(M1030-01-05) Enantiomeric and Regioisomeric Separation of Triacylglycerides in Liver Samples from Doxorubicin-Treated Rats Using LC-ESI-MS with Coupled CHIRALPAK IF3 Columns

Purpose: Accurate profiling of triacylglyceride (TG) isomers is essential for understanding lipid metabolism and its dysregulation in disease and response to drug treatment. While reversed-phase LC-MS (RPLC-MS) offers high sensitivity and detailed fatty acyl profiling, it lacks the ability to resolve regioisomers and enantiomers. These structural variants carry important biological information, particularly regarding stereospecific enzymatic processes. Enzymes such as GPAT-1 initiate TG biosynthesis by catalyzing the stereoselective acylation of glycerol-3-phosphate at the sn-1 position, influencing the enantiomeric ratio of TGs. Doxorubicin, a chemotherapeutic drug, induces oxidative stress and mitochondrial dysfunction, which can impair normal β-oxidation and thereby accumulate specific TG species and isomers in the liver. To address the limitations of conventional lipidomics, we aimed to develop a chiral LC-MS method capable of enantioselective and regioisomeric separation of TGs. This method was applied to liver homogenates from rats treated with doxorubicin, enabling a more detailed characterization of lipid biosynthesis and degradation. Our novel findings reveal shifts in specific enantiomeric ratios of TGs associated with doxorubicin treatment, offering deeper insight into lipid metabolic reprogramming under chemotherapeutic stress.

Methods: Two coupled columns of CHIRALPAK IF3 (250 × 2.1 mm, 3 µm) were used with isocratic elution using acetonitrile/water (98:2, v/v) at 0.15 mL/min at 35 °C using an Acquity-UPLC^TM (Waters) system. MS detection was performed using a Synapt G2-S Q-ToF^TM (Waters) in positive ESI mode. Post-column infusion of 50 mM ammonium formate enabled the detection of [TG+NH₄]⁺ adducts. Rats were divided into three groups: control (saline-injected, sacrificed after 6h), Dox 6h, and Dox 72h (sacrificed after 6 or 72 h post 10 mg/kg doxorubicin, respectively). Livers were homogenized and extracted using acetonitrile/methanol (10:90, v/v) at −20 °C, followed by reconstitution of the precipitate in isopropanol/acetonitrile (50:50, v/v) prior to analysis.

Results: Method optimization using one CHIRALPAK IF-3 column with acetonitrile showed that reducing flow-rate (0.30- 0.08 ml/min) and temperature (35- 25°C) improved enantiomeric resolution. The addition of 2% water to the mobile phase increased the separation of enantiomers further. The method was later used with two coupled CHIRALPAK IF-3 columns which achieved acceptable chiral separation in standards of TG enantiomers and regioisomers with varying unsaturation and chain lengths. However, due to runtimes of up to 110 min, the temperature was kept at 35°C to reduce retention time. Fully mono- or di-unsaturated TG enantiomers did not separate well on the system. In healthy doxorubicin treated rats, enantiomeric pairs such as sn-OPP (oleoyl-palmitoyl-palmitoyl)/ sn-PPO (palmitoyl-palmitoyl- oleoyl) and sn-LPP (linoleoyl-palmitoyl-palmitoyl)/sn-PPL (palmitoyl-palmitoyl- linoleoyl), revealed shifts in their corresponding enantiomeric ratio. sn-OPP and sn-LPP in the Dox 72h group were both upregulated compared to the Dox 6h and control group. Moreover, multivariate data analysis showed a decrease in TGs containing α-linolenic acid (18:3) and linoleic acid (18:2), along with increased levels of TGs containing highly polyunsaturated fatty acids (20:4, 20:5, 22:5, 22:6), in the Dox 72h group.

Conclusion: In conclusion, multivariate analysis showed a downregulation of TG consisting fatty acids 18:2 and 18:3 in combination with an upregulation of highly polyunsaturated fatty acids, 20:4, 20:5, 22:5 and 22:6. These results indicate an activation or upregulation of polyunsaturated fatty acid elongation and desaturation pathways, where essential FAs such as 18:2 and 18:3 are consumed to produce highly unsaturated fatty acids. Furthermore, our results show that the ratio between sn-OPP/sn-PPO as well as sn-LPP/sn-PPL is altered by the treatment of doxorubicin. The upregulation of sn-OPP and sn-LPP, suggest stereoselective effects of doxorubicin, particularly when palmitic acid is positioned at sn-2 and sn-3, suggesting an impaired GPAT-1 activity and altered β-oxidation in the liver. This study demonstrates that chiral LC-MS enables the detection of stereoselective changes in liver TG profiles following doxorubicin treatment, which are not observable with conventional RPLC-MS. These findings underscore the importance of resolving TG isomers and enantiomers to gain deeper insight into lipid metabolic alterations.

Acknowledgements: This reaserch was founded by The Swedish Re-search Council (Vetenskaprådet) 2023-04500 (P.I. Prof. Mikael Hedeland); and from Olga Jönssons Stipendiestiftelse VT24 (P.I. Jenny Nilsson). This animal study was approved by the local ethics committee for animal research (Dnr 5.8.18-00250/2022) in Uppsala, Sweden.

Mean ± SEM of enantiomeric ratio A) sn-OPP/sn-PPO and B) sn-LPP/sn-PPL of liver from doxorubicin treated rats. Control (saline injected, n=5), Dox 6h (doxorubicin treated and sacrificed after 6h, n=5), Dox 72h (doxorubicin treated and sacrificed after 72h, n=4) and quality control samples (QC) for variation from pooled homogenate, p-values **** <0.0001.

Logarithm of the fold change compared to the control group, represented by the zero line, of selected triacylglycerides (TG) with compositions of fatty acids 18:2 and 18:3 in liver tissue. The average of the fold changes is represented with squares and the confidence intervals by bootstrap resampling are represented by the horizontal bars. Each TG represent a chromatographic peak separated by time with two coupled CHIRALPAK IF-3 columns for enantiomeric and regioisomeric separation.

ogarithm of the fold change compared to the control group, represented by the zero line, of selected triacylglycerides (TG) with highly polyunsaturated fatty acids with 22:6 detected in liver tissue. The average of the fold changes is represented with squares and the confidence intervals by bootstrap resampling are represented by the horizontal bars. Each TG represent a chromatographic peak separated by time with two coupled CHIRALPAK IF-3 columns for enantiomeric and regioisomeric separation.