Specimen preparation and extraction
Enzymatic digestion of tissues
Enzymatic digestion involves the use of a robust proteolytic enzyme, such as subtilisin Carlsberg, to digest the tissue to yield an essentially aqueous matrix for extraction, and is suitable for general screening as well as for more specific analysis. It is simple, readily adaptable and provides a protein-free filtrate from which any drugs present may be extracted. It also provides enhanced extraction of many drugs compared with the tungstate, ammonium sulfate, or Stas–Otto methods. A suitable procedure is as follows: macerate 10 g of liver or other tissue with 40 mL of 1 M tris(hydroxymethyl)methylamine; add 10 mg of subtilisin Carlsberg and incubate in a water-bath at 50–60C for about 1 h, with agitation. Filter the digest through a small plug of glass wool to remove undissolved connective tissue. Aliquots of this digest may be substituted for the specified biological fluid in most routine screening procedures. The filtered digest has a pH of 8.0 to 9.5. This method is very useful for the analysis of sectioned injection marks. The superficial fatty skin layer is removed and the remaining muscle layer is analysed as above. If the injection was intramuscular and of recent origin, the drug concentrations should be greater than in a similar tissue sample that does not show an injection mark. If the injection was intravenous, such a distinction cannot be expected. As noted under Injection sites, a sample remote from the suspect injection site should also be analysed as a control for comparison purposes. The enzymatic digestion method’s superb ability to ‘liquidise’ solid tissue can be used for many purposes apart from drug analysis. The recovery of shotgun pellets and small bomb fragments in body tissues is one such application. The preparation of solutions for direct aspiration into atomic absorption instruments for the detection of some metals has also been studied (Lock et al. 1981), and its use for the detection of toxic anions and some pesticides also seems feasible. Drugs that form glucuronide conjugates may be hydrolysed with b-glucuronidase prior to extraction. For example, the determination of morphine in blood is often performed with and without the addition of b-glucuronidase to estimate the unconjugated (free) and ‘total’ (conjugated plus unconjugated) drug present, and therefore to aid interpretation. Other applications include enhanced detection of benzodiazepines and other drugs in blood and urine. A typical procedure for the enzymatic hydrolysis of glucuronides is to mix 1 mL blood or urine with internal standard and 1.5 mL buffer and then to add 100 μL of β-glucuronidase
obtained from Helix pomatia. Mix the solution and allow to incubate at 37Covernight (ca.16 h). After incubation, the pH of the solution is adjusted appropriately for solvent extraction or SPE of the drugs of interest. Acid hydrolysis may be used to cleave glucuronide conjugates. Although more rapid than enzymatic hydrolysis, this method is ‘harsher’ and should be restricted to acid-stable analytes such as morphine in urine. Acid hydrolysis of blood dramatically increases the amount of potentially interfering substances and may produce a denatured protein mass that reduces analyte recovery.
Liquid–liquid extraction still predominates in most laboratories. The choice of an appropriate solvent is often a matter of experience or tradition. The chosen solvent should ideally extract as much of the target analyte as possible, while minimising the extraction of endogenous substances. Ideally, a solvent should extract the target analyte with a reproducible efficiency of at least 50%, and preferably much higher. The pH of the specimen influences the extent to which acid and basic drugs are extracted. Addition of a weakly basic buffer, such as sodium borate (pH 9), favours the extraction of weakly basic drugs, as well as most neutral substances. Similarly, the addition of an acidic buffer, such as sodium dihydrogenphosphate, favours the extraction of acidic as well as neutral drugs. The majority of drugs of forensic interest are ‘basic’ in character, but are often present at relatively low concentrations in blood. It is therefore desirable to have an extraction scheme that incorporates a back-extraction step to eliminate or minimise the extraction of endogenous molecules. An example extraction scheme is shown in Figure 1. The sodium hydroxide will force the basic drugs into the lipid-soluble un-ionised form, allowing extraction into the chlorobutane. The chlorobutane is transferred to a fresh tube and the drugs are back-extracted into sulfuric acid. Neutral or acidic substances will remain in the upper chlorobutane layer, which may be pipetted or aspirated to waste. The remaining acid layer is then made basic by addition of sodium hydroxide, and the now un-ionised basic drugs re-extracted with chlorobutane. The upper solvent layer may then be removed and concentrated under nitrogen, prior to analysis by a suit able chromatographic method. This extraction scheme will give extracts that are relatively free of interfering substances. However, it should be noted that morphine and other amphoteric drugs cannot be detected by this method since the phenolic functional group will ionise at high pH, precluding extraction into the solvent. For amphoteric drugs, the basic phase should be less than pH 9.0, and preferably pH 8.0–8.5.
While a similar extraction scheme to that used for basic drugs (but with the additions of acid and base reversed to those of the scheme shown in Fig. 1), could be used for strongly acidic drugs, such a method does not efficiently extract weakly acidic drugs, such as the barbiturates, and neutral drugs, such as meprobamate. Conversely, simple addition of an acidic buffer to whole blood and extraction with a solvent result in the co-extraction of large amounts of endogenous lipid substances. Such extracts may be ‘cleaned up’ by partitioning between immiscible solvents of different polarities, such as acetonitrile and hexane, as shown in Figure 2. The more polar drugs tend to partition into the acetonitrile, whereas the endogenous lipids (fatty acids, sterols) tend to partition into the hexane. Solid-phase extraction (SPE) has been used for many years in clinical toxicology, although to a lesser extent for postmortem work. SPE usually results in better extraction efficiencies than liquid–liquid methods, especially for more polar analytes. One of the major obstacles to general acceptance in this field has been the difficult nature of postmortem specimens, which are often clotted and laden with solid material that can easily plug the fine material in SPE columns.
However, improvements in sample preparation techniques and SPE column technology have largely overcome these problems. Use of a good internal standard, appropriate dilution of the sample and centrifugation of residual solid material usually results in a solution that does not plug the columnand for which quantitativedetermin ation has not been compromised. Despite the inclusion of wash steps, the higher extraction efficiency of SPE columns compared with liquid–liquid extractions can sometimes result in dirtier extracts, although this need not be a problem if specific (e.g. MS) detection methods are used. Manufacturers of SPE columns readily provide sample extraction protocols.
Figure 1 Extraction pathway for strong bases.
Figure 2 Extraction pathway for acids and neutrals.
