Forensic Applications of New Analytical Technologies
By: Fiona Couper, Tom Gluodenis, Mark Jensen, Matthew Klee, Lawrence Neufield, Bruce Quimby, Lucas Zarwell, Jerry Zweigenbaum
Issue: April/May 2005
Untitled Document
Forensic investigations involve the discovery and characterization of evidence
that can be used to reconstruct a chronology of events associated with the
commision of a crime or other matters being adjudicated. As they become available,
increasingly sophisticated anyalytical tools and methods are being employed
to detect and discriminate evidence.
Multiply hyphenated techniques such as--gas chromatography/mass spectrometry
with retention time locking (GC/MS/RTL), liquid chromatography/time-of-flight
mass spectrometry (LC/MS/TOF), microfluidic-based capillary electrophoretic analysis
of mitochondrial DNA (mtDNA), and laser ablation inductively coupled plasma mass
spectrometry (LA/ICP/MS) --are able to uncover forensically germane information
by providing unprecedented levels of analytical selectivity and sensitivity,
extracting genetic signatures from previously overlooked biological sources,
and sequentially microdeconstructing samples so as to map the spatial variation
and concentration of elemental constituents. These new ranges of information
and rich data sets can contribute facts crucial to the reconstruction of events
and thereby increase the probability of an accurate finding in the matters under
investigation.
The term forensic science is associated with the application of analytical
tools and techniques in the discovery of evidence deemed relevant in the investigation
of a crime or in some other legal proceeding. Sources of evidence may be biological
samples from persons living or dead, physical objects and their disposition,
deposited trace materials, alteration or disruption of a setting, as well as
circumstantial references connecting these data. The evolving power and sophistication
of analytical instrumentation has made it possible to perform forensic investigations
at ever smaller size scales with greater sensitivity and finer ranges of differentiation.
As a result, it is now possible to routinely uncover evidence formerly indiscernible
or inaccessible.
Evidence delineated at the chemical level is particularly powerful because
the nature, relative abundance, and spatial disposition of atomic and molecular
constituents provide a distinctive signature that can identify an object or
substance, determine its source, and detect changes to its integrity resulting
from structural and surface alteration and/or contamination. Outcomes of the
forensic application of these technologies are reported daily in the media
and include the recent detection of dioxin poisoning of a political candidate,
the use of genomic DNA to exonerate the wrongly convicted, and the analysis
of unreported oil spills to determine the pollution source. This article reviews
recent advances in forensic analysis resulting from the application of increasingly
sophisticated analytical instrumentation and contingent methods.
Gas Chromatography/Mass Spectrometry with Retention Time Locking (GC/MS/RTL)
GC/MS/RTL is not a new development; rather it represents a key innovation in
an ongoing wave of “next generation” instruments and software
that have significantly improved legally defensible chemical analysis. In
GC/MS analysis, a combination of retention time (RT) and mass (m/z) spectra
data are used to identify and quantify target analytes. Given the complexity
of many forensic samples, there is a distinct possibility that target analytes
will co-elute with interfering compounds of similar RT values. The wider
the RT identification window used, the higher the number of potential hits
an analyst has to deal with. In addition to requiring wide RT windows, run-to-run
reproducibility of early GCs was not sufficiently precise to allow retention-time
tables to be ported from one instrument to another. This constraint limited
both the productivity of analysts as well as the overall productivity of
the laboratory. These limitations have been overcome with the introduction
of instruments equipped with electronic controls for all GC parameters that
enable settings for a method to be recorded and identically replicated from
one run to the next. The enhanced precision makes it possible to reproduce
run-to-run RTs to fractions of a second for a given method over the long
term. Locking a whole method requires only locking the RT of a single analyte,
designated as the locking standard, in the method. This technique can also
be repeated to lock the RTs on other instruments of similar configuration.
Since all instruments now produce nearly identical RTs, narrow identification
windows can be used, reducing the number of potential hits requiring validation
and providing much more efficient matches to a database of target RT values.
GC/MS/RTL has been employed in forensic toxicology, an area of forensic science
concerned with the identification and measurement of toxins in the human body.
One laboratory that has become expert in this application is the National Medical
Services (NMS) of Willow Grove, Pennsylvania. NMS utilizes a 5973 Network GC/MS
system with MSD Productivity ChemStation software, running the Drug Data Analysis
and Enhanced Data Analysis modes (Agilent Technologies, Palo Alto, CA). By
employing RTL, NMS has been able to dramatically increase throughput since
RTs are unvarying from one run to the next and in conjunction with m/z values,
conclusively identify a target analyte. By recording precisely locked RT values,
NMS has compiled an extensive target reference database of approximately 300
toxicological compounds with new compounds being added continually. Figure
1 and the accompanying table illustrate how RTL screener software automates
database searching and highlights both major and/or unusual hits in a chromatographic
run for analyst inspection and interpretation/verification.
Liquid Chromatography/Time-Of-Flight Mass Spectrometry (LC/MS/TOF)
Identifying analytes from a target list of RTs eliminates the requirement for
running individual standards for each unknown. But what if the peaks encountered
are truly unknown? Without a standard and a sufficiently accurate determination
of m/z, the possible number of candidate analytes fitting the data becomes
quite large. Making a conclusive identification under these circumstances
formerly required more extensive and time-consuming tandem MS analysis, including
the generation and interpretation of ion fragmentation data. This burdensome
requirement in performing forensic toxicological analysis is likely to grow
given the expansion in the number and variety of drugs being administered
by prescription or otherwise.
Until recently, the only way to determine a mass peak with sufficient accuracy
for conclusive identification was to run the sample with either a double focusing
magnetic sector mass spectrometer or Fourier Transform Mass Spectrometry (FTMS).
The magnetic sector instruments are not amenable to Atmospheric Pressure Ionization
(API)/LC/MS because of the high accelerating voltages. FTMS is a difficult
and expensive technique. Both require highly trained analysts to operate and
calibrate for accurate mass measurement. Thus this type of determination has
not been readily available to most laboratories.
The problem of achieving sufficient accuracy has been overcome by employing
liquid chromatographic separation in conjunction with new time-of-flight mass
spectrometer (MS/TOF) technology capable of mass accuracies better than 3 ppm.
This high level of mass resolution makes it possible to determine the empirical
formula of unknown analytes in full scan mode across the mass range of the
instrument. As a result, the number of potential candidates fitting the data
is greatly limited, thereby enhancing the capability of making a conclusive
identification, especially when pertinent ancillary information is considered.
A dramatic example of the improvement in detection capability afforded by
this new MS/TOF instrumentation is illustrated in a case involving the Washington
DC Medical Examiner’s GC/MS detection of an unknown analyte in a sample
of blood taken postmortem from a deceased person with cause of death unknown.
Figure 2 is an LC/MS/TOF spectrum of the sample showing the unknown peak. While
both GC/MS and LC/MS/TOF detect this peak, only the latter instrument is able
to assign mass values with sufficient accuracy to propose an empirical formula
(C15H15N4O) for the primary [M+H]+ ion. A search of the Merck Index yields
a single match, the antiretroviral drug, nevirapine1. A fragmentation spectrum
of the unknown peak produced fragment ions consistent with this assignment.
Microfluidic-Based Capillary Electrophoretic Analysis of Mitochondrial DNA
(mtDNA)
Advances in microscale fabrication have made possible the development of the
chemical microchip or lab-on-a-chip. The surface of these chips are manufactured
with in situ capillary channels, microvalves, and other miniaturized equivalents
of laboratory tools arranged in a configuration for carrying out one or a series
of laboratory operations. This technology is employed in an instrument designed
for the capillary electrophoretic separation and molecular size determination
of biomolecules such as DNA, RNA, and proteins, and is now being used in the
quality assessment of recovered mtDNA pursuant to sequencing.
For forensic purposes, the identification and sequencing of genomic DNA is
the “gold standard” in the identification and probabilistic linkage
of recovered biological material to its source. However, since each cell contains
only two copies of genomic DNA, limited sample availability or degradation
over time and/or from environmental exposure, may preclude the recovery of
genomic DNA in sufficient quantity and/or quality for forensic analysis. Under
such circumstances, workers are increasingly focusing on the analysis of mtDNA
as a surrogate for genomic DNA.
Unlike the two copies of genomic DNA present in each cell nucleus, the cytoplasm
of each cell contains multiple mitochondria and can yield about 1,000 copies
of mtDNA2. Because it is available in far greater abundance than genomic DNA,
mtDNA is more easily recovered, amplified, and sequenced than genomic DNA under
sample-limited and sample-degrading conditions. However, since mtDNA sequences
are maternally inherited, siblings and all maternal relatives have the same
sequence. For this reason, mtDNA sequences are not as conclusively probative
for identification purposes and this must be taken into consideration. Nevertheless,
the sequencing of mtDNA from degraded remains can help in the identification
of long deceased individuals or in linking biological evidence associated with
the commission of a crime to samples recovered from putative victims and suspected
perpetrators.
Before mtDNA can be employed for sequencing purposes, it must first be established
that the material is neither degraded nor contaminated, so as to prevent the
confounding of the sequencing process3. Quality and integrity can be quickly
determined by amplifying the mtDNA by means of the polymerase chain reaction
(PCR) and then separating and classifying the amplified product according to
molecular size by capillary electrophoresis. Figures 3A and 3B show respective
analyses for pure and contaminated or degraded mtDNA. If the sample is pure,
the analysis produces a single peak indicative of a single mtDNA molecular
size fraction. By contrast, contaminated or degraded mtDNA produces multiple
peaks. This quality control check is important if spurious results and false
assignments are to be avoided.
Laser Ablation Inductively Coupled Plasma Mass Spectrometry (LA/ICP/MS)
In general, the types of sample analyzed using the techniques discussed in
the prior sections of this article do not normally present formidable preparation
problems. However, since the sample is consumed in performing the analysis,
the evidence it represents is destroyed. In sample-limited situations, all
of the recovered material may be used in the analytical determination. This
precludes the possibility of conducting additional forensic analyses pursuant
to the current investigation or in future, should it be warranted. Some samples
are difficult to prepare because they are physically tough, inert, or inextricably
embedded in a matrix that resists dissolution by conventional treatments.
Moreover, it may often be the case that a complete forensic assessment requires
a determination not only of the nature and abundance of a sample’s
chemical composition, but also of the disposition of these constituents at
various locations on, about, or within the sample.
The technique of laser ablation inductively coupled plasma mass spectrometry
(LA/ICP/MS) overcomes the preparation problems associated with intractable
samples. It performs spatially localized sampling and analysis on a microscale,
often requiring <1µg of sample and thereby conserving evidence. Spatially
localized elemental constituents are discriminated at the ppb level enabling
the detection of trace chemical signatures that permit discrimination between
samples that otherwise might be indistinguishable.
Unlike other MS-based methods, LA/ICP/MS requires little or no time for sample
preparation to remove a matrix or otherwise render analytes of interest in
a form suitable for subsequent analysis. Instead, laser ablation at a predetermined “spot” on
the sample surface creates a microplume of sample material that is aerosolized
by a directed stream of inert argon gas, carrying it into the plasma where
it is decomposed, atomized, ionized, and then classified by the mass spectrometer.
The instrument is equipped with computer-controlled imaging and laser-positioning
and focusing, permitting the ablation “spot” size to be varied
between <5 µm to 300 µm. These controls make possible analyses
at microlocations within a sample. Comparative 3-dimensional assays may be
performed by applying identical laser spot sizes and ablation times at desired
sample surface locations. Figure 4 is a representation of laser ablation of
a sample.
Figure 5 illustrates the discriminatory capabilities of the LA/ICP/MS. In
this example, fragments of automobile headlight and window glass that might
be found at the scene of an accident or involved materially in some forensic
investigation may need to be identified with respect to supplier, automobile
manufacturer, and even the year of production. Refractive index measurements
do not provide sufficient discrimination. While traditional analytical methods
can provide a richer set of data, including the detection and quantification
of major and minor elemental constituents, the results may still be insufficiently
selective to differentiate samples of similar but not identical composition.
In the present case, only LA/ICP/MS provides the trace elemental signatures
that can discriminate between different glasses with such similar compositions.
Conclusion
The analytical tools and methods outlined here are quite dissimilar in design,
type of sample addressed, and results produced. What they do have in common
is a new level of sophistication in the way sample preparation and processing
is integrated into the analytical scheme. Moreover, the advanced electronics,
material fabrication, software design, and integration that these instruments
incorporate have enabled orders of magnitude improvement in detection sensitivity,
selectivity, and analytical throughput. When these capabilities are applied
forensically, it enables a more subtle discrimination and characterization
of material evidence. The new range of information extracted can be crucial
in validating, reducing uncertainty, or conversely disproving theories employed
in the reconstruction of a chronology for the events under investigation
and thereby help ensure that a correct finding results.
For more information about the topics discussed in this article visit the
Life Sciences/Chemical Analysis section of the Agilent Technologies website
(http://www.chem. agilent.com) and search on “forensics” as well
as on specific instruments, techniques, and applications.
Notes
1 It was independently determined that the deceased was infected with HIV/AIDS.
2 Mitochondria are the respiratory or energy-generating organelles of eucharyotic
cells. The construction and operation of the mitochondria are scripted by means
of resident DNA distinct from the cell’s nuclear genomic material. Unlike
the paired alleles of nuclear DNA, each of which is inherited from one parent,
mtDNA is inherited matrilineally.
3 Current FBI regional laboratory procedures stipulate that amplified target
mtDNA must be present in 10-fold excess above any unintended PCR products.
Failure to meet this purity requirement may render the target sequence unreadable
or may even result in erroneous nucleotide base assignments due to the presence
of sequence information arising from the presence of secondary PCR products.
More Information on Forensic Magazine April/May 2005 Article "Forensic
Applications of New Analytical Technologies”
Constance L. Fisher, Ph.D. a Forensic Examiner with the DNA II Unit of the
FBI Laboratory contacted Forensic Magazine with several clarifications. The
following notes are from the author, Mark Jensen of Agilent Technologies, Inc.
1. Reading Note #3, the reader might assume that FBI regional laboratories
follow procedures that differ from procedures used by FBI central laboratory.
Such an assumption is not necessarily true.
2. With respect to Figure 3B, in preparing our article we made assumptions
regarding the PCR QC process without adequately confirming details with the
FBI DNA Forensics Lab. We have since learned that the quality control procedure
used by the FBI is as follows:
For each mtDNA sample, the quantity of desired PCR product is determined (Figure
3A), and, in addition to this sample, two additional control samples are also
analyzed. These two additional control samples consist of a negative control
and a reagent blank. The quantity of DNA product in the negative control sample
and reagent blank sample are then compared to the quantity of target PCR product
determined in the mtDNA sample. The ratio of contaminant products of the same
electrophoretic migration time (bp length) in the negative control sample and
reagent blank sample, versus the quantity of desired PCR product in the mtDNA
sample must not exceed 10%.
The authors are grateful to Dr. Fisher for correcting any misunderstanding.
Fiona Couper, fiona.couper@dc.gov, is Chief Toxicologist, Office of the Chief
Medical Examiner, Washington, DC.
Thomas Gluodenis, tom_gluodenis@agilent.com,
is Marketing Manager for Homeland Security Products, Agilent Technologies,
Inc., Wilmington, DE.
Mark Jensen, mark_jensen@agilent.com, is Senior Applications Chemist, Agilent
Technologies, Inc., Wilmington, DE.
Matthew Klee, matthew_klee@agilent.com, is Technology Specialist, Agilent
Technologies, Inc., Wilmington, DE.
Lawrence Neufeld, lneufeld@new-wave.com, is Laser Ablation Products
Applications and Marketing Manager, New Wave Research, Inc., Fremont,
CA.
Bruce Quimby, bruce_quimby@agilent.com, is Senior Applications Chemist,
Agilent Technologies, Inc., Wilmington, DE.
Lucas Zarwell, lucas.zarwell@dc.gov, is Senior Forensic Toxicologist,
Office of the Chief Medical Examiner, Washington, DC.
Jerry Zweigenbaum, j_zweigenbaum@agilent.com, is Market Development
Specialist, Agilent Technologies, Inc., Wilmington DE.
