MICROBIAL IDENTIFICATION SYSTEM. The Automated Solution for Clinical & Environmental Labs

Transcription

1 MICROBIAL IDENTIFICATION SYSTEM The Automated Solution for Clinical & Environmental Labs

2

3 The MIDI Sherlock Microbial Identification System Automated Microbiology for the 21 st Century For over 20 years, clinical and environmental laboratories have relied on the automated Sherlock Microbial Identification System as an automated solution for the identification of bacteria. The Sherlock System represents an accurate, easy-to-use, low-cost solution for the identification of over 1,500 species of bacteria, including the 6 major bacterial agents of bioterrorism classified by the U.S. Centers for Disease Control (CDC). What makes the Sherlock System unique is that all bacteria are analyzed using the same procedure with no offline tests, gram stains or biochemical cards. Technology More than 300 fatty acids and related compounds have been found in the microbial cell membrane. The wealth of information contained in these compounds comes from both the qualitative differences (usually at genus level) and quantitative differences (commonly at species level). The Sherlock System identifies bacteria based on gas chromatographic (GC) analysis of extracted microbial fatty acid methyl esters (FAMEs). The fatty acids are extracted from pure cultures by a procedure that consists of saponification in dilute sodium hydroxide/methanol solution followed by derivatization with dilute hydrochloric acid/methanol solution to give the respective FAMEs. The FAMEs are then extracted from the aqueous phase by the use of an organic solvent. The extract is then analyzed by GC and sophisticated pattern recognition algorithms are used to match the unknown microbial GC-FAME profiles to the Sherlock libraries. The Sherlock software automates all analytical operations. Microbial Libraries Microbial fatty acid profiles are very unique from one species to another, and this fact has enabled MIDI to create large microbial libraries, as well as tools for tracking individual strains. The Sherlock bacterial libraries have more than 1,500 species entries. Some of the species that are discriminated well using the Sherlock System include those of Bacillus, Burkholderia, Pseudomonas, Gram-positive cocci and rods (such as coryneforms), Gram-negative non-fermenters (such as Acinetobacter) and unusual environmental and clinical organisms (including 6 bacterial agents of bioterrorism). MIDI Sherlock Microbial Identification System with the Agilent Technologies 6850 Model GC. Proven Accuracy The Sherlock System is an established microbial identification technology with proven accuracy over a wide range of bacteria. The Sherlock System has over 400 peer-reviewed journal articles to its credit and is a CDC Official Method for identification of aerobic bacteria (Method # 0801). For confirmation of Bacillus anthracis (anthrax pathogen), the Sherlock System has 2 distinctions: it is a U.S. Department of Homeland Security Official Method (AOAC Method # ) and is a U.S. FDA cleared method for B. anthracis confirmation. Other Key Features/Benefits of the Sherlock System Rapid Analysis sample extract to identification under 10 minutes Easy-to-use automated analysis, straightforward sample preparation, no GC experience needed Standardized one procedure for all bacteria with no offline tests or gram stains needed Common Reagents no proprietary test cards or media Inexpensive under $2.50 per sample in consumables Productivity Tools easy-to-use strain tracking and analysis tools Regulatory Compliance timesaving validation package and support for 21 CFR Part 11 Safety sample procedure kills the organism which increases user safety New in Sherlock System adds DNA Sequencing Capabilities Beginning in 2006, MIDI has incorporated DNA Sequence analysis into the Sherlock System. As a result, Sherlock users will now be able to identify unknown isolates by 16S (bacteria) and 28S (fungi/yeast) rrna sequence analysis in addition to the standard GC-FAME approach. The addition of DNA Sequencing will allow Sherlock users to created combined phenotypic (GC-FAME) gentotypic (DNA Sequencing) bacterial reports, an industry first. The addition of DNA Sequencing technology to the Sherlock System represents MIDI s continued pursuit to develop the most advanced microbial identification solutions available for the clinical and environmental markets.

4

5 Sherlock 6.0 Microbial Identification System Specification Sheet General Description The Sherlock Microbial Identification System identifies bacteria and yeast by gas chromatographic (GC) analysis of fatty acid methyl esters (GC-FAME). The Sherlock software, methods and libraries are combined with an Agilent Technologies 6890 or 6850 GC and Agilent ChemStation software for a complete automated microbial identification solution. Sherlock s pattern recognition algorithms, combined with its calibration mixture, standardize each instrument. This virtually eliminates the manual calibration adjustments associated with a GC. No chromatography knowledge or experience is required. Microbe Libraries Sherlock methods and libraries are available for the following applications. Environmental Aerobes The environmental library contains 695 species. The standard culturing media used is TSBA. The standard incubation is 28 o C for 24 hours. Clinical Aerobes The clinical library contains 430 species. Blood agar is the standard culturing media. The standard incubation is 35 o C for 24 hours. Anaerobes Two anaerobe libraries are available. One is for BHIBLA plate-grown anaerobes (135 species). The other is for PYG broth-grown anaerobes (590 species). Yeast The yeast library has entries for 190 species grown at 28 o C on Sabouraud Dextrose Agar for 24 hours. Bioterrorism/Biowarfare This library identifies 6 major bacterial agents of bioterrorism/ biowarfare, plus 30 closely related species. Combined with the Clinical Aerobe library, the Bioterrorism library is a powerful tool for confirming bacterial threat agents (Developed with the U.S. Army Medical Research Institute for Infectious Diseases, USAMRIID, Fort Detrick, MD). Low Costs Per Sample It costs under $3.00 USD per sample for all consumables. This includes reagents, gases, calibration standards, glassware, and culture media. Instrument Throughput Following a short preparation procedure (typically done in batches), the sample vials are loaded into the instrument s autosampler. The automated system takes over and quickly analyzes each sample. No additional incubation is needed at this point. Standard methods process 2 samples per hour on single channel instruments and 4 samples per hour on dual channel instruments. Rapid methods for Environmental Aerobes, Clinical Aerobes and Bioterrorism bacteria process 6 samples per hour on a single channel and 12 samples per hour on a dual channel instrument. This method has 2 times the detection sensitivity of the Standard methods. Sensitive methods for Anaerobes and Yeasts process 2 samples on a single channel and 4 samples on a dual channel instrument per hour. This method has 2 times the detection sensitivity of the Standard methods, and uses the same calibration standard as Rapid methods. Culturing Like all widely used confirmatory techniques, Sherlock requires pure microbial cultures. Using standard laboratory techniques, a primary isolation plate is incubated for 24 hours for a typical sample. If the primary plate appears to be a single organism, a small cross-section of cells is harvested and incubated for 24 hours on a secondary plate. If the primary plate appears to be a mixed culture, a colony or each type may be subcultured. Slow growing organisms will require longer incubation times. MIDI, Inc. Newark, Delaware Sample Preparation Using inexpensive reagents, available from almost any chemical supply house, a technician averages only 5 minutes per sample to prepare a batch of 30 samples. Each sample is prepared for analysis using a liquidliquid extraction in a single test tube. Harvesting a small quantity of cells from the culture plate is the most labor-intensive step. It will typically take 1 hour or less to harvest cells from 30 plates into 30 test tubes. The four-step liquid-liquid extraction process requires about 1½ hours or less for a batch of 30 samples. During the extraction process, approximately 35 minutes of wait time are available for the technician to do paper work and other tasks. The same sample preparation is used on all samples. It is not necessary to do a Gram stain or other offline tests before preparing and analyzing a sample. Bio-Safety Bio-Safety is enhanced because live organisms are not introduced into the instrument. The first step of the extraction procedure treats the cells with a sodium hydroxide solution for 30 minutes in a 100 o C water bath. After the first step, the technician is no longer working with live organisms. Laboratories that handle dangerous pathogens will typically perform the sample extraction in a BSL-3 lab and transfer decontaminated extracts to a non BSL-3 lab for instrument analysis. This allows the instrument to be maintained and serviced by technicians outside the BSL-3 lab.

6 Instrumentation A Sherlock system is composed of a Windows XP or 2000 based computer loaded with the MIDI Sherlock software and an Agilent ChemStation. The computer is interfaced to one of the following instruments: Agilent 6850 GC with single column and a 27-vial autosampler. Agilent 6890 GC with a single column and a 100-vial autosampler. Agilent 6890 GC with dual columns and a 100-vial autosampler. Analysis Relationships between samples can be explored using: Dendrogram plots Neighbor-joining trees Principal component analysis (PCA) with 2-D plots and histograms The graphics can be exported to Microsoft Office and other packages for further analysis and for research publications. Custom Libraries Using the optional Library Generation System (LGS), custom libraries can be created from your samples. Uses for custom libraries: Quality control of proprietary strains used in production processes Quickly recognize contaminants that reoccur in a facility or process Assign an identity to organisms that do not have a published taxonomy Catalog culture collections Research Data Export Data Export software exports sample data, fatty acid profiles, library match results, and other information to Excel spreadsheets and Access databases. There are many applications for custom reports and calculations created using Excel, Access, and other data analysis tools: Trend analysis Custom reports Summary reports for sample sets Microbe population studies Research and publications Data mining Strain Tracking Tracker/Cluster is an optional strain tracking package. It is a powerful tool that helps locate the source of a contamination. Tracker locates other samples that are likely to be the same strain as a sample of interest. Tracker searches for strain matches between the current sample and all previous samples. Uses include: Trend analysis Summary reports for sample sets Microbe population studies Research and publications Data mining Cluster, is a new module that automatically finds groups (clusters) of highly related samples. Relationships between clusters and samples can be explored using: Dendrogram plots Neighbor-joining trees 2-D Color-coded PCA plots Tracker and Cluster operate independently of sample identification, allowing unknown samples to be compared. 21 CFR Part 11 The optional Electronic Records and Signatures (ERS) package can be added to support compliance with FDA regulation 21 CFR Part 11. Provides access control based on Windows XP or Windows 2000 user passwords and group settings to authenticate users and determine their privileges. When configured with a Windows XP or Windows 2000 domain, users can be authenticated using their domain credentials. Sherlock requires the user to logon before granting access to controls and data. Packages all associated data, audit trails, logs and results in a secure electronic vault for storage on the local disk or on a remote file server. Supports two levels of electronic signature authority with notes added by the signer. Automatic inactivity logoffs. Supports security policies defined for the Windows XP/2000 system or domain, including lockout of accounts after a predetermined number of login failures. Sherlock DNA Sherlock DNA is an optional package, which allows for identification and analysis of microbial DNA sequences. Sherlock DNA comes with 16S rrna gene sequence libraries for bacterial identification and an optional 28S rrna add-on for fungi/yeast identification. The system is not limited to a specific DNA product. Custom libraries can be created from your samples. In addition Sherlock DNA makes it possible to have a combined fatty acid- DNA sequence report for a polyphasic approach to identification. Markets Using Sherlock Animal Science Bioremediation Biodefense / Bioterrorism Clinical Diagnosis Dental Research Entomology Epidemiology Food Microbiology Marine Science Medical Research Pharmaceutical QC Plant Pathology Soil Science Water Quality Taxonomy Studies The information in this publication is subject to change without notice. MIDI & MIDI Authorized distributors are the sole source for the Sherlock Microbial Identification System. Copyright 2005 MIDI, Inc. All Rights Reserved Date: 20 December 2005 MIDI, Inc. 125 Sandy Drive Newark, Delaware Phone: Fax: sales@midi-inc.com

7 Sherlock Tracker/Cluster Strain Epidemiology With Real-Time Results MIDI, Inc 125 Sandy Drive Newark DE (302) Scan thousands of samples automatically for closest matches Track strains quickly and easily-even unknowns Create your own data sets to track against No prior knowledge of the sample is required Track samples automatically after each run or in offline mode Detect contaminants and identify nosocomial outbreaks faster Tracker Fact Sheet Where and When have I seen this bacteria before? Description MIDI Inc. s Sherlock Microbial Identification System (MIS) has provided strain tracking as a manual capability of Sherlock s advanced Library Generation Software (See Technical Note #102). Sherlock Tracker is an add-on module for the Sherlock 4.5 and 6.0 versions that enhances Sherlock s strain-tracking ability. Tracker automatically compares the cellular fatty acid profile of a sample to a database of previously analyzed samples. A report is generated of the samples that are most similar. Tracker can track samples, whether or not the samples match known bacterial species in MIDI s extensive libraries. Comparisons can also be made automatically after each sample run, so that you will know immediately if you have seen the sample before. Database Creation With a few mouse clicks, the user can build a Tracker database. Data from Sherlock samples, even samples run on previous versions of Sherlock, may be quickly combined to create a Tracker database for searching. New samples may be automatically added to the database as Sherlock runs them. Online Tracker Reports Tracker may be configured to automatically generate reports as each Sherlock sample completes. Offline Investigation with Tracker Using the Tracker mode in Sherlock s CommandCenter, samples may be evaluated against Tracker databases, and new reports generated. Tracker Example In Figure 1, Tracker was used in house to compare the strain of anthrax found in Connecticut to MIDI s BIOTER1.0 library anthrax entries. The strain matched with the Ames strain at a Cutoff Factor of 3.000, indicating the same strain. Inquiry Match Figure 1- Tracking Anthrax in the Connecticut Case- Data courtesy of the Connecticut Department of Public Health Laboratory

8 Sherlock Tracker/Cluster Strain Epidemiology With Real-Time Results MIDI, Inc 125 Sandy Drive Newark DE (302) Introducing Cluster Cluster Fact Sheet Description Cluster is a new analytical tool available in Sherlock 6.0 with the Tracker option. Cluster automatically creates groups of similar samples. Two separate clustering algorithms are used allowing the user to easily create clusters and determine the relationship between these clusters. Uses Quality Control trending and tracking. In some cases a particular type of bacteria is seen consistently within a pharmaceutical process. Clustering allows identification of samples that fall into the group; Samples that fail to match the group may indicate a change in the pharmaceutical process. Figure 2 - Cluster Detail View Identification of new bacterial species. A cluster of closely related samples may indicate the existence of a new species. These samples can form the backbone for delineating the new species. Location of unexpected patterns in data. A cluster may contain a wholly unanticipated connection among samples, such as day of week, time of year, or processing technician. Once the cluster has been defined, the analyst can evaluate the associated samples to identify such patterns. Figure 3 - Principle Component Plot of Largest Clusters

9 Sherlock Data Export Advanced Sherlock Data Management Fact Sheet How do I create a customized report from my Sherlock identification data? Export Sherlock data to spreadsheets and databases with ease Generate custom reports for your facility or project Automatically gather and monitor calibration performance Description Sherlock Data Export is an optional extension to MIDI Inc. s Sherlock Microbial Identification System (MIS). Data Export can be used to export Sherlock information to a database or a spreadsheet. Once the information is exported to these industry-standard files, an array of analysis, reporting and charting tools may be used. Data Export creates Microsoft Access databases and Excel spreadsheets. Microsoft Office is not required to use Data Export- many packages accept databases and spreadsheets in these formats- however, to get the full value from Data Export, Microsoft Office is suggested. Customized Reports Data Export allows the Sherlock user to make customized reports, especially with the use of Excel and Access programs. Examples of customized reports (Figures 1-4) include: Calibration performance, hydroxyl fatty acid recoveries, best match reports and Similarity Index (SI) plots. Export 2-D plots to spreadsheets for publication purposes Figure 1- Sherlock Calibration Performance- this report gives a summary of calibration performance for each sequence run. Changes to the query and report could be made to report, for example, weekly calibrations instead of all sequences, or to include more or less information. Automatically monitor injection port liner and column performance to minimize downtime Hydroxyl Percentages Percent /5/01 6/19/01 7/3/01 7/17/01 7/31/01 8/14/01 8/28/01 9/11/01 10:0 2OH 16:0 2OH Low Percent Figure 2- Hydroxyl Fatty Acid Recoveries- the following chart shows percentages of 10:0 2OH and 16:0 2OH in calibration runs over time. In this case, the hydroxyl percentages stay above the cutoff of 2%. This type of report can be used to monitor injection port liner and column performance. Date

10 Figure 3- Result Summaries- with Data Export, a report can be created that has one line per sample with specific information selected for that sample, such as the the best match. Date ID_Nbr Samp_Id %Named Sim_Index Name 9/26/ UN-MIDI-B-1( Bacillus-megaterium-GC subgroup 9/26/ UN-MIDI-B-1( Micrococcus-luteus-GC subgroup 9/26/ UN-MIDI-B-1( Brevibacterium-linens* 9/26/ UN-MIDI-B-1( Bacillus-cereus-GC subgroup A* 9/26/ UN-MIDI-A( B Acinetobacter-radioresistens 9/26/ UN-MIDI-A( Acinetobacter-baumannii 9/26/ UN-MIDI-A( A Acinetobacter-radioresistens 9/26/ UN-MIDI-A( A Stenotrophomonas-maltophilia* 9/26/ UN-MIDI-A( F Stenotrophomonas-maltophilia* 9/26/ UN-MIDI-A( Paenibacillus-gordonae* Figure 4- Summary Statistics- in this example, a set of QC samples, all identified correctly as Stenotrophomonas maltophilia, is exported to Microsoft Excel using Data Export. The Similarity Index (SI) over time is plotted to look for any unusual trends. MIDI, Inc. Newark, Delaware The information in this publication is subject to change without notice. Copyright 2002 MIDI, Inc. All Rights Reserved Date: 03 November 2002 MIDI, Inc. 125 Sandy Drive Newark, Delaware Phone: Fax: midi@midi-inc.com

11 Sherlock Rapid Methods The Same Results 3x Faster General Description Rapid methods with corresponding libraries (databases) get the same results three times faster than their standard method counterparts. Rapid methods are 2.5 times more sensitive, reducing the number of cells needed for the analysis. Advantages Same day identification of isolates Samples requiring reanalysis can be processed immediately Fewer cells used for analysis reduces culturing time for slow growing organisms Easier harvesting of cells for analysis reduces labor Reduced injection port maintenance due to less material being injected Improved results for organisms containing mycolic acids Rapid Chromatography The rapid methods take advantage of the advanced electronic flow control features of the Agilent 6890 and 6850 gas chromatographs to obtain the same results using the same GC column in a fraction of the time. Figure 1 shows the chromatogram and Sherlock library search results for a Bacillus subtilis analyzed with the standard chromatographic method. Identical results are obtained much faster using the rapid method (figure 2). The two chromatograms are nearly identical except for the different time scales. The peak that elutes at minutes using the standard method elutes at minutes using the rapid method. The analytical run is 3.9 times faster. The total throughput, which includes the time for the GC to prepare for the next run, is approximately 3 times faster. System Requirements The rapid methods will not work with the 5890 GC or older versions of Sherlock. The following are required: Agilent 6890 or 6850 GC Sherlock version 4.5 or higher pa FID1 A, (A D) Matches: Library Sim Index Entry Name TSBA Bacillus-subtilis Bacillus-atrophaeus pa Figure 1- Standard Method for a B. subtilis sample *FID1 A, (A D) Fact Sheet Matches: Library Sim Index Entry Name RTSB Bacillus-subtilis Bacillus-atrophaeus Figure 2- Rapid Method for the Same B. subtilis m m

12 Rapid Libraries When the same sample is analyzed with both the Rapid and Standard methods, the match results for the Rapid library are nearly identical to those for the Standard library. The Rapid libraries were generated from the same data as the Standard libraries. The Rapid methods and libraries improve results for organisms, containing mycolic acids. Cleavage of mycolic acid side chains may result in fragments being randomly named as fatty acids by the Standard methods. Chromatography improvements let the Rapid method and library mostly ignore these fragments. Calibration Standard With the increased sensitivity, the Rapid methods require a new calibration standard. The Rapid Method Calibration Mix (MIDI part no A) is packaged in distinctive brown glass ampoules. Compatibility The Rapid methods use the same analytical column as the standard methods. This allows one instrument to run both Rapid and standard methods. Standard method samples can be mixed, in the same sequence, with Rapid method samples. The electronic pressure and flow control of the 6890 and 6850 allow Sherlock to establish the correct operating conditions for each method. Validation MIDI followed in-house procedures to validate the Rapid methods and libraries. The Rapid methods and libraries ship with a certificate of validation. Validation documents are available for inspection at MIDI s office in Newark, Delaware. MIDI, Inc. Newark, Delaware Operation Harvest only ca. 20mg wet weight of cells (2 to 2.5 times less than the 40 to 50mg for the standard method). Fewer cells should be used to avoid overloading the more sensitive GC analysis. The remainder of the sample preparation procedure is identical to the standard method. The same reagent volumes, reaction temperatures, and times are used. Enter the calibration mix and samples into the Sherlock Sequencer s sample table. Select the appropriate Rapid Method (RTSB50, RCLN50, or RBTR50) rather than one of the standard methods. Samples using standard methods (e.g. the Moore anaerobe method), with their calibration standards, can be entered in the same sequence. The analytical speed allows time during the workday for operators to review results and reanalyze samples that require dilution or concentration. Available Methods Rapid methods are currently available for the following applications: Environmental Aerobes The environmental library contains 690+ species. The standard culturing media used is TSBA. The standard incubation is 28 o C for 24 hours. Clinical Aerobes The clinical library contains 410+ species. Blood agar is the standard culturing media. The standard incubation is 35 o C for 24 hours. Bioterrorism This library identifies the 6 major bacterial agents of bioterrorism plus another 15 closely related organisms. Combined with the Clinical Aerobe library, the Bioterrorism library is a powerful tool for confirming a biological attack. (Developed with the U.S. Army Medical Research Institute of Infectious Diseases, Fort Detrick, MD, the Bioterrorism library is available free of charge to qualified owners of a MIDI Sherlock system.) The information in this publication is subject to change without notice. Copyright 2002 MIDI, Inc. All Rights Reserved Date: 31 October 2002 MIDI, Inc. 125 Sandy Drive Newark, Delaware Phone: Fax: midi@midi-inc.com

13 Sherlock ERS Electronic Records and Signatures Fact Sheet Is your Bacterial Identification System 21 CFR Part 11 Compliant? FDA 21 CFR Part 11 compliant software Data from each batch is stored in a single, secure file User authentication is required at key operational points Electronically sign each report and attach comments Create an audit trail of all entries and actions on an ERS file Description Sherlock Electronic Records and Signatures (ERS) is an optional extension to MIDI Inc. s Sherlock Microbial Identification System. Sherlock ERS permits controlled data access to authorized personnel. A core component to 21 CFR Part 11 compliance is the ability to link all related transactions to an electronic signature and to track any associated changes made to a secure file. Sherlock ERS preserves the data, methods, libraries and audit trail from a Sherlock sequence in a single, secure file. The Sequence Audit Log details all of the actions taken while the sequence was run, showing the date, the user name and the activity. Changes after the data have been collected, such as approval or rejection of a sample, are detailed in the Audit Log, which also displays the date, name and associated action. The Audit Log shows the old and new values for each change. Security In order to comply with the FDA s requirement that an electronic signature must employ at least two distinct identification components, such as a username and password, Sherlock ERS requires this information at every key operational point in the system. Sherlock ERS is integrated with the Windows NT and 2000 security models. Authorized ERS users and passwords are determined by the local or network system administrator, depending on the environment that the Sherlock system is employed in. Figure 1- Sherlock ERS View Regenerate original reports and view chromatograms ERS View Sherlock ERS permits review and verification of electronic records through a Windows -based interface. Authorized users are able to view sample information, reports, chromatograms and audit logs. In addition, select changes can be made to ERS files, depending on the level of authority the user has.

14

15 Sherlock Bioterrorism Library U.S. Department of Homeland Security Standard Method Fact Sheet The ONLY commercially available microbial identification system DHS-approved for confirmation of Bacillis anthracis Background The U.S. anthrax attacks in 2001 highlighted the need for accurate and reliable methods to identify Bacillus anthracis (the bacterium that causes anthrax), in order to better prepare the country against a future attack. The Department of Homeland Security (DHS), recognizing the need for a 3 rd party independent evaluation of existing anthrax detection and confirmation methods, partnered with AOAC INTERNATIONAL in June 2003 to fund a comprehensive validation of these methods. As a result of the 18-month long study, the MIDI Sherlock Microbial Identification System (using the Sherlock Bioterrorism Library) was the only commercially available system approved for confirmation of B. anthracis. The MIDI Sherlock System has since been granted AOAC Official Methods of Analysis SM status for confirmation of B. anthracis (AOAC # ), and is now the only microbial identification system in the world to have this recognition. Technology The MIDI Sherlock System is a laboratory-based system, with identifications done from pure culture. Cellular fatty acids (CFA) from bacteria are converted to fatty acid methyl esters (FAME) and then separated by gas chromatography (GC). Pattern recognition software is used to compare each chromatogram to MIDI s extensive microbial libraries, in order to identify the organism. Fatty acid patterns are so unique that they can often be used for strain tracking purposes as well. Industries using the MIDI Sherlock System include: biopharmaceutical companies, government agencies, hospitals, military installations, public health departments and university labs throughout the world. Bioterrorism Library The Sherlock Bioterrorism Library was co-developed by MIDI, Inc and the U.S. Army Medical Research Institute of Infectious Diseases (Fort Detrick, MD). The library consists of the following entries: Bacillus anthracis (Anthrax) Brucella melitensis (Brucellosis) Burkholderia mallei (Glanders) Burkholderia pseudomallei (Melioidosis) Francisella tularensis (Tularemia) Yersinia pestis (Plague) 15 near-neighbor species Proven Utility for Biodefense The MIDI Sherlock System is a laboratory-based system that can be used on a routine basis to identify commonly isolated bacteria from clinical and environmental sources, while at the same time proactively screening for potential agents of bioterrorism. Suspicion of a bioterrorism agent is not required for effective use of the system. With AOAC Official Methods SM status, laboratories will now be able to obtain a highly accurate identification system that can be used for both routine bacterial identifications and potential bioterrorism attacks. Confirmed Accuracy Validated method (AOAC # ) for confirmation of Bacillus anthracis Dual-Use Identifies 6 major bacterial bioterrorism agents plus 1,500 other bacterial species Safety Bacteria are killed during the sample preparation Inexpensive Under $3.00 per sample for consumables & common laboratory reagents High Throughput More than 200 samples per day Easy to operate No chromatography experience needed

16 Figure 1- Identifying Anthrax in the Connecticut Case- The Sherlock Bioterrorism Library was used to identify the anthrax bacterium from the 94-year-old woman in the Connecticut anthrax case in November The following report confirmed Bacillus anthracis in this case. The data is courtesy of the Connecticut Department of Public Health Laboratory. Created: 11/20/2001 9:24:11 AM Sample ID: K-BACI-ANTHR A (Connecticut case Calc. Method: BIOTER RT Response ECL Peak Name Percent E SOLVENT PEAK :0 ISO :0 ISO : :0 ISO :0 ANTEISO :1 ISO I :0 ISO :1w7c : ISO 17:1 w5c :1 ANTEISO A :0 ISO :0 ANTEISO :2 w6,9c :1 w9c 0.86 Matches: Library Sim Index Entry Name BIOTER Bacillus-anthracis-GC subgroup A The information in this publication is subject to change without notice. Copyright 2005 MIDI, Inc. All Rights Reserved Date: 20 May 2005 MIDI, Inc. Newark, Delaware MIDI, Inc. 125 Sandy Drive Newark, Delaware Phone: Fax: sales@midi-inc.com

17 Bacterial Bioterrorism Agent References May 2005 Bacillus anthracis (Anthrax) Abshire, T.G., Brown, J.E., Teska, J.D., Allan, C.M., Redus, S.L. and Ezzell, J.W. (2001). Validation of the use of gamma phage for identifying Bacillus anthracis. Abst 4 th Intl Conf Anthrax, p.30. Arnold, C.S., Jackoway, G., Mayo D. and Teska, J. (2003). Automatic Strain Identification of Bacillus anthracis by Fatty Acid Analysis. Abst 2003 ASM Biodefense Research Meeting, p 54. Bernard, K. (2002). The use of the BIOTER TM Data library as adjunct to the characterization of bacterial agents of Bioterrorism (BT). Abst 102 nd Gen Mtg Amer Soc Microbiol, p.101. Kim, W.Y., Song, T.W., Park, J. Y., Kim, C.M., Chung, S.I. and Choi., C.S. (2000). Analysis of cellular fatty acid esters (FAMEs) for the identification of Bacillus anthracis. J Kor Soc Microbiol 35, Osterhout, G., Allan, C., Redus S., Teska J., Romagnoli M., Dick J. and Sasser, M. (2002). Differentiation of Bacterial Agents of Bioterrorism From Routine Clinical Isolates Using Cellular Fatty Acid Analysis. Abst Intl Conf Emerg Infect Dis, p 34. Sasser, M., Kunitsky, C., Jackoway, G., Ezzell, J.W., Teska, J.D., Harper, B. and Parker, S. (2005). Identification of Bacillus anthracis from culture using gas chromatographic analysis of fatty acid methyl esters- AOAC Official Method SM J AOAC INTL 88 (1), Song, Y., Yang, R., Guo, Z., Zhang, M., Wang, X. and Zhou, F. (2000). Distinctness of spore and vegetative cellular fatty acid profiles of some aerobic endospore-forming bacilli. J Microbiol Meth 39, Teska, J.D., Allan, C.M., Redus, S.L., Coyne, S.R. & Ezzell, J.W. (2001). Identification of Bacillus anthracis using MIDI whole cell fatty acid analysis. Abst 4 th Intl Conf Anthrax, pp Teska, J.D., Coyne, S.R., Hartman, L.J., Allan, C.M., Christensen, D.R. & Ezzell, J.W. (2000). Bacillus anthracis identification based on two automated system databases. Abst 100 th Ann Mtg Amer Soc Microbiol, p.186. Uehling, M. (2002). Acting on anthrax- what one lab learned. CAP Today 16, 5-8. Brucella melitensis (Brucellosis) Bernard, K. (2002). The use of the BIOTER TM Data library as adjunct to the characterization of bacterial agents of Bioterrorism (BT). Abst 102 nd Ann Mtg Amer Soc Microbiol, p.101. Weyant, R.S., Moss, C.W., Weaver, R.E., Hollis, D.G., Jordan, J.G., Cook, E.C. and Daneshvar, M.I. (1996). Bacterial Identification by Cellular Fatty Analysis. In CDC s Identification of Unusual Pathogenic Gram-Negative and Facultatively Anaerobic Bacteria, second edn, pp Baltimore: Williams & Wilkins. Burkholderia mallei (Glanders) Srinivasan, A., Kraus, C.N., DeShazer, D., Becker, P.M., Dick, J.D., Spacek, L., Bartlett, J.G., Byrne, W.R. and Thomas, D.L. (2001). Glanders in a military research microbiologist. N Engl J Med 345, Weyant, R.S., Moss, C.W., Weaver, R.E., Hollis, D.G., Jordan, J.G., Cook, E.C. and Daneshvar, M.I. (1996). Bacterial Identification by Cellular Fatty Analysis. In CDC s Identification of Unusual Pathogenic Gram-Negative and Facultatively Anaerobic Bacteria, second edn, pp Baltimore: Williams & Wilkins.

18 Burkholderia pseudomallei (Melioidosis) Hsueh, P.R., Teng, L.J., Lee, L.N., Yu, C.J., Yang, P.C., Ho, S.W. and Luh, K.T. (2001). Melioidosis: An emerging infection in Taiwan? Emerg Infect Dis 7, Inglis, T.J.J., Aravena-Roman, M., Ching, S., Croft, K., Wuthiekanun, V. and Mee, B.J. (2003). Cellular fatty acid profile distinguishes Burkholderia pseudomallei from avirulent Burkholderia thailandensis. J Clin Microbiol 41, Inglis, T.J.J., Merritt, A., Chidlow, G., Aravena-Roman, M. and Harnett, G. (2005). Comparison of diagnostic laboratory methods for identification of Burkholderia pseudomallei. J Clin Microbiol 43, Weyant, R.S., Moss, C.W., Weaver, R.E., Hollis, D.G., Jordan, J.G., Cook, E.C. and Daneshvar, M.I. (1996). Bacterial Identification by Cellular Fatty Analysis. In CDC s Identification of Unusual Pathogenic Gram-Negative and Facultatively Anaerobic Bacteria, second edn, pp Baltimore: Williams & Wilkins. MIDI, Inc. Newark, Delaware Francisella tularensis (Tularemia) Bernard, K., Tessier, S., Winstanley, J., Chang, D. and Borczyk, A. (1994). Early recognition of atypical Francisella tularensis strains lacking a cysteine requirement. J Clin Microbiol 32, Bernard, K. (2002). The use of the BIOTER TM Data library as adjunct to the characterization of bacterial agents of Bioterrorism (BT). Abst 102 nd Ann Mtg Amer Soc Microbiol, p.101. Clarridge, J.E., Raich, T.J., Sjösted, A., Sandström, G., Darouiche, R.O., Shawar, R.M., Georghiou, P.R., Osting, C and Vo, L. (1996). Characterization of two unusual clinically significant Francisella strains. J Clin Microbiol 34, Weyant, R.S., Moss, C.W., Weaver, R.E., Hollis, D.G., Jordan, J.G., Cook, E.C. and Daneshvar, M.I. (1996). Bacterial Identification by Cellular Fatty Analysis. In CDC s Identification of Unusual Pathogenic Gram-Negative and Facultatively Anaerobic Bacteria, second edn, pp Baltimore: Williams & Wilkins. Yersinia pestis (Plague) Leclercq, A., Guiyoule, A., El Lioui, M., Carniel, E. and Decallonne, J. (2000). High homogeneity of the Yersinia pestis fatty acid composition. J Clin Microbiol 38, The information in this publication is subject to change without notice. Copyright 2005 MIDI, Inc. All Rights Reserved Date: 20 May 2005 MIDI, Inc. 125 Sandy Drive Newark, Delaware Phone: Fax: sales@midi-inc.com

19

20 Sherlock DNA Fact Sheet The name you trust in bacterial identification now offers identification by 16S rrna Gene Sequencing Extensive 16S rrna bacterial libraries, for both 500 base pair and 1500 base pair sequencing Extensive reporting options including: o Identification by percent difference o Concise Alignments o Neighbor Joining Trees o Accurate Root Neighbor Joining Trees (only available in the Sherlock system!) o Consolidated reporting of FAME and DNA results (only available in the Sherlock system!) Easy-to-use DNA Analysis tools Easy-to-use Custom Library Development tools

21 Sherlock Samples Page 1 Volume: DATA7 File: MIDITSB.DNA Samp Ctr: 6 ID Number: 5564 Type: Samp Bottle: 29 Method: DTSBA6 Created: 11/26/2003 3:00:30 PM Sample ID: QC-PAEN-GLUCA(V RT Response Ar/Ht RFact ECL Peak Name Percent Comment1 Comment E SOLVENT PEAK ---- < min rt :0 iso 1.38 ECL deviates Reference : ECL deviates Reference :0 iso 5.94 ECL deviates Reference :0 anteiso ECL deviates Reference : ECL deviates :0 iso 4.96 ECL deviates Reference :1 w11c 7.81 ECL deviates : ECL deviates Reference :0 iso 2.42 ECL deviates Reference :0 anteiso 6.70 ECL deviates Reference ECL Deviation: Reference ECL Shift: Number Reference Peaks: 8 Total Response: Total Named: Percent Named: % Total Amount: Matches: Library Sim Index Entry Name TSBA Paenibacillus-glucanolyticus Paenibacillus-alvei-GC subgroup A Paenibacillus-lautus 16S DNA: 538 base pairs D16S2 DNA Match Report Match %Diff Length Library Entry Name Paenibacillus-glucanolyticus Paenibacillus-lautus Paenibacillus-macerans Paenibacillus-amylolyticus Paenibacillus-pabuli Paenibacillus-azotofixans Paenibacillus-polymyxa Paenibacillus-apiarius Paenibacillus-curdlanolyticus Bacillus-chitinolyticus Cross Library Report: DNA Library: D16S2 FAME Library: TSBA6 DNA Ind %Diff Genus Species FAME SI FAME Ind paenibacillus glucanolyticus paenibacillus lautus paenibacillus macerans paenibacillus amylolyticus paenibacillus pabuli paenibacillus azotofixans paenibacillus polymyxa paenibacillus apiarius paenibacillus curdlanolyticus bacillus chitinolyticus paenibacillus alvei Sherlock Version 6.0 [S/N 12345] Printed: 11/9/2005 2:45:51 PM

22

23 Sherlock Software and Library Changes Sherlock 6.0 Software, Methods & Libraries vs. Sherlock 4.5 Software and 5.0 Methods & Libraries Sherlock 6.0 Software Additions Sherlock DNA Module 6.0- allows for the first automated fatty acid DNA sequencing (polyphasic) combined identification. 16S rrna (bacterial) and 28S rrna (fungal and yeast) data (from any DNA Sequencer) can be analyzed (identification and phylogenetic trees) as well as incorporated into a combined fatty acid- rrna sequencing identification report. There are also easy-to-use custom library development tools. Sherlock Advanced Analysis 6.0- new Sherlock Cluster combined with existing Sherlock Tracker. In Tracker, the user selects a seed sample (offline or online- after each sample analysis), and the system finds all samples that are close to the seed. Cluster automatically finds all groups of related samples to create sample clusters. Sherlock Advanced Analysis 6.0 vs. Sherlock Standard Analysis 6.0- One advantage of Sherlock Advanced Analysis (Tracker and Cluster) over Standard Analysis (Dendrogram and 2-D Plot) is that Standard Analysis requires the user to locate samples that might be related, whereas Advanced Analysis automatically locates related samples. With both Advanced and Standard Analysis, the user does not have to identify the sample. If the samples fail to identify with Sherlock (either because they are a species that we do not have in our library, a novel species that has never been named before or samples grown under different conditions that we specify), both Analysis & Advanced Analysis can still find relationships among the unknown samples. This is very useful for tracking unknown microbial contaminants. Sherlock 6.0 Software/System Enhancements Automatic instrument shutdown- saves gases by automatically shutting down the Gas Chromatograph when a sample batch is completed. Electronic Records enhancements- ERS sample-specific audit log and automatic transfer of ERS files to a networked server. Library Generation Software (LGS) is now split into Analysis and Library Generation- users that now want Analysis (2-D Plot and Dendrogram) capabilities do not have to purchase Library Generation as well. Sherlock 6.0 Method Improvements For the 6.0 methods, the peak-naming table for the Rapid and Standard Methods is identical- both peak-naming tables have the same set of named peaks, making comparisons between Rapid and Standard Methods easier.

24 Sensitive Anaerobe Methods (SMOORE6 & SANAER6) for Anaerobic Bacteria- the results are 2 times more sensitive (20mg vs. 40mg cells required) than the Standard Anaerobe Methods (MOORE6 & ANAER6). Useful for slow-growers. Sensitive Yeast Methods (SYEAST6 & SYSTCLN6) for Yeasts & Fungi- the results are 2 times more sensitive (20mg vs. 40mg cells required) than the Standard Yeast Methods (YEAST6 & YSTCLIN6). Useful for slow-growers. One Calibration Mix- for Rapid and Sensitive Methods, only one Calibration Mix is needed. Sherlock 6.0 Library Improvements Substantial improvements- in the Aerobic (Clinical and Environmental) Bacterial Libraries, as well as the Anaerobe Library. Peak naming table has been improved- allows for more accurate sample identifications. Improved library validation tools- for easier custom library creation. Addition of Sherlock DNA Libraries- 16S rrna for bacteria and 28S rrna for yeast & fungi. Both 500 & 1,500 base pair libraries are available for the 16S library. Sherlock 4.5 Software & 5.0 Methods & Libraries vs. Sherlock 3.0/3.1 Software and 4.0 Methods & Libraries Sherlock 4.5 Software Additions Sherlock Electronic Records and Signatures 4.5- allows for compliance with FDA 21CFR Part 11. Sherlock Data Export 4.5- allows one to export all Sherlock data reports to Microsoft Access and Excel Programs. Sherlock Tracker 4.5- with the Analysis features in Sherlock Library Generation Software 3.0/3.1, you needed to know which samples were of interest and select them for use in the particular visualization (dendrogram or 2-D Plot). With Sherlock Tracker 4.5, the Sherlock Software searches for related samples on its own. In the offline Tracker mode, the user selects a seed sample and the Sherlock Software identifies all samples (can be every sample on your hard disc) that are close to the seed. With the online Tracker mode, after each sample run, the Sherlock Software automatically identifies the closest samples. For many species, Tracker allows for presumptive strain tracking in near-real time. Sherlock 4.5 Software/System Enhancements Old DOS-based applications are replaced with new 32-Bit Windows-based programs- Softkeys are replaced by menus and toolbars. Dramatic improvement in ease of use. Look and feel of Microsoft Outlook.

25 WYSIWYG - what you see is what you get. When you select samples, you see the list on the screen. When you look at a sample, you see its profile and its matches right on the screen. When you create a dendrogram or 2-D Plot, you see the results right on the screen. You can then adjust the zoom and add/delete samples with a few keystrokes, making data analysis very easy. Printouts use "real" graphics instead of line print characters- The printouts can be saved to RTF, usable by Microsoft Word. Print Preview option- allows one to inspect large amounts of output before printing any pages. Automatic cataloging- no more manual cataloging of an invisible set of samples. More reliable interaction with the Agilent ChemStation- The DOS applications would occasionally lose connectivity with the ChemStation and require difficult system optimizations. More reliable interaction with other applications- we no longer limit the other applications that one can run- specifically it is totally acceptable to have the Microsoft Office suite on the same computer. On-screen chromatogram- the chromatogram for a sample can be seen on screen while inspecting Sherlock data for the sample. Sherlock 5.0 Method Improvements Rapid Clinical Aerobe Method (RCLIN5.0) for Clinical Aerobic Bacteria- the results are 3 times faster (9 min vs. 27 min GC runs) & 2 times more sensitive (20mg vs. 40mg cells required) than the Standard Clinical Aerobe Methods (CLIN5.0). Useful for slow-growers and higher sample throughput. Rapid Environmental Aerobe Method (RTSAB5.0) for Environmental Aerobic Bacteria- the results are 3 times faster (9 min vs. 27 min GC runs) & 2 times more sensitive (20mg vs. 40mg cells required) than the Standard Environmental Aerobe Method (TSBA5.0). Useful for slowgrowers and higher sample throughput. For the 5.0 methods, the peak-naming table was changed to eliminate high-variance features- the 15:0 feature had a high variance in the Standard Methods. Two other peaks were summed together in the Rapid Aerobe Methods for a similar reason (the sum had lower variance than the individual peaks). Those 5.0 modifications meant that the peak-naming table was not identical for Standard and Rapid samples. Sherlock 5.0 Library Improvements Improvements- in the Aerobe (Clinical and Environmental) Bacterial Libraries, as well as the Anaerobe Library.

26 The information in this publication is subject to change without notice. Copyright 2005 MIDI, Inc. All Rights Reserved Date: 01 December 2005 MIDI, Inc. Newark, Delaware MIDI, Inc. 125 Sandy Drive Newark, Delaware Phone: Fax: