Original Report

A Flow-Through Ultrasonic Lysis Module for the Disruption of Bacterial Spores Cynthia L. Warner,* Cynthia J. Bruckner-Lea, Jay W. Grate, Timothy Straub, Gerald J. Posakony, Nancy Valentine, Richard Ozanich, Leonard J. Bond, Melissa M. Matzke, Brian Dockendorff, Catherine Valdez, Patrick Valdez, and Stanley Owsley Pacific Northwest Laboratory, Richland, WA

Keywords: spore, Bacillus, ultrasound, sonication, sequential injection analysis, automation

A

n automated, flow-through ultrasonic lysis module that is capable of disrupting bacterial spores to increase the DNA available for biodetection is described. The system uses a flow-through chamber that allows for direct injection of the sample without the need for a chemical or enzymatic pretreatment step to disrupt the spore coat before lysis. Lysis of Bacillus subtilis spores, a benign simulant of Bacillus anthracis, is achieved by flowing the sample through a tube whose axis is parallel to the faces of two transducers that deliver 10 W cm2 to the surface of the tube at 1.4-MHz frequency. Increases in amplifiable DNA were assessed by real-time PCR analysis that showed at least a 25-fold increase in amplifiable DNA after ultrasonic treatment with glass beads, compared with controls with no ultrasonic power applied. The ultrasonic system and integrated fluidics are designed as a module that could be incorporated into multistep, automated sample treatment and detection systems for pathogens. ( JALA 2009;14:277–84)

INTRODUCTION The need for a rapid, automated system for the release of DNA from spores exists across a variety of disciplines. In recent years, the release of *Correspondence: Cynthia L. Warner, Pacific Northwest National Laboratory, Chemical and Biological Signature Science, P.O. Box 999, MS K4-12, Richland, WA 99352; Phone: þ1.509.372.4681; E-mail: [email protected] 1535-5535/$36.00 Copyright

 c

2009 by The Association for Laboratory Automation

doi:10.1016/j.jala.2009.04.007

biological agents as weapons and the expected recurrence of an attack have heightened the need for a means of rapidly identifying such materials. Delayed identification of biological threats can result in the inadvertent spread of the material and more illness and potentially greater loss in life because of delayed treatment. Quantitative real-time PCR is the most sensitive and specific method for the positive identification of threat agents such as Bacillus anthracis that causes anthrax. Unfortunately, organisms that form tough spore coats such as those of the Bacillus genera, hinder access to the nucleic acid targets, and thus reduce chances for trace detection in environmental settings. Therefore, it is necessary to develop methods to increase the quantity of amplifiable DNA from spores that are suitable for use in automated monitoring systems for biodetection. Bench laboratory methods for lysing cells and spores include disrupting cell membranes with detergent, heating, or mechanical bead-beating methods. The gentler heat and detergent methods are suitable for many cell types but are typically ineffective on spores. Without prior time-consuming treatments to germinate spores or otherwise soften their spore coats, mechanical bead-beating is generally necessary. Sonication methods have been described for spore1e6 and cell7e12 rupture using both low(w20 kHz) and high-frequency (1e5 MHz) ultrasound. Many of these methods still require some form of chemical treatment of the spores before sonication, and the results are variable. Nevertheless, ultrasonic approaches appear to be most promising for the development of lysis methods for automated fluidic biodetection systems. JALA

October 2009

277

Original Report To address these issues, our previous work focused on developing an ultrasonic flow-through lysis system for Bacillus spores.4 This bench-scale system, using conventional test and measurement equipment for operation, provided proof of principle experiments indicating that DNA availability was significantly increased. However, the unit involved significant size, complexity, cost, and safety issues that were unsuitable for general use or incorporation in automated pathogen detection systems. In this current report, we describe a flow-through dualtransducer ultrasonic spore lysis device configured as part of a completely automated module. It features a smaller form factor, lower power requirements, and lower cost. Samples require no chemical pretreatment before lysis, and the device can be configured for use in a modular fashion with other operational components if multiple sample processing steps are required. The design incorporates a sequential injection analysis (SIA)13,14 platform that allows precise aspiration and delivery of solutions (see Fig. 1B). This approach yields a flexible system for varying experimental parameters with the practical ability to process sample volumes from microliters to milliliters. Control programs automate sample aspiration, processing, and delivery, as well as subsequent system cleaning, disinfecting, and rinsing. During processing, the sample is precisely metered through the flow-through sample treatment chamber that is placed at the optimal location for coupling energy from the ultrasonic transducers (see Fig. 1C). We investigated several parameters for the operation of the lysis module to develop an automated protocol that provides a substantial increase in available DNA from Bacillus subtilis spores. Spore viability was also assessed in parallel. Quantitative real-time PCR is now used to more accurately assess increased DNA availability (compared with dilutionto-extinction techniques used previously). Real-time PCR

has emerged as the preferred method for such assessments and is also expected to be the method of choice for automated pathogen detection systems. Therefore, a lysis method that works well with real-time PCR is desirable. In addition, in this study we demonstrate that the ultrasonic power, and not just heat alone, is responsible for the observed increases in available DNA. Thus, this work provides improved characterization of the increase in DNA availability, determination of empirical parameters for operation, and demonstration of the efficacy of a simple dual transducer configuration, as part of a compact flow-through lysis module suitable for incorporation in automated instruments for biodetection.

MATERIALS

AND METHODS

Bacterial Spore Preparation The gram-positive, spore-forming bacteria, B. subtilis ATCC 49760 (Bacillus globigii) was cultured in tryptic soy broth (TSB, Difco, Fisher Scientific, Pittsburgh, PA). The culture was started from a 80 C 10% glycerol-preserved freezer stock. A small quantity (w10 mL) of freezer stock was added to 3.0 mL of sterile TSB in a 10-mL snap-cap tube. The tube was incubated at 30 C for w14 h while being agitated at 150 rpm on an orbital shaker. After the 14-h incubation, 150 mL of the vegetative cells were spread plated onto new sporulating medium (NSM) and incubated upside down for 3e5 days at 37 C. NSM is composed of 3 g L1 tryptone (Fisher Scientific), 3 g L1 yeast extract (Difco, Fisher Scientific, Pittsburgh, PA), 2 g L1 Bacto-agar (Difco, Fisher Scientific, Pittsburgh, PA), 23 g L1 Laboratory-Lemco Agar (Oxoid, Fisher Scientific), and 1 mL 1% MnCl2.4H2O (Sigma, Fisher Scientific).15 The cultures were checked microscopically for sporulation and harvested when O90%

Figure 1. (A) The computer controlled automated lysis module, (B) schematic of the fluidics system, and (C) detail of the transducer chamber. 278

JALA

October 2009

Original Report spores were present. The spores were washed from the plates with 10 mL of sterile deionized water (purified to 18 MU-cm on a Barnstead E-Pure water purification system), then centrifuged at successively higher speedsd1000 rpm (1e2 min), 5000 rpm (1e2 min), and 10,000 rpm for 6e7 min to separate the spores from the vegetative debris. The samples were decanted and washed 4e6 times with sterile E-Pure water to remove the vegetative cell debris. Microscopic evaluation was used to determine the number of washes needed to achieve O95% spore purity. The spores were then stored in water at 4 C. The spore counts were based on plate counts using six replicates. Before ultrasonic treatment, spore suspensions were washed four times with sterile water and re-suspended to a concentration of w107 spores mL1.

Determination of Spore Viability Plate counts were used to determine the efficacy of the ultrasonic treatment conditions. A 10-fold dilution series of the treated, control and stock spores was generated after each experiment. Each dilution was then spotted onto a tryptic soy agar (TSA) plate six times, with 20 mL of sample per spot. The plates were then incubated at 30 C for 24 h before counting the colony forming units (CFU). For each treatment condition, the following equation was used to determine the percent viability: (Ctreated/Ccontrol)  100, where Ctreated is the concentration in CFU mL1 of the ultrasound treated sample, and Ccontrol the concentration of the untreated sample. Standard propagation of error was used in the statistical analysis of viability data.

Ultrasonic Treatment of Spores The automated spore lysis module incorporates a SIA fluidics platform to enable automation of sample manipulation. The schematic of the entire fluidics manifold is illustrated in Figure 1B and includes a 1-mL syringe pump (Cavro Scientific Instruments, Sunnyvale, CA) and two 10-port selection valves (VICI Valco Instruments, Houston, TX). The minimum flow rate of the system with the 1-mL syringe pump and limits set by the instrument control software is 1 mL s1. Except for the sample, all reagents were delivered from disposable 50-mL Falcon tubes (BD Biosciences, San Jose, CA). All tubing used in the system is high-purity Teflon perfluoroalkoxy (PFA) (0.03099 i.d., Upchurch Scientific, Oak Harbor, WA) except for the sample treatment cell, which is custom machined from polyether ether ketone (PEEK). The flow system is controlled with a laptop computer and custom instrument control software written using LabWindows/CVI (National Instruments, Austin, TX). The ultrasonic system consists of an integral power supply, an RF signal source and amplifier, and two ultrasonic transducers. These components were obtained from a commercial humidifier (ReliOn Ultrasonic Humidifier, Model Number FHC-501/H-0565-0). The transducers in the system were 5/899 (15.8 mm) in diameter and operated at a frequency of 1.4 MHz. The transducers and associated components were

powered with the integral 120 VAC power supply (Astrodyne) and required 40 W. Power to the transducers (on and off) was at the command of the Programmable Logic Controller (PLC) via a serial switch. The PLC coordinated the transducer operation and the fluidics operation, based on parameters supplied by a laptop computer running an in-house program. The back surface of the transducer chamber was cooled by a water-cooling system (Thermaltake Bigwater 12-cm liquid-cooling system). The PLC (Direct Logic 06) and PLC thermocouple module were both purchased from Automation Direct. The PLC thermocouple module was used to monitor the transducer water bath temperature so that the ultrasonic transducer could be automatically turned off if the water bath temperature exceeded 82 C. The thermocouple used to monitor the bath temperature was purchased from Omega. The watercooling system, the two valves, LED indicator lights, and serial switch (B&B Electronics four-port serial switch) required a total of 480 W (power supplies from Astrodyne). The complete system, including the controllers for the fluidic valves and pumps, was packaged in a single 1299  1399  2099 box (Fig. 1A). The ultrasonic transducer chamber consists of two transducers with faces parallel to the axis of the flow cell, all submerged in a water-filled chamber. The transducers are positioned, as shown in Figure 1C, to focus energy on the same region along the axis of the flow cell. The transducers’ faces are oriented at a 35 angle on the horizontal as depicted in Figure 1C, defining an angle of 70 from transducer to flow cell to transducer. Each transducer is on a pivot with  six degrees of movement to allow optimal focusing on the flow-through treatment cell. The overall transducer power density of 10 W cm2 was determined using a calibrated hydrophone (Specialty Engineering Associates, Soquel, CA). The calibrated hydrophone was positioned so that it measured the acoustic power present at the tube surface. The flow cell is centered in the region determined to have the greatest concentration of ultrasonic power. The cell is mounted to the removable lid of the transducer chamber to allow easy access while maintaining exact positioning of the cell for all subsequent samples. Float glass beads (0.1 mm, Biospec Products, Bartlesville, OK) are retained in the cell using small filters fabricated from a bio-inert nylon mesh material (Biodesign Inc., Carmel, NY) with a 75-mm pore size. A typical ultrasonic experimental run consists of the addition of 80 mL of spore suspension to a Teflon sample inlet chamber. The execution of a software routine initiates aspiration of the sample and the delivery of the sample through the tubing to the PEEK treatment cell. The sample is preceded by air and followed by a 200-mL air segment to ensure there is no mixing and subsequent dilution of the sample during ultrasonic treatment. The sample is dispensed in 8-mL segments at 100 mL s1, and each segment is held over the transducer for an operational ‘‘treatment’’ time, chosen to be 50 s. This approach was selected because continuous flow at the pump’s minimum flow rate could not provide these residence times. After the entire sample has been treated, it is delivered to JALA October 2009 279

Original Report a sample collection tube and a system cleaning procedure consisting of 10% bleach and sterile water rinses commences. Typically, 100 mL of bleach is delivered through the tubing at 1 mL s1, followed by 3  100 mL of sterile water rinses at 10 mL s1. This cycle is repeated twice. A 1 M sodium hydroxide wash is carried out to remove residual spore and nucleic acid material from the glass beads. The sodium hydroxide solution is left in the presence of the beads for 2 min with the ultrasonic transducers on to enhance removal of residual materials. The beads can be used over a period of several months with daily use; and analysis of water blanks immediately after sample runs showed no carryover of live spores, as determined by culturing the blanks on TSA plates. The carryover of amplifiable DNA with the described wash conditions was at least five orders of magnitude less than the sample concentration, as determined by real-time PCR, and hence was negligible for the current studies.

Controls Each sample and corresponding control set was processed in triplicate. The controls that were processed in the system under all the same conditions as the treated sample (but without the ultrasound), were processed with every experimental sample set to account for loss of spores and nucleic acid because of the potential for adsorption to the tubing and filters. The efficacy of each experiment was always determined based on the samples’ specific corresponding controls because even with these washed spore preparations there was a significant amount of extraneous DNA associated with the spore coat that could vary between spore preparations. In addition, we performed an additional control experiment to determine if the mechanism of increased DNA availability was because of heat, or the synergistic action of ultrasound and heat. It was not possible to isolate the effects of ultrasound alone because ultrasound at megahertz generates significant heat that cannot be efficiently dissipated. Similar to the controls above, we used a water bath to replicate the temperatures typically observed in our experiments, and exposed the spores to these conditions. Plate counts were used to determine spore viability and real-time PCR was used to determine increased DNA availability.

Quantitative Real-Time PCR DNA availability after treatment was determined using an Applied Biosystems 7500 real-time PCR system equipped with a fast cycling thermal block. The primers and probe, selected to target the recA gene of B. subtilis, were designed using Applied Biosystems’ Primer Express software v2.0: (forward, 59 -GCGCCCGAGGACTTAAATC-39 ; reverse, 59 -TCCATGTCAAGAAACCGCC-39 ; and probe, 59 6FAMCGTAAAGGGCAGCCCGCAAGTAAGA-TAMRA 39 ), to form a 66-bp amplicon. This gene was chosen based on our previous work where dilution-to-extinction PCR was used to test our previous bench top cylindrically focused ultrasound system.4 280

JALA

October 2009

Real-time PCR was performed in an optically clear 96-well plate designed for the fast cycling block. Each well had a final volume of 20 mLd5 mL of sample template, 4 mL of nuclease free water (Ambion, Austin, TX), 10 mL TaqMan Fast Universal Master Mix (2) (Applied Biosystems, Foster City, CA), 900 nM of the forward and reverse primer, and 250 nM of the fluorogenic probe (0.25 mL of each primer and 0.5 mL of probe, respectively). PCR cycling conditions were 95 C for 20 s, and 40 cycles at 95 C for 3 s and 60 C for 30 s. Fluorescence data were collected at the end of each 60 C step.

Analysis of Quantitative Real-Time PCR Data A standard curve was generated for each real-time PCR reaction performed. DNA was extracted from a known quantity of vegetative cells and a series of 10-fold dilutions over a 5Log10 range were amplified. Data were plotted on a semi-log scale with the x-axis representing the range of concentrations (as Log10 values) and the Cycle threshold (Ct, the cycle number where the instrument determines that a positive amplification event has occurred (e.g., a positive change in fluorescence from baseline of 0.1 (DRn))) on the y-axis. For the standard curve, the instrument generates a line corresponding to a linear relationshipdy ¼ mx þ b where m is the slope of the line, x is the calculated cell concentration for a sample, b is the Ct value corresponding to a theoretical DNA concentration of 0, and y is the Ct value (measured by the instrument for a given threshold) for a given cell concentration. In optimized real-time PCR reactions m is typically between 3.2 and 3.3, and this was the case for all real-time PCR experiments reported here. This slope represents a PCR efficiency of 100%. The absolute value of the slope represents the number of cycles that correspond to a 10-fold (1 Log10) change in DNA concentration. For ultrasonically treated spores and corresponding controls run in the same PCR reaction plate that showed positive amplification, the corresponding cell quantity (x, expressed as a Log10) was calculated based on the measured Ct value (y) and equation of the standard curve. For each treatment listed in Table 1, three separate experiments, with corresponding controls, were run. Within each run, three real-time PCR reactions were performed for a total of nine real-time reactions for each treated and control sample, respectively. For propagation of error analysis, the nine measurements for both the treated and control samples were averaged, and the standard deviation was calculated. The change in a treated versus control samples was calculated by subtracting the average Ct of the treated sample from the control. This generated the DCt value for the treatment run. Greater positive DCt values indicated more available DNA.

RESULTS

AND DISCUSSION

Lysis Module Design and Approach The module shown in Figure 1 delivers the sample through a flow cell in the ultrasonic field created by the

Original Report Table 1. Bead versus no bead ultrasonic treatments with different treatment times Treatment time (s) 15 30 50 90 120

DCta 0.9  0.04 2.3  0.02 1.2  0.3 1.9  0.4 1.7  0.2

%Viableb 34  15 32  9 92 92 17  5

Beads with ultrasound

15 30 50 90 120

2.3  0.3 2.0  0.2 4.7  0.2 3.1  0.5 3.0  0.7

41  14 33  11 12 51 31

High temperature, no ultrasound

50 800

0.2  0.4 0.07  0.4

!0.001 !0.001

Sample set No beads with ultrasound

a DCt represents the difference in the averaged Ct values of control samples and the averaged Ct values of treated samples. Greater DCt indicates better performance. b Viability calculated using a dilution series comparing the treated and control sample viabilities as measured by cell culturing. Standard propagation of error analysis was carried out on all PCR and viability data.

two transducers. This arrangement was selected to simulate the use of a cylindrically focused ultrasonic transducer using two inexpensive planar transducers instead. The region of greatest ultrasonic intensity along the cell length is w5 mm, which for the 2-mm i.d. tube corresponds to a 16-mL sample treatment region. The 80-mL samples were advanced through the cell in a series of stopped flow events to provide adequate residence times for lysis. The sample was advanced 8 mL at a time, and held for an operational ‘‘treatment’’ time before the next flow event. This treatment time was varied, as we shall describe in the following sections, using the automated sequential injection flow system. Comparing the 16-mL active region with the 8-mL steps gives an approximate residence time for each portion of the sample that was twice the operational ‘‘treatment’’ time for each stopped flow. Initial experiments advancing the sample 16 mL at a time were not as effective at lysing spores as the chosen 8-mL steps (data not shown). The new two-transducer design in Figure 1 was compared with a similar single transducer configuration in multiple experiments using equivalent times and sample volumes.

The two-transducer system was approximately five times faster than the single transducer configuration at achieving the same increase in DNA availability as measured by realtime PCR (data not shown).

Lysis Conditions and Results A number of parameters were examined to determine operational conditions for effective lysis of spores in the flow-through module. Our primary metric was the increase in spore DNA availability as measured by real-time PCR, comparing results from treated versus untreated spores to determine DCt (see Materials and Methods). Spore viability measurements made in parallel, verified spore death resulting from exposure to the lysis conditions. The residence times in these experiments were varied by the operational ‘‘treatment time’’ per step in the stopped flow protocol. These studies evaluated the following: (1) residence time of spores being exposed to ultrasound, (2) ultrasonic treatments with or without the addition of float glass beads, and (3) investigation of heat without the application of ultrasound. Table 1 summarizes these data from the three different conditions. Ultrasonic treatment, with and without beads, increases the DCt values, decreases the viability, and causes clear disruption of the spore coat. Figure 2 shows SEM images of spore samples treated with and without beads, as well as a control sample without ultrasonic treatment. The application of ultrasound always heats the sample, raising the possibility that heat rather than ultrasound causes spore lysis. This issue was not resolved in our prior study. As illustrated in Table 1, the effects of heat alone resulted in virtually all loss of spore viability. However, we measured no increase in DNA availability. These temperature control runs (heating the sample to approximately 100 C) were carried out at the lowest and highest possible exposure times that a spore would be subjected to in the ultrasonic treatment cell. Therefore, our studies demonstrate that the mechanism of DNA release is most likely the result of disruption of the spore coat at megahertz frequencies, consistent with spore damage observed by SEM. However, because we were unable to apply ultrasound without heating the sample, it is possible that this mechanism may also involve a synergistic action of ultrasound and heat.

Figure 2. SEM images of (A) spores treated with ultrasound in the presence of beads, (B) spores treated with ultrasound but without beads, and (C) spores run through the system without ultrasound or beads. The treatment period was 50 seconds for these samples. JALA October 2009 281

Original Report In terms of ultrasonic lysis with and without float glass beads, the data in Table 1 also show that the DCt values are generally larger for treatment with float-glass beads compared with treatments without beads. Among the experiments with beads, a 50-s treatment time for each step in the flow provides the largest DCt value, and the remaining spore viability is about 1%. At shorter treatment times, the DCt values are lower and the viability is higher, indicating less spore disruption and less available amplifiable DNA compared with 50 s of treatment. At longer treatment times, the viability remains in the low single digits, indicating a high degree of spore disruption, but the DCt values are less than at 50 s. Accordingly, the best treatment time for the 8-mL stepwise procedure was found to be 50 s when beads are present, yielding a DCt value of 4.7  0.2. This time provided the greatest increase in DNA as measured by real-time PCR, and also the lowest sample viability (1%) compared with other treatment times with beads (see Table 1). These treatment conditions were selected for further use. The improved availability of amplifiable DNA under these conditions is graphically illustrated in Figure 3, where the ultrasonically treated sample has a lower measured Ct compared with the control with no ultrasonic power applied under otherwise similar conditions. We also investigated the size of the DNA molecules resulting from the ultrasonic treatment conditions we selected. Spore suspensions were run on an agarose gel as described in the Materials and Methods section. The image of the gel after the run (Fig. 4) shows a smear of nucleic acid material in the 100e600-bp range compared with the lane with standards. Given the real-time PCR amplicon length of 66 bp, the degree of fragmentation of the released DNA does not significantly diminish the quantity of amplifiable fragments.

Increase in Amplifiable DNA Availability In prior work we used a qualitative method, dilutionto-extinction PCR, to show that ultrasonic lysis increased DNA availability after spore lysis. In preliminary experiments with the current transducer configuration, we also used this method and observed similar results to our prior work. Both configurations (i.e., the prior single cylindrically focused transducer, or the current dual transducer geometry) yielded, qualitatively, ca. 100-fold increases in available DNA. Real-time PCR provides more precise estimates of increased DNA availability than dilution-to-extinction PCR because the Ct values are related back to a standard curve for DNA concentration. For these experiments, a true 100-fold (2 Log10) increase would represent a DCt of approximately 6.6. Our greatest DCt was 4.7 for the 50-s treatment with float glass beads. Thus, it is reasonable to estimate that increased DNA availability was O10-fold, but !100-fold. More precisely, the change in Ct value represents a 26e27fold (24.7 ¼ 26) increase in available DNA.

DISCUSSION Thus far, there are only a few reports on flow-through ultrasonic lysis devices for spores. Our laboratory4 described a bench top prototype flow cell consisting of a tube positioned to receive energy from a cylindrically focused ultrasonic transducer operating at 1.4 MHz. Spore disruption, spore death, and increases in amplifiable DNA were demonstrated using B. globigii spore preparations.4 Borthwick et al.3,8 described a flow-through sonicator using a tubular piezoceramic transducer at 267 kHz. In experiments with B. subtilis var. niger (previously B. globigii or BG) it was reported that this device disrupted spore coats somewhat but

Figure 3. (A) Real-time PCR amplification plot and (B) standard curve for 3 replicate experiments for spores treated using ultrasonic lysis with float glass beads for 50 seconds vs. the controls which did not receive ultrasonic treatment. The standard curve is used to quantify and determine fold increase of spore DNA availability. Efficiency of approximately 100% indicates that the slope of the line is the number of PCR cycles required to achieve a 10-fold increase in DNA availability. 282

JALA

October 2009

Original Report described simulates the use of a cylindrically focused transducer. These transducers are inexpensive and the supporting electronics are simpler than the laboratory test and measurement systems we have created in the past.4 Dual transducers clearly performed spore lysis more rapidly than a similarly configured single transducer system. These studies have established empirical parameters for flowthrough lysis, clarified the energy (thermal alone vs ultrasonic and thermal) and conditions that led to increased amplifiable DNA, and quantified the increased DNA availability using real-time PCR. The observed lysis results are clearly because of the application of ultrasound because controls in the absence of applied ultrasound did not yield an increase in amplifiable DNA. In our system, the use of float-glass beads enhanced the availability of amplifiable DNA. Potential uses for the device extend from routine laboratory bench top extractions to integration into fielddeployable biodetection systems. This lysis device could be combined with other automated units to capture and purify biological components (cells, spores, and released DNA)4,16e23 that would also allow the removal of interferences and inhibitors, if so desired, before amplification and detection. A flow-through lysis method is especially important for automated monitors that must operate autonomously, where use of batch methods and consumable cartridges is not feasible. Few flow-though lysis devices have been demonstrated. We have now demonstrated a 26-fold increase in amplifiable DNA based on real-time PCR measurements, for a true flow-through lysis module.

ACKNOWLEDGMENTS This work was supported by the Chemical and Biological Countermeasures

Figure 4. Example gel electrophoresis of a treated spore sample to determine the range of DNA fragmentation. The lanes contain the following samples: (1) MW ladder and (2) spores ultrasonically treated in the system with beads present.

Division of the Department of Homeland Security Science and Technology Directorate. The authors thank Bruce Arey for the SEM images, which were performed in the Environmental Molecular Sciences Laboratory, a national scientific user facility sponsored by the Department of Energy’s Office of Biological and Environmental Research and located at Pacific Northwest National Laboratory. Pacific Northwest National Laboratory is operated

did not break up the spores or reduce spore viability. Their approach was successful in increasing the availability of spore-coat antigens for immunoassay measurements.3 However, it was not focused on the challenges of increasing DNA availability from spores in environmental samples. Marentis et al.7 described a microfluidic minisonicator consisting of a channel microfabricated in glass and bonded to a quartz plate with a microfabricated gold/ZnO/gold transducer. Driving the transducer at 367 MHz transmitted acoustic waves through the quartz into the channel. Commercial B. subtilis var. niger spore samples, used as received, were pushed through the channel and the treated samples were assessed with real-time PCR.7 The analysis of the real-time PCR data was unconventional and the increase in DNA availability did not approach 90%. In contrast, we have developed a stand-alone ultrasonic lysis module that includes a flow-through approach with dual transducers. The two transducers arranged as we have

by Battelle Memorial Institute for the US DOE under contract DE-AC0676RLO 1830.

REFERENCES 1. Belgrader, P.; Hansford, D.; Kovacs, G. T. A.; Venkateswaran, K.; Mariella, R.; Milanovich, F.; Nasarabadi, S.; Okuzumi, M.; Pourahmadi, F.; Northrup, M. A minisonicator to rapidly disrupt bacterial spores for DNA analysis. Anal. Chem. 1999, 71(19), 4232e4236. 2. Belgrader, P.; Okuzumi, M.; Pourahmadi, F.; Borkholder, D. A.; Northrup, M. A. A microfluidic cartridge to prepare spores for PCR analysis. Biosens. Bioelectron. 2000, 14(10e11), 849e852. 3. Borthwick, K. A. J.; Love, T. E.; McDonnell, M. B.; Coakley, W. T. Improvement of immunodetection of bacterial spore antigen by ultrasonic cavitation. Anal. Chem. 2005, 77(22), 7242e7245. 4. Chandler, D. P.; Brown, J.; Bruckner-Lea, C. J.; Olson, L.; Posakony, G. J.; Stults, J. R.; Valentine, N. B.; Bond, L. J. Continuous spore disruption using radially focused, high-frequency ultrasound. Anal. Chem. 2001, 73(15), 3784e3789.

JALA October 2009 283

Original Report 5. Luna, V. A.; King, D.; Davis, C.; Rycerz, T.; Ewert, M.; Cannons, A.; Amuso, P.; Cattani, J. Novel sample preparation method for safe and rapid detection of Bacillus anthracis spores in environmental powders and nasal swabs. J. Clin. Microbiol. 2003, 41(3), 1252e1255. 6. Taylor, M. T.; Belgrader, P.; Furman, B. J.; Pourahmadi, F.; Kovacs, G. T. A.; Northrup, M. A. Lysing bacterial spores by sonication through

15. Nicholson, W. L.; Setlow, P. Sporulation, Germination and Outgrowth. Wiley: Chichester, UK; 1990. 16. Bruckner-Lea, C. J.; Stottlemyre, M. S.; Holman, D. A.; Grate, J. W.; Brockman, F. J.; Chandler, D. P. Rotating rod renewable microcolumns for automated, solid phase DNA hybridization studies. Anal. Chem. 2000, 72(17), 4135e4141.

a flexible interface in a microfluidic system. Anal. Chem. 2001, 73(3),

17. Bruckner-Lea, C. J.; Tsukuda, T.; Dockendorff, B.; Follensbee, J. C.;

492e496. 7. Marentis, T. C.; Kusler, B.; Yaralioglu, G. G.; Liu, S.; Haeggstrom, E. O.;

Kingsley, M. T.; Ocampo, C.; Stults, J. R.; Chandler, D. P. Renewable microcolumns for automated DNA purification and on-line amplifica-

Khuri-Yakub, B. T. Microfluidic sonicator for real-time disruption of eu-

tion: from sediment samples through PCR. Anal. Chim. Acta 2002, 469(1), 129e140. 18. Chandler, D. P.; Brown, J.; Call, D. R.; Grate, J. W.; Holman, D. A.;

karyotic cells and bacterial spores for DNA analysis. Ultrasound Med. Biol. 2005, 31(9), 1265e1277. 8. Borthwick, K. A. J.; Coakley, W. T.; McDonnell, M. B.; Nowotny, H.;

Olson, L.; Stottlemyer, M. S. Continuous, automated immunomagnetic

Benes, E.; Groschl, M. Development of a novel compact sonicator for

separation and microarray detection of E. coli O157:H7 from poultry

cell disruption. J. Microbiol. Methods 2005, 60(2), 207e216. 9. Fykse, E. M.; Olsen, J. S.; Skogan, G. Application of sonication to release DNA from Bacillus cereus for quantitative detection by real-time PCR. J. Microbiol. Methods 2003, 55(1), 1e10. 10. Joyce, E.; Phull, S. S.; Lorimer, J. P.; Mason, T. J. The development and evaluation of ultrasound for the treatment of bacterial suspensions. A study of frequency, power and sonication time on cultured Bacillus species. Ultrason. Sonochem. 2003, 10(6), 315e318.

carcass rinse. Int. J. Food Microbiol. 2000, 70, 143e154. 19. Chandler, D. P.; Brown, J.; Call, D. R.; Wunschel, S.; Grate, J. W.; Holman, D. A.; Olson, L.; Stottlemyre, M. S.; Bruckner-Lea, C. J. Automated immunomagnetic separation and microarray detection of E. coli O157:H7 from poultry carcass rinse. Int. J. Food Microbiol. 2001, 70(1e2), 143e154. 20. Chandler, D. P.; Holman, D. A.; Brockman, F. J.; Grate, J. W.; Bruckner-Lea, C. J. Renewable microcolumns for solid-phase nucleic acid

11. More, M. I.; Herrick, J. B.; Silva, M. C.; Ghiorse, W. C.; Madsen, E. L.

separations and analysis from environmental samples. Trends Anal.

Quantitative cell lysis of indigenous microorganisms and rapid extraction of microbial DNA from sediment. Appl. Environ. Microbiol. 1994,

Chem. 2000, 19, 314e321. 21. Chandler, D.P.; Posakony, G.J.; Bond, L.J.; Bruckner-Lea, C.J.

60(5), 1572e1580. 12. Waters, L. C.; Jacobson, S. C.; Kroutchinina, N.; Khandurina, J.;

Apparatus and method for ultrasonic disruption of a liquid. US patent 7,022,505, 2006.

Foote, R. S.; Ramsey, J. M. Microchip device for cell lysis, multiplex

22. Chandler, D. P.; Schuck, B. L.; Brockman, F. J.; Bruckner-Lea, C. J.

PCR amplification, and electrophoretic sizing. Anal. Chem. 1998,

Automated nucleic isolation and purification from soil extracts using

70(1), 158e162.

renewable affinity microcolumns in a sequential injection system.

13. Ruzicka, J. Tutorial review. Discovering flow injection: journey from sample

Talanta 1999, 49, 969e983.

to a live cell and from solution to suspension. Analyst 1994, 119, 1925e1934. 14. Ruzicka, J.; Pollema, C. H.; Scudder, K. M. Jet ring cell: a tool for flow

23. Grate, J. W.; Bruckner-Lea, C. J.; Jarrell, A. E.; Chandler, D. P. Automated sample preparation method for suspension arrays using renew-

injection spectroscopy and microscopy on a renewable solid support.

able surface separations with multiplexed flow cytometry fluorescence

Anal. Chem. 1993, 65(24), 3566e3570.

detection. Anal. Chim. Acta 2002, 478, 85e98.

284

JALA

October 2009