Patent Application
20220364157
The present invention relates to the field of multiplex based viral pathogen detection and analysis. More particularly, the present invention relates to detecting the presence of COVID-19 virus in patient and environmental samples.
The COVID-19 pandemic has increased awareness that viral infection can be an existential threat to health, public safety and the US economy. More fundamentally, there is a recognition that the viral risks are more dangerous and more complex than had been thought and will require new approaches to diagnostics and screening.
The next pandemic wave is expected to have more pronounced flu-like symptoms (seasonal influenza A and/or B) coupled with the COVID-19, or COVID-19 variants that will coexist with the Coronavirus already responsible for the common cold. These complexities are expected to pose significant challenges to public health and the healthcare system in diagnosing multi-symptom conditions accurately and efficiently.
The COVID-19 pandemic has also led to the realization of an additional level of complexity that the realization that human health and environmental contamination are linked in a fundamental way that affects collection efficiency and increases risk to the healthcare workers (1, 2). Alternatives to nasopharyngeal collection methods such as for example, saliva collection are needed to enable scalability among millions of individuals.
Q-RT-PCR technology has dominated COVID-19 diagnostics and public health screening. Independent of the test developer, Q-RT-PCR has been shown to have an unusually high false negative rate (15% up to 30%). As of May 2020, the CDC has recorded 613,041 COVID-19 tests. With a 15% false negative rate, approximately 91,956 people would thus be falsely classified as free of infection. Meta-analysis has shown that the false negative rate for Q-RT-PCR is high below day 7 of infection when viral load is still low. This renders Q-RT-PCR ineffective as a tool for early detection of weak symptomatic carriers while also lessening its value in epidemiology.
Thus, there is a need in the art for superior tools to not only administer and stabilize sample collection for respiratory viruses from millions of samples in parallel obtained from diverse locations including, clinic, home, work, school and in transportation hubs, but also to test multiple respiratory markers at the highest levels of sensitivity and specificity.
The present invention is directed to a method for detecting Coronavirus disease 2019 (COVID-19) in a sample. A sample is obtained and a total RNA isolated. A combined, reverse transcription reaction and an asymmetric PCR amplification reaction is performed on the isolated total RNA using at least one fluorescent labeled primer pair selective for a target nucleotide sequence in the COVID-19 virus, the fluorescent labeled primer pair comprising an unlabeled primer and a fluorescently labeled primer. Fluorescent labeled COVID-19 virus amplicons thus generated are hybridized to a plurality of nucleic acid probes, each attached to a solid microarray support. Each of the nucleic acid probes have sequence corresponding to a sequence determinant in the COVID-19 virus. The microarray is washed at least once and imaged to detect at least one fluorescent signal from the hybridized fluorescent labeled COVID-19 virus amplicons. The present invention is also directed to a related method further comprising calculating an intensity of the fluorescent signal for the COVID-19 virus amplicons, correlating with the number of COVID-19 virus genomes in the sample. The present invention is further directed to a related method for detecting at least one other non-COVID-19 virus in the sample by employing at least two fluorescent labeled primer pairs selective for the target nucleotide sequence in the COVID-19 virus and in the at least one other non-COVID-19 virus, and nucleic acid probes having a sequence corresponding to sequence determinants in the COVID-19 virus and the at least one of the other non-COVID-19 virus to perform the steps described above.
The present invention is also directed to a method for detecting a respiratory disease-causing pathogen in a sample. A sample is obtained, and total nucleic acids are isolated. A combined, reverse transcription reaction and an asymmetric PCR amplification reaction is performed on the isolated total nucleic acids using at least one fluorescent labeled primer pair selective for a target nucleotide sequence in at least one respiratory disease-causing pathogen, the fluorescent labeled primer pair comprising an unlabeled primer and a fluorescently labeled primer. At last one fluorescent labeled pathogen specific amplicon thus generated are hybridized to a plurality of nucleic acid probes, each attached to a solid microarray support. Each of the nucleic acid probes have sequence corresponding to sequence determinants in the pathogen. The microarray is washed at least once and imaged to detect at least one fluorescent signal from the hybridized fluorescent labeled pathogen specific amplicons. The present invention is also directed to a related method further comprising calculating an intensity of the fluorescent signal for the fluorescent labeled pathogen specific amplicons, correlating with the number of pathogen specific genomes in the sample.
The present invention is further directed to a method for detecting a Coronavirus disease 2019 (COVID-19) virus in a sample. A sample is obtained, and a total nucleic acid is isolated. A combined, reverse transcription reaction and an asymmetric PCR amplification reaction is performed on the isolated total nucleic using at least one fluorescent labeled primer pair selective for a target nucleotide sequence in the COVID-19 virus, the fluorescent labeled primer pair comprising an unlabeled primer and a fluorescently labeled primer in an excess over the unlabeled primer. Fluorescent labeled COVID-19 virus amplicons thus generated are hybridized to a plurality of nucleic acid probes each attached to a solid microarray support. Each of the nucleic acid probes have a sequence corresponding to a sequence determinant in the COVID-19 virus. The microarray is washed at least once and imaged to detect at least one fluorescent signal from the hybridized fluorescent labeled COVID-19 virus amplicons thereby detecting the COVID-19 in the sample. The present invention is also directed to a related method for detecting additionally, at least one non-COVID-19 virus in the sample by performing the combined reverse transcription and the asymmetric PCR amplification reaction on the isolated total nucleic acid using at least two fluorescent labeled primer pairs, each comprising the unlabeled primer and the fluorescently labeled primer selective for a target nucleotide sequence in the COVID-19 virus and the at least one non-COVID 19 virus. Fluorescent labeled COVID-19 and non-COVID-19 virus specific amplicons generated are hybridized to the plurality of nucleic acid probes having a sequence corresponding to the sequence determinant in the COVID-19 virus and the at least one non-COVID-19 virus. The present invention is also directed to a related method for detecting additionally, at least one bacterium in the sample by performing the combined reverse transcription and the asymmetric PCR amplification reaction on the isolated total nucleic acid using at least two fluorescent labeled primer pairs, each comprising the unlabeled primer and the fluorescently labeled primer selective for a target nucleotide sequence in the COVID-19 virus and the at least one bacterium. Fluorescent labeled COVID-19 virus and bacterium specific amplicons generated are hybridized to the plurality of nucleic acid probes having a sequence corresponding to the sequence determinant in the COVID-19 virus and the at least one bacterium. The present invention is further directed to a related method for detecting additionally, at least one fungus in the sample by performing the combined reverse transcription and the asymmetric PCR amplification reaction on the isolated total nucleic acid using at least two fluorescent labeled primer pairs, each comprising the unlabeled primer and the fluorescently labeled primer selective for a target nucleotide sequence in the nucleic acids from the COVID-19 virus and the at least one fungus. Fluorescent labeled COVID-19 virus and fungus specific amplicons generated are hybridized to the plurality of nucleic acid probes having a sequence corresponding to the sequence determinant in the COVID-19 virus and the at least one fungus.
As used herein, the term “a” or “an” when used in conjunction with the term “comprising” in the claims and/or the specification may mean “one,” but it is also consistent with the meaning of “one or more,” “at least one,” and “one or more than one.” Some embodiments of the invention may consist of or consist essentially of one or more elements, method steps, and/or methods of the invention. It is contemplated that any method described herein can be implemented with respect to any other method described herein.
As used herein, the term “or” in the claims is used to mean “and/or” unless explicitly indicated to refer to alternatives only or the alternatives are mutually exclusive, although the disclosure supports a definition that refers to only alternatives and “and/or.”
As used herein, “comprise” and its variations, such as “comprises” and “comprising,” will be understood to imply the inclusion of a stated item, element or step or group of items, elements or steps but not the exclusion of any other item, element or step or group of items, elements or steps unless the context requires otherwise. Similarly, “another” or “other” may mean at least a second or more of the same or different claim element or components thereof.
As used herein the phrase “lowest limit of detection (LLoD)” corresponds to the lowest number of genome copies capable of generating a measurable signal in the assay under consideration. For example, the LLoD corresponds to an analytical sensitivity of ˜0.3 copies/reaction and post extraction sensitivity of ˜3 copies/reaction.
As used herein, the term “about” refers to a numeric value, including, for example, whole numbers, fractions, and percentages, whether or not explicitly indicated. The term “about” generally refers to a range of numerical values (e.g., ±5-10% of the recited value) that one of ordinary skill in the art would consider equivalent to the recited value (e.g., having the same function or result). In some instances, the term “about” may include numerical values that are rounded to the nearest significant figure. For example, a fold excess of 3.6-fold to 8.8-fold is encompassed by about 4-fold to about 8-fold.
In one embodiment of the present invention, there is provided a method for detecting a Coronavirus disease 2019 (COVID-19) in a sample, comprising, obtaining the sample; isolating from the sample a total RNA; performing a combined reverse transcription and an asymmetric PCR amplification reaction on the isolated total RNA using at least one fluorescent labeled primer pair comprising an unlabeled primer, and a fluorescently labeled primer selective for a target nucleotide sequence in the COVID-19 virus to generate at least one fluorescent labeled COVID-19 virus amplicon; hybridizing the fluorescent labeled COVID-19 virus amplicons to a plurality of nucleic acid probes each having a sequence corresponding to a sequence determinant in the COVID-19 virus, each of said nucleic acid probes attached at a specific position on a solid microarray support; washing the microarray at least once; and imaging the microarray to detect at least one fluorescent signal from the hybridized fluorescent labeled COVID-19 virus amplicons, thereby detecting the COVID-19 in the sample.
In this embodiment, in one aspect, the sample is any sample obtained from a subject including, but not limited to a nasopharyngeal swab, nasal swab, mouth swab, and mouth wash (sample obtained by rinsing the subject's buccal cavity). A pooled sample obtained by combining two or more of these samples or by combining samples from multiple subjects may also be used. In another aspect of this embodiment, the sample is an environmental sample obtained from inanimate sources including but is not limited to an aerosol and a hard surface. In this embodiment, the aerosol samples are obtained using commercial air samplers such as for example a Coriolis Micro Air Sampler. In this embodiment, a sample from a hard surface is obtained using a swab. In either aspect of this embodiment, the viruses from samples obtained on swabs are dispersed in a liquid such as phosphate buffered saline. Aerosol samples are transferred into a volume of a liquid such as phosphate buffered saline.
In this embodiment, the COVID-19 virus is a Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV 2) or a mutated form thereof. In this embodiment, in some aspects, the sample is mixed with an RNA stabilizer such as for example, a chemical stabilizer that would protect the RNA from degradation during storage and transportation, prior to the RNA isolating step.
In this embodiment, total RNA is isolated from the COVID-19 virus and other potential contaminating pathogens and human cells. Any commercially available RNA isolation kits such as for example, a Quick-DNA/RNA Viral MagBead Kit from Zymo Research may be used for this purpose. The RNA thus isolated is used without further purification. Alternatively, an intact SARS-CoV-2 virus may be captured and enriched by binding to magnetic beads, using kits as for example that from Ceres Nanosciences (e.g. Ceres NANOTRAP technology), or by precipitation of the virus with polyethylene glycol (PEG), after which the enriched virus can be lyzed by heating with a “PCR-Friendly” lysis solution such as 1% NP40 in TE buffer and then used without additional purification.
In this embodiment, the COVID-19 virus RNA is used as a template in a combined reverse transcription/amplification reaction (RT-PCR). In this step, the nucleic acid sequences in the COVID-19 virus RNA are transcribed using a reverse transcriptase enzyme to generate COVID-19 complementary DNA (cDNA) that is amplified in the same reaction using COVID-19 virus selective fluorescent labeled primer pairs to generate fluorescent labeled COVID-19 virus amplicons. In this embodiment, fluorescent labeled primer pair comprises an unlabeled primer, and a fluorescently labeled primer in about 4-fold to about 8-fold excess of the unlabeled primer whereby, upon completion of the reaction, the fluorescently labelled amplicon will be primarily single stranded (that is, the reaction is a type of “asymmetric PCR”). In this embodiment, the fluorescent labeled primer pairs have forward (odd numbers) and reverse (even number) sequences shown in SEQ ID: 23 to SEQ ID: 28 (Table 1) and SEQ ID: 74 to SEQ ID: 80 (Table 37). Commercially available reverse transcriptase enzyme and buffers are used in this step. Controls including, but not limited to a RNAse P control having fluorescent labeled primer pair (forward primer SEQ ID: 43, reverse primer SEQ ID: 44) are also used herein. Any fluorescent label may be used, including, but not limited to CY3, a CY5, SYBR Green, a DYLIGHT™ DY647, a ALEXA FLUOR 647, a DYLIGHT™ DY547 and a ALEXA FLUOR 550.
Further in this embodiment, the fluorescent labeled COVID-19 virus amplicons generated are hybridized to a plurality of nucleic acid probes. The nucleic acid probes have a sequence corresponding to sequence determinants in the COVID-19 virus and have sequences SEQ ID: 45 to SEQ ID: 48 (Table 2) and SEQ ID: 85 to SEQ ID: 94 (Table 38). Controls including, but not limited to a RNAse P control nucleic acid probe (SEQ ID: 71 and SEQ ID: 72) and a negative control nucleic acid probe (SEQ ID: 73) are also used herein (Table 2). The nucleic acid probes are attached to a solid microarray support. The solid support is any microarray including but not limited to a 3-dimensional lattice microarray.
Further in this embodiment, after hybridization, unhybridized amplicons are removed by washing the microarray. Washed microarrays are imaged to detect a fluorescent signal corresponding to the fluorescent labeled COVID-19 virus specific amplicons to detect presence of the COVID-19 virus in the sample.
Further to this embodiment, the method further comprises calculating an intensity for the fluorescent signal. The calculated intensity is correlated with the number of COVID-19 virus specific genomes in the sample. The measured intensity is correlated with the number of COVID-19 virus specific genomes in the sample. Based on analysis of virus-free samples, an experimentally determined intensity threshold is established for the hybridization to each probe on the microarray, such that a fluorescent intensity above that threshold signifies the presence of SARS-CoV-2 viral RNA, while fluorescence intensities below the threshold signifies that SARS-CoV-2 was not detected.
Further to this embodiment, when the sample comprises in addition to the COVID-19 virus, at last one other non-COVID-19 virus comprising a Respiratory Syncytial Virus, a Middle East Respiratory Syndrome coronavirus (MERS-CoV), a Severe Acute Respiratory Syndrome Coronavirus (SARS-CoV), a 229E Coronavirus, a OC43 Coronavirus, a NL63 Coronavirus, a HKU1 Coronavirus an Influenza A virus or an Influenza B virus, the method further comprises detecting presence of both COVID-19 virus and the at least one non-COVID-19 virus by performing the amplification and hybridization steps described above for these viruses. In this embodiment, the fluorescent labeled primer pairs have forward (odd numbers) and reverse (even number) sequences shown in SEQ ID: 23 to SEQ ID: 42 (Table 1) and SEQ ID: 74 to SEQ ID: 84 (Table 37). and nucleic acid probe sequences SEQ ID: 45 to SEQ ID: 70 (Table 2) and SEQ ID: 85 to SEQ ID: 97 (Table 38). Controls including, but not limited to a RNAse P control having primer pair (forward primer SEQ ID: 43, reverse primer SEQ ID: 44) and nucleic acid probe (SEQ ID: 71 and SEQ ID: 72) and, a negative control nucleic acid probe (SEQ ID: 73) are also used herein.
In another embodiment of the present invention, there is provided a method for detecting a respiratory disease-causing pathogen in a sample, comprising obtaining a sample; isolating total nucleic acids from the sample; performing a combined reverse transcription and an asymmetric PCR amplification reaction on the isolated total nucleic acids using at least one fluorescent labeled primer pair comprising an unlabeled primer, and a fluorescently labeled primer selective for a target nucleotide sequence in the at least one respiratory disease-causing pathogen, to generate at least one fluorescent labeled pathogen specific amplicon; hybridizing the fluorescent labeled pathogen specific amplicons to a plurality of nucleic acid probes each having a sequence corresponding to a sequence determinant in the pathogen, each of said nucleic acid probes attached at a specific position on a solid microarray support; washing the microarray at least once; and imaging the microarray to detect at least one fluorescent signal from the hybridized fluorescent labeled pathogen specific amplicons, thereby detecting the respiratory disease-causing pathogen in the sample.
In this embodiment, in one aspect, the sample is any sample obtained from a subject including, but not limited to a nasopharyngeal swab, nasal swab, mouth swab, and mouthwash (sample obtained by rinsing the subject's buccal cavity). A pooled sample obtained by combining two or more of these samples or by combining samples from multiple subjects may also be used. In another aspect of this embodiment, the sample is an environmental sample obtain from inanimate sources including but is not limited to an aerosol and a hard surface. In this embodiment, the aerosol samples are obtained using commercial air samplers such as for example a Coriolis Micro Air Sampler. In this embodiment, a sample from a hard surface is obtained using a swab. In either aspect of this embodiment, the viruses from samples obtained on swabs are dispersed in a liquid such as phosphate buffered saline. Aerosol samples are transferred into a volume of a liquid such as phosphate buffered saline.
In this embodiment, the respiratory disease-causing pathogen is a virus, a bacteria, a fungi, or a combination of these. The sample may also comprise mutated forms of these pathogens. Examples of respiratory disease-causing viruses include, but are not limited to, Severe Acute Respiratory Syndrome Coronavirus 2 (COVID-19 virus), a Respiratory Syncytial Virus, a Middle East Respiratory Syndrome coronavirus (MERS-CoV), or a Severe Acute Respiratory Syndrome Coronavirus (SARS-CoV), or a 229E Coronavirus, or a OC43 Coronavirus, or a NL63 Coronavirus, or a HKU1 Coronavirus or an Influenza A virus or an Influenza B virus, an adenovirus, a bocavirus, a metapneumovirus, a parainfluenza and a rhinovirus. Examples of respiratory disease-causing bacteria include, but are not limited to, a Mycobacterium species (e.g. Mycobacterium tuberculosis), a Streptococcus species (e.g. Streptococcus pneumoniae), a Mycoplasma species, an Enterococcus species, a Haemophilus species, a Klebsiella species, a Moraxella species and a Corynebacterium species. Examples of respiratory disease-causing fungus include, but are not limited to, a Histoplasma species, a Coccidioides species, a Blastomyces species, a Rhizopus species, an Aspergillus species, a Pneumocystis species and a Cryptococcus species. In this embodiment, in some aspects, the sample is mixed with a nucleic acid stabilizer such as for example, a chemical stabilizer that would protect the nucleic acids from degradation during storage and transportation, prior to the isolating step.
In this embodiment, a total nucleic acids potentially comprising nucleic acids from the pathogen and contaminating human cells is isolated. Commercially available nucleic acid isolation kits such as for example, a Quick-DNA/RNA MagBead Kit from Zymo Research are used for this purpose. The total nucleic acids thus isolated is used without further purification. Alternatively, the pathogens may be captured using hydrogel chemistry (Ceres Nanosciences) or enriched using methods including, but not limited to centrifugation and polyethylene glycol (PEG), followed by lysis of the enriched pathogens by heating with a “PCR-Friendly” lysis solution such as 1% NP40 in TE buffer and the total nucleic acids used without additional purification.
In this embodiment, the isolated total nucleic acids are used as a template in a combined reverse transcription/amplification reaction (RT-PCR). In this step, the nucleic acid sequences in the pathogen are transcribed using a reverse transcriptase enzyme to generate pathogen specific complementary DNA (cDNA) that is amplified in the same reaction using pathogen selective fluorescent labeled primer pairs to generate fluorescent labeled pathogen specific amplicons. In this embodiment, fluorescent labeled primer pair comprises an unlabeled primer, and a fluorescently labeled primer in about 4-fold to about 8-fold excess of the unlabeled primer whereby, upon completion of the reaction, the fluorescently labelled amplicon will be primarily single stranded (that is, the reaction is a type of “asymmetric PCR”). In this embodiment, when the pathogen is a virus, the fluorescent labeled primer pairs have forward (odd numbers) and reverse (even number) sequences shown in SEQ ID: 23 to SEQ ID: 28 (Table 1) and SEQ ID: 74 to SEQ ID: 80 (Table 37). Commercially available reverse transcriptase enzyme and buffers are used in this step. Controls including, but not limited to a RNAse P control having fluorescent labeled primer pair (forward primer SEQ ID: 43, reverse primer SEQ ID: 44) are also used herein. Any fluorescent label may be used, including, but not limited to CY3, a CY5, SYBR Green, a DYLIGHT™ DY647, a ALEXA FLUOR 647, a DYLIGHT™ DY547 and a ALEXA FLUOR 550.
Further in this embodiment, the fluorescent labeled pathogen specific cDNA amplicons generated are hybridized to a plurality of nucleic acid probes. In this embodiment, when the pathogen is a virus, the nucleic acid probes have a sequence corresponding to sequence determinants in the pathogen and have sequences SEQ ID: 45 to SEQ ID: 48 (Table 2) and SEQ ID: 85 to SEQ ID: 94 (Table 38). Controls including, but not limited to a RNAse P control nucleic acid probe (SEQ ID: 71 and SEQ ID: 72) and a negative control nucleic acid probe (SEQ ID: 73) are also used herein (Table 2). The nucleic acid probes are attached at specific positions on a solid support. The solid support is any microarray including but not limited to a 3-dimensional lattice microarray.
Further in this embodiment, after hybridization, unhybridized amplicons are removed by washing the microarray. Washed microarrays are imaged to detect a fluorescent signal corresponding to the fluorescent labeled pathogen specific amplicons to detect presence of the respiratory disease-causing pathogen in the sample.
In this embodiment, the method steps for detecting the respiratory disease-causing virus, bacterium and fungus are concurrently performed in a single assay. This is advantageous since it enables streamlined detection of COVID-19 virus and the other pathogens in a single assay. Further in this embodiment, the methods described above may be used to detect in any combination, a COVID-19 virus, another virus, a bacterium, or a fungus.
Further to this embodiment, the method further comprises calculating an intensity for the fluorescent signal. The calculated intensity is correlated with the number of pathogen specific genomes in the sample. The measured intensity is a function of the number of pathogen specific genomes in the sample. Based on analysis of pathogen-free samples, an experimentally determined intensity threshold is established for the hybridization to each probe on the microarray, such that a fluorescent intensity above that threshold signifies the presence of pathogen nucleic acid, while fluorescence intensities below the threshold signifies that the pathogen was not detected.
In yet another embodiment of the present invention, there is provided a method for detecting a Coronavirus disease 2019 (COVID-19) in a sample, comprising obtaining the sample; isolating from the sample, a total nucleic acid; performing a combined reverse transcription and an asymmetric PCR amplification reaction on the total nucleic to generate fluorescent labeled COVID-19 virus amplicons using at least one fluorescent labeled primer pair selective for a target nucleotide sequence in the COVID-19 virus RNA, said fluorescent labeled primer pair comprising an unlabeled primer and a fluorescently labeled primer in an excess over the unlabeled primer; hybridizing the fluorescent labeled COVID-19 virus amplicons to a plurality of nucleic acid probes each having a sequence corresponding to a sequence determinant in the COVID-19 virus, each of said nucleic acid probes attached at a specific position on a microarray support; washing the microarray at least once; and imaging the microarray to detect at least one fluorescent signal from the hybridized fluorescent labeled COVID-19 virus amplicons, thereby detecting the COVID-19 in the sample.
In this embodiment, in one aspect, the sample is any sample obtained from a subject including, but not limited to a nasopharyngeal swab, nasal swab, mouth swab, and mouthwash (sample obtained by rinsing the subject's buccal cavity). A pooled sample obtained by combining two or more of these samples or by combining samples from multiple subjects may also be used. In another aspect of this embodiment, the sample is an environmental sample obtain from inanimate sources including but is not limited to an aerosol and a hard surface. In this embodiment, the aerosol samples are obtained using commercial air samplers such as for example a Coriolis Micro Air Sampler. In this embodiment, a sample from a hard surface is obtained using a swab. In either aspect of this embodiment, the viruses from samples obtained on swabs are dispersed in a liquid such as phosphate buffered saline. Aerosol samples are transferred into a volume of a liquid such as phosphate buffered saline.
In this embodiment, the COVID-19 virus is a Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV 2) or a mutated form thereof. In this embodiment, in some aspects, the sample is mixed with an RNA stabilizer such as for example, a chemical stabilizer that would protect the RNA from degradation during storage and transportation, prior to the RNA isolating step.
In this embodiment, a total nucleic acids potentially comprising nucleic acids from pathogens including the COVID-19 virus, and contaminating human cells is isolated. Commercially available nucleic acid isolation kits such as for example, a Quick-DNA/RNA MagBead Kit from Zymo Research are used for this purpose. The total nucleic acids thus isolated is used without further purification. Alternatively, the pathogens may be captured using hydrogel chemistry (Ceres Nanosciences) or enriched using methods including, but not limited to centrifugation and polyethylene glycol (PEG), followed by lysis of the enriched pathogens by heating with a “PCR-Friendly” lysis solution such as 1% NP40 in TE buffer and the total nucleic acids used without additional purification.
In this embodiment, the COVID-19 virus RNA is used as a template in a combined reverse transcription/amplification reaction (RT-PCR). In this step, the nucleic acid sequences in the COVID-19 virus RNA are transcribed using a reverse transcriptase enzyme to generate COVID-19 complementary DNA (cDNA) that is amplified in the same reaction using COVID-19 virus selective fluorescent labeled primer pairs to generate fluorescent labeled COVID-19 virus amplicons. In this embodiment, fluorescent labeled primer pair comprises an unlabeled primer, and a fluorescently labeled primer in about 4-fold to about 8-fold excess of the unlabeled primer whereby, upon completion of the reaction, the fluorescently labelled amplicon will be primarily single stranded (that is, the reaction is a type of “asymmetric PCR”). In this embodiment, the fluorescent labeled primer pairs have forward (odd numbers) and reverse (even number) sequences shown in SEQ ID: 23 to SEQ ID: 28 (Table 1) and SEQ ID: 74 to SEQ ID: 84 (Table 37). Commercially available reverse transcriptase enzyme and buffers are used in this step. Controls including, but not limited to a RNAse P control having fluorescent labeled primer pair (forward primer SEQ ID: 43, reverse primer SEQ ID: 44) are also used herein. Any fluorescent label may be used, including, but not limited to CY3, a CY5, SYBR Green, a DYLIGHT™ DY647, a ALEXA FLUOR 647, a DYLIGHT™ DY547 and a ALEXA FLUOR 550.
Further in this embodiment, the fluorescent labeled COVID-19 virus amplicons generated are hybridized to a plurality of nucleic acid probes. The nucleic acid probes have a sequence corresponding to sequence determinants in the COVID-19 virus and have sequences SEQ ID: 45 to SEQ ID: 48 (Table 2) and SEQ ID: 85 to SEQ ID: 94 (Table 38). Controls including, but not limited to a RNAse P control nucleic acid probe (SEQ ID: 71 and SEQ ID: 72) and a negative control nucleic acid probe (SEQ ID: 73) are also used herein (Table 2). The nucleic acid probes are attached to a solid microarray support. The solid support is any microarray including but not limited to a 3-dimensional lattice microarray.
Further in this embodiment, after hybridization, unhybridized amplicons are removed by washing the microarray. Washed microarrays are imaged to detect a fluorescent signal corresponding to the fluorescent labeled COVID-19 virus specific amplicons to detect presence of the COVID-19 virus in the sample.
Further to this embodiment, when the sample additionally comprises at least one virus that is not a COVID-19 virus, the step of performing the combined reverse transcription and the asymmetric PCR amplification reaction on the isolated total nucleic acid comprises using at least two fluorescent labeled primer pairs, each comprising the unlabeled primer and the fluorescently labeled primer selective for a target nucleotide sequence in the COVID-19 virus and the at least one non-COVID-19 virus to generate at least one fluorescent labeled COVID-19 virus specific amplicons and at least one fluorescent labeled non-COVID-19 virus specific amplicon; and the step of hybridizing comprises hybridizing the at least one fluorescent labeled COVID-19 virus specific amplicon and the at least one fluorescent labeled non-COVID-19 virus specific amplicon to the plurality of nucleic acid probes each having a sequence corresponding to the sequence determinant in the COVID-19 virus and the at least one non-COVID-19 virus.
In this embodiment, the non-COVID-19 virus is any virus including, but not limited to a respiratory disease-causing RNA or DNA virus. Examples of RNA viruses include, and are not limited to a Respiratory Syncytial Virus, a Middle East Respiratory Syndrome coronavirus (MERS-CoV), a Severe Acute Respiratory Syndrome Coronavirus (SARS-CoV), a 229E Coronavirus, a OC43 Coronavirus, a NL63 Coronavirus, a HKU1 Coronavirus an Influenza A virus, an Influenza B virus, a metapneumovirus, a parainfluenza, and a rhinovirus. In this embodiment, the fluorescent labeled primer pairs have forward (odd numbers) and reverse (even number) sequences shown in SEQ ID: 23 to SEQ ID: 42 (Table 1) and SEQ ID: 74 to SEQ ID: 84 (Table 37). and nucleic acid probe having sequences SEQ ID: 45 to SEQ ID: 70 (Table 2) and SEQ ID: 85 to SEQ ID: 97 (Table 38). Controls including, but not limited to a RNAse P control having primer pair (forward primer SEQ ID: 43, reverse primer SEQ ID: 44) and nucleic acid probe (SEQ ID: 71 and SEQ ID: 72) and, a negative control nucleic acid probe (SEQ ID: 73) are also used herein. Examples of DNA viruses include and are not limited to an adenovirus and a bocavirus.
Further to this embodiment, when the sample additionally comprises at least one bacterium, the step of performing the combined reverse transcription and the asymmetric PCR amplification reaction on the isolated total nucleic acid comprises using at least two fluorescent labeled primer pairs, each comprising the unlabeled primer and the fluorescently labeled primer selective for a target nucleotide sequence in the COVID-19 virus and the at least one bacterium to generate at least one fluorescent labeled COVID-19 virus specific amplicons and at least one fluorescent labeled bacterium specific amplicon; and the step of hybridizing comprises hybridizing the at least one fluorescent labeled COVID-19 virus specific amplicon and the at least one fluorescent labeled bacterium specific amplicon to the plurality of nucleic acid probes each having a sequence corresponding to the sequence determinant in the COVID-19 virus and the at least one bacterium.
In this embodiment, the bacterium is any bacterium including, but not limited to a respiratory disease-causing bacterium. Examples of bacteria include, and are not limited to a Mycobacterium species (e.g. Mycobacterium tuberculosis), a Streptococcus species (e.g. Streptococcus pneumoniae), a Mycoplasma species, an Enterococcus species, a Haemophilus species, a Klebsiella species, a Moraxella species and a Corynebacterium species.
Further to this embodiment, when the sample additionally comprises at least one fungus, the step of performing the combined reverse transcription and the asymmetric PCR amplification reaction on the isolated total nucleic acid comprises using at least two fluorescent labeled primer pairs, each comprising the unlabeled primer and the fluorescently labeled primer selective for a target nucleotide sequence in the COVID-19 virus and the at least one fungus to generate at least one fluorescent labeled COVID-19 virus specific amplicons and at least one fluorescent labeled fungus specific amplicon; and the step of hybridizing comprises hybridizing the at least one fluorescent labeled COVID-19 virus specific amplicon and the at least one fluorescent labeled fungus specific amplicon to the plurality of nucleic acid probes each having a sequence corresponding to the sequence determinant in the COVID-19 virus and the at least one fungus.
In this embodiment, the fungus is any virus including, but not limited to a respiratory disease-causing fungus. Examples of fungus include, and are not limited to a Histoplasma species, a Coccidioides species, a Blastomyces species, a Rhizopus species, an Aspergillus species, a Pneumocystis species and a Cryptococcus species.
In any of the above embodiments, the method steps for detecting the virus, the bacterium and the fungus are concurrently performed in a single assay with the COVID-19 virus detection steps described above. This is advantageous since it enables streamlined detection of COVID-19 virus and the other pathogens in a one assay. Further in this embodiment, the methods described above may be used to detect in any combination, a COVID-19 virus, another virus, a bacterium, or a fungus.
Also, in any of the above embodiments, the imaging step further comprises calculating an intensity for the fluorescent signal. The calculated intensity is correlated with the number of genomes of the virus, bacterium, and fungus in the sample. The measured intensity is a function of the number of such genomes in the sample. Based on analysis of pathogen-free samples, an experimentally determined intensity threshold is established for the hybridization to each probe on the microarray, such that a fluorescent intensity above that threshold signifies the presence of nucleic acids for the virus, bacterium or fungus, while fluorescence intensities below the threshold signifies that the virus, bacterium or fungus was not detected respectively.
Further to this embodiment, the imaging step further comprises calculating an intensity for the fluorescent signal. The calculated intensity is correlated with the number of genomes in the sample. The measured intensity is correlated with the number of genomes in the sample. Based on analysis of pathogen-free samples, an experimentally determined intensity threshold is established for the hybridization to each probe on the microarray, such that a fluorescent intensity above that threshold signifies the presence of pathogen specific nucleic acids, while fluorescence intensities below the threshold signifies that the pathogen was not detected.
Described herein is a method for detecting a COVID-19 disease in a sample such as a nasopharyngeal swab, a nasal swab, a mouth swab, a mouth wash, an aerosol, or a swab from a hard surface. The sample is mixed with a chemical stabilizer after sample collection. The stabilizer prevents RNA degradation during storage and transportation prior to RNA isolation. The isolated RNA is a total RNA preparation comprising viral and non-viral RNA including COVID-19 virus RNA that is used without further purification. This RNA preparation is used in a combined reverse transcription and asymmetric PCR amplification reaction to generate fluorescent labeled COVID-19 virus amplicons. The fluorescent labeled COVID-19 virus amplicons are hybridized to nucleic acid probes attached to a microarray. This method allows positive hybridization signals to be validated on each sample tested based on internal “mismatched” and “sequence specific” controls. Also described herein is a method for detecting presence of a respiratory virus disease-causing virus, bacterium and fungus in the sample using pathogen specific primers and nucleic acid probes and the same method steps described above. The method steps may be performed concurrently performed in a single assay, which is beneficial since it enables streamlined detection of COVID-19 virus and the other pathogens in a single assay. Any combination of COVID-19 virus, non-COVID-19 virus, bacterium, and fungus may be detected using this method.
In the embodiments described above, the microarray is made of any suitable material known in the art including but not limited to borosilicate glass, a thermoplastic acrylic resin (e.g., poly(methyl methacrylate-VSUVT) a cycloolefin polymer (e.g. ZEONOR 1060R), a metal including, but not limited to gold and platinum, a plastic including, but not limited to polyethylene terephthalate, polycarbonate, nylon, a ceramic including, but not limited to TiO2, and Indium tin oxide (ITO) and engineered carbon surfaces including, but not limited to graphene. A combination of these materials may also be used. The solid support has a front surface and a back surface and is activated on the front surface by chemically activatable groups for attachment of the nucleic acid probes. In this embodiment, the chemically activatable groups include but are not limited to epoxysilane, isocyanate, succinimide, carbodiimide, aldehyde and maleimide. These materials are well known in the art and one of ordinary skill in this art would be able to readily functionalize any of these supports as desired. In a preferred embodiment, the solid support is epoxysilane functionalized borosilicate glass support.
The nucleic acid probes are attached either directly to the microarray support, or indirectly attached to the support using bifunctional polymer linkers. In this embodiment, the bifunctional polymer linker has a top domain and a bottom end. On the bottom end is attached a first reactive moiety that allows covalent attachment to the chemically activatable groups in the solid support. Examples of first reactive moieties include but are not limited to an amine group, a thiol group and an aldehyde group. In one aspect the first reactive moiety is an amine group. On the top domain of the bifunctional polymer linker is provided a second reactive moiety that allows covalent attachment to the oligonucleotide probe. Examples of second reactive moieties include but are not limited to nucleotide bases like thymidine, adenine, guanine, cytidine, uracil and bromodeoxyuridine and amino acid like cysteine, phenylalanine, tyrosine glycine, serine, tryptophan, cystine, methionine, histidine, arginine and lysine. The bifunctional polymer linker may be an oligonucleotide such as OLIGOdT, an amino polysaccharide such as chitosan, a polyamine such as spermine, spermidine, cadaverine and putrescine, a polyamino acid, with a lysine or histidine, or any other polymeric compounds with dual functional groups which can be attached to the chemically activatable solid support on the bottom end, and the nucleic acid probes on the top domain. Preferably, the bifunctional polymer linker is OLIGOdT having an amine group at the 5′ end.
In this embodiment, the bifunctional polymer linker may be unmodified with a fluorescent label. Alternatively, the bifunctional polymer linker has a fluorescent label attached covalently to the top domain, the bottom end, or internally. The second fluorescent label is different from the fluorescent label in the fluorescent labeled primers. Having a fluorescent label (fluorescent tag) attached to the bifunctional polymer linker is beneficial since it allows the user to image and detect the position of the individual nucleic acid probes (“spot”) printed on the microarray. By using two different fluorescent labels, one for the bifunctional polymer linker and the second for the amplicons generated from the viral RNA being queried, the user can obtain a superimposed image that allows parallel detection of those nucleic acid probes which have been hybridized with amplicons. This is advantageous since it helps in identifying the virus comprised in the sample using suitable computer and software, assisted by a database correlating nucleic acid probe sequence and microarray location of this sequence with a known RNA signature in viruses. Examples of fluorescent labels include, but are not limited to CY5, DYLIGHT™ DY647, ALEXA FLUOR 647, CY3, DYLIGHT™ DY547, or ALEXA FLUOR 550. The fluorescent labels may be attached to any reactive group including but not limited to, amine, thiol, aldehyde, sugar amido and carboxy on the bifunctional polymer linker. In one aspect, the bifunctional polymer linker is CY5-labeled OLIGOdT having an amino group attached at its 3′terminus for covalent attachment to an activated surface on the solid support.
Further in this embodiment, when the bifunctional polymer linker is also fluorescently labeled a second fluorescent signal image is detected in the imaging step. Superimposing the first fluorescent signal image and second fluorescent signal image allows identification of the virus by comparing the sequence of the nucleic acid probe at one or more superimposed signal positions on the microarray with a database of signature sequence determinants for a plurality of viral RNA. This embodiment is particularly beneficial since it allows identification of more than one type of virus in a single assay.
DETECTX-RV enables screening for COVID-19 in nasopharyngeal swabs. The microarray has the capacity to test for multiple viral analytes in parallel DETECTX-RV is based on endpoint PCR (rather than qPCR) and is coupled to concurrent analysis of up to 144 distinct nucleic acid probes (rather than just 4 probes for qPCR). This enhanced test capacity enables concurrent testing of 3 different sites (N1,N2,N3) in the SARS CoV-2 genome and further, include a human RNA control (RNAse P). The testing may be performed in triplicate along with a panel of 8 viral controls, enabling confirmation of COVID-19 at a level of experimental specificity of over 10× compared to q-RT-PCR. The DETECTX-RV-V2 microarray differs from DETECTX-RV in the additional inclusion of the newly discovered S-D614G variant in the same assay and an additional amplification step. This microarray is suitable for fully automated testing capable of processing samples in a 96-well array plate format, or the higher throughput 384-well microarray plate format.
The following examples are given for the purpose of illustrating various embodiments of the invention and are not meant to limit the present invention in any fashion.
Tandem PCR (or RT-PCR, then Asymmetric PCR) Reactions to Enhance the Ability to Accurately Detect the Population Density (i.e. Molecules/μL) Near the Lowest Limit of Detection (LLoD)
The first of the two tandem reactions coverts segments of RNA genome into an abundance of amplified DNA. It is a type of Endpoint PCR reaction, such that the original RNA input is amplified 35 cycles, to form an Endpoint PCR product, wherein the input RNA target segments have been amplified to generate a maximal number of DNA amplicons.
The second PCR reaction, which may be a real-time or an endpoint PCR reaction, builds upon the first reaction such that if one or more molecules of DNA or RNA are input into the first reaction, that first PCR reaction produces an amplified DNA segment which has been amplified to yield a sample that may display up to a 10+6 fold increase in strand concentration within the amplicon product (
The second PCR reaction additionally tags the PCR amplified product with a Dye (e.g. CY-3), which enables amplicon detection after microarray hybridization. The second reaction is performed as an asymmetric PCR reaction, such that upon completion of this second “Endpoint” PCR reaction, the product is >90% single stranded (due to the asymmetry of the PCR reaction) with the single strand of interest being the only strand bearing the CY-3 dye probe. This asymmetry allows the product to be used for hybridization without the need for heat denaturation and avoids hybridization artifacts which are otherwise common.
If no RNA (or DNA) were input into the first reaction, none will be amplified (
Equation 1. PCR Reaction #1. Amplifies mass distinction between sample aliquots with (0) vs (>0)
DNA Copies in the original:
1 copy of genomic DNA Target→at 1×2M Copies of Product Targets (M=number of PCR Cycles)
0 copy of genomic DNA Target→at 0×2M Copies of Product Targets (M=number of PCR Cycles)
DNA target signal strength increases after PCR. 1 Copy→1×2M copies
Statistical occurrence of “Negative” events (Pr0) in aliquots of the original sample does not change as a result of PCR Reaction #1.
Pr(0)=exp-(<N>) before and after PCR
In medical diagnostics, food safety and other demanding applications, the LLoD is a crucial test parameter, which is defined by, and directly measured, in the context of samples where, for nucleic acids, the number of microbial or viral genomes in fluid solution are introduced as small (typically 1 μL) aliquots into the PCR reaction at levels so dilute that such single 1 μL aliquots will, via ordinary random sampling statistics, be expected to capture a significant number of “negative” aliquots, i.e. (0) copies of the original nucleic acid genome in each (see
The present invention serves to greatly increase the amplitude of the signal associated with the “positive” events (≥1 genomes per aliquot) relative to the “negative” events where, lacking a template for PCR, PCR does not occur (see Equation 1 and
For example, in the context of Poisson statistics, LLoD can be measured by counting the statistical likelihood of “negative” signals derived from the “negative” (N=0) aliquot events, relative to statistical occurrence of all “positive” events, i.e. signals obtained when (N=1+N>1). Using Poisson statistics, that (positive/negative) event ratio can be used to calculate the average population density (<N>) of nucleic acid target molecules in the original sample aliquot size. For instance, in an ideal assay near the LLOD, where <N>=1 per sample aliquot (e.g. 1 genome per 1 μL) Poisson statistics specify that the statistical likelihood of “positive” vs “negative” signals on repeat 1 μL aliquoting will approach 1-e1/e1≈2. Alternatively, when <N>=3 per aliquot, the ratio of (positive vs “negative) signals on repeat measurement would approach 1-e−3/e−3≈20 which is the standard definition of the LLoD defined by the FDA and the USDA food safety and medical diagnostics.
Thus, as seen in Equation #1 and
The present invention describes the use of a first PCR or RT-PCR reaction (as in
More formally, and in the context of the present invention, when the (n) nucleic acid target sites are distributed throughout the pathogen genome, each can be interrogated by a set (Pn) of at least 9 microarray probes, comprising at least three types of probe (in triplicate). The first probe type (sn) is perfectly matched to a sequence in target site (n) which is chosen to be unique to the pathogen. A second probe type (mn) is identical to (sn) but altered to include at least 10% of base changes to induce mismatches. In addition, there is created at least a third probe type (vn) which is intentionally made to be identical to a sequence in a closely related species variant and to differ in sequence relative to (sn) by at least 10% of base changes (See
The aggregated signal from all three probe types can be compared to each other to define a numerical value for the certainty that the hybridization signal (S) obtained from the pathogen (on probe sn) is statistically different from the hybridization signal (M) obtained on mn and also the signal (V) on vn; and wherein one such representative numerical value could comprise the relationship “Merit”=[S/(M+V)/2] where based on previous analysis of manufacturing and other sources of variance, “Merit” values at >10 would be significant of validated detection of the pathogen in any sample and values for “Merit” at <2 would indicate that the pathogen signal is not detected.
Nucleic acid-based microarray technology is based on the ability to mass produce DNA microarrays in a low cost a 1″×3″ glass slide format. This platform is used for DETECTX-RV and is scalable to 100,000 DETECTX-RV tests per month.
Briefly, viral RNA is extracted from a swab sample (see below) and taken through two Endpoint PCR reactions performed in tandem. The first PCR performs endpoint RT-PCR reactions on COVID-19 RNA to generate a set of primary DNA amplicons, each directed to one of several important regions of the COVID-19 genome N1, N2, N3. The primary DNA amplicons are used as a template for a second PCR reaction which additionally amplifies the primary product, while also applying a CY-3 fluorescent label to it. The second PCR is set-up as asymmetric PCR, a specialized version of Endpoint PCR, which produces a large excess of the CY-3 dye tagged strand of interest. The second PCR product is single stranded and can be used directly for microarray hybridization without clean-up or thermal denaturation. The workflow enables generation of 576 samples of microarray data/shift, which can be doubled with doubling upfront automation of RNA extraction.
The data is analyzed via AUGURY (Augury Technology company New York N.Y.), cloud-based automated software developed at PathogenDx, which can be implemented with modifications as appropriate. The software uses a basic logarithmic analysis to determine the results and is automatically processed and reported without any user interaction. Further, the cloud-based network capability enables data sharing with any number of testing labs needed to support national screening.
Based on the general principles described in background, a microarray test is described with a LLoD at about 1 viral genome per assay and as such more than 10× more sensitive than Q-RT-PCR. Such a >10× sensitivity enhancement enables the ability to detect and speciate COVID-19 at 100 virus particles per swab, which according to the literature is roughly 10× greater sensitivity than any known Q-RT-PCR reaction. Such LLoD performance is a direct result of 3 fundamental principles of tandem PCR coupled to microarray analysis.
Two 30 Cycle PCR Reactions Performed Serially, which Deliver, De Facto, 60 Cycles of Endpoint RT-PCR Amplification Prior to Microarray Analysis
RNA template input held constant, such a 2-step tandem RT-PCR+PCR reaction produces DNA amplicon (to support microarray hybridization) at a concentration that is >3 orders of magnitude greater than the amount of PCR amplified DNA which generates the Cq metric in Q-RT-PCR.
Nucleic acid analysis becomes more sensitive when interrogating multiple copy loci, e.g. rDNA in bacteria or fungi, because one genome becomes represented by (n) identical target nucleic acid strands. In the present microarray test, (n)=6 independent COVID-19 test loci were configured, distributed throughout the genome.
Near the LLoD, where aliquot sampling becomes stochastic (see Equations 1 and
Equation #2. The Analysis of Multiple Loci per Genome. The Effect on LLoD. Pr(0)=e−<N>, where <N>=the population average of events per sample aliquot=(n×<g>), where <g> is the average number of genomes per sample and n is the number of loci analyzed per genome.
LLoD is defined by the FDA as Pr(0)=1/20=e-<N>, thus at LLoD <N>=3 (3 average events per aliquot).
For 1 Locus analyzed per genome, (n)=1 and <N>=<g> and LLoD=3 genomes per aliquot (<N>=3).
For 6 Loci analyzed independently per genome, (n)=6 and <N>=6×<g> and LLoD=3/6=0.5 genomes per aliquot (n=6).
(n)=6 independent target loci per genome is easily obtained in a microarray test. In that important case, due to the multiple sampling (n=6) per genome, Equation #2 shows that the same 19/20 LLoD limit would be satisfied by a population average of 3/6=0.5 COVID-19 genomes per test. Thus, simultaneous measurement of (n)=6 independent target loci per COVID-19 test (each being an independent assay of the same genome) can be seen to reduce the LLOD 6-fold from about 3 COVID-19 genomes/assay to about 0.5 COVID-19 genomes/assay.
The widely used TaqMan qPCR technology, like all similar Molecular Beacon technologies, is based on deployment of PCR Primers (to amplify the RNA target) and the use of a Probe, to detect by hybridization, the PCR amplicon. Therefore, TaqMan Q-RT-PCR or its various Molecular Beacon equivalents are formally analogous to microarray hybridization analysis, which also relies on deployment of PCR amplification, then detection by hybridization.
Near the LLoD, the sensitivity of all nucleic acid tests (Q-RT-PCR and microarray hybridization included) become limited by the ability to distinguish “positive” signals from background, via the knowledge of a Threshold value to distinguish them from each other. In Q-RT-PCR, the analytical parameter of interest to define a “positive” signal is Cq. In microarrays, the analytical parameter is Relative Fluorescence Units (RFU).
For Q-RT-PCR, background discrimination is based on the setting of a “Threshold” which is based on accumulated historical data, or external “no template controls” run in the same batch, along with actual samples, but external to the sample aliquots themselves. Thus in Q-RT-PCR, in nearly all cases, including the current COVID-19 Q-RT-PCR assays, the crucial calculation of “positive” Cq values vs “negative” signals is performed by extrapolation of an external Threshold and not by direct reference to internal standards within the same sample aliquot (
As such, the capacity to detect positive signals, near the LLoD, where Cq values are typically at or greater than 35, becomes subject to sample-to-sample variation extrapolated from other measurements, rather than from the sample itself. Such extrapolation is a source of systematic calculational error to reduce the statistical certainty of distinguish “positive” from “negative” signals.
In the present Example, the DNA microarray test performs 144 individual hybridization tests in parallel for each sample aliquot tested. For each of the (n)=6 hybridization tests being used to detect COVID-19, 3 different “specific” probes are used to detect the presence of each of the (n)=6 viral cDNA targets/genome, along with 3 “mismatched” hybridization probes for each of the 6 target loci.
Thus, for the seminal parameter of importance to LLoD in microarray analysis (“positive” probe hybridization) the threshold, which defines the signal as being distinct from a “negative” is obtained for every sample by direct experimental numerical comparison. This is achieved by comparing a set of three “specific” vs three “mismatched” probes vs 3 or more species specific probes (
Testing of the microarray in this Example is focused on demonstrating that the LLoD for COVID-19 analysis is superior by an order of magnitude relative to that obtained by any of the known Q-RT-PCR assays. Such demonstration is done by third party testing on matched sample aliquots near the LLoD for microarray analysis relative to multiple commercial Q-RT-PCR COVID-19 tests.
Q-RT-PCR technology has been widely implemented to test for COVID-19 among patients. Q-RT-PCR has been shown to have significantly high false negative rates in the range of 15% up to 48%, requiring re-testing (with the same level of inaccuracy). Therefore, it is challenging to detect low viral loads for patients who are asymptomatic or pre-symptomatic (3,4).
False negative rates seen for Q-RT-PCR is due in part to the poor signal/noise ratio associated with Q-RT-PCR when it is implemented in the limit of low viral load and may be due to the nature of the principal Q-RT-PCR observable (Cq). It may also be due to poor control of RNA stability during and after collection.
Cq refers to the point at which PCR amplification of the COVID-19 genome produces enough product to be resolved from background (
The LLoD (Solution): DETECTX-RV, an Alternative Technology Platform. The problems associated with LLoD is well known in the detection of other pathogens. The nucleic acid-based microarray technology of the present invention obviates LLoD limitations. The nucleic acid-based microarray technology is based on the ability to mass produce DNA microarrays in a low cost a 1″×3″ glass slide format.
By inspection of typical Q-RT-PCR data (
A direct practical result of such saturation is that in many cases, such saturation allows samples to be pooled, as might be useful to expedite very large-scale epidemiological screening. See for instance, representative data as in
In this example, COVID-19 is the primary analyte, plus multiple coronavirus targets [SARS-CoV, MERS-CoV, CoV 229E, CoV OC43, CoV NL63, CoV HKU1] plus Influenza [type A and B] as species variants (Table 4).
The extra content available in the microarray format allows a very large panel of COVID-19 target sites (n=6) to be measured in parallel and in triplicate. The other six coronavirus targets and two influenza targets are included and are being used as both controls and as a universal screening tool for coronavirus and influenza.
For each of the n=6 unique SARS-CoV2 target loci [N1, N2, N3, ORF1ab, RNA-dependent RNA polymerase (RdRP), E] there are (2) microarray probes (Se), 12 specific probes in total, and 2 mismatched probes (mn) for each, with 10% of intentional base mismatching (i.e. there are 12 mismatched specificity probes). Relative to the twelve COVID-19 specific probes (Se), the 14 species specific controls (vn) are distributed among other coronavirus (SARS-CoV, MERS-CoV, CoV 229E, CoV OC43, CoV NL63, CoV HKU1). In that format, a Positive COVID-19 signal for any one of the set of six loci, deemed valid if it possesses a fluorescence signal strength of >10× background (>10,000 RFU) while at the same time and in the same microarray, the mismatched specificity probe (mn) generates a signal less than 2× background (<2,000 RFU).
The serial application of two PCR Endpoint reactions (RT-PCR, with Asymmetric PCR) creates a type of analysis which is “Binary” in the sense that, an aliquot of specimen which lacks RNA produces no measurable hybridization signal, under conditions where any input with at least 1 genome copy produces a signal of nearly constant, very large signal amplitude. Such behavior is shown graphically in
For LLoD analysis, the highly characterized COVID-19 standards (BEI) have been doped into N=30 separate pooled human nasal secretion samples (Lee Bioscience) along with 20 matched (negative) nasal secretion controls. Subsequent to RNA purification (Zymo kit) the resulting 50 RNA samples were subjected to PCR-Microarray.
As seen from
To confirm the intrinsic detection limit inferred from the LLoD analysis (
Comparison to matched Q-RT-PCR data (N1 target) shows performance typical of all such Q-RT-PCR tests. The data obtained below 5.5 copies per reaction becomes indistinguishable from the detection threshold (Ct≈35) as defined by negative controls. Thus, the detection limit for DETECTX-RV (<1 genome copy per reaction, is more than 5× lower than that of the present Q-RT-PCR assay. A summary of COVID-19 hybridization statistics is shown in Table 5.
Based on the substantial Signal/Background ratio obtained with DETECTX-RV near the LLoD (
Such 10-fold pooling is shown in Table 6, wherein a single sample near the LLoD (50 copies/ml) is mixed with an equal volume of 9 samples lacking COVID-19 RNA, yielding a net viral load of about 5 copies/ml. As seen in Table 6, all 3 COVID-19 markers are detected in each of the pooled samples tested. The data show both of the important attributes of “Binary” sample Collection. The signal strength at 5 copies/ml, is about 30,000 RFU, which is identical within experimental accuracy to the 50 copies/ml sample used for pooling (Table 6) and in turn identical within experimental accuracy to identical un-pooled samples at 100 copies/ml (
The Problem: The platform limit of Q-RT-PCR can be multiplexed to resolve four analytes in parallel, based on the four principal emission channels on most devices including CDC, LabCorp, Quest (N1,N2,N3, P). This limit may be exceeded as evidenced for Abbott (RdRp, N), Cepheid (E,N2,P) and Eurofins (N,P). However, the “maxed” capacity suggests that the known Q-RT-PCR assays will not be able to accommodate additional testing complexity, such as might arise if alternative COVID-19 clade variants were to emerge. A recent publication has suggested however, that a stable variant has been detected comprising a mutation in the spike protein S-D614G, which has been hypothesized to be more virulent.
Based on test content capacity alone, detection of both SARS-CoV2 and the S-D614G spike protein mutant will prove difficult to detect on the same q-RT-PCR test. Thus Q-RT-PCR might not be useful as a tool to screen for both variants.
The Solution. The process by which new coronavirus content can be added to DETECTX-RV is very efficient. It is based on the robust probe capacity of the arrays (144) and on the highly standardized methods of PCR primer design and microarray probe design (at one base pair hybridization specificity). As an example, the presumed importance of the S-D614G mutant was only recently published (Apr. 29, 2020). The variant comprises a SNP G-A transition converting Asp to Gly. New probes specific for the wild type and new variant along with a set of mismatched control probes were designed within a day, and submitted for fabrication, and were completed May 11, 2020. Microarray fabrication with these new probes was added to an otherwise identical DETECTX-RV microarray and were completed on May 15, 2020. In parallel, a pair of test amplicons were designed and produced by SGI methods possessing the wild type and new COVID-19 SNP. In parallel, 4 candidate PCR primer pairs have been designed. Probe selectivity was confirmed with the SGI template, and in parallel, inclusivity and exclusivity confirmed experimentally with the full panel of coronavirus research standards in-house from ATCC-BEI.
The Problem. While Q-RT-PCR can deliver adequate test specificity it is well-known that the TaqMan probe-template interaction does not adequately resolve SNPs in many cases (6,7) due to the fact that in a TaqMan assay, primer binding, probe binding and the Taq exonuclease activity must all occur at the same time and thus cannot be optimized independently. The problem is exacerbated in the present case (S-D614G) because the SNP generates a “run” of 3G's, which are difficult to accommodate.
The Solution. In this respect, the microarray technology of the present invention is beneficial as it has the capability of routinely generating “all or none” SNP discrimination due to uncoupling of probe binding from PCR. Further, a separate washing step is included for improved specificity. Here, a first set of hybridization tests are shown on a set of probe candidates to detect and resolve the SNP variants which define SARS-CoV2 Clade variation at the Spike protein (D-614G). Methods of probe design were used. Array manufacture was performed in the standard 12-well format, but all other aspects of probe formulation and deposition were identical to those deployed in the 96 and 384 well Plate formats. Six PCR primer pairs were designed and optimized for the standard Tandem PCR (RT-PCR+Labeling PCR) amplification process. Since both “sense” and “antisense” probes were tested, different asymmetric Labelling PCR reactions were deployed, which differed in which of the 2 PCR primers had the 5′-CY3 label in the second PCR reaction (labeling PCR).
To evaluate hybridization feasibility in 12-well slides, 50 probes candidates were printed on the slides to detect and resolve the 2 SNP variants which define SARS-CoV2 Clade variation at the Spike protein (D-614G). Proprietary methods of probe design developed at PDx (PathogenDx, Scottsdale, Ariz.) were used in the design. All aspects of probe formulation and deposition were identical to those used for 96-well and 384-well plates.
A PCR primer pair was designed and optimized for standard (tandem) 2-Step RT-PCR and labeling PCR. Since both “sense” and “antisense” probes were tested, different asymmetric labelling PCR reactions were deployed, which differed in which of the 2 PCR primers had the 5′-CY3 label. The template for this study was a pair of synthetic templates. Each template contained the defining SNP (A or G) embedded in the Wuhan reference sequence for the Spike protein.
Subsequent to standard hybridization and washing in the slide format (similar to 96-well and 384-well plate format), the two SNP variants were resolved with signal (relative fluorescence units, RFU) strength differences in the range from 15-40 (see Table 7) which approaches the theoretical limit for resolving SNPs by hybridization. For expediency, two “Sense Strand” and Two “Antisense Strand” candidates from the probe set were chosen for inclusion in the 384-well Plate “Mini-RV” print content. All four of these probes displayed very good sensitivity and SNP specificity. This Example conforms that the present tandem PCR (RT-PCR+labeling PCR) reaction coupled to microarray hybridization can cleanly resolve two SARS-Cov2 variants which differ by a single RNA base change the Spike protein.
The Problem. Scalability of test capacity is a huge challenge particularly during a pandemic. Assuming 1000 COVID-19 test sites throughput the US, this would require the ability for each site to support at least 10,000 tests/shift/site. At present Roche and Abbott lead the pack with Q-RT-PCR capacity, amounting to 250-900 tests/shift and 150 tests/shift respectively. This microarray technology supports population scale nucleic acid screening.
The Solution. In the present deployment of DETECTX-RV, the core array format (12×12) is deployed as 12 tests per slide. Eight such slides are routinely processed in parallel with ordinary fluid handling, thus allowing multiples of 96 tests in parallel.
At present, coupled to ordinary 96-well magnetic bead RNA purification, the hybridization steps are in all cases faster than the RNA preparation. Automation and system integration are deployed with industry leading partners. Technologies, which have been integrated for DETECTX-RV (
The full content of the original DETECTX-RV assay is described in Example 2 and Table 4. Table 8 shows a variant (DETECTX-RV-V2) of that Pan Coronavirus format. It is based on SARS-CoV2 analysis at (N1,N2) as in the original assay and differs in the inclusion of 2 new microarray probes and an additional RT-PCR primer pair to interrogate the recently described novel S-D614G mutant (5) in the same assay.
The streamlined DETECTX-RV-V2 assay deploys 12 microarray probes, which when printed in N=12 multiplicity, become a highly redundant 144 probe array suitable for printing in the present 12-well slide format, and in the more automation-friendly 96-well Society for Biomolecular Screening (SBS) format (
The 96-well late format (
The 96-well DETECTX-RV-V2 workflow has been integrated into off-the-shelf Tecan automation (Freedom Evo-2 100 Base) beginning with magnetic bead-based RNA extraction (Zymo) and ending with automated microarray hybridization and washing. The intervening PCR reactions are mediated by Tandem Thermo-ABI cyclers and imaging is performed on a Sensovation CCD based imager. Data generated is fed into AUGURY software discussed in Example 2 for autonomous plate reading, microarray data compilation and analysis.
The major strength of the DETECTX-RV-V2 technology is its large-scale public health application in any setting including at-home, at-work, healthcare institutions and transportation hub sample collection for diagnosis and detection of active and asymptomatic individuals. Current use of nasopharyngeal swabs is not suitable for such collection, due in most cases to the difficulty of sample collection and the instability of RNA on such swabs, using the currently used transport media of the day.
The Tecan robot or other commercial equivalents can process multiple 96-well plates in parallel, thus sample throughput of (6) 96-well microarray plates/shift is possible (
Deployment of DETECTX-RV-V2 enables 12,000 arrays/day in a 96-well array plate format providing a 360,000 arrays/month capacity. While DETECTX-RV-V2 will retain its 12×12=144 element structure (in 7 mm wells), the 384-well structure (3.5 mm wells) will accommodate a 6×6=36 probe array. The core probe content for SARS-CoV2 (N1, N2)S-D614G variant and Human RNase-P (P) internal control can all be included along with SARS CoV and MERS CoV as species specificity controls as 12 probes, printed in triplicate. The DETECTX-RV-V2 format may be modified to include pan Influenza A and Influence B probes to generate a targeted pan-respiratory virus test (DETECTX-RV-V3). The DETECTX-RV-V3 format has substantial benefits since it readily adapts to increase system testing throughput to more than 2,104 tests/shift, which exceeds existing commercial testing technologies and at the same time achieves a 3× reduction in test cost, from manufacturing & reagent economies of scale.
DETECTX-RV-V2 & DETECTX-RV-V3 are each manufacturable in 24 hours with a single printer in batches of 62 plates, comprising 6,000 arrays/day (96-well plate) 24,000 arrays/batch/day (384-well plate). Each printer completes two batches per 24-hour day.
The entire suite of Pan-Coronavirus content (Table 4) has been designed, developed and manufactured and is resident in the DETECTX-RV version of the assay. The full Pan-Corona Respiratory Virus content suite is validated using standardized viral reagents from ATCC-BEI, which are spiked into the same matrices (nasal and saliva). The test format employed for such expanded validation uses the LLoD and N=30 repeats testing protocols.
Early stages of COVID-19 clade development are in progress, which could be selected for stable changes in environmental durability, virulence or acute symptomology. PathogenDx monitors such data on a daily basis. At such time that solid evidence emerges for development of stable COVID-19 clade variants, new content was immediately added to DETECTX-RV (
The process by which new coronavirus content can be added to DETECTX-RV is very efficient due to the robust probe capacity of the arrays (144), and the highly standardized methods of PCR primer design and microarray probe design (at one base pair hybridization specificity).
If a new SARS-CoV2 subtype were identified in the literature, based on one or more regions with local sequence change in one of the domains already Interrogated in DETECTX-RV (N1, N2 or N3), PCR primer design would not change. The only modification is design of one or more new probes specific for the new variant added to the existing DETECTX-RV microarray. In parallel, a test amplicon would be produced by ordinary SGI methods possessing the new COVID-19 sequence markers. Probe selectivity would be confirmed with the SGI template, and in parallel, inclusivity and exclusivity confirmed experimentally with the full panel of coronavirus research standards in-house from ATCC-BEI.
On the other hand, if the new content were in regions not yet being interrogated, the process remains the same, with the added task of designing and fabricating a primer pair to amplify the COVID-19 region of interest. The primer design process occurs in parallel to probe design with a 2-week turnaround for the desired DETECTX-RV test modification.
The DETECTX-RV assay coupled to nasopharyngeal swab collection is presently being launched into CLIA certified labs for human diagnostics screening. Its oligonucleotide probe content (Table 4) comprises a 12×12 array, at present, with RNA targets comprising sites within a set of 10 respiratory viruses and a human RNA control (RNase P). Of these, SARS CoV2, SARS-CoV and SARS COV2 (mutation) support pandemic testing. The remainder of the test content (other coronaviruses and Influenza) are present as probes within the present 12×12 array and used as specificity controls.
In the Tandem, Asymmetric, Two-Step implementation of the present invention, DETECTX-RV workflow begins with viral RNA that had been extracted from a nasopharyngeal Swab Sample followed by two Endpoint PCR reactions in tandem. The first PCR, an “Enrich” PCR (
The DETECTX-RV workflow shown in
The COVID-19 pandemic has confirmed what many had known from field study of zoonotic disease: namely that the “Viral Transport Media” (VTM) used to collect virus on swabs, are poor stabilizers of viral RNA. To address this, a novel chemical stabilizer from GENTEGRA LLC (GTR) as well as inexpensive polymer stabilizers (PVS) along with well-known lab-based RNAse inhibitor (RNA-Shield) are used to allow for stable field collection of respiratory virus samples on swabs without refrigeration. Stabilized swab collection (COVID-19, Coronavirus and Influenza stability over one week at 30° C.) enables better clinical collection of nares swabs and saliva fluid also enables at-home nasal swab collection for population scale screening in centralized labs. Emphasis is to support very large-scale clinical collection (nares) plus at-home (lower nasal) collection.
A modified swab design that includes chemical stabilizers of viral RNA initiated in collaboration with GENTEGRA LLC enables samples to be transported at ambient temperature. This improved collection design may be employed with the DETECTX-RV-V2 platform to support very large-scale clinical collection and at-home collection.
The technologies for integration into DETECTX-RV are approved for in vitro diagnostics use for the type of workflow required for DETECTX-RV testing—RNA preparation via magnetic beads (Tecan) RT-PCR and PCR (Thermo Fisher Scientific), open architecture, ambient temperature binding and washing (Tecan) and microarray imaging (Sensovation AG). The AUGURY software has all functionalities in place to support DETECTX-RV data acquisition and analysis. Its capacity to manage and upload such data into a secure cloud network is also complete and fully validated for RUO use.
A “mouthwash” based saliva collection technology (QUIKSAL) is employed for collecting saliva samples. In a separate set of studies, 200 nasopharyngeal swabs are collected per the standard RevolutionDx and Lucid Lab protocols along with matched QuiKSal mouthwash collection from the same individual (400 matched swabs and Saliva). The swab and half of the mouthwash is analyzed in accordance with standard Q-RT-PCR workflow, while the remainder of the mouthwash was split and shipped at ambient temperature and −20 C in transport medium for analysis at PathogenDx on the DETECTX-RV-V2 microarray. The samples analyzed at PathogenDx have no associated personal identifiers or medical information other than the Cq values obtained from Q-RT-PCR testing at RevolutionDx.
Feasibility of Sample Pooling from Swabs and Saliva for Population Scale Screening
Pooling of swab and saliva samples among pre-symptomatic individuals is a powerful tool to enable contact tracing. This is established by the findings that demonstrated pooling of specimens with the highest COVID-19 load from at least 64 nasopharyngeal swab samples via Q-RT-PCR is free of false negatives when the input (positive) sample used for pooling is a clear, “strong positive” and characterized by a Cq value<30 (
Sample pooling is a powerful public health screening tool. However, for the most useful pooling levels (N≥10) for many COVID-19 positive samples (those with Cq>30) Q-RT-PCR generates an unacceptably high “Pooled False Negative Rate”. If that occurs, sample pooling in combination with Q-RT-PCR would not be adopted as a routine public health or industrial hygiene tool.
As shown in Table 6, data with contrived nasopharyngeal samples near the LLoD (at 50 genome copies/ml) suggests that DETECTX-RV may have the sensitivity needed to enable expanded pooling (N=10) with a reduced risk of false (pooled) negatives. Therefore, contrived samples are used to refine the sensitivity and specificity of N=10 pooling similar to that shown in Table 6, with technical emphasis on increasing cycle number from 30-35 in the Enrichment RT-PCR reaction. Pooling is then performed prior to RNA extraction, on the same swab and saliva samples freshly obtained. Raw samples (unstabilized swabs or stabilized swabs or stabilized saliva) are measured immediately by both DETECTX-RV-V2 and by Q-RT-PCR. Immediately upon identification of “true “positives”, the set is divided into quartiles, based on the semi-quantitative Q-RT-PCR data (that is, Very High, High, Medium, Low) for viral load based on the Cq value associated with each. 20 uL of each such positive sample are immediately mixed with 20 uL of 9 of the many “negatives” to yield 200 μL of pooled sample and transferred directly into Zymo RNA lysis buffer for freezing prior to RNA extraction. This approach permits the nasopharyngeal swab or saliva studies to yield up 40-80 unique N=10 pooled samples, where data for each pooled sample (Table 8) is directly compared to the “positive” from which it originated.
PathogenDx received 50 blinded nasopharyngeal swab samples in flash frozen Abbott Transport Media from Testing Matters Laboratory (TM Labs—Sunrise, Fla., CLIA certified) to evaluate the performance of the PathogenDx DETECTX-RV assay in comparison to the FDA-EUA approved Abbott Real-Time SARS-CoV2 qPCR assay.
Each of the 50 samples were collected on the same day/same time, one sample was collected from the right nostril and one from the left nostril. The two separate samples (each separately labelled and stored identically in transport medium) were taken back to TM Labs where one sample was flash frozen and shipped to PathogenDx and the second sample was processed and screened according to the Abbott Real-Time SARS-CoV2 qPCR assay FDA-EUA protocol. The results from the Abbott testing at TM Labs were shared after PathogenDx had screened the 50 samples using the DETECTX-RV assay.
The 50 matched samples that were sent to PathogenDx, arrived frozen on dry ice and were stored at −80° C. until use. The samples were thawed on ice and 400 μL of the 2 mL sample was used as the input for the Zymo Quick-DNA/RNA Viral MagBead purification. The purified RNA was then used to screen for SARS-CoV2 in these patient samples according to the PathogenDx product insert using the Promega AccessQuick RT-PCR system coupled to the PathogenDx PCR and the corresponding microarray test.
There were 50 total samples tested as well as the PathogenDx external positive and negative controls. Table 9 shows the results of the analysis.
Run 1—The DETECTX-RV assay demonstrated 100% concordance with the samples called positive (N=7) using the Abbott system and 93% concordance with the samples called negative (N=43) using the Abbott system. The PathogenDx, DETECTX-RV assay identified 2 additional samples as positive, that were identified as negative by Abbott testing and one additional sample as needing to be rerun. Measurements were repeated for all 9 samples of the samples that were identified as positive using the DETECTX-RV assay.
Run 2—The DETECTX-RV assay demonstrated 86% concordance (6 DETECTX-RV/7 Abbott) with the samples called positive (N=7) using the Abbott system. The one sample that was discordant was identified as a rerun on the second run. The rerun confirmed the positive signal from Run 1 for the two samples that were identified as negative by Abbott testing. The one previous sample that was identified as a rerun came back as negative on the second run. Table 10 summarizes the results from Run 1 and Run 2.
One of the greatest challenges in performing environmental monitoring of air and surfaces for viral contamination is the collection and stabilization of the viral RNA prior to analysis. To overcome this challenge, the strategy of utilizing a dilute RNA stabilization solution (from GENTEGRA LLC) called “ATA” here was evaluated at 1:40 dilution in 1×PBS and/or DNase/RNase Free Water for stabilization of COVID-19 RNA collected from surfaces on swabs or from the air into a fluid collection solution, using a device from Bertin Corporation, as an example.
To determine if the air is contaminated with bacteria, fungi, and/or virus air was collected using the Coriolis Micro Air Sampler from Bertin. In this utilization, the stability of viral RNA was evaluated during and up to 72 hours post collection. In this demonstration the GENTEGRA RNA stabilizer (“ATA”) was diluted at 1:40 dilution in 5 mL of 1×PBS, pH 7.2 or molecular grade water. Purified 5 μL of SARS-CoV2 RNA was spiked at 200,000 copies/μL directly into the collection cone and ran the instrument to dryness, which took ˜30 min, during which the spiked sample was exposed to the particulate contamination resulting from @2m3 of collected air input. Post air collection the dry viral RNA plus dried stabilizer and accumulated airborne contaminants were resuspended in 1 mL of molecular grade water and stored the samples at room temperature (0, 24, 48, and 72 hours) until RNA purification was performed. The RNA was extracted and purified using the Zymo Quick DNA/RNA Viral MagBead collection kit and the samples were ran on the DETECTX-RV assay by PathogenDx. RNA collection and stability for the entire 72 hour period as demonstrated in the
To determine if the surface is contaminated with a microbe (COVID-19 virus in the present example) surface swab samples were collected using nylon flocculated and rayon swabs. In this utilization, the stability of viral RNA during and up to 24 hours post collection was evaluated. In this demonstration, the “ATA” RNA stabilizer was diluted 1:40 in 5 mL of 1×PBS, pH 7.2 or molecular grade water. Purified 5 μL of SARS-CoV2 RNA was spiked at 200,000 copies/μL then applied it directly onto a stainless-steel surface. The swab was removed from its sterile case and three drops of the dilute “ATA” stabilizer were placed onto the swab to moisten it. The surface was swabbed to collect the viral RNA. The swab was placed directly back into the sterile container and allowed to sit at room temperature for 24 hours. Post surface collection and either (0 hrs) or (24 hrs) of ambient temperature swab storage, 1 mL of 1×PBS, pH 7.2 was added to the swab in the container and vortexed for 10 seconds. 400 μL of the resuspended viral RNA was removed for viral RNA preparation. The RNA was extracted and purified using the Zymo Quick DNA/RNA Viral MagBead collection kit and the samples were run on the DETECTX-RV assay, monitoring the fluorescence signal from the COVID-19 (N3) region. The positive control constitutes an identical matched, unprocessed spiked COVID-19 samples that had not been applied to the surface or gone through swabbing or storage. The data demonstrate RNA recovery and stability from a surface swab, subsequent to ordinary ambient storage of the swab for 24 hrs, as assessed by analysis via the present invention, as demonstrated in
The purpose of this study was to demonstrate that the present invention could be used to detect COVID-19 RNA in a novel oral rinse solution (QuiKSal from CLC Corporation) which had been spiked into it at clinically meaningful levels, then analyzed subsequent to several days of unrefrigerated ambient temperature storage, to emulate overnight shipping from point of collection to a central lab for COVID-19 analysis by the present invention. Two versions of QuiKSal were tested. One possessed a tracking Dye (SOW+) and one without the dye (SOW−).
The QuiKSal procedure asks the patient to swish 1 mL of the QuiKSal and spit the QuiKSal into the sterile storage container. The collection procedure was mimicked by spiking in a high and a low SARS-CoV2 RNA into 1 mL of QuiKSal. Eight 1 mL aliquots of Oral Rinse Solution were created, with and without SARS-CoV2 RNA spike. Two of the spiked sampled aliquots had 200,000 copies/mL (high) of a SARS-CoV2 standard (Integrated DNA Technologies) while the other six aliquots were spiked to 20,000 copies/mL (low). Following the addition of the RNA to the samples the samples were stored from 0 to 72 hours at room temperature to evaluate the stability of the RNA in the QuiKSal mouthwash. Following incubation, the RNA was isolated using the Zymo Quick-DNA/RNA Viral MagBead kit by removing 400 μL of the QuiKSal for sample preparation per the manufacturer's instructions. Following sample preparation, the samples were analyzed using the PathogenDx DETECTX-RV test, based on the teaching of the present invention.
Array data (
The present invention was capable of detecting COVID-19 RNA from the QuiKSal oral rinse with or without dye. COVID-19 RNA in that stabilized mouthwash was detectable via the present invention for up to 72 hours at room temperature.
96-well plates were printed with the hybridization probes under conditions optimized to eliminate dust and fiber contamination in the wells. Optical Inspection suggested that there are no measurable failures in printing (
A small number of validated clinical nasopharyngeal swab samples obtained from Boca Biolistics having 11 positives and 1 validated nasopharyngeal negative control were subjected to standard 2-step tandem RT-PCR (RT-PCR+PCR). Standard hybridization and washing were performed with an increase in hybridization (136 μL) and wash (200 μL) volumes, followed by imaging from the bottom of the fully assembled 96-well plate (plate #9901005001).
A small number of validated clinical nasopharyngeal samples obtained from Boca Biolistics having 11 positives and 1 validated nasopharyngeal negative control were subjected to standard 2-step tandem RT-PCR (RT-PCR+PCR). Samples were analyzed on the 96-well slide format with direct comparison to match samples on the 12-well slide format and are shown in Table 11. Signal intensity was optimized by increasing labelling PCR reaction volume and sample volume.
The probe content for the smaller version of DETECTX-RV (Mini-RV V1) was designed and is shown in Table 12.
SARS-CoV2 probe specificity and characteristics of probe prints were evaluated for clinical nasopharyngeal swab samples.
384-well test print—9980001001
SARS-CoV2 standard at 200 copies/reaction (Exact Diagnostics LLC)
Probes were printed in triplicate and amplified at 200 copies/reaction. A pooled PCR sample was created from 24 individual PCR reactions amplifying SARS-CoV2.
By increasing the amount of UV cross linking from 300 mJ to 500 mJ, the signal strength obtained for COVID-19 microarray analysis in the 96-well and 384-well plate format was comparable (Tables 13 and 14) to that obtained with 12-well slide hybridization as assessed by LLoD and clinical specimen analysis.
Methods to improve signal strength and overall performance were analyzed for 96-well plates (
Validation of Asymmetric One-Step RT-PCR using purified SARS-CoV2 RNA
1. 12-well glass slides—99030002 print series.
2. DETECTX-RV kit.
3. Purified SARS-CoV2 RNA (ATCC, NR-52285)
4. Labeling primers
To determine if the Tandem 2-step (RT-PCR+Labelling PCR) reaction can be combined to a single step (Asymmetric One-Step RT-PCR) to reduce assay times, first, different primer ratios (labeled:unlabeled) were used in the PCR reaction to establish optimal cycle number to achieve sensitivities similar to the 2-step reaction (LoD˜2 copies/reaction, 125 copies/mL)
Four different primer concentrations and ratios (labeled:unlabeled 4:1, 4:1, double concentration, 8:1, 2:1) were used. Three different cycling conditions were used over a dilution (500 copies/reaction=62,500 copies/mL to 2 copies/reaction=125 copies/mL) of purified SARS-CoV2 RNA. The purified SARS-CoV2 RNA was diluted in sterile water from 500, 250, 100, 50, 25, 10, 5 and 2 copies/reaction. NTC (No Template Control) and External extraction controls were also used. The PCR parameters were as follows;
Cycling conditions: 35, 40, 45
RT-PCR Program: 45° C., 45 min
PCR Program:
In conclusion, Asymmetric One-Step RT-PCR provides a slightly higher LLoD compared with the 2-step tandem RT-PCR (Asymmetric One-Step RT-PCR 5-10 copies/reaction=1250-625 copies/mL versus tandem RT-PCR 2 copies/reaction=125 copies/mL).
Using the same sample as used for the analysis shown in Example 11 and Table 13, a formal LLoD was obtained for the Asymmetric One-Step RT-PCR (Tables 15 and 16), which was determined to be relatively superior to that discussed in the previous section (Example 12, ‘Optimization 1’). The data in Tables 15 and 16 are identical within experimental accuracy to that observed using the 2-Step (RT-PCR+PCR) reaction.
Furthermore, data obtained for clinical sensitivity and specificity analysis using 30 Positive and 30 negative nasopharyngeal swab samples (Table 17) showed unaltered specificity and negative predictive value (NPV), but a preliminary reduction in sensitivity from 100% to 79% due to a general reduction in hybridization signal strength.
The reduction in signal strength was remedied by the following modifications:
Performance of the 384-well DETECTX-RV microarray was performed against a set of 30 positive nasopharyngeal swab samples. This analysis differed from the previous example (Example 12, ‘Optimization 2’) in that, RNA was freshly extracted and used immediately without freeze/thawing or storage, the primer concentration is increased 2-fold to 400 nM. Samples were evaluated based on average and standard deviation signal intensity, sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV).
An improved performance was observed with clinical isolates (Table 18) over the previous optimization described above (‘Optimization 2’). An improvement in clinical sensitivity was observed for all probes (range of clinical sensitivity, 88%-100%). The overall AUGURY readouts however report a specificity of 100% since AUGURY aggregates hybridization data from all three independent loci tests.
Thirty positive and 30 negative samples were used for LLoD analysis. Freshly prepared and negative NP-VTM samples from TriCore (New Mexico) samples doped with purified. SARS-CoV-2 RNA standard were used. All 60 NP-VTM samples were analyzed using the 4:1 asymmetric PCR primer ratio (at 2× higher concentration). in the Promega AccessQuick RT-PCR system. Each hybridization probe was analyzed individually to yield average and standard deviation of signal intensity (RFU), sensitivity, specificity, PPV and NPV. Tables 18 and 19 summarizes the results of the LLoD and the Clinical Sensitivity/Specificity analysis respectively.
The LLoD analysis (Tables 19 and 20) was found to be identical within experimental accuracy to the Asymmetric One-Step RT-PCR optimization obtain earlier ((‘Optimization 3’) as well as the 2-step RT-PCR reaction discussed above. Thus, these data confirm what was previously seen in 12 well slides, namely that the LLoD obtained with the Asymmetric One-Step RT-PCR protocol is identical, within experimental accuracy to that obtained via the Asymmetric, Tandem 2-Step RT-PCR reaction profile and is not affected by transition from 12 well to 96-well processing.
Analysis of the full set of 30+ and 30− TriCore samples, each previously analyzed via an industry standard Roche predicate q-rt-PCR assay (Tables 21 and 22) yielded clinical sensitivity of 100% for both the Asymmetric One-Step RT-PCR and Asymmetric Two Step RT-PCR methods, for the full set of 30 positive and 30 negative samples.
Clinical specificity at the local probe level on the other hand were not 100%, being about 94% for the Asymmetric One-Step RT-PCR method and 100% for the 2-Step RT-PCR method. Thus, the 2-Step method detected 3 false positives and the Asymmetric One-Step RT-PCR detected 2 false positives in the same set of 30 TriCore Negatives. This discordance was resolved by third party sequencing and it was shown that the positives detected by microarray analysis did in fact contain SARS-CoV-2, which had not been detected by the predicate Q-RTPCR assay.
A detailed analysis of individual probe Clinical Sensitivity revealed that 2 probes ([N1, 1.1], [N2, 1.3]) have generated 1 and 2 false negatives respectively. These 3 rare events did not affect the Overall AUGURY sensitivity, which remained at 100% because of its use of multiple probes to make a call.
Asymmetric One-Step RT-PCR performance with Clinical Isolates in the 96-well format has improved significantly over prior optimizations due to use of RNA extracted from fresh samples and 2× increase of Primer Concentration 2×. Sensitivity among all probes tested has yielded an increase in Average N1,N2 and N3 probe Clinical Sensitivity to about 97% (range=94%-100%). However, the aggregated AUGURY readouts obtained provide a 100% Sensitivity, since Augury aggregates hybridization data from all (3) independent loci tests.
Freshly collected NP/VTM samples (TriCore) matched with a complete set of Roche Cobas 6800 Q-RT-PCR Ct thresholds were used. Analysis was performed on 384-well plates using 30 positive and 30 negative clinical isolates in an RNAse P modified multiplex PCR reaction. The data obtained (Table 23) was in good agreement with that obtained for the 96-well format using previous RNAse P primers (Tables 18-22) and further provided a better RNAse P signal than previously observed (Tables 18-22).
Four sequencing primers with M13 tags were created and the amplicons generated using the Asymmetric One-Step RT-PCR method (Asymmetric One-Step RT-PCR) were analyzed by gel electrophoresis.
TriCore samples identified as “Negative” by Cobas but identified as “Positive” on multiple repeats of the DETECTX-RV assay were sequenced (third party sequencing, University of Arizona). The sequencing data shown in
Clinical samples (30 positive/30 negative NP/VTM, TriCore) tested using the Cobas 6800 platform were used as clinical reference samples to evaluate sensitivity and specificity of the DETECTX-RV assay using Two-Step Tandem and Asymmetric One-Step RT-PCR reaction methods. To confirm accuracy of the DETECTX-RV “Positive” readouts, Sanger sequencing was performed within the N Region, on all 6 discordant samples and one of the many “Positive” samples, which had been identified as “Positive” by both COBAS and DETECTX-RV. The results shown in Table 24 are in agreement with DETECTX-RV—all 6 discordant samples were identified by Sanger sequencing as containing measurable SARS-CoV-2 RNA. Sequence heterogeneity (N) in several of the negative samples was also observed.
Incorporation of Influenza Probes into the Array
Influenza A & B probes and primers were added to the 12-well array during analysis using the 2-step method. The hybridization data (Table 25) show that the influenza probes and primers may be used as such with no further refinement.
Reducing the time taken from obtaining the sample to performing the microarray analysis is expected to significantly increase the number of sample that may be processed per day, a factor that is critical during a pandemic. To establish the feasibility of bypassing the RNA isolation step in the method, experiments were performed using clinical mouthwash samples (QuikSal) from patients diagnosed with SARS-CoV2. The data in Table 26 shows the results of analysis in samples where the RNA extraction step was omitted. It was observed that raising the pH to the ordinary PCR range of pH 9.0 and adding Mg+2 to coordinate EDTA and citrate in the QuikSal, mouthwash enabled microarray analysis in crude samples with no further RNA purification.
RNA Extraction using Zymo Magnetic Beads and RNA loading onto PCR plates for RT-PCR was established using Tecan. Hybridization and Washing Automation for Asymmetric One-Step RT-PCR in 96-well format was completed for the Tecan and a first 96-well plate. It was run through with 20 positive Clinical Isolates (TriCore)+76 negative (water-only) samples. The corresponding 384 well software was also tested with clinical samples using a Tecan code modified for 384-well plate operation, capable of 384-well function with a 96-pipette head.
Approximate time elapsed to process 1×96 well slide
Total time˜9 h
Tip boxes required to process 1×96 well slide
Total tip boxes; 1 ml-7 boxes, 200 μl—6.5 boxes, 50 μI-2 boxes
Two full runs were performed with the Tecan EVO using the 96-well format;
Run 1. Comparison of automation versus manual: A series of contrived samples were created using irradiated SARS-CoV2 lysate in VTM. A checkered board pattern was created to evaluate the robotics and the potential for cross-contamination.
Results: The hybridization signals obtained (
Run 2. Clinical sample evaluation: Known positive (25 samples) and negative (22 samples) COVID samples from TriCore were employed, including 49 water blanks. A checkered board pattern was created as above to evaluate the robotics and the potential for cross-contamination.
Results: The hybridization signals obtained were found to be stronger than that observed in
Adding a RSV test to the previously discussed content (SARS-CoV2, Clade Variant and Influenza A, B) was considered be valuable in this analysis. To test this, a fast-track analysis for implementing a HRSV test with the SARS-CoV2 content was performed.
Assay Design. An established Q-RT-PCR assays for HRSV (Table 27) was modified using PDx design principles (Table 28) into a PDx format comprising a single RT-PCR reaction and 3 probes (Pan HRSV probe, Subfamily A probe, Subfamily B Probe).
Inclusivity analysis Table 29 shows an inclusivity analysis of the primers and probes for the Hu et al and the PDx assays using the following sequences—HRSV (taxid:11250), HRSV-A (taxid:208893) and HRSV-B (taxid:208895). The analysis revealed that PDx probes have adequate Inclusivity and well suited to distinguish HRSV A subtype from HRSV B subtypes.
1Hits increase when removing 3′TGA
2Hits increase changing 3′ TGTATT to TRTATT
3Hits increase when 1 mismatch allowed
4Hits increase when 3′-TTAA removed
Exclusivity. Table 30 shows that for both the Hu et al and the PDx assays using the following sequences—Homo sapiens (taxid:9606), HCoV229E (taxid:11137), HCoV-OC43 (taxid:31631), HCoV-HKU1 (taxid:290028), HCoV-NL63 (taxid:277944), MERS-CoV (taxid:1335626), Human metapneumovirus (taxid:162145), Human adenovirus sp. (taxid:1907210), HPIV-1 (taxid:12730), HPIV-2 (taxid:1979160), HPIV4 (taxid:1979161), Influenza A virus (taxid:11320), Influenza B virus (taxid:11520), Enterovirus (taxid:12059), Human parainfluenza virus 4b (taxid:11226), Streptococcus pneumoniae (taxid:1313), HRSV (taxid:11250), Rhinovirus (taxid:12059), Chlamydia pneumoniae (taxid:83558), Haemophilus virus HP2 (taxid:157239), Legionella pneumophila (taxid:446), Mycobacterium tuberculosis (taxid:1773), Streptococcus pyogenes (taxid:1314), Bordetella pertussis (taxid:520), Mycoplasma pneumoniae (taxid:2104), Pneumocystis jirovecii (taxid:42068), Candida albicans (taxid:5476), Pseudomonas aeruginosa (taxid:287), Staphylococcus epidermidis (taxid:1282), Streptococcus salivarius (taxid:1304), HPIV-3 (taxid:11216); and exclude: HCoV-SARS (taxid:694009), SARS-CoV2 (taxid:2697049).
As seen from Table 30, there is negligible experimental cross reaction with human DNA/RNA or any of the standard panel of respiratory pathogens required for analysis of Exclusivity in SARS-CoV-2 testing.
Feasibility. The calculations obtained suggests feasibility of implementing the HRSV assay capacity to the present 12 probe Mini-RV assay.
A hybridization script on the Tecan was upgraded to reduce reagent waste. A new hybridization script was tested and found to provide results equivalent to non-automated two-step RT-PCR (with labeling) as shown in Table 31. The script was edited for compatibility with plate processing ancillary equipment and the protocol used to run the Zymo kits.
A QC/QA test protocol was developed (for [S, N1, N2, N3, P, PanA, PanB). Multiple Tricore samples were pooled to generate a stock solution of purified clinically derived RNA for QC/QA. Table 32 summarizes the results from this analysis.
Homo
sapiens
Streptococcus
salivarius 62%
Pseudomonas
aeruginosa 67%
Streptococcus
salivarius 61%
Legionella
pneumophila 60%
Streptococcus
salivarius 64%
Streptococcus
pneumoniae 52%
1Exclusivity respiratory panel % complementarity (organism with closest match). Generally, <80% total complementarity requires no deeper analysis
2Surface Bound non-PCR oligos are not subjected to sequences other than amplimers generated from the PCR primers
Clinical sensitivity and specificity analysis performed on the Mini-RV content in 9985 array format using Asymmetric One-Step RT-PCR revealed a 100% sensitivity and 94% specificity for each of the 96-well and 384-well samples. Tables 33 and 34 shows an improvement in specificity for the SARS.COV2-N1-RE1.1 probe. Signal strength for the RNase-P control is also improved.
Influenza testing on Clinical Samples (NP/VTM) using Mini-RV, Asymmetric One-Step RT-PCR
Influenza Positive TriCore Clinical Samples (N/VTM) were used for clinical evaluation of Influenza A and B primer and probes in two positive Influenza A, validated on a respiratory panel (RESPAN, TriCore) and two positive Influenza B, validated on an Influenza A/B and RSV panel (FLURSV, TriCore), analyzed on Mini-RV slide format. Table 35 shows that Influenza A and B were detected in confirmed clinical samples via standard Asymmetric One-Step RT-PCR (Zymo), with a clear discrimination between Influenza A vs Influenza B.
Mouthwash/saliva samples were separated and evaluated by itself (MW-1), spiked with SARS-CoV2 viral lysate from ATCC (MW-2), or with SARS-CoV2 purified viral RNA from ATCC (MW-3). The mouthwash sample was taken through Zymo's RNA purification and amplified using the Asymmetric One-Step RT-PCR method. Amplicons were analyzed on Mini-RV format (12-well slides). Table 36 shows that SARS-CoV2 was detected in contrived mouthwash samples.
¶MW-2 - Mouthwash, spiked with SARS-CoV-2 Viral Lysate, 1 × 108 PFU (BEI. Wuhan)
§MW-3 - Mouthwash, spiked with SARS-CoV-2 purified RNA at 1000 copies total for input to RNA extraction (BEI. Wuhan)
The assay is based on RdRp and has an inclusivity of (NL63+OC43+229E+HKU1). Primers and Probes used for the assay are shown in Tables 37 and 38. In Silico analysis demonstrates that the primers and probes are specific for their targets and do not demonstrate off target interactions—less than 80% homology to any off-target sequence. Table 39 shows the exclusivity analysis using the following sequences—Homo sapiens (taxid:9606), HCoV-229E (taxid:11137), HCoV-OC43 (taxid:31631), HCoV-HKU1 (taxid:290028), HCoV-NL63 (taxid:277944), MERS-CoV (taxid:1335626), Human metapneumovirus (taxid:162145), Human adenovirus sp. (taxid:1907210), HPIV-1 (taxid:12730), HPIV-2 (taxid:1979160), HPIV4 (taxid:1979161), Influenza A virus (taxid:11320), Influenza B virus (taxid:11520), Enterovirus (taxid:12059), Human parainfluenza virus 4b (taxid:11226), Streptococcus pneumoniae (taxid:1313), HRSV (taxid: 11250), Rhinovirus (taxid:12059), Chlamydia pneumoniae (taxid:83558), Haemophilus virus HP2 (taxid:157239), Legionella pneumophila (taxid:446), Mycobacterium tuberculosis (taxid:1773), Streptococcus pyogenes (taxid:1314), Bordetella pertussis (taxid:520), Mycoplasma pneumoniae (taxid:2104), Pneumocystis jirovecii (taxid:42068), Candida albicans (taxid:5476), Pseudomonas aeruginosa (taxid:287), Staphylococcus epidermidis (taxid:1282), Streptococcus salivarius (taxid:1304), HPIV-3 (taxid:11216); and exclude: HCoV-SARS (taxid:694009), SARS-CoV2 (taxid:2697049). No complementarity (<80%) was observed compared to the full standard exclusivity panel. Inclusivity analysis (Table 40) using the sequences HCoV-NL63 (taxid:277944), HCoV-OC43 (taxid:31631), HCoV-229E (taxid:11137), HCoV-HKU1 (taxid:290028) MERS-CoV (taxid:1335626) showed >98% compared to GenBank (NL63+OC43+229E+HKU1) plus several other animal Coronavirus.
Homo
sapiens
Streptococcus
pyogenes 59%
Staphylococcus
epidermis 60%
Streptococcus
pneumoniae
Staphylococcus
epidermis
1Generally, <80% total complementarity requires no deeper analysis
2Surface Bound non-PCR oligos are not subjected to sequences other than amplimers generated from the PCR primers
Emory Test Samples. Emory contrived a sample with heat inactivated CoV2 virus (BEI standards) in VTM, covering a dilution series from 106 to 0 virus/ml. The sample was then shipped to PDx in double-blinded form. PDx performed the full manual process in the 96-well format (that is, Zymo RNA purification+One-Step PCR+Hybridization/Wash+Imaging+AUGURY). The results obtained were then reported to Emory, which was tasked with reporting concordance with the number of virus particles/ml originally added.
Training Samples. PDx completed validation and shipped to Emory, a full suite of stepwise training materials, “Imaging Test”; “Hybridization/Wash+Imaging Test”; “PCR+Hybridization/Wash+Imaging Test”. Two sets of blinded contrived samples were made from both γ-inactivated and heat inactivated reference standards (BEI) in pooled nasal fluid, diluted from 104 copies/ml to 10 copies/ml (Tables 41 and 42).
Emory testing of blinded contrived samples from PDx: Emory's data for the test samples provided by PDx, which required the full processing workflow (RT-PCR+Hybridization/Wash+Imaging+AUGURY) had readouts of the blinded PDx-contrived samples that were identical within experimental accuracy to that obtained independently at PDx.
PDx testing of double-blinded contrived samples from Emory: PDx's data readout on the double blinded samples provided by Emory (Tables 43 and 44) indicated a LoD of −1000 viral copies/ml for the contrived heat inactivated CoV-2 samples. It is interesting to note that the apparent LoD obtained by PDx is identical within experimental accuracy to that obtained by Emory and PDx (1000 copies/ml, Table 45).
It should be noted that if the PDx standard were relaxed to that used in most Q-RT-PCR assays, that is, probe detected comprising a positive readout, the LoD thus obtained by PDx
analysis on the Emory samples would be in the 10-100 copies/ml range, because reruns would have been identified as “positives” using the less stringent analytical standard (Tables 43).
It is also interesting to note that for the contrived samples prepared by PDx and measured by Emory (Table 42) and the contrived samples prepared by Emory and measured by PDx (Table 40) all samples lacking CoV-2 gave 100% negative results per PDx analysis (Tables 41, 42 and 44) indicative of desired specificity even when the samples were contrived to contain significant amounts of another coronavirus (OC43) as in the Emory contrived samples (Tables 44).
Improvements made in Hybridization/Washing of the Mini-RV array were implemented manually, but via pipetting that can map over directly to the Tecan.
Method—96-well plates
Access Quick RT-PCR
Step i. 45° C., 45 min, 1 cycle
Step ii. 94° C., 2 min, 1 cycle
Step iii. 94° C., 30 sec, 40 cycles
Step iv. 55° C., 30 sec, 40 cycles
Step v. 68° C., 30 sec, 40 cycles
Step vi. 68° C., 7 min, 1 cycle
Step vii. 4° C., ∞
Contrived negative samples TriCore Negative NP at 62.5 copies/ml.
Clinical LoD for Asymmetric One-Step RT-PCR in 96-Well Mini-RV approached matched 2-step PCR with slides (<62.5 copies/mL) (Tables 46 and 47). The LoB appeared to approach slides with manual operation.
Clinical Isolates. TriCore NP Positive and Negative. Analysis of clinical samples showed 91% specificity and 100% selectivity (N1, N2, N3, P) (Tables 48 and 49). Three false Negatives were found to coincide with clinical samples with Cq>30.
To test the potential of cross contamination of the negative sample wells with the positive samples various checkerboard patterns were tested in the Min-RV 96-well workflow. The goals of these experiments were:
Four different checkerboard patterns were tested in duplicate (8×96 data points) for pooled positive Boca NP swab samples (in VTM) extracted on the Tecan robot. Table 50 shows the four checkerboard patterns used with the positive wells shown in bold numerals. Tables 51 and 52 show the full representative data sets for Checkerboard patterns 2 and 3. These data revealed no well-to-well cross contamination across all 8 sets of experiments (602 negatives) and are summarized in Table 53.
To reduce time needed to perform the assay, temperature and time parameters in the Asymmetric One-Step RT-PCR were varied. The general change in the method steps were as follows:
A summary of 4 different RT-PCR reaction protocols is shown in Table 54. Results from these studies summarized in
This modification to the protocol also has additional advantages. Increasing the reverse transcription temperature to 55° C. makes the protocol compatible with a concurrent Uracil-DNA Glycosylase enzyme (UNG, Cod UNG from ArticZymes Technologies) reaction (see below). Further, reducing the time for heat denaturation from 30 sec to 20 sec reduces the Taq Thermal footprint during the RT-PCR reaction.
Mitigate Potential Assay Contamination Due to Low Copy Number RNA Sample Contamination by Ambient High Copy Number Amplicon Products from Previous Assays
Amplicon contamination has the undesired consequence of generating false positive results in the assay. This problem may be offset by the introduction of Uracil-DNA Glycosylase into the reverse transcription phase of the Asymmetric One-Step RT-PCR reaction. One of the requirements for using UNG is a reaction temperature of 55° C. As discussed above increasing the temperature from 37° C. to 55° C. during reverse transcription does not alter efficiency of the Asymmetric One-Step PCR (Table 54,
Further validation for employing a higher temperature for the reverse transcription was obtained using multiple clinical isolates (NP-VTM from TriCore) and contrived samples (in nasal fluid) titrated with gamma irradiated CoV-2 virion (BEI, 5,000 virion/ml to 500/ml).
Sample 1—Eight (8) Positive clinical samples.
Sample 2—Eight (8) Negative clinical samples.
Sample 3—Four (4) gamma irradiated virus (BEI). 5000, 1000, 500, 0 copies/mL
Three different RT-PCR conditions were tested with each of the above sample sets as shown in Table 55.
The data shown in Tables 56 and 57 confirms no change in N1 and N2 signals for these samples. Interestingly, the combination of 55° C. and a reduced, 20 min reverse transcription incubation step was statistically identical to 45° C. and 45 min, confirming that that the combined RT-PCR reaction can be performed about 25 min faster than the standard protocol.
Using an on-board plate shaker permits fluid phase mixing and laminar flow over the array surface during ambient temperature hybridization and washing, which helps reduce by at least 30%, the number of hybridization/wash steps in 96-well format. Two experiments were performed to test this.
Experiment 1: To test the effect of shaking on signals, 24 pooled positive samples were prepared (Boca) and tested under 3 separate hybridization conditions as follows:
X1—Plate remains static for 30 min hybridization incubation period.
X2—Plate is shaken at 1000 RPM for 30 min hybridization incubation period.
X3—Hybridization cocktail is mixed by pipetting up and down during hybridization period.
Results: Table 58 showed that condition X2 gave the highest average RFU across 8 wells on the appropriate probes, along with a lower standard deviation and lower background. These data reveal that shaking during hybridization improves signal strength when compared with the static hybridization method (X1) and the pipetting method (X).
Experiment 2: RNA extracted (Zymo kit) from contrived samples (gamma irradiated cell lysates+nasal fluid in RNA Shield™ reagent (Zymo research) was used as the first sample at 0.4-40 copies per reaction. SARS-COV2 RNA was used as a second sample at 1-100 copies per reaction. RT-PCR parameters described in Protocol C (Table 54) was used. Results: The data in
Evaluate Simplified Alternatives to Standard Magnetic Bead CoV-2 Purification from NP/VTM and Mouthwash
A CERES NANOTRAP (Ceres Nanosciences, Inc.) technology for RNA extraction was evaluated for reducing time and costs of raw sample processing over the Zymo Quick-DNA/RNA Viral technology.
Alternate methods for reducing assay time and costs during raw sample processing were tested including the CERES NANOTRAP (Ceres Nanosciences Inc.) and Chitosan Coated Magnetic Beads (Creative Diagnostics Inc). Specifically, compared to Zymo's Quick-DNA/RNA Viral method the CERES NANOTRAP method is 1.5 hours faster requiring ⅓rd of total manipulations, consumes 75% less consumables and may be automated for 96-well format.
A comparison between the Zymo method described earlier with the CERES NANOTRAP method was performed for contrived NP/VTM samples prepared by Emory (heat-killed Cov-2 virus from BEI, in VTM). The CERES NANOTRAP method (
Among all 3 SARS-CoV-2 probes tested, the sensitivity of the Mini-RV assay subsequent to CERES NANOTRAP is identical or superior to that obtained using Quick-DNA/RNA Viral method for sample processing.
Clinical samples (positive and negative NP/VTM samples) previously characterized at TriCore via the Roche, Cobas 6800 SARS-CoV-2 platform, were analyzed to generate a Q-RT-PCR based Cq values for each clinical isolate. All samples were subjected to viral capture and enrichment using CERES NANOTRAP, followed by direct heat lysis of the resulting viral pellet in 1% Triton-X-100 as described above. The lysate (5 μL) from each of the 61 samples was used as input without additional purification, in the Asymmetric One-Step RT-PCR, followed by Mini-RV hybridization analysis. Two types of analysis were performed on the hybridization data.
Analysis 1. Hybridization signals (RFU) from all Mini-RV probes in the positive and negative TriCore samples was used to generate mean and standard deviation for the LoB, which was then used to determine the RFU threshold to be deployed in analysis of the samples. The Clinical and Laboratory Standards Institute (CLSI) standard was applied in threshold determination. To account for user differences, LoB was modified using the equation;
LoB=(3*Standard Deviation)+Average
Using this threshold value (Table 62), clinical sensitivity, and specificity, PPV and NPV were calculated for each probe in the Mini-RV test, and in turn for the overall call generated by AUGURY from those multiplex probe data (Table 62).
Analysis 2. To facilitate analysis of the relationship between Q-RT-PCR signal strength (Cq) and the Ceres+Mini-RV signal strength (RFU), the data from TriCore, NP/VTM clinical positives was rank-ordered based on their Cobas Cq value—lowest Cq (highest viral load) at the top and highest Cq (lowest viral load) at the bottom (Table 63). Cq values from Roche Cobas 6800 for the negative samples is shown in Table 64.
The data show 100% clinical sensitivity and clinical specificity (Table 62, bottom row). Highest affinity Mini-RV probes (SARS.COV2-N2-RE1.3 and SARS.COV2-N3-RE1.1) remained positive even in the highest Cq (lowest viral load) positive samples, producing clearly defined calls, throughout (Table 62, columns 7,9). Additionally, the relationship between Q-RT-PCR (Cq) values and RFU signals (Table 63) manifest in the comparison of high affinity versus medium affinity Mini-RV probes enables microarray-based quantitation of Cov-2 RNA load.
The protocol used 200 μL of beads and elution in 100 μL of extraction buffer (0.5% TritonX-100 in water) and additionally, a wash step after the first pelleting step. Clinical sensitivity and specificity analysis of the CERES NANOTRAP Mini-RV technology using 30 Tricore (Cobas-Pos)
and 30 (Cobas-Neg) NP-VTM samples were 100% relative to the Cobas predicate. Probe threshold was calculated from LoB data obtained from the matched clinical negative samples using the formula;
Threshold=3 ×(STV)+Mean
where Mean is the Mean value of RFU signal and STV is one standard deviation about that mean.
In addition to clinical sensitivity and specificity analysis, determination of LoD in units of virions/ml, were performed using virus that were subjected to heat, radiative or chemical denaturation. On contrived samples distributed as PT standards (FDA's SARS-CoV-2 Reference Panel Comparative Data) the Roche Cobas Q-RT PCR assay delivered a LoD of 1,800 copies/mL. Thus, all 30 positive clinical samples studied here (TriCore, with Cobas Predicate) are expected to contain >1,800 copies/mL
As shown in
A finer dilution of the heat inactivated SARS-CoV-2 in VTM was tested at 5000, 3000, 2000, 1000 and 500 copies/mL (N=6 for each concentration). The samples were prepared in 500 μL of VTM and processed using the Ceres protocol with a final elution/lysis volume of 100 μL. The results showed 100% detection capability down to 500 copies/mL for N1 and N2, whereas a high background and variability for the N3 probe (
Based on the result from Experiment 1 above, additional LoD experiments were performed with increased sample number towards obtaining 95% positive results at 500 copies/mL. For this purpose, three sets of LoD samples were created at 3000, 1000, 500, 300, and 0 cp/mL (N=20 each) in VTM. Additionally, fresh vial of SARS-CoV-2 heat inactivated virus was used and prepared in VTM containing 10% glycerol prior to diluting to the concentrations tested. The results in
In the last experiment the hypothesis that freeze thaws might impact the background signal was tested. A large, pooled sample was created in which heat inactivated SARS-CoV-2 virus was diluted into VTM at 5000 cp/mL. The samples were aliquoted and stored at −80, −20, 4, and room temperature for 72 hours. The samples were then thawed or removed from the refrigerator and prepared using the CERES NANOTRAP beads followed by the DETECTX-RV protocol. No differences in background or signal strength were observed due to the different storage conditions or freeze thaws (
LoD Analysis of the CERES NANOTRAP Mini-RV Technology Pairing with Heat-Denatured CoV-2 (BEI) in VTM.
The clinical and LoD results presented in the previous example demonstrated excellent sensitivity and specificity and an LoD at or below 500 copies/mL. In addition, it was noted that there was variability in the baseline signal for the N3 probe. To further refine the protocol, pooling studies were undertaken and additionally, the platform was evaluated for multiplexed detection of Influenza A and B.
Experiment 1: A fully automated Ceres run was performed on the Tecan EVO150. In order to evaluate the run a checkerboard pattern (
Experiment 2: Three different lots of SARS-CoV-2 viral material (heat inactivated and gamma irradiated) were tested under different storage conditions (Table 65). A dilution from 30,000 to 1,000 copies/mL was used for each source material. The results shown in Table 65 demonstrate that absence of 10% glycerol displayed the lowest overall RFU and poor LoD followed by the heat inactivated virus stored in 10% glycerol. The best performing material was gamma irradiated lysates, which exhibited strong RFU signals down to 1000 copies/mL.
Experiment 3: To evaluate the ability to pool using the CERES NANOTRAP beads a series of positive samples with Ct values ranging from ˜15 to ˜35 by 5 Ct values was evaluated in relation to pooling 4 and/or 8 samples. To create the pooled samples, 100 μL of each sample was combined into a single tube such that, for example, a pool of 4 samples has a final volume of 400 μL. To each of the pooled samples was added 200 μL of CERES NANOTRAP beads and the pooled sample eluted into 100 μL of lysis buffer.
The results from this analysis demonstrate that with a pooling size of 4 or 8 samples with a Ct value of ˜30 is detectable (Table 66). The RFU values for that sample demonstrate linearity from the sample alone (10,000 RFU), 4:1 (-5,000 RFU), and 8:1 (-2500 RFU) starting at a Ct value of ˜25.
Experiment 4: To evaluate the ability of including Influenza A and B in the multiplexed array, the PCR conditions were modified to accommodate the incorporation of UNG. A comparison of the current room temperature (RT) condition at 45° to the 55°—conditions needed for UNG denaturation is shown in Table 67. The data shows that the change from 45° to 55° increases the RFU signals at the lower concentration without any adversely impacting signal strength. An LoD of 100 copies/mL was obtained using gRNA on the Zymo platform.
Experiment 5: Next, to test specificity of the platform for Influenza A and B, a series of samples were extracted using both Zymo and Ceres protocols. The samples were acquired through Tricore and were tested using the Respiratory Virus Panel by Real Time PCR (BioFire Diagnostics) with an LoD˜300 copies/mL for Influenza A (RESPAN). The results of this analysis shown in Table 68 demonstrate specificity within the assay. Table 69 shows that the CERES NANOTRAP beads capture/lysis/analysis protocol described above for CoV-2 (0.5 ml VTM+0.2 ml Ceres, magnetic bead isolation, elution & lysis in 0.1 ml) may also be adopted for detecting InfB signals on clinical positives that were greater than 8× the threshold obtained from matched clinical negatives.
¶threshold = 3292;
¶Threshold = 3292
¶SARS CoV-2 RNA;
§no template control
§no template control
§no template control
Repeat measurements (n=72) from a single pooled clinical negative sample (50 mls, TriCore NP-VTM) using the Ceres processing protocol and DETECTX-RV analysis were performed as 72 independent 0.5 ml Ceres extractions. The experiments were performed on multiple days over 2 weeks, followed by Ceres processing, RT-PCR and analysis in a 96-well format for N1 and N2 CoV-2 markers.
Clinical Matrix Samples for Threshold Analysis protocol:
1. Clinical matrix=TriCore negative clinical samples (NP-VTM, Cobas 6800)
2. 200 μL Ceres beads were added to 5004 of Contrived Clinical Matrix (N=10)
3. Samples were shaken for 10 mins.
4. Quick spin was performed before adding samples to a magnetic plate and removing supernatant.
5. 100 μL of 0.5% TritonX-100 was added to the samples.
6. Samples were shaken for 2 mins followed by heating at 95° C. for 10 mins.
7. Samples were centrifuged briefly before adding to magnetic plate.
8. Eluate was transferred to a PCR plate for storage.
8. Samples were run using standard RT-PCR cycling parameters.
9. Hybridization and Washing steps were performed in 96-well format.
10. Steps 1-9 were repeated for multiple days.
11. The threshold was calculated using the formula; Threshold=3×STD+RFU(blank)
The data in Table 70 reveals a defined average with no drift in threshold values over the 5 repeat measurements. The analysis also revealed the presence of occasional outliers—e.g. day 19 for N2 and day 22 for N1 (
In continuation of the in silico analysis of RSV described in Example 16, primer and microarray testing was performed. Table 71 summarizes the results from an analytical sensitivity dilution series—purified human RSV-B gRNA (BEI) diluted to 1000, 100 and 10 genome copies/PCR reaction. The data demonstrates excellent specificity with no measurable signal detected above background for either of the two RSV-A probes tested in the array (HSV-A, RE1.1, 1.2). The data also demonstrate excellent sensitivity for detection of N1 and N2 above a threshold defined by the (0) genome copy control down to 10 copies/reaction.
The method of detecting RNA virus comprises the following steps:
1) Recovery of viral RNA by capture of the virus from a fluid sample (analyte) by pipetting or centrifugation or binding of the pathogen to a solid phase such as an appropriate magnetic bead or column, followed by lysis of the captured pathogen and then in some cases additional purification of RNA from the virus by silica-based boom chemistry as routinely deployed in magnetic beads or columns.
2) RT-PCR of the viral RNA recovered, to generate PCR amplified cDNA amplicons that are further amplified using a suitable set of fluorescently labeled primers specific for the cDNA amplicons to obtain fluorescently labeled amplicons suitable for microarray hybridization.
3) Microarray hybridization of the resulting RT-PCR amplified DNA amplicons.
4) Analysis of the microarray hybridization patterns to detect the presence of viral analytes of interest.
The above method is extended to include DNA-containing pathogens including, DNA viruses, bacteria and fungus known to cause respiratory disease by accommodating capture and analysis of both RNA-containing and DNA-containing pathogens. Such a method comprises the following steps:
1) Use methods such as pipetting and centrifugation among others to capture concurrently, RNA viruses, DNA viruses, bacteria and fungus resident in the same clinical or environmental sample. Subsequent methods of lysis are then employed to enable concurrent lysis of all of the captured pathogens. Additional purification steps such as, silica-based boom chemistry as routinely deployed in magnetic beads or columns enables concurrent capture and purification of RNA and DNA from these pathogens.
2) Use of an appropriate panel of PCR primers enables reverse transcription of viral RNA to obtain cDNA followed by PCR amplification to concurrently amplify in the same reaction (single assay), cDNA and DNA from DNA viruses, bacteria and fungus to yield a set of amplicons that are further amplified using a suitable set of fluorescently labeled primers specific for each pathogen being queried to obtain fluorescently labeled amplicons suitable for microarray hybridization.
3) Concurrent microarray hybridization of the resulting fluorescent amplicons on the same microarray enables their analysis.
4). Analysis of the microarray hybridization patterns obtained is then used to concurrently detect presence of any or all of pathogens in a sample.
Rapid detection of respiratory disease-causing viruses including COVID-19 virus, other coronaviruses, Influenza A virus, Influenza B virus, RSV-A and RSV-B are crucial to controlling the COVID-19 pandemic. However, it is well known that there are other DNA containing respiratory disease-causing pathogens including DNA viruses like adenovirus and bacterial pathogens such as Mycobacterium tuberculosis and Streptococcus pneumoniae. The microarray-based detection methods described here are readily adaptable and extendable to detection of these DNA containing respiratory disease pathogens as well in a single assay. This is beneficial since it enables streamlined concurrent detection of COVID-19 virus concurrently with other respiratory disease pathogens in a single assay.
Experiment 1: LoD studies in contrived clinical negative samples (NP-VTM, TriCore) were performed using inactivated flu virus (ATCC, InFA (H1NI), and InFB (Hong Kong)). Particle density in the ATCC samples was measured in infectious units via PFU assay (i.e. CEID50/ml) which is approximately equal to particles/mL. In all cases, viral capture with Ceres beads was performed on the flu virus, followed by lysis and amplification of the lysate with the complete One Step RT-PCR master mix comprising the full N1,N2,N3,P,InFA,InFB multiplex described in previous reports. Hybridization was obtained in the 96-well format.
First, a clinical threshold was obtained using thirty-two (32) clinical negatives for the InF A and InF B probe content, using the formula;
Threshold (RFU)=3 ×(STV)+Mean
These data revealed low background values and as a result low thresholds (721 RFU for Inf A and 896 RFU for Inf B
Next, a preliminary range-seeking study was performed on contrived InF A and InF B samples prepared on a single batch of pooled clinical negatives that revealed that the LoD would be in the approximate range of 1,000 CEID50/ml for InF A and 100 CEID50/ml for InF B (Inf A, LoD=100-1000 CEID50; Inf B, LoD=1-10 CEID50). Based on this preliminary analysis, a more detailed LoD determination was performed that showed that the LoD for InF A is less than 400 CEID50/ml (
Experiment 2: An extension of the above studies was performed at multiple data points closer to the LoD. Dilutions of Inf A (H1N1) and Inf B (Hong Kong) were made in clinical matrix (45 mL VTM+5 mL pooled negative clinical sample) as shown in Table 73 and the method performed as described above for Experiment 1.
An extended clinical threshold was obtained by processing additional clinical negatives for the Inf A and Inf B probe content, using the formula used above. The extended data set revealed a statistically strong background mean and STD that is reproducible with a threshold value of 1259 RFU for Inf A and 6221 RFU for Inf B (
LoD studies were undertaken on QuikSal mouthwash negative clinical isolates. Briefly, three (3) oral rinse samples from healthy lab volunteers were pooled to generate a single 15 mL Cov-2 negative clinical sample. The pooled sample was then doped with gamma irradiated CoV-2 (BEI) ranging from 10,000 to 625 virus particles/mL (Table 75) Protocol:
The summary of the range analysis (Zymo vs Ceres processing) is presented in
As discussed above, the LoD for both N1 and N2 SARS-CoV-2 probes was <1000 virus copies/mL. Using an expanded titration to include 500, 800, 1,000 copies/mL it is observed (
The following references are cited herein:
This non-provisional application claims the benefit of priority under 35 U.S.C. § 119(e) of provisional applications U.S. Ser. No. 63/078,783, filed Sep. 15, 2020, and U.S. Ser. No. 63/000,844, filed Mar. 27, 2020, both of which are hereby incorporated by reference in their entireties.
| Number | Date | Country | |
|---|---|---|---|
| 63078783 | Sep 2020 | US | |
| 63000844 | Mar 2020 | US |
