MCROFLUIDIC METHODS FOR NUCLEIC ACID MONITORING

RELATED APPUCATION

This application claims the benefit of United States Provisional Application Serial Number 60/801 ,397, filed May 19, 2006, and United States Provisional Application Serial Number

60/861,969, filed December 1, 2006; both filed under 35 U.S.C. 119(e). The entire disclosure of the prior applications are hereby incorporated by reference.

FIELD OF THE INVENTION

The present invention pertains to the field of viral detection and microfluidics, in particular relating to monitoring of viral load during immunosuppressive or immunomodulatory therapy.

BACKGROUND OF THE INVENTION

Alt of the publications, patents and patent applications cited within this application are herein incorporated by reference in their entirety to the same extent as if the disclosure of each individual publication, patent application or patent was specifically and individually indicated to be incorporated by reference in its entirety.

There are a variety of important applications where a cost-effective and readily applied method of viral analysis could greatly improve the effectiveness of treatment. The art is in need of an integrated microchip-based means of detecting the presence of specific nucleic acids in a clinical sample at the clinical point of care. In the current era when potent immunosuppressive agents are used to prevent rejection in renal transplant recipients, high titer replication of BKV in the transplanted kidney has emerged as an important cause of graft dysfunction and graft loss. In the

context of renal transplantation, infection with BKV is almost universal in childhood and reactivation of the virus (from latency in the urinary epithelium) results initially in asymptomatic viruria (virus present in urine) that progresses to viremia (virus present in blood), subclinical nephritis and eventually, if unchecked, to BKV-associated interstitial nephritis or nephropathy (BKVN). Although the incidence of BKVN varies among transplant centers, up to 10% of renal transplant recipients develop BKVN, and 10-80% of these patients may suffer consequent loss of their graft.

In the absence of proven antiviral therapy for treatment, early identification of patients at risk of BKVN (before the onset of renal dysfunction) and subsequent modification of immunosuppressive regimens may significantly reduce damage. Although BK viremia is the best positive laboratory predictor of BKVN, viruria always precedes viremia, resulting in the recommendation that the detection of BK virura be used as an initial screening test. Both urine cytology and quantitative PCR for detection of BK DNA have been used for screening. Quantitative PCR to determine BKV load in urine and/or plasma has been shown to be sensitive and useful, but the need to purify DNA from urine or plasma introduces significant variability, and the cost of this test is relatively high. A recent international panel recommended that renal transplant patients be screened using a quantatitive measure of BKV load in urine every three months for the first two years post-transplant, and every 2-4 weeks once BKVN has been detected. Although benefit may be gained by routine screening at more frequent intervals, the screening frequency is limited by cost considerations. Therefore, there is a need for a less costly and more efficient on-chip PCR reaction using minute sample volumes for the sensitive and accurate detection of BKV DNA in unprocessed urine, thus avoiding DNA purification steps that increase the processing time, cost and variability of the assays. For the detection of BKV DNA, automated, low cost platforms that require human intervention only for introduction of raw urine to a microchip device would enable more frequent testing (at every clinic visit) and introduce real time, point of care monitoring, of BKV load in samples from renal transplant recipients. This is expected to facilitate early intervention, inform therapeutic strategies and prolong survival.

Microfluidic technology has evolved rapidly, but barriers remain that prevent widespread

application to diagnosis and monitoring, particularly for point-of-care use where cost, portability and ease of use are of greater concern than throughput. A remarkable range of microchip-based methods have been demonstrated; for example real-time nucleic acid sequence based amplification (NASBA) for cancer marker detection, PCR partially integrated with cell lysis, PCR-based assays, strand displacement amplification (SDA) and capillary electrophoresis (CE), and dielectrophoretic capture integrated with PCR, but the key development needed for the widespread implementation of molecular diagnostics is the effective integration of PCR with CE and fluid handling. Although Ramsey and others demonstrated on-chip integrations of PCR and CE (Woolley, A.T. et al. Anal. Chem 68:4081 (1996); Hong, J.W., et al Electrophoresis 22:328 (2001); Koh, C.G., et al Ana. Chem 75:4591 (2003); Khandurina, J. et al. Anal Chem 72:2995 (2000); Waters, L.C. et al. Anal Chem 70:5172 (1998)), such early work relied upon fluid control via electrokinetics, i.e. without valves. However, valves are critical to effective and reproducible analyses upon integrated systems. In recent years, extensive progress has been made in integrating valves with PCR/CE - of particular interest are the demonstrations by Mathies et al. (Lagally, E.T. et al. Anal Chem 76:3162 (2004)) and Landers et al. (Ferrance, J.P. et al. Analytica Chimica Acta 500:223 (2003)) of valves in integrated PCR/CE systems applied to diagnostics.

Although tremendous strides have been made, demonstrations to date have required some form of "off chip" sample preparation, analysis capability or extensive external instrumentation, thus hindering the use of these assays in a clinical setting. There are as yet no demonstrations of applications in a clinical, point-of-care setting where "gold standard" validation, cost effectiveness, reliability and ease of use have been shown.

SUMMARY OF THE INVENTION

The present art has suffered from the. inability to provide widespread monitoring of viral load in blood or urine, particularly for point-of-care use. This need is particularly relevant for patients undergoing immunomodulatory, immunosuppressive, or immunoablative therapy.

m one aspect, the present invention provides for a microfluidic device capable of providing fast,

accurate and reproducible point-of-care analysis of viral presence in a clinical sample. In one embodiment of the present invention, the clinical sample is selected from the group comprising urine, blood or tissue.

In another aspect, the present invention provides for a method of monitoring specific nucleic acids presence in a clinical sample. Ih another aspect, the present invention provides for method for monitoring viral nucleic acid presence in the blood or urine of a patient undergoing immunomodulatory, immunosuppressive, or immunoablative therapy. In one embodiment, the viral DNA monitored is JC virus. In a further embodiment, the viral DNA monitored is BK virus. In another aspect the present invention provides for an apparatus and method for detection and monitoring of JC virus in the blood or urine of a Multiple Sclerosis patient undergoing immunomodulatory, immunosuppressive, or immunoablative therapy. In a further aspect, the immunomodulatory, immunosuppressive, or immunoablative therapy is selected from the group consisting of natalizumab and rifuximab. The accompanying description illustrates preferred embodiments of the present invention and serves to explain the principles of the present invention

BRIEF DESCRIPTION OF THE FIGURES

FIGURE ϊ shows an exploded view of this tri-layer chip to illustrate the relative placement of the valves and the resistive element;

FIGURE 2 shows a cartoon of the tri-layer microchip;

FIGURE 3 shows representations of the side view of the tri-layer microchip comprising a top etched glass layer 301 (flow-layer), a bottom etched glass layer 303 (control layer), and a PDMS membrane 302 between these two glass layers, which is actuated (microcontroller controlled) by pressurized air or vacuum;

FIGURE 4 shows an electropherogram of an on-microchip PCR analyzed using the dual-layer microchip using two different sieving matrices (a) non-denaturing medium of GeneScan® polymer (5GS 10G) and (b) a denaturing POP6 sieving matrix;

FIGURE 5 shows quantitation of BKV on-microchip on the basis of serial dilution (a) concentration in sample: 1.78xlO ⁷ copies/ml (b) concentration in sample: 1.83x10 copies/ml;

FIGURE 6 shows quantitation of BKV on-microchip based on PCR cycle number;

FIGURE 7 shows detection of BKV on an integrated microchip;

FIGURE 8 shows on-microchip detection outcomes of varying concentration of viral loads;

FIGURE 9 shows the result of microchip-based amplification of JC, BK and Cytomegalovirus DNA as detected on SYBR stained polyacrylamide gels, JCV=positive JC control, B l=positive BK virus control;

FIGURE 10 shows the section of the Immunoglobulin-H gene, along with the appropriate location of the primers;

FIGURE 11 shows the single-step microchip RT-PCR integration demonstrating the capability of the present invention to detect transcripts using the β2 microglobulin (β2M) gene with primers designed to amplify a 243 bp fragment from total RNA from MM+ KMS-34 cell line: (a) On- chip run (without size standard), (b) On-chip CE run (with size standard), (c) Conventional thermal cycler positive control;

FIGURE 12 shows on-microchip single step RT-PCR performed to detect a region specific the CDR2/CDR3 of the Immunoglobulin VDJ for the MM positive patient Pt 1 , and resulting in a 177 bp PCR product: (a) On-chip run (without size standard), (b) On-chip CE run (with size standard), (c) Conventional thermal cycler positive control;

FIGURE 13 shows on-microchip single step RT-PCR performed to detect a region specific the CDR2/CDR3 of the Immunoglobulin VDJ for the MM positive patient Pt 2, and resulting in a

168 bp product: (a) Oii-chip run (without size standard), (b) On-chiρ CE run (with size standard), (c) Conventional thermal cycler positive control;

FIGURE 14 shows the microchip single-step RT-PCR detection to identify Norovirus starting with RNA isolated from the Norovirus positive patient, Pt-3: (a) On-chip run (without size standard), (b) On-chip CE run (with size standard), (c) Conventional thermal cycler positive control;

FIGURE 15 shows the microchip single-step RT-PCR detection to identify Norovirus starting with RNA isolated from the Norovirus positive patient, Pt-4: (a) On-chip run (without size standard), (b) On-chip CE run (with size standard), (c) Conventional thermal cycler positive control;

FIGURE 16 shows the results of the traditionally used three-stage PCR (denaturation, annealing and extension) reduced to a two-stage PCR (with annealing and extension combined), using isolated RNA from the KMS-34 cell lines: (a) On-chip run (without size standard), (b) On-chip CE run (with size standard), (c) Conventional thermal cycler positive control; and

FIGURE 17 shows an exploded view of the dual-layer chip is shown to illustrate the relative placement of the valves and the resistive element (length - 95 mm, width = 30 mm, height = 2.3 mm).

DETAILED DESCRIPTION QF THE PRESENT INVENTION

As used herein the term "immunomodulatory therapy" means the therapeutic administration of a compound or compounds which result in the response of the immune system to an immune challenge being different following therapy as compared to prior to therapy. The term immunomodulatory therapy includes, but is not limited to administration of a compound or compounds which: effect a switch from Tl to T2 immune response, effect a switch from T2 to Tl immune response, alter immune cell trafficking within the body, sequester immune cells to a

body organ or system, sequester immune cells away from a body organ or system or prevent certain immune activities.

As used herein the term "immunosuppressive therapy" means the therapeutic administration of a compound or compounds which result in the decreased ability of the immune system respond to an immune challenge. It is contemplated that an immunosuppressive therapy will also be an immunomodulatory therapy.

As used herein the term "immunoablative therapy" means the therapeutic administration of a compound or compounds which result in the removal, destruction, death or permanent inactivation of immune cells, a subset of immune cells, immune effector cells and/or their precursor cells. The term immunoablative therapy includes, but is not limited to, chemotherapy or radiation therapy which results in the death of immune cells and/or their precursors and administration of antibodies, humanized or otherwise, which results in the removal, destruction, death or permanent inactivation of immune cells and/or their precursor cells. It is contemplated that an immunoablative therapy will also be an immunomodulatory therapy,

As used herein the term "viral load" means the presence of viral nucleic acid in a sample, as determined on a quantitative, semi-quantitative or qualitative basis.

As used herein, the term "clinical sample" means a fluid or tissue originating from a human. The sample may either be unmodified, or alternatively the sample may be processed before introduction into the devices of the present invention. Processing is contemplate to include, but not be limited to, pH alteration, ion removal, ion addition, cell separation, cell purification, cell removal, protein removal or cell lysis; all of which give rise to a sample for analysis which would enrich the sample for the nucleic acid of interest, if present.

The use of immunomodulatory, immunosuppressive or immunoablative therapies are established for the treatment of autoimmune diseases. A number of therapies result in a deleterious alteration of the immune system, either intentionally or as an ancillary effect of the therapy. One skilled in the art would be capable of identifying those therapies which result in a modification of the immune system of a patient in which the immune system is compromised or otherwise

debilitated such that it is no longer functioning as it was prior to the administration of the therapy. Examples of diseases in which immunomodulatory, immunosuppressive or immunoablative therapies are relevant include but are not limited to aplastic anemia, ulcerative colitis, cancer, systemic lupus erythematosus, multiple sclerosis (MS), rheumatoid arthritis, myasthenia gravis and Crohn's disease. Immunosuppressive therapy usually accompanies organ transplants, including bone marrow and blood, and is frequently accompanied by increased viral loads for multiple viruses during, for example, acute or chronic rejection phases. Examples of immunomodulatory, immunosuppressive or immunoablative therapies include, but are not limited to, cyclophosphamide, methotrexate, azathioprine, mercaptopurine, dactinomycin, mitomycin C, bleomycin, mitramycin, antilymphocyte antibodies, antithymocyte antibodies, nataluzimab, rituximab, anti-CD25(IL-2) antibodies, anti-CD3 antibodies, basiliximab, daclizumab, tacrolimus, cyclosporine, sirolimus, Interferons such as interferon-beta, Tumor Necrosis Factor binding agents such as infliximab or etanercept or adalimumab, mycophenolate, FTY720 and glatiramer acetate.

As an example, in Multiple Sclerosis (MS) patients, it is common to use compounds which alter the immune system of the patient as a method to beneficially alter the course of this autoimmune disease. One example of an immunomodulatory therapy is the use of the MS therapeutic nataluzimab, which is a monoclonal antibody thought to block the migration of leukocytes to the site of inflammation and demyelination which gives rise to the observed symptoms of MS.

Recent reports have outlined the potential adverse effects that immunomodulatory therapies can have in MS patients, in particular the recent reports of Progressive Multifocal Leukoencephalopathy (PML) in MS patients undergoing concurrent treatments with interferon Beta- 1 α and the monoclonal antibody natalizumab (Langer-Gould, A. et al N Engl J Med 353:375 (2005)). Though the exact reasons for the observed increase incidence of PML in MS patients undergoing these therapies has not been established, it has been speculated that concurrent natalizumab and interferon therapy prevents the suppression of reactivated JC virus and the virus cannot be suppressed until the effects of the natalizumab wear off (Langer-Gould, A. et al N£«g/ Jλ.W353:375 (2005)). Therefore information on, and monitoring of, JC virus

activation, proliferation or presence in a patient, through the detection of a viremia or viruria is useful information for the patient and the patient's caregiver or medical practitioner, Similarly, recent studies have demonstrated that treatment with the monoclonal antibody therapeutic rituximab, has correlated with increased incidence of PML. Though not necessary to practise the present invention, it is hypothesized that the class of drugs giving rise to immunomodulatory, immunosuppressive, or immunoablative effects can result in increased incidence of virus mediated disease, including but not limited to PML caused by JC virus. The present invention provides for a means to routinely monitor viral load for JC, BK or other viruses before, during and after treatment.

JC virus is acquired in childhood and normally remains dormant in the bone marrow, kidney epithelia and spleen. The virus is spread to the central nervous system during a bout of viremia which occurs in the immunocompromised patient (Langer-Gould, A. et al N EnglJMed 353:375 (2005)). Therefore identification of a JC virus viremia during an immunomodulatory, immunosuppressive or immunoablative therapy, in this case treatment with natalizumab alone or in association with an additional immunomodulatory therapeutic such as interferon Beta- lα, will allow a treating physician to identify a patient at increased risk of PML, The treating physician may then choose to discontinue the immunomodulatory therapy; increase vigilance for such diseases including, but not limited to, PML; or administer prophylactic antiviral therapy such as Highly Active Antiviral Therapy (HAART) which has shown efficacy in treating of PML in the past (Garcia, D et al JCUn Microbiol 31:124 (1999)).

The presence of large amounts of JC viral nucleic acid in the cerebral spinal fluid (CSF) at the time of initial PML symptom presentation, absent HAART, has been shown to correlate with poor survival in patients (Bossolasco, S. et al Clin Infect Dis 40:738 (2005)). Therefore, there is significant utility, independent of the potential predictive information for PML obtained from monitoring JC viral burden in the blood or urine, in identifying the presence of JC virus nucleic acid in the blood or urine of a patient. Since large amounts of JC viral nucleic acid at the onset of PML indicates low survival, absent HAART, monitoring of JC viral nucleic acid (which correlates with JC viral load) is useful to identify JC viral replication or release of JC viral

nucleic acid into the blood or urine (for example, through lytic events in JC virus infected cells) which can be used to quickly identify, predict or corroborate potential PML events.

The hypothesized independence of JC virus viremia or viruria from disease presentation in patients undergoing immunomodulatory, immunosuppressive or immunoablative therapy does not diminish the utility associated with the monitoring of viral load in patients. The monitoring of viral load on a frequent and ongoing basis during the administration of the immunomodulatory, immunoablative or immunosuppressive therapy and following the end of the therapy, is further needed while the immune system is modified or otherwise compromised; this provides the medical practitioner with valuable insight into the state of the patient and the potential disease risks. In a preferred embodiment of the present invention, the presence and/or quantity of viral nucleic acid, and therefore the viral load, is determined through analysis of urine, tissue or blood and use of the methods and apparatus disclosed herein.

The increased risk recently observed with immunomodulatory, immunosuppressive or immunoablative therapies for MS, described above, is a risk commonly occurring in immunocompromised patients. The source of the immunocompromise may be an infection, such as Human Immunodeficiency Virus (HIV, a congenital disorder or an inherited disease, resulting in reduced T-cell or B-cell activity; it may also be a concurrent infection, for example measles, or a therapeutic treatment resulting in immunomodulatory or immunosuppressive effects, such as those discussed above. Examples of viral infections associated with reduced or otherwise altered immune activity, or "opportunistic" infections are known to those skilled in the art, include but are not limited to JC vims, BK virus, Herpes Simplex Viruses and Cytomegalovirus. One skilled in the art would be capable of identifying the appropriate primers to be used in place of the BK virus or JC virus specific primers, but otherwise using the apparatus and methods of the present invention.

Microfluidic microchip-based monitoring of viral presence in blood, urine or tissue holds promise for fast, low cost screening of patients. Miniaturization of the device lends itself to portability leading to pomt-of-care devices that are highly sensitive. Although significant further miniaturization of on-microchip PCR volumes is readily feasible, the current level of

miniaturization is optimal for detection of clinically relevant viral titers. Although single-use chips are inexpensive, if microchip reusability is clinically feasible, the assay cost may be further diminished. Additionally, with ongoing improvements in custom-made instrumentation, miniaturized PCR and nucleic acid detection will be performed in under three hours, and in the preferred embodiment of the present invention, under one hour. As such, patients could be monitored in a primary setting such as the clinic or even at home by untrained personnel. Inexpensive and frequent testing offers the possibility for rapid detection, faster clinical response to elevated viruria, and a more informed evaluation of the significance of viruria or viremia through more comprehensive surveillance. Although real time quantitative PCR offers more accurate quantitation than does the current microfluidic chip platform, exact quantitation may not be essential, since patient risk with lower levels of viruria is uncertain and is set at approximate cutoff values. Thus, the semi-quantitative approach of the present invention is likely adequate for use in the clinic. The present invention is the first demonstration of a miniaturized diagnostic procedure for BKV, JC virus, or viral load screening and the automated microfluidic approach described here offers significantly improved capability for close surveillance of renal transplant recipients, or patients otherwise under immunomodulatory, immunosuppressive or immunoablative therapy.

The system and method of the present invention is based on dual-layer and tri-layer chip architecture. The chamber for thermal cycling is etched in the fhiidic layer and is ~600 nL in volume. Although lower volume PCR is feasible, a 600nl volume is more suitable for detection at the commonly observed low levels of template, in clinical samples; particularly at early stages of disease. Thermal cycling is performed using a thin-film platinum resistive element, with resistive element and fluidic chamber geometry optimized for uniform temperature distribution using finite element modeling. The present invention demonstrates RT-PCR-CE integration using a tri-layer microchip and PCR-CE integration using a dual-layer microchip; with no manual inclusion of reagents between RT and PCR steps and PCR and CE steps, thus minimizing potential contamination and simultaneously reducing the complexity of the instrumentation for RT and PCR integration. Inclusion of CE as the detection strategy supports the use of many molecular diagnostics-based microchip tests, including mutation detection or •

sequencing-like applications. This chip architecture is compatible with the eventual elimination of external pressure and vacuum connections for the actuation of the valves. Furthermore, a single thin-film element that is simultaneously used as a resistive heater and a temperature sensor, and is more efficient than microchip systems requiring separate heating elements and temperature sensors. The microchips designed are re-usable, a useful strategy in a research setting that may also be valuable for some clinical applications. Using the system of the present invention, the detection of transcripts encoding the clonotypic immunoglobulin heavy chain VDJ of plasma cells from patients from multiple myeloma (MM), and the detection of viral RNA genomes (Norovirus), detection of viral DNA (including, but not limited to BK virus, JC virus) from clinical samples is possible.

The present invention demonstrates an entirely microchip-based system in which an application is developed, validated against conventional methods and then demonstrated using clinical samples. The present invention performs a nucleic acid analysis, including a viral load analysis, directly from samples of human bodily fluids, i.e. without off-chip sample preparation. The system is optimized to be insensitive to any PCR inhibitors resident in the sample. Its limit of detection allows performance of the testing over the clinically relevant range of viral titers. The present invention demonstrates integration of PCR and CE, and PC, CE and RT; for the amplification of viral or human derived DNA and separation of amplified PCR products in the same chip within a microfluidic platform.

The art is in need of a single-use or reusable assay with an optimal sample analysis volume for either individual or limited patient screening usage while using inexpensive, portable, peripheral instrumentation. The present invention provides a system capable of consistently detecting as few as 5 copies of the viral genome in a sample volume, and can detect as few as 1-2 copies, with sensitivity comparable to real-time PCR-based analysis, the current "gold standard". On- chip PCR can offer either a yes/no, quantitative or semiquantitative result that correlates well with parallel real-time PCR quantitation for the same samples. Further, preliminary costing suggests that on-chiρ testing for BKV will have a cost one-tenth of real time PCR, even in the current single-analysis paradigm. The present invention further contemplates simultaneous high-

throughput analysis in parallel channels on a single chip using the methods and system disclosed herein, and could further reduce the cost per test. As well, in combination with an external imaging/illumination and power source, the present invention provides a point-of-care device for molecular analysis that can be used in the clinic.

Microchip RT-PCR. All RT-PCR reactions were prepared in a total volume of 50 μl. The mixture included 25 μl of 2X reaction mixture (a buffer containing 0.4 mM of each dNTP, 2.4 mM MgSO4), 1 μl of the enzyme mixture comprising of Superscript® III RT and high-fidelity Platinum® Taq polymerase (Invitrogen Life Technology, Burlington, ON, Canada), 15-20 μl 5mM MgSO4, lμl of each forward and reverse primer (10 uM), 1 μg of RNA template, and including double distilled water to reach a 50 μl volume. The primers were labelled with VIC dye and synthesized by ABI (Applied Biosystems, Foster City, CA). Thermal cycling conditions using the microchip were 45°C for 30 min, 94°C for 2 min, 35 cycles of 94 ⁰ C for 15 s, 60°C for 30 s, 68 ⁰ C for 30 s, and a final extension time of 7 min at 68°C, subsequently the on- chip product was transferred to the CE microchannel209, 1707 for analysis.

On-chip PCR protocol. Clinical urine samples from renal transplant patients used in this study represented residual samples from patients followed at the University of Alberta Hospital, Alberta, Edmonton, that had been tested for BK viruria as part of routine care at the Provincial Laboratory for Public Health (Microbiology) and would have normally been discarded. AU samples were tested in an anonymous manner. Urine samples from renal transplant recipients were used directly with no prior nucleic acid purification or any other sample preparation step. 2 μl of urine sample (as diluted 1 in 10) was added to a final volume of 25 μl PCR mix for a final dilution of 1/100 by volume. The PCR reaction mixture contained a final concentration of IX PCR buffer, 2.8 μM MgCl ₂ , 200 nM dATP, dGTP, dCTP and dTTP, 1.2M betaine. Both the forward and reverse primers were found to be optimal at 30OnM. Amplification was carried out by IU Platinum® Taq DNA polymerase (Invitrogen Life Technology, Carlsbad, USA).

Thermocycling of the microfluidic chip was performed using custom-built instrumentation with: 300 s of denaturation at 94 "C, followed by 35 cycles of denaturation at 94 ⁰ C for 40 s, annealing at 60 ⁰ C for 50 ε, and extension at 72 ⁰ C for 40 s, and ending with an extension step of 72 ⁰ C for

180 s. After the chip reached room temperature, the mixture was unloaded by the actuation of the servo-motors and the amplified product allowed to mix with the electrophoresis buffer.

It is contemplated as part of the present invention that the on-chip PCR could be performed on DNA resulting from purification, preparation, or a reverse transcription procedures. It is therefore contemplated that the present invention may be used to detect both DNA or RNA viruses. The methods or preparation, purification or reverse transcription are known to those skilled in the art, and in particular are described in Sambrook et al. Molecular Cloning a Laboratory Manual Cold Spring Harbor Press 3 ed, (2001).

Sequencing of PCR product amplified on chip. Sequencing was performed to verify the PCR product amplified on-chip amplified product using capillary sequencing on the ABI 3100. The resulting sequence was identical to that predicted and also to that obtained from sequencing PCR product amplified from the same urine samples using the same primers in a conventional PCR reaction. The product sequence was SEQ ID NO. 3.

Conventional DNA fragment analysis of amplified PCR product. For DNA fragment analysis using CE and detection of Laser Induced Fluorescence (LIF), primers were labeled with VIC dye and synthesized by ABI (Applied Biosystems, Foster City, CA). For verification of the outcome of the microchip based CE, both the DNA fragment analysis and the sequencing were performed on the ABI 3100 Genetic Analyzer with POP4 polymer (ABI). On-chip amplified PCR product was denatured for 240 s at 96 ⁰ C and after rapid centrifugation, samples were immediately placed on ice for 300 s. The run was performed at 60 ⁰ C and 15kV with a run time of 1500 s following an injection performed with 1 kV for 22 s. The laser power was 12 mW and the run current was 100 μA. The results were analyzed using GeneMapper® v3.5 software (ABI).

Fluid handling. Servo-motor driven probes allow for pumping and valving of an internal PCR reservoir by the application of pressure to compress and close the PDMS microchannels. Here, the elimination of physical and electrokinetic means of fluid handling prevents any inter-run contamination and is also suitable for use with physiological fluids due to it being independent of

pH and viscosity changes. Further detail can be found described by Pilarski et al (Pilarski, P.M.et al. Journal of Immunological Methods 305:48 (2005)).

Chip architecture. FIG 1 shows the general tri-layer microchip structure (absent the CE component, for illustration purposes only), the top glass layer (fluidic layer) 110 was etched to form fluidic chambers 101, 206 and discontinuous channels 107, 207 for fluid handling. The bottom glass layer (control layer) 130 was etched to form valves at the discontinuities of the fluid layer. External vacuum and pressure lines were coupled to the microchip 102 and were controlled by a microcontroller-based circuitry to actuate the PDMS membrane 120 that forms valves 105, 205 (more folly described below), with three such valves in series acting together as a pump 202. The Pt resistive element 106, 204; acting as the heater and the temperature sensor, was patterned on the upper surface of the control layer 130. By design of the heater geometry and component placement, adequate spreading of the heat flux prior to heating of the PCR chamber was ensured, creating temperature uniformity within the PCR chamber. Holes 104 were drilled through the flow layer for accessing the heaters for electrical connection, and for coupling of the external pressurized air/vacuum ports. The microchip in operation was placed in a custom- built plexi-glass stage to clamp pressure/vacuum lines and electrode connections onto the microchip.

FIG 2 shows the integrated tri-layer microchip. The sample mix is prepared external to the chip and loaded into input well 201 of the chip. Subsequently, by appropriate sequence of actuations for the valves 202 abetting the PCR chamber 203, this RT-PCR mix is pumped into the reaction chamber for thermal cycling by regulation of the temperature of the platinum resistive element 204. Bubble-free loading of the fluid within the chip is consistently achieved, and this is planned for by design. After the genetic amplification is completed, the product is pumped out and into the input well 208 of the CE section of the microchip 209.

FIG 3 shows representations of the side view of the tri-layer microchip valves comprising of a top etched glass layer 301 (fluidic layer), a bottom etched glass layer 303 (control layer), and a PDMS membrane 302 between these two glass layers, which is being actuated (microcontroller controlled) by pressurized air or vacuum. The valves are normally closed when no external

vacuum is applied (FIG 3b). When external vacuum is coupled to a valve, (FIG 3a), the valve opens, providing continuity in the channel within the flow layer. When pressurized air is coupled to the valve (FIG 3b) the valve is sealed.

FIG 17 shows the dual-layer microchip architecture, the glass layer 1710 is irreversibly bonded to a layer of PDMS 1720 which contains the features of fluidic chambers 1701, 1702, 1703, 170S, 1706 and 1707; and discontinuous channels 1707 for fluid handling. The dual-layer microchip contains a sample loading well 1706, valving points 1704, an enclosed PCR chamber 1704, sample loading well 1702, buffer well 1701, sample waste well 1702, buffer waste well 1708 and CE channel (separation channel) 1707.

With the dual-layer chip design, servo driven probes at valving points 1704 allow for pumping and valving of fluids within the internal PCR reservoir 1705 by the application of pressure to compress and close the PDMS 1710 microchannels against the glass layer 1720. The elimination of physical and electrokinetic means of fluid handling prevents any inter-run contamination, and is also suitable for use with clinical samples.

Platinum thin film resistive element design and chip optimization of the tri-layer microchip. Platinum has a good thermal response and a resistivity that exhibits a highly linear dependence on temperature. Thus, when patterned, in addition to functioning as a heater, Pt can be used as a temperature sensor. Hence, thin-film resistive elements were designed to maintain uniform temperature within the PCR chamber, while simultaneously acting as a temperature sensor that provides necessary feedback for temperature regulation of the heater. Such a heater/sensor design eliminates the need for two resistive elements, greatly simplifying the microchip electrical interfacing and reducing space usage, an important benefit for a portable system. To achieve this, and to have the total resistance of the resistive element accurately reflect its temperature, the temperature distribution in the thin film was designed to be uniform by making use of finite element analysis (FEA). A ring geometry was established for the heater/sensor coupled with the optimal connection width to electrode pads yielded temperature uniformity (less then I ⁰ C temperature gradients) in the heater/sensor. IR imaging verified temperature distribution, and absolute temperature was calibrated using custom-built temperature

responsive crystals for different set temperatures (Hallcrest, Glenview, IL, USA). FEA placement of fluidic components in the vicinity of the heaters was also optimized to ensure the operation of other microchip components do not alter the temperature uniformity.

Temperature controller and thermal aspects of the tri-Iayer microchip. Microcontroller- based circuitry was used to operate the valves and pumps, as well as for controlling the thin film element for heating and for temperature sensing. Post-fabrication, the heaters were annealed prior to the calibration process and to all other operations, in order to minimize the stresses within the films that typically build up during fabrication. This annealing stage ensures highly repeatable performance at elevated temperatures. A DC current was applied to the resistive element, and the temperature of the element was subsequently computed from its resistance with apre-determined resistance vs. temperature function. With this temperature as the feedback, the applied current was controlled and adjusted to yield the desired temperature for thermal cycling. Within the microcontroller that is implemented in software, a PID controller was in operation to precisely regulate the current flow for thermal cycling within the PCR chamber. Proportional- derivative (PD) control was used to realize rapid transients between the PCR temperature stages, while proportional-integrative (PI) control was used to maintain a constant temperature during the stages themselves. The resistive element heats up almost instantaneously when the desired current is applied via the control circuitry, and therefore, the temperature cycling rate of the PCR chamber 203 was limited primarily by the thermal conductivity between the resistive thin film and the chamber and by cooling. Cooling was achieved via passive convection and hence is relatively slow, though other means of cooling known to the art; which includes but is not limited to Peletier cooling, thermally conductive heat sinks in direct or indirect thermal communication with the PCR chamber 203, is contemplated as part of the present invention.

Dual-layer Microfluidic chip fabrication. The PDMS/Glass chips are comprised of a 1.2mm thick layer of molded PDMS (Sylgard 184, Dow Coming, NC, USA) and a 1.1 mm thick borofloat glass (Paragon Optical Company, PA, USA) substrate. The soft-lithography replica molding approach of fabrication for PDMS is followed (Duffy, D.C., et al Anal Chem 70:4974 (1998).; Unger, M.A., et al. Science 288:113 (2000)). Briefly, first, a master with features in

AZ4620 photoresist (Shipley Microelectronics, MA, USA) is patterned. The chip designs were drawn in L-Edit v3.0 (MEMS Pro 8, MEMS CAP, CA, USA) and transferred to a chromium mask wafer using a pattern generator (DWL 200, Heidelberg Instruments, CA, USA). The 4 X 4" borofloat glass substrate was cleaned in a fresh Piranha solution and sputter coated with a layer of chromium to a thickness of ~200nm, spin-coated with AZ4620 photoresist at a spin speed of 500 rpm for 10 s and a spread speed of 2000 rpm for 25 s. A hotplate set at 100 °C was used to bake the wafer and then was placed in a box with a damp cleanroom wipe, light and moisture sealed, and left for ~2 h. UV exposure (30 s, 356 nm, and intensity of (19.2 mW/cm ) of the spin-coated substrate was performed through the chrome mask using a mask aligner (ABM Inc., CA, USA). The substrate was then chemically developed with AZ400K. that was diluted AZ400K (1:4) (AZ400K:H ₂ O) for about 80s resulting in the patterned mould. This was typically reused for ~20 times after which the glass substrate was reused after acetone cleaning. A single coating of this photoresist results in a feature height of ~14 μm. Posts made of aluminum or brass with radii of ~0.75 mm and heights of 1 mm were cleaned by a brief immersion in a cold Piranha, rinsed, dried and placed on the master at the locations for molding the PCR chambers. After mixing the pre-polymer and the curing agent in proportions of 1 : 10 by weight this was degassed (in vacuum oven at 20 in.Hg for 20 min) and thermally cured for 90 minutes at 90 ⁰ C. The active surfaces (features) of PDMS and the glass substrate were then exposed to oxygen plasma face-up in a reactive ion etch (RIE) (Plasmalab, Plasma Technology, Bristol, UK) with 80% O ₃ gas flow at a chamber pressure of 0.15 Torr, a power of 35 W and treated for 60 s. After oxygen plasma exposure, the surfaces of glass and PDMS were brought together and irreversibly bonded to form the microchip. To form the reservoirs, punched holes were made in the PDMS layer, these were 1.5 mm in diameter and able to contain a maximum volume of ~3.5 μl.

Tri-layer microchip fabrication. The microchip designs were drawn in L-Edit v3.0 (MEMS Pro 8, MEMS CAP, CA, USA) and transferred to a mask wafer using a pattern generator (DWL 200, Heidelberg Instruments, CA, USA). The 4 inch by 4 inch borofloat glass substrate (Paragon Optical Company, PA, USA) was cleaned in a hot Piranha (3:1 of H2S04:H2O2) and sputter coated with 20 μm of Cr and 200 μm of Au. HPR 504 photoresist was spin coated with a spin speed of 500 rpm for 10 s and a spread speed of 4000 rpm for 40 s. The photoresist coated

substrate was then baked in an oven set at 115"C for 30 minutes. UV exposure (4 s, 356 nm wavelength, and intensity of (19.2 mW/cm2) of the spin-coated substrate was performed through the chrome mask using a mask aligner (ABM Inc., CA, USA). The substrate was then chemically developed with Microposit 354 developer (Shipley Company Inc., Marlborough, MA, USA) for -25 s. Glass etch was performed using hydrofluoric acid (HF) at an etch rate of ~ 1.1 μm/min. The control layer was etched to 70 μm depth, and the flow layer 90 μm depth. Photoresist was removed by rinsing the substrate using acetone and isopropyl alcohol. Subsequently, Au and Cr (Arch Chemicals Inc., Norwalk, Connecticut, USA) etchants were used to strip the metal, the etch time being ~45 s for Au and ~30 s for Cr. Holes in the flow layer were drilled using a Waterjet for accessing both the flow and the control layers of the chip. The control layer requires by design the patterning of Pt films and this was performed via a lift-off technique. The metal-stripped etched glass was cleaned in fresh Piranha, and 20μm of Cr was then sputter deposited. Next, AZ 4620 photoresist was spin-coated for 10 s at a spread speed of 500 rpm and spin speed of 2000 rpm for 25 S, the substrate was soft-baked on a hot-plate for 90 s, and then hydrated for 2 h. The photoresist was them UV exposed for 30 s and developed using AZ 400K developer for ~180 s, after which the Cr was completely stripped. Then, 20 am of Ti and 220 nm of Pt were sputter deposited, and using lift-off, the Pt/Ti electrodes were defined on the control layer. To eventually realize the tri-layer microchip, the PDMS membrane was irreversibly bonded with the etched face of both the flow and the control layer.

Dual-layer microchip capillary electrophoresis (CE). Fragment analysis is performed within the crosstchanneJ CE section of the microchip using a modified .procedure. employed for the. glass-based chips (Vahedi, G. et al. Electrophoresis 25:2346 (2004)). Fragment analysis (CE) of the amplified PCR mix is performed within the microfluidic tool kit (μTK, Micralyne, Edmonton, Canada) (Vahedi, G. et al. Electrophoresis 25:2346 (2004)). The μTK provides similar the optical detection and high voltages needed to perform CE with confocal laser-induced fluorescence (LIF) detection. The LIF system uses excitation at 532 nm and detection at 578 nm. Further details on the system and its use can be found elsewhere (Ma, R. et al Electrophoresis 26:2692 (2005)). Briefly, conditioning of the chip involves flushing the chip with doubly

deionized water for a few minutes followed by rinse cycle with a running buffer Ix TBE (Tris Borate with EDTA). TBE was made with Tris (crystalline free base) and boric acid (Fisher Scientific, Fair Lawn, NJ, USA) and EDTA (Merck, Darmstadt, Germany). After thoroughly drying the channels, a non-denaturing GeneScan® polymer (Applied Biosystems, Foster City, CA) was loaded. The sample waste, buffer, and buffer waste wells were filled with 3 μl of IX TBE buffer withlθ% glycerol (w/w). Following amplification, a 0.3 μl aliquot PCR product was pumped from the enclosed PCR chamber into the open injection CE well where it was diluted in 0.1 X TBE to constitute a total volume of 3 μl. A pinch-off injection approach as in ma, R et al. (Ma ₅ R. et al Electrophoresis 26:2692 (2005)) with 0.4 kV was applied for 60 s with a separation voltage of 6 kV. LIF detection was performed at 76 mm from the channel intersection. Sizing was performed by simultaneously loading 0.3 μl of a DNA ladder GeneScan® 500 TAMRA (Applied Biosystems, Foster City, CA). Chip quality and consistency were monitored on a regular basis via a calibration procedure (Pilarski, L.M. et al. Journal of Immunological Methods 305:94 (2005)).

Tri-layer microchip capillary electrophoresis (CE) equipment. Fragment analysis of on-chip PCR product is performed within the cross-channel CE section of the integrated multi-layer microchip (FIG. 2) using a modification in the procedure for glass-only microchip (Vahcdi, G. et al Electrophoresis 25:2346 (2004)). Fragment analysis of the amplified PCR mix was performed within the microfluidic tool kit, μTK (Micralyne, Edmonton, Canada) (Ma, R. et al Electrophoresis 26:2692 (2005)), however this can also be readily performed within a portable CE system. The microchip platform and procedures described here can be readily integrated into a portable CE platform. The μTK provides the optical detection and high voltages needed to perform CE with confocal laser-induced fluorescence (LlF) detection with excitation at 532 nm, and detection at 578 ran. A pinch-off injection approach was used, with 0.4 kV applied for 60 s to inject the DNA sample and a separation voltage of 6 kV was used for separation through the CE channel 209. LIF-based detection was performed at 24 mm from the channel intersection. The on-chip PCR product was flushed out of the chip with 2 μl of 0.1X GABE (genetic analysis buffer with EDTA), 1.2 μl of HiDi formamide (ABI) and 0.5 μl of size standard (GS500) were

added, and the mixture was denatured for 4 min at 94 ⁰ C then rapidly cooled to ~4°C where it was kept for at least 10 minutes. A mix of 1 μl POP6 and 3 μl of 1 X GABE is loaded in the separation wells. To verify the identity of PCR products amplified on chip, both DNA fragment analysis and sequencing were performed on the ABI 3100 genetic analyzer.

Sieving matrices (Dual-layer and Tri-Jayer)

Both GeneScan® polymer (Applied Biosystems) and POP6 (Applied Biosystems) polymer are capable of being used for the present invention. Sizing was performed by simultaneously loading 0.3 μL of a DNA ladder GeneScan® 500 TAMRA (Applied Biosystems). Using GeneScan® polymer, the conditioning of the chip requires flushing the chip with doubly deionized water for at least 2 minutes, followed by rinsing with the running buffer IXTBE. TBE was made using Tris (crystalline free base) and boric acid (Fisher Scientific, Fair Law, NJ, USA) and EDTA (Merck, Darmstadt, Germany). GeneScan® polymer was mixed with glycerol (Sigma, St. Louis, MO, USA) and TBE buffer such that the final concentration of GeneScan® was 5% w/w and the final concentration of glycerol was 10% w/w. This dilution, referred to as "5GS 1OG" was used as the electrophoresis sieving matrix. A dilution of this solution 1 part in 10 parts deionized water is used for sample preparation and is referred to as 0. IxTBEl G. After thoroughly drying the channels, a non-denaturing 5GS10G polymer was loaded into the CE channel 209, 1707. The sample waste, buffer, and buffer waste wells 206, 1701, 1703, 1708 were filled with 3 μL of IxTBElG buffer with 10% glycerol w/w. Following amplification, a 0.3 μL aliquot PCR product was pumped from the PCR chamber 203, 1705 into the open injection CE well 208, 1702 where it was diluted in 0. IxTBEl G to constitute a total volume of 3 μL.

Using the denaturing POP6 polymer, first the polymer is heated to 67 ^C C for 10 minutes, before being loaded into the CE channel 209, 1707. Heating was done to reduce the viscosity of the POP6 polymer, as well as ensuring the complete dissolution of any unprecipitated urea, and facilitates the loading of the polymer within the CE channel 209, 1707. In the sample loading well 208, 1702 the PCR product was denatured for 4 minutes at 96°C and rapidly cooled to

approximately 4°C, and mixed with 1.2 μL of HiDi formamide (ABI), 1 μL of size standard (GS500) and 0.8 μL of l x genetic analysis buffer with EDTA to a total volume of 3 μL. Heating was implemented either through external application of heating and cooling to the sample well 208, 1702, or alternatively by addition of an additional heating element 204 adjacent to sample loading well 208, 1702. A mix of POP6 with of Ix GABE (1 :2 ratio) was loaded into the separation channel 209, 1707.

Patient samples. After institutional review board approval and informed consent, bone marrow (BM) samples were obtained at diagnosis or relapse of 2 patients with multiple myeloma (MM). BM was processed as previously described (Szczepek, K. et al Blood 92:2844 (1998)). With institutional review board approval, Norovirus was obtained from anonymous patients who provided stool samples to the Provincial Laboratories of Public Health, Edmonton. All samples were confirmed to be positive by conventional testing prior to testing on the microchip.

Microchip cleaning procedure for reusability. The microchips can be used indefinitely with the reusablity procedure presented here. The PDMS that is irreversibly bonded on both surfaces to glass was dissociated using Dynasolve 210 (Dynalloy, Indianapolis, IN, USA) when the microchips were left to soak for ~ 2 hours . This was followed by a hot Piranha cleaning (3 : 1 of H2SO4 to H ₂ O ₂ ) procedure to ensure that the disintegrated PDMS was cleaned from the surface. PDMS layers were then rebonded to the cleaned glass slides to re-use the chip. We have found that this re-usability procedure has no detectable ill-effect on the outcome of the PCR or on CE. The microchips can be used indefinitely with the reusablity protocol described here.

Example 1: Detection of BKV on dual-layer chip.

BKV is a small-enveloped virus within the polyomavirus family with a genome of approximately 5 Kbps of double stranded DNA37. The primers were designed to amplify a 293 bp region of the BKV structural protein VPl. The forward primer SEQ ID NO. 1 and the reverse primer sequence SEQ ID NO. 2 were used. The primers are located at nucleotide 433 on strain MM genome and position 2274 on the Dunlop strain genome. Analysis using the NIH BLAST database showed no cross reactivity with any other commonly present viral gene sequences. Specificity of these

primers was verified as described below.

To demonstrate proof-of-concept and optimize chip-based protocols, initial testing utilized a two chip system, having the PCR functionally on one microchip and the CE DNA fragment analysis of PCR product on a second chip with separation carried out using a Microfluidic Toolkit (μTK) as described herein and elsewhere (Vahedi, G. et al. Electrophoresis 25:2346 (2004)). The microfluidic chips are made of patterned poly(dimethyl)siloxane (PDMS) bonded to a glass substrate. Externally-actuated reusable diaphragm pumps and a pinch-off valve are integrated in the system (Pilarski, P.M. et al. Journal of Immunological Methods 305:48 (200S)) and these accurately control and manipulate fluids, particularly for the immobilization of reagents at elevated temperatures. This also minimizes potential cross-contamination, allows for effective inter-nur cleaning, and ensures a single system can be used for multiple runs on multiple chips. FIG. 17 shows a PDMS/glass hybrid microchip able to seamlessly perform integrated PCR and CE, with a PCR volume of ~2 μl on a single assay, an optimal volume for BKV load assessment. Raw urine samples having known titers of BKV (determined by real time PCR), were diluted 100-fold, manually added (0.24 μl of diluted urine and 2μl PCR master mixture) into the PCR sample loading well 1706 and driven into the PCR chamber 1705 by using the valve/pump 1704. During the 35 cycles of PCR, pinch-off valves 1704 immobilized the fluid until the amplified product was pumped to an open chamber and manually removed. For product sizing, a DNA ladder (size standard) was added to the product, followed by manual introduction to a glass CE chip for separation using the μTK. The amplified product was confirmed to be the expected 293 bp size as shown in the electropherogram of FIG. 4. In all cases, product detected on chip was also detected using the ABI 3100 DNA fragment analysis instrument for a "gold standard" comparison. To further validate the on-chip PCR of BKV DNA, sequencing of on-chip PCR product was carried out externally using the ABI 3100 and was shown to be correct, confirming the specificity of the on-chip reaction. A comparison was also made between the amplification of BKV from purified DNA and unprocessed urine, with equivalent results (data not shown), indicating that no loss of sensitivity occurs with the use of raw urine.

Chip-based CE was performed using two different sieving matrices in the injection and

separation channels of the glass microchip to ensure appropriate resolution both under denaturing conditions (POP6), and non-denaturing GeneScan® polymer. GeneScan® polymer is known to induce sequence dependent migration (FIG. 4a) in addition to size-based separations during migration which means that the placement of the product peak (293 bp) is not as expected with reference to the DNA ladder peaks. To address this, the same PCR products were verified by sizing on the ABI 3100 and by sequencing to confirm product identity. In the GeneScan® polymer, the DNA fragments migrate as double strands, the repeatability in positioning of the PCR product peak in the electropherogram relative to the DNA ladder peaks is hence used to indicate the occurrence of BKV. To further confirm that the correct product was amplified on- chip, a denaturing sieving polymer (POP6) was also used that is known to size the DNA fragments correctly. By using POP6, the fragment sizing is correct and the BK PCR product peak is positioned as expected with reference to the DNA ladder.

Example 2: Semi-quantitative detection of BK viral load on dual-layer chip.

Quantitative or semi-quantitative results are essential for determining whether follow-up is required as well as to monitor the progress of the disease during and following intervention. Two semi-quantitative techniques were employed to determine virus titers on chip: (a) patient urine samples were diluted from 1/10 to 1/10 ⁷ in steps of 10-fold dilutions, and (b) three aliquots of the same reaction mixture were cycled for 35, 25 or 15 cycles of PCR.

FIG. 5 shows representative serial dilution analysis of two urine samples with clinically defined BKV titers, one with a titer of 1.78xlO ⁷ copies/ml and one with a titer of 1.83 xlO ¹⁰ copies/ml. Overall, six samples were analyzed in dilution series, in duplicate. The x-axis represents the serial dilution factor and the y-axis represents the relative fluorescence units (rfu) after on-chip PCR runs were performed and analyzed using the ABI 3100. To increase sample throughput, PCR product was often manually removed from on-chip PCR wells and analyzed using the ABI 3100, which is capable of simultaneous analysis of 16 samples at a time.

For the first sample (FIG. 5a), detectable PCR product is amplified at the 1/100 (-360 copies/reaction) and 1/1000 dilution (-36 copies/reaction) but not at 1/10,000 (-3.6

copies/reaction), a dilution at which BKV templates have likely become limiting. For the second higher titer sample (FIG. 5b), product was amplified at all dilations tested, including at 1/10 ⁷ (~360 copies/reaction). Such a strategy would enable a degree of quantitation sufficiently precise for clinical use. The same results were obtained using either on-chip CE or the ABI 3100 as previously reported (Pilarski, L.M. et al. Journal of Immunological Methods 305:94 (2005))

Secondly, real-time PCR can be approximated on chip by analysis of product amplification (fluorescent intensity) as a function of increasing cycle numbers. FIG. 6 shows two representative samples. Progress in DNA amplification as captured by change in relative fluorescence (in ABI 3100) with cycle progression during PCR for different patient samples having different concentrations. Overall, five samples were amplified on-chip in duplicate, (a) Concentration of the sample: 1.78xlO ⁷ copies/ml, (b) Concentration of the sample; 1.83xlO ^io copies/mL. The x-axis represents the number of PCR thermal cycles performed prior to unloading the product from the chip and the y-axis represents the relative fluorescence units (rfu) after on-chip PCR runs and analysis on the ABI 3100 as indicated for FIG. 5. For the first sample (FIG 6a), PCR product is undetectable at 15 cycles but becomes detectable at 25 and 35 cycles. For the second (FIG. 6b), PCR product is detectable even at 15 cycles, consistent with the 10 ³ higher concentration of BKV.

FIGS. 5 and 6 indicate that the present microchip-based approach can be adapted for semiquantitative PCR strategies that could be incorporated on-chip in an automated manner. As one approach, by dispensing a predefined volume of the PCR mixture for analysis after a user- defined number of thermal cycles, repeated analysis at increasing cycle numbers could be performed. This type of quantitation can be evaluated for utility in the clinic and the extent of precision required for informed decision making, possibly providing a low-cost alternative to real time PCR.

Example 3: Integrated PCR-CE detection of BKV on dual-layer microchip.

Using the PCR cycling and separation conditions established above, detection of BKV on an integrated hybrid chip was implemented. A total of 10 different urine samples were tested in 30

different runs, on 20 different chips configured as shown in FIG. 17, with consistent results. Although both polymers shown in FIG. 4 were tested for CE, due to the relative ease of loading with GeneScan® polymer it was used for all subsequent analyses. FIG. 7 shows amplification of a 293 bp BKV PCR product using the integrated PCR-CE chip. Electropherogram with x-axis depicting the time in seconds when DNA fragments are detected as seen by the relative fluorescence intensity (rfu) peaks (y-axis). The 293 bp PCR product indicates of the presence of BK virus that was detected on the microchip using the PCR-CE integrated chip wherein both the functionalities of PCR and CE were performed seamlessly on a single assay. Resolution in base pairs provides a quantitative assessment of CE, with a lower number indicative of better separation. The resolution of this chip (FIG. 7) compares favourably with those of other demonstrations of CE performed on PDMS/Glass chips. Separation on the integrated chip had a resolution of 5-6 bps for DNA sizes bracketing the BKV PCR product.

Example 4: Sensitivity of BKV detection using dual-layer on-chip PCR.

Chip-based testing was performed on a panel of 55 randomly selected urine samples from renal transplant recipients, previously screened for BKV titre using routine clinical testing by real-time PCR. The 55 samples were tested in a total of 228 on-chip PCR reactions. FIG. 7 shows that product amplification is a function of the level of viruria, and that the detection level on-chip is consistent with the clinically established viral titre (y-axis of FIG. 8).

Viruria was detected in urine samples from renal transplant recipients randomly selected for on- chip PCR analysis. The number of recipients is indicated by the x-axis. Values on the y-axis were the BKV titers reported by the clinical testing laboratory (Provincial Laboratories, Edmonton, Canada) by quantitative PCR using LightCycler®. A total of 55 samples were tested at a dilution of 1/100 in a 2 μl PCR reaction volume, with a total of 228 replicate analyses performed on microfluidic chips. For low titer samples, multiple replicate tests were performed as indicated in results. Each sample was analyzed 2-10 times with reproducible results. As predicted from the expected random distribution of copies, consistently positive results were obtained for samples containing more than 10 copies of the BKV in the on-chip PCR reaction mixture (5x10 ⁵ copies/ml).

Below a known titre of 5x10 ⁵ copies/ml, the PCR reaction was sometimes positive and sometimes negative, likely reflecting a limiting number of viral templates in the 2 μl reaction volume (i.e. <10). For a series of 11 urine samples having known titres between 4xlO ⁴ to 5xlO ³ copies/ml, and having less than 10 copies of BKV in the PCR chamber, 29 out of 91 different reactions (32%) amplified detectable product, in a manner consistent with Poisson distributions of template. Each low-titre urine sample was tested 7-11 times and the urine was scored as positive if at least one test was positive; in practice, all of these urines scored positive for at least two tests. These results indicate that the limit of detection can be as low as 1-2 copies of BKV.

A set of 13 urine samples from transplant patients with clinically undetectable BKV were analyzed on chip in a total of 38 different reactions, with three positive results among replicate tests for two urines. This may reflect true positives in a Poisson distribution of templates that were below the levels required for detection with clinical real time PCR (as might be expected with microchip detection at the near-single copy level); 11/13 of these urine samples were consistently negative. These results demonstrate that the false positive rate for on-chip PCR is likely to be low or zero.

Finally, even though the PCR product was sequenced and found to be correct, and analysis using the NIH BLAST program indicated no cross reactivity with other DNA sequences, to further test the specificity of the BK primers, BK-negative urine samples from the transplant recipients were spiked with ~10 ⁴ copies each of JC virus, cytomegalovirus (CMV) and Epstein Barr Virus (EBV) DNA, since these can be present in urine from transplant recipients. This was followed by separate PCR amplification reactions using primers for BKV, JC, CMV or EBV. PCR reactions to amplify the spiked virus types were positive, confirming that viral templates had in fact been added to the BK-negative urines. All runs included negative controls for the PCR itself, which contained the reaction mixture and primers but had water instead of a urine sample. For the results disclosed herein, all accompanying PCR controls were negative. A set of 10 spiked urines were tested in 33 separate on-chip assays. For 7/10 spiked urines, BK product was undetectable. For 5/39 tests, a positive result was obtained at least once from 3/10 urine samples. Overall, for a total of 77 tests of transplant urines having clinically undetectable BKV levels,

90% of tests and 10/13 urines were negative. As expected, this rate was similar to that for the unspiked samples having clinically undetectable BKV. This indicates that the BK primers disclosed herein do not cross-react with other viral DNA templates.

An international panel of transplantation and nephrology experts recommend that further follow- up, such as renal biopsy, be performed if recipient viruria is equal to or greater than 10 ⁷ copies/ml and that recipients with viruria less than this may not require invasive intervention (Hirsch, H.H. et al Transplantation 79:1277 (2005)). The on-chip assay is approximately 10-fold more sensitive than the titre for triggering medical intervention. This high sensitivity coupled with the cost-effectiveness of a fully-developed device, would enable larger scale screening that would in turn facilitate trials to more comprehensively determine the clinical significance of low level viruria.

Example 5: Detection of JC Virus dual-layer microchip

Chip-based testing was performed on cloned JC virus samples using primers specific for the JC virus, SEQ ID NO. 4 and SEQ ID NO. 5. The resulting product of 253 base pairs was identified with serial dilutions of cloned JC virus when using the chip-based procedure described herein. No product was observed when using conventional non-chip based PCR procedures. FIG 9 shows a SYBR green stained acrylamide gel showing the product obtained form a chip-based amplification using the JC specific primers disclosed herein with the chip-based amplification process and apparatus of the present invention.

Example 6: Beta 2 Microglobulin (β2M) transcript identification using tri-layer microchip

To demonstrate the utility of the present invention with total RNA during the development of the microchip platform, we amplified and detected transcripts encoding p2-microglobulin (β2M), a housekeeping gene, expressed in all human cells (Pilarski, P.M. et al JImmuno Met 305:48(2005)). For the initial validation, RNA was isolated from KMS-34, a myeloma cell line. The primers SEQ ID NO. 11 and SEQ ID NO. 12, were designed to amplify a 243 bp fragment from RNA or from genomic DNA . The on-chip RT-PCR amplified a fragment of the

appropriate size, indicating successful RT-PCR (FIG. 11). On-chip PCR product was verified using acrylamide gels, and sequencing within the genetic analyzer (ABB 100).

Example 7: Microchip identification of clonotypic signatures using tri-layer microchip

MM is a cancer of the immune system characterized by a unique immunoglobulin gene rearrangement, the clonotypic IgH VDJ, and by IgH translocations that enable identification of the MM clone and are thus clinically valuable biomarkers. Point-of-care testing for such biomarkers has value for quantitative and real time monitoring of tumor burden in individual patients, as well as for monitoring response to treatment and detecting emerging relapse. Such detection techniques permit identification of malignant cells that would otherwise be clinically cryptic. The clonotypic IgH VDJ rearrangement was detected with primers specific to the complementarity determining regions (CDR2 and CDR3) of the clonotypic IgH VDJ for each individual patient (Szczkpek, AJ. et al. Blood 92:2844 (1998)).

FIG 10 shows the variable region (VH) contains framework regions (FRs) that maintain antibody structure and stability, and complementarity determining regions (CDRs) that vary in nucleotide sequence and confer specificity for distinct antigens, hence, patient specific primers are required to be designed. CDR3 is composed of diversity (DH) and joining (JH) regions and is the most variable of the CDRs. A portion of the constant (CH) region is shown. The remainder of the downstream CH region and immunoglobulin light chain V gene, have not been presented. Patient specific CDR primers are indicated as black arrows. Scale indicated at the top is in base pairs.

Two patients (Pt-I and Pt-2), which were MM positive, were tested for the CDR2/3 regions with positive amplification for both patients (FIG. 12 and FIG 13 respectively), and these tests were performed in triplicate. Primers used were SEQ ID NO. 6 and SEQ ID NO. 7 for patient 1, and SEQ EO NO. 6 and SEQ ID NO. 8 for patient 2. On chip RT-PCR product was detected using a glass CE chip with sizing using GS500 size standard. The product was confirmed using conventional DNA fragment analysis. The detection of genomic IgH VDJ provides a quantitative measure of tumour burden because each cancer cell has only one copy of the rearranged IgH

gene.

Example 8: ϊYi-layer microchip Norovirus detection

Norovirus (NV) is a single-stranded RNA virus with a genome of approximately 7.5 kbp, that causes acute gastroenteritis in humans. NV is transmitted by fecally contaminated food, vomit, and person-to-person contact. Tests involving electron microscopy (EM) are laborious and relatively insensitive, lmmuno EM or ELISA tests are limited to the viral subtypes detected by the antibodies used. Conventionally, and as recommended by the Center for Disease Control (CDC) both nucleic acid hybridization and RT-PCR assays are used to detect the NV genome in clinical and environmental specimens as they are highly specific. These genetic amplification techniques detect as few as 102-104 viral particles/ml in stool. The present invention demonstrates the potential for screening of patients one at a time, thereby holding promise for real time detection in the community rather than in a central diagnostic facility using high- throughput approaches.

Two patients (Pt-3 and Pt-4) were first tested conventionally by the Provincial Laboratories for Public Health, followed by testing using the newly developed microchip RT-PCR assay. The template for the on-chip tests was RNA isolated from stool samples. The intent here was to provide a yes/no test readout and a demonstration of the developed microfluidic platform, not precise quantitation as performed for other viral templates. The primer sets used in this microfluidic-based RT-PCR assay, SEQ ID NO. 9 and SEQ ID NO. 10, are highly conservative (universal), and in routine clinical use by the Provincial Laboratories (Edmonton, Canada). Primers were designed to pick up the most common infectivity strains of NV, and have been cross-verified by the other conventional techniques. NV was successfully detected by the on-chip RT-PCR system (patient 3, FIG. 14 and patient 4, FIG 15). All on-chip tests were performed in triplicate to ensure highly repeatable outcomes. For this initial testing, the microchip-amplified product was detected by DNA fragment analysis on a glass CE chip using a DNA ladder (GS500), as well as conventionally by sizing using gel electrophoresis and DNA fragment analysis using the ABI 3100 genetic analyzer. The product identity was confirmed by sequencing.

Example 9: Two-stage fast PCR using tri-Iayer microchip

In conventional PCR, using bench-top thermal cyclers, the emphasis is more on specificity, and to a lesser extent on product yield and with a minimum emphasis on the total reaction time. However, to realize a point-of-care device, reduced analysis time is important, and a necessity for wide acceptance of such devices. Microchip implementations have shown that the analysis time can be reduced by shrinking the reaction volumes to reduce thermal load, and thus speeding the temperature transitions (instrumentation modifications are underway for even more rapid transitions than this current demonstration). As yet however, no work has attempted to reduce the overall reaction time on-chip by optimizing the dwell times and the stages within PCR. Such implementations have recently been demonstrated within conventional thermal cyclers.

Using the present invention, it is possible to implement faster PCR by optimizing its biochemical processes. By optimizing (minimizing) the dwell times at each stage, the most optimized case leads to the elimination of the extension stage in the 3-stage PCR (FIG. 16). Together with published modifications in instrumentation, this could lead to even more rapid analysis times, provided each reaction is optimized to ensure that the primers anneal specifically to the target and that the Taq enzyme activity is sufficiently high with the modified conditions, as was done here.

The annealing and the extension steps of Hie regular 3-step PCR thermal cycling were combined to realize two-step PCR cycling comprising a denaturation step and a combined annealing/extension step. The reaction mix was as for the regular PCR (described above), however, the thermal program was modified as: 46°C initially for 10 minutes for the RT step, followed by 88°C for 1 min to deactivate the Superscript® III. This was followed by a 10 s denaturation at 88 ⁰ C and a 20s annealing/extension for 20 s, this was repeated for 40 cycles. A final extension of 1 min at 68°C was then realized.

Comparable results were observed for β2M using the 3-step PCR approach and the 2-step PCR approach (FIG 16). With the established PCR conditions (both thermal and biochemical), nonspecific amplification that might diminish the yield was not encountered. Hence, this means of

time and temperature optimization is a viable method for optimizing PCR to reduce the overall time from application of sample to test result. The product was confirmed by DNA fragment analysis on chip CE as well as conventionally using gel electrophoresis and DNA fragment analysis, and confirmed by sequencing.

While particular embodiments of the present invention have been described in the foregoing, it is to be understood that other embodiments are possible within the scope of the invention and are intended to be included herein. It will be clear to any person skilled in the art that modifications of and adjustments to this invention, not shown, are possible without departing from the spirit of the invention as demonstrated through the exemplary embodiments. The invention is therefore to be considered limited solely by the scope of the appended claims.