ReseaRch aRticle

Metabonomics has been defined as “quantitative measurement of multivariate metabolic responses of multicellular systems to pathophysiological stimuli or genetic modification” and has been primarily utilized to provide information for the widespread understanding of global metabolic regulation and biological system failure [1]. Initial work in the area of metabonomics was focused upon understanding the mechanisms of drug toxicity in a mammalian system [2–5]. The main aim of these studies was to collate information on biomarkers of nephrotoxicity, and hepatotoxicity, This information was subsequently used to build statistical models to monitor early safety assessment studies as a means to eliminate potentially undesirable compounds earlier in the drug-development process [6–8]. As part of these early studies, it was clear that the use of spectroscopic and spectrometric profiling of biological fluids could yield data that could result in the ability to discriminate between gender, age and diurnal variations [9,10].

As a result of these promising early studies more elaborate studies were undertaken, such as the consortium on metabolism and toxicology, to understand and build statistical models of toxicity [11,12]. These promising initial studies resulted in a dramatic increase in the interest and use of the metabonomic approach to investigate

disease models and dietary effects [13–17]. More recently, the use of data from high-resolution NMR and MS has aided decision making in both surgical and clinical settings [18,19].

Much of the initial metabonomics research employed either NMR or GC–MS as the primary platforms to acquire analytical data [20–22]. Although NMR and LC–NMR are used in drug metabolism studies, LC–MS is by far the most dominant analytical platform for the acquisition of qualitative and quantitative data in drug metabolism and PK studies because of the inherent sensitivity, speed and throughput of the technique. The application of LC–MS to metabonomics studies has provided complementary data to that obtained by NMR, often producing assays of greater sensitivity [23–25].

The vast majority of the LC–MS metabonomics methodologies employed the use of formic acid-based organo-aqueous gradient approaches, primarily due to the fact that it is the most common mobile phase combination used in LC–MS when the ionization mode utilized is ESI positive [26,27]. This could have been further compounded by the dearth of chromatographic stationary phases that were chemically and mechanically unstable in basic pH environments prior to the late 1990s. In a previous paper, we communicated the sensitivity benefits of using

Investigation of basic mobile phases with positive ESI LC–MS for metabonomics studies

Background: Accurate mass based LC–MS combined with statistical ana lysis is established as a core analytical technology for metabonomic studies. This is primarily due to the specificity, sensitivity and structural elucidation capabilities of the technology. The vast majority of these studies are performed using acidic-based mobile phases in combination with positive ESI mode LC–MS. Recent studies have investigated the use of highly basic pH mobile phases (>10 pH units) in bioanalytical studies that utilize positive ESI mode LC–MS. This non-traditional combination has been shown to improve analyte retention, chromatographic peak shape, and S/N for a variety of probe pharmaceutical compounds in biofluid samples. Results: The incorporation of basic pH mobile phases resulted in increased retention for analytes that where comparatively weakly retained by a traditional acidic-modified mobile phase. Increased resolution of isomers, which otherwise co-eluted under acidic conditions, was observed. Moreover, the implementation of basic pH mobile phases further allowed for the detection of complementary marker ions. Conclusion: Basic pH mobile phases utilized with positive ESI mode LC–MS have the potential for producing increased information from metabonomic studies and could lead to the detection of analytes that may prove to be valid biomarkers.

Paul D Rainville*1,2, Norman W Smith1, David Cowan1, Jeremy K Nicholson3, JP Shockcor2, St John Skilton2 & Robert S Plumb1,2,3

1King’s College London, School of Pharmacy, Stamford St, London, SE1 9NH, UK 2Waters Corporation, Milford, MA 01757, USA 3Imperial College, Department of Surgery & Cancer, London, SW7 2AZ, UK *Author for correspondence: E-mail: paul_rainville@waters.com

2833ISSN 1757-618010.4155/BIO.12.258 © 2012 Future Science Ltd Bioanalysis (2012) 4(23), 2833–2842

For reprint orders, please contact reprints@future-science.com

basic pH mobile phases for the ana lysis of the benzodiazepine drug alprazolam analyzed under positive ESI mode [28], and, in a subsequent larger bioanalytical study for a significant number of compounds varying in structure, molecular weight, pKa and hydrophobicity [29].

In this study, we present the first known investigation into the utilization of basic pH mobile phase with reversed phase chromato-graphy coupled with positive ESI LC–MS for metabonomic ana lysis. Urine samples were collected from rats that were treated with the model hepatotoxin, hydrazine, and were utilized to determine any benefit of utilizing this nontraditional mobile phase combination for positive ESI LC–MS. Chromatographic resolution, analyte response and retention were all compared with that obtained when using the same column and MS conditions with a traditional acidic mobile phase. An additional goal of this study was to compare the differences in the detection of analytes between the traditional acidic mobile phase LC–MS ana lysis with the nontraditional basic mobile phase LC–MS ana lysis.

Experimental�� Chemicals

Optima grade acetonitrile (HPLC grade) was purchased from Fisher Scientific (Hampton, NH, USA), formic acid (spectroscopic grade) and 0.1 M ammonium hydroxide solution were purchased from Sigma-Aldrich (MO, USA). Distilled water was purified ‘in-house’ using a MilliQ system Millipore (MA, USA).

�� Animal studiesRat urine samples were obtained from a study of hydrazine toxicity conducted as part of a major toxicology project [30]. All animal experiments were conducted according to appropriate national guidelines. In this study, male Sprague-Dawley rats were randomly allocated to groups: control, low and high dose (n = 10 per group). Here, we only consider the control and high-dose groups. The high-dose group received hydrazine hydrochloride in 0.9% saline administered orally at 90 mg/kg. Dose vehicles alone were administered to matched control animals. The time of dosing was defined as 0 h. Rats were housed in individual metabolic cages under controlled temperature, humidity and light cycles. Urine samples were collected for time intervals from -16 to 168 h. Data for the high-dose group used for this paper were obtained

from 58 samples, of which 20 corresponded to predose time points (showing no toxic response) and 38 samples to time points known to demonstrate a high-toxic response. Data for the control group were obtained from n = 54 control samples.

�� LCThe separations were performed on a 2.1 × 100 mm ACQUITY UPLC 1.7 µm BEH C18 column using an ACQUITY Ultra-Performance LC Chromatography™ System (Waters® Corporation, MA, USA). The column was maintained at 40°C and eluted with a linear acetonitrile/aqueous gradient over 10 min at 500 µl/min, starting at 5% acetonitrile and rising to 95% over the course of the gradient. The aqueous mobile phase was either 0.1% formic acid or 0.1% ammonium hydroxide. The column eluent was directed to the mass spectrometer for ana lysis.

�� MSMS was performed on a Waters Micromass® Xevo™ G2 QTOF mass spectrometer (Waters Micromass, Manchester, UK) operated in positive ESI mode with ‘W-Optics’, the resolving quadrupole set to a wide pass mode, the collision cell was set to alternate between a collision energy of 5 and 25 eV every 60 ms with a 20 ms interscan delay. Leucine enkephalin was employed as the Lockspray™ solution at a concentration of 200 ng/ml.

�� Statistical analysisData were analyzed using the MarkerLynx XS™ applications manager (Waters, Elstree, UK); this application manager integrates peaks in the LC–MS data by using ApexTrack™ peak detection. The LC–MS data were peak-detected and noise-reduced in both the LC and MS domains such that only true analytical peaks were further processed by the software (e.g., noise spikes are rejected). A list of the intensities of the peaks detected was then generated for the first sample, using the retention time (RT) and m/z data pairs as the identifier for each peak. An arbitrary number is then assigned to each of these RT–m/z pairs in order of elution (one, two, three, four and so on). This process was repeated for each run; once this was completed, the data from each LC–MS ana lysis in the batch were then sorted such that the correct peak intensity data for each RT–m/z pair was aligned in the final data

Key Terms

Metabonomics: Quantitative measurement of multivariate metabolic responses of multicellular systems to pathophysiological stimuli or genetic modification.

Biomarkers: Endogenous small molecule or protein that indicates a change in expression or state of a protein that correlates with the risk or progression of a disease, or with the susceptibility of the disease to a given treatment.

Safety assessment studies: The use of cell or small animal-based studies to determine the toxicological profile of new candidate drugs prior to administration to human volunteers.

Principal component ana lysis: Mathematical procedure utilizing an orthogonal transformation to convert a set of observations of possibly correlated variables into a set of values of linearly uncorrelated variables called principal components. The highest ranking principal components are those that contribute most significantly to the variation observed in the dataset. It is mostly used as a tool in exploratory data ana lysis and for making predictive models.

ReseaRch aRticle | Rainville, Smith, Cowan et al.

Bioanalysis (2012) 4(23)2834 future science group

table. The ion intensities for each peak detected are then normalized, within each sample, to the sum of the peak intensities in that sample with these normalized peak intensities then multiplied by 10,000. The resulting 3D data, peak number (RT–m/z pair), sample name and ion intensity were analyzed by principal components analysis (PCA) using mean centering and Pareto scaling.

Results & discussion�� Basic pH mobile phase with positive

ESI MSThe benefits of basic pH mobile phases in reversed-phase chromatography are well known for compounds containing basic groups. Operation at a pH above the pKa of these compounds allows them to be chromatographed as neutral species, resulting in greater, more reproducible retention, and more symmetrical peak shapes. Furthermore, the use of basic pH helps to overcome the effect of peak tailing.

Although the use of pH stable stationary phases, such as zirconium and porous graphitic carbon [31], had been described earlier in the 1980s and 1990s, their use had not become widespread due to poor chromatographic performance, lack of reproducibility and unpredictable chromatographic elution characteristics compared with silica-based materials [32].

The routine application of basic pH mobile phases was made possible by the development of organo-silica-based stationary phases in the late 1990s [32]. These new phases relied on the incorporation of a methyl group into the core silica structure, reducing phase dissolution and allowing operation in the range of 9–10 pH units. More recently, the development of a bridged-ethyl silica hybrid stationary phase has resulted in greater phase stability allowing continual operation at pH 10–11.

Delatour and Leclercq, showed the potential for incorporating basic pH mobile phases for use in pharmaceutical ana lysis by positive ESI

100

0.39

0.81

0.65

0.35

1: TOF-MS ES+BPI

6.50e6

1: TOF-MS ES+BPI

6.50e6

0.81

0.65

0.35

1: TOF-MS ES+BPI

6.50e6

0.70 3.30

3.44

3.44

1.09

1.61

4.01

4.57

4.084.83

5.18

5.82

5.91

5.91

6.17

6.87

10.17

10.03

6.90

7.01

7.25

7.65

8.26

8.05

8.54 8.64

8.68 8.96

9.489.58

9.589.72

Time (min)

Time (min)

1.96

1.40 1.61

1.47 2.31

2.69

2.97

02.00 4.00 6.00 8.00 10.00

Sig

nal

inte

nsi

ty (

%)

100

0

2.00 4.00 6.00 8.00 10.00

Sig

nal

inte

nsi

ty (

%)

0.39 6.50e6

0.70 3.30

3.44

1.09

1.61

4.01

4.57

5.185.91

6.17 10.17

6.90

7.018.26

8.68 8.96

9.489.581.96

1.47 2.31

A

B

50

50

0

0

Figure 1. Positive ESI LC–MS chromatogram of control rat urine. Using an (A) acidic and (B) basic mobile phase.

Basic mobile phases with positive ESI LC–MS for metabonomics studies | ReseaRch aRticle

www.future-science.com 2835future science group

mode LC–MS in 2005 [33]. Similarly, in 2008, Peng and Farkas reported the first in-depth study utilizing basic pH mobile phases for the ana-lysis of over 40 basic drug substances analyzed with positive ESI mode LC–MS. In this work, the authors reported an actual increase in MS response for the basic probe pharmaceutical compounds [34].

�� Application to metabonomic studyPlasma and urine contain a signif icant proportion of basic compounds, so the ability to improve retention and separation for these compounds in reversed-phase LC–MS would be of significant benefit [35]. Greater separation of analytes from the ‘salt’ components eluting in the void of the column is also likely to reduce ion suppression and improved assay reproducibility. This is critical in metabonomics studies as any variation in analyte response could be interpreted as the result of a toxicological or disease state response. The use of other forms of chromatography, such as hydrophilic interaction LC, has yielded promising results in this area [35]. However, this form of chromatography is

less effective than reversed phase at dealing with the broad range of analyte polarities present in biological fluids.

The data displayed in Figure 1 illustrate two chromatograms obtained for the ana lysis of control rat urine, using either an acid or basic mobile phase. In these two examples the gradient duration, organic solvent and gradient steepness have been kept constant. The top chromatogram represents the acidic mobile-phase separation and the bottom chromatogram the separation that is obtained from a basic mobile phase separation. A qualitative review of the data shows that there is a significant difference in the elution profile across the whole of the chromatogram.

The use of the basic mobile phase not only increased analyte retention but also improved analyte sensitivity and peak resolution. The data displayed in Figure  2 illustrate the increased analyte response obtained from the two ions detected within the mass window 150.077 +/- 50 mDa. The top chromatogram is the separation with the acidic mobile phase and the bottom chromatogram is that obtained from the basic mobile phase. The two chromatograms axes are linked for intensity. The peak eluting with a RT of 3 min in the basic separation demonstrates a response almost 50% greater than the same analyte in the acidic separation, where it elutes with a RT of 1.82 min. The same phenomenon is also observed for the smaller peak in the two chromatograms eluting with a RT of 2.08 and 0.81 min in the basic and acidic mobile-phase systems, respectively. Due to the fact that many analytes in biological fluids are basic in nature, upon exposure to an acidic mobile phase these analytes will be present in a charged state. This charged state will, in effect, make the analyte more polar and, therefore, less likely to retain on a reversed-phase column. However, when basic analytes are exposed to a basic mobile phase, they exist in a neutral state and are more likely to be retained by reversed phase.

The use of a basic mobile phase also, in some cases, provided enhanced resolution for isobaric analytes. The separation of such analytes allows the statistical ana lysis to correctly process them as two discrete ions. This fact is illustrated in Figure 3. In this example, the single peak eluting with a RT of 10.96 min in the acidic system (top chromatogram) is resolved into two discrete peaks eluting at 9.72 and 10.03 min in the basic system. The accurate mass of the two peaks detected, m/z = 358.269, was

Time (min)

Time (min)

1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00

Sig

nal

inte

nsi

ty (

%)

0

100

1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00

Sig

nal

inte

nsi

ty (

%)

0

1001: TOF-MS ES+150.077 0.05Da

2.63e6

1.82

0.81

1: TOF-MS ES+150.077 0.05Da

2.63e6

3.00

2.08

A

B

50

50

0

0

Figure 2. Extracted ion chromatogram of rat urine in positive ion mode, m/z = 150.077. Using an (A) acidic and (B) basic mobile phase.

ReseaRch aRticle | Rainville, Smith, Cowan et al.

Bioanalysis (2012) 4(23)2836 future science group

subjected to a search on multiple databases, but did not return any potential compounds. The data shown in Figure 3 clearly illustrate that the spectra obtained from the two discrete peaks obtained from the basic separation were significantly cleaner and identified the two peaks as isomers, whereas the spectra obtained from the acidic separation was equally as informative and is due to the fact that the statistical analysis does show separation of

the control and dosed groups, as did the traditional acidic LC–MS separation. The extra resolution is a result of extra selectivity imparted onto the separations system as a result of employing a basic pH mobile phase. This improved resolution can be beneficial not only in metabonomic studies, but also in metabolite identification studies, especially in the light of the recent MIST guidelines that emphasize the need to study toxic species [36].

Figure 3. Extracted ion chromatogram of rat urine in positive ion mode, m/z = 358.269. Using an (A) acidic and (B) basic mobile phase.

Time (min)

Time (min)

2.00 4.00 6.00 8.00 10.00 12.00

Sig

nal

inte

nsi

ty (

%)

0

100

2.00 4.00 6.00 8.00 10.00 12.00

Sig

nal

inte

nsi

ty (

%)

0

100

1: TOF-MS ES+358.269 0.05Da

1.12e7

1: TOF-MS ES+358.269 0.05Da

6.59e6

10.96

10.03

9.72

8.69

m/z100 120 140 160 180 200 220 240 260 280 300 320 340

%

0

100 137.1378

133.0941

121.1064

340.3968312.3660

155.1484166.0916 283.2675

248.1723184.0785 228.2735199.1541 282.2840284.2706

313.3692 358.2611

m/z100 120 140 160 180 200 220 240 260 280 300 320 340

%

0

100 137.1379

133.0997

155.1485

358.2617

201.0517156.1517

162.1222 299.1893279.0978

202.0550 257.1744 333.2055

m/z100 120 140 160 180 200 220 240 260 280 300 320 340

%

0

100 137.1380

133.1006

155.1484

201.0519

156.1522

358.2621279.0980204.0914

274.0968 339.2717321.2616

50

50

0

0

A

B

Basic mobile phases with positive ESI LC–MS for metabonomics studies | ReseaRch aRticle

www.future-science.com 2837future science group

�� Model toxin study samplesThe effect of utilizing basic mobile-phase combin-ations in an LC–MS-based metabo nomics study was investigated using urine samples obtained from a toxicology study with a model toxin, hydrazine, dosed to rats. The ana lysis of these samples was previously reported by Crockford et al. [30]. In this paper, the authors describe the use of high-resolution chromatography, accurate mass QTOF-MS system combined with advanced correlation statistical ana lysis to obtain structural, as well as higher level biological information relating to changes in metabolic pathway activity and connectivity. For the present study, we have reanalyzed a subset of these samples using both the acidic and basic aqueous-organic reversed-phase chromatography systems. Again, as with the previous control of urine, the samples were analyzed using high-resolution sub 2-µm porous particle LC–MS with QTOF detection in positive ion mode. The data displayed in Figure 4 illustrates the separations obtained from the LC–MS ana lysis of the resulting urine samples from the hydrazine study, in this case, subject 20, control animal

predose. The upper chromatogram illustrates the data obtained from the acidic separation and the lower chromatogram is that obtained from the basic LC–MS ana lysis. The data clearly demonstrate that the large peak eluting at 0.46 min in the acid chromatogram has either been significantly shifted in the basic separation or resolved into several peaks. The samples were analyzed using the PCA method and the orthoganal projections to latent structures class model described by Wiklund et al. [37]. The data displayed in Figure 5 show that the PCA plots PC1 versus PC2 for the ana lysis of the 0 and 48 h samples using either the acid mobile phase (Figure 5A) or basic mobile phase (Figure 5A). The statistical ana lysis of the two datasets showed that both systems clearly separated the 0 h samples from the 48 h samples with a very similar pattern. However, it is clear from the data ana lysis that the ions demonstrating the greatest contribution to the variance in the data differed between the two chromatographic systems, irrespective of the actual retention time. The data displayed in TAble 1 list the top ten ions contributing to variation in the data for

100 0.46

0.63

1.56

1.782.58 2.97

3.59

4.18

4.67

4.765.28

5.65 6.78 7.07

7.581: TOF-MS ES+

BPI3.78e6

0.46

0.63

1.56

1.782.58 2.97

3.59

4.18

4.67

4.765.28

5.65 6.78 7.07

7.581: TOF-MS ES+

BPI3.78e6

1: TOF-MS ES+BPI

1.04e6

00 2.00 3.00 4.00 5.00 6.00 7.001.00

0.65

0.81

1.26

1.33 1.63

1.70

2.22

2.91

3.00

3.45

3.94

3.70 4.40

4.064.67

5.44 5.666.08

6.19

Time (min)

Time (min)

Sig

nal

inte

nsi

ty (

%)

100

00 2.00 3.00 4.00 5.00 6.00 7.001.00

Sig

nal

inte

nsi

ty (

%)

50

50

A

B

1: TOF-MS ES+BPI

1.04e6

0.65

0.81

1.26

1.33 1.63

1.70

2.22

2.91

3.00

3.45

3.94

3.70 4.40

4.064.67

5.44 5.666.08

6.19

Figure 4. Positive ESI LC–MS rat urine following the oral administration of hydrazine at 60 mg/kg. Using an (A) acidic and (B) basic mobile phase.

ReseaRch aRticle | Rainville, Smith, Cowan et al.

Bioanalysis (2012) 4(23)2838 future science group

-80

-110 -100 -90 -80 -70 -60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60 70 80 90 100

-70

-60

-50

-40

-30

-20

-10

0

10

20

30

40

50

60

70

80

PC(2)

PC(1)

R10R10R10R10R10

R11R11R11R13R13R13R13R13

R22R22R22R22R22R25R25R25

R8R8R8R8R8

R20R20R20R20R20

R10R10R10R10R10

R11R11R11

R13R13R13R13R13

R22R22R22R22R22

R25R25R25

R8R8R8R8R8

R20R20R20

EZinfo 2 - acidic_run edited(M3: PCA-X) - 2011-05-20 09:57:53 (UTC-5)

Predose

48-h postdose

Replicate response

-80

-110 -100 -90 -80 -70 -60 -50 -40 -30 -20 -10 0 10 20 30 40 50 60 70 80 90 100 110

-70

-60

-50

-40

-30

-20

-10

10

20

30

40

50

60

70

80

PC(2)

PC(1)

EZinfo 2 - basic_run3 (M3: PCA-X) - 2011-05-20 09:53:50 (UTC-5)

Predose

48-h postdose

Replicate response

0R10R10R10R10R10

R11R11R11R13R13R13R13R13

R22R22R22R22R22R25R25R25

R8R8R8R8R8

R20R20R20R20R20

R10R10R10R10R10

R11R11R11

R13R13R13R13R13

R22R22R22R22R22

R25R25R25

R8R8R8R8R8

R20R20R20

A

B0 h48 h

0 h48 h

Figure 5. Principal components analysis of positive ion LC–MS analysis of rat urine following the oral administration of hydrazine at 60 mg/kg. Using an (A) acidic and (B) basic mobile phase.

Basic mobile phases with positive ESI LC–MS for metabonomics studies | ReseaRch aRticle

www.future-science.com 2839future science group

both the 0 and 48 h samples. It is likely that the different datasets would provide different complementary data. This point can be further illustrated graphically in Figure 6. In this data, we can see the acid and basic separations, as well as a trend plot for the marker ion 962.3001: from the data in Figure 6 it can be seen that this ion

was detected in the basic LC–MS separation while remaining undetected with the traditional acidic LC–MS separation. The data displayed in TAble 2 compares the RT and signal intensity obtained for a select set of extracted signals (ions) previously identified in the work of Crockford et al. on hydrazine toxicity, as well as selected other commonly reported markers of toxicity compounds, in both the acidic and basic mobile phases. As can be seen from the data, the RT was increased for some of the marker ions under basic conditions and the signal intensity increased for 50% of ions under basic conditions (all measurements were made under the same positive ESI conditions). Not unsurprisingly, the acidic compounds, such as huppuric acid and pantothenic acid, were retained to a greater extent under acidic conditions; however, even when the retention (K

g) did not increase under

basic conditions, the signal response often did. For example, biomarker candidate m/z = 170.06 had a reduced RT under basic conditions, but the signal intensity increased by a factor of five.

Table 1. Comparison of the LC–MS marker ions detected with acidic and basic mobile phase in the analysis of hydrazine-dosed rats.

Mobile phase pH 2.5; 0 h

Mobile phase pH 2.5; 48 h

Mobile phase pH 10; 0 h

Mobile phase pH 10; 48 h

287.863 431.688 284.948 309.044

162.66 295.731 215.744 267.815

144.131 160.858 182.489 234.191

144.044 141.272 167.834 211.589

139.497 129.423 163.578 156.26

117.805 123.879 160.262 143.76

117.787 113.722 160.199 140.175

114.18 109.387 111.56 137.354

104.861 100.739 110.745 131.268

287.863 431.688 106.291 309.044

2.00 4.00 6.00 8.00 10.00 2.00 4.00 6.00 8.00 10.00

Sig

nal

inte

nsi

ty (%)

00

1002.97

0.391.47

2.31

4.01

3.30

5.18

4.57

6.905.91

9.49

8.958.697.01

8.08

9.62

9.79

Time (min)Time (min)

Sig

nal

inte

nsi

ty (%)

0

100 0.81 10.03

1.61

5.82

2.04 3.442.69

4.104.83

8.54

8.057.63

7.25

9.72

9.58

0

5050

A B

Samples

Sig

nal

inte

nsi

ty (%)

01 2 3 4 5 6 7 8 9 10 11 12 13 14

100

Marker (7.44,962.3001) 9.116e-002

50

C

Figure 6. Positive ion LC–MS analysis of rat urine. Using an (A) acidic and (B) basic mobile phase. (C) Trend plot for the marker ion m/z 962.3001 at retention time 7.44 min.

ReseaRch aRticle | Rainville, Smith, Cowan et al.

Bioanalysis (2012) 4(23)2840 future science group

ConclusionThe application of basic mobile phases coupled with positive ESI LC–MS to the ana lysis of biological samples in metabonomic studies shows great promise. It provides complementary data to that obtained with traditional acidic-based mobile phases and results in an increased number of ions detected overall. For many analytes, the use of a basic mobile phase increases analyte retention and results in an enhanced response in the mass spectrometer. During this study, it was also noted that by changing the pH to a basic system it was possible to resolve compounds that had previously co-eluted on an acidic reversed-phase-based system. The ana lysis of a real sample set of control rats and rats dosed with hydrazine revealed that both the acidic and basic systems were capable of differentiating between the 0 and 48 h time points.

Future perspectiveThe application of LC–MS to metabonomics studies has provided analytical data resulting in an understanding of disease state progression and the efficacy of medicines. The incorporation of non-traditional mobile phase combinations in LC–MS ana lysis may lead to the detection of potential biomarker compounds that will lead to a greater understanding of the health of the global population, as well as identify the most appropriate treatment. The use of LC–MS for this application area will continue to grow,

mainly due to the richness of information present in the datasets and compatibility with the samples.

Financial & competing interests disclosureThe authors have no relevant affiliations or financial involvement with any organization or entity with a finan-cial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert t estimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.

Table 2. Comparison of the retention of reported biomarkers of toxicity under acidic and basic LC–MS conditions.

Compound m/z Relative signal intensity

Kg

Acid Base Acid Base

Biomarker candidate 127.04 1 0.5 1.7 4.2

Biomarker candidate 149.04 1 ND 1.7 ND

Biomarker candidate 156.03 1 0.7 0.1 0.3

Biomarker candidate 156.06 1 0.4 0.9 0.3

Biomarker candidate 170.06 1 5.0 8.5 7.7

Pantothenic acid 220.12 1 1.1 6.0 4.8

Hippuric acid 180.06 1 1.5 7.5 3.6

Hydrazine toxicity marker 402.1 1 2.7 7.0 6.3

Hydrazine toxicity marker 265.05 1 2.9 0.3 0.3

Hydrazine toxicity marker 218.11 1 0.2 10.6 4.7

Kg: Gradient retention factor (T-T

0/T

0); ND: Not detected.

Executive summary

�� Basic pH LC–MS provides complementary data to that obtained from acidic pH LC–MS for metabonomic studies.

�� The use of a basic pH does not detrimentally affect the sensitivity of the positive ESI mode LC–MS separation.

�� Basic pH LC–MS can allow for increased retention of analytes, therefore increasing the ability for the detection and identification.

�� The basic pH separation mode allowed for the separation of isomers previously unresolved when using traditional acidic conditions.

ReferencesPapers of special note have been highlighted as:� of interest

1 Nicholson JK, Lindon JC, Holmes E. ‘Metabonomics’: understanding the metabolic responses of living systems to pathophysiological stimuli via multivariate statistical analysis of biological NMR spectroscopic data. Xenobiotica 29(11), 1181–1189 (1999).

�� Provides a good definition of the term metabonomics.

2 Robertson DG, Bulera SJ. High-throughput toxicology: practical considerations. Curr. Opin. Drug Discov. Devel. 3(1), 42–47 (2000).

3 Holmes E, Nicholls AW, Lindon JC et al. Chemometric models for toxicity classification based on NMR spectra of biofluids. Chem. Res. Toxicol. 13(6), 471–478 (2000).

�� Comprehensive description of chemometrics for toxicology.

4 Robertson DG, Reily MD, Sigler RE, Wells DF, Paterson DA, Braden TK.

Metabonomics: evaluation of nuclear magnetic resonance (NMR) and pattern recognition technology for rapid in vivo screening of liver and kidney toxicants. Toxicol. Sci. 57(2), 326–337 (2000).

5 Holmes E, Nicholson JK, Tranter G. Metabonomic characterization of genetic variations in toxicological and metabolic responses using probabilistic neural networks. Chem. Res. Toxicol. 14(2), 182–191 (2001).

6 Hellmold H, Nilsson CB, Schuppe-Koistinen I, Kenne K, Wärngård L. Identification of end points relevant to detection of potentially

Basic mobile phases with positive ESI LC–MS for metabonomics studies | ReseaRch aRticle

www.future-science.com 2841future science group

adverse drug reactions. Toxicol. Lett. 127(1–3), 239–243 (2002).

7 Robertson DG, Reily MD, Albassam M, Dethloff LA. Metabonomic assessment of vasculitis in rats. Cardiovasc. Toxicol. 1(1), 7–19 (2001).

8 Holmes E, Antti H. Chemometric contributions to the evolution of metabonomics: mathematical solutions to characterising and interpreting complex biological NMR spectra. Analyst 127(12), 1549–1557 (2002).

9 Major HJ, Williams R, Wilson AJ, Wilson ID. A metabonomic analysis of plasma from Zucker rat strains using gas chromatography–mass spectrometry and pattern recognition. Rapid Commun. Mass Spectrom. 20(22), 3295–3302 (2006).

10 Plumb RS, Johnson KA, Rainville P et al. The detection of phenotypic differences in the metabolic plasma profile of three strains of Zucker rats at 20 weeks of age using ultra-performance liquid chromatography/orthogonal acceleration time-of-flight mass spectrometry. Rapid Commun. Mass Spectrom. 20(19), 2800–2806 (2006).

11 Lindon JC, Keun HC, Ebbels TM, Pearce JM, Holmes E, Nicholson JK. The consortium for metabonomic toxicology (COMET): aims, activities and achievements. Pharmacogenomics 6(7), 691–699 (2005).

12 Lindon JC, Nicholson JK, Holmes E et al. Summary recommendations for standardization and reporting of metabolic analyses. Nat. Biotechnol. 23(7), 833–838 (2005).

�� Provides a good definition of agreed terms and best practice for metabolic analysis.

13 Jennen D, Ruiz-Aracama A, Magkoufopoulou C et al. Integrating transcriptomics and metabonomics to unravel modes-of-action of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) in HepG2 cells. BMC Syst. Biol. 5(1), 139 (2011).

14 Verwaest KA, Vu TN, Laukens K et al. (1)H NMR based metabolomics of CSF and blood serum: a metabolic profile for a transgenic rat model of Huntington disease. Biochim. Biophys. Acta 1812(11), 1371–1379 (2011).

15 Nicholson G, Rantalainen M, Maher AD et al. Human metabolic profiles are stably controlled by genetic and environmental variation. Mol. Syst. Biol. 7, 525 (2011).

16 Gika HG, Theodoridis GA, Earll M, Snyder RW, Sumner SJ, Wilson ID. Does the mass spectrometer define the marker? A comparison of global metabolite profiling data generated simultaneously via UPLC–MS

on two different mass spectrometers. Anal. Chem. 82(19), 8226–8234 (2010).

17 Wu B, Peng WJ, Wang PJ et al. In vivo 1H magnetic resonance spectroscopy in evaluation of hepatocellular carcinoma and its early response to transcatheter arterial chemoembolization. Chin. Med. Sci. J. 21(4), 258–264 (2006).

18 Kinross JM, Holmes E, Darzi AW, Nicholson JK. Metabolic phenotyping for monitoring surgical patients. Lancet 377(9780), 1817–1819 (2011).

�� Overview on how metabonomics can be used in a clinical setting.

19 Patel VM, Ashrafian H, Ahmed K et al. How has healthcare research performance been assessed? A systematic review. J. R. Soc. Med. 104(6), 251–261 (2011).

20 Glinski M, Weckwerth W. The role of mass spectrometry in plant systems biology. Mass Spectrom. Rev. 25(2), 173–214 (2006).

21 Wiklund S, Johansson E, Sjöström L et al. Visualization of GC–TOF-MS-based metabolomics data for identification of biochemically interesting compounds using OPLS class models. Anal. Chem. 80(1), 115–122 (2008).

22 Zhang L, Liu X, You L et al. Benzo(a)pyrene-induced metabolic responses in Manila clam Ruditapes philippinarum by proton nuclear magnetic resonance ((1)H NMR) based metabolomics. Environ. Toxicol. Pharmacol. 32(2), 218–225 (2011).

23 Plumb RS, Granger JH, Stumpf CL et al. A rapid screening approach to metabonomics using UPLC and oa-TOF mass spectrometry: application to age, gender and diurnal variation in normal/Zucker obese rats and black, white and nude mice. Analyst 130(6), 844–849 (2005).

24 Plumb RS, Stumpf CL, Gorenstein MV et al. Metabonomics: the use of electrospray mass spectrometry coupled to reversed-phase liquid chromatography shows potential for the screening of rat urine in drug development. Rapid Commun. Mass Spectrom. 16(20), 1991–1996 (2002).

25 Geier FM, Want EJ, Leroi AM, Bundy JG. Cross-platform comparison of Caenorhabditis elegans tissue extraction strategies for comprehensive metabolome coverage. Anal. Chem. 83(10), 3730–3736 (2011).

26 Castro-Perez JM, Kamphorst J, DeGroot J et al. Comprehensive LC–MSE lipidomic analysis using a shotgun approach and its application to biomarker detection and identification in osteoarthritis patients. J. Proteome Res. 9(5), 2377–2389 (2010).

27 Shockcor JP, Holmes E. Metabonomic applications in toxicity screening and disease diagnosis. Curr. Top. Med. Chem. 2(1), 35–51 (2002).

�� Overview of the application of metabonomics to a toxicology screening.

28 Mather J, Rainville PD, Potts WB 3rd, Smith NW, Plumb RS. Development of a high sensitivity bioanalytical method for alprazolam using ultra-performance liquid chromatography/tandem mass spectrometry. Drug Test. Anal. 2(1), 11–18 (2010).

29 Rainville PD, Smith NW, Cowan D, Plumb RS. Comprehensive investigation of the influence of acidic, basic, and organic mobile phase compositions on bioanalytical assay sensitivity in positive ESI mode LC–MS/MS. J. Pharm. Biomed. Anal. 59, 138–150 (2012).

�� Investigation of the benefits of high pH mobile phases for bioanalytical LC–MS.

30 Crockford DJ, Holmes E, Lindon JC et al. Statistical heterospectroscopy, an approach to the integrated analysis of NMR and UPLC–MS data sets: application in metabonomic toxicology studies. Anal. Chem. 78(2), 363–371 (2006).

31 Ross P, Knox JH. Carbon-based packing materials for liquid chromatography: applications. Adv. Chromatogr. 37, 121–162 (1997).

32 Neue UD, Kele M, Bunner B et al. Ultra-performance liquid chromatography technology and applications. Adv. Chromatogr. 48, 99–143 (2010).

33 Delatour C, Leclercq L. Positive electrospray liquid chromatography/mass spectrometry using high-pH gradients: a way to combine selectivity and sensitivity for a large variety of drugs. Rapid Commun. Mass Spectrom. 19(10), 1359–1362 (2005).

34 Peng L, Farkas T. Analysis of basic compounds by reversed-phase liquid chromatography–electrospray mass spectrometry in high-pH mobile phases. J. Chromatogr. A 1179(2), 131–134 (2008).

35 Spagou K, Wilson ID, Masson P et al. HILIC–UPLC–MS for exploratory urinary metabolic profiling in toxicological studies. Anal. Chem. 83(1), 382–390 (2011).

36 US FDA. International Conference on Harmonization/Guidance for Industry Safety Testing of Drug Metabolites. Washington, DC, USA (2008).

37 Wiklund S, Johansson E, Sjöström L et al. Visualization of GC–TOF-MS-based metabolomics data for identification of biochemically interesting compounds using OPLS class models. Anal. Chem. 80(1), 115–122 (2008).

ReseaRch aRticle | Rainville, Smith, Cowan et al.

Bioanalysis (2012) 4(23)2842 future science group