Patent Application
20110259744
1. Field of the Invention
The present invention provides methods for detecting analytes such as proteins, peptides, nucleic acids, ligands, antigens, lipids, enzymes, and other molecules in simple and complex systems. the present invention provides hydrogels for separating bound and free analytes and their optical analysis. More specifically, the invention provides new methods for rapid clinical uses such as for pap smears, diagnosis of sexually transmitted diseases, diagnoses of skin cancers, diagnosis of oral cancers and monitoring lymphocytes.
2. Description of the Background
The disclosures referred to herein to illustrate the background of the invention and to provide additional detail with respect to its practice are incorporated herein by reference and, for convenience, are numerically referenced in the following text and respectively grouped in the appended bibliography.
A device that can be used to monitor gene expression rapidly in single cells would have several important applications. For example, surgeons often rely on histological methods to distinguish tumor and normal tissues during surgery to remove cancers. These methods serve well when the morphology of the abnormal and normal cells is readily distinguished. Unfortunately, the borders of many tumors are not always well defined and do not provide clear landmarks that can be used to guide surgery. Further, it may be difficult to gauge the characteristics of the tumor even after sections have been stained with histological dyes. This can lead to unnecessary surgery during efforts to remove all the cancerous tissue. Indeed, some surgery for breast cancer involves removing lymph nodes to stage the cancer even though there often is no evidence that this additional surgery will be of significant benefit. Application of a technique that has the ability to monitor gene expression in these frozen sections would have considerable application during surgery to guide the procedure. It would also be useful to guide the type of therapy that is to be used following surgery.
Recent advances in genetics have provided the basis by which physicians and scientists have gained new insights into cell function. Bioinformatic analysis suggests that humans have 30-40 thousand genes [1;2] that are transcribed, spliced, and edited to yield 100 thousand mRNAs detected as expressed sequence tags [3]. This information has permitted the design of microarrays capable of monitoring thousands of gene products at one time [4;5]. Microarray technology is being applied widely to characterize changes in gene expression patterns that are associated with various tumors and with the prognosis of tumor therapy [5-7]. Indeed, there is considerable hope that the results of these studies will enable a more accurate classification of tumors and thereby guide the choice of therapy following surgery. One benefit of this may be a reduction in unnecessary chemotherapy or radiotherapy [5], procedures that often make patients ill and that may even be a source of malignancies later in life [8].
Further technical advances in measurements of gene expression products are required to take full advantage of the new information that is being made available from microarray measurements. Tumors are often quite complex and contain endothelial cells, fibroblasts, lymphocytes, and other cell types in addition to transformed cells. Microarray analyses of whole tumor tissues detect expression products of these cell types simultaneously [4;5], a phenomenon that confounds the association of particular gene expression patterns with specific tumor cells. These analyses can be further compromised by the presence of different types of tumor cells within the sample. Nonetheless, despite this complexity, gene expression patterns detected in some tumors are correlated highly with five-year survival rates [5] and this information can be used to facilitate tumor classification, the major parameter used to decide how patients are treated.
The massive amount of data obtained during microarray analysis is extremely valuable but it is confounded by the presence of gene products that have been obtained from multiple cell types. It can also be time-consuming to obtain and, because it contains so much information, can be difficult to interpret accurately. Results of array analyses indicate that it not necessary to monitor the expression of all possible genes to classify the tumor accurately [5;9]. In fact, as exemplified by findings made from studies of colon carcinomas, a majority of which have a preponderance of mutations of the APC and p53 genes [10], it appears that analysis of relatively few gene products would be adequate to classify tumors. The types of genes to be monitored can be determined by taking advantage of information that is usually known at the time of surgery, such as the location of the tumor (i.e., mammary gland, prostate, colon, lung, brain, etc.). The technology described here permits one to measure the expression of several gene products in single cells of frozen sections that are routinely prepared during surgical procedures. By focusing on genes whose expression has been found in microarray and other analyses to be most characteristic of a given tumor type, it will be possible to classify the tumor accurately. The devices taught here permit this information to be determined in a rapid fashion and can be used to form the basis of instant decisions needed for patient care.
The cells in a cancer have altered properties that enable them to evade apoptotic mechanisms that normally limit cell growth. Some of these include checks on the integrity of their genome and, when these are lost or become non-functional, cancer cells tend to accumulate mutations that make them more aggressive. Since not all the cells of a tumor have the same mutations, the tumor can be heterogeneous. The heterogeneity of some tumors may even be due to the fact that they have originated from several different cells, not just a single cell. Thus, to classify the tumor accurately, it is best to assess gene products from individual cells so that the degree of heterogeneity can be ascertained. It is also important to detect the existence and location of even a small number of cells that have reduced sensitivity to natural regulatory mechanisms. The ability to do so would enable pathologists and surgeons to learn if the tumor contains cells that have characteristics indicative of a more advanced stage of cancer as well as to learn where they are within the tumor. If this information were available at the time of surgery, it would enable the surgeon to tailor the surgical procedure appropriately for each patient. For example, the absence of these cells might indicate that it would not be essential to remove nearby or distant lymph nodes that are not part of the tumor. In contrast, the presence of a few advanced cells in a small otherwise unremarkable tumor might be grounds for more extensive surgery. Thus, it would be desirable to have a sensor that could quantify gene expression rapidly in single cells of frozen sections obtained at the time of surgery. Furthermore, this information should also affect the choice of post-surgical treatment such as chemotherapy and/or radiation therapy.
The therapeutic benefits of identifying cells that have altered genotypes and/or phenotypes that lead to pathological states have been recognized for many years. The need to classify these cells has led to developments of several methods for examining cells that range from simple staining procedures to highly refined approaches for identifying specific genes and gene products within the cell. Increased knowledge of cell function offers a greatly expanded number of markers that can be used to assess the pathological status of single cells.
Several methods have been developed to study gene function in individual cells. Fluorescence Activated Cell Sorting (FACS) methods have permitted individual cells to be isolated from complex cellular mixtures based on the use of antibodies to a single surface protein. This method requires disrupting tissues into their component cells, which is a time-consuming process that makes FACS analysis poorly suited for use as a routine surgical procedure. Techniques such as Fluorescent in situ Hybridization (FISH) are sufficient to detect single genes within cells of a tissue. The most sensitive of these techniques require considerable tissue preparation, however, and are not sufficiently rapid for routine use during surgery. Furthermore, the intrinsic fluorescence in cells and other factors often contribute to high background. This makes it essential to perform several time-consuming internal controls without which it would be impossible to interpret the analysis. Other properties of fluorescence, such as the ability of adjacent fluorophores to interact with one another, a process known as Fluorescent Resonant Energy Transfer (FRET), have been used to facilitate analyses of gene expression. For example it has been found that fluorescent oligonucleotides can be used to detect mRNA products of single genes cells based on the abilities of the oligonucleotides to bind to adjacent portions of the mRNA [11]. Nonetheless, these techniques can be plagued by the high intrinsic fluorescence of cells. While it is possible to circumvent this problem using time-resolved methods [12], this increases the complexity of the method substantially at the expense of assay sensitivity. In addition, there is a need to get the fluorophores into the cells where they can interact with the mRNA. Thus, this approach is not practical for routine examination of tissue sections. Efforts have also been made to monitor gene products using fiber optic techniques [13]. These methods are also not applicable to tissue sections and suffer from a very slow response time.
In summary, knowledge of the gene products that are associated with different pathologies is accumulating rapidly. The public availability of the sequence of the human genome and advances in microarray technology have permitted the simultaneous semi-quantitative measurements of large numbers of gene products. Array procedures have been used to characterize changes in gene expression in several types of normal and abnormal tissues. Indeed, comparisons of gene expression patterns in tumor tissues with tumor recurrence and long-term survival of patients following surgery, chemotherapy, and/or radiation have enabled predictions about the types of therapies that are most likely to be beneficial [4]. As noted earlier, array procedures are not readily adapted to analyses of single cells. Consequently, the data generated by application of this technique are confounded by the presence of analytes in non-tumor cells as well as by the fact that many tumors contain different types of abnormal cells. This makes it difficult to associate gene expression with particular cells in even a semi-quantitative fashion. Furthermore, array analysis is time-consuming and not suited for the rapid estimation of gene expression while the patient is in the operating room. Measurements of gene expression in single cells within the tumor would be of considerable value for classifying the tumor, a key component used to make informed decisions about the extent of surgery and subsequent therapies. It would also be applicable during research to learn which gene expression products are most likely to have predictive value. Finally, it would also be useful for studies of cell function during complex processes such as those that occur during development and cellular differentiation.
The present invention provides sensor devices for detecting an analyte in a sample in which an analyte is bound to a detection reagent to form a bound complex, wherein the device comprises:
(a) a sample (5) comprising an ionic analyte and a detection reagent in a conductive fluid, wherein the detection reagent has a net charge different from the analyte;
(b) a first permeable polymeric hydrogel plate (3) and a first spacer plate (8), which plates provide a compartment for the sample;
(c) an anode (1) juxtaposed to the outside of the first hydrogel plate and not in contact with the sample;
(d) a cathode (9) juxtaposed to the outside of the first spacer plate and not in contact with the sample;
(e) a voltage generator (10) to apply an electric potential to the anode and cathode; and
(f) a detector (11);
wherein the bound complex formed from the analyte and detection reagent is detected by the detector (11) because the bound complex has a charge that causes it to migrate in a direction opposite from that of the unbound analyte when the electric potential is applied;
wherein the improvement comprises the first permeable polymeric hydrogel plate (3) and the first spacer plate (8) further contain an analytical surface (12) and a focusing optical frame component (13) that causes light to pass across the analytical surface in a total internal reflection mode.
The present invention also provides methods for detecting an ionic analyte in a sample in which an analyte is bound to a detection reagent to form a bound complex, comprising the steps of:
(a) a sample (5) comprising an ionic analyte and a detection reagent in a conductive fluid, wherein the detection reagent has a net charge different from the analyte;
(b) a first permeable polymeric hydrogel plate (3) and a first spacer plate (8), which plates provide a compartment for the sample;
(c) an anode (1) juxtaposed to the outside of the first hydrogel plate and not in contact with the sample;
(d) a cathode (9) juxtaposed to the outside of the first spacer plate and not in contact with the sample;
(e) a voltage generator (10) to apply an electric potential to the anode and cathode; and
(f) a detector (11); and
The present invention also provides sensor devices for detecting and quantifying a gene product in a cell or tissue section sample by employing an analysis reagent that binds to the gene product to form a detectable product comprising:
(a) a first and second coated plate, wherein the plates are parallel to each other and are coated with a conductive material;
(b) a first and second conductive plate, wherein the plates are parallel to each other and are juxtaposed over the coated plates of (a);
(c) a first conducting tape connecting a first end of the coated plates of (a) and the conductive plates of (b) and a second conducting tape connecting a second end of the coated plates of (a) and the conductive plates of (b);
(d) a first gasket insulator insulating a first end of the coated plates of (a) and the conductive plates of (b) and a second gasket insulator insulating a second end of the coated plates of (a) and the conductive plates of (b);
(e) a voltage generator connected to the first and second conductive plates to apply an electric potential to the conductive plates; and
(f) a detector;
wherein the first and second coated plates provide a compartment for a cell or tissue section sample and a conductive fluid and an analysis reagent is provided in the sample or tethered to a surface of the first or second coated plate such that when the voltage generator applies an electric potential to the conductive plates, the detector will detect the interaction between charged materials within the cell or tissue section sample, migrating towards either surface of the coated plate, and the analysis reagent.
The sensor device may further comprise a heating means to heat the sample prior to, or during, detection of the sample and may further comprise a cooling means to cool the sample prior to, or during, detection of the sample. The detector may be a fluorescence, luminescence, colorimetry, or total internal reflection illumination detector or may detect by phase contrast microscopy, bright field microscopy, darkfield microscopy, differential interference contrast microscopy, confocal microscopy, or epifluorescence microscopy. The electrical potential may be applied perpendicular to the coated plate and may be constant or varied such that the overall effect is to have each plate have a net charge, such that charged analytes in the tissues will migrate to one plate. The electrical potential may also be applied perpendicular to the coated plate and may be alternated such that there is no net charge on either plate, such that charged analytes will oscillate back and forth in the central space away from either plate where they interact with analysis reagents.
The present invention also provides methods for detecting and quantifying a gene product in a cell or tissue section sample by employing an analysis reagent that binds to the gene product to form a detectable product, wherein the analysis reagent is tethered to a surface of a sensor device, comprising the steps of:
(a) a first and second coated plate, wherein the plates are parallel to each other and are coated with a conductive material, and an analysis reagent is tethered to a surface of the first or second coated plate;
(b) a first and second conductive plate, wherein the plates are parallel to each other and are juxtaposed over the coated plates of (a);
(c) a first conducting tape connecting a first end of the coated plates of (a) and the conductive plates of (b) and a second conducting tape connecting a second end of the coated plates of (a) and the conductive plates of (b);
(d) a first gasket insulator insulating a first end of the coated plates of (a) and the conductive plates of (b) and a second gasket insulator insulating a second end of the coated plates of (a) and the conductive plates of (b);
(e) a voltage generator connected to the first and second conductive plates to apply an electric potential to the conductive plates; and (f) a detector; and
The present invention further provides methods for detecting and quantifying a gene product in a cell or tissue section sample by employing an analysis reagent that binds to the gene product to form a detectable product, wherein the analysis reagent is soluble in the sample, comprising the steps of:
(a) a first and second coated plate, wherein the plates are parallel to each other and are coated with a conductive material;
(b) a first and second conductive plate, wherein the plates are parallel to each other and are juxtaposed over the coated plates of (a);
(c) a first conducting tape connecting a first end of the coated plates of (a) and the conductive plates of (b) and a second conducting tape connecting a second end of the coated plates of (a) and the conductive plates of (b);
(d) a first gasket insulator insulating a first end of the coated plates of (a) and the conductive plates of (b) and a second gasket insulator insulating a second end of the coated plates of (a) and the conductive plates of (b);
(e) a voltage generator connected to the first and second conductive plates to apply an electric potential to the conductive plates; and
(f) a detector; and
The detector may be a fluorescence, luminescence, colorimetry, or total internal reflection illumination detector or may detect by phase contrast microscopy, bright field microscopy, darkfield microscopy, differential interference contrast microscopy, confocal microscopy, or epifluorescence microscopy. The electrical potential may be applied perpendicular to the coated plate and may be constant or varied such that the overall effect is to have each plate have a net charge, such that charged analytes in the tissues will migrate to one plate. The electrical potential may also be applied perpendicular to the coated plate and may be alternated such that there is no net charge on either plate, such that charged analytes will oscillate back and forth in the central space away from either plate where they interact with analysis reagents. The gene products may be nucleic acids or proteins. The analysis reagent may be a biotin-streptavidin conjugate or may be a molecular beacon. Preferably, a mixture of molecular beacons labeled with the same fluorophore is employed to detect a mixture of gene products associated with a tumor class. A second molecular beacon may be employed as an internal control. Preferably, a first molecular beacon is employed to detect a control gene product and a second molecular beacon is employed to detect a gene product of experimental or diagnostic interest, wherein the first and second molecular beacons are each labeled with a different fluorophore that emits at a different wavelength so that the first and second molecular beacons can be simultaneously analyzed. The control gene product may be β-actin. The transparent plates may be coated with indium tin oxide or tin dioxide.
The present invention provides a rapid, sensitive, and accurate method that can be used to measure nearly any analyte. In particular, the method can be employed to visualize the relationship between gene expression and tissue morphology. The method utilizes an electrical potential to promote the movement of the analyte from one site to another causing the analyte to be concentrated in the region where the measurement can be made. By controlling the electrical potential it is possible to concentrate materials from tissue samples, electrophoresis gels, or any other media at a sensor surface and thereby enhance the sensitivity and the speed with which measurements can be made. Furthermore, the electrical potential can be used to reduce non-specific interactions that occur during analysis and thereby facilitate measurement accuracy. The electrical potential can also be used to alter the chemistry of the analyte. Thus, it is possible to reduce or oxidize the sample. This will also increase the specificity and accuracy of the device. The electrical potential can also be used to alter the chemistry of the analyte and the sensor surface, and to immobilize sensor molecules at the surface via covalent bonds, coordination or physical adsorption. Analysis occurs by the specific interaction between the material that has migrated towards the surface of the plate and reagents that are attached to the plate or that are held near the plate surface. Because this analysis does not alter the relative positions of cells or other factors that are being analyzed, it permits the identification of analytes that are associated with specific cell types or with specific portions of the material being analyzed. The sample may also be reduced or oxidized to increase the specificity and accuracy of the device. The method permits decisions to be made by physicians and pathologists at the time of the procedure and facilitates analysis by persons less skilled in these tasks, such as technicians who do the preliminary reading of Pap tests and other analyses that are preformed in high volume on a routine basis. The information will also be useful for making decisions regarding treatments after the procedures are completed.
In one embodiment, the present invention can be used to measure gene expression products in tissue sections. These gene products can be nucleic acids, such as messenger and other RNAs, or proteins such as enzymes and transcription factors. The method proposed for use with tissue sections involves placing the tissue sections or cells, including those taken at time of surgery, between two transparent plates or slides that have been coated with a material that conducts electricity or that can be made to conduct electricity. When an electric potential is placed on either side of the tissue, charged materials within the tissue can be made to migrate towards either plate. Those with a net positive charge will migrate towards the cathode and those with a net negative charge will migrate towards the anode. The electrical potential on the transparent plates, which serve as electrodes, can be constant or varied in a variety of fashions. When the potential is constant or when it is varied such that the overall effect is to have each plate have a net charge, charged analytes in the tissues will migrate to one electrode. When the potential is alternated such that there is no net charge on either plate, charged analytes will oscillate back and forth in the central space away from either electrode where they interact with detection reagents.
The method is not limited to tissue sections but can be applied to detect other agents and may not require the use of two slides. Metabolites that are altered as the result of changes in gene expression may also be detected.
In a second embodiment, the sample is recognized by a binding agent in an interaction that occurs in solution and that can take place at either surface of the sensor device, in the vicinity of either surface, or away from either surface. The complex that is formed has a different electrical charge than the binding agent. Application of an electrical potential across the plates results in the migration of the complex towards one of the plates where it can be measured. Since the binding agent and the complex have different charges, it is possible to separate the binding agent from the complex, a phenomenon that can be employed to reduce measurement noise. When operated in this fashion, the device can be used to monitor any interaction that leads to a change in charge. This includes enzyme reactions in which enzymatic activity leads to a change in the charge of the substrate.
The actual measurement will be made when the charged materials reach the surfaces of one or both plates. In most cases, the measurement will depend on a change in fluorescence. There are two basic methods of examining fluorescence. In one method, a fluorophore will be attached to the surface. Migration of the analyte to the surface will cause an increase in fluorescence or a decrease in fluorescence of the bound detection reagent. For example, binding of the analyte to a molecular beacon would increase its fluorescence. While one could take advantage of a decrease in fluorescence caused by quenching, energy transfer, or even destruction of the surface fluorophore (e.g., by proteolysis or nuclease digestion), this would be less sensitive due to the fact that it would have a high background. The second method of detection depends on the ability of the analyte to cause the migration of a fluorophore to the surface. In this case, the fluorophore detection reagent is either uncharged or charged in a way that would cause it to migrate to the side of the device that is not being examined. Binding of the analyte to the fluorophore would change its net charge and cause it to migrate to the surface that is being examined. Since the charge on a fluorophore could also be changed by cutting the fluorophore or modifying it (i.e., adding a phosphate), this procedure would also permit detection of enzymes. This method could readily be used with quantum dots, fluorophores that are nearly indestructible and that are very bright.
Both methods have their advantages. The second method is preferable because it does not require surface labeling (a task that can require difficult chemistry), it enables the use of much higher reagent concentrations of reagents, and it can produce very low background because of the physical separation of the materials that occurs after electrophoresis. Advantages of the first method include the fact that the bound and free analytes are not separated, permitting detection of lower affinity interactions, and it can be used with a larger number of optical techniques. Indeed, since the fluorophore is attached to the surface, there is no need to use optical techniques that limit illumination to the surface.
The sensor device can also be heated and/or cooled to facilitate interactions between the reagents or even amplification of the analyte (i.e., by PCR). Fluorescence on the surface may be monitored using Total Internal Reflection Methods (TIRF), including TIRF microscopy (TIRFM) using methods that are well known in the art. A lens-based method has also been devised for extending these measurements. Another procedure for monitoring surface fluorescence involves the use of two photon methods. In these methods, photons that have insufficient energy to excite the sample individually are directed at the surface at the same time. When the photons reach the surface, the sum of their energies will excite the sample, enabling it to be detected. Another procedure that can be used is the employment of a lens that has a shallow depth of field that can be focused on the surface. Colorimetric methods can be also used, i.e., when the analyte-detection complex reaches the surface, it causes the appearance of a color.
When tissue sections are to be examined, it will be useful to have a method that can be used to scan the tissue sections automatically, freeing the surgeon or pathologist from spending time finding regions of greatest interest. Once these are detected by their fluorescence, they can be examined manually.
As set out above, detection of the interactions between the analytes and reagents may be carried out using fluorescence techniques although other visual methods including colorimetry and luminescence can be applied as well. One of the most useful techniques for detecting nucleic acid gene expression products such as mRNA employs molecular beacons. These can be attached to the surface of the sensor plate using a variety of methods. One of the most convenient involves attaching biotinylated molecular beacons to surfaces that have been coated with streptavidin. In this method, the beacon is synthesized as a biotin derivative by standard methods such as those employed by companies specializing in molecular beacon synthesis including IDT Technologies, Inc., Coralville, Iowa 52241, USA. Attachment of the biotinylated molecular beacon to the surface of the plate can be performed by attaching it to streptavidin that has been attached to the surface of the plate. Attachment of streptavidin to surfaces is well known in the art and can accomplished by reacting it with biotin derivatives that are covalently attached to the plate or by permitting it to interact with bovine serum albumin-biotin conjugates such as those obtained from Sigma Chemical Co., St. Louis, Mo. 63195, USA that have been adsorbed to the plate surface. Introduction of a charge between the plates of the sensor device promotes migration of the mRNA from the tissue to the positively charged surface of the sensor. This can be facilitated by the introduction of small quantities (i.e., 0.1-2%) of non-ionic detergents such as octylglucoside, which disrupt the plasma membranes that surround the cells in the tissue sections. It can also be facilitated by varying the charge on the plate surface in a fashion that prevents the negatively charged nucleic acid from sticking directly to the plate surface. Interaction of the mRNA gene products with the molecular beacon, a process that can be made to be highly specific by design of the molecular beacon using methods that are standard in the art will lead to increased fluorescence. Since this will be immediately above or below the material being analyzed, the amount of fluorescence will be roughly proportional to the amount of nucleic acid within cells or other local portions of the material being tested.
It is not necessary to use fluorescent reagents that are covalently attached to the surface of the sensor for analysis. For example, mRNA can be monitored using peptide nucleic acids (PNA), which are analogs of nucleic acids that have the sugar-phosphate backbone replaced by peptide bonds. PNA have the same binding specificity as nucleic acids and can be designed using the same principles as are well known in the art to construct oligonucleotides that interact with nucleic acids. PNA are superior to nucleic acids for measurement in the sensor, however, because they lack the strong negatively charged phosphate backbone structures characteristic of nucleic acids. Thus, PNA are essentially neutral in physiological buffers and do not have a great propensity to migrate to either surface of the measuring device. When they bind to mRNA or other nucleic acids, the complex becomes negatively charged due to the negatively charged backbones of the part of the complex derived from the nucleic acid. Thus, the complex will migrate towards an anode. If the PNA are made to contain a fluorophore, formation of the complex will cause the fluorophore to migrate towards the anode where it can be readily detected using TIRFM, confocal microscopy, microscopic techniques that employ two or three photons to excite the sample, or by use of an objective that has a very shallow depth of field. If the fluorophore that is attached to the PNA is positively charged, unbound PNA molecules will migrate towards the cathode. Thus, by measuring fluorescence at the anode, it is possible to detect and quantify specific mRNA gene products in samples.
While nearly any procedure capable of detecting fluorescence can be used to detect the material, it is often most useful to perform the technique in an optical microscope. In cases where the background fluorescence that may be present in tissues and tissue sections is found to limit the sensitivity of the technique, one can also apply microscopic techniques such as TIRFM a device that is constructed specifically for this purpose and that is readily adapted to routine use. TIRFM is a very sensitive procedure that permits studies of single molecules and has even been used to investigate the folding of single molecules of RNA [14]. TIRFM takes advantage of a physical characteristic of electromagnetic radiation that occurs when light contacts surfaces that differ in refractive index. In TIRFM a beam of light is passed through a material of high refractive index such that it reaches an interface with a material of lower refractive index. When the angle of irradiation is below a value known as the critical angle, all the light is reflected back into the material of high refractive index. “ . . . Some light can be detected at the surface. This light is available to excite fluorophores that are attached to the surface high refractive index. material and is responsible for the ability of TIRFM to illuminate material that is on or very near the surface of the TIRFM sensor (i.e., the material of high refractive index). As a consequence of the physical principle that underlies TIRFM, the unwanted background light that results from the intrinsic fluorescence of tissue samples that is often a problem for other types of fluorescent microscopy is virtually eliminated. This high signal-to-noise ratio is responsible for the ability of TIRFM to detect and quantify trace amounts of material in the face of an overwhelming amount of non-specific contaminating debris.
Use of TIRFM can also permit use of the sensor for analysis under conditions in which the reagents that are being used to detect the analyte are not necessarily attached to the sensor surface. Thus, when fluorescent PNA are added to the tissue sections that have been treated with agents such as non-ionic detergents that disrupt the integrity of the cell membrane but not the overall architecture of the tissue, they will interact with nucleic acid gene products (i.e., mRNA and other RNA polymerase derived nucleic acids). Application of an electric potential will cause the fluorescent PNA-RNA hybrid complexes to migrate to the sensor surface where they can be detected. Since multiple PNA can be employed and since multiple fluorophores can be employed, this technique permits simultaneous measurement of many different analytes, a significant advantage during studies to identify gene expression products.
Addition of an electrochemical potential to the TIRFM can increase the sensitivity and the speed of analysis further. Application of a thin layer of indium tin oxide (ITO), tin dioxide (SnO2), or several other metals does not affect its ability to be used for TIRFM at near ultraviolet or visible light wavelengths. Application of an electrical potential to the metal coating can be used to enhance the concentration of material at the sensor surface. This can increase the sensitivity of detection as well as the speed with which the measurements can be made. For example, by varying the electrical field on the TIRFM sensor surface, it is possible to facilitate the migration of nucleic acid oligomers to the surface of the sensor where they can hybridize with others that are on the sensor surface. The presence of an electric field can also facilitate the release of mRNA from tissue sections by disrupting the plasma membranes, a process known as electroporation. This will enhance the migration of mRNA towards the anode sensor surface. It will also facilitate interactions of mRNA with other agents such as PNA. When appropriate fluorophores such as molecular beacons are attached to the sensor surface, it is possible to use this principle to selectively measure nearly any gene product in single cells. Since tissue sections are applied directly on the sensor surface during surgery, this procedure results in a rapid and quantitative analysis of gene products within cells and will permit distinguishing the expression patterns cells within the tissue.
Several different types of fluorophores have been incorporated into molecules than can be used for detection and companies such as Molecular Probes, Eugene, Oreg. and Integrated DNA Technologies (IDT), Coralville, Iowa market them. One of the most useful properties of fluorophores is their ability to undergo resonance energy transfer (RET), also known as fluorescent resonance energy transfer (FRET). RET between adjacent fluorophores occurs when the adsorption spectrum of one overlaps the fluorescence spectrum of the other. According to principles first established by Förster [15], the amount of RET between two fluorophores varies as the inverse of the distance between them to the sixth power. Thus, RET will be nearly quantitative when the fluorophores are adjacent and virtually undetectable when the fluorophores are separated by as little as 100 Å and, in many cases, even less. During RET, energy from the fluorophore that adsorbs light at shorter wavelengths is transferred to that of the fluorophore whose adsorption spectrum overlaps the emission spectrum of the first fluorophore. This leads to a reduction in the amount of light emitted from the first fluorophore and an increase in the amount of light emitted from the second fluorophore. The reduction of light emitted by the first fluorophore can be used to estimate the distance between the fluorophores. It can also be used to assess the formation of a complex between two molecules that are labeled with fluorophores that are capable of undergoing RET. RET between two fluorophores usually leads to a change in the spectrum of light that is emitted. Measurements of the emission spectrum are also useful for quantifying the distance between the two fluorophores and have been widely used to monitor enzyme reactions, such as that seen in the presence of β-lactamase. RET is also useful for quantifying analytes as well as interactions between ligands and receptors. Its uses for these purposes are well known.
Not all molecules that adsorb light fluoresce. When RET occurs between a fluorophore and non-fluorescent molecule, the latter will quench the fluorescence of the fluorophore. When the fluorophore and the quenching molecule are sufficiently close to one another, all or nearly all the fluorescent energy will be quenched and little or no light will be emitted. This property is particularly useful for detecting analytes that disrupt contacts between the fluorophore and the quenching molecule since the amount of light that is emitted will be directly proportional to the amount of the analyte. In the absence of analyte, none of the light will be emitted, resulting in a very low assay blank. This property led to the development of “molecular beacons” [16], hairpin shaped molecules designed for the measurement of nucleic acids. In the absence of analyte, the end of the molecular beacon that contains the fluorophore is held adjacent to the end of the molecular beacon that contains the quenching molecule by hydrogen bonds similar to those responsible for the hybridization of nucleic acids. When these interactions are disrupted by the binding of a second molecule of nucleic acid, the distance between the fluorophore and the quenching molecule exceeds that needed for RET and the fluorescence becomes readily visible. By combining RET and TIRFM, it is possible to enhance the desirable properties associated with each technology, thereby facilitating the measurements of analytes. The combined sensitivity of RET and TIRFM has permitted studies of single molecules [14].
In a preferred application of the device, the application of an electric field causes the analyte to migrate to the sensor surface where it interacts with an immobilized molecular beacon or other fluorophore. This results in a change in fluorescence of the immobilized fluorophore. Molecular beacons are particularly well suited for use in this device since their fluorescence increases upon interaction with nucleic acids in a highly sensitive and predictable fashion. One of the limitations of this type of sensor is the need to attach the agent to the sensor surface. This requires additional steps in sensor construction and can be limited by the amount of material that can be attached to the surface. While these limitations are usually not severe, they can increase the costs of sensor construction. A wide range of chemistries is available for attaching materials to the surface of sensors used in the device and reagents for doing so are available from several companies including United Chemical Technologies, Inc., 2731 Bartram Road, Bristol Pa. 19007. Furthermore, it is possible to increase the “depth” of the surface considerably by attaching compounds such as dextran that can serve as additional attachment points.
It is not necessary to attach the detection reagent to the surface to operate the device, however, and another preferred embodiment of the sensor is based on the use of soluble detection reagents. These can have considerable advantages to the use of surface bound material. First, since soluble reagents are not coupled to the sensor surface, their use facilitates sensor design by eliminating the surface-coupling step. Second, they can often be used in massive excess, a phenomenon that can increase the sensitivity and speed of detection. Third, they can be designed in a manner that prevents them from reaching the surface unless they have interacted with the analyte. This can reduce the background fluorescence observed in the absence of analyte. Indeed, the excess reagent can be designed such that it will migrate away from the sensor surface during analysis, a phenomenon that can minimize the background further. Fourth, interaction of the detection reagents and the analyte can take place away from the surface, which minimizes artifacts caused by surface phenomena. These include non-specific adsorption to the surface, which can prevent interactions between the analyte and the detection reagent. While these can also be minimized by varying the potential on the surface of the device, this adds an additional complication to the analytical procedure. Fifth, these reagents are readily adapted to use with quantum nanodots, fluorophores that are not readily photobleached and that have a very high quantum efficiency. Quantum nanodots can be purchased from the Quantum Dot Corporation, 26118 Research Road, Hayward Calif. 94545, USA. Furthermore, quantum nanodots can be excited at short wavelengths and have narrow fluorescence spectra. This permits the simultaneous detection of multiple analytes following excitation with only a single laser beam, a major advantage in analysis of gene expression where it is desirable to observe many gene products at one time.
The need for analytes to reach the sensor surface before they can be observed, a property of TIRFM that facilitates distinguishing specific from non-specific interactions, can result in slow response times. This can also reduce the sensitivity of TIRFM, particularly if the substance to be measured is prevented from reaching the sensor surface. Gene expression products such as mRNA or proteins that are held in tissue sections would not be expected to section such that charged analytes are driven to the surface of the sensor where they can be detected. The application of a charge perpendicular to the tissue section also reduces lateral diffusion of the gene products thereby increasing the likelihood that the fluorescence observed is associated with the cell that is expressing the gene. In addition, by varying the charge, it is possible to accelerate interactions between surface molecules and to reduce non-specific binding.
TIRF can also be monitored without the use of a high-magnification microscope lens. In this case one loses the spatial resolution needed to identify individual cells within a sample. Nonetheless, there are times when it useful to monitor light emission over a large areas, such as during efforts to scan the perimeter of a tumor to determine if the edges have been removed during surgery. There are few limits to the size of the TIRF sensor and it is envisioned that sensors of sizes other than those used commonly by pathologists will be of value for the technique.
Measurements of TIRFM can be done at several different magnifications through the use of an objective prism. High magnification TIRFM using commercially available 60× and 100× microscope objectives can currently be accomplished using devices that have been specifically designed for this purpose. Useful equipment for this purpose can be purchased from Nikon microscope dealers such as Micron Optics, 240 Cedar Knolls Road, Cedar Knolls, N.J. 07927 USA. In these devices, a laser beam is directed through the objective, an oil layer, and a thin coverslip of approximately 0.17 mm. These devices are excellent for visualizing fluorescence in tissue samples. When used with differential interference optics (DIC), these microscopes can also be used to monitor the cells from which the analytes are derived.
Due to the high power of the objective lenses that are used in the commercial microscopes for TIRFM, it is difficult to scan tissue sections in a rapid fashion. There is a need for lower power TIRFM that can also be used with the sensor. As taught here, this is met by designing a new method for illuminating the samples. The use of this strategy to monitor a broad image permits much more rapid scanning of the sample.
Data collection can be made using a charge coupled device (CCD) camera or related cameras of sufficient sensitivity, many of which are available commercially and are available from microscope dealers such as Micron Optics. Intensified CCD cameras are also available that are much more sensitive. These can also be purchased from most microscope dealers. Measurement of light intensity can also be done using photomultipliers that are attached to one of the optical ports on most high quality microscopes. One useful instrument that has been designed for this purpose can be purchased from C&L Instruments, 314 Scout Lane, Hummelstown, Pa. 17036 USA.
Even with the use of low power objectives, it is often desirable to scan the surface of the sensor. This permits one to detect gene products in subsets of tissue sections and thereby distinguish normal and pathological tissues. This process can be accomplished manually by moving the microscope stage that holds the sensor. It can also be accomplished automatically using computer driven stages that are available from most microscope dealers. By combining the use of computer driven stage movements and data collection, it is possible to devise an image of the entire sensor surface at high resolution. The operator can then examine those regions of particular interest, a time saving feature of the method.
The analytical techniques taught here are not restricted to the analysis of nucleic acids, although this will be an important use. For example it is possible to measure proteases by permitting them to cleave specific substrates that are attached to the sensor surface. One such method involves the preparation of peptides that contain a fluorophore and a quencher. Proteolysis of the peptide liberates the fluorophore from the quencher, resulting in enhanced fluorescence. Proteolysis can also remove charged components of the substrate that permits it and its attached fluorophore to migrate to the sensor surface for observation. Similarly, the technique can be applied to the measurements of kinases and phosphatases, enzymes that alter the phosphorylation status and hence the charge of an analyte. Changes in the charges of fluorescent kinase and phosphatase substrates can be used to promote migration of the substrates to a sensor surface where they can be measured. This forms the basis for the enzymatic analyses of these agents as well.
It is not essential to use fluorescent techniques for detection of the analytes that are to be measured. Enzymatic analytes can be often be detected by virtue of their enzymatic activity which can lead to the deposition of colored reagents on the surface of the sensor.
As setout above, the present method can also be used to measure changes in the charge of any fluorescent material caused by interaction with an analyte, including a binding molecule or an enzyme. It can also be caused by a cascade of events such as multiple enzyme-coupled reactions.
The present invention is further illustrated by the following examples, which are not intended to limit the effective scope of the claims. All parts and percentages in the examples and throughout the specification and claims are by weight of the final composition unless otherwise specified.
A Sensor Device to Monitor Gene Expression in Frozen Tissue Sections in which the Analysis Reagents are Tethered to One Surface of the Device During the Entire Analytical Procedure.
There are two principle methods that can be used to detect negatively charged RNA polymerase generated gene products using the device illustrated in
Attachment of detection reagents to the sensor surface can be done by a variety of methods. One of the most convenient is to use a biotin-streptavidin conjugation procedure. In this method a biotin moiety is attached to the surface directly by chemically attaching a biotin derivative to a properly derivatized surface or indirectly by adsorbing a bovine serum albumin biotin complex to the sensor surface. The biotinylated surface is then reacted with streptavidin, a protein that contains four biotin binding sites. Binding of streptavidin to the surface creates a biotin binding site on the surface, which can be used to immobilize biotinylated detection reagents such as biotinylated molecular beacons. Incorporation of biotin into the beacons can be done at the time they are synthesized. For example the beacon illustrated in
Many methods for preparation of chemically biotinylated ITO surfaces are well known in the art. One method that is useful involves cleaning ITO coated slides by treating them with H2O/H2O2/NH3 in a ratio of 10:2:0.6 at 55° C. for 75 minutes followed by baking them in a vacuum oven at 165° C. for 150 minutes to remove water. The slides are then cooled in dry nitrogen and treated with 0.5% 3-aminopropyltrimethoxysilane in toluene. Both reagents can be obtained from Sigma-Aldrich, St. Louis, Mo. They are then washed with methanol and the resulting surface amino groups are biotinylated by reacting the slides with a biotin analog that is reactive with amino groups such as biotinamidocaproate, N-hydroxysuccinimidyl ester obtained from Molecular Probes, 29851 Willow Creek Road, Eugene, Oreg. 97402.
The chemically cleaned slides can also be treated with other agents that permit them to be derivatized with thiol, aldehyde, and other groups that facilitate conjugation with biotin containing and other compounds. They can also be treated with agents that cause them to be derivatized with polyethylene glycol (PEG) and PEG derivatives that can be purchased from Shearwater Corp. (U.S.), 1112 Church Str., Huntsville, Ala. 35801. They can also be treated with reagents such as Sigmacote obtained from Sigma, that renders the surface hydrophobic and that facilitates the adsorption of biotinylated serum albumin.
Introduction of an electrical potential across the ITO or other metal coated slides used to fabricate the optically transparent chamber walls will cause negatively charged gene products such as mRNA to migrate towards the anode where they can interact with detection reagents such as molecular beacons. Indeed, molecular beacons are preferred detection reagents since they usually have low background fluorescence in the absence of analyte and can be designed to interact specifically with predetermined gene products using methods well known in the art. Indeed, companies that specialize in the synthesis of DNA and molecular beacons such as IDT DNA Technologies, Inc. offer a service in which they assist in the design of properly functioning beacons.
The molecular beacon will become much more fluorescent when it binds the analyte for which it has been designed, a phenomenon that causes the shape of the beacon to be altered and that displaces the quenching agent from the fluorophore. For the mRNA to interact with the beacon, it must travel from the cellular milieu to the anode sensor surface. This is facilitated by the presence of the electric potential. Interaction of the mRNA with the molecular beacon can be enhanced by varying the potential used to cause migration of the gene product to the anode. A diagram representing a typical polarization pattern that can improve the interaction of the mRNA and the beacon is illustrated in
While a single molecular beacon can be used during analysis, it is usually preferable to employ at least two different beacons, one of which is intended to serve as an internal methodological control. This beacon can be made to detect gene products such as β-actin that are found in abundant amounts in most cells and whose expression is not changed significantly during most pathologies. The other beacon can be made to detect products that are of experimental or diagnostic interest and should be labeled with a fluorophore that emits at a different wavelength to permit its simultaneous analysis with the control beacon. The finding that the ratios of these gene products change provides strong indication that significant changes in gene expression have occurred within the tissue. Furthermore, many tissue sections will contain more than one cell type. Another control would be to compare the expression of actin in each cell type with the expression of the gene product that is associated with a pathological condition.
The choice of the gene products to be measured for experimental or diagnostic purposes will depend on the results of preliminary studies or of published microarray analyses, many of which are already known to those familiar with the art. Furthermore, it may be desirable to monitor multiple gene products of diagnostic interest at the same time. For example, as noted earlier, microarray analysis has indicated that several different gene products are associated with specific types of breast carcinomas. By using mixtures of beacons that are labeled with the same fluorophore and that recognize several gene products associated with tumor class one can increase the chances of detecting this type of tumor. This is because the interaction of any or all of these gene products with these beacons will be associated with a particular fluorescent emission spectrum. By labeling pools of beacons that recognize gene products associated with a different type of tumor with a fluorophore that has a different emission spectrum, it is possible to detect and classify pathological cells derived from more than one class within the tumor or to more accurately classify the tumor type, a significant advance in diagnostic practice. Since analysis can be done on sections obtained at the time of surgery, use of the sensor makes it possible for the surgeon and pathologist to modify the surgical procedure in the most appropriate fashion for the patient during the procedure.
There are two principle advantages that accrue from operating the sensor using detection reagents that are attached to its surface. The first is simplicity of analysis. Since the detection reagents are physically separated from the tissues throughout the procedure, it is not necessary to use methods that limit fluorescence excitation to the anode or cathode. Thus, while procedures such as TIRFM and multiple photon excitation can be used to monitor interactions between the beacons and the gene products on one sensor surface, the fact that the beacons are found only on this surface means that these techniques are not required. Indeed, it is often possible to use standard fluorescence microscope techniques when the background illumination can be adequately controlled. This reduces the costs of the instrumentation required. And second, use of surface bound fluorophores does not require physical separation of bound and non-bound analytes. This permits monitoring of low affinity interactions. While this is not a problem with the molecular beacons, it can be an issue for other types of analytical procedures such as interactions between fluorophores and surface bound proteins.
The advantages of using immobilized detection reagents can be offset by several factors including difficulties in attaching them to the surface, limits to the amount of material that can be attached to the surface, effects on ligand recognition caused by their attachment to the sensor surface, the need to employ organic dyes that can photobleach, and the influence of non-specific interactions. The latter can often be minimized by the use of agents such as bovine serum albumin and polyethylene glycol to block these interactions. The limitation on the number of groups that can be placed on the sensor surface can be offset in part by increasing the surface area by coating it with dextran and other agents that serve as attachment sites. These techniques are all well known to those familiar with the art.
A Sensor Device to Monitor Gene Expression in Frozen Tissue Sections in which the Analysis Reagents Move with the Gene Products to the Anode During Analysis.
The second preferred embodiment, the device shown in
A diagram outlining the mechanism by which this sensor operates is shown in
There are several advantages to detecting analytes using soluble reagents that can be separated in an electric field. First, there is no need to attach them covalently to the sensor surface. This simplifies the design of the device. Second the fluorophores migrate to the sensor surface only when they have formed a complex with the analyte, a phenomenon that provides an intrinsic mechanism to limit background fluorescence. In fact, since the PNA-fluorophore complex can be made to have a weak positive charge, molecules that are not bound to the mRNA gene products will migrate away from the sensor surface. As a result, a massive reagent excess can be used within the device without causing an unacceptable increase in background noise. The fact that a larger amount of these reagents can be used in the device also increases its sensitivity and the speed with which it can be operated. Finally, as will be noted in later examples, the mechanism that underlies this analytical approach can be used to monitor gene products other than nucleic acids.
These advantages of using soluble reagents for analysis of nucleotide based gene products are offset in part by the requirement that illumination be limited to the anode sensor surface. One practical approach for doing this is to use devices that illuminate the surface by total internal reflection. This limits illumination to the surface of the sensor used for detection. Equipment for TIRFM is commercially available from microscope dealers who handle instruments made by either Nikon or Olympus. Instruments purchased from these companies are limited to relatively high power objectives, however (i.e., 60× and 100×). This can make it difficult to scan rapidly an entire sensor surface. There are other strategies for performing TIRFM that can be used with lower power objectives. These involve illuminating the sample through a prism such as that shown in
Another means of illuminating the anode surface is to use two or three photon microscopy or confocal microscopy. In the former approach, the anode surface would be illuminated such that that single photons are unable to excite the sample. Focusing the illumination source on the sensor surface to cause it to be illuminated “simultaneously” by two or more photons provides sufficient energy to obtain fluorescence emission. The major limitation to the routine use of this type of illumination is its high cost.
Separation of the bound and free detection reagents is done by application of the electric field, which causes the bound detection reagent to migrate to the anode when the complex is negatively charged or to the cathode when the complex is positively charged. The rate at which the analyte will reach the surface will depend on the difference in potential between the plates, the frequency with which the potential on the plates is changed, the size and charge of the analyte, and factors that may limit its ability to migrate to the surface of the plate. Variations in the electric field can be very useful for causing the complex to form. Thus, by alternating the electric field, one can cause charged analytes to migrate back and forth within the region of the sensors. This creates a mixing effect that can enhance interactions between the analytes and the detection reagents that facilitate formation of the complexes.
The sensor described in
The sensor surfaces are coated with ITO, SnO2, or other conducting or semi-conducting materials that are also optically transparent at the wavelengths to be used. This is done to enable an electric potential to be developed between these two surfaces. While this is a preferable means of designing the electrical components of the sensor since it permits both the optical and electrical components to be combined, workable sensors can be envisioned that would contain conducting grids or membranes in place of one or both of these surfaces.
The device outlined in
Several methods can be used to deliver an electrical potential to the surface of the plates. In one procedure, the entire plate is coated with ITO or other conducting metal. When this is placed on a metal wire or other conducting surface, it will permit the introduction of an electrical potential on all portions of the plate, including that in contact with the sample. Another method of connecting the conducting surface of the plate to the wire or conducting surface must be used when only one surface of the plate that contacts the sample is coated with ITO or conducting metal. Use of plates having only a single coated surface can facilitate the optical transmission of the device, a property that is often critical at ultraviolet or near ultraviolet wavelengths. One means of making the appropriate electrical contact involves placing a wire directly on the metal surface of the plate. This approach suffers from the difficulty of maintaining sufficient contact between the wire and metal coating on the surface to facilitate uniform electrical conduction, particularly when the device is subjected to repeated handling. To circumvent this, one can glue the wire to the metal coated surface using material obtained from Delta Technologies Limited, 13960 North 47th Street, Stillwater, Minn. 55082, USA. Alternatively, one can place a thin strip of metal on the conducting surface of the plate. This can also be glued in place. A preferred material for this can be purchased from Schlegel Systems, Inc., Rochester, N.Y. 14623, USA. One thin strip that is particularly useful is their Conductive Anti-Tarnish Copper Tape which comes in a variety of widths, contains one sticky surface, and is heat stable at 121° C., making it autoclavable. This permits construction of sterile sensors that can be used as cell culture growth chambers. The resistance between these tapes and that of the ITO surface of glass microscope slides purchased from Delta Technologies is less than 1 ohm.
The need to prevent electrical contacts between the two plates of the device shown in
Electrical contacts between the sensor and the sample occur through a conducting fluid. This can be nearly any dilute buffer that is capable of conducting electricity. The pH of the buffer should be chosen to render the analyte charged such that it migrates towards the surface that is to be observed. This includes the surface that coated (Example 1) or that to which the analyte-detection complex will migrate (Example 2). The type of buffer to be used in the connecting fluid will vary with the sample being analyzed. Analysis of RNA transcripts can be analyzed using most neutral buffers, often with EDTA, a divalent cation chelator that can reduce RNase activity. The use of a conducting fluid that contains a small amount of 0.3-1% agarose is often helpful for maintaining the alignment of the analyte and cells in the tissue section. Agarose that is suitable for this use, including low temperature melting forms, can be obtained from many commercial suppliers including FMC 191 Thomaston St., Rockland, Me. 04841 (USA).
Following the addition of the sample and conducting fluid, the two component surfaces of the sensor device are then joined to create a “sandwich” such that their conductive surfaces are brought into contact with the fluid. In this position each conductive surface of the sensor contacts the conducting fluid and, in some cases, the sample. Each surface is separated from the other by the insulating membrane as shown in
Sensors that can be Heated and Cooled.
ITO and other metal coatings have a significant resistance depending on their thickness. For most applications the thickness and hence the electrical resistance of these layers will not be a major concern unless it impedes the optical clarity of the sensor since relatively little current flows through the sensor during its operation. The passage of larger amounts of current through metal coatings can be used to heat the sensor, however, and a preferred means for doing this using glass slides that are metal coated on both surfaces is shown in
Heating the sensor prior to, or during, its operation can facilitate analysis. Heating prior to analysis can help disrupt the cell membranes in the tissue, thereby enhancing migration of the analytes to the sensor surface and/or facilitating interactions between the analytes and the detection reagents. Heating can also contribute to the specificity of nucleic acid detection. For example, the temperature stabilities of oligonucleotides as a function of ionic strength are well known. Single base changes can result in a substantial change in the stability of an oligonucleotide pair. By heating the sensor surface, the interactions between mRNA and the molecular beacons or PNA can be controlled accurately. Brief heat treatment can also disrupt the molecular beacons in a transient fashion, enabling them recognize their “ligands” more rapidly.
Heating can also be used to examine the quality of the sensor surface before use. For example, when sensors that contain molecular beacons are heated above the beacon melting temperature, they will fluoresce. By measuring the amount and uniformity of fluorescence observed, one can monitor the quality of the coating. Since operation of the beacons is reversible, they will return to their non-fluorescent conformation when the sensor is cooled. Heating can also be used to distinguish non-specific and specific interactions during the analysis of mRNA and other nucleic acid hybridization assays when the sensor is used in the fashion described in Example 2. As the sensor is warmed, non-specific interactions between mRNA and the fluorescent PNA will be disrupted, preventing the transport of the PNA to the sensor surface. Precise control of sensor temperature can thereby facilitate identification of single base pair mismatches. This may be particularly helpful in identifying cells that contain mutations in only one allele.
It is also possible to incorporate mechanisms for cooling the sensor. Methods for doing this can be as simple as mounting the sensor on a Peltier heating/cooling stage or as complex as passage of a cooled fluid in a chamber that can be constructed beneath the lower sensor plate or above the upper sensor plate. By altering the temperature of the sensor, it is envisioned that it can be used for polymerase chain reaction analyses that can amplify the analytes being studied.
The sensor has been designed to be operated in the presence of an applied voltage. While it is conceivable that some analysis can be obtained in the absence of an electrical potential, the benefits of using an applied voltage greatly facilitate analysis sensitivity and speed. Application of an electrical potential to the device can accelerate the movements of analytes to the sensor surface, depending on their charges. This will result in enhanced speed and sensitivity of the measurements. The presence of an electrical potential can also cause disruption of cells and thereby permit detection of analytes that would otherwise be prevented from reaching the sensor surface. Many analyses can be performed under constant voltage conditions. It is not necessary for the voltage across the sensor be constant, however, and it will often be preferable to vary the voltage using patterns, shown in
Variations in the electric field can also facilitate analyses when the sensor is used as described in Example 2. In this case a variation in the surface charge similar to that in
The ability of the sensor to detect protein gene products can also be enhanced by the use of the electric potential. By operating the sensor at the appropriate pH, it is possible to separate protein isoforms that may otherwise interact with the same detection molecule. Many proteins can be phosphorylated, a phenomenon that also results in a shift in their isoelectric points. Thus, even if two proteins are recognized by the same fluorophore, they can be distinguished if one migrates towards the sensor surface and the other migrates away from the sensor surface at the pH at which the sample is being measured. They can also be distinguished if they are oxidized differently when they come into contact with the metal oxide coating.
When the sensor is assembled correctly, the sample will be contained within a small chamber the thickness of the gasket. It is possible to attach thin tubes or needles that act as “ports” to access the interior of the chamber within the gasket. One means of doing this is simply is to insert needles through the gasket. This permits perfusion of substances through the device. Furthermore, it is possible to utilize both surfaces of the device for observation. The cell can be used to rapidly optimize electrical polarization parameters for promoting interactions between the analyte and the sensor surface or materials attached to the sensor surface. Thus, in addition to its use as a sensor per se, it can be used to optimize the parameters needed for analysis of tissue sections in the device to be employed for this purpose such as that in
Several methods are available for monitoring analytes in the sensor using TIR. As noted earlier TIRFM systems can be purchased from Nikon and Olympus Corporations. These enable illumination of the sample through either 60× or 100× high numerical aperture objectives that are in optical contact with coverslips that contain the samples. Use of these TIRFM systems requires that the surface used for analysis be a coverslip having a thickness of approximately 0.17 mm. They also require the use of an immersion oil to make optical contact between the objective and the coverslip.
Several other types of TIR illumination can be used for examining the sample. A preferred illuminator has the design shown in
The illuminator functions by passing light from a laser through a rectangular lens having planar and convex surfaces. This lens is in optical contact with a triangular prism that is in optical contact with a 0.17 mm coverslip as shown. The prism can also be replaced by a cube as indicated by the broken lines in
The illuminator and the sensor are placed upon a microscope stage in a holder designed to keep the illuminator next to the side of the sensor. It is important that the illuminator not be joined permanently to the sensor, however. Microscopic observation across the width of the sensor is accomplished by moving the illuminator and the sensor in tandem as shown in
Use of the Device with Standard Light Microscopy.
The design of the device permits its use with standard light microscope techniques including phase contrast microscopy, bright field microscopy, darkfield microscopy, differential interference contrast microscopy, confocal microscopy, and epifluorescence microscopy. In most of these uses, the sample is illuminated by light that passes roughly perpendicular to the plane of the sensor. This permits examination of the entire sample, not just that portion that is adjacent to the sensor surface. By comparing the images obtained using these techniques with those obtained by TIRFM, it is possible to identify specific cells that contain the analytes being observed during TIRFM even though it is not possible to observe the entire cell using TIRFM.
The tissue sections can also be stained to increase the contrast between various cell types or organelles. This can be done using non-fluorescent dyes prior to TIRFM. It is also possible to use fluorescent dyes prior to TIRFM if the dye recognizes a substance to be analyzed or if the dye can be excluded from the evanescent field by application of the electric field. The advantage of using a dye before performing TIRFM is that it will facilitate correlating specific cell types with the location of the fluorescence. In some cases, however, it may not be possible to stain the tissue prior to TIRFM. In this case, it may be necessary to remove the non-sensor surface from the device to gain access to the tissue section. This can be facilitated by including a small layer of gauze between the non-sensor surface and the tissue section to prevent sticking of the surface to the tissue.
In some cases it will also be useful to employ the electrical potential that can be generated by placing a charge on the sensor surface to remove excess stain from the tissue section, thereby reducing the time needed for staining and clearing the background. This can be done by placing the sensor surface and its attached tissue section in a bath and applying a low voltage across the sensor surface and the bath.
Use of Photobleaching within the Device.
One of the limitations of using fluorescence to study gene expression is related to the number of fluorophores that can be distinguished at one time. Photobleaching can expand the measurement range, however. For example fluorescein and Alexa Fluor488 have about the same fluorescence spectra. The former is much more readily photobleached, making it possible to distinguish analytes that are labeled with fluorescein from those labeled with Alexa Fluor488 by the differences in the rates at which they are photobleached. The combined use of organic dyes and quantum nanodots, which are nearly impossible to photobleach should extend this technique further.
Another use of the device is for measurements of enzyme levels in tissue samples. Many cancers have different levels of extracellular and intracellular proteases and these can be readily distinguished by use of fluorophores that contain protease cleavage sites. Cleavages at these sites by the actions of the specific proteases will cause the release of a quencher from the fluorophore resulting in fluorescent light emission. One of the advantages of the device described here is that it is possible to use the electrical potential to cause proteins and other molecules that are not nearly as negatively charged as mRNA and nucleic acids to migrate to a different sensor surface than the nucleic acids. This will permit simultaneous analysis of mRNA and proteins in the same sample. Application of similar approaches will permit the measurement of any type of enzyme reaction that can lead to the appearance or disappearance of fluorescence.
The ability of the sensor to detect differences in the net charge of a molecule can also be used in assays of kinases and phosphatases, enzymes that alter the phosphorylation status and charge of a molecule. For example it is possible to prepare fluorescent peptides that are substrates for various protein kinases. The presence of kinase activity in the sample can cause the fluorescent peptide analog to migrate to the anode whereas the non-phosphorylated analog may fail to migrate or may migrate to the cathode at the pH employed in the conducting fluid buffer. This will permit cell specific analysis of these important cellular enzymes, many of which have been implicated in tumorigenesis.
The ability of the sensor to detect differences in charge can also be used to detect protease activity. Fluorescent protease substrate can readily be designed such that proteolysis will change the ability of the fluorophore to migrate to either the anode or the cathode, where it is readily detected. This can be accomplished by adding charged amino acid residues to the substrate, which are then cleaved by the protease.
Binding of small fluorophores to proteins or larger macromolecules results in a loss of molecular mobility. When the small molecules are labeled with fluorophores, this will result in a change in fluorescence polarization that is readily detected. The device illustrated in
As noted earlier, data obtained using microarrays suggest that many mRNA will be elevated at the same time in cancerous and malignant cells. This phenomenon can contribute to the sensitivity of the device. Molecular beacons that are specific to multiple mRNA are coupled to the surface of the sensor surface as in Example 1. When these are labeled with the same fluorophore, they will detect the increase in any of these mRNA. Similarly, some populations of mRNA decrease in cancerous cells. By mixing these and labeling them with a different fluorophore than used in beacons to monitor mRNA whose expression is found to be unchanged and with a different fluorophore that used in beacons designed to monitor mRNA whose expression is found to be increased, it is possible to increase the sensitivity of the method. As noted earlier, it is also possible to make use of both surfaces of the device to increase the numbers of analytes that can be monitored. Similar types of mixtures can be employed for analysis of gene transcription produces using the sensor as described in Example 2.
The principles shown in the device illustrated in
Microtiter plates are often used for analysis and the application of an electrical potential to this assay format can facilitate analysis. For example, it can be used to increase concentration of an analyte at the plate surface. It can also be used to reduce the concentration of an analyte at the plate surface. Many of the applications of the sensors except for those that involve tissue sections can be transferred to a microtiter well plate format. These include enzyme assays and nucleic acid assays. Several formats can be used to build microtiter plates that can be used with electrical potentials. One of these formats is illustrated in
Sensors with Permeable Optical Polymers (Polymeric Hydrogels)
One of the limitations of the sensor shown in
The voltage limitation of the sensor can be overcome by replacing the glass optical components of the sensor with permeable optical polymers (polymeric hydrogels) that are permeable to ions and placing the polymer between the sample and the electrodes as shown schematically in
The use of polymeric materials has another major advantage as well. It permits the design of components that include aqueous solutions that can be stored in sealed pouches. This frees the operator from having to add water or buffers. This is important because it lessens the potential for mistakes to be made. When tissue sections are being made during surgery, time is of the essence. The fewer operations that are required, the less likelihood that mistakes will be made. Furthermore since, the fluid components are within the gels, it reduces the chances that bubbles will be introduced between the tissue sections and the components of the sensor when assembling sensor components, a process likely to be done manually by the person cutting the tissue sections.
The overall principles that underlie the operation of a polymer-based sensor are the same as those that are responsible for the operation of the sensor in
The use of polymers in the design of the sensor in
The anode sensor assembly contains the polymer that will be the primary site of analysis when RNA gene products are to be examined from tissue sections since this is the direction in which these gene products will migrate during electrophoresis. This is identified as component #3 in
When the sensor is being used to measure RNA and no other gene products, virtually all the measurements will be made on the part of the sensor shown as component #3 in the anode assembly (
It should also be noted that sensors can be made with molecular weight cut off devices by inserting a piece of dialysis tubing between the polymers. When these are placed between components #2 and #3, all the high molecular fluorescent species will be collected at a surface that can be made very thin to permit better detection (c.f., component #2a,
The sensor also contains other components that are not optical polymers or even polymers but that are present to facilitate delivering an electrical potential to the sensor. Components #1 and #9, which serve as the anode and cathode, respectively, are designed to create an electrical potential across the device. Components #2 and #8 can be incorporated into the anode and cathode as shown in
The final steps in the construction of the anode and cathode assemblies involve layering the polymers illustrated as components #3 and #4 and components #6 and #7 on the anode and cathode respectively. This is shown in
Use of the sensor device requires only a few simple steps. Either the anode or cathode assembly pack is opened at the time of sectioning or a section is placed directly on the exposed gel. When this is opened just prior to use, there should be sufficient moisture to give good contact of the tissue section to the gel. It is important that no air be trapped between these sections, however, since this can interfere with RNA or other analyte extraction from the tissue. A few drops of sterile water can be added at this time to avoid this problem, if needed. Once the section has been placed on the anode or cathode assembly it is covered by a cathode or anode assembly, which is placed on top of the section such that its gel contacts the section. It is a good practice to begin with either the anode assembly since it is easy to see how the tissue section contacts the polymer and since this contact is the most important. Then, one adds the cathode assembly such that its gel side faces the tissue section. Again, a few drops of water might be needed, but this should not be necessary if the assembly package is opened at the time of use and if it has stayed hydrated.
Once the sensor sandwich has been assembled, it is ready for the electrophoresis step. The sensor sandwich is inserted into the electrophoresis chamber as diagrammed in
Following electroporation and electrophoresis, the sample is ready for visualization. This is done by removing the sandwich from the electrophoresis box and, in the case of negatively charged analytes such as RNA, observing the fluorescent material that is collected in the portion of the sensor in components #2a or #3 (
Sensors with Peptide Nucleic Acids (PNA)
A desirable detection reagent for a nucleic acid is a molecule that has bases that are held in an ordered fashion such that they can form Watson-Crick base pairs with nucleic acids and that lacks the negative charges in the backbone atoms that hold the bases in order. This is because the negatively charged phosphates of nucleic acids exert a repulsive effect on formation of the oligonucleotide duplex. By replacing the negatively charged phosphate atoms with atoms or groups of atoms that have either no charge or that have positive charges, one can devise detection reagents that will have high affinity for specific oligonucleotide sequences. Indeed, the affinities of these can be greater than that of nucleic acids for complementary nucleic acids.
PNA are molecules capable of forming Watson-Crick base pairs with nucleic acids that have a peptide backbone. Because they lack the negatively charged sugar-phosphate backbones found in RNA and DNA, hybrids of RNA-PNA and DNA-PNA are known to be highly stable [24]. PNA can be constructed to be essentially uncharged, negatively charged, or positively charged simply by incorporating amino acids into their backbones by standard peptide synthesis chemistry. PNA have also been labeled with fluorophores [18,21] and used to detect nucleic acids by fluorescence in situ hybridization (FISH). PNA are not the only structures that can be used for this purpose, however. Agents in which the phosphate is replaced by sulfur or carbon are also useful.
When PNA are bound to nucleic acids, they can alter its mobility [20] in gels or in capillary electrophoresis tubes [17]. The ability of DNA to change the charge of PNA such that its migration in an electric field is reversed has not been employed, however. This is a particularly important property for use in a sensor of the type taught here in which it is preferable for the electrophoretic migration distance to be relatively short. Binding of uncharged or positively charged PNA to RNA or DNA will cause it to become negatively charged. As a consequence, the complex will migrate in the opposite direction from the uncomplexed PNA in an electric field. This can be used to separate bound PNA from non-bound PNA. If the PNA is labeled with a reagent such as a fluorophore, a radioisotope, biotin, or other molecule that does not cause it to acquire a net negative charge, then binding of the labeled PNA to nucleic acids will cause it to be separated from the non-bound PNA. This provides a very useful and simple tool for the identification of nucleic acids. Further, this permits the labeled PNA to be employed at very high concentrations, which facilitate its interactions with nucleic acids without increasing the background signal when the signal is measured by a technique such as total internal reflection fluorescence of TIRFM. In addition, this property can be used to cause nucleic acids or other charged materials to migrate into areas where they can be assembled into complexes.
Because PNA have a peptide backbone and can be synthesized similar to peptides, it is possible to incorporate several different types of labels into them. For example, it is possible to add cysteine residues to PNA that will permit labeling of the molecule with fluorescent probes that react with thiols or that can be made to react with thiols. Many such probes are available from Molecular Probes, Eugene, Oreg. It is also possible to incorporate lysine molecules into PNA. This will give them a positive charge or serve as a labeling site for amino reactive agents. These are also available from Molecular Probes in a wide variety of absorption and emission wavelengths. One can incorporate arginine residues into PNA to alter their charges as well. PNA have also been labeled with histidine residues [24]. The pK of the imidazole moiety of histidine can have a favorable influence on the migration of PNA in an electric field that has a pH gradient. For example, at low pH, histidine is positively charged. At high pH, it becomes uncharged. A PNA that contains histidines will tend to migrate away from an anode when it is in a low pH environment. Its mobility will be reduced as it reaches a higher pH environment due to the loss in charge. Thus, one can easily devise conditions in which histidine labeled PNA migrate away from an anode until they reach a region of an electrophoresis chamber in which their migration becomes slow. One use of this is to drive the PNA to a region of the chamber away from the anode but prevent them from migrating to a region where they would be unable to react with oligonucleotides. PNA that have bound to oligonucleotides will migrate back towards the anode away from their non-bound counterparts.
The design of PNA is relatively straightforward and is based on the notion of Watson-Crick base pairing [24]. The fact that PNA are uncharged or can be made positively charged enables them to invade short RNA-RNA duplexes found in most gene expression products. Increasing the temperature of the device can facilitate this. The usual length for the hybridization reaction is 16-25 bp. The only other considerations in designing the PNA relate to the solubility of the molecule. Long uncharged PNA are generally not soluble and are not well suited for use in the sensor. Positively charged PNA are much more soluble and much better suited for the measurements with the sensor, particularly if their charges can be modulated as a function of pH, e.g., by addition of residues such as histidine when they are employed at pH values in the range of 6-8.
A key to the operation of the sensor is its ability to maintain a very low background. This enables the detection of trace quantities of RNA analytes. As just discussed, the use of PNA and the ability to reverse the migration of labeled PNA molecules in an electric field is one means of maintaining a low background. Another method of reducing the background is to use PNA that have a hairpin conformation similar to that found in molecular beacons. In the PNA hairpin conformation [23], which is found before the PNA is complexed with an oligonucleotide, the fluorophore at one end of the PNA is quenched by a molecule that is attached to the other end of the PNA by resonance energy transfer. This occurs due to the proximity of the fluorophore and the quencher, which are near one another only when the PNA has a hairpin conformation. Binding of the PNA to RNA causes the hairpin to become linear, which results in the fluorophore being moved from the quenching agent. As a result, the fluorescence becomes visible and can be observed. Since formation of the hairpin shape does not alter the isoelectric point of the PNA before it is bound to RNA, the hairpin shaped PNA will also migrate away from the anode. This will change when it interacts with RNA, however, the time that the fluorescent PNA-RNA complex will be migrating towards the anode. These movements are illustrated schematically and described in
Another important means of reducing the background fluorescence is the use of total internal reflection optics. By restricting the illumination to the components of the sensor that contain the fluorescent RNA-PNA* complexes, it is possible to prevent illuminating the uncomplexed PNA*, which would contribute to the background. It is desirable to illuminate only component #3 in the sensor. This can be done if component #3 is transparent to the illuminating radiation, if component #3 has a higher refractive index than component #4, and if component #3 is illuminated at an angle less than the critical angle. This can be calculated from Snell's law from the refractive indices of components #3 and #4.
The requirement for total internal reflection illumination of component #3 can be met using polymers that have been designed for the construction of soft contact lenses that are intended for long use. These have been designed to be sufficiently porous to enable air and fluids to pass through the lens where it can reach the cornea. Further, their refractive index is sufficient to bend light needed for vision correction. The refractive index of these materials has been shown to permit their use for total internal reflection fluorescence [22] as would be expected from their refractive indices.
There are several materials that have been used to construct contact lenses. Two of the most common are HEMA (hydroxyethylmethacrylate) and HEMA-MAA (HEMA-methacrylic acid). Commercially available lenses of the former have a refractive index of approximately 1.437 and contain 42% water. Commercially available lenses of the latter have a refractive index of approximately 1.407 and contain 55% water. The latter are also much more permeable and have significantly larger pore sizes. Thus, these materials are suited for both total internal reflection in aqueous buffers in which the refractive index is approximately 1.33-1.37 and for electrical conduction needed for electrophoresis. Both types of polymers can be readily molded and made sufficiently thin for use in the device and are commonly made in sizes in the range of 0.2 mm. Due to the desirability of having the most resolution possible, it is important that the thickness of component #3 be kept relatively small, in the order of 0.2 mm. The thickness of component #4 should also be kept small, but this is not as important as that of component #3, which is the component that will be illuminated. Since component #4 is not to be illuminated, its composition is much less critical than that of component #3. In fact the composition of component #4 can be virtually any soft gel that can be molded into a shape that will fit between component #3 and the tissue section. The critical features of component #4 are that it permit migration of RNA, PNA*, and RNA-PNA* complexes and that it have a lower refractive index than component #3 to permit component #3 to be illuminated by total internal reflection fluorescence. Thus, it is even possible to use a low percentage polyacrylamide or agarose gel for component #4. The use of polyacrylamide also permits the incorporation of immobilines into the gel during polymerization [25,19]. These should be chosen to buffer the local pH such that PNA* will be positively charged and will migrate towards the tissue section and away from component #3. The immobiline to be chosen, if one is to be used, would depend on the design of the PNA*, which will depend on the RNA to be monitored. In general, it is most useful to chose an immobiline that will buffer the pH of component #4 to be at least 0.1 pH unit less than the pH of the PNA*. The pH of the solution that is in components #1-3 should also be lower than the pH of the immobiline in component #4.
When RNA is the only cellular constituent to be analyzed, the composition of components in the cathode assembly is not nearly as critical as those of components #3 and #4. In general, components #6 and above should be at a pH that is equal to or greater than that of component #4. These can be fabricated of polyacrylamide or HEMA-MAA. When component #7 is to be used for total internal reflection, it is better to construct it of HEMA. It will be subjected to the same considerations as those discussed next for component #3. Note, that when an immobile is not used, the buffer throughout the sensor should have a pH that lower than that of the pH of the PNA*.
The design of component #3 should be considered carefully since this is the component of the sensor that will be illuminated and used for detecting the sample. As a rule, component #3 should be a hydrogel having an optical density greater than that of the buffer on either side of it and greater than that of component #4 to permit its illumination in a total internal reflection fashion and since it should be capable of transmitting an electrical current. This is a property that is also found in most soft contact lens hydrogels such as those that contain HEMA. Methods for preparing polymeric hydrogels containing HEMA and other substances are well known in the art and more than 700 patents related to the fabrication of these types of polymers were obtained by searching the United States Patent data base with the terms “HEMA” and “contact lens.” Particularly useful U.S. Pat. Nos. 6,447,118, 6,552,103, 6,582,631, and 6,623,747, which describe methods for molding and modifying hydrogels that can be used to prepare component #3 in the sensor using an appropriate mold. It should be appreciated that nearly any hydrogel material that is has a refractive index that is sufficiently greater than the buffer to be used to permit total internal reflection of light, that has the ability to conduct an electrical current, and that is optically clear at the wavelengths of light used for illumination and fluorescence will be appropriate for use in sensor component #3 and for use in sensor component #7 when the latter will also be used for analysis and illuminated by total internal reflection.
Several aspects of component #3 can influence on the operation of the sensor. For example, when PNA having positive charges are used to detect RNA, it is useful to make the surface of component #3 positively charged. This will facilitate migration of non-complexed fluorescent PNA away from the surface of component #3 into areas of component #4 that will not be illuminated by total internal reflection fluorescence. When the sensor is used to detect RNA, it is also useful to fabricate component #3 from a hydrogel that has a smaller pore size. This will enable component #3 to behave in a semi-permeable fashion and thereby prevent the RNA-PNA* complex from migrating through it. This will avoid-the need to attach materials to component #3 that are capable of binding nucleic acids or to use a semi-permeable membrane such as component #2a (
Following completion of the electrophoresis, it is necessary to detect the fluorophores that are bound to the surface of component #3 or, if the pores of component #3 are sufficiently large, that have traversed component #3 and accumulated on component #2a. This can be accomplished using an illuminator that is focused on the side of component #3 as seen in
Once the fluorescent image of the gene products has been captured, components #8 and #9 can be removed (if they have not already been removed) and the remainder of the sensor sandwich can be transferred to a light microscope. This permits visual inspection of the tissue section, if desired. Alignment of the section with the fluorescence image can be made by comparing the position of component #3 while the section is on the microscope stage with that while it was on the fiber optic. This is because the position of the tissue section will remain constant with regard to the position of component #3.
The sintered materials that can be used to make components #2 and #8 can be obtained from SPC Technologies Ltd., 1 Raven's Yard, Nethergate Street, Harpley, Norfolk, PE31 6TN, UK.
Since a primary use for the sensor will be microscopy, the example shown here is constructed from microscope slides. There is no reason that larger or smaller sensors cannot be made, however. The sensor can also be constructed of 1 mm thick slides, a common size for microscopy, coverslips that are 0.17 mm thick, a common size for microscopy, or from a combination of the two. Indeed, since the device is likely to be observed by TIRFM, a preferred construction would involve the use of a coverslip for the portion of the sensor most likely to be viewed using TIRF. When RNA gene expression products are to be examined, this will be the anode. The view in
The sensor contains at least one and preferably two optically transparent components. These are covered with a tape that is folded around the sensor as indicated in the first image of the top view. Other methods of attaching the electrical contacts will also work, but this design was chosen for its robustness, high conductivity, and ease of construction. Note that the conducting tape lies along the top and bottom of the entire sensor surface to facilitate even electrical contact with the metal oxide layer and the brass conductor. The edge is not coated throughout most of the slide, however, leaving it available for TIRF illumination. There are other means of attaching the tape such as running it along the metal oxide layer and folding it back around the ends. The method of attaching the tape does not matter to the function of the sensor, provided that the edge of the plate will permit TIRF illumination, should this type of illumination be used during analysis. Shown below the slide is the structure of the conductor and the gasket. Basically, each has a rectangular shape that enables it to contact the conducting tape without blocking the ability of the user to observe the contents of the sensor, e.g., tissue sections.
Note that the conducting tape is shown as in a semi-transparent fashion. It does not cover the edge of the sensor plates (slides) for most of the length of the sensor. This is the portion of the sensor that will be used for TIRF illumination, should the illuminator described later be used for visualization of the analytical results. Several different types of visualization can be used, as noted in the text.
The surface area illuminated on the sensor would depend on the curvature of the cylindrical element and its distance from the sensor surface. Only the surface facing the sample would elicit fluorescence. A cutoff filter would need to be placed between the sensor and the detector to distinguish light of different colors—for example from different quantum nanodots.
As shown, the illuminator would be held adjacent to the sensor surface such that both would move side to side as a unit. The sensor could be moved forward and backward relative to the illuminator. This would permit different “slices” of the sensor to be observed.
Microtiter well plates that contain conducting surfaces can be constructed in a variety of methods. The only requirement is that two electrically conducting surfaces be able to contact fluids within the well. One method of constructing a plate in which all the wells will be at the same potential is shown in this
In this arrangement, each well will be at the same electrical potential. An alternate mode of constructing the plate top can be used to create plates in which the electric potential in each well can be controlled separately. One way of doing this is to use a top that lacks a conductive layer. A separate wire is inserted through the top into each well. When the microtiter plate is closed, the wire will make electrical contact with the wells.
It should also be noted that it is not necessary for the top of the microtiter plate to contain electrodes. To prepare a device that can be used in an open format, an electrically conducting surface is sputtered on the molded plastic layer that is used to form the walls of the wells to completely coat its inside and outside surfaces. An insulating layer is then coated on the bottom of this molded piece before it is glued to the metal-coated bottom.
1. Electrode and electrode holder
2. Spacer to separate electrode and holder from optical surface (optional depending on the design of the electrode holder). This can be made of a hydrogel, sintered polypropylene, or other porous substances.
2a. Semi-permeable membrane to trap analytes
3. Polymer or other material used for optical analysis
4. Polymer or other material to used as a spacer and to facilitate mixing—separates optical analysis surface from sample.
5. Sample
6. Polymer or other material used as a spacer (optional, permits additional analyses)
7. Polymer or other material used as a spacer (optional, permits additional analyses)
8. Spacer to separate electrode and holder from optical surface (optional depending on the design of the electrode holder). This can be made of a hydrogel, sintered polypropylene, or other porous substances.
9 Electrode and holder.
One aspect of the invention provides hydrogels similar to those used to make contact lenses that can be used in a sensor because the hydrogels are suitable for electrophoresis and optical refraction and capture of reagents. The other aspect of the invention is the sensor itself and will depend on how the sensor is used. The sensor is designed to be user friendly in that the user does not need to add any fluids. For this reason, the electrodes need to be built into the sensor. In other uses, the user can add the fluids. In this case the electrodes do not need to be built into the sensor per se, but can be built into the electrophoresis box.
The word “bound” reflects the idea of “change” as well as “binding.” Interaction of the detection reagent with the analyte leads to a change in the direction of its migration in an electric field. Electrodes do not need to be attached to the sensor per se unless the device is to be constructed such that the user does not need to add fluid. A spacer would still be required to keep the component #3 from touching the electrode to permit bubbles to escape the device. The device as shown is useful for analyses that are located at different spatial positions in an analyte such as a tissue section.
(1) International Human Genome Sequencing Consortium. Initial sequencing and analysis of the human genome. Nature 2001; 409:860-921.
(2) Venter C J et al. The sequence of the human genome. Science 2001; 291:1304-1351.
(3) Moseley M R. Current trends in differential expression proteomics: isotopically coded tags. Trends in Biotechnology 2001; 19:S10-S16.
(4) Dhanasekaran S M, Barrette T R, Ghosh D, Shah R, Varambally S, Kurachi K, Pienta K J, Rubin M A, Chinnaiyan A M. Delineation of prognostic biomarkers in prostate cancer. Nature 2001; 412(6849):822-826.
(5) 't Veer L J, Dai H, van de Vijver M J, He Y D, Hart A A, Mao M, Peterse H L, van der K K, Marton M J, Witteveen A T, Schreiber G J, Kerkhoven R M, Roberts C, Linsley P S, Bernards R, Friend S H. Gene expression profiling predicts clinical outcome of breast cancer. Nature 2002; 415(6871):530-536.
(6) Monni O, Hyman E, Mousses S, Barlund M, Kallioniemi A, Kallioniemi O P. From chromosomal alterations to target genes for therapy: integrating cytogenetic and functional genomic views of the breast cancer genome. Semin Cancer Biol 2001; 11(5):395-401.
(7) Mousses S, Wagner U, Chen Y, Kim J W, Bubendorf L, Bittner M, Pretlow T, Elkahloun A G, Trepel J B, Kallioniemi O P. Failure of hormone therapy in prostate cancer involves systematic restoration of androgen responsive genes and activation of rapamycin sensitive signaling. Oncogene 2001; 20(46):6718-6723.
(8) Quarmby S, West C, Magee B, Stewart A, Hunter R, Kumar S. Differential expression of cytokine genes in fibroblasts derived from skin biopsies of patients who developed minimal or severe normal tissue damage after radiotherapy. Radiat Res 2002; 157(3):243-248.
(9) Rew D A. DNA microarray technology in cancer research. Eur J Surg Oncol 2001; 27(5):504-508.
(10) Simpson R J, Dorow D S. Cancer proteomics: from signaling networks to tumor markers. Trends in Biotechnology 2001; 19:S40-S48.
(11) Liggett S B, Caron M G, Lefkowitz R J, Hnatowich M. Coupling of a mutated form of the human beta 2-adrenergic receptor to Gi and Gs. Requirement for multiple cytoplasmic domains in the coupling process. J Biol Chem 1991; 266:4816-4821.
(12) Tsuji A, Sato Y, Hirano M, Suga T, Koshimoto H, Taguchi T, Ohsuka S. Development of a time-resolved fluorometric method for observing hybridization in living cells using fluorescence resonance energy transfer. Biophys J 2001; 81(1):501-515.
(13) Liu X, Tan W. A fiber-optic evanescent wave DNA biosensor based on novel molecular beacons. Anal Chem 1999; 71(22):5054-5059.
(14) Zhuang X, Bartley L E, Babcock H P, Russell R, Ha T, Herschlag D, Chu S. A single-molecule study of RNA catalysis and folding. Science 2000; 288(5473):2048-2051.
(15) Lakowicz J R. Principles of fluroescence spectroscopy. second ed. New York: Kluwer Academic/Plenum Publishers, 1999.
(16) Tyagi S, Kramer F R. Molecular beacons: probes that fluoresce upon hybridization. Nat Biotechnol 1996; 14:303-308.
(17) Basile, A., A. Giuliani, G. Pirri, and M. Chiari. 2002. Use of peptide nucleic acid probes for detecting DNA single-base mutations by capillary electrophoresis. Electrophoresis 23:926-929.
(18) Chen, C., Y. K. Hong, S. D. Ontiveros, M. Egholm, and W. M. Strauss. 1999. Single base discrimination of CENP-B repeats on mouse and human Chromosomes with PNA-FISH. Mamm. Genome 10:13-18.
(19) Hirano, H., H. Kawasaki, and H. Sassa. 2003. Two-dimensional gel electrophoresis using immobilized pH gradient tube gels. Electrophoresis 21:440-445.
(20) Jansen, K. and E. Richelson. 2000. Detection of peptide nucleic acids in tissue extracts of treated animals by gel mobility shift assay. J. Biochem. Biophys. Methods 42:31-34.
(21) Kim, D. H., Y. K. Hong, M. Egholm, and W. M. Strauss. 2001. Non-disruptive PNA-FISH protocol for formalin-fixed and paraffin-embedded tissue sections. BioTechniques 31:472, 475-472, 476.
(22) Pokidysheva, E. N., I. A. Maklakova, Z. M. Belomestnaya, N. V. Perova, S. N. Bagrov, and V. I. Sevastianov. 2001. Comparative analysis of human serum albumin adsorption and complement activation for intraocular lenses. Artivicial Organs 25:453-458.
(23) Pokidysheva, E. N., I. A. Maklakova, Z. M. Belomestnaya, N. V. Perova, S. N. Bagrov, and V. I. Sevastianov. 2003. Comparative analysis of human serum albumin adsorption and complement activation for intraocular lenses. Artif. Organs 25:453-458.
(25) Ray, R. and B. Norden. 2000. Peptide nucleic acid (PNA): its medical and biotechnical applications and promise for the future. FASEB J. 14:1041-1060.
(26) Zuo, X. and D. W. Speicher. 2002. Comprehensive analysis of complex proteomes using microscale solution isoelectrofocusing prior to narrow pH range two-dimensional electrophoresis. Proteomics 2:58-68.
In another embodiment, the present invention provides hydrogels for separating bound and free analytes and their optical analysis. More specifically, the invention provides a modification of a sensor that 1) enables one to employ higher voltages without the danger of losing the sample, 2) has a more easily used electrophoresis chamber, 3) is a more convenient illumination device for analysis, 4) enables the sensor for micro-array analysis, 5) enables the sensor to be used as a replacement for qRT-PCR, and 5) permits new rapid clinical uses such as for pap smears, diagnosis of sexually transmitted diseases, diagnoses of skin cancers, diagnosis of oral cancers and monitoring lymphocytes.
In one embodiment, the invention provides a “first plate” wherein an illumination hydrogel (IH) now contains both an analytical surface and a focusing component that causes light to pass across the analytical surface in a total internal reflection mode. This is a significant development in which these components are separate. This modification simplifies the analysis and is very user friendly. The phrase “first plate” now acts as a “sensor platform.” In the present invention, this is the hydrogel layer that is closest to the fluorescence detection device and the invention provides a method in which one can build the sensor platform with a very thin analytical surface and can also be made such that parts are derived from different materials. The novel redesigned “first plate” into an illumination hydrogel (IH) having an optical frame (OF) and analytical surface (AS) provides several benefits. It insures that the optical components of the device are aligned with its analytical components. This eliminates potential problems that a user would have in alignment and gives the manufacturer greater ability to control the quality and reproducibility of the analysis. It creates a “buffer” chamber that simplifies the electrophoresis step. The buffer chamber also facilitates stacking and alignment of other hydrogels used in analysis, thereby making the sensor more versatile. These can be added by the user as required for different assay configurations simply by placing them in the buffer chamber. The redesign also makes it easier for the manufacturer to coat the analytical surface with reagents used for array analyses.
In a second embodiment, the invention provides a “first plate” that can be charged, a factor that will prevent or reduce over-electrophoresis, a phenomenon that occurs when one runs the gel too far. For example, a charged first plate would act in the same way as an ion exchange resin to trap the charged analyte. In the case of a nucleic acid analyte, which has a negative charge, the first plate should be positive. This can be made by adding different hydrogel components that have an appropriate charge as outlined in the specification. The advantage of this embodiment is that one can make the first plate very thin, which puts it close to the fiber optic or other detection device.
In a third embodiment, the invention provides a sensor can be used in an array format. This can be done by applying the sample in different wells, by applying the detection reagent in different wells, or by applying the detection reagent to the first plate or a second plate that is very close to the first plate using a reversible bond. Non-limiting examples include a disulfide bond, a pH sensitive bond, and the like.
In a fourth embodiment, the invention provides for the use of a pH sensitive bond. This makes it possible to attach detection reagents to gels of various charges. There are several key aspects here. It is a very convenient way to make the array; it is easily disrupted by altering the pH and should be fully reversible; one can make the surface negatively charged, which in the case of DNA and RNA would prevent or reduce non-specific binding; when reversed, the PNA-fluorophore will be positively charged and will migrate away from the surface if the surface originally has a small positive charge (many —NH2 groups). For example a drug containing an amine group can be covalently bonded to a polymer containing an anhydride group at pH>8 and released from the polymer at pH<7. This embodiment may be used to provide a device such as a test strip or a simple dot like assay on a strip.
In a fifth embodiment, the invention provides for a new design of the electrophoresis chamber. A bubble deflector is provided which makes it possible to do the entire electrophoresis in a simple manner. Before, one had to slide the sensor components into a box, run the electrophoresis in an upright position, and take the device out of the box and put it on the detector that was to keep the bubbles away. The new design prevents bubble interference automatically. Moreover the design of the first plate now makes it possible to add buffer to the sensor platform itself
In a sixth embodiment, the invention provides for the design of the measurement chamber. Combined with the sensor platform makes carrying out the assay easy. One may use a thin Teflon-AF layer as a very low refractive index medium to keep the fiber optic from inhibiting the total internal reflection and for protecting the fiber optic. The box itself is designed to keep the background to an absolute minimum so that single molecules can be detected.
In addition, a negatively charged detection reagent that has a cleavable bond that removes the element that enables it to be bound to the surface by its charge. This would cause it to become positively charged and leave the surface by washing or in the presence of an electric field. An example of this would be a tiny Dowex bead. This might also facilitate fabrication because one would only need to add the bead to the surface. One could add a large amount of detection reagent in a very small spot. The analyte-detection reagent complex would stick to the surface due to its charge and therefore be easily monitored. Alternatively, one could a use a detection reagent that had a pH sensitive element, e.g., a poly-his tail. At high pH, it would adhere to an NTA resin on the surface even though the overall charge of the material is positive. Lowering the pH would make it dissociate unless it was bound to a nucleic acid analyte.
Complexes formed by the binding of the analyte to detection reagents are separated from the non-bound detection reagents by differences in the charges of the bound and free detection reagents. There are many methods that can be used to do this, including the use of ion-exchange materials, e.g., papers, films, membranes or beads coated with negative or positively charged groups, or by electrophoresis. When the detection reagent is fluorescent, a preferred method for separation involves the use electrophoresis on hydrogels wherein the fluorescence of the bound detection reagent can be observed directly and quantitatively. Due to the low background fluorescence when the approach is employed using the abilities of hydrogels to be used for total internal reflection (TIR), this approach is highly sensitive and can permit the detection of a single molecule of analyte and quantification of large numbers of analyte molecules. The use of electrophoretic methods for separation can be improved further by employing a hydrogel that can trap the free detection reagent or preferably the analyte-detection complex. The latter is preferred because it enables one to monitor analyte under conditions having little or no fluorescent background. The use of a hydrogel having a charge opposite that of the analyte-detection complex (
As noted above, a positively charged hydrogel can be employed to prevent a negatively charged sample (e.g., a nucleic acid) from running through the IH. A working sensor can be prepared in which the only component is an IH or other light transmitting material, e.g., a porous glass such as Vycor. However, for many analyses a preferred arrangement consists of one or more gel layers in addition to the IH (
It should also be noted that any additional gel layers that are used in the device do not need to be fabricated at the same time as the IH. Indeed, additional gels can be added to the IH or other gel layers at any time. This makes the device much more flexible and as will be discussed, this procedure can be used to fabricate an IH that is very thin.
When one is analyzing a negatively charged analyte, e.g., RNA, one employs a positively charged detection reagent such as a PNA with an attached fluorophore. The PNA, which is the part of the detection reagent that has high specificity and affinity for specific nucleic acid sequences is coupled to a fluorophore and either the PNA or the fluorophore component is made to have one or two positive charges, e.g., a guanidinium group. The detection reagent contains many fewer positive charges than the number of negative charges on the nucleic acid analyte to which it will bind. As the positively charged detection reagent migrates towards the negative electrode it encounters the negatively charged analyte, which is migrating in the opposite direction towards the positively charged electrode. Interactions between the analyte and detection reagent cause the formation of the analyte-detection reagent complex. Since the number of negative charges in the analyte exceeds the number of positive charges in the detection reagent, the overall charge of the complex that is formed is negative and the complex migrates towards the positively charged hydrogel away from the non-bound detection reagent. This permits the use of a detection reagent excess; a condition that insures all the analyte will become labeled. When the analyte-detection reagent complexes encounter the positively charged hydrogel, their migration slows or ceases. Use of a positively charged hydrogel is particularly useful for analytes that are small and that have high mobility, e.g., micro RNA, since it prevents them from running through the IH. It should be noted, however, that the presence of an excess charge in the hydrogel will tend to increase the amount of water in the hydrogel due to the fact that the charges tend to repel one another, often making the gel swell. This will reduce the refractive index of a poorly crosslinked IH, a phenomenon that has the potential to interfere with steps in the analysis that depend on TIR. This problem can be eliminated by increasing the concentration of reagents used to fabricate and/or crosslink the IH as well as keeping its charge density as low as necessary to prevent sample loss.
The differences in charge between the analyte-detection complex and the non-bound detection reagent also enable the separation to be done by permitting the complex to adhere to a membrane or other surface that is oppositely charged. For example, it is possible to detect and quantify a protein that is positively charged at the pH being employed using a fluorescent detection reagent, e.g., a fluorescent antibody—that is or has been engineered to have a weak opposite charge such that the protein-antibody complex that forms had a different charge than the free detection reagent at the pH being used. Separation of the bound and unbound detection reagent would involve monitoring the material that is bound to the surface. A potential problem with this type of assay is the tendency for it to have a high background. In the preferred assay, the oppositely charged reagents are separated using an electric field. Thus, if the complex were stable at the pH used, the material used to trap the analyte-detection complex could be subjected to an electric field to reduce the background. In principle, the complex could also be stabilized by reacting it with a crosslinking reagent before application of the electric field. Clearly, this type of procedure is more complex than the preferred procedure that involves the use of hydrogels and electrophoresis.
In principle, the use of charge differences between the analyte-detection complex and the non-bound detection reagent could also be done using any charged surface that would bind a charged complex of analyte-detection reagent differently than the free detection reagent. Furthermore, it would be possible to employ amplification strategies wherein the detection reagent is tethered to an enzyme or other reagent that could be used to measure the detection reagent-analyte complex. Signal amplification has been to measure material bound to microtiter plates or other surfaces such as membranes used in Western Blotting. Because these assays are indirect however, low concentrations of analyte are more difficult to detect and quantify accurately. These problems are avoided using a method in which binding measurements are made directly, e.g., without additional amplification. For this reason, the use of electrophoretic separation using hydrogels that can be illuminated by total internal reflection (TIR) coupled with fluorescence (total internal reflection fluorescence, i.e., TIRF) is preferred. Furthermore, when the electrophoretic separation is done using thin hydrogels, the analysis can be very rapid, easily quantified, have little background, and applied to tissue sections, soluble samples, or arrays of samples. Direct assays that do not require amplification steps to detect an analyte are also more rapid than the common assays used to detect RNA by quantitative reverse transcript polymerase chain reaction assays (qRT-PCR), which involves several steps, requires many controls, and is more complex.
Use of this method enables one to obtain spatial resolution, thereby permitting analytes expressed in one tissue to be quantified separately from those that are expressed in an adjacent tissue. This technology was developed to monitor gene expression products in tissue sections, biopsies, cultured cells, bacteria, and other related samples. Use of any technique that depends on fluorophores is hampered by the fact that there are a limited number of fluorophores having colors that can be readily distinguished. This potential limitation can be circumvented in part using multiple excitation sources, filters, photobleaching, fluorescence resonance energy transfer, and fluorophores that have different lifetimes, but it is often difficult to circumvent completely. When using this technique to analyze biopsies or to observe samples taken at the time of surgery to determine the most appropriate treatment regimen, it is possible to use the same fluorophore for several analytes. This enables analyses that can provide yes or no answers. For example, in the cases where one wants to determine if a tumor is expressing one or more oncogenes or has stopped expressing one or more suppressor genes, it is possible to monitor several gene expression products simultaneously using a mixture of PNAs labeled with the same fluorophore to monitor a few oncogene products and a mixture of PNAs labeled with a different fluorophore to monitor a monitor a few suppressor gene products. As a control, one would monitor the expression of one or more gene products that are known not to change using a PNA labeled with a third fluorophore. These analyses could be done simultaneously on a tissue section or biopsy sample using a mixture of all the PNA reagents. The three fluorophores would be quantified by exciting them using different lasers and using filters to distinguish the fluorescent colors, by using photobleaching, and/or by monitoring their fluorescence lifetimes. An elevated ratio of oncogene expression products relative to that of suppressor gene products compared with the control gene products would suggest that further analyses be considered. Likewise, at the time of surgery, the finding that a gene that is associated with metastasis is expressed in a tumor would suggest a need for further possible surgery, radiation, or chemotherapy whereas its absence would indicate that such additional treatment might do more harm than good.
Other significant uses for this type of rapid combined assay include the analysis of pap smears in which papilloma viruses that are associated with cervical cancer are monitored simultaneously with mixtures of PNA that are labeled with a single fluorophore to distinguish those papilloma viruses that are associated cervical carcinoma from others that are unlikely to cause cancer. This would enable one to detect one or more papilloma viruses with a single assay and the presence of any of these virus strains would be cause for additional testing. Using a mixture of different colored fluorophores, it would also be possible to simultaneously test for the presence of gene expression products or their mutants, e.g., those from myc, ras, or other oncogenes—that are associated with carcinomas or pre-carcinomas. These tests could be done in a doctor's office at the time of the patient visit, thereby reducing anxiety caused by the long time for the current histological analysis. Cytological assays, the current method of performing these tests, are not very accurate and often result in additional patient visits, a phenomenon that is very stressful for many women. Using a mixture of PNA labeled with a different fluorophore could also be done simultaneously to detect the presence of microorganisms associated with sexually transmitted diseases. The rapidity and sensitivity of the assay technique described here would also facilitate the diagnoses of skin cancer observed by dermatologists as well as potential cancers of the mouth seen by dentists. These analyses are also usually done by histology. The rapid accurate measurement of gene expression products in tissues taken by the physician in the office would reduce the time needed for analysis and improve diagnostic accuracy significantly.
Still another important use for this technique involves research studies wherein one seeks to observe and quantify specific differences in tissue expression, e.g., during development—to identify stem cells, changes in stem cells, cancer stem cells, or to identify potential interactions between cells. The ability to detect gene expression products in tissue samples is useful for learning which genes are expressed in individual cells of tissue sections. The technique would enable one to monitor three or four gene expression products to be monitored simultaneously in multiple tissue sections. This would be particularly useful for analyzing any gene expression product including micro-RNA (miRNA) in tissue sections. Due to their short lengths miRNA are difficult to quantify by other methods and their analysis usually requires several steps, a phenomenon that reduces the accuracy of measurement.
As noted in Example 18, it is difficult to use a large number of fluorophores simultaneously for multiple samples. Even the use of nano-dots would not circumvent this problem during efforts to detect and quantify as few as a dozen or more individual analytes at the same time. This limitation can be circumvented using array methods, which permit one to quantify multiple gene products simultaneously using only one or a few different fluorophores. Array methods can be used to detect and quantify the same analyte in multiple samples as well as to detect and quantify multiple analytes in the same sample. There are several ways to do this. For example, to simplify manufacturing and use wherein a user will be analyzing or comparing the same analyte in clinical samples from many patients, e.g., in a drop of blood, it would be desirable to use a sensor device in which the detection reagent has been incorporated uniformly into the IH or other hydrogel. This insures that a uniform amount of detection reagent is present throughout the hydrogel. This could be done while the gel is being polymerized or, since the detection reagent is charged; it could be accomplished using electrophoresis to cause it to penetrate into the gel after the gel has been polymerized. This “preloaded” hydrogel can be of high refractive index, e.g., the gel to be used as the IH, or low refractive index. If it is of low refractive index, it is placed adjacent to the IH on the surface that will be distant from the FO. If the “preloaded” hydrogel is of the same or higher refractive index as the IH, it is separated from the IH by a placing a low refractive index hydrogel on the surface that will be distant from the FO. The thickness of this “separating” hydrogel must be sufficient to guarantee that the IH retains its ability to be illuminated by total internal reflection. For practical reasons, a useful thickness of this is 100 micrometers or more although it can be as little as a few micrometers. The samples to be analyzed can be separated by any convenient method, including spotting them on different sites on the hydrogel that has been loaded with detection reagent or preferably by placing them into wells (
The procedure just described is most useful when one wants to analyze multiple samples of the same analyte since it is possible to control the concentration of detection reagent in the gel at the time of manufacture. This also has the potential of reducing human operator error or of being automated using robotics. For many assays in which many different analytes are to be monitored, however, another approach is more useful. As noted earlier, it is difficult to use more than a few multiple fluorophores simultaneously due to the overlaps in their excitation and fluorescence emission spectra. Thus, while the approach just described is useful for the commercial production of hydrogels that are labeled with one or a few detection reagents, a more general approach is often preferred. In this approach, one mixes the analyte with the detection reagent before analyzing the mixture and adds the mixture directly to the gel, or more preferably to a sample gel that has multiple wells (
This type of analysis has considerable advantages over the use of techniques such as quantitative reverse transcription polymerase chain reaction (qRT-PCR) in which an RNA sample is converted to DNA by the use of reverse transcriptase and the sample is then quantified by PCR. qRT-PCR requires several internal controls and takes longer. Furthermore, since it is not necessary to use reverse transcriptase for analyzing RNA samples using the sensor, it provides a rapid direct quantitative analysis of individual RNAs in the sample.
The approach described in Example 18 illustrates methods for simultaneously analyzing a small number of gene expression products or other analytes in the same or different samples. However, an alternate approach is required for simultaneous identification and quantification thousands of gene expression products within a complex mixture such a tissue or cell extract. Due to the fact that tissues usually contain complex mixtures of cells, it would be desirable to use a procedure that could be performed with a small amount of sample to maximize the likelihood that one could distinguish differences in various tissue components.
In this assay format the sample is mixed with the array to permit binding of each analyte to the tethered detection reagent and it may be difficult to insure good mixing between the array when very small sample volumes are employed. Because a few different fluorophores can be used to label different detection reagents, it is possible to prepare a arrays in which the density of the detection reagents is high. Differences in fluorescent color or lifetime seen during analysis would permit analytes that are adjacent to be resolved since they would be distinguished by their colors or lifetimes. The use of high density arrays minimizes the physical size of the array, which would enable one to have good mixing during the binding phase, especially when the sample size is small. Furthermore, since this type of array analysis does not require analytes such as RNA to be treated other than to be released from the tissue, it is expected to be much more accurate than assays in which RNA in the sample is converted to DNA and amplified. When materials from two different tissues or two different patients are to be compared, the ratio of fluorescence in each spot of the two or more samples will reveal differences in gene expression between the samples.
There are many ways that this type of analysis can be preformed. One preferred way to do this is to apply a thin coating of a low refractive index hydrogel to the surface of the IH, e.g., amino activated agarose (Pierce). The chemistry of amino activated agarose enables it to form a covalent bond with nearly any reagent that has a free amine. When treated with cysteine or another amine containing thiol compound, the amino activated agarose coating will react with the amine group of cysteine or the other reagent and thereby create a surface that is coated with free thiol (SH) groups. Under mild oxidizing conditions these will form a disulfide bond with thiol groups that are incorporated into PNA. The location of the SH group in the PNA, e.g., at one end of the molecule—would be chosen such that it did not interfere with its ability to bind nucleic acids with high specificity and affinity. When the thiolated PNA detection reagent is added to different spots on the thiolated surface under oxidizing conditions, the detection reagent will become covalently bound to the surface by a disulfide bond. This leads to a grid-like array of detection reagents having a specificity for different nucleic acid analytes in the sample. To eliminate DNA in the sample, which would interfere with analysis of RNA, one can treat it with RNase free DNase before incubating the sample with the matrix. The array is permitted to interact with the analyte to permit the gene expression product to bind to the detection reagent, a phenomenon that can be facilitated by electrophoresis of the sample in multiple directions, rocking the grid from side to side, using gentle sonication, or any other technique that permits the mixture of analytes in the sample to interact with detection reagents bound to the array. Non-specific binding, if any, can be reduced by subjecting the gel to electrophoresis at an increased temperature during which weakly bound RNA will migrate away from the IH and away from the grid towards a positive electrode. It is also possible to eliminate non-specific RNA binding using RNase or S1 nuclease, which will cleave regions of RNA that are not hybridized to the PNA. Following the binding reaction, during which specific nucleic acids in the sample, e.g., miRNA or RNA, bind to specific known predetermined detection reagent on the surface and after any non-specific interactions are reduced or eliminated, the polarity of the electrodes is reversed such that the detection-reagent complexes migrate towards the array and the IH in the presence of an excess amount of reducing agent to disrupt the disulfide bond that covalently attaches the detection reagent to the array while continuing the electrophoresis. This will cause non-bound detection reagent to migrate towards the negative electrode and the analyte-detection reagent complex to migrate towards the positive electrode where it will enter the IH or adhere to its surface. A control for the ability of the reducing agent to remove a disulfide bound positive fluorophore would consist of a fluorescently labeled PNA that is unable to bind RNA. If cleavage is complete, all the fluorescence at this site should be lost during analysis. In contrast, the fluorescence of a fluorophore that is linked to the matrix by a method that does not involve disulfide bond should remain constant. If the first control remains fluorescent, one would subject the IH to a second round of electrophoresis in a reducing buffer. As will be noted later, the optical device to be used in this analysis is designed to permit repeated electrophoresis and fluorescence analysis.
During this analysis, the analog-detection reagent complex migrates to the IH and becomes located within the IH or immediately adjacent to its surface. Since the analysis is designed such that the complex will migrate only a very short distance, there will be little or no diffusion that will cause spot widening. In contrast migration of the smaller non-bound detection reagent away from the IH should be fast and complete. Since fluorescence will be observed only when the analyte-detection reagent complex is within the IH or on its surface, migration of the unbound detection reagent of only several micrometers will be sufficient to make non-bound detection reagent invisible during analysis using an IH in which light is transmitted through its AS by TIR.
Following reduction and electrophoresis, one monitors fluorescence of the HI using TIRF. The advantage of this approach is that it is direct, quantitative, and fast. A problem with current array analyses methods, which are indirect, is that they are often unable to detect less than a 50% change in the level of expression of genes in a sample. This is a significant issue when one is attempting to quantify gene expression in tumors, during development, or in any instance when one is searching for quantitative differences that can be used for clinical diagnosis or for research purposes.
A major advantage of the methods outlined here is that they provide very sensitive assays in which one does not need to label an analyte before analysis. There are several ways this can be used to take advantage of the methods of analyte identification and quantification taught here. For example, it is possible to create free detection reagents that have or can be made to have charges different from complexes formed by analyte-detection reagent complexes. Indeed, it would be expected that any means by which the charge of the analyte-bound detection reagent and the non-bound detection reagent could be made to differ upon binding of an analyte would enable it to be identified and quantified. Here I illustrate how this principle could be used when a detection reagent is modified after it has been used to bind an analyte. This is particularly useful for monitoring analytes that become bound to surface-bound detection reagents, including those in arrays or microarrays as exemplified using any pH sensitive reagent, e.g., dimethylmaleic anhydride (DMMAn), that could be used to covalently link a detection reagent to a surface. DMMAn, which was used to protect amines during peptide synthesis, was developed many years ago and has been shown to become attached to DMMAn when it is co-polymerized with vinylpyrrolidone (VPD). The DMMAn in the DMMAn-VPD surface would bind amine-containing PNA fluorescent detection reagents covalently at pH8 or above. Formation of the covalent bond eliminates the positive charge of the amine in the detection reagent. When the pH is reduced to 6, a change that breaks the pH-sensitive covalent bond between the DMMAn and the amine in the detection reagent, the detection reagent would be released and its positive charge restored. When a membrane that contains a fluorescent PNA detection reagent bound to its surface through a DMMAn-amine bond at pH8 or above is incubated with a solution that contains nucleic acid analytes, the PNA portion of the detection reagent will bind the analytes with high affinity and specificity. When the pH is reduced to 6, all the amine labeled PNA-fluorophores will be released from the membrane. Those that have formed a complex with a negatively charged nucleic acid analyte will be negatively charged whereas those that did not form such a complex will be positively charged. During electrophoresis using an IH that was either neutral or positively charged, the non-bound analyte that became positively charged after it had been released from the surface would migrate towards the negatively charged electrode. The complex of nucleic acid analyte and detection reagent, which would have an overall negative charge, would migrate towards the positively charged electrode and become separated from the positively charged non-bound detection reagent. This will enable the analytes to be analyzed and quantified by the fluorescence of the bound complex using TIRF. Furthermore, if the overall charge of the PVD membrane were positive at low pH, repulsive ionic interactions between the membrane and the positively charged non-bound detection reagent would facilitate its removal from the membrane during washing steps. In contrast, ionic attraction between the positive charge on the membrane and the negative charge of the analyte would facilitate binding of the analyte bound detection reagent with the membrane. This would permit the detection, identification, and quantification of specific analytes in the material being analyzed. Thus, this type of procedure also has the potential for use in a dip-stick approach to nucleic acid identification and quantification. Although an example of this procedure has been described only for analyses of nucleic acids, it is adaptable to any analyte that has a charge at the pH being used for analysis provided that a detection reagent can have or be made to have a different charge under the analysis conditions. It should also be noted that in the case of nucleic acid analyses, both RNA and DNA will bind to PNA. Prior to analysis, these analytes can be distinguished by treating them with DNase or RNase, respectively, both of which are commercially available.
This approach can be extended further by using the DMMAn to alter the surface charge of a membrane as well as to tether the detection reagent. When negatively charged amine containing compounds and amine containing fluorescent PNA detection reagents are attached to a positively charged surface that contains DMMAn, the surface would become negatively charged and acquire the ability to bind specific nucleic acid analytes. The ionic repulsion between the negatively charged surface and negatively charged nucleic acid analytes would reduce or prevent non-specific interactions between them. It would not prevent the very strong interactions between the nucleic acids and the tethered PNA, however, to which the analytes would bind with high specificity. Reducing or preventing non-specific binding of nucleic acids to the surface would be expected to facilitate the analysis of very small mixtures of nucleic acids by reducing non-specific interactions between the nucleic acids and the surface thereby enabling analytes to be captured by the PNA. At low pH, the negative charges on the surface would be released, restoring its positive charge. The low pH would also cause the release of all the detection reagents. Those detection reagents that had formed a complex with the negatively charged analyte would remain bound by ionic interactions to the newly positively charged surface; detection reagents that had not bound analyte would be positively charged and repelled from the surface. This would facilitate the separation of analyte-detection reagent complexes from free detection reagents in a single step, i.e., altering the pH. Use of this procedure in an electric field during electrophoresis and an IH gel would facilitate separation of analyte detection reagent complexes and non-bound detection reagent and facilitate identification and quantification of specific analytes. It would also facilitate analyses based on dip-stick approaches.
Still another approach would be to attach negatively charged groups to the detection reagent by a pH sensitive bond, a disulfide bond, or other cleavable bond. When these bonds are broken, the free detection reagent would become positively charged whereas the analyte bound detection reagent would remain negatively charged due to its specific binding of a nucleic acid analyte. This would enable the separation of complexes formed between analyte bound detection reagent and the free detection reagent. An advantage of this approach is that it would enable detection reagents to be spotted on a positively charged surface and to be held there by ionic interactions until the pH is lowered.
A preferred illumination system is one in which a charged or uncharged hydrogel serves as both the electrophoresis device and as an illumination device. Examples of principles that can be used to form such “illumination hydrogels” are illustrated in
The use of hydrogels having equal or higher optical density to the IH can be used in the apparatus, but it would be necessary to remove them if one were to use internal reflection during analysis. It would be obvious that removing these gels would permit the IH to be illuminated by total internal reflection or another illumination system. However, this would require an additional step, i.e., removing the other hydrogels. Moreover, the use of a different illuminating system would have the tendency to create a higher background due to the need to take greater steps to filter the excitation light energy, a phenomenon that total internal reflection reduces to a minimum, i.e., due to incorrect laser alignment or light scattering. Light scattering can be minimized by using IH that do not contain particulate matter contamination.
A convenient chamber that is used for analysis contains four illumination regions and is either square (
Multiple arrangements of laser light sources can be used to excite the fluorophores in the complexes formed by the analyte(s) and detection reagent(s) (
An alternate approach to illuminating the IH is to employ a laser that has its output attached to a single mode fiber (
The IH is used 1) for separation of analyte-detection compound complexes from free detection reagent and 2) for focusing light onto an analytical surface by TIR. Due to the high sensitivity of fluorescence techniques and their established uses for single molecule detection, they are preferred for detecting and quantifying complexes of the analyte bound to its detection reagent. To collect the most fluorescent light and thereby achieve the highest sensitivity, the preferred IH is designed such that the analyte-detection reagent complexes become located near the device used to quantify the emitted fluorescent light, e.g., a fiber optic (FO) coupled directly to a sensitive camera or an objective lens (OL) attached to a microscope and a sensitive camera. In a preferred design of the sensor, the component used for electrophoretic separation is also used for fluorescent analysis.
These dual functions of the primary sensor component can met by fabricating it from hydrogels or any material that is capable of transmitting fluorescent light and that can be used for separating the free and bound complex. While in principle, this component could also be fabricated from porous Vycor glass, it is more likely to be a molded reusable or disposable hydrogel. When the primary sensor component is a hydrogel that is used for both electrophoresis and optical measurements, it is termed the illumination hydrogel (IH). The IH can be fabricated by any well-established technique such as one to make optical hydrogels for use as contact lenses. As diagrammed in
Preparation of a mold used to fabricate the IH. There are many ways to prepare an IH, but one that worked satisfactorily is to make it using a mold. A procedure to make an IH mold with a minimum of tools is described here. To make a mold that yields a square IH, one cuts a rod or tube of known diameter, e.g., between 1 and 6 mm—into 4 equal lengths. In the case of a 6 mm diameter rod, the length used was approximately 49 mm. Knowledge of the rod diameter enables one to calculate the distance between the OF and AS needed for the circular surface of the OF to fabricate a hydrogel of known optical density that will focus parallel light rays on the AS. These calculations depend on the radius of the rod and the refractive index of the IH that is to be used and are well known in the art of optics. While it might be best to use an optical surface that has a non-circular curvature, a rod having a spherical diameter has been found to work well enough for fabrication of an IH that can be used to illuminate all but very thin AF by TIR.
Four equal length pieces of the circular rod were glued end to side as shown in
The second IH mold component was prepared as shown in
Next, the first hydrogel and its attached components were placed into a box that had been fabricated from plywood that had been covered with a thin layer of masking tape (
When assembled together, the two components produced by this molding process created a void that was the shape of the IH. The last step in building the mold was to create the filling groove in a corner of the first mold segment at a site comparable to that of indicated region in
The non-transparent gray colored epoxy mold that was created to test the design parameters of the IH can be used to prepare the analytical hydrogels needed for operation of the sensor. One of the features of this mold is the position of the interface between its two components. The mold was designed to permit the molded IH, i.e., the uncolored space within the mold of
There are several obvious improvements that can be made in the IH that would facilitate its use and manufacture. First, the illuminated surface of the device designed here is circular, which was satisfactory for use with the thickness of the AS in the device created here. However, it would be possible to create a thinner AS that is illuminated uniformly by altering the curvature of the OF slightly using standard optical calculations by persons or ordinary skill in the art of mold design and manufacture. For high throughput manufacture that would be required for commercial use, it would be preferable to use a transparent mold that would facilitate quality control by enabling the processes of filling the mold to be observed, e.g., to insure that bubbles were not present and that the hydrogel was uniform. Furthermore, it would be much better to use a non-stick surface such as a thin coating of Teflon or Teflon-AF on the surface of the mold components that contact the hydrogel to facilitate removal of the hydrogel from the mold. Another obvious improvement would be the fabrication of a rigid plastic holder to facilitate handling of the IH, particularly as the AS is made thinner. An alternative to this is to use a flexible polyethylene holder such as that described later for holding the IH and other hydrogels during electrophoresis and fluorescence measurements.
As the AS is made thinner to increase the sensitivity and speed of analysis, it would become more difficult to mold the AS uniformly and to remove it from mold without tearing it. This problem can be solved by fabricating the IH in two steps; the first of which would be to create the OF and the second would be to add the AS. This can be done by placing the molded OF on a lower refractive index hydrogel, e.g., a hydrogel that often serves to hold the analyte before analysis or to separate other hydrogels from the IH or on a Teflon block. Either of these would serve as a “template” for the formation of the AS (
Nearly any device capable of performing electrophoresis can be used for separating the analyte-detection reagent complex from the non-bound detection reagent. However, some envisioned uses of the device will be occur in clinical settings during routine office visits for biopsies by physicians, dentists, nurses, and others who are not skilled in laboratory techniques. The device could also be used for a variety of other uses—e.g., as a replacement for qRT-PCR quantification of ribonucleic acid expression in tissues and as an alternative to current methods of microarray analysis. Therefore, a simple electrophoresis chamber that was designed specifically for these and other types of analyses would be preferred. The electrophoresis system outlined here, which is designed to work with electrophoretic separation and optical measurements, requires little end-user knowledge of sample handling, electrophoresis, or fluorescence analysis.
The preferred electrophoresis chamber permits this operation to occur in a vertical format, which makes it simpler for a user since it enables the user to transfer the IH and items in its buffer well to the chamber while they are in a flexible handler. Thus, there is no need for the user to handle the IH once it is in the flexible handler. Furthermore, there is little chance of leaks since the lower chamber can be fabricated from one piece of transparent plastic. Buffer is added to the buffer well of the IH a component that is one piece. There are no screws, gaskets, or other devices required and the handler and the associated IH cannot be lowered beyond the appropriate spot in the lower buffer chamber by the presence of the shelf.
As described here, many uses of the sensor will be for applications performed using a fiber optic (FO), an objective lens (OL), or both in the fashion commonly used in an endoscope. These will deliver light to a highly sensitive camera (e.g., a CCD or an EMCCD) or photomultiplier tube (PMT). This format enables the analysis to be performed over an area viewed through the fiber optic at the limit of detection of the camera. When a camera of sufficient sensitivity us employed, e.g., a CCD or EMCCD, the limit of detection will be single molecules. The major advantage of using a FO instead of a microscope and its objective lens is that the FO enables a much larger area to be observed without moving the sample or the camera. An objective lens offers greater resolution and is well suited for analyses in which higher resolution is required. The PMT is well suited to detection of photons and, in some enablements, can also give rise to an image.
The preferred design of the device for monitoring fluorescence (
The major difficulty expected during the measurement of samples having low fluorescence with a very sensitive light detection system will be caused by stray light that can be derived from extraneous sources, e.g., room light, or from light that passes through the sensor and bounces off the interior of the analysis chamber. This has the potential to increase the background, an unwanted problem when single molecule sensitivity is desired. For this reason the measurements will be made in a box that blocks the entry of light, e.g., a box fabricated from Deldrin, a black plastic that is easily machined and that prevents the passage of light. The inside of the box would also be coated with black felt or other rough light absorbing material that will reduce light reflection off the smooth surface of Deldrin. The post that will be used to stabilize the FO and to create a surface that aligns the IH should also be fabricated from a material such as Deldrin to reduce stray light. By causing the OF to be illuminated through an adjustable slit mounted on part of the lid used to cover the measuring device, the light that reaches the OF can be adjusted. A filter can be placed in front or behind this slit that blocks light wavelengths other than those being used to illuminate the IH. This will prevent most light from outside the illumination box from entering.
Normally, since a FO or OL will be used for collection of fluorescent light, only those photons that pass through the AS will be at an angel that permits them to reach the camera. These can be prevented from entering the camera using a black absorbing plastic “shield” that is the same shape as the buffer chamber in the IH. This will prevent photons from reaching the camera by passing through the AS and would be particularly effective when analysis is conducted in a light tight box. Another source of background illumination is the light that passes through the IH and that is emitted within the box or that is reflected off the front surface of the OF used to focus the light onto the AS. Much if not all of this will be prevented by the box or the black plastic used to replace the buffer in the IH. However, in the unlikely event that this does not reduce the background sufficiently to permit single fluorophore analysis, this light can be reduced or eliminated by placing a light absorbing cap on the OF opposite that used for illumination. Furthermore, for analyses that will employ only a single OF for illumination, the three remaining OF can be fabricated from a hydrogel that contains a light absorbing material that absorbs all the light that passes through the AS by TIR. This will eliminate virtually all the stray light derived from the sensor. If two OF are to be used for illumination, they should be adjacent so that the remaining two OF can also be fabricated from light absorbing material. It is anticipated that fabricating one or more OF from light absorbing material is a measure that would be required only under the most extreme circumstances, i.e., when only a few molecules of analyte are present. At this level, the shot noise in the camera is more likely to contribute more to the difficulty in monitoring single molecules of fluorophore. Shot noise is normally minimized by a cooling system that is included in the camera.
There are other steps that can also reduce or eliminate light that has the potential to interfere with that emitted by very small amounts of an analyte. This light would be derived primarily from imperfections in the AS that restrict its ability pass all light by TIR or that are caused by light scattering of unwanted material in the AS. Since the emitted fluorescent light has a longer wavelength than the light that is used to illuminate the sample, the latter can be removed using a filter that is placed between the FO and the AS. This filter should be coated with Teflon AF or some other low refractive index material to prevent the loss of TIR.
As can be seen in
As noted from the similarities in the design features of the electrophoresis chamber and the fluorescence analysis device, it has been contemplated that both processes could be done simultaneously in the same apparatus. For example, the box that surrounds and protects the camera could be made to contain lower buffer. The well in the IF has already been designed to contain buffer. The surface of the Teflon AF could be coated with a transparent conducting metal such as indium tin oxide (ITO), which would serve as the lower electrode. A buffer layer of 100 micrometers between the ITO surface and the bottom of the AS would be sufficient to permit current flow though a charged hydrogel, particularly if the buffer were to be replaced continuously by perfusion. These obvious modifications would extend the range of uses for the device.
Throughout this application, various publications have been referenced. The disclosures in these publications are incorporated herein by reference in order to more fully describe the state of the art.
While the invention has been particularly described in terms of specific embodiments, those skilled in the art will understand in view of the present disclosure that numerous variations and modifications upon the invention are now enabled, which variations and modifications are not to be regarded as a departure from the spirit and scope of the invention. Accordingly, the invention is to be broadly construed and limited only by the scope and spirit of the following claims.
This continuation-in-part application claims priority from PCT/US2003/031486, filed 3 Oct. 2003, which is a continuation-in-part application of PCT/US2003/13538, filed 30 Apr. 2003, which application was filed as U.S. Pat. No 10/962,003 on 8 Oct. 2004, on which pending divisional was filed as application Ser. No. 12/148,243, filed 17 Apr. 2008.
| Number | Date | Country | |
|---|---|---|---|
| Parent | PCT/US2003/031486 | Oct 2003 | US |
| Child | 12800574 | US | |
| Parent | PCT/US2003/013538 | Apr 2003 | US |
| Child | PCT/US2003/031486 | US |
