Kidney transplantation is a life-saving method for patients with end-stage renal dysfunction that enables higher survival rates and patient quality of life compared to dialysis treatment. Unfortunately, an ongoing organ shortage has led to the rapid growth in the number of patients on kidney waiting lists (Swanson et al., 2020; Warmuzińska et al., 2022). This growing gap in supply and demand has led clinicians to explore the possibility of using kidneys recovered from extended criteria donors (ECD) or donation after circulatory death (DCD) donors; however, recipients of ECD kidney grafts tend to have worse outcomes than those receiving organs from standard criteria donors, including being at a higher risk of delayed graft function (DGF) and primary nonfunction incidence (Urbanellis et al., 2020; Warmuzińska et al., 2022). Hence, strategies for reducing preservation injury and monitoring graft function are of intense interest. At present, static cold storage (SCS) and hypothermic machine perfusion (HMP) are the most common preservation methods applied in clinical settings. SCS involves submerging the kidney in a cold preservation fluid and then placing it on ice in an icebox; in contrast, HMP entails using a device to pump cold preservation fluid through the renal vasculature, which has been demonstrated to be more effective at preserving marginal and DCD grafts compared to SCS (Lindell et al., 2013; Urbanellis et al., 2020; Warmuzińska et al., 2022). Normothermic ex vivo kidney perfusion (NEVKP) is a novel dynamic preservation strategy applying a perfusion solution’s circulation through the kidney. NEVKP conditions are developed to replicate physiological conditions as closely as possible to reduce cold ischemia damage and improve graft outcome (Resch et al., 2020; Hosgood et al., 2021). While several studies have attained promising results indicating NEVKP’s superiority over SCS, this application is still in the experimental stage (Kaths et al., 2017; Urbanellis et al., 2020). One notable problem is the need for more accurate methods of evaluating graft quality and assessing donor risk, especially concerning marginal grafts. A kidney’s suitability for transplantation is determined based on detailed parameters, including the donor’s medical history, visual assessment, and examination results (Warmuzińska et al., 2022). The visual evaluation of donor organs is frequently fundamental in decision-making. However, although macroscopic inspection can help diagnose tumors and anatomical changes, this method is subjective and depends on the transplant team’s experience level (Dare et al., 2014). Consequently, pretransplant biopsies remain the gold standard for identifying donor kidney injury. Histological examinations are often applied selectively, mainly in marginal grafts, and the frequency of performed biopsies varies between medical facilities and countries (Dare et al., 2014; Moeckli et al., 2019; Warmuzińska et al., 2022). Moreover, the use of biopsies is hampered by two significant limitations: the low reproducibility of results between on-call pathologists and their time-consuming nature (Azancot et al., 2014). Additionally, the number of allowable biopsies during the transplantation procedure is usually restricted to a single sampling due to their invasiveness, which limits their application for capturing dynamic changes and time-series analyses (Bojko, 2022). Therefore, new representative organ-quality assessment methods are needed to increase the number of organs available for transplantation. In this study, we assess the viability of solid-phase microextraction (SPME) chemical biopsy as a method for evaluating the impact of hypothermic and normothermic preservation techniques on the lipidomic profile of the kidney. The small diameter of the SPME probe (∼200 μm) enables minimal invasiveness and allows for several samplings of the same organ without damaging the tissue. Furthermore, SPME combines sampling, extraction, and metabolite quenching into a single step, which makes it a valuable tool for on-site analysis. Finally, SPME’s low invasiveness enables its application for monitoring of changes in the organ throughout the entire transplantation procedure, beginning with its removal from the donor’s body, through its preservation, and ending with its revascularization in the recipient’s body (Łuczykowski et al., 2023).

The proposed method was employed to investigate changes in the lipidomic profiles of kidneys during transplantation and preservation. Principal component analysis was employed to confirm the quality of the instrumental analysis for all combinations of chromatographic separation and ionization mode. As shown in Supplementary Figure S1, the pooled QC samples formed a tight cluster, thus confirming the good quality of the analytical performance. Using all four blocks of LC-HRMS analysis, 128 lipid species belonging to 14 lipid classes were annotated with level 2 confidence in metabolomics compound identification. A UpSet plot representing the number of compounds annotated in each analytical block is shown in Figure 2. A list of the identified lipid species is provided in Supplementary Table S1.

The analysis of the results was divided into four parts: 1) analysis of how warm ischemia influenced the lipidomic profiles of the kidneys; 2) comparison of the three organ-preservation methods; 3) monitoring changes across time; and 4) investigation of the influence of transplantation procedure on kidney grafts.

Organ-preservation methods have garnered significant interest in graft quality evaluation, advanced organ monitoring, and the treatment of transplanted kidneys during machine perfusion. This study further explores these trends and, to the best of our knowledge, it is the first to apply SPME to compare SCS, NEVKP, and HMP preservation methods in a porcine DCD autotransplantation model. Moreover, this study highlights the alterations in the lipidomic profile during warm ischemia and perfusion, as well as after transplantation. The minimally invasive SPME chemical biopsy allows for the repeated direct sampling of the organ, as it does not require any tissue collection. Thanks to this undisputed advantage, it was possible to validate the PLS-DA models and identify significantly differentiating lipids using a relatively small number of animals. Additionally, this study employed four LC-HRMS analyses to cover a wide range of lipids. The analysis showed that this approach allowed the identification of more lipids than was possible using only one type of chromatographic separation. The largest number of unique compounds were identified with the use of RP and HILIC separation in positive ionization mode. In negative ionization mode, using HILIC and RP separations, 9 and 4 unique compounds were identified, respectively. Therefore, based on the results obtained and taking into account the time-consuming nature of the analyses, the consumption of organic reagents, the amount of data obtained, and the benefits of using additional LC-HRMS analyses in future studies, it is worth considering limiting the number of analyzes used, thus choosing the best compromise between advantages and disadvantages.

A thorough analysis of kidney graft tissue during warm ischemia indicated increased levels of CARs, LPCs, and LPEs. CARs are established mitochondrial biomarkers in neonatal screening; however, they are not routinely used beyond this screening, despite the growing evidence of their biomarker potential among disorders such as diabetes, sepsis, cancer, and heart failure (Breit and Weinberger, 2016; McCann et al., 2021). Moreover, alterations of CARs related to kidney disease, including acute kidney injury, progression of chronic kidney disease (CKD), and diabetic nephropathy, have been previously reported in several reports (Breit and Weinberger, 2016; Hocher and Adamski, 2017; Afshinnia et al., 2018; Kordalewska et al., 2019; Andrianova et al., 2020). In this study, significantly elevated levels of four long-chain CARs (CAR 14:0, CAR 14:1, CAR 14:2, CAR 18:2) and one medium-chain CAR (CAR 12:0) were mainly observed after 45 min of warm ischemia. CARs are crucial for transporting long-chain fatty acids into the mitochondria to ensure the β-oxidation process (Breit and Weinberger, 2016; Kordalewska et al., 2019). Thus, the increase of CARs observed during warm ischemia may be related to mitochondria dysfunction. Indeed, a similar accumulation of long-chain CARs in ischemic tissue has been reported elsewhere, and it is thought that this accumulation inhibits oxidative phosphorylation, induces mitochondrial membrane hyperpolarization, and stimulates the production of reactive oxygen species (ROS) (Liepinsh et al., 2016; Andrianova et al., 2020). LPCs and LPEs are the products of the phospholipase-induced hydrolysis of PCs and PEs (Xu et al., 2015; Law et al., 2019). These compounds play a role in cellular signal transduction, tumorigenesis, angiogenesis, immunity, atherosclerosis, cancer, and neuronal survival (Kordalewska et al., 2019). Previously, higher levels of LPCs have been associated with oxidative stress and pro-inflammatory effects (Kordalewska et al., 2019). Moreover, similar to the results obtained in this study, Xu et al. (2015) observed higher pretransplant levels of LPCs (LPC 16:0, LPC 18:0) and LPEs (LPE 16:0, LPE 18:0) in donors after circulatory death compared to donors after brain death. Conversely, Rao et al. found higher levels of LPC 18:0, LPC 26:6, LPE 16:0, and LPC 18:0, along with a corresponding reduction in LPC 20:0, LPC 20:4, LPE 22:0, and LPE 24:6, 24 h after IR-induced acute kidney injury (Rao et al., 2016). The inconclusive results obtained in recent studies may be related to the complexity of the enzymatic cascade involved in LPC metabolism (Law et al., 2019).

Chemometric analysis revealed significant differences between the SCS, HMP, and NEVKP preservation methods, with further in-depth analysis demonstrating that the method’s preservation temperature has a greater impact on the lipidomic profile than its mechanical character. Higher levels of CARs, PCs, ether-linked PCs, ether-linked PEs, phosphatidylinositols (PIs), TGs, most LPCs and LPEs, and longer-chain PEs were observed in the hypothermic group, while higher levels of ceramides (Cer), PSs, and shorter-chain PEs were observed in the normothermic group. Numerous SMs differentiated the preservation methods, but no dominant trend was observed. As mentioned above, increased levels of long-chain CARs may be related to mitochondria dysfunction and increased production of ROS (Liepinsh et al., 2016; Andrianova et al., 2020), and higher levels of LPCs may be associated with oxidative stress and pro-inflammatory effects (Kordalewska et al., 2019). Moreover, higher levels of LPEs were observed in DCD livers compared to livers from donors after brain death (Xu et al., 2015), and increased LPE levels were observed in CKD progression (Kordalewska et al., 2019). Rao et al. (2016) and Solati et al. (2018) have reported elevated levels of PCs in mice and rat ischemia/reperfusion (I/R) models. Similar to our results, they observed increases in PC 34:1, PC 34:2, PC 36:1, PC 36:2, PC 36:3, PC 36:4, PC 38:3, PC 38:4, PC 38:5, PC 38:6, PC 40:4 and PC 40:5. Additionally, Solati et al. reported increased levels of PIs 6 h and 12 h after I/R, while in this study higher levels of PI 36:2 and PI 38:4 were observed in the hypothermic preservation group (Solati et al., 2018). Ether-linked PCs and ether-linked PE were present at higher levels in the hypothermic group compared to the normothermic group. In Rao et al.’s study, significantly elevated levels of a small number of ether-linked phospholipids were observed 6 h after renal I/R (vs. control animals); notably, levels of these ether phospholipids were correlated with plasma creatinine. However, 24 h after I/R, only PC O-38:1 was still elevated, while ether-linked PEs were reduced at this later time point (Rao et al., 2016). Ether lipids differ from their diacyl counterparts, allowing them to contribute unique structural characteristics to biological membranes. Moreover, previous findings suggest that ether lipids play a role in various biological processes, including cell differentiation, cellular signalling, and oxidative-stress reduction (Dean and Lodhi, 2018). While alterations of ether lipids have been reported in several disorders, including neurodegenerative disease, cancer, and metabolic disorders (Rao et al., 2016; Dean and Lodhi, 2018), the molecular mechanism underlying their role in these pathologies remains unclear. Higher levels of three TGs (TG 44:1, TG 56:6, TG 56:7) and three longer-chain PEs (PE 38:4, PE 38:5, PE 40:4) were also observed in the hypothermic group. Similar elevations were reported by Afshinnia et al. (2018) in their study of lipid profiles in CKD. Their results indicated that higher quantities of long polyunsaturated lipids are related to CKD progression. Elsewhere, Solati et al. (2018) observed an increase in longer-chain PEs and a reduction in shorter-chain PEs 24 h after I/R, which aligns with the trend observed for the hypothermic group in our study. The alterations described above suggest the NEVKP method’s beneficial effects with respect to kidney grafts. The lower accumulation of lipids in NEVKP kidneys, especially those with pro-inflammatory properties, results in improved graft function following NEVKP compared to hypothermic preservation methods. Additionally, the reductions of CAR 12:0 and CAR 14:1 across perfusion time points may also be indicative of NEVKP’s beneficial effects.

A much smaller number of compounds differentiating the HMP, NEVKP, and SCS methods were identified post-perfusion (vs. during preservation). However, after reperfusion, the trend in changes was similar to that observed during perfusion, with most of the significant changes differentiating NEVKP from other hypothermic methods, rather than from HMP and SCS.

A comparison of the kidneys during reperfusion and on POD3 revealed more significant changes in the SCS group compared to the NEVKP group. In particular, the SCS group exhibited elevated levels of ether-linked PEs and PC P-38:4 on POD3. As noted above, Rao et al. reported higher levels of PC O-38:1, PE O-42:3, and PE O-40:4 6 h after I/R, but only PC O-38:1 remained elevated in I/R at 24 h. However, more research is required to investigate the molecular mechanisms underlying the role of ether lipids in kidney disease. Additionally, in the SCS group, lower levels of SMs were identified on POD3. Previous findings have shown an increase in SMs 24 h after I/R in animal models (Rao et al., 2016; Solati et al., 2018). In contrast, Zhao et al. (2014) observed lower SMs levels in an acute graft rejection group after transplantation, while Tofte et al. (2019) demonstrated that higher levels of specific SMs were associated with a lower risk of end-stage renal disease and all-cause mortality in type 1 diabetes. These findings indicate that lower levels of SMs may be related to impaired graft function. The results of this comparison (donor vs. POD3) provide further evidence that NEKVP enables improved graft function compared to SCS.

The last step of untargeted analysis entailed a comparison of the lipidomic profiles at baseline (donation) and reperfusion and at baseline and POD3. The largest number of statistically significant compounds was identified in the donor-POD3 comparison for the SCS group. While the trends in changes were similar for the SCS and NEVKP groups, the alterations were usually more noticeable in the SCS group. The observed alterations mainly concerned lipids described previously in this study. Higher levels of CARs, a few LPCs and LPEs, several PCs and PEs, and ether-linked PCs on POD3 were observed for both the SCS and NEVKP groups. However, more compounds among these lipid classes were statistically significant for the SCS group. As discussed above, the elevated levels of these lipids may be related to I/R injury, mitochondrial dysfunction, pro-inflammatory effect, and/or oxidative stress. Significant reductions in SM levels on POD3 were observed only for the SCS group. As previously noted, prior findings have shown a connection between reduced SM levels and acute graft rejection (Zhao et al., 2014).

Despite these promising results, this study has several limitations. The first limitation is the small sample sizes. However, since SPME enables multiple analyses over time, it is possible to acquire a large number of samples without requiring more animals. Another limitation was that the obtained results were not compared with routinely assessed clinical parameters; as such, the analysis was not as comprehensive as it might have otherwise been. A final limitation of this study design is the relatively short follow-up period of 3 days post-transplantation. Unfortunately, this interval was necessary, as the porcine experimental setup makes it difficult to investigate over a longer follow-up period of months, or even years.

SPME chemical biopsy followed by LC-MS/MS analysis enables the minimally invasive and repeated sampling of the same tissue. As such, this method was successfully applied to track alterations in a graft throughout the entire transplantation procedure, and to compare kidney lipidomic profiles during storage with different preservation methods. As a result, we observed that the preservation temperature has a more significant impact on the lipidomic profile of the kidney than the preservation method’s mechanical characteristics. Higher levels of CARs, PCs, ether-linked PCs, ether-linked PEs, PIs, TGs, most LPC and LPE, and longer-chain PEs were observed in the hypothermic preservation group, which may be related to I/R injury, mitochondrial dysfunction, pro-inflammatory effect, and oxidative stress. Hence, the obtained results suggest that the use of NEVKP can have a beneficial effect on graft function.