Lysolecithin Attenuates LPS-Induced Acute Liver Injury in Weaned Piglets by Inhibiting M1 Macrophage Polarization via the mTOR–Glycolysis Pathway

TL;DR

Lysolecithin reduces liver damage in piglets caused by bacterial toxins by inhibiting harmful immune responses and improving liver metabolism.

Contribution

This study reveals that lysolecithin protects the liver by modulating macrophage polarization and the mTOR–glycolysis pathway.

Findings

Lysolecithin supplementation ameliorates LPS-induced liver injury in piglets.

Lysolecithin inhibits M1 macrophage polarization by modulating the mTOR–glycolysis pathway.

Lysolecithin restores metabolic gene expression related to glycolysis and the TCA cycle.

Abstract

Liver injury poses a significant health burden and is frequently initiated by bacterial infections. This study examined whether lysolecithin—a natural bioactive lipid—could confer hepatoprotection against bacterial toxin-induced damage by modulating macrophage activity. Using a weaned piglet model, we assessed the effects of dietary lysolecithin supplementation in animals challenged with lipopolysaccharide. Results indicated that lysolecithin significantly attenuated lipopolysaccharide-induced liver injury, improved hepatic histopathology and function, restored immune cell homeostasis, and optimized cellular energy metabolism. Moreover, lysolecithin was found to regulate a key intracellular signaling pathway involved in macrophage polarization. Collectively, these findings suggest that lysolecithin alleviates liver injury by fine-tuning immune responses and metabolic pathways. This study holds translational relevance, offering novel perspectives for developing potential hepatoprotective agents or therapeutic strategies against liver-related diseases.

Macrophage polarization is widely recognized as a pivotal role in the maintenance of liver homeostasis. Lysolecithin (LPC) has previously been associated with hepatoprotective effect. This study aimed to determine whether LPC protected against lipopolysaccharide (LPS)-induced liver injury by modulating macrophage polarization. Twenty-four piglets were allocated in a 2 × 2 factorial design, involving dietary supplementation (0 vs. 0.01% LPC) and immunological challenge (saline vs. LPS). Animals were euthanized 4 h post-injection, and liver tissues were harvested for analysis. Our findings showed that LPS challenge induced significant liver damage, which was ameliorated by LPC supplementation, as evidenced by improved histological and functional outcomes. LPC counteracted the LPS-induced dysregulation of mRNA expression related to macrophage polarization, including pro-inflammatory markers (IL-6, IL-1β, TNF-α, IFN-γ, iNOS, and CD80) (p < 0.05). Furthermore, LPC restored the expression of key metabolic genes involved in glycolysis and the TCA cycle (HK2 and IDH) (p < 0.05). Mechanistically, LPC normalized the activation of the mTOR signaling pathway by modulating both mRNA and protein levels of mTOR, S6K1, and HIF-1α (p < 0.05). These findings suggest that LPC attenuates LPS-induced liver injury by influencing metabolic and inflammatory pathways, potentially through the inhibiting M1 polarization mediated by glycolysis-related mTOR signaling pathways. Targeting macrophage polarization by LPC may represent a promising therapeutic strategy for inflammatory liver conditions.

Figures

Full text

1. Introduction

The liver serves as a vital metabolic organ and primary site for detoxification. However, when exposed to pathogen invasion, toxin exposure, or stress conditions, the liver triggers excessive release of inflammatory cytokines, leading to inflammatory responses and hepatic dysfunction [1].

The disruption of immune homeostasis constitutes a core pathological mechanism driving liver inflammatory injury and disease progression. Macrophages, as the predominant immune cell population in the liver, play essential roles in pathogen clearance, inflammation modulation, and tissue homeostasis maintenance. Under pathological conditions, not only resident macrophages but also recruited macrophages infiltrate the diseased tissue. These cells undergo polarization into distinct phenotypes—primarily M1 and M2—in response to microenvironmental cues, thereby fulfilling diverse functional roles [2,3]. Classically activated M1 macrophages produce high levels of pro-inflammatory cytokines to combat infection, while alternatively activated M2 macrophages promote inflammation resolution through anti-inflammatory mediators [4,5]. For instance, Jin et al. [6] reported that LPS-induced acute liver injury in mice is associated with enhanced M1 and suppressed M2 polarization. Similarly, Liu et al. [7] demonstrated that IL-34 alleviated immune-mediated liver damage by promoting M2 macrophage polarization, underscoring the importance of macrophage phenotypic balance in hepatic inflammation. Although LPS strongly induces M1 polarization, it can simultaneously trigger a compensatory anti-inflammatory response, leading to the upregulation of certain M2 markers such as IL-10 and TGF-β as a physiological feedback mechanism to limit excessive tissue damage [8,9]. Therefore, maintaining an appropriate balance between M1 and M2 polarization is critical for preserving hepatic health.

Macrophage polarization is closely linked to cellular metabolic reprogramming, particularly involving glycolysis and the tricarboxylic acid (TCA) cycle, processes regulated by signaling pathways such as mTOR and LKB1 [10,11]. When LPS-inducible glycolysis is attenuated, macrophages tend to shift toward the M2 phenotype, accompanied by a reduced inflammatory response. M2 macrophages primarily rely on oxidative phosphorylation (OXPHOS) fueled by the TCA cycle to meet their ATP demands. In contrast, M1 polarization is supported by reactive oxygen species (ROS) and glycolytic intermediates [12]. Under conditions of energy stress, the LKB1–SIRT1 axis inhibits mTOR activity and glycolysis, thereby promoting the TCA cycle and OXPHOS [10,11]. Thus, nutrient-sensitive metabolic reprogramming represents a promising target for modulating macrophage polarization.

Lysolecithin (LPC), a highly amphiphilic endogenous molecule, acts as a potent natural emulsifier. Dietary LPC supplementation in weaned piglets effectively emulsifies dietary fats—especially saturated fats—generating smaller fat micelles. This process markedly increases the interfacial area available for pancreatic lipase, thereby substantially enhancing fat digestibility [13]. In addition to its role in lipid digestion, LPC promotes the shift from pro-inflammatory M1 to anti-inflammatory M2 macrophage phenotypes, contributing to inflammation resolution, fibrosis attenuation, and tissue repair [14]. Nevertheless, whether LPC modulates macrophage polarization via metabolic pathways, such as glycolysis and the TCA cycle, remains unclear.

In this study, we established an LPS-induced model of hepatic injury in piglets to investigate whether LPC mitigated liver injury by modulating macrophage polarization via glycolysis and TCA cycle-associated signaling. Our results provide new theoretical insights into the hepatoprotective mechanisms of LPC.

2. Materials and Methods

2.1. Animals and Husbandry

All animal experimental procedures involving weaned piglets were approved by the Animal Care and Use Committee of Wuhan Polytechnic University (Wuhan, China) (Approval No. WPU201907002). A total of 24 crossbred weaned barrows (Duroc × Large White × Landrace), with an average body weight of 8.62 ± 0.63 kg, were purchased from Aodeng Agriculture and Animal Husbandry Technology Co., Ltd. (Tianmen, China). The piglets were housed in groups of pens (1.8 m × 1.1 m), equipped with plastic slatted flooring. Throughout the study, animals were vaccinated in accordance with the farm’s standard immunization protocol. The ambient temperature was maintained at 27 °C during the first week and gradually reduced by approximately 1 °C per week thereafter. All pigs had ad libitum access to feed and water.

2.2. Experimental Design

In this study, 24 piglets were first divided into 2 groups (12 piglets in each group) using a randomized complete block design. One group was fed a control diet (Ctrl), and the other was fed a 0.01% LPC diet. [Lysolecithin (≥90% purity, as acetone—insolubles) was obtained from Centree Bio—Tech Co., Ltd. (Wuhan, China)]. The feeding period lasted for 28 days. Respectively, the diets were formulated to meet or exceed the nutrient requirements for weaned pigs according to the recommendation by NRC (2012) (Table 1), aiming to meet or exceed the nutritional requirements of weaned piglets. The composition and nutrient contents are shown in Table 1. On day 28 of the experiment, the piglets were randomly divided into four groups (n = 6) using a 2 × 2 two-factor design (diet treatment: Ctrl vs. LPC; immunological challenge: saline vs. LPS), namely: (1) control group (Ctrl); (2) LPC group (0.01% LPC) [15,16]; (3) LPS group; (4) LPC + LPS group (0.01% LPC + LPS). Subsequently, piglets in the corresponding groups were intraperitoneally injected with 100 μg/kg LPS (E. coli serotype O55:B5; Sigma Chemical Inc., St. Louis, MO, USA) or an equal volume of normal saline. According to our previous study, 24 piglets were sacrificed 4 h after injection of LPS or saline, by which time LPS had induced liver injury and an inflammatory response [17].

2.3. Blood and Liver Sample Collection

Prior to euthanasia of piglets, blood samples were collected from piglets using 10 mL vacuum tubes and centrifuged at 3000× g for 10 min at room temperature to obtain serum. The serum was aliquoted and stored at −20 °C until further analysis. Following blood collection, piglets were humanely euthanized via intramuscular injection of pentobarbital sodium (80 mg/kg body weight). Liver samples were then collected. A portion of liver tissue (0.5 cm^3^ segments) was fixed in fresh 4% formaldehyde/phosphate-buffered saline for 24 h for histological analysis. The remaining liver tissue was cut into smaller pieces, rapidly frozen in liquid nitrogen, and stored at −80 °C for further analysis.

2.4. Morphology Analysis

Following 24 h of fixation, liver samples were dehydrated in a graded ethanol series, cleared in xylene, embedded in paraffin wax, and sectioned into 5 µm-thick slices using a microtome (Vazyme Biotech Co., Ltd. Nanjing, China).The sections were subsequently stained with hematoxylin and eosin (H&E) for histological examination, and the histological images were acquired using a high-resolution digital pathological scanner (Leica Aperio AT2. Leica Microsystems GmbH, Am Leitz-Park 5, 35578 Wetzlar, Germany) with a magnification of 400×. Original images were shown in Supplementary Figure S1.

2.5. Biochemical Measurements of Serum Samples

Serum levels of total protein (TP), albumin (ALB), alanine aminotransferase (ALT), aspartate aminotransferase (AST), alkaline phosphatase (ALP), and gamma-glutamyltransferase (GGT) were measured colorimetrically using commercial assay kits (Nanjing Jiancheng Bioengineering Institute, Nanjing, China) in accordance with the manufacturer’s protocols.

2.6. mRNA Abundance Analysis

Total RNA was extracted from liver samples using Trizol reagent (Vazyme Biotech Co., Ltd., Nanjing, China). Complementary DNA (cDNA) was synthesized from total RNA using the PrimeScript^®^ RT reagent kit (Vazyme Biotech Co., Ltd. Nanjing, China), and quantitative real-time PCR was performed with SYBR^®^ Premix Ex Taq™ (Tli RNaseH Plus) (Vazyme Biotech Co., Ltd., Nanjing, China). The relative mRNA expression levels of target genes were normalized to the reference gene β-actin and calculated using the 2-ΔΔCt method, with results expressed relative to the saline-treated SO group. Primer sequences are listed in Supplementary Table S1.

2.7. Protein Expression Analysis

Total protein was extracted from liver tissue using a commercial total protein extraction kit (Jiangsu KeyGEN BioTECH Corp., Ltd., Nanjing, China). Quantified protein samples were separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), transferred onto PVDF membranes, and subsequently incubated with primary and secondary antibodies (see Supplementary Table S2 for antibody details). The blots were developed using Clarity Max Western ECL Substrate (Bio-Rad Laboratories, Hercules, CA, USA), and protein band intensities were quantified with the Image Lab Software version 6.1 (Bio-Rad Laboratories, Hercules, CA, USA). Original images were shown in Supplementary Figure S2.

2.8. Statistical Analysis

The experimental data were analyzed using a 2 × 2 factorial design via the General Linear Model (GLM) procedure in IBM SPSS Statistics (version 20.0.0; IBM Corp., Armonk, NY, USA). The model accounted for the main effects of dietary treatment (0 vs. 0.01% LPC), immunological challenge (saline vs. LPS), and their interaction. In cases of a significant (p ≤ 0.05) or trending interaction (p < 0.10), post hoc comparisons were performed using Duncan’s multiple range test. Data are presented as mean ± SEM. Differences were considered statistically significant at p ≤ 0.05, and p < 0.10 was considered indicative of a statistical trend.

3. Results

3.1. Liver Morphology

The administration of LPS induced significant liver damage, as evidenced by (a) hepatic sinusoidal congestion and (b) inflammatory cell infiltration. In contrast, treatment with LPC markedly attenuated these histopathological alterations (Figure 1).

3.2. Serum Biochemical Parameters

Serum biochemical parameters, including TB, ALB, AST, ALT, ALP, and GGT, are summarized in Table 2. LPS challenge significantly decreased the concentrations of TB and ALB (p < 0.05), while increasing the activities of AST, ALT, ALP, and GGT (p < 0.05). Significant interactive effects between LPC supplementation and LPS challenge were observed for AST, ALT, and ALP activities (p < 0.05). Specifically, dietary LPC supplementation significantly attenuated the LPS-induced elevations in AST, ALT, and ALP activities (p < 0.05), but had no significant effect on these parameters in saline-treated piglets.

3.3. Macrophage Polarization in the Liver

The mRNA expression levels of genes associated with macrophage polarization are shown in Table 3. Compared with the saline-treated group, LPS challenge significantly up-regulated the hepatic mRNA expression of pro-inflammatory markers (IL-6, IL-1β, TNF-α, IFN-γ, iNOS, and CD80) as well as anti-inflammatory markers (IL-10, TGF-β, and ARG1) (p < 0.05). Significant interactions between LPC supplementation and LPS challenge were observed for the mRNA expression of IL-6, IL-1β, TNF-α, IFN-γ, iNOS, and CD80 (p < 0.05). Dietary LPC supplementation significantly reduced the expression of these pro-inflammatory genes in LPS-challenged piglets, while no significant effects were observed in saline-injected piglets.

3.4. Glycolysis and TCA Cycle Status in the Liver

The mRNA expression levels of genes related to glycolysis and the TCA cycle are summarized in Table 4. Compared with the saline-treated group, LPS challenge significantly increased the hepatic mRNA expression of HK2 (p < 0.05), but decreased the expression of PK, PDH, CS, IDH, and DLST (p < 0.05). Dietary LPC supplementation tended to increase the mRNA abundance of IDH (p < 0.10). A significant interaction between LPC supplementation and LPS challenge was observed for HK2 expression (p < 0.05). LPC supplementation significantly down-regulated HK2 expression in LPS-challenged piglets, whereas no significant effect was found in saline-injected piglets.

3.5. mTOR Signaling Pathway

The mRNA and protein abundances of genes related to the mTOR signaling pathway, including mTOR, S6K1, 4EBP1, HIF-1α, LKB1, and SIRT1, are shown in Table 5 and Figure 2. Compared to piglets treated with saline, pigs challenged with LPS had lower mRNA abundance of mTOR, S6K1, 4EBP1, and LKB1 (p < 0.05), and tended to have lower mRNA abundance of HIF-1α and SIRT1 (0.05 < p < 0.10) in the liver. There was an interaction between LPC and LPS challenge for mRNA abundance of S6K1, HIF-1α, and SIRT1 (p < 0.05). Feeding with LPC upregulated mRNA abundance of S6K1, HIF-1α, and SIRT1 among LPS-injected piglets; however, there was no significant effect among saline-injected piglets.

Compared to piglets treated with saline, pigs challenged with LPS had lower protein abundance of mTOR and p-mTOR (p < 0.05) in the liver. There was an interaction between LPC and LPS challenge for protein abundance of mTOR and p-mTOR (p < 0.05). Feeding with LPC upregulated protein abundance of mTOR and p-mTOR among LPS-injected piglets; however, there was no significant effect among saline-injected piglets.

4. Discussion

The liver serves as a central organ for detoxification and is highly vulnerable to pathogenic infections and immune stress [18]. Within the liver, LPS binds to Kupffer cells—the resident macrophages—and activates them, leading to the excessive release of pro-inflammatory cytokines such as TNF-α, IL-1β, and IL-6 [19]. LPC, a bioactive lipid, regulates macrophage polarization during immune responses, inflammation, and tissue remodeling [20]. Moreover, LPC modulates intestinal immune responses, helping to maintain immune homeostasis and alleviate post-weaning intestinal inflammation [19].

In this study, we investigated the protective effects of dietary LPC supplementation against LPS-induced liver injury in weaned piglets. Histopathological analysis revealed that LPS challenge caused significant hepatic damage, characterized by nuclear dissolution, inflammatory cell infiltration, and nuclear pyknosis. Dietary LPC supplementation markedly alleviated these LPS-induced histopathological alterations. He et al. reported that High-energy diets significantly elevated liver ether extract content and pathological injury score relative to low-energy diets, whereas LPC supplementation exerted a significant ameliorative effect on the liver pathological injury score [21]. Consistent with our findings, Murch et al. also reported that LPS-induced structural damage to liver tissue could be mitigated by LPC supplementation [13]. Together, these results indicate that LPS challenge induces structural liver injury, which can be effectively attenuated by LPC.

Plasma biochemical components are primarily derived from nutrients absorbed through the gastrointestinal tract, rendering them reliable indicators of an animal’s health status and metabolic profile [17]. ALB and TB indirectly reflect immune competence [22]. In our study, LPS challenge decreased serum TP and ALB concentrations in piglets. AST, ALT, ALP, and GGT serve as important biomarkers for hepatic health. These enzymes are released into circulation upon hepatocyte damage or inflammation, leading to significantly elevated serum concentrations, thus providing diagnostic value for liver dysfunction [23]. Similarly to our data, Oliver et al. reported that LPS challenge significantly increased ALT and AST content in rats [13]. Here we found that LPC significantly alleviated LPS-induced reduction in ALT, AST and AKP, which demonstrated that dietary addition of LPC alleviated LPS-induced hepatic dysfunction. Liver injury is often accompanied by oxidative stress. Studies have shown that LPC can alleviate oxidative stress in the liver of weaned piglets and thus protect liver health [15]. Consistent with our findings, Cai et al. [24] reported that LPC reduced serum AST in 21-day-old broilers and ALT in 42-day-old broilers. Consistent with previous findings, our results showed that LPS increased serum AST, ALT, ALP, and GGT activities, while LPC supplementation mitigated the LPS-induced elevations in AST, ALT, and ALP, indicating hepatoprotective effects of LPC.

Macrophages represent the predominant immune cell population in the liver [25]. Classically activated M1 macrophages primarily secrete pro-inflammatory cytokines—such as IL-1β, IL-6, TNF-α, and iNOS—thereby driving inflammatory responses [26]. In contrast, alternatively activated M2 macrophages produce anti-inflammatory mediators, including IL-10, TGF-β, and Arg-1, which contribute to inflammation resolution [27]. Under physiological conditions, a dynamic balance between M1 and M2 phenotypes is maintained; however, pathological stimuli can disrupt this equilibrium, promoting inflammatory liver diseases [28]. For instance, Orecchioni et al. [29] reported that LPS challenge in mice significantly up-regulated M1 marker expression (TNF-α, IL-6, and iNOS) while down-regulating M2 markers. In the present study, LPS increased the mRNA expression of both M1 and M2 polarization markers in piglet livers, while dietary LPC supplementation attenuated the up-regulation of M1-associated genes. The concurrent elevation in M2 markers observed here may reflect a compensatory anti-inflammatory response during acute hepatic injury. Supporting our findings, Zhao et al. demonstrated that LPC suppressed M1 polarization in LPS-challenged macrophages via modulation of the SIRT1/NF-κB and p38 MAPK pathways [20]. Additionally, Assunção et al. reported that LPC promoted PPARγ-dependent M2 macrophage polarization, characterized by up-regulated expression of Arg-1 and IL-10 [30]. Collectively, these results indicate that LPC alleviates liver injury by modulating macrophage polarization toward an anti-inflammatory phenotype, primarily through suppressing M1-related pro-inflammatory cytokine expression.

Accumulating evidence highlights the pivotal roles of glycolysis and the TCA cycle in macrophage polarization. When macrophages are polarized to the pro-inflammatory M1 phenotype by stimuli such as LPS and IFN-γ, their metabolism shifts from oxidative phosphorylation to aerobic glycolysis. This metabolic switch is mainly regulated by the mTOR pathway. Aerobic glycolysis provides rapid energy supply and biosynthetic intermediates for M1 macrophages to support the production of pro-inflammatory cytokines like TNF-α and iNOS. For example, Ran et al. [31] demonstrated that inhibiting HK2 suppresses LPS-induced M1 polarization. Similarly, Zhai et al. [32] reported that a soluble death receptor 5-Fc fusion protein dose-dependently lowered HK2 mRNA levels in M1 macrophages, thereby blocking glycolysis-dependent polarization and reducing IL-1β and TNF-α secretion. Consistent with these findings, our study revealed that LPS up-regulated HK2 expression while down-regulating other key enzymes in glycolysis and the TCA cycle. Dietary LPC supplementation counteracted the LPS-induced elevation in HK2 mRNA. Furthermore, LPC has been shown to promote immune-cell differentiation via the glycolytic pathway in contexts such as psoriasis and atherosclerosis [33,34]. Research on psoriasis has reported that LPC can activate STAT1, thereby recognizing and binding to the promoters of glycolytic enzymes GCK and PKLR. Additionally, the LPC/G2A axis directly promotes Th1 differentiation, which is dependent on LPC-induced glycolytic activity, rather than the mTOR–glycolysis axis. In atherosclerosis models, LPC was found to promote the M2 polarization of macrophages through the AMPK pathway, which is different from the regulatory effect of LPC on macrophage polarization in our liver injury model. Several in vitro studies illustrated that LPC stabilized M1 polarization and proinflammatory cytokine expression in human macrophages via the G protein coupled receptor (GPCR) G2A [35]. Conversely, schistosomal-derived LPCs facilitated M2 polarization of murine macrophages through a peroxisome proliferator-activated receptor- γ (PPARγ)-dependent pathway [30]. Thus, how LPC affects macrophage polarization may depend on the cellular environment and distinct properties of LPC including concentration, acyl chain length, and saturation and whether it is bound to proteins or not. Together, these results suggest that LPC alleviates hepatic inflammation by modulating glucose metabolism—specifically through suppressing HK2-mediated glycolysis—which subsequently reduces M1 polarization and pro-inflammatory cytokine production.

To further elucidate the effect of LPC on the LPS-induced imbalance between hepatic M1 and M2 polarization, we examined its role in regulating the mTOR signaling pathway. Both the mTOR and LKB1 signaling pathways are critically involved in regulating macrophage metabolism, including glycolysis and the TCA cycle [36]. In our study, LPS challenge significantly downregulated the hepatic mRNA expression of mTOR, S6K1, 4EBP1, and LKB1. Dietary LPC supplementation attenuated the LPS-induced suppression of S6K1, HIF-1α, and SIRT1 expression. Consistent with the mRNA findings, LPS also reduced both total and phosphorylated mTOR protein levels, effects that were ameliorated by LPC. Vergadi et al. [37] demonstrated that mTOR signaling not only influences glycolytic activity but also directs macrophage polarization; specifically, mTOR inhibition promotes an M1 pro-inflammatory phenotype and enhances cytokine secretion. Our results align with these observations, suggesting that LPC helps sustain mTOR activation, thereby restraining M1 polarization and mitigating inflammatory liver injury. Collins et al. [38] reported that macrophage-specific mTORC1 signaling deletion in C57BL/6 mice led to a marked enhancement of M1 macrophage function, as validated by complementary in vitro and in vivo assays. In our current study, it is possible that the protective effects of LPC on liver integrity were closely associated with regulating glycolysis (particularly HK2) through the mTOR signaling pathway.

This is the first study to report that dietary LPC supplementation can attenuate LPS-induced liver injury in weaned piglets by inhibiting M1 macrophage polarization. We also identified a new mechanism by which LPC inhibits M1 macrophage polarization through the mTOR-mediated glycolysis pathway, which provides a new theoretical basis for the immunomodulatory function of LPC.

In our study, dynamic variations in the expression of hepatic pro-inflammatory mediators, metabolites, and signaling molecules likely occur over time. Earlier time points (1–2 h) may not fully reflect the polarization of macrophages, while later time points (8–12 h) may involve the compensatory repair mechanism of the liver, which could mask the regulatory effect of LPC on macrophage polarization. Consequently, measurements taken at a single time point (4 h) may not be sufficient to fully elucidate the roles of these mediators and molecules in LPS-induced liver injury. Therefore, sampling across multiple time points would be necessary to analyze their dynamic changes at both mRNA and protein levels, a direction that warrants further experimental investigation.

5. Conclusions

Dietary supplementation with LPC attenuates LPS-induced hepatic injury in piglets, likely through the inhibiting of M1 polarization mediated by glycolysis-related mTOR signaling pathways.