Introduction

Cytochrome P450 (CYP) monooxygenases metabolize arachidonic acid (AA) to four regioisomeric cis-epoxyeicosatrienoic acids (EETs)—14,15-, 11,12-, 8,9-, and 5,6-EET—each consisting of a pair of enantiomers: 14(R),15(S)-/14(S),15(R)-EET, 11(R),12(S)-/11(S),12(R)-EET, 8(R),9(S)-/8(S),9(R)-EET, and 5(R),6(S)-/5(S),6(R)-EET, respectively [1] (Fig. 1). EETs are generated in all tissues [1, 2], sometimes representing the dominant eicosanoid group [3], and display a broad array of biological actions including modulation of electrolyte transport processes [1, 4], potentiation of vascular cell proliferation [5], inhibition of platelet aggregation [1, 6, 7] and vascular cell migration [8], and partially opposite vasoregulative activities from vasodilatation to vasoconstriction [1, 2, 9–14]. In the last decade EETs gained particular interest as putative chemical correlates of the vasodilatative endothelium-derived hyperpolarizing factor (EDHF) in the coronary vasculature [15–17] stimulating proliferation of vascular cells [18] and as anti-inflammatory agents preventing leukocyte adhesion to the vascular wall [19]. EETs are generated stereospecifically by various CYP isoforms from many sources (e.g., 1A1, 1A2, 2A1, 2B1, 2B2, 2C11, 2C12 from rat liver; 2C8, 2C9 from human liver; 2B4 from rabbit lung) [1, 20–27] and their actions are elicited by specific optical isomers: 14(R),15(S)-EET is a stereospecific inhibitor of cyclooxygenase [6], 11(R),12(S)-EET is a potent renal vasodilator [28], and 8(S),9(R)-EET is a renal vasoconstrictor [24], whereas their optical antipodes, 14(S),15(R)-EET, 11(S),12(R)-EET, and 8(R),9(S)-EET, respectively, are inactive.

Fig. 1
figure 1

Formation of EET enantiomers from AA via CYP epoxygenase

Full size image

Despite the multitude of contradictory reports on the physiological actions of EETs and clear evidence for both stereospecific biosynthesis and activity, in most studies they are not assessed by chiral analysis, meaning that no correlation between the occurrence of a specific enantiomer and its putative biological action was possible. Consequently, inconsistent findings regarding the physiology of EETs, as, e.g., the longtime controversy about their putative role as EDHF, might be elucidated by correlating the physiological function with the generation of stereoisomers, rather than with that of regioisomers. The lack of chiral EET analysis is obviously a result of the fact that previous methods are extremely labor intensive and/or time consuming and not sensitive enough. These techniques comprise multistep procedures susceptible to substance loss, like low-yield derivatization prior to individual chiral resolution and quantitation [29, 30] or direct, i.e., without previous derivatization, but likewise individual separation of each regioisomer into the corresponding pair of enantiomers either with a different mobile phase [31, 32] or with the same mobile phase at different flow rates [32] instead of direct and simultaneous chiral resolution of all eight enantiomers within one chromatographic run. Additional proneness to loss of the thermolabile EETs results from the use of high-temperature detection modes. In all cases, quantitative sensitivity is dramatically reduced by the use of large-inner-diameter (4.6 mm) conventional chiral columns with low chromatographic signal intensity instead of capillary columns (i.d. < 500 μm) with typical high quantitative sensitivity. Accordingly, there was compelling rationale to develop a time-saving, direct, simple one-step and highly sensitive capillary chiral analysis as a tool to facilitate and improve work in this area by enabling routine systematic and sensitive assessment of the distribution of free EET antipodes as a prerequisite for new insights into the biological functions of EET enantiomers.

We present a method comprising (i) solid-phase extraction (SPE), (ii) reversed-phase (RP) HPLC purification, (iii) highly sensitive capillary tandem column chiral-phase liquid chromatographic (CapTC-CP-LC) nonoverlapping separation of all eight underivatized EET enantiomers within one run using a setup of two serially coupled columns followed by both online (iv) photodiode array (PDA) and gentle electrospray ionization (ESI) tandem mass spectrometric (MS2) detection in the negative ion mode for identification and quantitation of eluting analytes in the low picogram range. Application of this one-step CapTC-CP-LC-PDAD-ESI-MS2 technique of high quantitative sensitivity (owing to the use of capillary columns) to complex biological material like lung and kidney homogenate demonstrated the robustness, reliability, and sensitivity of this method in routine high-sample-throughput profiling of endogenous free EET enantiomers (“enantioprofiling”), i.e., EETs not esterified into cellular membrane phospholipids, typically occurring at very low concentrations in biological sources. The method can flexibly be used for efficient preparative purification of EET enantiomers in the microgram range with large-i.d. columns.

Experimental

Standards, reagents, and solvents

The racemates (±)14,15-, (±)11,12-, (±)8,9-, and (±)5,6-EET were from Biomol (Hamburg, Germany), Cayman Chemical Company (Ann Arbor, USA), and Cascade Biochem (Reading, UK). The stereoisomers 14(R),15(S)-/14(S),15(R)-EET, 11(R),12(S)-/11(S),12(R)-EET, 8(R),9(S)-/8(S),9(R)-EET, 5(R),6(S)-/5(S),6-(R)EET were from Cascade and WAK Chemie (Bad Homburg, Germany). The antipodes 14(R),15(S)-/14(S),15(R)-EET were also from Sigma (Munich, Germany). The deuterated standards [2H8]14,15-, [2H8]11,12-, [2H8]8,9-EET, and [2H8]5,6-EET-Me and the internal standard (IS) eicosapentaenoic acid (EPA) were from Biomol. (±)14,15-Epoxy-5Z,8Z,11Z,17Z-eicosatetraenoic acid (designated throughout this study simply as EETE) was from Cayman. EET standards and EPA were checked for purity and concentration as described elsewhere [33, 34]. The antioxidant butylated hydroxytoluene (BHT), triethylamine (TEA), diethylamine (DEA), acetic acid (AcOH), and formic acid (FA) were from Sigma. Ethylenediaminetetraacetic acid tetrasodium salt, tetrahydrate (EDTA Na4·4H2O) was from E. Merck (Darmstadt, Germany). Krebs–Henseleit buffer supplemented with 5% hydroxyethyl starch (KHB) was from Serag Wiessner (Naila, Germany). Water was purified with a Milli-Q system (Millipore, Eschborn, Germany). Methanol was from Burdick & Jackson (Muskegon, MI, USA), acetonitrile was from J.T. Baker (Deventer, Netherlands), and isopropyl alcohol (iPA) was from Fluka AG (Buchs, Switzerland). Ethanol (EtOH) and n-hexane (nHex) were from Riedel-de Haën (Seelze, Germany). The solvent mixtures used for SPE are listed in the Electronic Supplementary Material (Table S1).

Analytical procedures and equipment

Sample extraction

Cleanup and concentration were performed by multistep SPE as displayed in the Electronic Supplementary Material (Table S2). For the assessment of recoveries (quality controls) we used tissue homogenate (lung and kidney from rabbit and rat) as biological sample matrix and KHB as EET-free sample matrix. Aliquots (2 ml each) of KHB in homogenizer tubes were supplemented with 60 μl FA, 56 μl BHT 1% in methanol, 40 μl of solution E (Electronic Supplementary Material Table S1) and spiked with mixtures of known amounts of enantiomerically pure EET standards and BHT. Constant amounts of deuterated EETs and/or EPA as IS were added. Alternatively, whole blood-free rinsed rabbit/rat lungs or blood- and urine-free rinsed rabbit/rat kidneys were quickly cut into small pieces. Minced tissue samples (2 g each) were placed into 5-ml chilled homogenizer borosilicate glass tubes, each containing 60 μl FA, 56 μl BHT 1% in methanol, 40 μl of solution E, known amounts of enantiomerically pure EET standards, and constant amounts of deuterated EETs and/or EPA as IS and subjected to mechanical homogenization for 10 min at 1,500 rpm employing a Potter S homogenizer with PTFE plunger mounted on a stainless steel shaft, all from Sartorius AG (Göttingen, Germany). Genuine biological samples (for assessment of endogenous enantiomers from tissue homogenate) were processed in the same manner, except for supplementation with EET standards.

Homogenates (2 g each of quality controls or genuine samples) and 2-ml aliquots of KHB were supplemented with 0.8 ml of solvent mixture F, vortexed, and the contents were transferred to 10-ml glass tubes. Homogenizer tubes were rinsed with 2 ml of solvent mixture F, and the contents were combined with the original sample. The diluted homogenate samples were vortexed for 10 min and centrifuged for 60 min at 4,460 × g and 4 °C.

Prior to SPE, 2.4 ml of supernatant was diluted with 17.6 ml of ice-cold KHB and applied to the conditioned extraction cartridges filled with end-capped octadecyl silyl (ODS) silica (Chromabond; C18 ec, 200 mg sorbent) mounted in a vacuum manifold, all from Macherey-Nagel (Düren, Germany). After the final elution step, the collected samples were dried gently under a stream of N2 and subjected to purification by a modified version of a previous RP-HPLC procedure (see “EET purification by RP-HPLC” in the Electronic Supplementary Material).

TC-CP-LC-PDAD-ESI-MS2 procedures and instrumentation

Stereoisomers were separated with a setup consisting of two consecutively connected columns: an amino column (Grom-Sil 300 Amino-4PR, L = 60 mm, particle size 3 μm) followed by a chiral column (Chiralcel OD-H, L = 250 mm; filled with cellulose tris(3,5-dimethylphenyl carbamate) on a 5-μm silicagel substrate). The following column sizes were used:

  1. (a)

    Capillary columns (i.d. = 300 μm or 150 μm) for CapCP-LC analysis, and

  2. (b)

    Large-i.d. columns, i.e., narrowbore columns (i.d. = 2.0 mm) or conventional columns (i.d. = 4.6 mm) for preparative chiral purification of enantiomers.

All amino columns and the capillary/narrowbore chiral columns were from Grom Analytik + HPLC (Rottenburg-Hailfingen, Germany). The conventional chiral columns were from Daicel Chemical Industries, Niigata, Japan, and Chiral Technologies Europe, Illkirch, France. The same mobile phase consisting of 99.7:0.21:0.09:0.015 (v/v/v/v) nHex, iPA, EtOH, and AcOH was used with all column sizes. Flow rates were 2.8 μl/min, 8 μl/min, 380 μl/min, and 1.8 ml/min with 150-μm-i.d., 300-μm-i.d., 2-mm-i.d., and 4.6-mm-i.d. columns, respectively.

The instrumental setup used for chiral analysis consisted of two main parts: (i) a 1100 Series capillary LC unit (Agilent Technologies Deutschland, Waldbronn, Germany) interfaced with (ii) an Esquire 3000+ ion trap mass spectrometer (Bruker Daltonik, Bremen, Germany).

  1. (i)

    The CapLC unit consisted of a micro autosampler, a fraction collector, a micro pump, and a photodiode array detector (PDAD). This setup could readily be adapted to (a) CapCP-LC analysis employing capillary columns and to (b) preparative HPLC conditions employing large-i.d. columns, as follows:

    1. (a)

      For CapTC-CP-LC-PDAD-ESI-MS2, the autosampler and PDAD were equipped with an 8-μl sample loop and a nano flow cell (500-nl volume, 10-mm path length), respectively.

    2. (b)

      For preparative chiral purification of deuterated and nondeuterated EET enantiomers, autosampler and PDAD were equipped with a 40-μl sample loop and an analytical flow cell (10-μl volume, 10-mm path length), respectively.

  2. (ii)

    The ionization process of analytes necessary for ESI-MS2 detection was assisted, as described elsewhere [35–37], by post-column mixing of the nonpolar water-free LC effluent with a polar makeup solution consisting of a 50:34:16 (v/v/v) mixture of iPA, DEA, and AcOH prior to reaching the ESI source. The makeup solution was infused into the LC effluent of the 300-μm-i.d. capillary columns at a flow rate of 10 μl/h via a tee connector integrated into the fused-silica connecting tubing between PDAD and MS using a syringe pump (Cole-Parmer Instrument Company, Vernon Hills, IL, USA).

N2 was used as nebulizing and drying gas, and He was used as collision gas. Nebulizing gas pressure, drying gas flow, and drying gas temperature were set to 12.0 psi, 6.0 l/min, and 200 °C, respectively. Capillary voltage, collision energy in terms of fragmentation amplitude, isolation width, and mass cutoff were set to 2,800–3,200 V, 0.58–0.67 V, 1.5–2.0 Da, and m/z 85, respectively.

Prior to injection of 8-μl aliquots into the CapTC-CP-LC-PDAD-ESI-MS2 system, biological or quality control samples dried after RP-HPLC purification were redissolved in 40 μl of pure hexane or mobile phase containing 30 μM BHT. Calibration standard mixtures were redissolved and diluted with the same solvents, and supplemented with constant amounts of IS and EETE. The eluting enantiomers were subjected to dual online spectrophotometric and tandem mass spectrometric analysis employing the consecutively interfaced PDAD and ion trap. PDAD provided full UV spectra (190 to 400 nm) of eluting analytes and allowed checking for peak purity and subtraction of possible coeluting material. The ion trap was operated in the negative ion mode employing multiple reaction monitoring (MRM) for the simultaneous acquisition of MS2 spectra and traces of coeluting nonisomeric analyte precursor ions.

The relative retention index (RRI) for chiral chromatographic characterization of EET enantiomers by comparison of their retention times (RT) to those of the reference substances EPA and EETE was calculated by the following equation:

$${\text{RRI}} = \frac{{{\text{RT}}\left( {{\text{analyte}}} \right) - {\text{RT}}\left( {{\text{EPA}}} \right)}}{{{\text{RT}}\left( {{\text{EETE}}} \right) - {\text{RT}}\left( {{\text{EPA}}} \right)}}$$

The elution order of EET enantiomers was assigned by comparison with the chromatographic properties of enantiomerically pure commercial standards as well as enantiomers prepared from commercial racemic standards using the method of Zhang and Blair [31] and/or the preparative version of the present approach.

Thus, triple analyte identification was performed: chromatographically by RRI, and by both UV and MS2 spectra of the peaks detected on the optical and MS2 traces, respectively.

Quantitation was performed by IS method using standard mixtures which contained:

  1. (i)

    EPA as a common IS when UV and MS2 or UV detection alone was employed, or

  2. (ii)

    Deuterated EET standards and EPA as IS when MS2 detection alone was used.

Biological matrix

The distribution of free EET enantiomers was assessed in tissue homogenate from hypoxic lungs as well as from control lungs and kidneys of rabbits and rats (see “Physiological experiments” in the Electronic Supplementary Material).

Results and discussion

Direct and simultaneous CapTC-CP-LC-PDAD analysis

The presented chiral separation technique was developed by starting with the chiral chromatographic conditions described by Zhang and Blair [31] for underivatized EETs, i.e., use of a conventional (4.6-mm i.d.) Chiralcel OD column and resolution of each pair of enantiomers with a separate mobile phase consisting of different proportions of nHex, iPA, and AcOH.

The concept of one-step chiral analysis was to obtain direct (i.e., without EET derivatization), simultaneous, and highly sensitive CapTC-CP-LC separation of all four regioisomers into the corresponding pairs of enantiomers without peak overlapping in one chromatographic run followed by online, i.e., direct and thus sensitive tandem ESI-MS2 identification and quantitation of the eluting antipodes. Direct and simultaneous nonoverlapping chiral resolution within one run was a result of column choice and setup, development of an appropriate mobile phase, and use of optimal column temperature (CT) as detailed in “Method development strategy” in the Electronic Supplementary Material. Optimization trials were performed with a standard mixture consisting of the four racemic EET regioisomers, EPA, and EETE. Eluting compounds were monitored at 204 nm. Ultimately, injection of the standard mixture on the tandem column setup—nonchiral followed by chiral column—resulted in both regioisomeric and enantiomeric separation, i.e., simultaneous and consecutive resolution of all four pairs of enantiomers within one chromatographic run without diastereoisomeric peak overlapping (Fig. 2a).

Fig. 2
figure 2

Simultaneous separation of the four racemic underivatized EET regioisomers into the corresponding consecutively eluting nonoverlapping pairs of enantiomers in one chromatographic run as a result of tandem capillary (i.d. = 0.300 mm) column coupling [a nonchiral Grom-Sil Amino-4PR column for effective regioisomeric separation followed by a Chiralcel OD-H chiral column for simultaneous enantiomeric resolution, both with CT set to 26 °C] and isocratic normal-phase elution using a 99.7:0.21:0.09:0.015 (v/v/v/v) mixture of nHex, iPA, EtOH, and AcOH as mobile phase at a flow rate of 8 μl/min. An 8-μl aliquot of a mixture consisting of the four racemic regioisomeric EET standards, EPA as an IS and a reference substance for RRI calculation, and EETE as a second reference substance for RRI calculation were injected into the CapTC-CP-LC-PDAD-ESI-MS2 system. Online PDAD enabled monitoring of CapTC-CP-LC-PDAD chromatograms at 204 nm and identification of eluting analytes by online spectra plot on the peak apex (see inset) as well as peak purity control-assisted quantitation (a). Serial interfacing of the PDAD with an ESI ion trap mass spectrometer enabled online monitoring of CapTC-CP-LC-ESI-MS2 chromatograms of the eluting compounds in the negative ion mode employing MRM. MS2 analysis was performed by identification and quantitation of substance peaks on the EIC MS2 traces defined for the precursor anions at m/z 319 (EET enantiomers) (b, e, f, g, and h), 301 (EPA) (c), and 317 (EETE) (d). The regiospecific EIC MS2 trace definitions of the neighboring pairs of enantiomers 14(S),15(R)-EET/14(R),15(S)-EET (e) and 11(S),12(R)-EET/11(R),12(S)-EET (f) as well as 8(R),9(S)-EET/ 8(S),9(R)-EET (g) and 5(R),6(S)-EET/5(S),6(R)-EET (h), respectively, contained no common fragments, thus enabling accurate assessment of the enantiomeric ratios even under conditions of coelution between the regioisomeric pairs 14,15-EET and 11,12-EET as well as 8,9-EET and 5,6-EET, respectively, e.g., due to aging-dependent alteration of the amino column. The nonregiospecific definition of the EIC MS2 trace in b included all signals from the regiospecific EIC MS2 trace definitions in panels e, f, g, and h. Peak designation: 1 = (±)14,15-EET; 1a = 14(S),15(R)-EET; 1b = 14(R),15(S)-EET; 2 = (±)11,12-EET; 2a = 11(S),12(R)-EET; 2b = 11(R),12(S)-EET; 3 = (±)8,9-EET; 3a = 8(R),9(S)-EET; 3b = 8(S),9(R)-EET; 4 = (±)5,6-EET; 4a = 5(R),6(S)-EET; 4b = 5(S),6(R)-EET; IS = EPA; EETE

Full size image

CapTC-CP-LC-ESI-MS2 analysis

The molecular carboxylate ions of the EETs, EETE, and EPA (designated a) yielded noncharacteristic fragments due to neutral loss of CO2 (b = a − CO2), H2O (a − H2O), CO2 and H2O (b − H2O), and characteristic fragments due to the position of the epoxide function and double bonds (Table S3). The MS2 spectra are shown in Electronic Supplementary Material Fig. S3 with the characteristic fragments designated c, d, e, f etc. The online MS2 spectra obtained by fragmentation of the molecular ions of the EETs (a at m/z 319) under normal-phase conditions were similar to those generated under reversed-phase conditions [34, 38–42].

The MS raw data acquired with the online interfaced ion trap operated in the negative ion mode using MRM were used, as described elsewhere [38, 39], for the definition of extracted ion chromatograms (EIC MS2) of the eluting analytes by summing the intensities of substance-specific signals from the MS2 spectra of the precursor ions (Fig. 2b–d). Moreover, regiospecific EIC MS2 trace definitions (Fig. 2e–h) enabled chiral analysis even under conditions of overlapping diastereoisomeric peaks, i.e., when the regioisomers 14,15- and 11,12-EET or/and 8,9- and 5,6-EET are poorly resolved. These EIC MS2 trace definitions contain a restricted selection of regioisomer-specific signals, so that the traces of the neighboring regioisomers 14,15-EET and 11,12-EET (EIC 219; 205; 175; 113; 99 MS2 (319) and EIC 208; 195; 179; 167; 163; 149; 135 MS2 (319), respectively), and those of 8,9-EET and 5,6-EET (EIC 221; 179; 155; 151; 139; 127; 123 MS2 (319) and EIC 219; 205; 191; 129; 115; 99 MS2 (319), respectively) do not share common signals. Thus, the present tandem-mass-spectrometry-based technique enabled individual identification and quantitation of each regioisomer by substance-specific EIC MS2 traces, regardless of isomeric matrix interferences.

Preparative purification of EET enantiomers

Preparative purification of optical isomers in the microgram range from commercial racemates was performed with the same mobile phase as for CapCP-LC and the corresponding setup of large-i.d. columns, as specified in the Experimental. Chromatographic selectivity and the influence of CT on chiral resolution was similar under capillary, narrowbore, and conventional LC conditions. Thus, the UV chromatograms obtained with large-i.d. columns (not shown) corresponding to the three separation modes described in “Method development strategy” in the Electronic Supplementary Material, i.e., (i) CT = 26 °C, (ii) CT = 4 °C, and (iii) CT = 36 °C, were similar to the CapCP-LC traces in Electronic Supplementary Material Figs. S2c, d, and e, respectively. Consequently, for purification with maximal recovery of the 14,15-, 11,12-EET, and EETE enantiomers, mode (ii) at CT = 4 °C was used. The second EETE antipode was used as reference substance for chiral chromatographic RRI calculation. Mode (iii) at CT = 36 °C was used for efficient purification of the 8,9- and 5,6-EET antipodes.

Assessments of calibration and recovery

Calibration of UV and MS signal performed by IS method, employing EPA as a common IS for all analytes, yielded linear regression functions (R 2 > 0.998) for all EET stereoisomers (exemplified for MS2 signal calibration in Electronic Supplementary Material Fig. S4).

SPE followed by RP-HPLC purification and CapTC-CP-LC-PDAD-ESI-MS2 was applied to biological samples and to EET-free KHB, both spiked with known amounts of nondeuterated enantiomerically pure standards, and deuterated enantiomers and/or EPA as IS (see the Experimental), yielding linear plots and recoveries >70% for all EETs (Electronic Supplementary Material Table S4). Comparison of recoveries obtained by adding constant amounts of EPA and deuterated EETs as IS before SPE, prior to any manipulation, to both the biological matrix and KHB, showed no significant differences (<RSDmax = 9%), demonstrating that EPA and deuterated EETs were equally suitable for serving as the IS. Moreover, the ratios between the recovery of EPA and the recoveries of all enantiomers remained unchanged during sample preparation. Thus, we decided on the more convenient use of EPA as an IS for both MS2 and PDA detection and, together with EETE, as a reference substance for the calculation of the chiral RRI.

With 300-μm-i.d. columns, the limits of quantitation (LOQ) for MS2 detection of EET enantiomers, defined as the on-column analyte amount at a signal to noise (S/N) ratio >12, were 30 pg, 30 pg, 40 pg, and 60 pg for the 14,15-, 11,12-, 8,9-, and 5,6-enantiomers, respectively. With 150-μm-i.d. columns, the LOQ for MS2 detection were four times lower. Overall, the LOQ for PDAD were higher than MS2 detection results by a factor of 30. RSD < 9%, n = 6.

The actual recoveries were checked with quality control samples freshly prepared prior to SPE of a sample batch as specified in “Quality control samples” in the Electronic Supplementary Material.

Characteristics of CapTC-CP-LC-PDAD-ESI-MS2 analysis

Although separation of EET enantiomers with Chiralcel columns goes back to the end of the 1980s [29] and has successfully been used for elucidation of the stereoselective biosynthesis and physiology of EETs [6, 12, 21–28], these methods have a series of disadvantages with two main detrimental consequences—(a) very long overall analysis time and (b) low quantitative sensitivity—rendering them unpractical for sensitive and reliable routine chiral analysis with high sample throughput.

The inherent disadvantages of previous methods employing individual conventional CP-HPLC separation of each enantiomer were overcome by the present tandem column one-step chiral analysis, which provides the highest quantitative sensitivity owing to the use of capillary columns, and displays the following main advantages:

  1. (1)

    High sample throughput compared with previous chiral methods for profiling of EET enantiomers, thus enabling routine analysis,

  2. (2)

    LOQ in the low picogram range, and

  3. (3)

    Reliable identification and quantitation by online and dual complementary UV spectrometric and tandem MS2 detection (see specification in “Features of the developed CapTC-CP-LC-PDAD-ESI-MS2 technique” in the Electronic Supplementary Material).

Profiling of EET enantiomers in rabbit and rat lungs and kidneys by one-step CapTC-CP-LC-PDAD-ESI-MS2 analysis

Application of one-step CapTC-CP-LC-PDAD-ESI-MS2 to homogenate from hypoxically ventilated rabbit lungs, control rabbit, and rat lungs and kidneys revealed source-specific chiral chromatographic profiles (Fig. 3; the corresponding MS2 spectra of endogenous EETs are depicted in Electronic Supplementary Material Fig. S5) and enantiomeric ratios of endogenous free EETs (Table 1).

Fig. 3
figure 3

Chromatographic profiles of EET enantiomers from hypoxic rabbit lung homogenate obtained by CapTC-CP-LC-PDAD-ESI-MS2. Homogenate samples were subjected to SPE followed by RP-HPLC purification and finally analyzed by isocratic one-step CapTC-CP-LC-PDAD-ESI-MS2 as described in Fig. 2. Eluting compounds were assessed by integration of the peaks detected at 204 nm by online PDAD (a) with analyte identification by spectra plot on the peak maximum (inset in panel a) and subsequent tandem MS2 detection with the online interfaced ion trap in the negative ion mode employing MRM. Integration and MS2 analysis of substance peaks were performed on the EIC MS2 traces defined for the precursor anions at m/z 319 (EETs), 301 (EPA), and 317 (EETE) (b, c, and d, respectively), as well as on the regiospecific EIC MS2 traces defined for the peak pairs 14(S),15(R)-EET/14(R),15(S)-EET (e), 11(S),12(R)-EET/11(R),12(S)-EET (f), 8(R),9(S)-EET/8(S),9(R)-EET (g), and 5(R),6(S)-EET/5(S),6(R)-EET (h) (see also Fig. 2e–h). For assessment of the enantiomeric ratios, the data obtained by peak integration on both the nonregiospecific EIC MS2 trace defined for all EET regioisomers (b) as well as on the regiospecific EIC MS2 trace definitions (e, f, g, h) were employed. Owing to optimal regioisomeric separation, the results obtained with these two types of EIC MS2 trace definitions were virtually identical. Peak designation is the same as in Fig. 2

Full size image
Table 1 Profiling of free EET enantiomers (“enantioprofiling”) generated in rabbit and rat lung and kidney
Full size table

Although, as described [39], hypoxic challenge of rabbit lungs for 4 h elicited the enhanced generation of all EET regioisomers, chiral analysis exhibited unchanged enantiomeric EET ratios, except for 14,15-EET (Table 1), thus suggesting only partial stereospecific EET contribution to sustained hypoxic pulmonary vasoconstriction. This finding does not exclude enantiospecific contribution of the other EETs too to hypoxic pulmonary hypertension with vascular remodeling due to the induction of specific CYP isoenzyme expression under conditions of prolonged (16–24 h) or chronic (up to 1 week) hypoxia [43, 44].

Remarkably, assessment of the distribution of free EET enantiomers in control lungs and kidneys of rabbits and rats (Table 1) revealed both species- and organ-specific enantiomeric EET generation profiles. As the preferential formation of certain EET enantiomers is indicative of enzymatic rather than autooxidative generation and each CYP isoenzyme is known to display its characteristic enantiomer generation profile [23, 26], the results presented in Table 1 demonstrate both the biosynthetic origin of the EETs as well as species and organ specificity of CYP isoform expression.

Conclusions

Chiral analysis by CapTC-CP-LC nonoverlapping separation of all eight underivatized EET antipodes within one chromatographic run followed by online dual complementary PDA and tandem MS2 detection enables highly sensitive, simultaneous, direct, and reliable identification and quantitation, thus providing comprehensive and rapid information about the enantioselective distribution of free endogenous EETs, typically occurring at very low concentrations, at once and with minimal effort but maximal sample throughput. Application of capillary-chiral one-step analysis to complex biological sources demonstrated its usefulness as a convenient and attractive method, which provides in a minimum of time the complete enantioselective EET generation profile at a glance without any laborious and/or time-consuming multistep procedures. Highly sensitive and routine “enantioprofiling” of endogenous EETs enabled by CapTC-CP-LC, and the use in physiological experiments of exogenous enantiomers obtained by the preparative version might be a prerequisite for new insights and elucidation of contradictory findings regarding EET physiology in future studies. To the best of our knowledge, these tasks cannot be accomplished by presently available methods.