Patent Application
20060210967
The present application includes a sequence listing on an accompanying compact disk containing a single file named 272918US59SDSt2-5.txt, created on Jul. 1, 2005, 639 KB in size, and additionally labeled:
“Inventors: Brian K. Agan, Eric H. Hanson, Russell P. Kruzelock, Baochuan Lin et al.
Invention: “Re-Sequencing Pathogen Microarray”
The entire contents of that accompanying compact disk are incorporated by reference into this application.
1. Field of the Invention
The present invention provides pathogen detection by use of DNA resequencing microarrays. Preferably, the present invention provides for simultaneous detection of multiple pathogens. The present invention also provides resequencing microarrays and microarray chips for differential diagnosis and fine-scale discrimination between closely related pathogens present in a biological sample. The present invention further provides methods of detecting the presence and identity of pathogens present in a biological sample. The invention enables diagnosis and surveillance of known pathogen sequences and pathogens that may be identified due to unanticipated sequence variations, as well as mixtures of such pathogens. Resequencing, combined with several amplification strategies, allows simultaneous clinical diagnosis and performance of traditional surveillance assays for serotyping, antibiotic resistance profiling, genetic drift/shift analysis, forensics, and rapid detection of biological terrorism events.
2. Discussion of the Background
As we move through the biotechnology age fostered by the human genome project a premium has been placed on the development of high throughput methodologies to obtain and analyze sequence information. To meet this demand, the multifunctional DNA microarray platform has gained notoriety leading to an explosive growth in application methods using the same.
More importantly, the evolution of world events and the emergence of bioterrorism in mainstream society have led to a growing sentiment amongst the scientific community and lay people alike that new, rapid, and accurate techniques for biological threat identification and eradication must be developed. The concept of a microarray used for broad-spectrum pathogen identification has considerable and obvious appeal to both medical practice and national defense. It is within this framework that the present inventors have endeavored.
Heretofore, for the purpose of pathogen identification, approaches generally rely on the ability of immobilized “probe” DNA sequences on the surfaces of microarrays to hybridize with complementary genomic “target” that is uniquely identifying of a particular category or specific strain of microbial pathogen. Various microarray technologies have been developed for this purpose, varying in the density of probes and the time ranges required for assay completion.
One technical challenge for pathogen detection with microarrays arises due to the difficulty in obtaining samples with a sufficient quantity of pathogen nucleic acid. Thus, for a majority of sample types, some sort of target amplification will likely be required to provide sufficient copies of pathogen gene markers for detection by microarray hybridization. Unfortunately, conventional methods for this amplification do not scale well in comparison to the number of probes that can be placed on a microarray chip. However, the most commonly employed means of providing sufficient quantities of genomic target to detect hybridization relies upon genotypic identification methods that utilize molecular biology-based techniques, such as the polymerase chain reaction (PCR). These techniques offer several potential advantages over conventional microbiological approaches. Nucleic acid amplification strategies base pathogen identification on the detection of genetic information contained within the organism, such that culturing the organism is not required.
Although PCR-based assays are sensitive, accurate, and rapid, these methods also introduce a new set of problems. As successful identification depends almost entirely on appropriately chosen primer sets, as PCR-based testing requires assumptions about the exact sequences pertaining to the identity of the target organism(s). Consequently, there is a critical need for advanced diagnostic systems that can detect both assumed and unanticipated pathogen sequences. DNA microarrays, which enable the simultaneous interrogation of thousands of genetic elements, address this crucial need. Here, the term “microarray” refers to any type of planar substrate or solid beads presenting a high multiplicity (102 to 106) of individual sites, each presenting nucleic acid probes designed to selectively capture complementary strands of target (i.e. pathogen or host) nucleic acid.
However, the majority of pathogen identification microarrays described in the literature is prepared using oligonucleotides that are robotically spotted onto derivatized glass surfaces (typically 3×1 inch microscope slides). This approach allows the most flexibility with regards to the size of the oligonucleotides that are deposited, ranging from 20-mers to cDNA PCR products of several thousand base pairs (bp). With few exceptions, the detection event is an increased level of fluorescence originating from a spot following hybridization of a fluorophore-labeled target nucleic acid.
Short (14-25 mer) oligonucleotides, immobilized inside acrylamide pads, have been applied extensively to pathogen identification (Strizhkov et al., 2000; Vasiliskov et al., 1999) in a collaborative effort between Argonne National Lab (DOE, USA) and the Engelhard Institute of Molecular Biology (Moscow, RU) under the leadership of Andrei Mirzabekov. In addition, low-density microarrays (several hundred features per 3×1 inch microscope slide) have been used for determination of drug resistance determinants (Volokhov et al., 2003). One distinguishing aspect of this body of work is the use of three-dimensional polymer matrices for probe immobilization instead of two-dimensional planar surfaces.
More recently, Cherkasova et al have described the use of glass-immobilized short oligonucleotide spotted microarrays to map poliovirus mutations using overlapping 14-25 mer probes (Cherkasova et al., 2003). Two variations of this approach have been used: (1) Microarrays for Resequencing and Sequence Heterogeneity (MARSH) assay, and (2) Microarray Analysis of Viral Recombination (MAVR) assay. MARSH uses a set of overlapping (at half length) nucleotide probes for individual gene sequences. Hybridizations patterns allow the detection of single point mutations or substitution/deletion events to a resolution of half probe lengths (e.g. 7-10 bp) but does not allow for exact determination of position(s) or the nature of the mutation. Accordingly, conventional DNA sequencing technologies must be employed subsequently to determine these changes. MAVR uses organism-specific oligonucleotide probes that cover the entire genome at ˜150 nt spacings and is used to detect large scale genetic recombinations.
The DeRisi group at UCSF pioneered the use of long (70-mer) oligonucleotide probe microarrays for broad-spectrum pathogen identification (Wang et al., 2002; Wang et al., 2003). The use of long (70 nt) oligonucleotides bears implicit advantages and disadvantages. One advantage is that higher degrees of sensitivity can usually be achieved with 70-mer probes compared to shorter ones (e.g. 20-25 mers). However, specificity is reduced because 70-mer target/probe hybridizations are generally insensitive to significant numbers (e.g., 7-10) of single base mismatches, whereas shorter probes provide much greater sequence specificity.
DeRisi's group described the use of spotted microarrays having 1,600 different 70-mer oligonucleotide probes to identify a variety of viruses responsible for common respiratory infections (Wang et al., 2002). The probes were selected for each pathogen using an algorithm that located discriminatory sequences from a list of known viral genomes. A serial combination of a previously described (Bohlander et al., 1992) method and subsequent PCR/Klenow fragment-based amplification was used to achieve non-biased amplification of both viral RNA and DNA, allowing generation of sufficient amounts of target amplicons for successful microarray hybridization and detection via fluorescent label. (N.B. This protocol was placed into the public domain via the DeRisi lab website.(http://derisilab.ucsf.edu)). The time required from sample preparation to obtained result was approximately 24 hours. Because exact sequence information was not attainable from such arrays, pathogen identifications were made on the basis of a hybridization pattern that could be empirically determined for each pathogen or strain. In a related report from the same group (Wang et al., 2003) similar microarrays were prepared using highly conserved sequences in an effort to capture as many microbial species as possible from a sample. Following physical removal of the pathogen sequences from the microarray, the sequences are cloned and sequenced using conventional DNA sequencing technologies. No measure of analytical/clinical sensitivity or specificity for pathogen detection in clinical specimens was provided in the work from the DeRisi group.
In contrast to the above-mentioned approaches using spotted microarrays, Affymetrix, Inc. (Santa Clara, Calif.) uses high-density probe fabrication technology to construct “tiled” microarrays using 4 probes each in both the sense and anti-sense directions for each nucleotide base to be resequenced. Thus, single base substitutions are directly detected by the hybridization pattern (for additional information see Affymetrix CustomSeq design manual). Several groups described the use of tiled microarrays for pathogen genotyping. (Kozal et al., 1996) utilized this type of microarray to measure mutational drift in HIV while Gingeras et al (Gingeras et al., 1998; Troesch et al., 1999) used a tiled array of 65,000 oligomer probes to resequence and accurately identify 70 clinical isolates of 27 mycobacterial species and 15 rifampin-resistant M. tuberculosis strains. More recently, Andersen et al. (Wilson et al., 2002b) described the use of tiled Affymetrix microarrays for the identification of biological warfare agents. Their approach relied entirely on the use of specific PCR reactions performed in parallel to generate sufficient pathogen target DNA for microarray hybridization. In all cases listed above, specific PCR primers were used to amplify DNA targets prior to microarray hybridizations, through the use of conserved primer sites, or in the work of Andersen et al. (Wilson et al., 2002a; Wilson et al., 2002b), by carrying out up to ˜150 different PCR reactions in multi-well format and pooling the amplicons.
U.S. Pat. No. 6,228,575 B1 describes the same data as described by Gingeras (Gingeras et al., 1998) and Troesch (Troesch et al., 1999). In this patent, which is incorporated herein by reference in its entirety, target pathogen sequences are tiled onto arrays. Because several types of variations (esp. insertion/deletion or frequent multiple substitutions) in pathogen sequence can perturb hybridization patterns, Gingeras et al. used differential measures of specific pathogen hybridization patterns to identify individual mycobacterial variants. That is, identifications required a priori knowledge of a differential hybridization pattern that was empirically determined in ground truth experiments.
As stated above, there is a critical need for advanced diagnostic systems that can detect known and pathogen genomic sequences as well as variations of those sequences. More particularly, there remains a critical demand for DNA microarray techniques that are fast and reliable, but are free from the systemic bias flowing from the specific PCR based methods that have heretofore been employed.
It is an object of the present invention to provide the following preferred embodiments:
The above objects highlight certain aspects of the invention. Additional objects, aspects and embodiments of the invention are found in the following detailed description of the invention.
A more complete appreciation of the invention and many of the attendant advantages thereof will be readily obtained as the same becomes better understood by reference to the following Figures in conjunction with the detailed description below.
Unless specifically defined, all technical and scientific terms used herein have the same meaning as commonly understood by a skilled artisan in enzymology, biochemistry, cellular biology, molecular biology, bioinformatics, and the medical sciences.
All methods and materials similar or equivalent to those described herein can be used in the practice or testing of the present invention, with suitable methods and materials being described herein. All publications, patent applications, patents, and other references mentioned herein are incorporated by reference in their entirety. In case of conflict, the present specification, including definitions, will control. Further, the materials, methods, and examples are illustrative only and are not intended to be limiting, unless otherwise specified.
It is well known that DNA microarrays might be used to analyze the sequence of target nucleic acids from pathogens for diagnostic or surveillance purposes. The effects of oligonucleotide probe sequence selection, length, positioning on surfaces, the physical chemistry of binding, and labeling techniques, have been the topics of vigorous research in the past ten years. However, heretofore, the approaches taken have almost exclusively involved the use of single oligonucleotide probes (13-70 mers) that are specifically designed to hybridize to-a single recognized pathogen target sequence with high specificity (13-25 mers) or to a longer stretch of pathogen target with a lesser degree of specificity (70 mers). These microarrays are commonly known as spotted microarrays, but the same content of a spotted microarray can also be embodied in other forms besides two-dimensional planar surfaces, one example being “bead arrays” (Ferguson, Steemers & Walt, 2000).
There have been many fewer descriptions of pathogen identification and characterization using Affymetrix resequencing microarrays. Kozal (Kozal et al., 1996) measured mutational drift in HIV and Gingeras (Gingeras et al., 1998; Gingeras et al., 2001), and Troesch (Troesch et al., 1999) identified mycobacterial species and Single Nucleotide Polymorphisms (SNPs) relating to antibiotic resistance using resequencing arrays. Wilson et al. (Wilson et al., 2002a) used the same types of arrays for bacterial identification using ribosomal RNA and for detection of multiple biowarfare agents (Wilson et al., 2002b).
The use of resequencing microarrays for simultaneous pathogen identification and surveillance has fundamental advantages over single oligonucleotide approaches because detailed target sequence information becomes directly available as part of the raw microarray data. By definition, spotted microarrays having 102 to 104 oligonucleotide probes are incapable of resolving subtle sequence variations for more than 10-1000 base pairs of total target sequence, even if the array were designed with a tiling strategy. Thus, spotted oligonucleotide microarrays that facilitate identification of a broad spectrum of pathogens at an individual strain or SNP level (Cherkasova et al., 2003; Wang et al., 2002; Wang et al., 2003) require that the pathogen target sequence be determined using conventional DNA sequencing technology, requiring specialized equipment, personnel, and time periods of several days.
Accordingly, the present invention generally relates to the detection and identification of bacterial, viral and protozoan pathogens and virulence markers using DNA microarray technology. The present invention also relates to the design, test, and analysis of resequencing microarrays having defined regions that can be used to assign numeric detection probabilities for a large number of specific pathogen genotypes and mixtures of pathogens. Furthermore, it relates to methods required for the processing of both simple and complex (e.g. clinical and environmental) samples for said microarrays.
More specifically, the present invention relates to the use of high plurality resequencing microarrays as a novel class of devices for purposes that would include diagnosis of infectious disease agents and pathogen surveillance. In particular, it pertains to DNA microarrays that employ a large number (hundreds to thousands) of selected “prototype” nucleic acid sequence regions (typically 250-2000 base pairs each) from target genomic sequences to detect and identify a logarithmically larger number (thousands to tens of thousands) of both unrelated and closely related (strain level) microbial pathogens. More-particularly, the invention relates to the design of microarrays fabricated to allow a resequencing of the target nucleic acids using a “tiled” probe approach. Yet even more particularly, the invention concerns the design of Affymetrix resequencing microarrays using tiled probes that will allow for analysis of up to 105 to 106 base pairs of target sequence.
In a particular embodiment, the present invention provides a method for simultaneously assaying for a large number of pathogens by detecting their nucleic acid sequences and subjecting the detected sequences to an automated similarity search against public domain and private databases. This object is accomplished by a non-obvious adaptation of a commercial technology (Affymetrix CustomSeq™). The program employed, CustomSeq™, was designed for detection of Single Nucleotide Polymorphisms (SNPs) by resequencing target DNA. This so-called SNP-detection calls for: (1) an error rate in base calling that is much lower than the naturally occurring frequency of mutation (in humans approximately (1 per 108 base pairs), (2) amplification of the purified starting material, at a concentration of 106 genomic copies/microliter, prior to microarray hybridization using specific PCR primers, and (3) combination and processing multiple chips for replicate samples by an algorithm that restricts base calls to those having a degree of confidence at which the presence of low frequency SNPs can be deduced. The present inventors demonstrate that Affymetrix resequencing technology can be adapted for insertion into an integrated system for high multiplicity infectious disease diagnostics and pathogen surveillance in a time scale and level of sample preparation complexity that are enabling for point-of-care diagnostics applications. This system is substantively different from the intended use of the Affymetrix technology, and is not obvious to the typical skilled practitioner of microarray technology.
At present, there have been few, but relevant literature descriptions of application of Affymetrix resequencing technology to pathogen identification. From the design, experimental, and analysis standpoints, the present invention has substantive advantages over the prior art in pathogen characterization using resequencing. Kozal (Kozal et al., 1996) measured mutational drift in HIV and Gingeras (Gingeras et al., 1998; Gingeras et al., 2001), and Troesch (Troesch et al., 1999) identified mycobacterial species and SNPs relating to antibiotic resistance using resequencing arrays. In each of these cases, optimized sequences were selected for tiling, based on alignments of the possible target sequences. Specific PCR primers were designed to amplify targets for hybridization. Unknown clinical isolate identifications were made using pattern recognition algorithms based on empirically determined differential hybridization patterns to the arrays. As such, this approach would depend on the amplification and hybridization of the full length of the target sequences, and would not be amenable to: (1) contributions of confounding non-specific binding resulting in lost base calls, (2) incomplete hybridization across the length of the target due to low target concentration or gaps caused by low homology, and (3) compromised hybridization integrity caused by non-biased (total) amplification of unknown target sequences from a clinical or environmental sample.
The method proffered by the present inventors does not share the same limitations as those disclosed in the prior art using resequencing microarrays. Moreover, technology enablements for at least one intended use, namely the simultaneous detection of a large number of diverse pathogen species, have only recently occurred with the introduction of Affymetrix microarrays (18×18 micron features) for CustomSeq™ RPMV1 chip and higher density (8×8 micron features) for Respiratory Microarray Version 2 (RPMV2), allowing 29.7 kb and 300 kb, respectively, to be tiled for resequencing. However, the most significant improvement offered by the present invention is the use of the increased density chips with sequence length-independent similarity searches (BLASTN), which affords that many fewer assumptions must be made in advance of selecting sequences for tiling. Furthermore, the use of length-independent similarity searches (BLASTN) removes the constraint that a particular known subsequence be fully resequenced, making the approach more resistant to variations in target concentration and contributions from nonspecific binding leading to lost base calls.
Thus, in the present invention, sequences selected for tiling are “prototypes” in the sense that a single gene or subsequence may represent a much broader class of organism species and subspecies (alternatively types, strains, variants, or mutants). The resulting method is robust with respect to minor variations in the genotypes of individual pathogens and strains, and enables detection and probable identification among a plurality of candidate pathogens that may not be explicitly represented in the design of the experimental chip layout, including the tiled regions.
The chip layout also takes advantage of partially redundant tile sets (these were discouraged by the manufacturer's design guidelines and by the manufacturer during design), both intragenic and intergenic variation within a single pathogen strain, and between multiple similar or diverse pathogen types. However, the present inventors have determined that the redundancy is important to elevate confidence in results and to minimize likelihood of false positive and false negative results. Advantages of this design/analysis approach will permit incorporation of both conserved and hyper-variable regions of genomes, facilitating group, type, and strain level identification.
There are only a few literature reports describing generic amplification strategies for microarray analysis of microbial pathogens. Wang et al. (Wang et al., 2002) described a multi-step process for amplification of RNA pathogen targets followed by sequential PCR and Klenow fragment-based amplifications, prior to hybridization on 70 mer oligonucleotide arrays. The time and/or number of technical steps required for this amplification were not specified but a subsequent paper from the same group (Wang et al., 2003) reported it to be approximately 24 hours. It was also not specified which of the amplification steps resulted in enhanced detection efficiency. A recent report from our group (Vora et al., 2004) describes a variety of nonspecific nucleic acid amplification techniques, individually and in combination, for hybridization to 70 mer oligonucleotide arrays. There are no known reports of prior art in the area of nonspecific amplification of DNA targets, either purified or in a complex mixture (e.g. clinical sample), for hybridization to a resequencing microarray.
The present invention generally advances the art of generic amplification for pathogen detection by microarrays in several ways: (1) it details specific methods for total amplification of purified or highly enriched pathogen nucleic acids for presentation to a high density short (25 mer) Affymetrix resequencing array, which is neither previously described nor obvious, since the methods are a large departure from the prescribed Affymetrix protocols, and (2) it describes novel alternative approaches to the use of enzymatic processes, competitive hybridizations, and magnetic bead-based subtractive and enrichment steps to reduce background and subsequent non-biased (i.e. total) amplification and microarray hybridization.
The present invention is embodied by a specific set of design and processing methods that enable broad-scale pathogen identification and characterization by utilizing resequencing microarrays. Specifically, the present invention allows for the precise, sensitive, and high confidence identification of a large multiplicity (thousands) of pathogens in a single assay.
In an embodiment of the present invention is a resequencing DNA microarray chip of multiple groups of oligonucleotide primers of a length ranging from 13 to 70 nucleotides (preferably 25 nucleotides, although it is possible and is within the scope of the present invention to use primer lengths corresponding to each integer value within this recited range) immobilized to a solid phase support, wherein each group of oligonucleotide primers is selected to span a particular region of the reference sequence, occupying a discrete area of the array (e.g. a tile), and comprising at least four sets of primers arranged in a parallel fashion on the chip: 1) a first set that is exactly complementary to the reference sequence; and 2) three additional sets of primers, each of which is identical to the first set of primers but for the nucleotide at a central position, which is different in each of the three sets such that all four conventional nucleotide bases are present on said array.
The present invention further provides methods for processing complex clinical samples (e.g. nasal wash) requiring minimal nucleic acid isolation/amplification step(s).
The present invention is distinct from the overwhelming majority of microarray-based pathogen detection schemes because it uses high-density “tiled” microarrays to determine the actual sequences of pathogen genetic targets. It is further unique from other resequencing pathogen identification strategies in a number of important areas, including incorporation of: (1) a high multiplicity of disparate pathogen “prototype” target regions that exhibit little or no discernible cross-hybridization or interference with one another, (2) a high sequence redundancy within closely related pathogens that allows higher confidence identification of specific strain (e.g. adenoviruses or influenza viruses), (3) “prototype” regions of large tiled segments representative of a class of pathogens allowing for the precise identification of specific pathogen strains and the use of specially-designed software to parse and arrange sequence fragments for presentation to similarity search (e.g. BLAST) algorithms, allowing discrimination of pathogen mixtures and recombination events between pathogens, instead of the more restrictive tile selections and differential algorithm described in the most closely-related prior art (U.S. Pat. No. 6,228,575), (4) minimally-biased nucleic acid amplification strategies that allow precise, high-confidence pathogen target resequencing without significant interference or cross-hybridization, and (5) sample processing methodologies that allow the resequencing array to be used in conjunction with minimally-biased nucleic acid amplification strategies on complex clinical samples.
The combination of these methods allows the simultaneous detection and identification of a high multiplicity of pathogen(s) from a clinical sample by a single qualified technician within a period of 24 hours, but preferably within 4 hours, more preferably with 2 hours, most preferably within 30 minutes.
Accordingly, owing to its embodiments, the present invention supports: (a) routine diagnosis of infection in a clinical setting within several hours of sample collection, (b) simultaneous interrogation of the sample for indications of a rare infectious event (e.g. unanticipated pathogen, antibiotic resistance pattern or biological warfare agent), (c) routine molecular pathogen surveillance, (d) vaccine quality control and (e) monitoring of genetic changes of a pathogen as a result of natural genetic variations, drug treatment, intentional manipulation, or other events.
High-Density Resequencing Microarrays
High-density microarrays (HDMs) are fabricated by light-directed combinatorial synthesis of DNA oligomers (Kozal et al., 1996). The DNA oligomers synthesized on these sites typically have lengths of 20-30 bases. Through subsequent improvements to the method using high-resolution semiconductor photoresists, Affymetrix has demonstrated fabrication of HDMs having individual features with resolutions approaching 1 μm2, enabling probe feature densities of 10-100 times greater than that demonstrated in the RPMV1. To date, HDM designs relevant to pathogen identification have been based on a “tiling” strategy. Accordingly, four probes of equal length are synthesized for each base in both the sense and antisense directions, requiring that a total of eight 25-mer probes are used for each base pair in a given reference sequence. One probe in each direction (sense and antisense) exactly complements the reference sequence while three others have a single base mismatch at the position of the interrogated base. Thus, a tiled HDM can effectively allow the target nucleic acid to be “resequenced”.
In this manner, the base calls of the unknown target are interrogated at each of one of four possible base positions (one of every four possible base pairings varied at the number 13 position in the tiled 25 mer probes), allowing a direct read of the target sequence from the corresponding positions across the array. In the case of CustomSeq arrays, the GCOS (Version 1.1) software is used to reduce the raw image (.DAT) file to a simplified file format (.CEL file) with intensities assigned to each of the corresponding probe positions. Finally, the GDAS (Version 2.0) software is used to apply an embedded version of the ABACUS (Cutler et al., 2001) algorithm to produce an estimate of the correct base calls, comparing the respective intensities for the sense and antisense probe sets. One of the available export file types from GDAS is the FASTA-formatted base calls made for each tiled region of the resequencing array.
HDMs of the type described above have been used to identify pathogen species and detect drug resistance-conferring mutations in a series of in vitro experiments using cultured microorganisms, including HIV (Kozal et al., 1996). Troesch et al. (Troesch et al., 1999) designed HDMs to discriminate between 54 different to detect Mycobacterium species and Mycobacterium tuberculosis rifampin resistance. A tiled array of 65,000 oligomer probes was used to accurately resequence 70 clinical isolates of 27 mycobacterial species and 15 rifampin-resistant M. tuberculosis strains. More recently, sequence-specific identification of F. tularensis and Y. pestis was demonstrated in environmental samples using tiled HDMs (Wilson et al., 2002b). Both of these general approaches rely on specific hybridization patterns based on ground-truth (control) measurements. Moreover, the authors have provided no direction on how a quantitative comparison might be made against closely-related or unanticipated organisms for starting concentrations that may vary by six orders of magnitude.
Array Types
The present invention is developed using Affymetrix CustomSeq resequencing microarrays. For a discussion of resequencing microarrays, the artisan is directed to U.S. Pat. No. 6,228,575, which is incorporated herein by reference in its entirety. However, the present invention is not conceptually limited to microarrays produced using that specific fabrication strategy. In principle, resequencing can be performed on the scale described with any technology that is capable of producing microarrays with sufficient feature density to allow this approach. Theoretically this can be accomplished using an oligonucleotide printing technology, but it is more likely accomplished using a photolithographic approach. Whereas the Affymetrix resequencing chips are based on sequential lithographic steps using a separate mask that corresponds to each step, an alternative approach could use maskless lithography (Albert et al., 2003; Nuwaysir et al., 2002) or by nanolithographic methods (Ginger, Zhang & Mirkin, 2004). More generally, any method of producing a plurality of oligonucleotide probes for the purpose of determining target sequence would be applicable, even bead “arrays” that are not in a 2-dimensional format (Ferguson et al., 2000).
The probes themselves could be comprised of variants of DNA, namely RNA or oligomeric peptide-nucleic acids (PNA). The probes can be made sensitive to enzyme digestion, then subject to subsequent handling. In a preferred embodiment, the probes will incorporate dUTP instead of dTTP, making them sensitive to uracil-DNA-glycosylase. This will make them amenable to selective degradation following the capture of target. Moreover, within the scope of the present invention it is also possible to immobilize RNA and obtain complementary sequence recognition thereof. Immobilization of RNA would require chemical stabilization of the RNA. In a more general sense, the probes can be made from chemically modified nucleic acids that would make them more or less susceptible to subsequent chemical processing steps.
Array Design
This present invention, using the RPMV1 microarray, illustrates the ability of tiled prototype sequences to identify a wide variety of specific pathogen strains without assuming that specific hybridization patterns are required for specific pathogen identification. In the current apparatus, the prototypes for tiled regions, particularly those for adenoviruses 4, 5, and 7, were empirically chosen to be representative of adenovirus subgroups E, C, and B, respectively.
A more preferred and systematic approach to the design would involve the use of multiple sequence alignments to produce consensus sequences, where consensus sequences are defined as those representing the most frequent bases at the alignment position. In a preferred embodiment, the alignment algorithm will produce a hierarchical phylogenetic tree for target gene sequences from individual pathogen strains or for a group or family of pathogens. A consensus sequence will first be formed for members of each node or group of nodes on the tree, whose distance measures fall within a threshold, using an appropriate algorithm (Lee, 2003). The actual target sequences would then be compared individually to the consensus sequence, and the effects of hybridization to a tiled microarray region defined by the consensus sequence would be simulated. Known effects such as insertion or deletion of bases, as well as the effects of multiple base substitutions within an oligonucleotide probe region would define rules for the simulated hybridization. Subsequent analysis of the resultant hybridization and base call pattern would then indicate the suitability of a proposed tile for acting as a prototype for a given range of pathogens. This process would be reiterated until the most suitable set of prototype regions were defined for coverage of a given set of pathogens. In one very preferred embodiment, this process would be used to maximize the efficiency of space usage on a chip, resulting in the greatest reduction in the overall physical dimensions of a resequencing microarray, and thus, the greatest reduction in manufacture cost per microarray.
In one preferred embodiment, the resequencing array would be designed in combination with another less complex solid phase capture device (array, gel, or otherwise) that that would be used initially to queue the resequencing array. For example, an array comprised of a multiplicity of long oligonucleotide probes would be used initially for the detection of pathogen by recognizing conserved sequences within a family of pathogens. The resequencing array could then be used to interrogate variable regions associated with the conserved probes on the initial array, providing detailed sequence information on the pathogen. In a very preferred embodiment, the sample preparation would be common for both the solid phase capture device and the resequencing microarray. In yet another very preferred embodiment, the initial array would be used to capture target that would be isolated and then amplified using non-biased amplification techniques for presentation to the resequencing array.
Overall Design Philosophy for the Intended Use
This invention leverages pathogen genome information from various public and/or private sources to, design, fabricate, evaluate, validate and integrate an advanced diagnostics platform as part of an effective biodefense surveillance and operational medicine system. It is an integral facet of the Epidemic Outbreak Surveillance (EOS) program biodefense model that the system would most effectively be enabled through the adoption of a widely distributed device that would find utility in routine diagnostics of infectious diseases, most particularly in diagnosis of infectious respiratory disease (see the Defense Sciences Board Summer 2000 Study Report and the Health Sciences Biodefense System (HSBS) final briefing). It is important that the device (e.g. microarray) will provide a cost-effective alternative to conventional approaches to the diagnosis, management, and surveillance of infectious diseases, most particularly respiratory infections. It is equally important that the device is supported with design and analysis informatics, and to ensure that decision quality information from that device will be transmissible and interpretable by a variety of care providers, public health officials, and decision makers. Thus, it is a critical objective of this invention that the device (e.g. microarray) be a key component of an integrated system comprised of a local point-of-care diagnostic device that provides automated, two-way data sharing between health care providers, public health officials, and decision makers (this a maybe a cross-reference to the business model patent that is in the queue). The invention described herein might fulfill this role in at least two ways: (1) through reduction in array size (i.e. lower cost), process automation, and the availability of portable hardware for processing resequencing arrays, the invention could become the object of the point-of-care device itself, and (2) in the event of lower-cost or easily automated microarray alternatives, the resequencing array can be a higher echelon component in a diagnostics/surveillance pipeline. In the latter case, the lower-cost alternative device will provide for initial sample processing, pathogen target enrichment, amplification, decision information at the point-of-care, while the resequencing capability provides for a sequential testing capability by facilitating a more detailed interrogation of the sample when desired.
General Strategy for Resequencing Microarray Chip Design
According to the present invention, the process of designing resequencing microarray chips is carried out by selecting pathogen genomic sequences having sequence properties that make them unique to a small number (ideally one) of pathogens, or are highly conserved, allowing them to detect many types of microbial species at the family or genus level, or are moderately conserved and selected as “prototype” regions. Prototype. regions will have an intermediate level of sequence homology across a group of microbial species and allow for both efficient hybridization and unique identification of most or all of a subtype of pathogenic species. The design strategy for layout of resequencing tiles includes leveraging from alignment of similar sequences and application of a consensus probe sequence for tiling to the chip. The consensus sequence might not be identical to any of selected pathogens, but will interact with a plurality of similar pathogen genomes. The patterns of perfect and imperfect matching of real pathogen genome sequences with the consensus sequence will provide diagnostic discrimination power.
In a particular embodiment of the present invention is a general protocol for designing resequencing microarray chips that permit the identification and designation of pathogens present in a sample (e.g., a partially purified sample, a purified sample, an enriched sample, a biological sample, etc.). The various aspects of this design and validation protocol are embodied in the following 6 “phases.”
Phase 1: Pathogen Identification:
a. Pathogen list (e.g. Adenovirus; Influenza; Streptococcus pyogenes)—provided by pathogen experts or public domain
b. Genus/species (new pathogen i.e. Severe acute respiratory syndrome (SARS) variant of corona virus)
c. Species/subspecies (Epidemiologic tracking; forensics)
d. Pathogen unique segments (cross hybridization issues)
e. Issues of genetic shift/drift (e.g. Influenza, HIV)
f. Drug resistance markers
g. Pathogenicity-related genes or virulence markers (useful for diagnostic and prognostic purposes)
h. Markers for genetically engineered traits
i. Plasmid DNA sequence (Bluescript, PUC etc.)
a. Common respiratory pathogens (and close relatives)
b. Biothreat agents: (identified by Centers for Disease Control)
c. Consideration of background or commensal organisms in:
a. Sequence Homology (relative to pathogen identity)
b. Functional Sequences (relative to pathogenicity and patient management)
a. Identify Gene Accession Number(s)
b. BLAST search: (Inclusion/Exclusion criteria)
c. Pragmatic Issues
a. Define chip “real estate” (total sequences represented)
b. Refer to primary objectives for prioritization criteria. Key issue include:
a. Design controls
b. Develop pathogen chip annotation schema:
c. Chip Validation Experimental Framework:
The following table (Table 1) represents a preferred (but not limiting) set of pathogens (both viral and bacterial) that may be used within the context of the present invention as designed for the detection and diagnosis of common respiratory pathogens:
For the intended use of a broadly distributed respiratory diagnostic device with built-in surveillance capability for agents of bioterrorism, the list of pathogens chosen for inclusion onto the chip would also include those selected from the U.S. Centers for Disease Control (CDC) Category A, B, and C bioterrorism agents. These are most notably, but not limited to, include:
Bacillus anthracis (targets: lethal factor, protective antigen)
Yersinia pestis
Smallpox (variola major)
Francisella tularensis
Viral hemorrhagic fevers (filoviruses [e.g. Ebola, Marburg] and arenaviruses [e.g. Lassa, Machupo]
CDC Category B
Brucella abortus (2308 B37), Brucella melitensis (F6145), Brucella suis (A44)
Burkholderia mallei (Glanders)
Burkholderia pseudomallei (Meliodosis)
Psittacosis (Chlamydia psittaci)
Typhus fever (Rickeusia prowazekii)
Viral encephalitis (alphaviruses [e.g., Venezuelan equine encephalitis, eastern equine encephalitis, western equine encephalitis])
CDC Category C
Emerging infectious diseases such as Nipah virus and hantavirus
A complete and updated listing of these agents may be found on the CDC website (http://www.cdc.gov/).
For purposes of illustration of the present invention, two resequencing microarray chips (RPMV1 and RPMV2) will be described herein below:
Resequencing Respiratory Pathogen Microarray Version 1 (RPMV1 Chip)
RPMV1 was made using a high-density Affymetrix microarray fabrication process having an individual probe feature size of 18×18 microns. At this density, 29.7 kb of total pathogen target sequence was tiled for resequencing. The fabrication was performed as part of a pre-production beta-test of an Affymetrix commercial product (CustomSeq) that was intended for SNP detection in an arbitrary collection of sequences.
The following overall design strategy was used for the RPMV1:
The Affymetrix CustomSeq design protocol was followed. While there were a variety of ongoing efforts within the project consortium to non-specifically amplify targets for microarray analysis, the present inventors made every effort to tile sequences on the array that could interrogate conventional (specific primer pair) PCR amplicons. Whenever possible, primer pairs were developed or adapted for conserved sequences that flanked the variable regions to be interrogated with tiled microarray probe sets. This allowed for the ability to: (1) directly amplify with an existing hardware platform (e.g. RAPID Light Cycler, Idaho Technologies), (2) provide control measurements for comparison with generic or total amplification strategies that are on the immediate horizon. In general, our strategy was to match the hybridization patterns with specific strains of organisms and to detect subtle variations in sequence that corresponds to pathogenicity and drug resistance.
Adenovirus (double-stranded DNA virus) types 4, 5, and 7 were designated as “prototypes” for the E, C and B subgroups, respectively. Specifically, the present inventors postulated that resequencing on the prototype-tiled regions would allow detection and identification of subtle sequence variations between the subgroup members. Three target gene regions were selected, specifically from E1A, hexon, and fiber genes. However, it was not postulated nor anticipated which regions or which sections of any of the tiles would allow unique identification.
Thirteen adenovirus genomes were completely sequenced as part of the Epidemic Outbreak Surveillance program. The names, accession numbers, and sources for these are listed in Table 6 of the Examples section. Multiple sequence alignments were performed to determine variable regions of the E1A, hexon, and fiber genes that were flanked by conserved regions that could be used to amplify multiple adenoviruses with a single set of degenerate primers (Lin et al., 2004). These common regions for E1A, hexon, and fiber genes were obtained for each of the 3 prototype adenoviruses that are associated with respiratory illness: 7 (subgroup B), 5 (subgroup C), and 4 (subgroup E) were submitted to Affymetrix as part of the 29.7 kb total pathogen target sequence for tiling on the RPMV1 microarray.
Adenovirus Taxonomy:
Sub-Group B: 3, 7, 11, 14, 21, 34,-35 and 50
Sub-Group C: 1, 2, 5, and 6
Sub-Group E: 4
The present inventors made the hypothesis that if they tiled three genes (E1A, Fiber and Hexon) on the array for prototype sub-group representatives, types 7, 5, and 4 (for subgroups B, C, and E, respectively), they would be able to identify any of the fully-sequenced types (listed above) by variations in the hybridization patterns that map to their sequence differences.
Influenza A and B viruses, which are negative-polarity single stranded RNA viruses (ssRNA), were represented with prototype regions for hemagglutinin (HA) neuraminidase (NA) and matrix (M). These genes were represented for three types of Influenza A (H1N1, H3N2 and H5N1) and Influenza B. Influenza is one of the best examples of a prototype model system, as hundreds, if not thousands of influenza strains have been at least partially sequenced, and most have been sequenced for the hemagglutinin and neuraminidase segments.
Prototype Influenza HA, NA and M genes were chosen from strains that were either identical of closely related to the three vaccine strains recommended for the Northern hemisphere by the World Health Organization;
These sequences were publicly available from the Los Alamos National Laboratory influenza Internet database. The present inventors postulated that the sequence calls made for unknown Influenza A or B on the prototype tile regions would allow the identification of that target if it was sufficiently similar to allow similarity search-based querying.
The remainder of RPMV1 was populated with tiles for a variety of common respiratory pathogens, the first set being viruses:
These viral pathogens had relatively little sequence available compared to adenovirus and influenza, complicating efforts to perform analogous tests for use of prototype sequences to identify a large number of related strains.
Common bacterial pathogens were also chosen:
In addition, the following plasmid-conferred antibiotic resistance genes were represented on the RPMV1 chip:
The following biological threat agents were also included on the RPMV1 chip:
Lassa, Machupo])
In addition to the hybridization controls prescribed by Affymetrix (see CustomSeq protocol), it was considered necessary to include internal process controls to verify target isolation, reverse transcription (RT) and cDNA/DNA amplification. Thus, we included a specialized vector containing gene inserts from the plant Arabidopsis thaliana (Yang et al., 2002) and provided tile regions on the array for several genes inserts from that vector.
For all targets other than adenovirus and influenza, the present inventors selected targets from the open literature (e.g., NCBI GenBank) for genes containing diagnostic (via PCR) regions and tiled that diagnostic region or a larger coding region containing that region. The full listing of all target genes, PCR primer sites, and primers for the RPMV1 are shown in Table 8 (infra) and the attached Sequence Listing. The sequences submitted for tiling and chip fabrication are summarized in Table 7 (infra) and the attached Sequence Listing.
However, the present invention is not limited to the specific sequences and/or microorganisms (bacteria or virus) listed above. It is well within the purview of the skilled artisan to tailor the “pathogen chip” to meet the particular requirements of the intended field test. Such tailoring may be on a gene-level or on a microorganism level. Clearly, the artisan would be greatly assisted in the efforts by the general protocol above.
Resequencing Microarray Chip Version 2 (RPMV2 Chip)
RPMV2 is built on a high-density Affymetrix microarray having an intended feature size of 8×8 microns. At this density, approximately 300 kb of target sequence information may be tiled for resequencing.
The general layout strategy described in Tables 2-4 (infra). The RPM V2 design allowed for roughly 10 times the sequence content of the RPMV1, and thus allowed inclusion of a large number of pathogens, which included expanded representations of all the pathogens present on the RPMV1, numerous other respiratory pathogens, and virtually the entire set of CDC Category A, B, and C bioterrorism agents. The full listing of the RPMV2 content is provided in Example 8. The RPMV2 design expanded molecular epidemiology capability by enabling inclusion of probes, such as the emm types for Streptococcus pyogenes. This feature will provide the epidemiologist/public health professional an ability to determine a likely index disease case and to study transmission of the typed pathogen in enclosed populations (ails, dormitories, nursing homes, military barracks, etc.). These probes may also predict pathogen virulence and clinical presentation of disease in the population.
The number of pathogens represented can also be expanded with increasing probes per unit area of microarray. The chips of the present invention take advantage of high feature density to simultaneously probe and discriminate among a large number of candidate pathogens that may be present in a single clinical specimen.
It is also important to note that gene expression markers can conceptually be tiled onto the array in the same manner as genomic markers. Thus, sequences representing pathogen transcripts (RNA templates coding for protein synthesis) can be simultaneously positioned on arrays with those allowing resequencing of genomic markers. This would enable the microarray to detect not only the presence of a pathogen, but would also allow determination of viability, since transcriptional markers would only be present in intact pathogens.
Streptococcus
Streptococcus pyogenes
Streptococcus
pneumoniae
Staphylococcus aureus
Mycoplasma
pneumoniae
Bordetella pertussis
Chlamydia
Chlamydia pneumoniae
Chlamydia psittaci
Neisseria meningitidis
Bacillus
Bacillus anthracis
Bacillus cereus
Bacillus thuringensis
Bacillus subtillus
Yerslnia pestis
Francisella tularensis
Burkholderia
Burkholderia mallei
Burkholderia
pseudomallei (Meliodosis)
Burkholderia cepacia
prowazekii)
Brucella
Brucella abortus (2308
Brucella melitensis
Brucella suis (A44)
Brucella canis
Brucella ovis
Brucella neotoma
Arcanobacterium
hemolyticum
Haemophilus influenzae
Moraxella catarrhalls
Mycobacterium
tuberculosis
Clostridium
Clostridium botulinum
Clostridium perfringens
Clostridium tetani
Coxiella burnetil
Cryptosporidium
parvum
E. coli 0157:H7
Ricinus communis
Salmonella
Salmonella enterica
Salmonella typhimurium
Shigella
Shigella dysenteriae
Shigella flexneri
Vibrio cholerae
Legionella
pneumophilia
Histoplasma
capsulatum
Blastomyces
dermatiditis
Coccidioides immitis
Corynebacterium
Corynebacterium
diphtheriae
Streptococcus
Streptococcus pyogenes
Streptococcus pneumoniae
Staphylococcus aureus
Mycoplasma pneumoniae
Bordetella pertussis
Chlamydia
Chlamydia pneumoniae
Chlamydia psittaci
Neisseria meningitidis
Bacillus
Bacillus anthracis
Bacillus cereus
Bacillus thuringensis
Bacillus subtillus
Bacillus globigii
Yersinia pestis
Francisella tularensis
Burkholderia
Burkholderia mallei
Burkholderia pseudomallei
Burkholderia cepacia
prowazekii)
Brucella
Brucella abortus (2308
Brucella melitensis (F6145)
Brucella suis (A44)
Brucella canis
Brucella ovis
Brucella neotoma
Arcanobacterium
hemolyticum
Haemophilus influenzae
Moraxella catarrhalis
Mycobacterium
tuberculosis
Clostridium
Clostridium botulinum
Clostridium perfringens
Clostridium tetani
Coxiella burnetii
Cryptosporidium parvum
E. coli 0157: H7
Ricinus communis
Salmonella
Salmonella enterica
Salmonella typhimurium
Shigella
Shigella dysenteriae
Shigella flexneri
Vibrio cholerae
Leglonella pneumophilla
Histoplasma capsulatum
Blastomyces dermatiditis
Coccidioides Immitis
Corynebacterium
Corynebacterium
diphtheriae
Chip Size 293826
The sequences submitted for RPMV2 tiling and chip fabrication were based on the Affymetrix instruction file summarized in Table 5, which corresponds to the sequences appearing as SEQ ID NOs: 59-421.
Sample Preparation
In any embodiment of the present invention, the target nucleic acids (DNA and/or RNA) may be contained within a biological sample. The term “biological sample”, as used herein, refers to a sample obtained from an organism or from components (e.g., cells) of an organism. The sample may be of any biological tissue or fluid. Alternatively, the sample may be one taken from the environment (air, soil or water). Frequently the sample will be a “clinical sample” which is a sample derived from a patient. Such samples include, but are not limited to, sputum, nasal wash, nasal aspirate, throat swab, blood, blood cells (e.g., white cells), tissue or fine needle biopsy samples, urine, peritoneal fluid, visceral fluid, and pleural fluid, or cells therefrom. Within the context of the present invention it is preferred that the clinical sample be a nasal wash, nasal aspiration or a throat swab. In a particularly preferred embodiment the clinical sample is a nasal wash. Biological samples may also include sections of tissues such as frozen sections taken for histological purposes, or from non-human animal sources, plants, or environmental sources such as water, air, or soil.
In order to be detected using microarrays, the target nucleic acids may require some degree of processing. For the intended purposes, this will involve one or more of the following target processing steps: (1) isolation, (2) enrichment for target sequences of interest, (3) amplification, (4) labeling, and (5) hybridization. Preferred embodiments corresponding to each of these processing strategies are described below; however, the present invention is not intended to be limited thereto. To this end, the skilled artisan would readily appreciate alternative methods corresponding to the aforementioned processing strategies both based on those commonly in use and as described in U.S. Pat. No. 6,638,717, U.S. Pat. No. 6,376,191, U.S. Pat. No. 5,759,778, U.S. Pat. No. 6,268,133, and U.S. Pat. No. 6,613,516 (each of which are incorporated herein by reference in its entirety).
Target Nucleic Acid Isolation
In an embodiment of the present invention, the target nucleic acids (RNA and/or DNA) to be assayed are isolated prior to amplification of the same. Methods of isolating nucleic acids are well known to the skilled artisan.
In a preferred embodiment, the target nucleic acid isolation may be performed using a MasterPure™ DNA Purification Kit (Epicentre Technologies, Madison, Wis.) ethanol purification method (per manufacturer's instructions). In another preferred embodiment, the target nucleic acids will be isolated using a rapid microcentrifuge technique, as in the use of Xtra Amp Kit (XTRANA, Inc. Broomfield, Colo.). In yet another preferred embodiment, the nucleic acids will be isolated using an automated device for said purpose, such as a GeneXpert (Cepheid, Sunnyvale, Calif.) or using a robot for magnetic bead-based isolation (e.g. Qiagen or Beckman).
A variety of other commercial products are available that are geared towards purification and concentration of nucleic acids from complex matrices. In addition to the methods described above and in the invention, alternatives include:
QIAamp DNA Mini Kit (Qiagen)
(For Purification of Genomic, Mitochondrial, Bacterial, Parasite, or Viral DNA)
The QIAamp DNA Mini Kit simplifies isolation of DNA from human tissue samples with fast spin-column or vacuum procedures. DNA binds specifically to the QIAamp silica-gel membrane while contaminants pass through. PCR inhibitors such as divalent cations and proteins are completely removed in two efficient wash steps, leaving pure DNA to be eluted in either water or a buffer provided with the kit. QIAamp DNA technology yields genomic, mitochondrial, bacterial, parasite, or viral DNA from human tissue samples ready to use in PCR and blotting procedures.
RNeasy Mini Kit (Ambion)
The RNeasy Mini Kit allows efficient purification of total RNA from very small amounts of tissue or cells. Total RNA is easily purified from animal cells or tissues, Gram-positive or Gram-negative bacteria, or yeast. RNeasy technology simplifies total RNA isolation by combining the stringency of guanidine-isothiocyanate lysis with the speed and purity of silica gel-membrane purification.
UltraClean™ Tissue DNA Kit (Mo Bio Laboratories, Inc.)
Fresh or frozen tissue samples are homogenized using bead-beating technology to lyse the cells. Lysates are loaded onto a silica spin filter. During a brief spin, the DNA selectively binds to the silica membrane while contaminants pass through. Remaining contaminants and enzyme inhibitors are removed by a wash step. Pure DNA is then eluted into certified, DNA-free Tris buffer.
UltraClean™ Tissue RNA Kit (Mo Bio Laboratories, Inc.)
Prepared fresh or frozen tissues are homogenized using a tissue homogenizer or mortar and pestle in the presence of lysis solutions. The RNA is captured on a silica membrane spin filter while contaminants are passed through the filter by centrifugation. The filter is washed to remove any remaining contaminants and salts. The RNA is then eluted into certified RNase-free water (provided). RNA is of high quality and is ready for any downstream applications.
Wizard® Genomic DNA Purification Kit (Promega)
The Wizard® Genomic DNA Purification Kit is designed for isolation of DNA from white blood cells, tissue culture cells and animal tissue, plant tissue, yeast, Gram-positive and Gram-negative bacteria. The Wizard® Genomic DNA Purification Kit is based on a four-step process. The first step in the purification procedure lyses the cells and the nuclei. For isolation of DNA from white blood cells, this step involves lysis of the red blood cells in the Cell Lysis Solution, followed by lysis of the white blood cells and their nuclei in the Nuclei Lysis Solution. An RNase digestion step may be included at this time; it is optional for some applications. The cellular proteins are then removed by a salt precipitation step, which precipitates the proteins but leaves the high molecular weight genomic DNA in solution. Finally, the genomic DNA is concentrated and desalted by isopropanol precipitation.
SV Total RNA Isolation System (Promega)
The SV Total RNA Isolation System provides a fast and simple technique for the preparation of purified and intact total RNA from tissues, cultured cells and white blood cells using this membrane-based purification system. The system incorporates a DNase treatment step directly on the membrane of the minicolumn. Purification is achieved without the use of phenol:chloroform extractions or ethanol precipitations, and there is no DNase carryover in the final RNA preparation.
RNAqueous Technology (Ambion, Inc.)
RNAqueous kits can be used to purify total RNA from many different tissues and cells. Cells or tissue are disrupted in a guanidinium thiocyanate solution; this chaotropic agent effectively lyses cells and inactivates endogenous ribonucleases. The lysate is then diluted with an ethanol solution and applied to an RNA-binding glass fiber filter. Proteins, DNA and other contaminants are removed in three rapid washing steps, and the bound RNA is then eluted in concentrated form.
Nucleic Acid Isolation Robotics
In addition to those methods described above, several vendors (e.g. PSS BIO Instruments, Roche Diagnsotics, Qiagen, Caliper) manufacture both small (benchtop) and/or high-throughput liquid handling robots and associated reagents that can be used instead of the manual methods described above. In a preferred embodiment, one or more such robots and their associated reagensts will be used to automatically isolate nucleic acids for the subsequent processing (background removal and amplification).
Target Amplification
One set of technical challenges for pathogen detection with microarrays arises because of the difficulty in obtaining samples with a sufficient quantity of pathogen nucleic acids. Thus, for a majority of sample types, some sort of amplification will likely be required to provide sufficient copies of pathogen gene markers for detection by microarray hybridization. Multiplex PCR as a microarray preparative step is practically limited to tens of different primer pairs, not thousands, and increasing numbers of primer pairs give rise to varying numbers of spurious amplicons. However, discrimination occurs when the labeled amplicons are required to hybridize to specific probes on the array surface and, fortuitously, spurious amplicons may not detected by the microarray assay (Chizhikov et al., 2001).
Multiplex PCR
Conserved (degenerate) multiplex PCR serves to reduce the systemic bias introduced with specific PCR protocols by designing a series of primers selected to target the conserved regions flanking species-specific variable regions of a gene(s) to be probed. In the examples of the present application, the strategy for E1A, fiber, and hexon genes has been illustrated; however, the skilled artisan may expand this strategy to target any gene that is conserved across a broad spectrum of species, but still has species-specific variable regions. To this end, candidate genes and the specific regions (conserved and variable) can be readily identified by global or local homology searches (i.e., sequence alignment).
Applicants describe below a general strategy for the selection of targets and primer design:
As used herein the term “primer” (and generally appreciated in the art) refers to an oligonucleotide that is capable of acting as a point of initiation of polynucleotide synthesis along a complementary strand when placed under conditions in which synthesis of a primer extension product that is complementary to a polynucleotide is catalyzed. Typical polynucleotide synthesis conditions include the presence of four different nucleotide triphosphates or nucleoside analogs and one or more enzyme to catalyze polymerization (e.g., a DNA polymerase and/or reverse transcriptase) in an appropriate buffer (“buffer” includes substituents which are cofactors, or which affect pH, ionic strength, etc.), and at a suitable temperature.
To facilitate specific PCR, amplification primers typically range from 13 to 35 nucleotides, preferably from 20 to 25 nucleotides. The nucleotide sequence of the primer must have sufficient sequence complementarity to the target sequence in order to facilitate hybridization. Although the degree of complementarity will depend in large part on the length of the primer, the degree of complementarity generally is at least 80%, preferably at least 90%, more preferably at least 95%.
For random PCR protocols, the preferred primer length is from 6 to 10 nucleotides. The sequence would ideally include all permutations of a hexameric (2 permutations) to a decameric (210 permutations).
For the case of “prototype” region design, the preferred scenario would be to have conserved primers that flank variable regions of target genes. This was the case in RPMV1 for adenovirus and has been adopted by present inventors for Influenza A (using conserved 3′ and 5′ ends of Influenza A segments for hemagglutinin (HA) neuraminidase (NA) and matrix (M). This concept can be extended to any of a large number of pathogen types because highly conserved regions are ubiquitous in nature and degenerate primers can be designed for those sequences.
When targets are not selected as “prototypes”, the amount of real estate on the chip can be greatly reduced compared to prototype regions. The objective in these cases is not necessarily to identify a particular strain or sub-variant of the species, but rather to allow for enough base calls to be made for an unambiguous statistical estimate demonstrating that the sequence corresponds to the pathogen of interest and not to a closely related species or non-virulent strain. With this chip design of multiple, small targets, it is not feasible to design and optimize conditions for the large number of specific PCR reactions that would be needed and total amplification becomes the optimal amplification strategy.
Homology, sequence similarity or sequence identity of nucleotide or amino acid sequences may be determined conventionally by using known software or computer programs such as the BestFit or Gap pairwise comparison programs (GCG Wisconsin Package, Genetics Computer Group, 575 Science Drive, Madison, Wis. 53711). BestFit uses the local homology algorithm of Smith and Waterman, Advances in Applied Mathematics 2: 482-489 (1981), to find the best segment of identity or similarity between two sequences. Gap performs global alignments: all of one sequence with all of another similar sequence using the previously described methods (Needleman & Wunsch, 1970). When using a sequence alignment program such as BestFit, to determine the degree of sequence homology, similarity or identity, the default setting may be used, or an appropriate scoring matrix may be selected to optimize identity, similarity or homology scores. Similarly, when using a program such as BestFit to determine sequence identity, similarity or homology between two different amino acid sequences, the default settings may be used, or an appropriate scoring matrix, such as blosum45 or blosum80, may be selected to optimize identity, similarity or homology scores.
Targets may also be amplified using a combination of specific PCR reactants in what is referred to as “multiplexed PCR”. In this strategy, PCR primer pairs for all target regions on the RPM are combined into one reaction miture. This is a preferred method in cases where one or more of the pathogens is present in too low a concentration to be detected using random amplification strategies alone (described below).
Random Amplification Strategies
Despite the sensitivity and specificity associated with PCR amplification, the inherent bias and limited throughput of this approach limits the principal benefits of downstream microarray-based applications. As successful identification depends almost entirely on appropriately chosen primer sets, all PCR-based testing requires a priori knowledge pertaining to the identity of the contaminating organism. As such, there remains a critical need for advanced diagnostic systems that can rapidly screen clinical and environmental samples without bias for specifically anticipated sequences. The advantage of microarray-based detection is that it can combine unbiased (e.g. random) nucleic acid amplification strategies with the subsequent discriminatory capability of microarrays, resulting in high sensitivity, specificity and throughput capacity.
The inventors have addressed the aforementioned issues through the use of alternative approaches and have investigated four front-end amplification strategies: random primed, isothermal Klenow polymerase-based, 029 DNA polymerase-based and multiplex PCR (Vora et al., 2004).
Klenow-based amplification—The Klenow fragment of DNA polymerase I can be used for isothermal, random amplification of target DNA of unknown sequence with random octamers oligonucleotide primers. For the present invention, this can be performed using the BioPrime® DNA Labeling System (Invitrogen). Biotinylated amplicons are generated (following the manufacturer's recommended labeling protocol with minor modifications). This process is typically performed at 37° C. for 4 hours. Klenow amplification results in some high molecular weight amplification products, but the vast majority of amplicons are 50-450 basepairs (bp) in length. As such, these amplicons are suitable for direct hybridization to DNA microarrays.
Random PCR amplification—Random PCR (rPCR) amplicons can be generated using the 2.5× Random Primers Solution and 10× DNTP mix from the Invitrogen BioPrime DNA Labeling System. The reaction components typically include: 1× PCR buffer (Qiagen), 2.5 mM MgCl2, 1× DNTP Mix (containing biotin-14-dCTP), 5 μl of 2.5× random octamers and 5 U Taq DNA polymerase (Qiagen). The amplification reactions are performed using the following conditions for 35 cycles: 94° C. for 30 sec, 26° C. for 2 min, and 72° C. for 1 min.
φ29-based amplification—Isothermal φ29 DNA polymerase-based amplification with random hexamers can be performed using the TempliPhi™ 100 Amplification Kit (Amersham Biosciences Corp., Piscataway, N.J.). Biotinylated amplicons are generated following the manufacturer's recommended protocol with minor modifications (14.4 μl amplification reactions contained 0.4 μl enzyme mix and 3 μl of 350 μM biotin-14-dCTP and are incubated at 30° C. for 16 h).
The majority of both the rPCR and φ29 DNA polymerase amplification products are too large to migrate into electophoretic gels (>1500 bp). Large amplification products hybridize poorly to immobilized probes on two-dimensional surfaces due to spatial and steric constraints (Vora et al., 2004). Thus, the rPCR and φ29 polymerase amplification products are digested with a DNase I fragmentation buffer to achieve amplicon sizes that are comparable to the Klenow and multiplex amplicons prior to hybridization.
Tandem Amplification
Increased sensitivity to pathogen targets can be achieved using a combination of the random amplifications strategies described above. Tandem [Klenow+Klenow] and [φ29+Klendow] tandem random amplification strategies can provide better sensitivity to enriched pathogen targets than multiplex PCR. The same random amplification strategies are also able to detect diagnostic genomic targets in spiked environmental water samples containing a 63-fold excess of contaminating DNA (Vora et al., 2004). The results presented underscore the feasibility of using random amplification approaches and begin to systematically address the versatility of these approaches for unbiased pathogen detection from environmental sources.
Combination of Target Enrichment and Random Amplification
Random amplification methods result in amplification of non-target DNA as well as target DNA. Thus, the inventors describe herein a variety of methods for enrichment of pathogen diagnostic target sequences in advance of random amplification steps. This is especially important to improve sensitivity and/or specificity of detection of analytes in complex matrices such as clinical (e.g. nasal wash or throat swab) or environmental (e.g. water, soil, air collector) samples.
Complex samples (clinical and environmental) contain significant, and sometimes overwhelming amounts of unknown genomes and genomes not of interest (non-specific background). One approach to reduce the non-specific background is to expose the nucleic acids to enzymatic processes that selectively digest the background genomic sequences. This process is described below within the examples.
Another strategy for reducing non-pathogen genomic content is to use background genomic sequences from a comparable complex sample known to be negative for the organism(s) of interest and use it to directly remove background genomic sequences. Such “subtractive” methods can include: (1) immobilization of the background genomes on solid phase surfaces such as gels or beads, followed by hybridization with the test sample to absorb background genomes from the samples. One can also use unlabeled background genomes during hybridization of targets on microarray to block the effects of non-specific binding due to the labeled background genomes. This blocking approach has the advantage in that it does not require additional steps during assaying. Both such strategies are also described in the Examples section.
Alternatively or concurrently, one can also use strategies to selectively enrich for the pathogen target nucleic acids prior to random amplification. One possibility for direct selection (also known as positive selection) from samples is to use immobilized probes on a solid phase support (magnetic beads or gel matrices) to selectively enrich for genomic targets of interest. Ideally, the probes on the solid support should not be detectable on the array, as they would result in false signals in the event that the enrichment molecules are carried over or inadvertently amplified. As such, the probes on the beads should select for target genomes via an adjacent region or slightly overlapping with the region of the analytes' genomes to be queried on the microarray. If there is a slight overlap between the probe sequences and the analytes' sequences that is detectable on the array, then this queried region should be masked out during in silico analysis. In the event that it is desirable that the enrichment probes have homologous sequence to that on the microarray, then those probes should be comprised so as to not undergo subsequent amplification, or be made susceptible to selective enzymatic digestion prior to amplification. One can use combinations of the above and other strategies if necessary according to the specifics of the sample and applications of interest.
Alternative Subtraction or Enrichment Methods
Immunoprecipitation is another way to enrich the pathogens' nucleic acids. Antibodies for pathogens of interest could be mixed with clinical samples, then precipitated with anti-IgG to pull down pathogens, thereby removing background genomic DNA. This will be of particular importance when it is desirable to correlate specific genomic traits with a single pathogen, such as antibiotic resistance markers or indicators of intentional genetic manipulation.
Size exclusion is another method by which to subtract or enrich pathogen DNA. Exemplary size exclusion methods include: gradient centrifugation, column, or centrifugal filter units. Gradient centrifugation or column separation methods are time-consuming and require special set-up in the laboratory. The use of “centrifugal filter units” for separation of high- from low-molecular-weight species suffers from inconsistent and, often, low recovery rates. Each of these methods requires a large volume of starting material. Other size exclusion methods include: flow cytometry or electrophoresis methods, such as fluorescence-activated cell sorting (FACS) or with a special electrophoretic chip used in an Agilent bioanalyzer.
General Target Enrichment and Amplification Strategies
The present inventors have described methods for subtraction of “normal” human and microbial nucleic acids from nasal wash specimens. This same principle can also be applied to samples of any origin (clinical or environmental) where a representative “normal” mixture of nucleic acids can be collected, pooled, and prepared as a subtractive reagent.
A variety of approaches can also be applied to enrich for the pathogens or specific targets that are tiled on the array prior to performing a total amplification strategy. One such approach would use immobilized variants of the sequences that are tiled on the array. These variants would be bound to a solid phase component used for separation (beads, matrices, etc.) and serve to enrich for targets that will be amplified without being amplified themselves. Making the enrichment probes susceptible to enzymatic degradation can do this or they might be comprised of modified nucleic acids that would not be amplifiable.
A more preferred embodiment would utilize probes that recognize sequences that are adjacent to target gene sequences and not represented on the tiled regions of the chip. In this manner, their amplification in a non-biased amplification strategy would not result in artifacts.
As conventional DNA sequencing technologies can be considered as part of an overall information collecting process in basic research, so can the described resequencing approach be considered a component of an overall pathogen detection/characterization scheme. Wang et al. (Wang et al., 2003) describe a method for using a spotted DNA microarray comprised of long oligonucleotides (70-mers) that recognized conserved sequences, which are within a family of pathogens. Target sequences bound to these locations were isolated and used in a conventional DNA sequencing approach to allow further characterization. In an analogous manner, a very preferred embodiment would involve the use of the resequencing array to replace the steps related to alternative forms of DNA sequencing, thus providing specific pathogen characterization within hours instead of days and enablement of an effective biodefense system.
The method of the present invention preferably does not make use of specific sequences for amplification (PCR). The invention illustrates that there are alternate methods for enrichment of pathogen nucleic acids, for example: using solid phase support separations, prior to applying reduced bias, and isothermal (e.g. random-primed Klenow or strand displacement) amplification. In a preferable embodiment, a skilled technician in a conventional laboratory setting would be capable of performing time- and cost-effective sample preparation using a minimal set of automated steps to perform microarray experiments. In a highly preferred embodiment, a minimally skilled technician (medical technologist or medic) in a field environment (medical level 1) would be capable of performing manual separation/enrichment of pathogen target nucleic acids using a handheld instrument and perform isothermal amplification of pathogen targets with few reagents and technical steps.
Ampification and Hybridization Following Background Subtraction
In addition to random RT-PCR using primer D: GTITCCCAGTCACGATCNNNNNNNNN (SEQ ID NO: 573), and Primer E: GTTTCCCAGTCACGATC (SEQ ID NO: 574) (Kessler et al., 2004), analogous, previously described primer sets (Wang et al., 2002), variants of those primer sets, and/or random (6-9 mers) to synthesize first strand cDNA from RNA viruses (pathogens) may be employed. A subtractive hybridization protocol could be employed after first strand cDNA synthesis is complete to reduce the amount of background DNA in clinical samples before subjecting complex (i.e. clinical or environmental) samples to one or more DNA amplification step(s).
One method is direct subtraction of the background genomic DNA from amplified products (first strand cDNA synthesis, then total amplification) in hybridization solution with COT-1 human DNA (which consists largely of rapidly annealing repetitive elements). Another method is bead-based subtraction of background genomic DNA from clinical sample after first strand cDNA synthesis and prior to the DNA amplification step.
The third method is the combination of the above methods. In this case, the human genomic background DNA from complex samples would be subtracted using bead-based subtraction after first strand cDNA synthesis and prior to the DNA amplification step(s). Subsequently, the background DNA could be further subtracted from complex sample in hybridization solution with sequences designed to selectively capture human DNA and RNA. The DNA amplification step itself could be aimed at amplifying the cDNA products formed in the random RT step specifically, non-specifically, or through a combination of methods to amplify both the specific primer site-labeled RT-PCR products as well as pathogen genomic DNA targets and other non-cDNA targets that did not arise from the random RT-PCR step. One approach for doing so would be to ligate a specific primer sequence to the pathogen DNA genomic targets (this could be the same primer as used attached to cDNA products or a separate one), allowing a single PCR step to amplify all RNA and DNA pathogen targets.
RNA Substraction
Where separate processing pathways exist for detection of RNA and DNA pathogens the following may be used. For development of the RNA processing pathway, spiked and clinical (nasal wash and throat swab) Influenza A positive specimens were processed using different methods (random, universal, and multiplexed PCR). However, to achieve acceptable levels of sensitivity with random amplification approaches, human messenger and ribosomal RNA should be subtracted using the MICROBEnrich™ kit (Ambion, Inc., Austin, Tex.) in accordance with the manufacturer's instructions. A preferred random amplification strategy may be empolyed adapted from a previously-described method (Wang et al., 2002).
It was found by the present inventors that by employing this method positive detection of H3N2 and H1N1 Flu A detection in 16/19 different culture-positive Flu A nasal wash and throat swab specimens at concentrations as low as 0.45-3.75 pfu/150 μL sample aliquot may be achieved. It was further determined it was possible to detect Flu A in 6/8 specimens in the low concentration range following subtraction of human RNA.
Using random amplification approach, without background subtraction, the flu sensitivity is about 0.25 pfu/μl (1 ng/μl) in spiking nasal wash. In clinical samples, the sensitivty without subtraction is estimated to be about 1-10 pfu/μl.
DNA Substraction
Random amplification for DNA samples may be performed with either bacteriophage φ29 DNA polymerase or modified random amplification protocol from previously published paper (Wang et al. 2002; Wang et al.2003). Briefly, DNA amplification utilizing bacteriophage φ29 DNA polymerase with random hexamers can be conducted according to the instruction of GenomiPhi™ DNA Amplification Kit (Amersham Biosciences Corp., Sunnyvale, Calif.). The amplified products are then ethanol precipitated according to manufacture recommended protocol. DNA amplification utilizing modified random amplification may be performed with initial round of DNA synthesis with Sequenase™ version 2.0 DNA polymerase (United States Biochemical, Cleveland, Ohio) using primer D, followed by PCR amplification with primer E. For RNA amplification, viral samples are then amplified by a modified version of a random PCR protocol (Wang et al. 2002; Wang et al. 2003; Kessler et al. 2004). Briefly, 10 μl of total RNA can be reverse transcribed by using primer D and superscript III reverse transcriptase (Invitrogen Corp. Carlsbad, Calif.), and was then amplified by PCR with primer E.
Convergence of RNA and DNA Pathways
The combination of the RNA and DNA pathways is described in
Data Acquisition and Processing
Affymetrix/Microarray Apparatus
In general, a “microarray” is a linear or two-dimensional array of preferably discrete regions, each having a defined area, formed on the surface of a solid support. The density of the discrete regions on a microarray is determined by the total numbers of different target polynucleotides to be detected on the surface of a single solid phase support, preferably at least about 102/cm2, more preferably at least about 104/cm2, even more preferably at least about 106/cm2, and still more preferably at least about 108/cm2. As used herein, a DNA microarray is an array of oligonucleotide primers placed on a chip or other surfaces used to detect and/or isolate target polynucleotides. Since the position of each particular group of primers in the array is known, the identities of the target polynucleotides can be determined based on their binding to a particular position in the microarray.
One embodiment of the invention utilized standard Affymetrix hardware (Agilent GeneChip Scanner (phased out) and the Affymetrix Scanner 3000 workstation and Fluidics Station 450. In principle, the described invention does not require this equipment. Given the manufacturer's intended uses of the GeneChip system for quantitative gene expression profiling and high confidence SNP detection, the existing apparatuses are not optimized for the dynamic range of signal intensities or background interferences inherent to the present invention. Thus, a preferred embodiment incorporates the use-of image acquisition methods, including the use of spatial frequency filtering and image enhancement, taking advantage of the inherent regularity of the feature dimensions to impose a filtering algorithm (e.g. edge enhancement, convolution, etc.) that allows better contrast of noisy images.
Affymetrix CustomSeq Protocol
In an embodiment of the present invention, sample processing is accomplished by employing the standard Affymetrix CustomSeq™ protocol. Generally, this method entails: (a) amplification of DNA probe-containing materials, (b) pooling and quantitation of amplified product, (c) fragmentation and labeling of the amplified product, (d) target hybridization, and (e) washing, staining, and scanning of the hybridized target. A detailed description of the Affymetrix CustomSeq™ protocol can be found in the product manual and protocol guide, which are available from the manufacturer and are incorporated herein by reference.
In the standard Affymetrix CustomSeq™ protocol, step (a) entails either long-range PCR or short-range PCR, with long-range PCR being the preferred amplification strategy. For each of these amplification strategies, the manufacturer recommended PCR DNA polymerases are Taq variants, AmpliTaq Gold DNA polymerase (short-range PCR) and LA Taq DNA polymerase (long-range PCR). Although not specifically recommended by the manufacturer, any DNA polymerase may be employed for step (a) so long as the DNA polymerase used is a high fidelity DNA polymerase.
To facilitate DNA amplification step (a), the standard Affymetrix CustomSeq™ protocol employs specific PCR primers. However, the use of specific PCR primers significantly limits the broad application of the inventive technique due to the introduction of systemic bias flowing from the specific PCR based methods that have heretofore been employed. Accordingly, in a preferred embodiment of the present invention step (a) of the Affymetrix CustomSeq™ protocol is replaced with an alternate amplification strategy, such as multiplex PCR, total amplification (GenomiPhi™), or random RT/PCR. These alternate strategies are discussed hereinabove. Conditions for optimal PCR amplification for each of the manufacturer recommended PCR strategies, as well as the preferred strategies of the present invention, can be determined by routine experimentation by the skilled artisan.
Since variability exists between PCR reactions, the Affymetrix CustomSeq™ protocol states that assay performance may be compromised if amplicon concentration in the hybridization varies by more than two fold. Therefore, step (b) of the Affymetrix CustomSeq™ protocol entails pooling PCR reactions and spectrophotometrically quantifying the same to ensure equimolar application of sample to the microarray.
However, the present invention poses several advantages over the SNP detection tailored protocols of the Affymetrix CustomSeq™ method. In particular, the use of the increased density chips with sequence length-independent similarity searches (BLASTN) in the present invention affords that fewer assumptions must be made in advance of selecting sequences for tiling. Furthermore, the use of length-independent similarity searches (BLASTN) removes the constraint that a particular known subsequence be successfully resequenced, making the approach more resistant to variations in target concentration and contributions from nonspecific binding leading to lost base calls. Accordingly, within the present invention step (b) of the Affymetrix CustomSeq™ protocol is optional and may be omitted.
Following DNA amplification, the DNA molecules obtained thereby are too long to hybridize with the short probes on the array surface. Accordingly, step (c) of the Affymetrix CustomSeq™ protocol involves fragmentation and subsequent labeling of the fragments with a fluorescent substrate. The method and reagents for fragmentation and labeling are not particularly limiting; however the label must be compatible with the detection apparatus for the resequencing microarray. To this end, the manufacturer recommended reagents and conditions may be employed.
Alternative Variations of the Affymetrix Protocols
Fluorescent labels that may serve to be advantageous for the methods described herein, as these are routinely used with automated instrumentation for simultaneous high throughput analysis of multiple samples, include the Cy fluorophores, the rhodamine based fluorophores: TARAM, ROX, JOE, and FAM; the BigDye™ fluorophores (Applied Biosystems, Inc.), the dansyl group, fluorescein and substituted fluorescein derivatives, acridine derivatives, coumarin derivatives, pthalocyanines, tetramethylrhodamine, Texas Red™, 9-(carboxyethyl)-3-hydroxy-6-oxo-6H-xanthenes, DABCYL™, BODIPY™, and ALEXA™ fluorophores (Molecular Probes, Eugene, Oreg.)
Additionally, there are a variety of labels other than fluorophores that will be suitable and perhaps preferable for a variety of situations. These labels include, but are not limited to: resonance light scattering (RLS) particles (In Vitrogen, Carlsbad, Calif.), quantum dots (Quantum Dot Corp.) and other nanoscale particles having desirable optical qualities.
The target hybridization (step (d)) may be performed as described in the Affymetrix CustomSeq™ protocol. The highlight of this step is that the sample containing the fragmented and labeled DNA is denatured by a high temperature (e.g., 85-100° C., preferably 95° C.) incubation followed by a hybridization temperature (e.g., 45° C.) equilibration. Once the DNA-containing sample has equilibrated the sample is applied to the resequencing array. The manufacturer recommends conducting the hybridization reaction for 16 hours; however, as stated above, the method of the present invention does not have its hands tied to enhance sensitivity to make the appropriate base calls. Therefore, shorter incubation times are appropriate for target hybridization. Within the context of the present invention, the target hybridization incubation time may range from a short time of 15 minutes to a long of 24 hours. Clearly, it is contemplated that this range of times embodies each of the intermediate times as if they were explicitly stated. Preferable times worth noting are 15 minutes, 30 minutes, 1 hour, 2 hours, 4 hours, 12, hours, and 16 hours.
The last step of the Affymetrix CustomSeq™ protocol entails washing the hybridized array using the Affymetrix Fluidics Station and scanned using the Agilent GeneArray™ Scanner. This instrument simply automates what would otherwise be manually performable labeling and rinse steps. Thus, any instrument that would be capable of delivering and withdrawing milliliter quantities of labeling and rinse media on a timed basis would be a suitable alternative. The invention described herein will be amenable to any subsequent hardware variations offered by Affymetrix. In addition, data acquisition from the types of resequencing microarrays described herein may be obtained from any manufacturer of equipment for microarray processing.
Bioinformatics Issues Concerning Pathogen Detection
Depending on the endpoints used for microarray-based detection of pathogens, the emphasis of bioinformatics issues is very different. Bioinformatic tools are indispensable for the efficient design and selection of specific complementary nucleic acid probe sequences for microarray development. For example, target pathogen genomic nucleic acid sequences are often amplified prior to microarray analysis and bioinformatics clearly has a role in the design of primers (assessing Tm/Ta, secondary structure, self-complementarity, and specificity issues) for assaying genes considered specific to an organism and strain (Kampke, Kieninger & Mecklenburg, 2001). These same assessments must also be made for microarray probe design.
During the initial stages of experimental design, it is assumed that primers and probes to genetic signatures associated with a target pathogen are specific to that pathogen or family of pathogens. It would follow that the generation of an amplicon or positive hybridization reaction using specifically designed primers or probe, respectively, would indicate the detection of the designated molecular trait from the target pathogen; however, this is not necessarily true. Bacterial and viral ‘genetic promiscuity’, the propensity of microorganisms to exchange genetic material, creates difficulties in developing single species or strain specific probes (Ochman, Lawrence & Groisman, 2000). Thus, preferred primer and probe design methodologies require the use of bioinformatic tools to: (a) perform multiple sequence alignments between different organisms or strains and design appropriate primers with the appropriate biochemical properties, (b) compare these sequences with those deposited in sequence databases to determine the present uniqueness of particular sequences and the potential for cross-reactivity, and (c) infer the probability of target specificity based on the level of genetic conservation and evolutionary relatedness with other pathogenic and non-pathogenic species whose primary genetic sequence has not yet been elucidated.
A very important bioinformatics aspect of the disclosed invention involves the assembly, annotation and selection of pathogen diagnostic targets into database(s) for incorporation into microarray design, as well as the concomitant task of relating detection events on the microarray to such database(s). An advantage of the present invention is that the information contained in the publicly available databases is ever increasing, thus further adding to the robust nature of the present invention. The present invention describes a process of manually selecting pathogen target sequences from the published literature (e.g. GenBank) and/or ascertaining an empirically determined diagnostic target sequence from published literature. The described approach has the advantage that a consortium of scientists, each possessing sufficient “domain expertise” for each of a large number of disparate pathogen species, can provide relevant, pathogen diagnostic sequence information that can be incorporated into an automated array design process without specific regard to specific probe, reagent, amplification, and sample preparation methods.
In one very preferred embodiment, the requisite domain expertise for each of a large number of unrelated pathogens will be maintained in an up-to-date fashion through a web-portal enabled database. Thus, an extended consortium, comprised of individual researchers of specific pathogens, would be able to provide the latest annotated target sequence information via a “pathogen page” formatted web portal, analogous to the “molecular page” model adopted by the Alliance for Cellular Signaling (AfCS). The AfCS database then maintains an otherwise incomprehensible amount of specific information on thousands of molecules involved in intracellular signaling cascades. In this format, individual researchers without specific knowledge about individual signaling molecules can access detailed parameters that can be used in numerical simulations of signaling events. Thus, in another very preferable embodiment, the annotated target sequence data for individual pathogens is organized into an automated data pipeline in which will impose user-defined design constraints (e.g. number of probe features, number of pathogen targets, the levels of sensitivity and specificity required for array performance, etc.) upon the total information content of a pathogen database, allowing automated, optimal target selection and submission of those targets to a vendor in a format necessary for microarray fabrication.
In yet another very preferred embodiment, the selected target sequences determined by the previous process will be correlated with the data that is collected in actual use of the microarray, such that metrics for probability and quality can readily used for decision-making. Two preferable approaches for performing such automated pipelining of data and algorithms are VIBE (Visual Integrated Bioinformatics Environment) software (Incogen, Inc., Williamsburg, Va.) and iNquiry (BioTeam, Boston, Mass.) which are representative of a class of integrated bioinformatics environments that could be used to equal effect for the intended purpose.
Data Acquisition
Raw sequence data from the resequencing microarray chips is provided by the Genetic Data Analysis Software version 2.0 (GDAS) packaged with the microarray reader from Affymetrix.
The Affymetrix resequencing array contains a defined number of probe cells or features. During scanning, the software divides each feature into subunit squares or pixels (3×3 μm). Each feature contains many copies of a unique 25-base oligonucleotide probe of defined sequence, while a series of eight features query a specific site in a known reference sequence. Four features interrogate the sense strand and contain probes that are identical except for the central base which is A, C, G, or T and four features interrogate the anti-sense strand and contain probes that are identical except for the central base which is A, C, G, or T.
GDAS uses the cell intensity data to make base calls for every base position represented on the resequencing array. Under the manufacturer setting for GDAS, the algorithm uses the intensity data from multiple samples to improve its calling accuracy and assigns a quality score for each call.
GDAS base calling is based on a previously described base-calling algorithm, ABACUS, detailed in (Cutler et al., 2001)). The model assumes that the pixel intensities of a feature are independently and normally distributed. The algorithm computes the estimated mean background and variance for the sense and anti-sense strand features. The base-calling algorithm also specifies models for the presence or absence of various genotypes in the sample (haploid or diploid). A variety of base calling algorithm parameters can be defined by the user (GDAS operator's manual/user's guide, Affymetrix) to obtain a trade-off between base calling percentage and accuracy.
Additional information regarding the GDAS algorithm and the parameters that can be modified is available in the GDAS user manual, which is incorporated herein by reference in its entirety. A description of the parameters is found in the GDAS version 2.0 manual on pages 207-217. The recommended (default) settings for GDAS are “conservative” settings that focus on the highest level of accuracy. In contrast, the objective of the present invention is to increase the percentage of base calls. To achieve this objective, the present inventors adjusted the parameters to allow highly permissive base calls (increased percentage) as listed below:
“Permissive” Base Calling Algorithm Settings”
Filter Conditions
Algorithm Parameters
Final Reliability Rules
The settings above are significant in the present application because the base call algorithm is set up by default to sacrifice the number of base calls made in order to make the most accurate calls (i.e., for SNP detection). In the present application, the technique is less concerned about achieving the same degree of accuracy as required for SNP detection but instead expanding the number of calls made so that the longest possible stretches of contiguous sequence are produced by GDAS while maintaining necessary specificity.
It is to be understood that within the scope of the present invention, the above-listed permissive settings can be altered individually or in toto as desired by the practitioner to obtain an optimal sensitivity/specificity agreement. In addition, it is to be understood that the settings above are exemplary and that each setting may be altered by 10% or more (parameter dependent) without altering the desired result of the present invention.
Resequencing Pathogen Identifier (REPI) and Alternatives, Modifications, Developments
Also according to the invention, it is the sequence information derived from a base-calling algorithm, as applied to the microarray hybridization pattern that is used to identify individual pathogens. Preferably, the sequence of target sequences determined by the resequencing probes is used to query a database using a similarity search algorithm. More preferably, the algorithm uses commonly used local alignment (e.g. Smith-Waterman, BLASTN) sequence alignment algorithms to statistically determine the probability that a given target sequence corresponds to a specific sequence in a database record (Korf, Yandell & Bedell, 2003). Even more preferably, a custom algorithm that determines subsequences that are most suitable for producing meaningful similarity searches against database records determines the set(s) of sequences that are submitted for similarity search automatically. Yet even more preferably, the automated subsequence-parsing algorithm is the Resequencing Pathogen Identifier (REPI) algorithm described in this invention and the sequence database records will be in both the public (e.g. GenBank) and private domain. Variants of nucleic acid sequence similarity search algorithms that are suitable for use in the intended invention include, but are not limited to: Washington University BLAST (WU-BLAST), NCBI-BLAST, FastA, MPsrch, Scanps, and BestFit (Korf et al., 2003).
REPI Alternatives and Variants
In the described invention, REPI (Resequencing Pathogen Identifier) software (see U.S. Application Ser. No. 60/609,918 filed on Sep. 15, 2004, and U.S. Application Ser. No. 60/631,460, filed on Nov. 29, 2004, which are incorporated herein by reference in their entirety) was used to ascertain which base call subsequences from the CustomSeq/GCOS/GDAS process would likely return significant BLAST results through the use of a customized sliding window algorithm. Subsequently, REPI automatically returns BLAST outputs to the end user that allow probabilistic assignments to the likelihood that a given set of base calls correspond to a particular microbial sequence. This low-level software functionality is analogous to the “kernel” of UNIX or UNIX-derived computer operating system, in that all higher-level functions and user interfaces must pass though it for resequencing chip analysis.
The low-level functionality provided by REPI will be central to a number of higher bioinformatics tasks that will utilize discontinuous segments of nucleic acid, or even amino acid sequence. In the following examples, the present inventors provide data showing that sequence fragments can be linked automatically to individual pathogens. In several more preferred embodiments, this approach can be refined to better discriminate between mixtures of pathogens and genetic recombination between pathogens. In one very preferred embodiment, the analysis software would allow for automatic detection of overlapping or homologous sequence fragments on different tiled regions of the array, allowing inference of a mixture of pathogens. In an even more refined embodiment, the analysis software would determine that the sequence outputs from different tiled regions are not overlapping but correspond to contiguous sequence that may be used to infer a genetic recombination event.
For example, a co-infection of two strains of a virus may produce a recombinant with a gene that is homologous with one virus strain except for the 5′ end, which has been substituted with the corresponding section of gene for the other virus strain. When this new recombinant virus genome is hybridized on a resequencing microarray, it produces signal from the corresponding pieces of both regions. One would need to have an assembly algorithm to construct a “model” of the pathogen showing which parts might fit together to form an entire target. If the two have significant overlap, one might conclude that there is probably a mixture. But if there were no overlap, there would remain a possibility that there is a recombinant. The degree of overlap (or lack of) could be affected by low concentrations of target with correspondingly smaller amounts of the tiles being filled in. This same principle can be applied even more readily, and with greater impact, on viruses where the recombination is a steady and recurring event, as in the case of influenza, where recombinations between viral segments result regularly in the formation of new viral strains. In fact, this described functionality in REPI will be essential for the distinction of pathogen mixtures versus recombination.
In another very preferred embodiment, REPI algorithms will allow for the analysis of transcriptional markers (e.g. RNA) that have been resequenced using the presently described type of microarray (via hybridization of RNA or complementary cDNA). In a method analogous to that described above for inference of genomic recombination events, transcriptional sequences may also be assembled to determine pathogen viability and transcriptional editing events that can serve as markers for infection.
Another REPI Alternative (Estimating the Amount of Pathogen Target in a Sample)
Not only is the present inventive approach able to distinguish between mixtures of pathogens and recombination events within a given pathogen (described elsewhere herein) it would also be of great value to provide the end user with an estimate (quantitation) of the relative amount of pathogen that was detected in the resequencing microarray assay. In particular, this would be of great utility when the clinician (technician) attempts to assign cause and effect when multiple pathogen genomic signatures are detected.
Two types of data may be used for this purpose. The first is the absolute intensity of the hybridization signals on the chip. A non-linear relationship exists between the amount of target in solution and the amount that actually hybridizes and the resulting signal. However, an estimate of the amount of target nucleic acid in the sample could be made by comparison with a standard curve prepared under control conditions. The signal intensity data is readily available from the .CEL file in the Affymetrix data hierarchy, and although the content of the .CEL files were not used in this disclosure, the output of REPI could easily be modified to include the intensity values of the .CEL files. Secondly, the percentage of base calls, both as a percentage of the total tile region size and as a percentage of base calls within a selected subsequence satisfying the sliding window algorithm, could be used as a measure of concentration. Our results show that both of these percentage metrics decrease with decreasing target concentration, although the correct pathogen can still be identified.
General Utility for Pathogen Detection
In a preferred embodiment, the invention described herein will be used for the routine diagnosis and surveillance of common respiratory pathogens in a clinical setting (at or near point-of-care). Readily obtainable samples (e.g. nasal wash, nasal swab, throat swab, sputum, or blood) will be processed in a simple manner to produce nucleic acid isolates that are obtained using an adsorptive process, enriched for pathogen-specific targets, amplified using a non-biased (e.g. total) amplification method or multiplexed PCR method, and hybridized on the resequencing microarray for a minimal amount of time prior to washing and imaging. The overall process will be sufficiently simple such that a skilled technician (medical technologist level) will be able to perform the assay without a significant interruption in their routine work pattern. Base calls will be made using the custom algorithms or using the steps specified by the vendor. REPI, or some variant thereof, will be used to automatically parse the base calls made by the microarray, and provide the end-user (e.g., physician, health care provider, public health officer, or other decision-makers) with decision-quality information for management (e.g., diagnostic, treatment, prognostic and outbreak control/containment measures) of the infectious pathogen(s) that are causative of the disease symptoms and complications. This analysis would occur locally through the use of an embedded sequence database that would be queried by REPI (e.g. local dedicated BLAST server). In addition to providing a routine diagnostic functionality, the microarray will also carry markers for highly improbable (i.e. bioterrorism) pathogens that would be cause for involvement of others, namely public health officials. However, it is understood that a nasal wash or throat swab may not be the optimal sample type for diagnosis of bioterrorism agents and that a separate sample type may be needed.
Also within the scope of the present invention, which further demonstrate the utility of the microarrays and methods of the present invention, include:
Scenario 1:
Patient arrives to medical facility with T>100.5 and respiratory symptoms. Nasal wash and/or throat swab are taken. Pathogens which can be identified by this route include the commonly occurring pathogens including those listed in Table 1. The presence of fever has been found to be an important criteria for isolating respiratory pathogens by culture and literature demonstrates that pathogens are typically present at peak titer during febrile periods.
For the bioterrorism agents, little information is available in the literature on infectious titers in respiratory secretions after intentional release of a biological agent. It is suspected that an aerosol release of an agent of bioterrorism would lead to detectable titers of organism in the first 24 hours post-exposure. In individuals presenting after the first 24 hours post-exposure, the microarray would serve the purpose of identifying common pathogens that might be otherwise erroneously suspected of being cases from the BT/BW agent. A proportion of individuals exposed to a significant aerosol release will develop symptoms rapidly and will retain the BT/BW agent in the nares for detection. For a disease, such as smallpox, the incubation period for onset of symptoms is much longer, but the virus can be isolated from throat culture for days.
Scenario 2:
The individual has been ill for days and did not initially present to a medical clinic, but chose to self-medicate. Patient may not have initially noted fever, but now has fever and respiratory symptoms and is concerned about lack of resolution. Health Care Practitioner (HCP) sees patient and determines appropriate clinical sample. If patient appears more ill, then chest X-ray may be acquired. Common organisms causing lower respiratory tract infection include, but are not limited to: Mycoplasma pneumoniae, S. pneumoniae, C. pneumoniae, L. pneumophila, S. pyogenes, Influenza A/B, RSV, Parainfluenza, S. aureus, SARS. Amongst the BT/BW agents of concern, an abnormal chest x-ray can be found with B. anthracis, Smallpox, Y. pestis, F. tularensis. For the common respiratory organisms, nasal wash/throat swab can still be appropriate sample techniques, although some practitioners will choose to send a sputum samples. For the BT/BW agents, sputum and blood can be appropriate sample techniques.
Scenario 3:
Ongoing outbreak of a known agent. The microarray can continue to be useful for detecting cases of other common pathogens, but may also be used to screen for known BT/BW agent using any type of appropriate specimen. Other assays may prove less expensive, but the microarray can provide forensic information and antibiotic resistance data at the time of pathogen identification.
Overall Business Model
The resequencing DNA microarray and associated devices (to include other types of lower content microarrays or alternative sequence detectors) will be but one class of elements of an integrated pathogen diagnostic/surveillance system. This system will be comprised of diagnostic, informatic, and epidemiologic components. At the diagnostic level, the RPM (and ancillary devices) will provide a rapid and cost-effective methodology for providing a diagnosis, patient-specific treatment information, and prognostic information (based on virulence and resistance markers) of infectious respiratory disease. This will represent a shift in diagnostic emphasis for the provider from clinical suspicion of one or a few pathogens (e.g., Mycoplasma pneumoniae and other pathogens with a presentation of “walking pneumonia” would be tested by obtaining bacterial cultures, presence of antibodies to that pathogen, etc.) to symptom-based and broad differential testing by multiplexed DNA microarrays based on the symptom(s) and the organ system or systems effected. The single highly multiplexed DNA arrays will additionally provide diagnostic information on which group of pathogens to treat (e.g., bacteria, viruses, parasitic, and fungal pathogens) producing substantial return on investment by reducing the number of prescriptions based on suspicion of causal agents (e.g., prescribing antibiotics for a viral respiratory infection will have no effect). Additional diagnostic applications can be created to cover all pathogens effecting an organ or organ systems (e.g., respiratory, genitourinary, central nervous system, etc.), syndromes (e.g., fever of unknown origin evaluated in the hospital for two weeks without identifying the cause to rule out all potential infectious disease causes), symptoms (e.g., all pathogens causing a urticaria or rash that itches, or cough, etc.), around collections of pathogens or threats present in geographic locations, or functional categories (e.g., groupings based on all the organisms known to have antimicrobial resistance in hospital or other confined settings like prisons or nursing homes). This highly multiplexed differential diagnostic testing will be done by determining the causative agent(s) of infection, antimicrobial susceptibility of the agent(s), and genomic sequence-specific clues as to the nature of the agent such that the maximum number of opportunities is identified to increase the efficacy and efficiency of clinical management.
However, because of the high number of specific molecular queries that are being made, a large number of pathogens can be detected, each at previously unattainable levels of detail. As such, a large number of assays (culture, serotyping, and PCR confirmation) that are subsequently and infrequently performed as part of “surveillance” activities are now being performed simultaneously with diagnosis, obviating the need for burdensome and expensive tasks currently required at the clinical level for surveillance support. This would have great relevance to diagnosis and surveillance of variants of RNA viruses (e.g. Influenza and SARS) in a manner that would be prohibitively complicated when relying on specific oligonucleotide probe sets.
In the specific embodiment described here, it will be highly advantageous to have diagnostic microarrays, the fabrication of which will not rely on the availability of a large number of target sequences and a means to fabricate arrays using them. More importantly, it will be critical not to be restricted to the assumption that the target sequences are invariant. Diagnostic DNA microarrays will identify specific but unanticipated genomic variants of a model pathogen, without requiring re-design of specific oligonucleotide probes and array re-fabrication. This will be critical to the ability to characterize the cause(s) of infectious disease outbreaks in a time-effective manner. For example, such microarrays could be used to rapidly detect new variations of influenza or SARS virus without requiring that the pathogens be isolated, cultured, and sequenced using conventional approaches; a process that would require weeks to months if the pathogen were readily cultivable.
The present invention also embraces applications for detecting a mixture of pathogens, especially when there is no preliminary evidence to suggest that the mixture might be interrogated by specific reagents (e.g. PCR primers). Thus, the present invention provides a means to impact the ability to determine the complementary roles of interacting pathogens in disease etiology. However, in applications such as viral or bacterial stock quality control and assessment of viral vaccine production, which involves the intentional mixture of field strain and cultivable viruses to produce recombinants that culture well. Thereby, the present invention enables the presentation of the correct the appropriate epitope(s) for vaccine efficacy.
The informatics component of the system will provide the necessary components to allow local (point-of-care), automatic microarray data analysis as well as coordinate multi-directional information transfer. “Upward” flow of information will entail the transfer of specific sequence base calls from the resequencing chip, preferably in FASTA format, and all associated local processing results, to local, regional, national and international levels. “Lateral” flow of information will involve the exchange of specific sequence base calls and associated local processing results to other local point-of-care medical facilities. “Downward” flow of information is defined as a provision of national level data integration to regional and local health officials.
Like clinical samples, environmental samples may contain small amounts of target nucleic acids in a high genomic background of unknown origin. But unlike clinical samples of a given type, the background found in an environmental sample (e.g. soil, water, or collected from an aerosol particle collector) might show a more heterogeneous composition depending on the geographical location, season, and environmental conditions. Accordingly, the aforementioned amplification, enrichment and/or subtraction strategies may be employed to obtain reliable base calling.
Forensic and Environmental Applications
The amount of detailed sequence information provided by the RPM will be enabling for various applications other than medical diagnosis and surveillance. Thus, the capability of the apparatus extends to forensic fingerprinting of specific pathogen strains. The capability enables preemptive diagnosis of etiologies of infectious disease, as alternative to conventional practice of corroborative diagnostic analysis. In the case of an intentional infection, poisoning, or bioterrorism event, the resequencing pathogen detection microarray would allow for detailed strain identification that could be used to determine the possible originator of the event and to allow for rapid mitigation of the event (e.g. determination of infectious capacity, antimicrobial resistance, or engineered modifications to an organism) by implementing targeted public health containment measures directed by having the pathogen identified to the strain as the completed first step in the outbreak investigation process.
The present invention further suggests methods and processes to automate and optimize the iterative and adaptive design, fabrication and validation of arrays, including derivative sub-arrays. In a very preferable embodiment, an enterprise level, a consortium of experts on individual pathogens would maintain a web portal-enabled database. The consortium would maintain pathogen target sequences for identification and virulence.
The same technology described in the present invention can be used for non-clinical samples, including those collected from air, water, soil or surface swabs. The only modifications necessary to those described in the present invention will be those necessary for nucleic acid extraction and background nucleic acid removal, if a subtractive approach followed by generic amplification is a desired approach.
Multiple Pathogen Surveillance in a Population
The invention further provides a specific implementation that validates its capabilities in a real world operational setting. This implementation relates to epidemic outbreaks of acute respiratory disease involving common and less common etiologies, while simultaneously evaluating the possible presentation of hostile pathogenic agents (rare if ever) in individual encounters. Thus, the invention provides a means to perform near real time surveillance of a plurality of infectious pathogens involved in an infectious outbreak. Such surveillance may be validated and eventually become operational in a “real world testbed”. In a preferred embodiment, the real world testbed is a human population that regularly encounters a variety of respiratory pathogens. In a preferred embodiment, the population is comprised of military personnel at an installation or base. In a very preferred embodiment, the population is comprised of active duty military personnel.
End User-Specified Applications
The integrated process of microarray design and assay will require only that sequences be provided to a microarray manufacturer and not involve the design of ancillary reagents (e.g. for specific PCR). Thus, an end user who has no detailed knowledge of genomics or would be capable of defining qualitative attributes of a microarray-based assay, and an automated bioinformatics pipeline would be used to select suitable target gene subsequences for submission to a resequencing microarray manufacturer. This would allow rapid deployment of a new micorarray design for a specific geographic location, theater of operations. Thus, the integrated design/analysis capability that is enabled by the present invention will generalize to other envisioned applications besides those listed herein.
Adenovirus Sequences
In an additional embodiment of the present invention are the genomic sequences of thirteen adenovirus strains, which were not known as of the date of the present invention. The thirteen adenovirus strains are: Ad3, Ad3FS_navy, Ad4, Ad4vaccine, Ad4FS_navy, Ad4FS_AF, Ad5FS, Ad7, Ad7FS_navy, Ad7 vaccine, Ad16, Ad1, and Ad21. These genomic sequences have been assigned the GenBank accession numbers shown in Table 6 appearing in the Examples. The full GenBank records, including partial annotation, for each of these sequences are found in the Sequence Listing attached herewith.
A “polypeptide” as used herein is understood to mean a sequence of several amino acid residues linked by peptide bonds. Such amino acids are known in the art and encompass the unmodified and modified amino acids. In addition, one or more modifications known in the art such as glycosylation, phosphorylation, etc may modify the polypeptide.
The term “isolated” means separated from its natural environment. This term is intended to also embrace the terms “purified” (100% pure) and “substantially purified” (at least 90% pure).
The term “polynucleotide” refers in general to polyribonucleotides and polydeoxyribonucleotides, and can denote an unmodified RNA or DNA or a modified RNA or DNA.
The term “homologous” as used herein is understood to mean sequence similarity between two or more polynucleotides or proteins from the same species or from a different species. Within the meaning of this term, said two or more polynucleotides (or proteins) are homologous when at least 70%, preferably at least 80%, most preferably at least 90% of the nucleotide base (amino acid) composition of a candidate sequence corresponds to the sequence according to the invention. According to the invention, a “homologous protein” is to be understood to retain at least 50%, preferably at least 75%, more preferably at least 85%, most preferably at least 95%, of the activity of the activity of the sequence of the present invention. As used herein “corresponds” is to be understood to mean that the corresponding amino acids are either identical or are mutually homologous amino acids. The expression “homologous amino acids” denotes those that have corresponding properties, particularly with regard to their charge, hydrophobic character, steric properties, etc. The same terminology can be used to describe DNA or RNA sequence homology for gene sequences that encode the corresponding proteins.
The term “homologous fragment” as used herein is understood to mean two or more polynucleotides or proteins from the same species or from a different species. In this context, it is contemplated that a fragment is homologous when it shares at least 40% identity to a fragment having at least 50 amino acids. Preferably, homologous fragments share at least 50% identity to a fragment having at least 50 amino acids. More preferably, homologous fragments share at least 60% identity, at least 70% identity, at least 80% identity, at least 90% identity, or at least 95% identity to a fragment having at least 50 amino acids. Accordingly, homologous fragments are included within the scope of the present invention. For homologous polynucleotides, it is understood that that the same homology ranges are envisioned in the present inventions but over a range of up to 1000 nucleotides, inclusive of all integers (i.e., 150, 250, 300, 500, 750, etc.).
Homology, sequence similarity or sequence identity of nucleotide or amino acid sequences may be determined conventionally by using known software or computer programs such as the BestFit or Gap pairwise comparison programs (GCG Wisconsin Package, Genetics Computer Group, 575 Science Drive, Madison, Wis. 53711). BestFit uses the local homology algorithm of Smith and Waterman, Advances in Applied Mathematics 2: 482-489 (1981), to find the best segment of identity or similarity between two sequences. Gap performs global alignments: all of one sequence with all of another similar sequence using the method of Needleman and Wunsch, J. Mol. Biol. 48:443453 (1970). When using a sequence alignment program such as BestFit, to determine the degree of sequence homology, similarity or identity, the default setting may be used, or an appropriate scoring matrix may be selected to optimize identity, similarity or homology scores. Similarly, when using a program such as BestFit to determine sequence identity, similarity or homology between two different amino acid sequences, the default settings may be used, or an appropriate scoring matrix, such as blosum45 or blosum80, may be selected to optimize identity, similarity or homology scores.
The present invention also relates to polynucleotides that contain complete genes that are identified by open reading frames. Examples of the preferred genes embraced by the present invention include E1A, hexon, and fiber. The sequences of E1A, hexon, and fiber, as well as other preferred polynucleotide sequences within the scope of the present invention are found in the attached Sequence Listing. The present invention also embraces fragments of said genes and polynucleotides and fragments thereof that can be obtained by screening by means of the hybridization of a corresponding gene bank with a probe which contains the sequence of said polynucleotide or a fragment thereof, and isolation of said DNA sequence.
The present invention also relates to coding DNA sequences that result from degeneration of the genetic code. Moreover, one skilled in the art is also aware of conservative amino acid replacements such as the replacement of glycine by alanine or of aspartic acid by glutamic acid in proteins as “sense mutations” that do not result in any fundamental change in the activity of the protein, i.e. which are functionally neutral. It is also known that changes at the N- and/or C-terminus of a protein do not substantially impair the function thereof, and may even stabilize said function.
Polynucleotide sequences according to the invention are suitable as hybridization probes for RNA, cDNA and DNA, in order to isolate those cDNAs or genes, which exhibit a high degree of similarity to the probe sequence.
Polynucleotide sequences according to the invention are also suitable as primers for polymerase chain reaction (PCR) for the production of DNA that encodes an active enzyme.
Oligonucleotides such as these, which serve as probes or primers, can contain more than 30, preferably up to 30, more preferably up to 20, even more preferably at least 15, and most preferably at least 13 successive nucleotides. Oligonucleotides with a length of at least 40 or 50 nucleotides are also suitable.
Hybridization protocols are known in the art and are disclosed, for example, in Sambrook et al., Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory, N.Y. (1989). However, as used herein, stringent hybridization conditions are those conditions which allow hybridization between polynucleotides that are 75%, 80%, 85%, 90%, 95%, or 98% homologous as determined using conventional homology programs, an example of which is UWGCG sequence analysis program available from the University of Wisconsin (Devereux, Haeberli & Smithies, 1984). Typically, stringent conditions will be those in which the salt concentration is less than about 1.5 M Na ion, typically about 0.01 to 1.0 M Na ion concentration (or other salts) at pH 7.0 to 8.3 and the temperature is at least about 30° C. for short probes (e.g., 10 to 50 nucleotides) and at least about 60° C. for long probes (e.g., greater than 50 nucleotides). Stringent conditions may also be achieved with the addition of destabilizing agents such as formamide. Exemplary low stringency conditions include hybridization with a buffer solution of 30 to 35% formamide, 1 M NaCl, 1% SDS (sodium dodecyl sulphate) at 37° C., and a wash in 1× to 2×SSC (20×SSC=3.0 M NaCl/0.3 M trisodium citrate) at 50 to 55° C. Exemplary moderate stringency conditions include hybridization in 40 to 45% formamide, 1 M NaCl, 1% SDS at 37° C., and a wash in 0.5× to 1×SSC at 55 to 60° C. Exemplary high stringency conditions include hybridization in 50% formamide, 1 M NaCl, 1% SDS at 37° C., and a wash in 0.1×SSC at 60 to 65° C.
Specificity is typically the function of post-hybridization washes, the critical factors being the ionic strength and temperature of the final wash solution. For DNA—DNA hybrids, the Tm can be approximated from the equation of Meinkoth and Wahl, (Meinkoth & Wahl, 1984): Tm=81.5° C.+16.6 (log M)+0.41 (% GC)−0.61 (% form)−500/L; where M is the molarity of monovalent cations, % GC is the percentage of guanosine and cytosine nucleotides in the DNA, % form is the percentage of formamide in the hybridization solution, and L is the length of the hybrid in base pairs. The Tm is the temperature (under defined ionic strength and pH) at which 50% of a complementary target sequence hybridizes to a perfectly matched probe. Tm is reduced by about 1° C. for each 1% of mismatching; thus, Tm, hybridization and/or wash conditions can be adjusted to hybridize to sequences of the desired identity. For example, if sequences with approximately 90% identity are sought, the Tm can be decreased 10° C. Generally, stringent conditions are selected to be about 5° C. lower than the thermal melting point (Tm) for the specific sequence and its complement at a defined ionic strength and pH. However, severely stringent conditions can utilize a hybridization and/or wash at 1, 2, 3, or 4° C. lower than the thermal melting point (Tm); moderately stringent conditions can utilize a hybridization and/or wash at 6, 7, 8, 9, or 10° C. lower than the thermal melting point (Tm); low stringency conditions can utilize a hybridization and/or wash at 11, 12, 13, 14, 15, or 20° C. lower than the thermal melting point (Tm). Using the equation, hybridization and wash compositions, and desired Tm, those of ordinary skill will understand that variations in the stringency of hybridization and/or wash solutions are inherently described. If the desired degree of mismatching results in a Tm of less than 45° C. (aqueous solution) or 32° C. (formamide solution) it is preferred to increase the SSC concentration so that a higher temperature can be used. An extensive guide to the hybridization of nucleic acids is found in Current Protocols in Molecular Biology, Chapter 2, Ausubel, et al., Eds., Greene Publishing and Wiley-Interscience, New York (2000).
In this invention “primer” or “probe” means a polynucleotide, especially an oligonucleotide, that is produced synthetically or biologically and includes a specific nucleotide sequence and permits hybridization to a section containing the target nucleotide sequence.
Defined primers or probes, as well as all other oligonucleotides and polynucleotide of the present invention, may be produced by any of several well-known methods, including automated solid-phase chemical synthesis using cyanoethyl-phosphoramidite precursors. Other well-known methods for construction of synthetic primers/oligonucleotides may, of course, be employed. J. Sambrook, E. F. Fritsch and T. Maniatis, Molecular Cloning 11 (2d ed. 1989).
The primers used to amplify the sample nucleic acids may be coupled to a detectable moiety. A preferred example of such a detectable moiety is fluorescein, which is a standard label used in nucleic acid sequencing systems using laser light as a detection system. Other detectable labels can also be employed, however, including other fluorophores, radio labels, chemical couplers such as biotin which can be detected with streptavidin-linked enzymes, and epitope tags such as digoxigenin detected using antibodies. The primers may be modified whereby another nucleotide is added to, removed from, or substituted for at least one nucleotide in the oligonucleotide. Introduction of known labels such as radioactive substances, enzymes, fluorescence substances, etc. after synthesis of oligonucleotide is also included therein.
Similarly, the probes/oligonucleotides used to hybridize with the polynucleotides coding for the polypeptides of the invention, for example for the purpose of detection of such a polynucleotide, may be coupled to a detectable moiety.
As used herein, the term “enhancement” means increasing the intracellular activity of one or more enzymes in a plant cell and/or plant that are encoded by the corresponding DNA. Enhancement can be achieved with the aid of various manipulations of the bacterial cell. In order to achieve enhancement, particularly over-expression, the number of copies of the corresponding gene can be increased, a strong promoter can be used, or the promoter- and regulation region or the ribosome binding site which is situated upstream of the structural gene can be mutated. Expression cassettes that are incorporated upstream of the structural gene act in the same manner. In addition, it is possible to increase expression by employing inducible promoters. A gene can also be used which encodes a corresponding enzyme with a high activity. Expression can also be improved by measures for extending the life of the mRNA. Furthermore, preventing the degradation of the enzyme increases enzyme activity as a whole. Moreover, these measures can optionally be combined in any desired manner.
A gene can also be used that encodes a corresponding or variant enzyme with a high activity. Preferably the corresponding enzyme has a greater activity than the native form of the enzyme, more preferably at least in the range of 5, 10, 25% or 50% more activity, most preferably more than twice the activity of the native enzyme.
The above written description of the invention provides a manner and process of making and using it such that any person skilled in this art is enabled to make and use the same, this enablement being provided in particular for the subject matter of the appended claims, which make up a part of the original description.
As used herein, the phrases “selected from the group consisting of,” “chosen from,” and the like include mixtures of the specified materials.
Where a numerical limit or range is stated herein, the endpoints are included. Also, all values and sub-ranges within a numerical limit or range are specifically included as if explicitly written out.
The above description is presented to enable a person skilled in the art to make and use the invention, and is provided in the context of a particular application and its requirements. Various modifications to the preferred embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments and applications without departing from the spirit and scope of the invention. Thus, this invention is not intended to be limited to the embodiments shown, but is to be accorded the widest scope consistent with the principles and features disclosed herein.
Having generally described this invention, a further understanding can be obtained by reference to certain specific examples, which are provided herein for purposes of illustration only, and are not intended to be limiting unless otherwise specified.
Materials and Methods
Table 6 (below) lists the adenovirus strains cited in the following examples. GenBank accession numbers refer to the genomic sequence numbers assigned to each strain. These sequences were not available to the public at the time of invention and form an embodiment of the present invention, as well as fragments of the same.
ATCC = American Type Culture Collection (Manassas, VA)
NHRC = Naval Health Research Center (San Diego, CA)
NIH = National Institutes of Health (Bethesda, MD)
The full GenBank records, including partial annotation, for each of these are found in the Sequence Listing attached herewith.
Influenza Content
The prototype influenza types represented by tiles on RPMV1 are:
Influenza A virus (A/New Caledonia/20/99) H1 HA gene for hemagglutinin 1
FluAHA3 Influenza A virus (A/Fujian/411/02) H3N2 gene for hemagglutinin 3
FluAHA5 Influenza A virus (A/HongKong/156/97/H5N1), hemagglutinin 5
FluANA1 Influenza (A/Chile/1/83), neuraminidase 1
FluANA2 Influenza A virus (A/Panama/2007/99/H3N2) NA gene for neurarninidase 2
FluAMATRIX Influenza virus A/NWS/33/H1N1 matrix protein (M)
FluBNA Influenza B virus (BlYamagata/16/88), neuraminidase glycoprotein genes
FluBHA Influenza B virus (B/Yamanashi/166/98) hemagglutinin 1 subunit (HA)
FluBMATRIX Influenza B virus (BlYamagata/16/88) M1 matrix protein (M)
The Accession numbers for each of the foregoing sequences, as well as the remaining pathogen target sequences, obtained from the public domain and personal communications, are listed in Table 9.
DNA sequences were provided to Affymetrix for creation of the resequencing microarray chip (RPM Version 1 chip) utilized in the following examples. Submission of the DNA sequence and instruction files to Affymetrix were in accordance with the manufacturer instructions CustomSeq™ Array Protocol and product literature. Probe lengths were nominally 25-nucleotides long and contained a variable (interrogation point) central nucleotide for each of four possible variants (A, C, T or G) in both the sense and antisense directions.
The target genes selected for the RPMV1 pathogens listed above are described in the version 1 layout shown in Table 8 and the Sequence Listing along with the respective PCR primers used for amplification of the same. The sequences submitted for tiling and chip fabrication were based on the Affymetrix instruction file summarized in Table 7, which corresponds to the sequences appearing as SEQ ID NOs: 1-58. The corresponding “instruction file” lists the alias designator (e.g. FluAHA5) for each tile region and provides a “FASTA” formatted target gene sequence (this can be all or part of the complete target gene).
Streptococcus
pneumoniae
Streptococcus
pneumoniae
Mycoplasma
pneumoniae
Neisseria
meningitidis
Neisseria
meningitidis
Bordetella pertussis
Bordetella pertussis
Chlamydia
pneumoniae
Chlamydia
pneumoniae
Chlamydia
pneumoniae
Streptococcus
pyogenes
Streptococcus
pyogenes
Streptococcus
pyogenes
Streptococcus
pyogenes
Bacillus anthracis
Bacillus anthracis
Bacillus anthracis
Francisella
tularensis
Francisella
tularensis
Yersinia pestis
Yersinia pestis
Arabisopsis thaliana
Arabisopsis thaliana
Streptococcus
pneumoniae
Streptococcus
pneumoniae
Mycoplasma
pneumoniae
Neisseria
meningitidis
Neisseria
meningitidis
Bordetella pertussis
Bordetella pertussis
Chlamydia
pneumoniae
Chlamydia
pneumoniae
Chlamydia
pneumoniae
Streptococcus
pyogenes
Streptococcus
pyogenes
Streptococcus
pyogenes
Streptococcus
pyogenes
Bacillus anthracis
Bacillus anthracis
Bacillus anthracis
Francisella
tularensis
Francisella
tularensis
Yersinia pestis
Yersinia pestis
Arabisopsis thaliana
Arabisopsis thaliana
The instruction file references the same alias designators, but further specifies the actual subsequences that are actually to be tiled onto the array. In most cases, this represents the entire sequence from the sequence file but in others (e.g. FluBMATRIX) only nucleotides 1-362 of the total sequence were used for tiling. Table 9 provides the information presented in the instructions file.
Table 9 (below): Instruction file (specific file) of the RPM V1 design specification. The alias designator is a name assigned to each discrete “tile” region (e.g. a geometric region of the microarray containing all probe combinations required for resequencing of a stretch of pathogen genome sequence). The pathogen, sequence accession number, and tile size for each tile region are shown. The instructions, which were processed in Affymetrix submission format, appear in Table 7. The instruction file references complete or partial sequence of the complete target genes that are found in SEQ ID NOs: 1-58 appearing in the attached Sequence Listing.
The chip design team at Affymetrix used the combination of the information above and the corresponding sequence file information to generate the layout of the chip.
It is important to note that the tiling strategy dictates that the first 12 and last 12 sequences from each discrete tile region that are not queried by the tiling strategy of the resequencing microassay, since they are used as components of the first and last 25-mer probes that are varied at the number 13 positions.
The sequences used for the adenovirus regions (Ad4, Ad5, Ad7 and Ad7 vaccine) of the chip were all derived from early drafts of genomes sequenced by the present inventors. The GenBank submission files that correspond to the genomes used for tiled regions are listed in Table 6. Because the sequences submitted to Affymetrix for the prototypes were based on early drafts of the genomes, there were discrepancies observed between those early sequences and the final sequences submitted to GenBank. A list of those discrepancies is given in the Table 10:
This relatively small number of discordances did not interfere with the ability of the chip to make base calls that could be associated with the correct organism, except in one specific case of adenovirus type 4, described in the Examples. Overall, the resequencing microarray technique of the present invention corroborated the accurate (refined) final sequence of the tiled genes with respect to base substitutions, validating the unanticipated robustness of our method.
Degenerate PCR Primers Design
The objective of primer selection to support conserved (degenerate) multiplex PCR is to design primers that target the conserved regions flanking species-specific variable regions of E1A, fiber, and hexon genes. In general, this method may be applied to any organism, as conserved sequences within a species are a ubiquitous in nature. These target genes were selected based on their function and location within the linear adenoviral genome. E1A is located at the 5′ end of the adenoviruses genome and encodes a trans-acting transcriptional regulatory factor that is necessary for transcriptional activation of early genes. The hexon and fiber genes, which are located in the middle and 3′ end of the adenovirus genome, encode antigenic determinants ε and γ respectively, which determine the viral serotype. Thus, detection and serotyping of ARD-causing adenoviruses can be effectuated by targeting the nucleic acid determinants that give rise to the serotype. Thereby, the primers provided specific amplification within the adenovirus while the variable regions supplied serotype-specific characters for proper species identification.
The primers used for conserved (degenerate) multiplex PCR in the following examples are based on a global alignment of E1A, fiber, and hexon gene sequences, respectively, available from GenBank (GenBank accession numbers are given in parentheses): E1A-AdB (NC—004001), AdC (NC—001405), Ad3 (AF492352), Ad4 (M14918), Ad7 (X03000);fiber-Ad2 (AJ278921), AdS (M18369), Ad3 (X01998), Ad4 (X76547), Ad7 (M23696), Ad16 (U06106), Ad2l (U06107); hexon-Ad3 (X76549), Ad4 (X84646), Ad6 (AF161560, X67710, Y17245), Ad7 (AF053087, X76551), Ad16 (X74662), Ad21 (AB053166). The global sequence alignment for primer design of the E1A gene used the E1A gene sequences from Ad3, Ad4, Ad7, Ad21, AdB, and AdC serotypes. The global sequence alignment for primer design of the fiber gene used fiber gene sequences from Ad2, Ad3, Ad4, AdS, Ad7, Ad16, and Ad21 serotypes. The global sequence alignment for primer design of the hexon gene used the hexon gene sequences from Ad3, Ad4, Ad6, Ad7, Ad16, and Ad21 serotypes.
Subsequent to the global sequence alignment, primer pairs were selected based on their ability to amplify E1A, fiber, and hexon genes of serotype 3, 4, 6, 7, 16 and 21 (data not shown). Table 11 shows the primer pairs employed in the following Examples.
Multiplex Degenerate Primer PCR Protocol
The primers pairs (Lin et al., 2004) were evaluated in various multiplex combinations to obtain amplification of adenovirus serotype 3, 4, 6, 7, 16, and 21. PCR was performed in 50 μl volumes containing 20 mM Tris-HCl (pH 8.4), 50 mM KCl, 200 μM each of dNTPs, 200 nM AdB1, AdB2 primers; 300 nM AdE1A-F, AdE1A-R primers; 400 nM Adfib-F3, Adfib-R3 primers; 500 nM Adhex-F7, Adhex-R5 primers, 2 U of Platinum Taq DNA polymerase (Invitrogen), and 106 copies of DNA templates. The amplification reaction was performed using a Peltier Thermal Cycler-PTC225 (MJ Research) with preliminary denaturation at 94° C. for 3 minutes, then followed-by 40 cycles of denaturation at 94° C. for 30 seconds, annealing at 50° C. for 30 seconds, the extension at 72° C. for 40 seconds, and a final extension at 72° C. for 10 minutes. The generic multiplex PCR assay could amplify all three genes of all six ARD causing adenoviruses even though the hexon gene of Ad4 showed very weak band on agarose gel.
Total Amplification Protocol
Total amplification was conducted using the commercially available GenomiPhi™ DNA Amplification Kit (Amersham Biosciences) to representatively amplify linear genomic DNA (hereinafter referred to as “GenomiPhi total amplification”). The amplification method employed in the GenomiPhi total amplification method utilizes bacteriophage Phi29 DNA polymerase enzyme to exponentially amplify single- or double-stranded linear DNA templates by strand displacement amplification. The starting sample was not quantified, but rather was used directly.
Raw sequence data from the resequencing microarray chips is provided by the Genetic Data Analysis Software version 2.0 (GDAS) packaged with the microarray reader from Affymetrix. GDAS base calling is based on a previously described base-calling algorithm (Cutler et al., 2001). Each of the FASTA output files containing the base calls obtained from the GDAS software was analyzed using specialized software (REPI) that the present inventors developed.
In the case of the present invention, the sequence output of GDAS is most often a scattered mixture of contiguous sequence calls (A, T, C or G) that are interspersed with varying amounts of no-calls (n's) where the GDAS software does not make a base call due to lack of amplification, weak hybridization signal on the chip and/or high background hybridization caused by non-specific binding (Cutler et al., 2001). An example output of the GDAS output for the Adenovirus 4 prototype sample for the Ad4FIBER tile region is shown below (SEQ ID NO: 429):
Similarity search algorithms such as BLAST (Korf et al., 2003) allow the use of n's (analogously to a wildcard) but the inclusion of too many n's results in an unacceptable degree of ambiguity. In that case, no meaningful BLAST results will be returned.
The REPI software was designed to parse the output of the FASTA file selecting and editing to mold sequence data into a format suitable for sequence similarity searches using the NCBI BLASTN algorithm. To accomplish this objective, REPI functions through a series of filters modifying the data as little as possible while extracting usable, “BLASTable” data from the FASTA files. Due to the nature of the resequencing microarray the sequences often contain large amounts of non-base calls (n's). BLAST is unable to return significant similarity for sequences with a large amount of non-base calls. Therefore the original sequences must be filtered to extract those portions that are most likely to return a significant similarity.
The first filter that a prospective sequence encounters is a control check. The control sequence incorporated into the microarray is specifically designed to be a nonsense sequence; therefore, it will never return a significant similarity. Next the sequence is evaluated for “BLASTable” data. Here a sliding window algorithm is used, the window-size parameter is entered by the user and represents the number of base calls the algorithm will evaluate at one time. The sequence is evaluated starting from the first base-call the window slides along the sequence searching for the first area containing relevant data, this is evaluated using a scoring method where all valid bases are given a score of one and all n's are given a score of zero if the score is greater than or equal to a predetermined threshold (here 25% is used) the program marks the start of this window as the beginning of usable data. Once the start of usable data is determined the program reverses its parameters and begins searching for the end of the useable data, the score must now be less than or equal to the threshold. For each sequence REPI searches for the largest continuous string(s) of usable data, which will represent that sequence's initial subsequence.
This subsequence is then trimmed for beginning and trailing n's. Trimming is necessary because the previous filter uses the starting position of the starting window as the beginning of the subsequence and the last position of the ending window as the end of the subsequence, therefore all though that window's score was acceptable there may be n's leading or trailing the subsequence. The next filter the subsequence goes through is a length evaluation. Subsequences longer than 50 nucleotides are allowed to continue, subsequences shorter than 20 nucleotides are discarded, and subsequences between 20 and 50 nucleotides are re-evaluated as follows. Due to the length of these subsequences they are rescored using the same scoring system described earlier. Subsequences with greater than 60% non-base calls are discarded; all others are allowed to continue to be searched with the BLAST algorithm against GenBank, or one's own modified database.
Once the similarity search is complete REPI computes a number of statistics on the subsequence including the subsequence percentage of the target sequence, the subsequence length, the number of subsequence base calls, and the percentage of subsequence base calls. The subsequence percentage of the target sequence and the subsequence length shows what portion of the target pathogen gene was identified. The subsequence length and percentage of subsequence base calls allow us to monitor the filtering algorithm filters as well as the GDAS threshold parameters. REPI saves all statistical results returned from the BLASTN algorithm allowing the user to manipulate which results are displayed in the graphical user interface.
In the examples provided, REPI was interfaced to a local BLAST (NCBI GenBank) database (contained on an Apple G5 single processor (1.8 Ghz) computer with 4.5 GB of random access memory) via a CGI (Perl) interface. Displayed results included all database sequences within an expect value (E-value) threshold of 1.0e-9. The E-value represents the number of alignments expected at random given the size of the search space, the scoring matrix, and the gap penalties; the lower the E-value the less likely the database sequence similarity matches was in fact a random identification. By definition e=2.71828182845904523536028747135.
The REPI output is comprised of the (BLASTable) subsequence names, lengths, E-values, and bits scores are displayed for each subsequence in descending order of bit scores. The name is reported as the GenBank record's FASTA definition line and includes the sequence length. The score is the normalized score computed from the scoring matrix and gap penalties, the higher the score the greater the similarity.
The REPI output of the example listed above is shown below. For each “BLASTable” subsequence, REPI returns (in descending order of bit score ranking) all GenBank data records having expect values of <1.0 e-9. The highest bit score is achieved for the adenovirus 4 prototype (AY594253), which is genetically indistinguishable from the Ad4 vaccine strain (AY594254) across this stretch, while lower bit scores suitably distinguish field strains from Air Force and Navy training sites (SEQ ID NO: 430).
Subsequence Percentage of Target: 98%
Subsequence Length: 1215
Number of Subsequence Base Calls: 1020
Percentage of Subsequence Base Calls: 84%
The application Java Archive (.jar) files for the REPI program are generated and processed in accordance with the description provided in U.S. Application Ser. No. 60/609,918 filed on Sep. 15, 2004, and U.S. Application Ser. No. 60/631,460, filed on Nov. 29, 2004, which are incorporated herein by reference in their entirety.
Raw sequence data from the resequencing microarray chips is provided by the Genetic Data Analysis Software version 2.0 (GDAS) packaged with the microarray reader from Affymetrix. GDAS base calling is based on a previously described base-calling algorithm, ABACUS, detailed previously (Cutler et al., 2001). A variety of base-calling algorithm parameters can be defined by the user (GDAS operators manual) to obtain a trade-off between base calling percentage and accuracy.
A description of the parameters is found in the GDAS manual on pages 207-217. The recommended (default) settings for GDAS are “conservative” settings that focus on the highest level of accuracy. In contrast, the objective of the present invention is to increase the percentage of base calls. To achieve this objective, the present inventors adjusted the parameters to allow highly permissive base calls (increased percentage) as listed below:
“Permissive” Base Calling Algorithm Settings
Filter Conditions
Algorithm Parameters
Final Reliability Rules
The settings above are significant in the present application because the base call algorithm is set up by default to sacrifice the number of base calls made in order to make the most accurate calls (i.e., for SNP detection). In the present application, the technique is less concerned about achieving the same degree of accuracy as required for SNP detection but instead expanding the number of calls made so that the longest possible stretches of contiguous sequence are produced by GDAS, while maintaining specificity.
In a comparative approach to sample preparation for analysis by the RPM V1 microarray, a starting concentration of 106 genomic copies of Adenovirus 4 prototype (AY594253) per microliter were amplified using either degenerate adenovirus PCR primers (Lin et al., 2004) or GenomiPhi isothermal methods. For the total amplification experiments, DNA was isolated from cultured adenoviruses and aliquoted at concentrations of 106 copies per microliter. DNA was amplified using a total amplification strategy (GenomiPhi, Amersham), then processed in accordance with the standard Affymetrix CustomSeq™ protocol (available from manufacturer).
Images of the hybridized microarrays are shown in
Based on the corresponding GDAS outputs (using “permissive” settings), REPI identified the following list of top-ranked returns for the conserved PCR amplification strategy for the E1A, Fiber, and Hexon-1 tile regions of the V1 RPM microarray (note: omitted from this list are several adenovirus type 4 GenBank records having nearly identical sequences as the Ad4 prototype and Ad4 vaccine strain).
In each case shown above, the adenovirus type 4 prototype (AY594253) and vaccine strain (AY594254) returned the highest expect values and bit scores for hybridization of the adenovirus type 4 prototype amplicons obtained by conserved PCR amplification. This is expected because the prototype strain was used for vaccine production and the sequences were determined to be identical. In each case, the returns were distinguished in order of similarity from the closely related Air Force (AY599837) and Navy (AY599835) adenovirus 4 field strains.
Listed below are the corresponding outputs following GenomiPhi amplification of the adenovirus type 4 prototype instead of conserved PCR (note: not shown are several closely-related adenoviruses with bit scores higher than the Air Force and navy field strains):
Based on the results evidenced by
With this one exception, it is also important to note that irrespective of which amplification strategy was employed, i.e., conserved (degenerate) multiplex PCR or GenomiPhi total amplification, the proper Ad4 calls were made (except in case of Ad4FIBER with GenomiPhi due to tile sequence errors). The consistency between these methods is important as the conserved primers may have a more direct application to serotyping members of the same family of pathogens from complex samples without the need for selective enrichment.
In the present example, hybridization and base calling was assayed as a function of the initial concentration of the target subsequent to conserved (degenerate) multiplex PCR. In addition, this example also compares two different base-calling strategies within the GDAS software: (a) “permissive” (described in Specification) and (b) “conservative” (default) settings. The biological sample utilized in this example was the adenovirus 4 prototype.
Hybridization and base calling were assayed as a function of the initial concentration of the target subsequent to conserved (degenerate) multiplex PCR. A dilution series of the adenovirus 4 prototype (ATCC) was prepared having 105, 103, and 101 genomic copies per microliter. To this end, the Affymetrix CustomSeq protocol was followed, except that samples aliquoted from the aforementioned dilutions were amplified using the conserved (degenerate) multiplex adenovirus primer strategy for E1A, fiber, and hexon genes.
Tables 18-20 demonstrate the ability of the RPMV 1 chip with REPI analysis to detect the appropriate Adenovirus type 4 target (not specific to strain) depending on whether “Conservative” (default GDAS) or “Permissive” (from Example 1) base call settings were used. A positive detection was counted when REPI returned one or more GenBank (or local) database record(s) for an Adenovirus type 4 target as the highest BLAST bit score value(s). In each case, “length” corresponds to the number of base pairs in the subsequence selected by REPI as having satisfied the conditions of the sliding window algorithm.
*indicates degraded template; n/d indicates “not determined” due to insufficient base calls
*indicates degraded template;
n/d indicates “not determined” due to insufficient base calls
Across the range of dilutions, the “Permissive” base call settings produced GDAS outputs that were used by REPI to consistently yield higher subsequence lengths and bit scores than those outputs using the default GDAS base call settings, for SNP detection. In several cases, the “Permissive” base call settings resulted in a sufficient number of base calls for REPI to detect the target whereas the default settings did not. This example demonstrates that by lowering the restrictions on base calling by GDAS and coupling the output to the REPI algorithm, higher sensitivity and discriminatory power (strain identification) among pathogens is achieved.
Of particular note here is that in some cases, there was evidence contained in the image, GDAS output, and REPI analyses, of unintentional contamination of the degenerate primer cocktail with trace amounts of Adenovirus 7 and Adenovirus 5 (confirmed by specific real-time PCR). Because there was no significant cross-hybridization between the tile regions for the prototype adenoviruses, this contamination did not result in a perturbation of the base calling or results in the Adenovirus 4 tile regions. This demonstrates the robustness of the methods described herein to perform quantitative separation of mixtures of closely related pathogens.
It is noteworthy that the Andersen group at Lawrence Livermore National Laboratory (Wilson et al., 2002b) described the use of an Affymetrix resequencing chip that could detect a low concentration (˜101 copies) following specific PCR amplification. However, it was not disclosed or suggested how this sensitivity was defined or to what extent those results were influenced by hybridization of closely related microbial species. Rather, the emphasis in that paper was the percentage of probe pairs that could be used, not the actual sequences as they were called by the chip. Therefore, the artisan would not have any expectation resulting from the disclosure. Nor would the skilled artisan have inferred this possibility from the previous patent (U.S. Pat. No. 6,228,575) or related publication (Gingeras et al., 1998) since these references fail to disclose or suggest robustness of the prescribed methods of pattern recognition to concentration differences, errors in the definition of tile region sequences, or other forms of interference.
This example is the same as Example 1, except the resequencing microarray chip (RPMV1) described above in Preparation Example 1 was probed with adenovirus type 5 field strain (Ad5FS; AY601635) or adenovirus type 7 prototype (AY594255) using either (a) conserved (degenerate) multiplex PCR or (b) GenomiPhi total amplification (
Also, without exception, each subsequence for the adenovirus type 7 tile regions (Ad7E1A, Ad7FIBER, Ad5HEXON-1 and AdSHEXON-2) listed the top “hit” (highest BLAST bit score) as the correct prototype strain for adenovirus type 7 (Ad7; Accession number AY594255) when using either conserved PCR or total (GenomiPhi) amplification. In addition, GenomiPhi amplification led to base calls from both the Ad4E1A regions and the Ad7vaccine regions, both resulting in top bit scores assigned to the correct Ad7 prototype strain.
The results obtained for the adenovirus type 5 and adenovirus type 7 prototypes confirmed those for adenovirus type 4. Specifically, the results shown in FIGS. 3A-D show that GenomiPhi total amplification allowed the entire tiled region (increased sensitivity) to be resequenced compared to the more limited region resulting from conserved (degenerate) multiplex PCR. More importantly, these results confirm that irrespective of which amplification strategy was employed, i.e., conserved (degenerate) multiplex PCR or GenomiPhi total amplification, the proper Ad5 or Ad7 calls were made.
The sample for this example was provided by the Naval Health Research Center (San Diego). The archived sample was a nasal wash from a basic trainee who had previously been immunized for adenovirus (using the Ad4/Ad7 vaccine) and who subsequently was diagnosed with adenovirus infection and febrile respiratory illness. Strains associated with infection in spite of previous vaccination are broadly referred to as “breakthrough strains”. In this case, the presumption was that a single variant form of adenovirus was the causative agent. The sample described here is referred to as Ad4-5 Breakthrough because of inconclusive culture and serological assays that indicated properties of both adenovirus types 4, while sequencing of the hexon gene (one of antigenic determinants of adenovirus serotype) indicated properties of adenovirus type 5.
The sample was processed according to the multiplex degenerate primer PCR protocol described in Preparation Example 1. This approach was demonstrated to (Lin et al 2004) successfully amplify hypervariable regions on hexon and fiber genes (as well as a relatively variable E1A region) for all adenoviruses associated with febrile respiratory infection. Otherwise, the standard Affymetrix CustomSeq protocol was used.
The resultant hybridization pattern is shown is
Selected REPI return values for Adenovirus 4,5 Breakthrough Strain on RPMV1. This “strain” was determined by several independent means to be a mixture of adenovirus type 5 and a B subgroup member, most likely being adenovirus type 21. This was borne out in the combined hybridization results and REPI analysis, with adenovirus type 5 being confirmed on all adenovirus 5 tiles, while the B subgroup prototype (adenovirus 7; AY594255) tiles returned the highest bit scores for a adenovirus type 21, a different member of the B subgroup.
Conventional DNA sequencing revealed that the complete adenovirus 5 genome was recovered and that other sequences were present that could not be assembled because an Adenovirus 5 model assumption was required. However, a significant amount of corroborating evidence was amassed to support the conclusion of an Ad5/Ad 21 co-infection, including similar results obtained in parallel experiments with 70-mer oligonucleotide arrays (Lin et al 2004) showing evidence of an Ad 5/Ad 21 mixture or co-infection, and sequencing of amplicons using conventional approaches that supported the presence of both Ad5 and Ad 21. An extended study conducted subsequently of adenovirus breakthrough strains (in collaboration with NHRC San Diego) showed that a variety of such strains were actually comprised of mixtures of lesser adenoviruses (manuscript in preparation).
There are several significant aspects and advantages implicit in this discovery:
Influenza A positive clinical samples of unknown serotype were collected during the 2002-2003 and 2003-2004 flu seasons and provided to Dr. Zheng Wang as frozen nasal washes by Dr. Elizabeth Walter.
Samples were processed for pathogen chip analysis as follows:
EPICENTRE MasterPure™ DNA Purification Kits (Madison, Wis.) were used to extract total nucleic acids from 50 Ill of nasal wash (sample NW20031114-05-02) without RNase treatment. Sample NW20031114-05-02 was collected on Nov. 14, 2003 at Lackland Air Force Base (San Antonio, Tex.). The patient was vaccinated on Nov. 10, 2003. The total nucleic acids were suspended in 20 μl nuclease free H2O. Two-step RT-PCR was employed to amplify each of the viral gene segments.
Briefly, the RNA in 4 μl total nucleic acids was transcribed into cDNA by using SuperScript™ III Reverse Transcriptase (Invitrogen, Carlsbad, Calif.) according to the protocol provided by the manufacturer and 100 pmol Uni3 primer in 20 μl. The RT (reverse transcription) reaction was performed at 42° C. for 1 hr and then was inactivated at 70° C. for 15 min. 2 μl of the RT-reaction was used for cDNA amplification. The cDNA was amplified by using TaqPlus Long System (Stratagene, La Jolla, Calif.) according to protocol provided. Two different PCR conditions were used for amplification. For universal PCR, universal primers uni3 and uni5 were used to amplify all eight segments of hemagglutinin (HA), neuraminidase (NA) and matrix (M) (see Hoffman et al, 2001). For multiplex PCR, segments were amplified by mixing three segment specific primer pairs (Bm-HA-1/Bm-NS-890R, BA-Na-1/Ba-Na-1413R and Bm-M-1/Bm-M-1027R). The final concentration of Mg2+-ions was 2 mM and final primer concentration was 1 μM. PCR condition for universal amplification was: 94° C. for 2 min followed by 29 cycles of 94° C. for 1 min, 40° C. for 2 min. 72° C. for 3 min, and final extension at 72° C. for 10 min. The multiplex PCR condition was basically same except that the annealing temperature was raised to 58° C. PCR products were purified by Qiagen PCR Purification kit. 500 ng of universally amplified PCR products and 1000 ng of multiplex PCR products were applied to two V1 Pathogen chips for hybridization according to Affymetrix protocol, respectively.
Primer Sequences (from Hoffman et al, 2001)
The chip scan results are shown for each of the two amplification protocols: (a) universal primer PCR and (b) multiplex PCR (
For this experiment, the present inventors obtained sequence for the HA gene of the Flu strain that evaded the 2003-2004 vaccine and caused widespread illness, namely Fujian/411/2002 (note: only the HA sequence was available for Fujian/411/2002) and incorporated that into the database searched by REPI. Each of the two amplification methods produced effectively the same results, although bit scores varied slightly. Specifically, Fujian 411/2002 returned the highest bit score for FluAHA3 in the REPI output files. Moscow 10/99, the strain used for vaccination in 2003, did not return a REPI output for FluAHA3.
The highest BLAST bit score for HA3 was achieved for (InfluenzaA/Fujian/411/2002_Hemagglutinin—104), the H3N2 etiologic agent of the 2003-2004 Influenza outbreak in the United States that evaded protection by vaccination. An indistinguishable viral sequence for HA3 also achieved the same bit score. Over 500 GenBank records were returned for HA3, all having expect values of 0.0 and bit scores ranging from those shown above to 995.636. The vaccine strain, (A/Panama/2007/99(H3N2)), was not in the list of returned Influenza A strains in the HA3 region. Note: sequence for InfluenzaA/Fujian/411/2002_Hemagglutinin—104 was only available for HA, so other regions are not compared here.
Although not shown, the present inventors have been able to identify the H1N1 outbreak strain from the 2002-2003 flu season.
The Affymetrix CustomSeq protocol specifies that the hybridization step be at least 16 hours to facilitate maximum hybridization. For the intended purpose of a diagnostic capability, this could be prohibitively long. Therefore, the applicability and efficiency of the present methodologies to short hybridization times was assessed.
In the present example hybridization and base calling were performed, using samples that were amplified by specific PCR reagents and then hybridized for periods of either 1 hour, 30 minutes, or 15 minutes on the resequencing microarray chip (RPMV1) described above in Preparation Example 1. For this comparative study the Affymetrix CustomSeq protocol was followed with the exception of varying hybridization time. In the example data shown (
The results appear in
The corresponding GDAS and REPI analyses show that the number of base calls resulting is fewer in the case of the 15 minute hybridization compared to that for the 16 hour hybridization (See Tables 26-31). For example, the SPYERMB tile region produced a subsequence length of 219 bp (98% of tile region, 95% base calls made within the subsequence) for a 16 hour hybridization, while discontinuities in the hybridization pattern from a 15 minute hybridization resulted in fragmentation of the tile into two subsequences of lengths 100 bp (45% of tile region, 99% base calls made within subsequence) and 87 bp (39% of tile region, 99% base calls made in subsequence).
Enterococcus faecium
Streptococcus pyogenes
Strepto-
coccus
Streptococcus
The tables above show the comparative top rankings for the three different antibiotic resistance markers described in Example 6 for 16 hour versus 15 minute hybridization times. Although the SPYERMTR tile region did not produce a sufficient number of base calls to allow identification following a 15-minute hybridization, both the SPYERMB and SPYMEFAE returned high bit scores for the same variants of the antibiotic resistance markers comprising the tile regions (identified in Table 9).
However, the REPI output for each of the three tile regions showed that the highest bit scores in each region were the same (except for SPYERMTR which lacked sufficient base calls for the 15 minute hybridization), although both the bit scores and expect values were different in each case. Similar results were obtained for 30 minute and 1 hour hybridizations, with an increase in the number of base calls made with increasing hybridization times. Further, this example clearly illustrates the robustness of the method to make fine scale discrimination between targets with a range of different hybridization patterns.
A variety of methods might be employed to reduce the amount of background human DNA in clinical samples when using total amplification. One method is direct subtraction of the background genomic DNA from amplified products in hybridization solution with COT-1 fraction human genomic DNA that consists largely of rapidly annealing repetitive elements. Another method is bead-based subtraction of background genomic DNA from clinical sample prior to total amplification. A third method is the combination of the above methods, the genomic background DNA from clinical sample were subtracted using bead-based subtraction prior to total amplification, subsequently, the background DNA might be further subtracted from the clinical sample in hybridization solution with COT-1 human DNA.
Methods for
A nasal wash was collected from one clinical sample and the total nucleic acid was isolated using MasterPure™ DNA purification kit (Epicentre). The nucleic acid was predigested with McrBc (an endonuclease that cleaves DNA containing methylcytosine on one or both strands) at 37° C. for one hour, and then heated to 65° C. to denature the enzyme. Subsequently, the sample was split to two aliquots and amplified with GenomiPhi DNA amplification kit (Amersham Biosciences). The amplified products were fragmented and labeled according to Affymetrix CustomSeq protocol.
For this comparative study the Affymetrix CustomSeq protocol was followed except for the following changes. The clinical sample was hybridized to the resequencing microarray chips (RPMV1) either with standard hybridization solution according to Affymetrix CustomSeq protocol (
Methods for
The present example describes bead-base subtraction and the combination of bead-base subtraction with subtractive hybridization methods for increasing the pathogen specific signal-to-noise ratio when using total amplification.
COT-1 human DNA, consisting largely of rapidly annealing repetitive elements, was biotinylated at 3′-end using terminal transferase (Tdt) (NEB) to generate Biotinylated Subtraction Probes (BSPs). An aliquot of ˜400 ng of BSPs was incubated at 95° C. for 10 minutes in 4×SSC and 0.2% SDS with nucleic acid extract of clinical nasal wash (diagnosed as Ad4 positive clinical sample) predigested with McrBC at 37° C. for one hour, then slowly cooled to 65° C. to allow co-hybridization with the HMB for one hour. After hybridization, the BSP-HMB complexes were incubated at room temperature for 30 minutes at 1×PBS, 0.1% BSA with BcMag.Streptavidin Magnetic Beads (SMBs, Bioclone). The beads were separated from supernatant using a magnetic separator. The supernatant was then transferred to a new tube and the magnetic beads were washed once with 133 PBS, 0.1% BSA. The supernatants were combined and precipitated with ethanol. The DNA pellet was resuspended with nuclease free water (Ambion) then split to two aliquots and amplified with GenomiPhi DNA amplification kit (Amersham Biosciences) according to the manufacture recommended protocol. Briefly, 1 ml DNA template was denatured with 9 μl of sample buffer at 95° C. for 3 minutes, the cooled down to 4° C. In a separate tube, 1 μl of enzyme was mixed with 9 μl of reaction buffer, and then the mixture was added to the denatured DNA template. The reaction was then performed at 30° C. for 18 hours at Thermal Cycler-PTC225 (MJ Research Inc., Reno, Nev.).
For this comparative study, the amplified products were hybridized to the resequencing microarray chips (RPMV1) either with standard hybridization solution according to Affymetrix CustomSeq protocol (
Methods for
The combination of bead-base subtraction and subtractive hybridization procedure was repeated with double amount of initial material and the result appearing in
Subtractive Hybridization Results
The results of the subtractive hybridization approaches are shown in
However, through the combined use of a bead-based subtraction prior to amplification and co-hybridization with COT-1 human DNA, enough base calls could be made to identify and rank the adenovirus 4 field strains (Navy and Air Force without discrimination) higher than those of the vaccine and prototype adenovirus 4 strains in each of three subsequences identified in the Ad4HEXON-1 tile region (
Moreover, by performing the same set of combined steps using 2 microliters of starting sample material instead of 1 microliter (
A listing of the sequence tiles for the RPM V2 chip is listed below in (Table 35). This represented an approximate 10-fold increase in the content of the RPM V1 chip.
This example is to present application of newly modified random RT-PCR protocol on isolated Influenza A (Fujan/411/2002 strain) nucleic acids spiked into normal nasal wash samples. A random RT-PCR protocol has been applied by DeRisi's lab to amplify viral pathogens for microarray interrogation (Wang et al., 2003). Modification to this protocol was recently reported for amplifying cultured influenza viruses (Kessler et al., 2004). However, the modified protocol was not reported to amplify viral targets in clinical samples.
The Influenza A virus cultures (H1N1 & H3N2) were provided by the Air Force Institute for Operational Health (AFIOH) and sent to Virapure, LLC for amplification and quantification prior to the study. According to the Certificate of Analysis, the virus was amplified for two rounds on early passage MDCK cells purchased from ATCC exclusively for this project, followed by virus amplification in the presence of 1 μg/ml TPCK treated trypsin. At 48 hours after infection, the virus supernatants were collected and adjusted to contain 3% sucrose, 20 mM Tris pH 7.1 and pH stabilizer. Prior to delivery, Corning screw top cryovials were filled with 200-300 μl aliquots of supernatant. The titration of each sample vial was performed according to procedure AM110, Plaque Forming Assay of Influenza Virus. A titer of 1.2×107 PFU/ml was obtained for Influenza A H3N2, K0717 SV5/SV40 RMK and 5×106 PFU/ml for Influenza A/California/2935/03/H1N1.
To generate a clinical collection control, a nasal wash (0.9% Normal Saline) was obtained from an asymptomatic member of the Clinical Division of the Advanced Diagnostic Laboratory at Lackland AFB. From each Virapure pre-quantified titer, a 10-fold/6-series dilution of Influenza-spiked nasal aspirate was generated beginning with and including the Virapure stock of each strain. A total of 100 μl of each influenza-spiked nasal wash were processed from the dilution using the MasterPure Total Nucleic Acid Extraction distributed by EPICENTRE. During the protocol, 150 μl of 2× T and C Lysis Solution containing Proteinase K was added to each 100 μl dilution of the spiked nasal wash, vortex thoroughly and incubated at 65° C. for 15 minutes. Next, the samples were placed on ice for 5 minutes followed by the addition of 150 μl of MPC Protein Precipitation Reagent. The samples were vortex mixed vigorously for 10 seconds. The mixture was then precipitated by centrifugation at 10,000×g for 10 minutes and the supernatant transferred to a microcentrifuge tube. To the recovered supernatant, 500 μl of isopropanol wer added and the tube inverted (30-40) times.
The mixture was centrifuged at 4° C. for 10 minutes and the isopropanol decanted without dislodging pellet. Following, the pellet was rinsed twice with 75% ethanol and dried at room temperature. Once dried, the nucleic acid pellet was resuspended in 35 μl of TE Buffer. Each sample was properly labeled and placed in dry ice prior to shipment to the Naval Research Laboratory.
Each of three 100 μl normal nasal wash aliquots was spiked with a serial of dilution (ranging from 105 pfu to 101) of influenza A H3N2 viral cultures. Total nucleic acids were extracted with EPICENTRE MasterPure™ DNA Purification Kits (Madison, Wis.), as in Example 5, and suspended in 40 μl H2O. Random RT-PCR was applied as follows, according to that described (Kessler et al., 2004):
2 μl of total nucleic acid was reverse transcribed by using 1 μl primer D (40 pmol/μl; 5′-GTT TCC CAG TCA CGA TCN NNN NNN NN; SEQ ID NO: 573) and 1 μl SuperScript™ III Reverse Transcriptase (50 units/μl; Invitrogen, Carlsbad, Calif.) in 20 μl. The RT (reverse transcription) reaction was performed at 42° C. for 1 hr and then was inactivated at 70° C. for 15 min. Then 10 μl RT products were amplified with 1 μl primer E (100 pmol/μl; 5′-GTT TCC CAG TCA CGA TC; SEQ ID NO: 574) and 2.5 units TaqPlus Long polymerase (5 units/μl, Stratagene, La Jolla, Calif.) in 50 μl. PCR was conducted with 35 cycles of 94° C. for 30 s, 40° C. for 30 s, 50° C. for 30 s, and 72° C. for 160 s, followed by an additional 7 minutes at 72° C. PCR products from 105, 103 and 101 pfu flu spiking samples were applied to three different V1RPM pathogen microarrays for hybridization according to Affymetrix Customseq protocol.
The V1RPM chip scans for the three spike-in samples, 105, 103, and 101 pfu, amplified by this random RT-PCR method, are shown in
Because of success of random RT-PCR in this experiment, it is believed that this protocol will work in flu clinical samples and other pathogenic viruses.
This example is significant because it shows that a completely non-biased amplification method, derived from a June 2001 public domain protocol may be used as a preparative step for resequencing microarrays, resulting in PCR-level sensitivity (101 target copies). Although this amplification protocol has been used for several published examples of long oligomer (70-mer) microarray-based assays, it is not obvious that it would suffice for resequencing microarray approaches. Numerous modifications and variations on the present invention are possible in light of the above teachings. It is, therefore, to be understood that within the scope of the accompanying claims, the invention may be practiced otherwise than as specifically described herein.
In summary, the following pathogens were detected/discriminated from laboratory and/or clinical samples on RPMV1, (in aggregate representing >180 chips tested) with 100% concordance of identification and input identity, and with no type cross-hybridizations, except to parts of sequence tiles representing homologous sequences between type variant strains, in:
Selected examples of RPMV1 microrarray results for prepared non-infectious laboratory samples of Variola Major (smallpox), Ebola Virus, B. anthracis, Y. pestis, and F. tularensis have been performed, but not shown here. Variola Major (smallpox), Ebola Virus, and F. tularensis samples were prepared using specific PCR primers. B. anthracis and Y. pestis samples were amplified using GenomiPhi reagents (as described in Example 3).
The following tables present the sequence features and putatitve gene products for the sequenced adenoviruses appearing in Table 6 and SEQ ID NOs: 575-587.
For the products designated above in Tables 39-51, the present inventors note that it is routine in the art by referring to the universal genetic code to translate the nucleic acid sequence identified in the “location” column to the corresponding amino acid sequence. As such, the amino acid sequences designated in the “product” column have not been explicitly listed.
The present application claims priority to U.S. provisional Application Ser. No. 60/590,931, filed on Jul. 2, 2004, U.S. provisional Application Ser. No. 60/609,918 filed on Sep. 15, 2004, U.S. provisional Application Ser. No. 60/631,437 filed on Nov. 29, 2004, U.S. provisional Application Ser. No. 60/631,460 filed on Nov. 29, 2004 and U.S. provisional Application Ser. No. 60/691,768 filed on Jun. 16, 2005. This application is also related to U.S. non-provisional application Ser. No. ______, titled “Computer-Implemented Biological Sequence Identifier System and Method,” filed along with this application on Jul. 2, 2005. The entire contents of these applications are incorporated herein by reference.
The United States Government owns rights in the present invention pursuant to funding from the Defense Threat Reduction Agency (DTRA; Interagency Cost Reimbursement Order (IACRO #02-4118), MIPR numbers 01-2817, 02-2292, 02-2219, and 02-2887), the Office of the U.S. Air Force Surgeon General (HQ USAF SGR; MIPR Numbers NMIPR035203650, NMIPRONMIEPRO35203881, NMIPRONMIPRO35203881), the U.S. Army Medical Research Acquisition Activity (Contract # DAMD17-03-2-0089), the Defense Advance Research Projects Agency (DARPA; MIPR Number M189/02), and the Office of Naval Research (NRL Work Unit 6456).
| Number | Date | Country | |
|---|---|---|---|
| 60590931 | Jul 2004 | US | |
| 60609918 | Sep 2004 | US | |
| 60631437 | Nov 2004 | US | |
| 60631460 | Nov 2004 | US | |
| 60691768 | Jun 2005 | US |
