Lakhasly

standards (IS), for example, for cortisol [31].The hydrolysis is itself a process that requires careful validation, since enzymatic hydrolysis varies in efficiency depending upon the chosen enzyme (e.g., b-d-glucuronidase), with regards to the cleavage of glucuronide and sulfate con- jugates.Urine steroid profiling by GC-MS is primarily concerned with the analysis of polar steroid metabolites and their sulfate/ glucuronide conjugates, such as pregnanetriol, androsterone and tetrahydrocortisol, and their excretion patterns and ratios.The common ionization method for GC-MS, electron impact ioniza- tion, is a 'hard' technique that causes exten- sive molecular fragmentation, generating mass spectra that are 'fingerprints' providing detailed structural information.Use of instruments with high mass resolution may enhance the distinction of targeted steroids that are isobars for nonsteroidal compounds at nominal mass, but offers no distinction of steroid isobars unless unique secondary fragment ions can be identified.Concentrations of endogenous steroids rela- tive to these interfering compounds vary in dif- ferent matrices, so that matrix-specific validation is required before applying one analytical tech- nique to an alternative matrix [41].Endogenous steroids are transported around the body extensively (>95%) bound to nonspecific plasma proteins (e.g., albu- min) and to specific steroid-binding proteins 2554 Bioanalysis (2011) 3(22) future science group

Challenges & benefits of endogenous steroid analysis by LC-MS/MS |Sampling A number of factors must be considered prior to sampling, including age, sex, body mass, fasting/ nonfasting, pubertal stage, menstrual cycling, pregnancy and diurnal/circadian rhythm [41,48], as these will heavily influence the reference inter- vals and interpretation of any findings.When used in MS/MS mode, an ion-trap instrument enables the distinction of hundreds of newly recognized components in urine from newborns with congenital adrenal hyperplasia (CAH) [35], leading to the establishment of the best diagnostic markers [36].Analyses have been developed to measure free hormone concen- trations (typically using equilibrium dialysis [42] or ultrafiltration [43]), although the majority of analytical methods measure total serum concen- trations after disruption of any protein binding.Whilst common isotopically labeled analytes are now available at reasonable prices (e.g., in the UK, 1 mg trideuterated testoster- one costs ~UKGBP114 [~US$185 at time of press]; LGC Standards, Teddington, UK), some stable isotope-labeled IS are either very expensive (as they require custom synthesis) or, for some minor metabolites, not easily available.A reliable method for the analysis of estrogens in serum from males and postmeno- pausal females, for example, should have a LOQ of the order of 1 pg/ml (~3 pmol/l, 1 ppt).In an example from vitamin D analysis, Bunch et al. used postcolumn infu- sion of isotopically labeled analyte as a surro- gate to quantify matrix effects in the analysis of 25-hydroxyvitamin D metabolites [62].For example, cortisol [31] has been used clinically in the diagnosis and moni- toring of endogenous steroid disorders in serum [32,33] and has been used in reference centers for more than 30 years for urinary steroid profiling.There is little dif- ference when using plasma or serum for steroid analysis, although gel separation tubes may cause problems for some hydrophobic/lipophilic steroids such as progesterone.It has been used clinically in the diagnosis and monitoring of endogenous steroid disorders using serum [32,33] and has been in use in refer- ence centers for more than 30 years for urinary steroid profiling.A relatively cheap benchtop system with a single-mass fil- ter (quadrupole or ion-trap) enables the routine distinction of approximately 100 steroid compo- nents.GC-MS of urinary steroid metabolites requires extensive sample preparation (extraction from the matrix, conjugate hydrolysis, clean-up and two-step derivatization).Isobaric steroids are, however, difficult to distin- guish since they show similar secondary fragmen- tation, thus chromatographic resolution is essen- tial.Samples for steroid analysis are relatively stable once col- lected [41], although tubes used for collection and during extraction should be evaluated to ensure steroids do not adsorb to their surfaces [49,50].future science group For LC-MS/MS, matrix effects (either ion suppression and/or ion enhancement) caused by coextracted interfering compounds in the MS source are well-documented [53-57], and should always be evaluated when implementing new methods [58-61].Sample preparation for such analyses therefore requires a hydrolysis step (enzymatic or acidic) to cleave the conju- gated moieties, prior to chemical derivatization.Precision of quan- tification is also relatively poor, almost certainly related to sample instability within the GC injector, but this can be overcome if isotopically- labeled IS are included.The ions observed are influenced by eluent and buffer composition [39].Challenges ??

standards (IS), for example, for cortisol [31]. It has been used clinically in the diagnosis and monitoring of endogenous steroid disorders using serum [32,33] and has been in use in refer- ence centers for more than 30 years for urinary steroid profiling. For example, cortisol [31] has been used clinically in the diagnosis and moni- toring of endogenous steroid disorders in serum [32,33] and has been used in reference centers for more than 30 years for urinary steroid profiling. This approach is still vital for the identification of steroid metabolic disorders and definition of previously unrecognized types [34]. A relatively cheap benchtop system with a single-mass fil- ter (quadrupole or ion-trap) enables the routine distinction of approximately 100 steroid compo- nents. When used in MS/MS mode, an ion-trap instrument enables the distinction of hundreds of newly recognized components in urine from newborns with congenital adrenal hyperplasia (CAH) [35], leading to the establishment of the best diagnostic markers [36]. GC–MS has recently been used to quantify 69 steroids in maternal blood, cord blood and amniotic fluid [37].
GC–MS of urinary steroid metabolites requires extensive sample preparation (extraction from the matrix, conjugate hydrolysis, clean-up and two-step derivatization). Precision of quan- tification is also relatively poor, almost certainly related to sample instability within the GC injector, but this can be overcome if isotopically- labeled IS are included. By contrast, LC–MS/MS of serum permits rapid analysis of small samples with minimal preparation, and sample injection in the liquid phase. The common ionization method for GC–MS, electron impact ioniza- tion, is a ‘hard’ technique that causes exten- sive molecular fragmentation, generating mass spectra that are ‘fingerprints’ providing detailed structural information. Optimization of one set of source conditions, with standardization of the ionizing voltage at 70 eV is all that is required [38]. High-resolution capillary GC provides a very reproducible order of elution for a given liq- uid phase, with a precision of retention time of approximately 0.05%. These enable single com- ponents to be identified by comparison with pure standards with an exceptional level of confidence. By contrast, LC–MS/MS ionization techniques are ‘softer’, providing fewer distinctive fragment ions. The ions observed are influenced by eluent and buffer composition [39]. Chromatography columns for LC give poorer peak resolution, although this is currently a very active field of development [40]. Injected samples may contain
high concentrations of interfering material. Fragmentation experiments in LC–MS/MS are used to enhance compound identification. Isobaric steroids are, however, difficult to distin- guish since they show similar secondary fragmen- tation, thus chromatographic resolution is essen- tial. Use of instruments with high mass resolution may enhance the distinction of targeted steroids that are isobars for nonsteroidal compounds at nominal mass, but offers no distinction of steroid isobars unless unique secondary fragment ions can be identified. Conditions in the ionization source must be optimized for each component. The following section discusses these and other limitations that must be considered when setting up LC–MS/MS-based steroid assays.
Challenges
„ Pre-analytical challenges
When using LC–MS/MS, as for IA and GC– MS, the pre-analytical aspects of sample collec- tion must be remembered. This is particularly relevant in the analysis of steroid hormones, where concentrations are influenced by a number of factors, including circadian rhythms. When offering a service to analyze and interpret con- centrations of these hormones, it is important that appropriate instructions on sample collec- tion are offered, and suitable request forms are available for service users.
Expected concentration ranges
& protein binding
Serum concentrations of some endogenous steroids are very low [27]. A reliable method for the analysis of estrogens in serum from males and postmeno- pausal females, for example, should have a LOQ of the order of 1 pg/ml (~3 pmol/l, 1 ppt). This is complicated by the fact that other endogenous steroids, which may interfere with such analyses, may be present at significantly higher concentra- tions. Concentrations of endogenous steroids rela- tive to these interfering compounds vary in dif- ferent matrices, so that matrix-specific validation is required before applying one analytical tech- nique to an alternative matrix [41]. A further con- sideration is that concentrations of some steroids, which are low in ‘normal’ samples, may be grossly elevated (many orders of magnitude greater than ‘normal’) in some disorders, as described above for CAH. Reliable assays therefore require a large linear dynamic range. Endogenous steroids are transported around the body extensively (>95%) bound to nonspecific plasma proteins (e.g., albu- min) and to specific steroid-binding proteins
2554
Bioanalysis (2011) 3(22)
future science group

Challenges & benefits of endogenous steroid analysis by LC–MS/MS | Review
(e.g., sex hormone-binding globulin, cortisol/ corticosteroid binding globulin). Concentrations of biologically active (‘free’) hormone are therefore even lower still, and are affected by concentra- tions of these binding proteins, which may vary considerably between individuals. Analyses have been developed to measure free hormone concen- trations (typically using equilibrium dialysis [42] or ultrafiltration [43]), although the majority of analytical methods measure total serum concen- trations after disruption of any protein binding. Separate analysis of the binding proteins them- selves then allows for an estimation of the free concentration by calculation [44,45]. However, a number of equations/algorithms exist and these can yield variable results [46,47].
Sampling
A number of factors must be considered prior to sampling, including age, sex, body mass, fasting/ nonfasting, pubertal stage, menstrual cycling, pregnancy and diurnal/circadian rhythm [41,48], as these will heavily influence the reference inter- vals and interpretation of any findings. Samples for steroid analysis are relatively stable once col- lected [41], although tubes used for collection and during extraction should be evaluated to ensure steroids do not adsorb to their surfaces [49,50].
Hemolysis and lipemia are known to affect some IAs [41,51], but should also be evaluated for additional matrix effects in LC–MS/MS methods, particularly those where less extensive sample preparation is used. There is little dif- ference when using plasma or serum for steroid analysis, although gel separation tubes may cause problems for some hydrophobic/lipophilic steroids such as progesterone. Clot activators in blood tubes were shown to interfere with one m/z transition in selected reaction monitoring (SRM) mode during testosterone analysis [52].
„ Analytical challenges
Sample preparation
Sample preparation has a significant bearing on the results of LC–MS/MS analyses [39]. This is particularly true for quantitative analyses, where accuracy and precision of data may be affected by:
„ Any differences in analyte recovery;
„ The amount of interfering compound(s) remaining in extracted samples (matrix effects);
„ The formation of adducts. future science group
For LC–MS/MS, matrix effects (either ion suppression and/or ion enhancement) caused by coextracted interfering compounds in the MS source are well-documented [53–57], and should always be evaluated when implementing new methods [58–61]. The evaluation of matrix effects is complicated when a suitable blank matrix is not available. In an example from vitamin D analysis, Bunch et al. used postcolumn infu- sion of isotopically labeled analyte as a surro- gate to quantify matrix effects in the analysis of 25-hydroxyvitamin D metabolites [62]. Internal standardization as a means to overcome some of the problems associated with matrix effects is important for qualitative work, although this in itself is not infallible [63–67]. IS themselves should be evaluated as causes of ion suppres- sion/enhancement for the analyte and vice versa. The purity of labeled compounds must also be considered. Whilst common isotopically labeled analytes are now available at reasonable prices (e.g., in the UK, 1 mg trideuterated testoster- one costs ~UK£114 [~US$185 at time of press]; LGC Standards, Teddington, UK), some stable isotope-labeled IS are either very expensive (as they require custom synthesis) or, for some minor metabolites, not easily available.
There exists no universal sample preparation technique for the analysis of steroids [68]. Protein precipitation, liquid–liquid extraction and SPE procedures have all been successfully used to extract steroids from a range of biological flu- ids [68]. It is clear that any sample preparation technique will be dependent largely upon the matrix in which analytes are to be measured. Urine will contain a different range of steroids to plasma or serum. Urine steroid profiling by GC–MS is primarily concerned with the analysis of polar steroid metabolites and their sulfate/ glucuronide conjugates, such as pregnanetriol, androsterone and tetrahydrocortisol, and their excretion patterns and ratios. Sample preparation for such analyses therefore requires a hydrolysis step (enzymatic or acidic) to cleave the conju- gated moieties, prior to chemical derivatization. The hydrolysis is itself a process that requires careful validation, since enzymatic hydrolysis varies in efficiency depending upon the chosen enzyme (e.g., b-d-glucuronidase), with regards to the cleavage of glucuronide and sulfate con- jugates. Likewise, chemical hydrolysis can vary depending on the chosen reaction conditions. When using LC–MS/MS for analysis of urine steroids, steroid conjugates can be analyzed directly, negating any hydrolysis step [69–71]. Th