| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Abstracts of Oak Ridge Posters |
1 Nanosphere, Inc., Northbrook, IL 60062
aauthor for correspondence
Staphylococcus aureus (SA) is one of the most important human pathogens, causing both nosocomial and community-acquired infections (1)(2). The occurrence of methicillin-resistant SA (MRSA) has increased steadily worldwide and now accounts for a substantial portion of all staphylococcal infections in US hospitals (3). To develop preventive measures, a rapid screening method, along with accurate and timely identification of MRSA, is essential. The existing techniques for doing so are either time-intensive (culturing of bacteria on selective media), relatively insensitive (use of latex agglutination), or expensive and easily susceptible to operator error (such as PCR).
We describe a method designed for clinical laboratories using oligonucleotides conjugated to gold nanoparticles. We avoid radioactivity, fluorescence, or target amplification (such as PCR), and use a simple and rapid hybridization-based approach in a microarray format, with ClearReadTM technology to detect specific genomic sequences (4).
The ClearRead procedure involves a two-step process: the first involves the hybridization of target to oligonucleotides conjugated to gold nanoparticles as well as oligonucleotides attached to a solid matrix; the second step involves the catalytic deposition of silver on the gold nanoparticle, providing a sixfold amplification of signal (Fig. 1A
) (4)(5). The differentiation at isothermal hybridization temperatures is a result of the sharp melting transitions characteristic of nanoparticle probes (5). Previous methods based on this property (6)(7) have required the use of PCR and have not directly assayed for genomic DNA. Our assay, on the other hand, requires minimal amounts of genomic DNA (
500 ng, or 108 DNA molecules) and has been used to reliably identify MRSA from liquid cultures, based on the detection of the mecA and tuf genes. Resistance to methicillin is mediated by the presence of penicillin-binding protein 2a, encoded by the mecA gene (8)(9). We are also able to identify polymorphisms in the tuf gene (10) that are characteristic for SA.
|
We applied our technology to screen a staphylococcal clinical panel, after growth in liquid medium, and found excellent correlation with culture results as well as standard coagulase tests (100% correlation). Our assay typically uses a hybridization of 30 min, and our proprietary detector, the Verigene IDTM (Nanosphere, Inc.). The Verigene ID takes advantage of the strong light-scattering properties of the silver-enhanced gold nanoparticles (11), thereby eliminating the need for the high-powered lasers and photomultiplier tubes typically used in instrumentation for the detection of fluorescent signals against a glass surface. Light-emitting diodes in combination with a complementary metal oxide semiconductor sensor are used for detection. The technology uses the slide itself as a waveguide, which allows the particle labels to scatter the evanescent field, giving even distribution of excitation light across the entire surface of the glass slide. This illumination methodology for the detection of hybridization events was first described by Stimpson et al. (12) as a means of performing single-nucleotide polymorphism analysis through measurement of the real-time melting behavior of DNA from a surface. Instead of using white light, the Verigene ID has been optimized to collect specific wavelengths to suppress background contributions. The imaging system allows sensitive detection of the proprietary Nanosphere gold labels. We also performed studies demonstrating the ability of the system to differentiate MRSA samples with as little as 5 min of hybridization time.
Oligonucleotide synthesis was performed with phosphoramidite chemistry on an Applied Biosystems Expedite 8909 DNA synthesizer. Modified oligonucleotides were prepared and loaded on the 13-nm diameter gold nanoparticles as described previously (13)(14) and were then designated as "probes". Oligonucleotides that were attached to the microarray surface were designated as "captures". Each microarray slide contained 10 test wells, such that 10 different samples could be tested per slide. Each test well had three test spots for mecA, three test spots for tuf, and nine negative-control spots. The hybridization cassette had 10 individual, spatially separated hybridization wells. An image of this hybridization cassette can be seen at www.nanosphere-inc.com. The specific oligonucleotide sequences used were 5'-TTCCAGATTACAACTTCACCA-3' (mecA capture) (4), 5'-GCACTTGTAAGCACACCTTCAT-3' (mecA probe), 5'-CCATTCTTCTCAAACTATCGT-3' (tuf capture), 5'-TTCTATTTCCGTACTACTGACGTAACT-3' (tuf probe), and 5'-ATCGACTCCCGCAGACACCTTCTC-3' (negative-control capture). The negative-control sequence was derived from the human mutant methylene tetrahydrofolate reductase gene (15) after BLAST searches to ensure lack of cross-hybridization.
Arrays were printed on either NoAb (NoAb Biodiscoveries) or CodeLink (Amersham Biosciences) modified microscope slides by use of a Genomic Solutions Prosys Gantry (Genomic Solutions) with either SynQuad noncontact dispensing nozzles or Stealth SMP3 (Telechem International) split pins. Each spot on each array ranged from 200 to 400 µm in diameter after printing. Regardless of slide type or dispensing method, amine-modified oligonucleotides were suspended in 150 mmol/L sodium phosphate, pH 8.5, at 100 µmol/L. Slides were arrayed at low humidity (relative humidity <30%) and subsequently rehydrated in a humidity chamber (relative humidity >70%) for 18 h. Slides were then dried, washed to remove excess oligonucleotides, and stored in a cabinet desiccator (relative humidity <20%) until use.
We received 48 Staphylococcus isolates as swabs from a local hospital microbiology laboratory, where they had been identified as 29 coagulase-negative Staphylococcus (CoNS) and 19 S. aureus. They were inoculated into tryptic soy broth (TSB) and grown overnight at 37 °C. A loopful from the overnight culture was grown for 24 h at 37 °C on tryptic soy agar containing 50 mL/L sheep blood and on mannitol salt agar containing 6 mg/L oxacillin. Colony morphologies and hemolytic patterns were recorded for each sample. Eight samples showed colonies with mixed hemolytic patterns. Twelve samples (2 typed as CoNS and 10 typed as S. aureus) showed strong growth on oxacillin-containing agar. For those 12 samples, a loopful of cells representing multiple colonies was picked from the MSA-oxacillin plate, grown overnight in TSB at 37 °C, and frozen with glycerol at –80 °C. The remaining samples were picked from blood agar plates and grown and frozen similarly. These frozen cultures were used to inoculate TSB for growth of cells for DNA isolation and growth on blood agar for use in phenotyping. Confirmatory testing was conducted with the OXOID Staphytect Plus Kit (Oxoid), using conditions recommended by the manufacturer, or by growth in TSB containing various concentrations of oxacillin (0, 0.25, 0.5, 2.0, and 4.0 mg/L) for 21 of the samples. Cells grown in TSB were lysed with achromopeptidase, and DNA was isolated by use of the QIAGEN Genomic DNA 20/G protocol.
DNA samples were blinded and screened by our ClearRead technology in a microarray format. We hybridized 500 ng of purified genomic DNA from each sample for 30 min in a buffer containing 200 mL/L formamide, 5x standard saline citrate, 0.5 mL/L Tween 20, and a multiplex mixture of mecA and tuf nanoparticle probes (at 100 and 250 pmol/L, respectively), at 40 °C (n = 36 for each sample) in a final volume of 50 µL. Slides were washed in 0.5 mol/L NaNO3, and signal was developed for 3 min at room temperature by use of our proprietary silver development solution [prepared at Nanosphere, Inc, but similar in composition to an equimolar solution of reagents from Sigma (cat. nos. S5020 and S5145)]. Slides were scanned and imaged with the Verigene reader, with an exposure time of 199.8 ms, and data were analyzed by use of JMP software. Intensity values for every test spot were collected. Each sample was hybridized on 12 separate microarray slides, 1 test well per slide, to make a total of 36 separate data points for each sample (12 microarray slides x 3 spots per slide).
A threshold was generated using the mean intensity values + 3 SD of the nine negative-control spots per well. A positive response was defined as an intensity value above the threshold for that sample well.
The success rate was 100% compared with the results obtained from bacterial culturing in the presence of oxacillin as well as from coagulase testing; all MRSA, methicillin-sensitive SA (MSSA), and MR/MS non-SA (MRCoNS and MSCoNS) were correctly identified (8 of 8, 9 of 9 and 29 of 29 samples, respectively).
Purified genomic DNA, extracted from culture, from MRSA, MSSA or from methicillin-sensitive S. epidermidis (MSSE) were screened as described previously. Briefly, 500 ng of purified genomic DNA was hybridized for 5, 15, or 25 min in a multiplex mixture of mecA and tuf nanoparticle probes at 40 °C. Slides were washed, signal was developed, spots were scanned and imaged, and data were analyzed as described above. Each sample was separately hybridized on four different microarrays, and each microarray had six specific captures for the tuf gene, making a total of 24 individual data points per target per condition (4 slides x 6 specific captures; n = 24 each). As before, a threshold was generated by use of the mean intensity values + 3 SD of nine negative-control spots per well. A sample was designated as positive if the specific intensity values were above the threshold value for each individual well.
The results of this experiment are shown in Fig. 1B
for the tuf gene. After 5 minutes of hybridization at 40 °C, MRSA and MSSA were clearly differentiated from MSSE with use of 500 ng of genomic DNA (equivalent to 108 molecules of DNA; also the approximate number of bacteria present in a single bacterial colony), a mock hybridization conducted with target replaced by water (designated as no-target), and the Verigene ID. These values are comparable with those reported previously by Jenison et al. (16), although they used a much longer hybridization time (2 h) for detection of an equivalent copy number of mecA molecules from genomic DNA. We also have preliminary evidence (data not shown) that hybridization can be completed in 1 min, depending on the amount of genomic DNA target used. We attribute the difference in hybridization rates to the increased sensitivity provided by direct detection of gold nanoparticle probes (4), because Jenison et al. (16) indirectly detected their biotin-labeled detector probes by use of a sandwich anti-biotin-horseradish peroxidase system, as well as the fact that we use a three-dimensional slide substrate (17).
In conclusion, we have demonstrated technology that uses genomic DNA, eliminates the need for generating PCR target, is specific, does not require fluorescent labels, and uses a low-cost detector. Our use of gold-nanoparticle probes makes the assay ultrasensitive (that uses the equivalent of DNA from a single isolated colony), and slides can be archived after processing because the gold-derived signal is not susceptible to photobleaching, in sharp contrast to fluorescence-based technologies.
Although we report reliable detection and differentiation of MRSA with use of 500 ng of genomic DNA, the sensitivity of our assay is considerably greater, extending at least out another two orders of magnitude, to at least 2.5 ng (5 x 105 molecules) of unamplified bacterial genomic DNA with no loss of specificity and without any alterations in buffer composition or assay conditions (our unpublished data). These results compare very favorably with other studies (3)(4)(18), in spite of the fact that we use complex genomic DNA in our assays. In developing a biologically relevant assay for detection and differentiation of MRSA, we have not been attempting to push the bounds of sensitivity. The combination of reliable specificity, high sensitivity, and low cost make this technology a viable option for widespread clinical use.
Acknowledgments
We thank Sung Jon Lee for help with bacterial culturing and DNA extraction and Hao An for assistance in sequence design. We acknowledge the intellectual contributions of James Storhoff and Susan Hagenow. We also acknowledge the manufacturing and engineering groups at Nanosphere, Inc., for technical support.
Footnotes
Editor’s Note: This poster received a "Critical and Point-of-Care Testing" Poster Award at the Oak Ridge Conference.
References
G) polymorphism, double heterozygosity methionine synthase 2756 AG/methionine synthase reductase (MTRR) 66 AG, and elevated homocysteinemia are three risk factors for having a child with Down syndrome. Am J Med Genet 2003;121A:219-224.
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  L.-X. Zhu, Z.-W. Zhang, C. Wang, H.-W. Yang, D. Jiang, Q. Zhang, K. Mitchelson, and J. Cheng Use of a DNA Microarray for Simultaneous Detection of Antibiotic Resistance Genes among Staphylococcal Clinical Isolates J. Clin. Microbiol., November 1, 2007; 45(11): 3514 - 3521. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L.-X. Zhu, Z.-W. Zhang, D. Liang, D. Jiang, C. Wang, N. Du, Q. Zhang, K. Mitchelson, and J. Cheng Multiplex Asymmetric PCR-Based Oligonucleotide Microarray for Detection of Drug Resistance Genes Containing Single Mutations in Enterobacteriaceae Antimicrob. Agents Chemother., October 1, 2007; 51(10): 3707 - 3713. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. V. Baptista, M. Koziol-Montewka, J. Paluch-Oles, G. Doria, and R. Franco Gold-Nanoparticle-Probe-Based Assay for Rapid and Direct Detection of Mycobacterium tuberculosis DNA in Clinical Samples. Clin. Chem., July 1, 2006; 52(7): 1433 - 1434. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
