH-sensitive transducer. Specificity of the 4-MBA-based bio-sensor in PBS

This particular study evaluates the specificity of the Anti-PSA/4-MBA sensor toward its targeted antigen, i.e. PSA. To achieve this, we performed two sets of sensing experiments with one set of which involving the Anti-PSA/4-MBA sensors being exposed to PSA prepared in PBS, while the other set to an equal molar of BSA (Bovine Serum Albumin) also prepared in PBS. In each experiment, a reference SERS spectrum was first measured with the anti- PSA 4-MBA sensor submerged in PBS, i.e. without PSA or BSA, after which the sensor was partially dried before the application of the analyte of interest. Here, BSA was used as the control analyte. All the anaJytes (PSA and BSA) were made to react with the sensing surface for 30 min at room temperature before Raman readings. Typical SERS spectra derived from these two experiments are shown in Figs. 12a and 12b. Fig. 12a shows changes in the SERS spectra of Anti-PSA 4-MBA before and after the application of PSA, and Fig. 12b shows SERS response to BSA analyte. Fig. 13 is a bar chart showing the responses of Anti-PSA/4-MBA to the above analytes in terms of the intensity ratio (11080cm- 1/I1 180cm-1 ). As can be seen, maximum response occurs only when the sensor is exposed to PSA, i.e. the sensor is specific toward PSA.

Specificity of the 4-MBA-based bio-sensor in human blood plasma This particular study evaluates the specificity of the Anti-PSA/4- BA sensor toward its targeted antigen, i.e. PSA, in human blood plasma (HBP). To achieve this, we repeated the above sensing experiments in HBP. We again performed two sets of sensing experiments with one set of which involving the Anti-PSA/4- MBA sensors being exposed to PSA-spiked HBP, while the other set to PBS- spiked HBP. PSA-spiked HBP was prepared by adding 2 pi of PSA in 18 μΙ HBP. PBS-spiked HBP was prepared by adding 2 μΙ of PBS in 18 μΙ HBP. In each experiment, a reference SERS spectrum was first measured with the anti- PSA/4-MBA sensor submerged in HBP, i.e. without PSA or BSA added, after which the sensor was partially dried before the application of the analyte-spiked blood plasma. Here, the PBS-spiked HBP was used as a control sample. All the spiked plasma samples were made to react with the sensing surface for 30 min at room temperature before Raman readings. Typical SERS spectra derived from these two experiments are shown in Figs. 14a and 14b. Fig. 14a shows changes in the SERS spectra of Anti-PSA/4-MBA before and after the application of PSA-spiked plasma, and Fig. 14b shows SERS response to PBS- spiked plasma. Fig. 15 is a bar-chart showing the responses of Anti-PSA/4-MBA to the above plasma samples in terms of the changes in the intensity ratio (Ii080cm-i/li i80cm-i ) before and after the application of spiked plasma samples. As can be seen, PSA-spiked HBP induces a positive response, while PBS-spiked HBP a negative response. This study demonstrates the potential of the current sensor for direct human blood plasma analysis.

OTHER APPLICATIONS

Although a large emphasis has been put on the detection and measurement of proteins/biomarkers in a biological context, the embodiments of the current invention can also be used for detecting and measuring chemical agents (such as chloride and magnesium) as well as micro-organisms (such as bacteria and viruses).

It is anticipated that the embodiments of the present invention could not only greatly assist in facilitating early treatment but also reduce healthcare costs. ³ Furthermore, embodiments of the present invention are capable of removing the need for secondary and tertiary antibodies. It is thus less time-consuming as less washing steps and incubation time is required! Furthermore, the embodiments exploit the narrow peak-widths and photo-stability of SERS spectra and are less susceptible to interference by background signals. They are also capable of detecting multiple analytes simultaneously as each analyte is usually associated with a unique SERS spectrum and hence a unique peak in the combined SERS spectrum.

REFERENCE

1 . Micheal FS; Wayne MK; Andre LR; Joseph AC; Li HX; Claus FE; Linwah Y; Paul LL; Li W; Shixi XL; Carman J; Wiliam HW; S, D. Clin. Cane. Res. 2001 , 7, (3), 607.

2. Hood, L; Heath, J. R.; Phelps, M. E.; Lin, E. Sci. 2004, 306, (640-3).

3. Levenson, V. V. Biochimica et Biophysic Acta 2007, 1770, 847-56.

4. Raman, C. V.; Krishnan, K. S. Nat. 1928, 121 , 501 .

5. Jeanmaire, D. L.; Van Dyune, R. P. J. Electroanal. Chem. 1977, 84, (1 ), 1 -20.

6. Albrecht, M. G.; Creighton, J. A. J. Am. Chem. Soc. 1977, 99, 5215.

7. Reischman, M.; Hendra, P. J.; McQuilan, A. J. Chem. Phys. Lett. 1974, 26, 123.

8. Kneipp, K.; Kneipp, H.; Itzkan, I. ; Dasari, R. R.; S. , F. M. Chem. Rev. 1999, 99, (10), 2957-76.

9. Moskovits, M. J. Raman Spectro. 2005, 36, (6-7), 485-96.

10. Liang, E. J.; Kiefer, W. J. Raman Spectro. 1996, 27, (12), 879-85.

. Lemarchand, B. T.; Vannotti, A. Experientia. 1965, 21 , (6), 353-6.

12. Waggoner, P. S.; Craighead, H. G. Lab. Chip 2007, 7, 1238-55.

13. Nishihira, A.; Ozaki, H.; Wakabayashi, M.; Kuwahara, M.; Sawai, H. Nucleic. Acids Symp. Ser. 2004, 48, 135-6.

14. Michota, A., Bukowska, J. J. Raman Spectrosc. 2003, 34, 21 - 25.

15. Lee, S.B., Kim, K. Kim, M.S. J. Raman Spectrosc. 1991 , 22, 8 1 - 817.

16. Yohannan Panicker, C , et al. J. Raman Spectrosc. 2010, 41 , 944 - 951 17. Leopold N. et al. J. Raman Spectrosc. 2010, 41 , 248 - 255