The forensic science

Introduction of title

The terms of “forensic science” cover those professions which are involved in the application of social and physical science to the criminal justice system. Forensic experts are expected to find out the details of method used, to substantiate the choice of the applied technique and to give their unbiased conclusions. Therefore, the methods applied in forensic laboratories should assure a very high level of reliability and must be subjected to extensive quality assurance and rigid quality control programs. The legal system is based on the belief that the legal process results in justice. This has come under some question in recent years. Of course, the forensic scientist cannot change scepticism and mistrust single-handedly. He or she can, however, contribute to restoring faith in the judicial processes by using science and technology in the search for facts in civil, criminal and regulatory matters.

However, there are various chromatographic techniques used in forensic science since long time ago. Chromatography is probably the most powerful and versatile technique available to the modern analyst. In a single step process it can separate a mixture into its individual components and simultaneously provide an quantitative estimate of each constituent. Chromatography is a separations method that relies on differences in partitioning behavior between a flowing mobile phase and a stationary phase to separate the components in a mixture.

Separation of two sample components in chromatography is based on their different distribution between two non-miscible phases. The one, the stationary phase, a liquid or solid, is fixed in the system. The other, the mobile phase, a fluid, is streaming through the chromatographic system. In gas chromatography the mobile phase is a gas, in liquid chromatography it is a liquid The mobile phase is forced through an immobile, immiscible stationary phase. The phases are chosen such that components of the sample have differing solubilities in each phase. A component which is quite soluble in the stationary phase will take longer to travel through it than a component which is not very soluble in the stationary phase but very soluble in the mobile phase. As a result of these differences in mobilities, sample components will become separated from each other as they travel through the stationary phase. Different distribution of the analytes between mobile and stationary phase results in different migration velocities.

As all chromatographic separations are carried out using a mobile and a stationary phase, the primary classification of chromatography is based on the physical nature of the mobile phase. The mobile phase can be a gas or a liquid which gives rise to the two basic forms of chromatography, namely, gas chromatography (GC) and liquid chromatography (LC). The stationary phase can also take two forms, solid and liquid, which provides two subgroups of GC and LC, namely; gas-solid chromatography (GSC) and gas-liquid chromatography (GLC), together with liquid solid chromatography (LSC) and liquid chromatography (LLC).

There are few common examples as below :

1. Gas chromatography (GC)

   It applies to volatile organic compounds. The mobile phase is a gas and the stationary phase is usually a liquid on a solid support or sometimes a solid adsorbent.

2. High-performance liquid chromatography (HPLC)

A variation of liquid chromatography that utilizes high-pressure pumps to increase the efficiency of the separation.

3. Liquid chromatography (LC)

Used to separate analytes in solution including metal ions and organic compounds. The mobile phase is a solvent and the stationary phase is a liquid on a solid support, a solid, or an ion-exchange resin.

4. Size-exclusion chromatography (SEC)

Also called gel-permeation chromatography (GPC), the mobile phase is a solvent and the stationary phase is a packing of porous particles.

5. Thin-layer chromatography (TLC)

A simple and rapid method to monitor the extent of a reaction or to check the purity of organic compounds. The mobile phase is a solvent and the stationary phase is a solid adsorbent on a flat support.

Among all the chromatographic techniques, thin-layer chromatography is the easiest method to demonstrate in a classroom how chromatography is used to separate materials.

Case Study 1:

Extraction and analysis of flunitrazepam/7-aminoflunitrazepam in blood and urine by HPLC and GC-MS

Introduction:

Flunitrazepam is a drug which is famous used in the area of drug-facilitated sexual assaults (DFSA) (Madea et al., 2004). Flunitrazepam solution is odourless, tasteless and colorless. These features, together with its potential use with ethyl alcohol have often caused the drug to be used an incapacitating agent in rape or robbery. (Calhoun et al., 1996.; Ledray and Amerg, 1996 .; Anglin et al, 1997) It belongs to most potent hypnotic benzodiazepines and prescript in most European countries, in Australia, South Africa and Latin America. In US, flunitrazepam is not approved. Nevertheless, the drug is smuggled into the US and its metabolite has been identified several times in clinical samples during drug screening (Valentine et al., 1996; Woods et al., 1996; Saum and Inciardi, 1997). The drugs in its parent form has been administered orally and by intravenous injection in 0.5-2 mg. Previous published methods for the extraction of flunitrazepam and its metabolites have used both of liquid-liquid and solid-phase extraction (Robertion and Drummer, 1995).

Flunitrazepam undergoes reduction to 7-aminoflunitrazepam(7-AF) After acetlyation the metabolites are excreted in urine (Baselt and Cravey, 2004). 7-AF was the most abundant metabolites among others and could be found up to 72 hours after ingestion of 1mg of drug, whereas the parent drug was not detectable in urine (Salamone et al., 1997). Instrumental methods of analysis have used electron-capture detection gas chromatography (Cano et al.,1997), and without prior derivatization (deSilva and Bekersky, 1974), high performance liquid chromatography (Sumitrapura et al.,1982) with mass spectrometry (LeBeau et al.,2000). Gas chromatography-mass spectrometry (GC-MS) using chemical derivatives such as BSTFA (Salmone, 1997) and MTBSTFA (United Chemical Technologies).

All assay methods for flunitrazepam and its metabolites show some limitations. Benzodiazepine immunoassays often lack the sensitivity required to detect flunitrazepam metabolites in urine (Manchon et al., 1985; Rohrich et al., 1994; Huang and Moody, 1995). The on-line enzyme hydrolysis may improve the detectability of 7-AF with immunoassay (Drummer et al.,1993; Beck, 1997). Whole blood radioimmunoassay, specific for flunitrazepam, showed no crossreactivity for 7-AF (West et al., 1995). Gas chromatography (GC) with sensitive, but unspecific electron capture detection (ECD) allows detection of flunitrazepam, but not 7-AF. In contrast to GC, all substances involved may be separated by high-performance liquid chromatography (HPLC) without any derivatization. The weak points of HPLC with UV detection were: questionable specificity and low sensitivity (the limit of quantitation was 10 mg/l (Boukhabza et al.,1991; Guichard et al., 1993; Robertson and Drummer, 1995). The purpose of the present study is to identify a solution to problem of extraction, detection and quantification of this drug and its metabolite in whole blood and urine that other workers in the field have identified in that flunitrazepam is easily converted into metabolite(s)

Procedure :
Sample pretreatment
Calibrators and controls

Stock solutions of flunitrazepam, 7-aminoflunitrazepam and nitrazepam (10µgmL-1) were made up separately in 10mL volumetric flasks. This was performed by the addition 100µL of each solution (1mgmL-1) to distilled water and making up to the mark with the same. Working solutions (1000ngmL-1) of each drug were made by serial dilution in separate 10mL volumetric flasks.

Whole blood or urine

Nitrazepam (IS) (100µL of 1000ngmL-1) was added to samples of human drug free whole blood/urine (1mL) in 16mm-125mm screw top test tubes. Drug free bovine blood and post mortem human urine was employed. Both sources of blood had previously treated with sodium fluoride. To control stability of the drugs, the analytes were added to the blood/urine at the point of analysis (as storage may lead to conversion of the parent to the metabolite (Robertson and Drummer, 1998). For the calibrators, to each sample was added a known amount of flunitrazepam and 7-aminoflunitrazepam (0, 10, 25, 50 and 100ng). In the case of the controls, in addition to the nitrazepam (100ng), flunitrazepam and 7-aminoflunitrazepam were added at a level of 40ngmL-1 . All determinations (calibrators and controls) were carried out in duplicate. The tubes were vortex mixed (approximately 1min). To each sample was added 10mL of distilled water. The tubes were again vortex mixed for approximately 1min. The tubes were centrifuged at 2000-g for 10min. The supernatant liquid was applied to previously conditioned solid phase columns.

Extraction

The SPE columns were placed in numbered slots in the manifold. Each column was conditioned with 1- 3mL of methanol followed by 1- 3mL of distilled water. These were allowed to percolate through the sorbent under gravity. The level of distilled water was held just above the solid phase sorbent bed to prevent it from drying out.

The supernatants were loaded onto the conditioned solid phase sorbent and allowed to pass through with the aid of gravity. After the samples were drawn through the solid phase columns, the sorbent was washed with 1- 3mL of distilled water, this was followed by 1- 3mL of 1M of acetic acid. The columns were dried under full vacuum for 5min. The cartridges were then washed with 1- 3mL of hexane after which they were dried under full vacuum for 5min.

Elution process prior to high performance liquid chromatography (HPLC)

In liquid chromatography, the eluant is the liquid solvent. The drugs and internal standard were eluted from the columns using an organic solvent mixture consisting of ethyl acetate/methanol (80:20). The volume of elution solvent employed was 2- 3mL. The eluants were collected in screw top test tubes (10mL capacity) at a rate of approximately 1mLmin-1. The combined extracts were evaporated to dryness under a gentle stream of nitrogen at 40°C. The residue was dissolved in 0.1% (v/v) aqueous TFA (100µL). This solution was transferred to an auto sampler vial containing a low volume insert (200µL) for analysis by HPLC.

Process employed prior to gas chromatography-mass spectrometry

The residue was dissolved in ethyl acetate (25µL) and 25µL of Pentafluoropropionic acid (PFPA) was added as the derivatizing reagent after the collection and evaporation of the combined eluants. Next, this solution was heated at 80°C for 30min. After cooling, the PFPA was evaporated to dryness at 40°C under a gentle stream of nitrogen. The residue was dissolved in 50µL of ethyl acetate. This was transferred to an autosampler vial containing a low volume insert (200µL) for analysis by GC-MS.

Result :

Discussion :

Altough there are low levels of flunitrazepam and 7-aminoflunitrazepam from whole blood/urine, the use of butyl endcapped solid phase columns give an selective solution to the problem of extracting them. The extraction relies on the hydrophobic interaction of the sorbent and the analyte solution. By applying a shorter alkyl chain, example C4 bonded to a silica structure, the polarity of the solid phase sorbent is increased. This works to aid the process of sorbing the drugs onto the solid phase column.

In the literature, the pKa values of flunitrazepam and nitrazepam are reported as 1.8 and 3.2 respectively (Moffat et al., 2004). Information relating to the pKa of 7-aminoflunitrazepam is not easily available. From this data, it is a natural assumption that all three analytes are acidic in nature. The pH modification of the blood and urine samples by distilled water make it easier to extract the drugs from the bio-fluids efficiently. It also assists the smooth flow of the solution through the column.

The use of liquid chromatography with photodiode array (LC-PDA) shows the ability of this system to separate and detect these drugs at low levels. It also expands the possibility of opening up this technology to this area of forensic analysis. In this method, 250nm was chosen for detection wavelength for the analytes rather than 220nm. It is because 220nm would have increased the sensitivity but not the selectivity. In this case, we emphasized the selectivity.

Gas chromatography coupled to mass spectrometry(GC-MS) employed in selected ion monitoring (SIM) is routinely used in forensic toxicological analysis. The application of this technique to the procedure shows that much lower levels of detection can be achieved by using this method of extraction.

The linear range of this method of extraction was found to be linear from 0 to 100ngmL-1. To measure the accuracy and precision of the method spiked samples were analyzed. In the analysis of spiked controls (40ngmL-1) blood samples were analyzed in duplicate. The values obtained by HPLC were: 34 (±5)ngmL-1 for the flunitrazepam and 48 (±5)ngmL-1 for the 7-aminoflunitrazepam.

The values obtained by GC-MS for spiked blood controls (40ngmL-1) were reported as: 37 (±4)ngmL-1 for the flunitrazepam and 45 (±4)ngmL-1 for the 7-aminoflunitrazepam. The recovery of flunitrazepam was 83% (±4%) and 87% (±4%) for 7-aminoflunitrazepam.

The limits of detection and quantification (LOD, LOQ) of the extraction method were determined using HPLC by undergoing a series of standards containing flunitrazepam and 7-aminoflunitrazepam with the internal standard (both unextracted and extracted from blood samples using solid phase columns) over a range of concentrations (5-100ngmL-1) to observe the lowest level at which the drug(s) could be detected. They were run on gas chromatography-mass spectrometry in selected ion monitoring mode with the addition of a decrease of five-fold in the lowest concentration level (5ngmL-1 of flunitrazepam and 7-aminoflunitrazepam). A detection level of 1ngmL-1 was achieved for both drugs (flunitrazepam and 7-aminoflunitrazepam) using gas chromatography-mass spectrometry in selective ion monitoring mode.

Analysis of a CAP proficiency test sample (urine) was found to be positive for 7-aminoflunitrazepam by Enzyme-linked immunosorbent assay (ELISA). The data supplied by the College of American Pathologists (CAP) indicated that the sample contained 63ngmL-1 of 7-aminoflunitrazepam (FTC-A Forensic Toxicology (Criminalistics), 2002). Analysis of the sample by this procedure using HPLC in 2004 revealed a level of 51 (±5)ngmL-1 of 7-aminoflunitrazepam. No flunitrazepam was detected. This showed that the method is valid for both internal and external controls.

It has been noted by earlier workers (Robertson and Drummer, 1998) that nitrobenzodiazepines, such as flunitrazepam are not only difficult to analyze cause of the low concentration of the drug that present in blood/urine samples (especially in DFSA cases which are discovered late after certain event) but also due the parent drug can break down to form the metabolite even after the sample has been drawn from the body and stored appropriately. In this situation, the concentrations of both flunitrazepam and the 7-aminoflunitrazepam are reduced even further.

Case study 2:

Analysis of explosives using high performance liquid chromatography with UV absorbance and photo-assisted electrochemical detection.

Introduction

Explosives are the chemical compounds or mixtures that will, on application of an external stimulus such as heat, shock, friction or ignition, undergo rapid chemical decomposition. The chemical reaction causes sudden releases of large amount of energy due to liberation of gas and temperature. The pressure thus release out equally in all directions.

Although there are mechanical and nuclear explosives, chemical explosives are the most commonly used. A mechanical explosive is one in which a physical reaction is produced, like that caused by overloading a container with compressed air. While nuclear explosives, which produce a sustained nuclear reaction, are by far the most powerful, but their use is restricted due to the over powerful explosion. They normally used in certain industrial operation in oversea.

The main chemical explosives include black powder, dynamite, nitroglycerine and trinitrotoluene (TNT). A chemical explosive can be gaseous, liquid or solid. Nitroglycerin and dynamite succeeded black powder as the chief explosives. An Italian chemist, Ascanio Sobrero discovered nitroglycerin in 1846. The Swedish scientist, Alfred Nobel explored dynamite in 1867, the original explosives being a mixture of 75 % nitroglycerin and 25 % ghur (a porous, absorbent material that made the product easier to control and safer to use). Later, ammonium nitrate was used in dynamite which is more safer to use and cheaper to produce.

In order to analyse content of different explosive, several detection have been coupled to HPLC for the determination of explosives, including UV absorbance, refractive index(RI), mass spectrometry (MS) and dc amperometry (Yinon and Zitrin, 1993; Yinon 1990; Yinon, 1999; Krstulovic and Brown, P.R., 1982; Martens and Frankenberger Jr, 1900, Krull et al., 1983; Hilmi et al., 1999). However, RI and UV detectors are lack of sensitivity because of the presence of interfering compounds in environmental.

Ignoring its inherent flows, HPLC-UV is accepted and commonly used by the U.S. Environmental Protection Agency (EPA) for the determination of explosives in ground water and soil (EPA Method 8330). A ground water sample size of 1liter must be subjected to solvent extraction and be salted out and evaporated down to 5ml prior to analysis. Then samples be run on a C18 bonded-phase column and, subsequently, on a cyano bonded-phase column for confirmation of analyte identification. Reductive amperometry has been applied to the analysis of explosives to gain the selectivity. But it is suffer from lack of sensitivy and excessive noise, which is attributable to the reduction of dissolved oxygen present in the mobile phase and sample.

Post-column photochemistry in HPLC as a general derivatization approach for improved UV, fluorescence, and electrochemical detection has been used for a variety of analytes (Krull and LaCourse, 1986). Photochemical derivatization provides the advantages of a reagent-free system, including the elimination of mobile phase restrictions, pulsations due post-column delivery, inadequate mixing of post-column reagents, and matrix effects due to chemical reagents [8] I.S. Krull and W.R. LaCourse, Reaction Detection in Liquid Chromatography, Marcel Dekker, New York (1986) p. 326.(Krull and LaCourse, 1986). Photochemical reactors have variety of bulbs at different wavelengths and knitted reactor coils of various lengths and diameters. HPLC-photo-assisted electrochemical detection (HPLC-PAED), formerly referred to in the literature as HPLC-h?-EC, involves first the separation of the analytes of interest, followed by the photolytic generation of a new species that can then be detected electrochemically (Krull and LaCourse, 1986). For nitro compounds, it has been reported that photolysis results in bond cleavage to generate inorganic nitrite (NO2-) which is then oxidized at a glassy carbon electrode (no other species have been identified at this time) (Krull and LaCourse, 1986; Krull et al., 1983; Bratin et al., 1984; Krull et al., 1984; Krull et al., 1984; Krull et al., 1989; Selavka et al.,1986). Photo-assisted electrochemical detection (PAED), the detection scheme investigated here, operates in the oxidative mode and at applied potentials where dissolved oxygen is not a problem. Hence, PAED exhibits higher sensitivity than reductive amperometry.

The combination of HPLC-PAED known for its inherent sensitivity and selectivity for organic nitro compounds (Krstulovic and Brown, 1982; LaCourse, 1997), with on-line solid-phase extraction (SPE) for environmental samples in a compatible platform (Marple and LaCourse, 2003). The function of SPE is fractionate and concentrate the analytes of interest based on simple chromatographic principles. SPE eliminates sample handling and reduces dramatically (in combination with the increased sensitivity of PAED) the amount of sample required for analysis from 1L (required by Method 8330) to 2mL. It is this technique that makes the platform unique and compatible with the on-site analysis of explosives.

Methods
Reagents and solutions.

Purified water were used to made all solutions. Water was purified by using a reverse osmosis system coupled to multi-tank/ultraviolet ultrafiltration stations. All solvents are HPLC grade. All solvents were filtered with a Fisher vacuum filtration apparatus utilizing a 0.2µm PTFE membrane filter.

The standard solutions were purchased as mixtures denoted as either calibration mix 1 or calibration mix 2 at concentrations of 1000µg/mL each in 1mL acetonitrile. Table 1 lists the name, peak number, abbreviation, EPA classification (mix A or B), and Restek classification (1 or 2) of all the explosives of EPA Method 8330. Standard solutions were stored in a refrigerator at 4°C. All stock solutions were prepared daily. Ground water was obtained from Columbia Technologies (Baltimore, MD) and stored at 4°C until use.

Procedure

The glassy carbon electrode was cleaned daily prior to use. The electrochemical cell is disassembled, and the working electrode is detached for polishing. A small amount of electrode polishing compound was poured onto a POLYPAD Gemstone Polishing Pad and placed the pad on a flat surface. The electrode was rubbed in a figure eight motion in the polishing compound for approximately 30s. It is important to hold the electrode flat against the pad to avoid rounding of the block which would result in cell leakage. All polishing compound were rinsed off from the working electrode with copious amounts of deionized water. Next, methanol was rinsed to remove oils deposited from the polishing compound. Follow up with a final rinsed of deionized water. All traces of polishing compound been removed because trace particulates on the electrode surface will alter electrode response. The cell was resembled and place it back into the LC-30 oven.

Precaution steps

Nitro explosives are known toxins and carcinogens, so they should be handled with gloves in a fume hood. Skin, eye contact and ingestion must be avoided.

Result and discussion
Model compounds

The main compounds used in this experiment contributed from 14 nitro explosives. They are emphasis on RDX, TNT, and Tetryl which are commonly found at contaminated sites. EPA Methods 8330 utilizes UV detection at 254nm exploiting the inherent optical activity of each aromatic explosive. However, HMX and RDX both of them do not respond as well to this method because only compounds contain at least one nitro (-NO2) will show good reading by PAED.

System Optimization
Optimization for direct injection method

To optimize the PAED detection potential, the technique of hydrodynamic voltammetry (HDV) at a constant flow rate of 1.0mL/min was performed at the buffer pH values of 3.5,4.5 and 5.5 over a range of 0.2-1.0V. The resulting data was plotted as S/N (analytical signal-to-noise ratio) versus applied potential, as shown in Fig. 2A for HMX (-), RDX (cdots, three dots, centered), Tetryl (—), and TNT (··-··). From 0.7V the signal-to-noise start increasing slowly. The analytes show a maximum signal-to-noise ratio at 1.0V.. Beyond 1.0V, however, the analyte response begins to decrease while the noise increases, thus decreasing the signal-to-noise ratio. Based on the result, HMX and RDX show the higher signal-to-noise raiot. It is due to the fact the N-N bonds in nitramines are much weaker than the C-N bonds of nitro aromatics), and the photolytic cleavage of N-N bonds is much more efficient. As result more nitrite was formed from nitramines than from nitro aromatics. Therefore higher S/N ratio detected for HMX and RDX. Varying the pH showed little effect, so a pH of 4.5 and applied potential of 1.0V were chosen as optimum values for PAED.
Full-size image (12K)

Here, the potential used was 1.0V and pH=4.5, and the flow rate was varied between 0.4 and 1.4mL/min based on Fig. 2B. At flow rates below ca. 0.8mL/min, chemical degradation of electroactive species occurs due to the residence time in the photoreactor is too long (Krull, I.S et al., 1983). The analytes show a maximum signal-to-noise ratio at 1.0mL/min, and all faster flow rates do not allow enough time for the generation of the maximum amount of photoproducts. All nitro compounds responded similarly to those shown in Fig. 2B, and an optimum flow rate of 1.0mL/min was chosen.

Fig. 2.(A) Hydrodynamic voltammograms for HMX (-), RDX (cdots, three dots, centered), Tetryl (—), and TNT (··-··) at pH=4.5. Mobile phase: 50% methanol in 20mM acetate buffer, pH=4.5; flow rate: 1.0mL/min; guard column: Phenomenex SecurityGuard with 4mm-3.0mm C8 cartridge; column: C18, 5µm, 4.6mm-250mm; column oven temperature: 30°C; electrode: 1.0mm glassy carbon; reference electrode: Ag/AgCl. (B) Plot of electrochemical S/N vs. flow rate for HMX (-), RDX (cdots, three dots, centered), Tetryl (—), and TNT (··-··). Mobile phase: 50% methanol in 20mM acetate buffer, pH=4.5; guard column: Phenomenex SecurityGuard with 4mm-3.0mm C8 cartridge; column: C18, 5µm, 4.6mm-250mm; column oven temperature: 30°C; electrode: 1.0mm glassy carbon; reference electrode: Ag/AgCl; applied potential: 1.0V vs. Ag/AgCl.

Optimization of on-line SPE

In order to perform on-line SPE, sample is loaded into the 2mL injection loop, and the prep injector is electronically actuated, which allows the SPE solvent to flow through the loop and carry the sample to the SPE column. The prep injector is returned to the load position after an optimized “wash” time that concentrates the analyte and rinses off potential contaminants, while the 6 port valve is turned to the inject position simultaneously. This procedure allows the HPLC mobile phase to backflush through the precolumn and elutes the analytes of interest onto the analytical column. This valve remains open throughout the entire chromatographic run. At the end of the run, a 5min wash with 80% methanol cleans the SPE column, which is followed by a 5min equilibration with “wash” solution prior to the next injection.

A wash solvent of 7.5% methanol in a solution of 20mM sodium acetate trihydrate (pH=4.5) and 0.5M sodium chloride was determined to give the greatest sample cleanup. As in the “salting out” extraction process of EPA Method 8330, sodium chloride was added to increase the retention of the explosives from the matrix into the C18 phase of the SPE column. The methanol was needed to wet the C18 phase, and the sodium acetate buffer was added to matrix match the wash solvent and the HPLC mobile phase.

Direct injection of explosives

The UV and PAED chromatograms under optimized separation and detection conditions are shown in Fig. 3 (A: UV; B: PAED). The use of two detectors show three distinct modes of selectivity. First, inherent to all chromatographic methods, is the selectivity afforded by the chromatography and the comparison of retention times of standards versus analytes in a sample. This is due to the nitro compounds cannot be oxidized directly, no detectable signal for the compounds of interest when the lamp in the photochemical reactor is turned off. Next, the second mode of selectivity is that the analyte must be photoreactive and produce oxidizable products to be detected. Finally, the use of two detectors allows for the determination of response ratios for standards and those for analytes in a sample.

Fig. 3.Optimized separation of explosives: (A) UV at 254nm chromatogram; (B) PAED chromatogram. Mobile phase: 50% methanol in 20mM acetate buffer, pH=4.5; flow rate: 1.0mL/min; guard column: Phenomenex SecurityGuard with 4mm-3.0mm C8 cartridge; column: C18, 5µm, 4.6mm-250mm; column oven temperature: 30°C; electrode: 1.0mm glassy carbon; reference electrode: Ag/AgCl; applied potential: 1.0V vs. Ag/AgCl.
View Within ArticleFull-size table

View Within Article

Upon examination of the two chromatograms (Fig. 3A and B), the expected increased sensitivity achieved by PAED for the nitramines (HMX and RDX) and other compounds such as Tetryl and TNT are easily seen. The figures of merit for the model compounds are presented in Table 3A. These were determined using the direct injection method with the 100µL injection loop, and all solutions were made in deionized water. When comparing limits of detection, PAED (LODs 0.007-3µg/L) is more sensitive than UV detection (LODs 0.9-5µg/L) for almost all of the model compounds. For some of the more common explosives, there is an approximate 1000- and 100-fold increase in sensitivity for HMX and RDX, respectively, and an approximate 100- and 10-fold increase in sensitivity for Tetryl and TNT, respectively, over UV detection. Furthermore, the PAED limits of detection are much lower here than those previously described for h?-EC . (Yinon, 1999; Krull et al., 1983; Krull and LaCourse,1986; Krull et al., 1983; Bratin et al., 1984; Krull et al., 1984; , [12] I.S. Krull, C. Selavka, X-D. Ding, K. Bratin and G.A. Forcier, Curr. Sep. (1984) (5), p. 57.Krull et al., 1989; Selavka et al.,1986)

The slope of the line (m) and the y-intercept (b), were included to show the calibration sensitivity of the detection methods and that there is no system bias in the method, respectively. And also, R2 values determined by linear regression analysis show that UV and PAED are of comparable linearity over the concentration range tested, which is at least four orders of magnitude in both cases. The percent relative standard deviation (%R.S.D.), determined by seven injections at the approximate limit of quantitation for each explosive, ranged from 0.80 to 3.41% for UV detection and 0.46 to 6.70% for PAED, all under 15% as required by the Resource Conservation and Recovery Act (RCRA) were described in Section 3.5.

On-line SPE analyses

A large background was present on the electrochemical detector that completely overwhelmed any signal from the explosives present when a real ground water sample was injected on the existing system, as shown in Fig. 4A. This is due to the electroactive species present in the groundwater (e.g., urea, salt etc.). PAED limits the analyzation of these species, therefore on-line SPE was chosen for reasons stated earlier. When those electroactive species eliminated, the high background was virtually eliminated, as seen in Fig. 4B.

Table 3B displays the figures of merit for the nitro explosives in deionized water using the SPE system. Preconcentrated limits of detection ranged from 0.0007 to 0.4µg/L for PAED and from 0.04 to 0.4µg/L for UV detection. It should be noted that on-line SPE could be used with Method 8330 alone to increase its sensitivity without incorporating PAED, but PAED should be used for Method 8330 enhancement. This concentration factor will turn the need of 1L of ground water sample to 2mL water sample. Linearity is retained when using SPE, with an average R2 value of 0.99962 (PAED and UV). The %R.S.D.s ranged from 1.17 to 5.38%, comparable to those by direct injection and still within the Resource Conservation and Recovery Act (RCRA) guidelines.

Conclusions

The instrumentation for determining explosives in ground water has been developed, optimized, and validated. It provides the advantages of increased sensitivity and selectivity over EPA Method 8330. On-line SPE allows for on-site analysis compatibility, reducing the required sample amount from 1L (EPA Method 8330) to 2mL, and performing sample preparation on-line, thus minimizing sample handling and allowing for “real-time” analyses. The developed platform is a unique analytical tool with increased sensitivity and selectivity, enabling faster, more accurate site assessment.

Case Study 3:

Determination of opiates, amphetamines, and cocaine in human hair by gas chromatographic-mass spectrometric (GC-MS) confirmation.

Introduction

The importance of hair analysis in drug testing has grown rapidly in recent years because of this technique gives long-term information on drug use, and gives complementary information to other biological matrixes (urine or blood) analyses and may offer crucial data in evaluating, interpreting and concluding with the obtained results. Nowadays, hair analysis is routinely used as a powerful tool for the detection of drug use, not only in forensic science but also in clinical toxicology or in traffic medicine. The wide uses of this kind of analysis is due to the progress of separation techniques and the increased sensitivity and selectivity of analytical instrumentation, which able to detect small amount of drugs contained in hair. Morphological, serological and chemical examination of human hair for forensic and medical expertise was initiated some years ago. A single human hair sometimes is the only remain at the crime scene. In most cases, it serves to confirm or exclude a possible suspect. Because hair does not decompose like body tissues and fluids, it is the main reason that has been used by forensic pathologists to determine the circumstances of death.

Several months after the last intake, the drugs can be detected. They enter hair roots from the capillaries and are embedded in the hair stalk, which grows at a rate of approximately 0.9-1.2 cmmonth-1 (Harkey and Henderson, 1989). Therefore hair can be used as a “calendar” of past exposure to drugs. Drugs can be detected in hair tissue weeks or months after intake. Exogenous compounds are incorporated into hair tissue at the root. They reach the growing hair matrix from capillary blood surrounding the hair germination centre, from skin-gland secretions (Henderson, 1993) and, in some cases, form the external environment (Smith and Kidwell, 1997). The low metabolic activity of the hair shaft, and the protection exerted by the hair matrix components, contribute to the stability of the embedded compounds. Although contamination of the hair by drugs present in the environment (Cone et al., 1991), by hair bleaching, and by hair dyeing (Skopp et al., 1997) may affect the accumulation of chemicals in the hair.

In biological matrix, hair gives some advantages which is, can be easily obtained without violating individual privacy and it can be stored and transported without specific precautions due to its stability. Aware of the disadvantages of urine as biological specimen for drugs of abuse (Bosomworth, 1993; James and Moore, 1997) and the advantage of hair, we extended our routine urine analysis (Karacic and Skender, 2000) to hair analysis. Commonly, the papers on hair analysis is deal with cocaine, followed by opiates and amphetamines (Nakahara, 1999).

Materials and methods
Reagents and standards

Morphine sulphate, codeine, heroin hydrochloride, methadone hydrochloride, cocaine hydrochloride, and methaqualone hydrochloride, 6-acetylmorphine, amphetamine, methamphetamine, MDA, MDMA, MDEA, propionic acid anhydride, heptafluorobutyric acid anhydride (HFBA) pyridine, and Bond Elut Certify columns.

Hair samples

Hair samples were collected from 36 young people ranging from 16 to 22 years of age. Screening tests of drug abuse were positive for most of them. About 5mm in diameter of hair was cut from close to the scalp at the vertex posterior area, folded in aluminium foil, and the proximal and distal ends marked. We analysed samples 2-4cm long.

Hair analysis

The hair was washed twice in dichloromethane for 2min at room temperature. It was dried before cut into very small pieces of less than 1mm, and 50mg was used for analysis. It was not possible to analyse for all drugs at the same time.

Morphine, codeine, heroin, 6-acetylmorphine, methadone and cocaine

Methanol (2ml) as an extraction solvent and 200ng of methaqualone in methanol solution of 2µg/ml as internal standard (IS) were added to 50mg of hair in a 10ml screw-cap tube. The samples were incubated for 18hours in a 40°C water bath. The methanol was then collected, the remaining hair was rinsed with 0.5ml methanol, and both fractions were evaporated to dryness at 40°C under a stream of nitrogen.

Clean-up procedure and derivatisation

Solid-phase extraction was used to purify hair extracts prior to analysis. Bond Elut Certify columns were conditioned with 2ml of methanol and 2ml of a 0.1M phosphate buffer at pH 6.0. After methanol evaporation, the dry residue of the hair extract was added to 2ml of 0.1M phosphate buffer at pH 6.0 and was poured into the conditioned columns. The sample passed very slowly through the column. Then, 2ml of deionised water, 1ml of 0.1M acetic acid, and 2ml of methanol were added in that sequence. The cartridges were dried under full vacuum for 5min and eluted with a 1ml (3-) mixture of dichloromethane:2-propanol:ammonium hydroxide (80:20:2, v/v/v). The eluents were collected in glass tubes and evaporated to dryness at 40°C under a stream of nitrogen. 100 µl of pyridine and 30µl of propionic acid anhydride were added to the residues and the tubes were capped, vortexed, and heated at 60°C for 30min. Followed evaporation to dryness, reconstitution in 100µl of ethyl acetate, and GC/MS analysis.

Amphetamine, methamphetamine, MDA, MDMA, and MDEA

1 µl of 1M sodium hydroxide was added to every 50mg hair sample. Then the samples were hydrolysed for 20min at 70°C and cooled. Followed extraction with 1ml (2-) of ethyl acetate and evaporation to dryness in the presence of a 100µl mixture of methanol:hydrochloric acid (99:1, v/v). 50 µl of ethyl acetate and 50µl of HFBA were added to the dry residues and the tubes were capped, vortexed, and heated at 60°C for 30min. Followed evaporation to dryness, reconstitution in 100µl ethyl acetate, and GC/MS analysis.

Each batch of samples (A, B) included standards for drug abuse/metabolites, negative control, and genuine positive sample.

Standard preparation

Stock solutions containing 2µg/ml of (A) morphine sulphate, codeine, heroin hydrochloride, 6-acetylmorphine, methadone hydrochloride, and cocaine hydrochloride and (B) amphetamine, methamphetamine, MDA, MDMA, and MDEA, were prepared in methanol and stored at -20°C. Standard calibration curves were obtained through the described methods using 20, 50, 100, 200, 400, and 800ng of the stock solution (A) or (B), 200ng of methaqualone as IS (only for A), and 50mg of blank control hair, previously washed and cut into very small pieces. Blank control hair samples were obtained from coworkers in laboratory.

Limit of detection (LoD) for the analyte was determined by decreasing concentrations of spiked samples until the response equalled signal-to-noise ratio (S/N) of 3.

GC/MS analysis

For this experiment, the analysis was performed using a Varian 3400 CX GC with Saturn ion trap mass spectrometer (mass selective detector, MSD). The chromatographic column was RTX-5 (5% diphenyl-95% dimethyl polysiloxane, 30m 0.25mm i.d.). For the analysis of morphine, codeine, heroin, 6-acetylmorphine, methadone, and cocaine (A), the initial column temperature of 50°C was held for 1min, then programmed to 300°C at 50°C/min and held for 6min. For the analysis of amphetamine, methamphetamine, MDA, MDMA, and MDEA (B), the initial column temperature of 50°C was held for 1min, then programmed to 225°C at 20°C/min and held for 1min, then programmed to 260°C at 50°C/min and held for 1min.

Ultra-pure grade helium was used as the carrier gas at a flow rate of about 1ml/min. Septum-equipped programmable injector (SPI) was used; the initial temperature of 40°C was held for 0.1min, then programmed to 280°C at 200°C/min and held for 8min. The transfer line temperature was 260°C.

The MSD was operated in the selected ion-monitoring mode. Analytes (A) were identified and quantitated using a comparison with the retention times and relative abundance of three confirming ions to methaqualone (IS). The external standard method of quantitation was used for amphetamines (B). Quantitative calculations were automated by Saturn® software. For each drug the following ions were used: methadone, m/z Image, 309, 165; cocaine, m/z Image, 82, 303; codeine propionyl, m/z Image, 282, 341; heroin, m/z Image, 327, 369; 6-acetylmorphine propionyl, m/z Image, 268, 383; morphine dipropionyl, m/z Image, 268, 397; and amphetamine HFBA, m/z Image, 240, 91; methamphetamine HFBA, m/z Image, 210, 118; MDA HFBA, m/z Image, 162, 240; MDMA HFBA, m/z Image, 254, 210; MDEA HFBA, m/z Image, 135, 162; and methaqualone, m/z Image, 250, 91. The underlined ions were used for quantitation.

Results and discussion

Possible conversions of heroin to 6-acetylmorphine and morphine, and 6-acetylmorphine to morphine during the extraction were checked by adding 200ng of heroin or 6-acetylmorphine (three times each) to 50mg of the blank sample and analyses were performed as described. Heroin hydrolysed up to 18% to 6-acetylmorphine and up to 1% to morphine, and 6-acetylmorphine up to 6% to morphine. This statement agree with the findings of Kintz et al. (1998).

The hair samples only washed twice in dichloromethane because the third wash was always negative, although the two previous washes were positive (Kintz and Margin, 1995)

Methanol turns to be the solvent of choice for extracting heroin and 6-acetylmorphine (Cone et al., 1993; Rothem and Pragst, 1995), although acidic extraction (Kintz and Margin, 1995; Jurado et al., 1995) and enzymatic digestion (Hold et al., 1998; Moeller et al., 1993) are also used for the extraction of the same substances from hair. Kintz and Cirimele (Kintz and Cirimele, 1997) found that the best recoveries for amphetamine, MDA, and MDMA were observed after alkaline hydrolysis of hair.

Different of derivatisation reagents are used in the analysis of drugs of abuse (Segura et al., 1998). A mixture of propionic acid anhydride and pyridine was found very convenient and superior to N,O-bis(trimethylsilyl) trifluoroacetamide (BSTFA) for derivatisation of codeine, 6-acetylmorphine and morphine. Namely, propionylation formed very stable derivatives with good GC properties. Another advantage of propionylation over BSTFA is that it does not decrease the resolution power of the capillary column, as it evaporates before analysis. HFBA is often recommended for derivatisation of amphetamines (Thurman et al., 1992).

Predominant analytes were 6-acetylmorphine (18subjects,50%) and morphine (16 subjects, 44.4%). In most subjects, the presence of 6-acetylmorphine (specific heroin metabolite) confirmed heroin consumption which is consistent with their statements. It showed both 6-acetylmorphine and morphine in 16 subjects,. The 6-acetylmorphine/morphine ratio (mean: 2.84; median: 2.11; range: 0.69-8.26) was similar to that found by Moeller et al. (Moeller et al., 1993) although individual concentrations were very different and the latter study involved far more subjects. Higher concentrations of 6-acetylmorphine than those of morphine are usual for consumers of heroin. Only rarely does the morphine concentration exceed the concentration of 6-acetylmorphine (Nakahara, 1999; Kintz et al., 1998; Moeller et al., 1993). In the result observed, only two such cases among our subjects (No. 11 and 34) and they concerned colored hairs. The morphine/6-acetylmorphine ratio (mean: 0.55; median:0.44; range: 0.15-1.45) is in accordance with results of Gaillard and Pepin (Gaillard and Pepin, 1997). Heroin was found in only eight hair samples, which is consistent with literature (Cone et al., 1993).

Kintz et al. (Kintz et al., 1998) found no codeine in subjects to whom heroin was administered in the controlled study and explained that result with high heroin purity. Cocaine was present in four hair samples and in two together with heroin. Of amphetamines predominant drug was MDMA, mostly alone. In two samples no drug was found (subjects 6 and 16) and in two only 6-acetylmorphine was present without morphine (subjects 22 and 28). The reasons of decreasing tendency of the hair drug content because of cosmetic treatment (Cirimele et al., 1995; Potsch and Skopp, 1996; Jurado et al., 1997; Skopp et al., 1997). Examples, bleaching, colouring, or permanent waving were found to affect the stability of incorporated drugs and to cause alterations of the fibres at an ultra-structural level. This may result in partial or complete loss of drug substances, depending on particular drug molecule and on its concentration prior to cosmetic treatment.

Conclusion

These methods have been validated and found acceptable for the analysis of drugs of abuse in hair. Despite uncertainties related to the analysis of died or otherwise treated hair, the methods were able to establish drug abuse. Despite the small number of subjects, this method can be used to indicate the trend in drug abuse among people in the world.

Case study 4 : Detection of cyanobacterial hepatotoxins by thin-layer chromatography and application to water analysis

Introduction

Blooms cyanobacteria (blue-green algae) have been identified in fresh and brackish water bodies all over the world. Certain species of cyanobacteria are potential of producing toxin, thus cause a significant water quality problem,.

There are three groups of toxic compounds associated with cyanobacteria: lipopolysaccharide endotoxins, neurotoxins, and hepatotoxins (McElhiney and Lawton, 2004). Lipopolysaccharide endotoxins are elements of the gram-negative cell wall, and therefore, common to all cyanobacteria and the certain strains of cyanobacteria will produce neurotoxins and hepatotoxins. These toxin may release into water.

In this experiment, the most vast studied group of cyanobacterial toxins are the hepatotoxic cyclic peptides which involved the microcystins and nodularins. Microcystins is the species belongs to the genera Microcystis, Anabaena, Oscillatoria (Planktothrix), Nostoc, and Anabaenopsis. While nodularin has a similar structure and mode of toxicity with microcystin. , Nodularia spumigena is the only cyanobacteria may extract nodularin (Sivonen and Jones, 1999).

Microcystin in water bodies may cause severe harm in wild and domestic animals worldwide (Sivonen and Jones, 1999). Consumption of contaminated drinking water or activities such as swimming easily to be exposed to microcystins (Kuiper-Goodman et al., 1999). In other way, intoxication may be through the consumption of contaminated foods. As examples, the toxins is accumulate in certain species of freshwater mussel (Eriksson et al., 1989; Vasconcelos, 1995), and fish (Carbis et al., 1997). However, other studies have also stated that consumable plants is anohter route for exposure to microcystins (Codd et al., 1999; McElhiney et al., 2001).

In animals, cyanobacterial hepatotoxins inhibit protein phosphatases (PP) 1 and 2A in the liver (Yoshizawa et al., 1990). Acute toxicity is different due to the species of animal and physiological condition (Kaya, 1996). If chronic exposure to microcystins it may cause tumour growth (Nishiwaki-Matsushima et al., 1992) and exposure to nodularin cause cancer (Ohta et al., 1994). A guideline set by World Health Organisation (WHO) set the concentration value of 1µg/l of microcystin-LR in drinking water (Falconer et al., 1999). The first reported fatal acute poisonings of humans by cyanobacterial hepatotoxins occurred in Caruaru, Brazil, in 1996, which cause death of 50 people due to the use of microcystin-contaminated water in hemodialysis treatment ( Pouria et al., 1998; Jochimsen et al., 1998). So there is monitoring raw and treated waters for cyanobacterial hepatotoxins exists in all countries to avoid water poisoning.

High-performance liquid chromatography (HPLC), coupled with ultra violet (UV) is the common method used to detect cyanobacterial hepatotoxins in water samples (Lawton et al., 1994) or mass spectrometric detection ( Tsuji et al., 1994; Kondo and Harada, 1996), enzyme-linked immunosorbent assays (ELISA) ( Chu et al., 1990; Nagata et al., 1995; McDermott et al., 1995), and PP inhibition assay ( An and Carmichael, 1994; Lambert et al., 1994). For the screening analysis of cyanobacterial bloom samples, thin-layer chromatography (TLC) was applied ( Pelander et al., 1996). But, it is less using due to high-cost effective technique.

The derivatization of purified cyanobacterial hepatotoxins to coloured or fluorescent products was investigated. An analytical method for raw and treated water samples, utilizing one of the derivatization reactions is described. In this case study, 38 water samples investigated by protein phosphatase inhibition assay (PP) and enzyme-linked immunosorbent assay (ELISA) and then analysed blind by the TLC.

Experimental
Materials

Purified toxins microcystin-LR (MCYST-LR), microcystin-RR (MCYST-RR), microcystin-YR (MCYST-YR), microcystin-LA (MCYST-LA) and nodularin . All reagents used in the derivatization studies include 4-methoxybenzaldehyde (anisaldehyde), 3,3,5,5-tetramethylbenzidin (TMB) and polyethylene glycol. Water was Alpha-Q purified. In the solid phase extraction studies the following sorbents were used: Isolute C18(EC) (1g), and Isolute ENV+ (200mg), LiChrolut RP-18 (500mg), LiChrolut RP-18e (500mg), and LiChrolut EN (200mg), PolarPlus C18 (1g), and SDP (200mg). The natural water samples were collected from various environments during the summers of 1997 and 1998, and frozen as such.

Procedure

At the concentration of methanol at 0.5µg/ml, pure toxins were dissolved, and further dilutions made as needed. The quantity of each toxin applied to the plate was 0.5µg in the derivatization studies. In order to obtain sufficiently clear print-outs for the photographs, an amount of 1µg of each toxin was added. Detection limits of pure MCYST-LR were determined in three concentration series ranging from 1 to 25 ng (1, 2, 5, 10, 15, 20, and 25 ng), 10 to 250 ng (10, 25, 50, 100, and 250 ng), and 100 to 700 ng (100, 250, 500, and 700 ng). For the first screened the range of detection limit was began with the 100 to 700 series, followed by a series with lower concentrations when appropriate. The analysis was repeated on the selected range as duplicate, resulting in triplicate analysis in three different days. Two different people was evaluated the detection limit of the amount of toxin producing a derivative visible to the eye. All applications to plates were performed by an Automatic TLC sampler III in the spray mode. The application volume was 10µl and bandwidth 4mm. Plates were developed in twin-trough chambers for 20-10cm plates at ambient temperature. The mobile phase consisted of water-ethyl acetate-n-propyl alcohol (2+5+3) with 5% acetic acid addition (Pelander, A. et al., 1997). The plates were dried under a stream of warm air after development. In the derivatization reactions, all the reagents were sprayed on plates with pressurized air employing a spraying device. A TLC Plate Heater III was used to heat the plates in the derivatization studies. A Zymark TurboVap LV evaporator was used to evaporate the solid phase extraction eluates.

Derivatization reactions
Anisaldehyde

A quantity of 0.8ml of concentrated sulphuric acid and 50µl of anisaldehyde to a mixture of 8.5ml of methanol and 1.0ml of glacial acetic acid were added to prepare spray solution. The plate was heated at 100°C for 5min after sprayed with the solution and, then further sprayed with 40% polyethylene glycol (PEG 400) solution in methanol to enhance the fluorescence. The effect of the temperature was tested between 75 to 150°C in increments of 25°C.

Antimony(V)-chloride

By adding 1ml of antimony(V)-chloride to 4ml of tetrachloroethylene, the spray solution was prepared. The developed plate was sprayed with the solution and heated at 150°C for 10min, and further sprayed with the PEG 400 solution and effect of the temperature was tested at 125, 150 and 175°C.

N,N-DPDD and N,N,N’,N’-TPDD

The quantity of 100mg of N,N-DPDD or N,N,N’,N’-TPDD was added into a mixture of 5ml of methanol, 5ml of water and 0.1ml of glacial acetic acid to produce spray solution. Then, the developed plate was exposed to chlorine gas for 20min in a glass tank. By mixing equal volumes of 10% hydrochloric acid and 5% potassium permanganate solutions in a beaker, chlorine gas was generated. After the exposure the plate was aired in a stream of warm air for 15seconds and evenly sprayed with the reagent solution.

Manganese(II)-chloride

The spray solution was prepared by dissolving 20mg of manganese(II)-chloride in 3ml of water, 3ml of methanol and 0.2ml of concentrated sulphuric acid. The plate was sprayed with the solution and heated for 10minutes at 150°C. The effect of the temperature was studied from 50°C to 200°C in increments of 50°C.

Sulphuric acid

The plate was sprayed with 40% aqueous sulphuric acid solution and heated or 10minutes at 100°C, then further sprayed with PEG 400 solution. The effect of the temperature was studied from 50°C to 125°C in 25°C increments.

Vanillin

The spray solution was produced by dissolving 50mg of vanillin in a mixture of 1ml of glacial acetic acid, 0.5ml of concentrated sulphuric acid and 8.5ml of methanol. The plate was sprayed and heated at 100°C for 5min and next sprayed with the PEG400 solution. The effect of the temperature was tested at 100 and 150°C.

Phosphomolybdic acid

The spray solution was prepared by dissolving 200mg of phosphomolybdic acid in ethanol and adding 0.4ml of concentrated hydrochloric acid. The plate was sprayed and heated for 10minutes at 150°C. The effect of the temperature was tested at 100 and 150°C.

TMB

The spray reagent was prepared by dissolving 8mg of TMB in 1.5ml glacial acetic acid, 25ml of water and 50mg of potassium iodide were added. The plate was then exposed to chlorine gas, as stated for N,N-DPDD and sprayed with the reagent.

Water sample pretreatment

A amount of 500ml samples of natural brackish water fortified with MCYST-LR at a concentration of 1µg/l. The fortified samples was undergo Solid Phase Extraction(SPE) and then performed with Isolute C18(EC) (1g) cartridges which conditioned with 10ml of methanol followed by 10ml of purified water. The cartridges were washed with 10ml of water after the application of the samples, followed by 10ml of 20 % methanol, and the toxins eluted with 5ml of methanol. Next, the eluent was then evaporated until dry in a 60 °C water bath. The next step is residue reconstituted with 30µl of 70 % methanol.

By using N,N-DPDD in the water analysis method, sample volume was 50ml and LiChrolut RP-18e (500mg) cartridges were employed for the SPE. The extraction procedure was matched with previously described, except only 10ml of water was used in the wash step. Spiked reference standards were prepared by adding MCYST-LR, MCYST-RR and NODL to nontoxic fresh raw water from Lake Päijänne in concentrations of 1 and 5µg/l, and the standards were treated identically to the samples. Two standards at the higher concentration (tracks 1 and 16 on the edges of the plate) and one standard at the lower concentration (track 9 in the middle) were applied to each plate.

Result :

Visualization reactions

There is a total of 17 potential visualization reactions for the cyanobacterial hepatotoxins were tested in the previous studies, of which nine produced either colored, fluorescent or both. The detection limit of MCYST-LR in each reaction is shown in Table 1 and the products are described in Table 1 and presented in Fig. 2 and Fig. 3.

Applicability of the reactions for water analysis

Visualization reactions for water analysis was studied in natural raw water samples braced with MCYST-LR. Matrix interferences were observed in all reactions, even severe with the fluorescence mode. The reactions met the WHO guideline value which is 1µg/l of MCYST-LR were those with anisaldehyde (visible mode), N,N-DPDD, N,N,N’,N’-TPDD, and phosphomolybdic acid. Detection limits with N,N-DPDD and N,N,N’,N’-TPDD were ten times better than other reactions. N,N-DPDD was chosen due to its higher specificity and lesser background interference than N,N,N’,N’-TPDD.

Sample pretreatment for the N,N-DPDD reaction

By using the Isolute C18(EC) solid phase extraction cartridges, it was detected severe interference for MCYST-LA and MCYST-RR. In order to find a cartridge type with less interference, reversed phase and polymer-based adsorbent materials were tested, and LiChrolut RP-18e was chosen for further studies. With the end-capped materials, the recovery of MCYST-RR was systematically better.

LiChrolut RP-18e cartridges did not produce the first interfering spot, however present in all cartridges and could not be removed at the second interfering spot. The interference co-eluted with MCYST-LA, and consequently MCYST-LA had to be excluded from the water analysis study. The character of the interfering spots remained not clear.

Thirty-eight fresh and brackish water samples were tested by ELISA and PP previously were analyzed blind with the N,N-DPDD method. The results has shown in Table 2. In samples 1-11, 14, 16-23, 25-29, and 33-38, TLC results were in agree with the ELISA and PP results. For the sample 12, concentration obtained by ELISA (0.7µg/l) was below the detection limit of the TLC method (1µg/l). By ELISA and PP, the concentrations obtained different from 0.2 to 1.3µg/l. Sample 15 contained 0.5µg/l MCYST by ELISA, so this is not detected by the PP method (0.6µg/l). For samples 13 and 24, they appeared false positive results, which appeared to have elevated contents (>5µg/l) of MCYST-RR by TLC. While for samples 31 and 32, MYCST were not detected by TLC due to the sample volume is less than the recommended volume (50ml).None of methods detect the availability of MYCST in sample 32. The last three samples, 35-38 were brackish water, and 37 and 38 contained significant algal material. Samples 37 and 38 had more severe background interference in all four samples. However, nodularin still detected in these samples, which is the expected toxin in brackish water samples.

Discussion :

The derivatization could occur in microcystin and nodularin at the N-methyldehydro-Ala double bond of microcystin, the 2(Z)-dehydrobutyric acid double bond of nodularin, and the Adda diene structure for microcystin and nodularin. The experiment were failed to produce fluorescent products by derivatizing the carboxylic groups, so Adda diene structure was considered the most favourable site for the derivatization.

In this experiment, the effect of temperature were varied for each reaction, thus the temperature range was not uniform(refer to procedure). Heating was necessary for the reactions with anisaldehyde, antimony(V)-chloride, manganese (II)-chloride, sulphuric acid, vanillin, and phosphomolybdic acid. Among these reagants, anisaldehyde and sulphuric acid were the most sensitive for temperature changes.

By over-spraying with viscous lipophilic or hydrophilic substances, TLC possible to enhance and stabilize fluorescence. Except for the manganese(II)-chloride the fluorescence enhancement step was necessary with all reactions.

By using N,N-DPDD , the sample volume could be reduced from 500ml to 50ml because the detection limit of pure MCYST-LR was only 10 ng, which is lesser than other reagent when operating at the WHO guideline toxin concentration of 1µg/l. Therefore, this can makes the sample handling, storage and pretreatment simpler and faster.

The present method is qualitatively more informative compared to ELISA and PP, where results are shown as MCYST-LR equivalents, as the identification of individual toxins is possible. Unfortunately, very less toxins are available commercially. But, in this study some parts were detected by TLC, identified as unknown microcystins due to their Rf-values, based on Rf-data obtained in the earlier studies (Pelander et al., 1997). They were not given names, but were likely demethylated variants of MCYST-LR and MCYST-RR. TLC results were agree also with the concentrations measured by ELISA and PP. A comparison with mass spectrometric structure elucidation would be needed to evaluate the present method,

The most widely techniques used in research and for routine analysis of these toxins is HPLC with detection using either photodiode array or mass spectrometry. HPLC can give precise result for both quantitative and qualitative data. But, the method is technically demanding and expensive, and its result often depends on the availability of a range of toxin standards. HPLC detection of the toxins in natural samples can be extremely time consuming due to the collection on on-site sampling followed by sample processing and analysis in the laboratory. Biological-based assays give a simpler and less expensive alternative way. The sensitivity and selectivity given by ELISA offered a accurate estimation of microcystins with minimum sample processing. However, the commercially available ELISA kits only capable of determining toxicity in terms of microcystin-LR equivalence.

In order to prevent possible exposure ways for these toxins into humans and animals, we need to analyze a wide range of sample matrices for microcystins, involving plant and animal tissues. Due to the interferences of sample matrix components, the detection and quantification of microcystins can be complicated. Recently, a commercially available ELISA kit has recently been used to determine the presence of microcystins-LR in the tissues of edible fish and crustaceans, indicating a possible threat to human health (Magalhães, V.F., Marinho, M.M., Domingos, P., Oliveira, A.C., Costa, S.M., Azevedo, L.O. and Azevedo, S.M.F.O., 2003. Microcystins (cyanobacteria hepatotoxins) bioaccumulation in fish and crustaceans from Sepetiba Bay (Brasil, RJ). Toxicon 42, pp. 289-295. Article | PDF (316 K) | View Record in Scopus | Cited By in Scopus (59)Magalhães et al., 2003).

Conclusion

The results presented by TLC water analysis method are comparable to ELISA and PP results. The advantages of the TLC method would be cheaper price and manual operation. Besides, TLC analysis is also easy to perform therefore suitable for small-scale laboratories with simple resources. However, the most widespread conventional screening method for cyanobacterial hepatotoxins is probably ELISA. On the other hand, no sample preconcentration is needed in ELISA.

The detection limit was chosen referring to the WHO guideline of 1µg/l of MCYST-LR, and sample volume adjusted on that basis. With larger sample volumes, lower detection limits can be reached.