Skip to main content
NCBI home page
Current Research in Toxicology logoLink to Current Research in Toxicology
. 2025 Nov 5;9:100269. doi: 10.1016/j.crtox.2025.100269

Novel dose- and time-dependent toxicity biomarkers of fentanyl/carfentanyl: uncovered by urine-plasma metabolomics for forensics

Wanting Xie a,b,1, Shuo Yang a,1, Xin Wang a, Jinting Liu a, Wen Gao b, Yan Shi a,⁎
PMCID: PMC12648698  PMID: 41311705

Graphical abstract

graphic file with name ga1.jpg

Open in a new tab

Keywords: Fentanyl, Carfentanyl, Metabolomics, Biomarker, UHPLC-HRMS

Highlights

  • •

    Used UHPLC-HRMS metabolomics to analyze plasma/urine from fentanyl/carfentanil-treated rats, tracking temporal metabolic changes.

  • •

    To clarify fentanyl analog toxic mechanisms amid rising intoxications.

  • •

    Identified oxidative stress, energy dysregulation, and taurine as a potential fentanyl-like abuse biomarker.

Abstract

Fentanyl analogs present a significant global risk to public health and safety due to their high abuse potential and related mortality. However, the absence of a structured research framework focused on their clinical pharmacology and toxicology has led to a scarcity of studies on the mechanisms underlying the toxicity of and addiction to these substances, particularly regarding the rapidly emerging new fentanyl analogs. In this study, we employed a nontargeted metabolomics strategy. By conducting multivariate statistical analysis on ultrahigh performance liquid chromatography–high resolution mass spectrometry (UHPLC-HRMS) data from rat urine and plasma to elucidate the metabolic disruptions induced by fentanyl and carfentanyl. Adult male SD rats were randomly assigned to four groups: low-dose fentanyl group, high-dose fentanyl group, low-dose carfentanyl group, and high-dose carfentanyl group. All groups underwent continuous tail vein injection of toxicants for 5 consecutive days. Plasma and urine samples were collected from the rats before the first administration and at different time points after the last administration, followed by detection and analysis. Specifically, we aimed to elucidate the dose- and time-dependent metabolic toxicity of fentanyl and carfentanyl through a nontargeted metabolomics strategy. Oxidative stress, immunosuppression, and energy dysregulation were identified as core toxic effects of fentanyl and carfentanyl, tightly linked to perturbations in taurine and glutathione pathways, taurine regulating immune function and neuronal homeostasis, glutathione maintaining redox homeostasis. Succinic acid and taurine were confirmed as common biomarkers in urine and plasma. Succinic acid showed decreased plasma levels and increased urine levels, directly indicating fentanyl and carfentanyl-induced energy metabolism impairment. Taurine exhibited similar dysregulation, reflecting drug-induced immunosuppression and neuronal excitability abnormalities. These findings provide critical references and experimental support for studies on fentanyl-related hazards, toxicological mechanisms, and forensic detection.

1. Introduction

Among the new synthetic opioids that continue to emerge in illicit markets, fentanyl analogs pose a particularly serious threat to public health and safety due to their serious global abuse and high mortality rate (Shafi et al., 2022, Stanley, 1992, Jeal and Benfield, 1997, Benyamin et al., 2008, Burns et al., 2018). Fentanyl and its main analogs, such as fentanyl, sufentanyl, and remifentanyl, are used medically as an adjunct to anesthesia, sedation, and anesthesia in surgery, whereas others, such as carfentanyl and orfentanyl, are often spiked into other illicit drugs, such as heroin, or used as heroin substitutes because of their high potency (Coopman et al., 2016, Suzuki and El-Haddad, 2017, Armenian et al., 2018, Frank and Pollack, 2017, Dayer et al., 2019). On April 1, 2019, China announced the listing of a whole class of fentanyl-like substances as controlled substances. This listing of fentanyl analogs introduced institutional safeguards to deal with the complex and volatile drug market, but it also created higher requirements and challenges for the actual detection and supervision of opioid-type drugs. Unfortunately, due to a lack of systematic clinical pharmacology and toxicology research, little information is available regarding the mechanisms underlying the toxicity of fentanyl substances or the development of drug dependence on them.

Drug addiction is now being hypothesized to be a chronic form of metabolic disease that is caused by a combination of genetic risk factors and substance abuse (Courtwright, 1997). This theory now provides important direction for studies on the mechanisms of drug addiction, relapse, and withdrawal by pointing to explorations of the metabolic processes in the body. In view of the increasingly serious drug abuse issues and the need for research progress, one promising new research strategy is metabolomics, a relatively new field of research in “omics” technology. The aim of metabolomics is to use qualitative and quantitative analyses of endogenous compounds in living organisms to identify changes associated with certain stimuli of interest, such as drug intake (Sindelar and Patti, 2020, Zhang et al., 2020). For drug abuse studies, nontargeted metabolomics analysis is particularly useful, as it describes the metabolism of a drug in various organisms or tissues through the identification of all endogenous metabolites. The nontargeted approach, which is highly sensitivity nonbiased, is therefore particularly suitable for the panoramic analysis of metabolites, and has been widely used to reveal the dynamics of metabolite changes under different conditions (Steuer et al., 2022).

One commonly used nontargeted metabolomics tool is liquid chromatography-mass spectrometry (LC-MS) because this method can simultaneously detect multiple compounds with high sensitivity and good reproducibility (Szeremeta et al., 2021). Not surprisingly, metabolomics has become a valuable tool for research on fentanyl-like compounds in different areas, such as studies on toxicity mechanisms (Li et al., 2022, Dhummakupt et al., 2019), drug metabolites (Goggin et al., 2017, Montesano et al., 2021), and forensic intake (Amante et al., 2021). Among the various biological substrates used in metabolomics, urine remains one of the preferred substrates due to the ease and noninvasive nature of collection and to the presence of an abundance of metabolites (Fu et al., 2014).

In the present work, we used a nontargeted LC-MS metabolomics approach, combined with multivariate statistical analyses to investigate metabolic disruption of urine and plasma by fentanyl and carfentanyl in rats. To improve the accuracy of the results, we also tested plasma samples for cross-validation. Although serum and plasma have similar metabolic characteristics, many differences still occur between the two, and a good plasma collection under strict standard operating procedures can provide a relatively accurate snapshot of the metabolic state of the blood (Liu et al., 2018). This study aims to address two critical gaps: (1) the lack + of systematic data on metabolic disruptions caused by fentanyl and carfentanyl at clinically relevant doses, and (2) the identification of cross-validated biomarkers in both urine and plasma. To this end, it reveals that the two substances induce dose- and time-dependent toxicity associated with oxidative stress and energy metabolism dysregulation, with key underlying metabolic networks including the taurine pathway and succinic acid-linked energy pathways, and succinic acid and taurine emerging as potential cross-validated biomarkers. These findings thus fill the aforementioned gaps and provide a basis for further exploring the toxicological mechanisms of fentanyl-related substances and developing forensic detection strategies.

2. Material and methods

2.1. Materials and reagents

Fentanyl (CAS: 437-38-7) and carfentanyl (CAS: 59708-52-0) reference standards were purchased from the Shanghai Yuansi Standard Science & Technology Co., Ltd. (Shanghai, China). All standards were certified reference materials with purity greater than 99 % based on their HPLC profiles and UV and MS data. LC-MS grade methanol and acetonitrile were purchased from Honeywell International (New Jersey, USA). LC-MS-grade formic acid was obtained from Sigma-Aldrich. Inc. (New Jersey, USA). All water used in this study was ultrapure water prepared using a Millipore AFS-10 water purification system (Billerica, MA, USA).

2.3. Sample collection

Approximately 500 µL of blood was collected from the orbital venous plexus into anticoagulant tubes containing EDTA at 0 h before the initial drug administration (control group) and again at 6 h, 1 d, 3 d, and 15 d after the last administration. Blood samples were collected from the orbital venous plexus using heparinized capillary tubes. Rats were restrained with gentle cervical compression to induce orbital congestion, and the capillary was inserted 4–5 mm into the medial canthus at a 45° angle. Approximately 0.5 mL of blood was collected per session. After bleeding, pressure was applied to the puncture site for 30 s. The blood was centrifuged at 4 °C and 10,000 rpm for 5 min, the upper plasma sample was separated. Rats were placed into metabolic cages for 12 h for urine collection. Urine samples were centrifuged at 13,000 rpm for 3 min and the supernatant was taken for analysis. The urine collection times were before the initial administration (control group, labeled as 0 h) and after the blood samples were collected at 6 h, 1 d, 3 d, and 15 d after the last drug administration. The plasma and urine samples were dispensed into tubes, snap-frozen in liquid nitrogen for 15 min, and stored at −80 °C prior to metabolomics analysis.

2.4. Sample preparation and analysis

The collected plasma and urine samples were thawed on ice, and the metabolites were extracted from 100 µL of each sample using 400 µL of pre-cooled methanol. The extraction mixture was then vortexed, mixed, and placed in a refrigerator at −20 °C for 30 min to precipitate the proteins in the samples. After centrifugation at 20,000 g for 15 min, the supernatants were transferred to new tubes and vacuum dried. The dried samples were redissolved in 100 μL 50 % methanol, which was chosen to maximize metabolome coverage of both polar and nonpolar metabolites (Lindahl et al., 2017, Manier and Meyer, 2020) and experimentally verified to result in symmetric peaks and stable retention times, centrifuged at 20,000 g for 15 min, and the supernatant was taken for ultrahigh performance liquid chromatography–high resolution mass spectrometry (UHPLC-HRMS) analysis. Pooled quality control (QC) samples were also prepared by combining 10 μL of each extraction mixture.

2.5. Chromatography and spectrometry

The UHPLC-HRMS analyses were carried out utilizing a Vanquish Flex UHPLC system (Thermo Fisher Scientific, Bremen, Germany) coupled to an Orbitrap Exploris 240 mass spectrometer (Thermo Fisher Scientific, Bremen, Germany). Chromatography was performed using an ACQUITY UPLC T3 column (100 × 2.1 mm, 1.8 μm, Waters, UK) with linear gradient elution at a flow rate of 0.3 mL/min using 5 mmol/L ammonium acetate and 5 mmol/L acetic acid in water (solvent A) and acetonitrile (solvent B). The gradient elution conditions were as follows: 2 % solvent B for 0–0.8 min, 2–70 % solvent B for 0.8–2.8 min, 70–99 % solvent B for 2.8–5.3 min, 99 % solvent B for 5.3–7.0 min, 99–2 % solvent B for 7.0–7.2 min, and 2 % solvent B for 7.2–10.0 min. The column temperature was maintained at 40 °C.

The Orbitrap Exploris 240 mass spectrometer was operated in both positive and negative ion modes with a source temperature of 350 °C. The ion source gas 1 and gas 2 pressures were set at 50 PSI. For the positive mode, the ion spray floating voltage was set at 5,000 V, and for the negative-ion mode, the ion spray floating voltage was set at −4,500 V. The MS data were acquired in the data-dependent acquisition (DDA) mode. The mass scan range was 60–1,200 Da, and the top 5 signal ions with signal accumulation intensities above 5,000 were selected for secondary fragmentation scanning. Additionally, to assess the stability of the LC-MS system throughout the entire acquisition period, a QC sample, prepared by pooling all the samples, was analyzed after every 10 experimental samples.

2.6. Data processing and statistical analysis

The UHPLC-HRMS raw data files were converted into mzXML format and then processed using the XCMS (Smith et al., 2006), CAMERA (Kuhl et al., 2012), and metaX (Wen et al., 2017) toolbox implemented with the R software (v.4.1.2). The acquired MS data pretreatments, including peak picking, peak grouping, retention time correction, second peak grouping, and annotation of isotopes and adducts, were performed using XCMS software. Each ion was identified by combining retention time and m/z data. The Kyoto Encyclopedia of Genes and Genomes (KEGG, http://www.kegg.com/) and the Human Metabolome Database (HMDB, http://www.hmdb.ca) were used to annotate the metabolites by matching the exact molecular mass data of the samples to those from the database to identify the physicochemical properties and biological functions of the metabolites.

The potential metabolites were screened based on their variable importance in the projection (VIP) values and Student’s t-test. Values of VIP > 1 and p < 0.05 were considered statistically significant. Furthermore, to reduce false positives inherent in multiple comparisons, the false discovery rate (FDR) was controlled using the Benjamini-Hochberg procedure, with an FDR threshold of <0.05 applied to finalize the selection of differential metabolites. For these finalized differential metabolites, to further characterize their trends across groups, the normalized data were visualized as heatmaps. Additionally, enrichment analysis of their associated metabolic pathways was determined using the KEGG database.

3. Results

3.1. Urine metabolomics analysis

3.1.1. Metabolic alterations in urine

Urine nontargeted metabolomics data were collected by UHPLC-HRMS in both positive and negative modes. The total ion chromatogram (TIC) can reflect the separation of all metabolites in the liquid chromatography, and control the overall mass spectrometry signal intensity of the samples. As shown in Fig. 1A and C, the response intensities and retention times of the peaks essentially overlapped, indicating that the instrument was stable throughout the experiment and the variation caused by errors was small. We then evaluated the distribution of the detected metabolite ions, as shown in Fig. 1B and D. In the graphs, retention time is the horizontal coordinate and m/z is the vertical coordinate, each point represents a substance, and the color indicates the density of the substance in the region (a darker color corresponds to the a higher feature number). Additionally, to oversee the efficacy of the analytical methodologies employed within this research and to segregate high-caliber data, QC samples were intermittently infused throughout the data acquisition phase.

Fig. 1.

Fig. 1

Open in a new tab

Overlapping spectra of total ion flow maps (negative ion mode: A; positive ion mode: C) and metabolite m/z-rt distributions (negative ion mode: B; positive ion mode: D) of urine samples.

Examination of the PCA score plots (depicted in Fig. 2) confirmed that the QC samples exhibited cohesive clustering under each analytical technique, signifying that the data were both robust and reproducible and warranted inclusion in subsequent statistical evaluations. The partial least-squares discriminant analysis (PLS-DA) score plots for the global metabolomics data at 6 h, 1 d, 3 d, and 15 d post-administration of fentanyl and carfentanyl demonstrated that the urinary metabolite profiles of the rats in the treatment groups were markedly distinguished from those of the control group under each set of experimental conditions. However, no significant segregation was observed that was dependent on the concentration of the administered substances (as depicted in Fig. 3). Furthermore, the permutation tests used to validate the OPLS-DA model did not show overfitting, as the R2 and Q2 values for were lower for all permutation classes than for the original class, indicating a high robustness and predictive power of the model. We further identified the different metabolic phenotypes by plotting heat maps (Fig. 4). With the exception of the 3 d carfentanyl low-dose group, the clustered bars on the heat maps for each group showed a clear separation between the experimental and control groups, suggesting significant differences in metabolites between the groups. These results suggest that fentanyl and carfentanyl are capable of altering urinary metabolism.

Fig. 2.

Fig. 2

Open in a new tab

PCA scores plots of urine (Shadow: QC Gathering).

Fig. 3.

Fig. 3

Open in a new tab

PCA scores plots of urine.

Fig. 4.

Fig. 4

Open in a new tab

Heatmap showing changes in urinary metabolism after exposure to different concentrations of fentanyl and carfentanyl.

3.1.2. Investigating the significant metabolites and pathways

The results presented in section 3.1.1 were used to select potential variables as biomarkers based on VIP >1, p < 0.05 and fold change >2. Statistically significant changes were found in 96 metabolites in the ULF group, 96 in the UHF group, 98 in the ULC group, and 74 in the UHC group when each group was compared with the control group. Table S1 shows the results. Ingestion of fentanyl or carfentanyl interferes with several metabolic pathways in urine, including lipid and free fatty acid metabolism and amino acid conversion. It is worth noting that in ULF and UHF groups, the metabolite M-toluic acid increased at the 6 h, 1 d, 3 d and 15 d time points. In ULC group, the metabolites M-toluic acid and cis-muconic acid increased, and the metabolite psoralen showed a decreasing trend at those same time points. In the UHC group, the metabolites benzoic acid and M-toluic acid were increased at the 6 h, 1 d, 3 d and 15 d time points, whereas the metabolites adenine and oxypurinol showed a downward trend at those time points, shown in Fig. 5. The metabolic pathways of pyrimidine metabolism and cysteine and methionine metabolism were disturbed in both low-dose and high-dose fentanyl groups. Taurine and hypotaurine metabolism and vitamin B6 metabolism were disturbed by both low and high doses of carfentanyl.

Fig. 5.

Fig. 5

Open in a new tab

Peak area of urine putative biomarkers.

The identified metabolites were subjected to pathway analysis using MetaboAnalyst 4.0 software. KEGG enrichment analysis of urine differential metabolites are shown in Fig. 6. Typical KEGG pathways in urine samples are shown in Table 1. Among the various pathways, pyrimidine metabolism, glutathione metabolism, cysteine and methionine metabolism, and tyrosine metabolism could be distinguished between the ULF rats and untreated control rats. Pyrimidine metabolism, taurine and hypotaurine metabolism, cysteine and methionine metabolism, vitamin B6 metabolism, and pentose and glucuronate interconversions could be distinguished between UHF rats and untreated control rats. Tyrosine metabolism, phenylalanine, tyrosine and tryptophan biosynthesis, taurine and hypotaurine metabolism and vitamin B6 metabolism could be distinguished between ULC rats and untreated controls. Glyoxylate and dicarboxylate metabolism, cysteine and methionine metabolism, taurine and hypotaurine metabolism, and vitamin B6 metabolism could be distinguished between rats in the UHC group and untreated control rats.

Fig. 6.

Fig. 6

Open in a new tab

KEGG enrichment analysis of urine differential metabolites (A: ULF; B: UHF; C: ULC; D: UHC).

Table 1.

Typical KEGG pathways in urine samples.

Group Name Pathway Name Match Statusa Pb −log(p)c Holm pd FDRe Impactf
ULF Pyrimidine metabolism 3/39 0.07 1.19 1.00 0.36 0.11
Glutathione metabolism 2/28 0.14 0.84 1.00 0.64 0.11
Cysteine and methionine metabolism 2/33 0.19 0.73 1.00 0.75 0.16
Tyrosine metabolism 2/42 0.27 0.57 1.00 0.94 0.14
UHF Pyrimidine metabolism 3/39 0.07 0.19 1.00 0.52 0.11
Taurine and hypotaurine metabolism 1/8 0.18 0.75 1.00 0.95 0.43
Cysteine and methionine metabolism 2/33 0.19 0.73 1.00 0.95 0.16
Vitamin B6 metabolism 1/9 0.20 0.70 1.00 0.95 0.49
Pentose and glucuronate interconversions 1/19 0.37 0.43 1.00 1.00 0.17
ULC Tyrosine metabolism 3/42 0.05 1.30 1.00 0.37 0.14
Phenylalanine, tyrosine and tryptophan biosynthesis 1/4 0.08 1.10 1.00 0.52 0.50
Taurine and hypotaurine metabolism 1/8 0.15 0.82 1.00 0.85 0.43
Vitamin B6 metabolism 1/9 0.17 0.77 1.00 0.85 0.49
UHC Glyoxylate and dicarboxylate metabolism 2/32 0.10 1.00 1.00 0.79 0.11
Cysteine and methionine metabolism 2/33 0.11 0.97 1.00 0.79 0.16
Taurine and hypotaurine metabolism 1/8 0.13 0.89 1.00 0.79 0.43
Vitamin B6 metabolism 1/9 0.14 0.84 1.00 0.81 0.49
Open in a new tab
a

Match Status: Number of differential metabolites matched to the pathway.

b

P: Uncorrected p-value from pathway enrichment analysis.

c

-log(p): Negative logarithm (base 10) of p (larger values = stronger significance).

d

Holm p: p-value corrected via Holm’s multiple comparison method.

e

FDR: False Discovery Rate.

f

Impact: Pathway impact value.

3.2. Plasma metabolomics analysis

3.2.1. Metabolic alterations in plasma

Plasma nontargeted metabolomics data were also collected by UHPLC-HRMS in both positive and negative modes(as depicted in Fig. 7). The reliability of the analytical techniques utilized in this study and the superior quality data were confirmed by periodically introducing QC samples during the data collection process. The PCA score plots, as illustrated in Fig. S1B, indicated that the QC samples formed a tight cluster under each analytical method, verifying that the data were both robust and reproducible and justifying their use in subsequent statistical analyses (as depicted in Fig. 8). As shown in Fig. 9, the PLS-DA scores for plasma nontargeted metabolomics data obtained at 6 h, 1 d, 3 d and 15 d after fentanyl and carfentanyl administration showed significant differences in the plasma metabolites of rats in the administration group compared to the control group. The replacement test of PLSDA model revealed no overfitting, confirming that the model had high robustness and predictive ability. We further identified the different metabolic phenotypes by producing heat maps, as shown in Fig. 10. Except for the PLF_3d and PHC_6h groups, the cluster bars on the heat maps showed obvious separation between the administration group and the untreated control group, indicating significant differences in metabolites between the groups. These results suggested that fentanyl and carfentanyl can alter plasma metabolism in rats.

Fig. 7.

Fig. 7

Open in a new tab

Overlapping spectra of total ion flow maps (negative ion mode: A; positive ion mode: C) and metabolite m/z-rt distributions (negative ion mode: B; positive ion mode: D) of plasma samples.

Fig. 8.

Fig. 8

Open in a new tab

PCA scores plots of plasma (Shadow: QC Gathering).

Fig. 9.

Fig. 9

Open in a new tab

PLS-DA scores plots of plasma.

Fig. 10.

Fig. 10

Open in a new tab

Heatmap showing changes in plasma metabolism after exposure to different concentrations of fentanyl and carfentanyl.

3.2.2. Investigating the significant metabolites and pathways

Statistically significant changes were found in 91 metabolites in the PLF group, 93 in the PHF group, 79 in the PLC group, and 95 in the PHC group when each group was compared with the untreated control group. Table S2 shows the results. Notably, in the PLF group, the metabolite pelargonic acid increased at the 1 d, 3 d, and 15 d time points. In the PHF group, the metabolite lysophosphatidyl inositol 18:2 (lysoPI 18:2) showed an increasing trend at the 6 h, 1 d, 3 d and 15 d time points. In the PLC group, the metabolite 4-guanidinobutanoic acid increased at the 6 h, 1 d, 3 d and 15 d time points and the metabolite (S)-homostachydrine decreased at the 6 h, 1 d, 3 d and 15 d time points. In the PHC group, the metabolites citric acid and (S)-homostachydrine decreased at the 6 h, 1 d, 3 d and 15 d time points. Ingestion of fentanyl or carfentanyl interfered with several plasma metabolic pathways, including amino acid, lipid, and riboflavin metabolism, as shown in Fig. 11. The pathways of sulfur metabolism and ether lipid metabolism were disturbed in both the low-dose and high-dose fentanyl groups. Glycerolipid metabolism, ether lipid metabolism, histidine metabolism, and riboflavin metabolism were disrupted in the carfentanyl low dose and high dose groups.

Fig. 11.

Fig. 11

Open in a new tab

Peak area of plasma putative biomarkers.

The identified metabolites were also subjected to pathway analysis using MetaboAnalyst 4.0 software. KEGG enrichment analysis of the plasma differential metabolites are shown in Fig. 12, while the typical KEGG pathways in plasma samples are shown in Table 2. Among these pathways, riboflavin metabolism, sulfur metabolism, and ether lipid metabolism were able to distinguish between PLF rats and untreated control rats. Sphingolipid metabolism, sulfur metabolism, taurine and hypotaurine metabolism, ether lipid metabolism, and tryptophan metabolism were able to distinguish between rats in the PHF group and untreated control rats. Riboflavin metabolism, glycerolipid metabolism, histidine metabolism, and ether lipid metabolism could distinguish between PLC rats and untreated control rats. Citrate cycle (TCA cycle), riboflavin metabolism, ascorbate and aldarate metabolism, glycerolipid metabolism, histidine metabolism, and ether lipid metabolism were able to distinguish between rats in the PHC group and the untreated controls.

Fig. 12.

Fig. 12

Open in a new tab

KEGG enrichment analysis of plasma differential metabolites (A: PLF; B:PHF; C:PLC; D:PHC).

Table 2.

Typical KEGG pathways in plasma samples.

Group Name Pathway Name Match Statusa Pb −log(p)c Holm pd FDRe Impactf
PLF Riboflavin metabolism 1/4 0.05 1.27 1.00 1.00 0.50
Sulfur metabolism 1/8 0.10 0.98 1.00 1.00 0.32
Ether lipid metabolism 1/20 0.24 0.62 1.00 1.00 0.20
PHF Sphingolipid metabolism 2/32 0.17 0.76 1.00 1.00 0.22
Sulfur metabolism 1/8 0.17 0.76 1.00 1.00 0.32
Taurine and hypotaurine metabolism 1/8 0.17 0.76 1.00 1.00 0.43
Ether lipid metabolism 1/20 0.38 0.42 1.00 1.00 0.20
Tryptophan metabolism 1/41 0.63 0.20 1.00 1.00 0.14
PLC Riboflavin metabolism 1/4 0.64 1.20 1.00 1.00 0.50
Glycerolipid metabolism 1/16 0.23 0.63 1.00 1.00 0.10
Histidine metabolism 1/16 0.23 0.63 1.00 1.00 0.22
Ether lipid metabolism 1/20 0.28 0.55 1.00 1.00 0.20
PHC Citrate cycle (TCA cycle) 2/20 0.07 1.17 1.00 1.00 0.12
Riboflavin metabolism 1/4 0.08 1.08 1.00 1.00 0.50
Ascorbate and aldarate metabolism 1/10 0.20 0.71 1.00 1.00 0.41
Glycerolipid metabolism 1/16 0.30 0.53 1.00 1.00 0.10
Histidine metabolism 1/16 0.30 0.53 1.00 1.00 0.22
Ether lipid metabolism 1/20 0.35 0.45 1.00 1.00 0.20
Open in a new tab

aMatch Status: Number of differential metabolites matched to the pathway.

b

P: Uncorrected p-value from pathway enrichment analysis.

c

-log(p): Negative logarithm (base 10) of p (larger values = stronger significance).

d

Holm p: p-value corrected via Holm’s multiple comparison method.

e

FDR: False Discovery Rate.

f

Impact: Pathway impact value.

4. Discussion

This study examined the metabolomics inferred by the UHPLC-HRMS biochemical profiles of urine and plasma from rats subjected to fentanyl-induced and carfentanyl-induced toxicity. This approach enabled a holistic evaluation of changes occurring at the metabolite level, through which we first identified three core toxic effects of the two drugs: oxidative stress, immunosuppression, and energy metabolism dysregulation. Each of these effects was tightly linked to disruptions in key metabolic networks, including the taurine pathway that regulates immune function and neuronal homeostasis, the glutathione pathway that maintains cellular redox balance, and succinic acid-linked energy metabolism that acts as a central node of the tricarboxylic acid cycle. We also confirmed succinic acid and taurine as potential cross-matrix biomarkers detectable in both urine and plasma, and their dysregulation is marked by decreased plasma levels and increased urinary excretion. Based on these results, we conclude that fentanyl and carfentanyl exposure can lead to dose-dependent and time-dependent toxicity, and that the observed metabolic changes shed new light on the mechanism of fentanyl analog toxicity.

4.1. Discovery and profiling of potential biomarkers

The following eight metabolites were considered potential urinary biomarkers following administration of 1/100 LD50 fentanyl: 2-ketobutyric acid, L-tyrosine, L-methionine, glycine, orotic acid, cytidine, uracil, and 5,6-dihydroxyindole. The following eight metabolites were considered potential urinary biomarkers following administration of 1/40 LD50 fentanyl: 2-ketobutyric acid, uracil, orotic acid, taurine, xylitol, L-methionine, and cytidine. The following five metabolites were considered potential urinary biomarkers following administration of 1/100 LD50 carfentanyl: vanillylmandelic acid, pyridoxal, L-tyrosine, 5,6-dihydroxyindole, and taurine. The following six metabolites were considered potential urinary biomarkers following administration of 1/40 LD50 carfentanyl: L-methionine, 2-ketobutyric acid, citric acid, pyridoxal, glyceric acid, and taurine.

The following three metabolites primarily were potential plasma biomarkers following administration of 1/100 LD50 fentanyl: plasmenyl-PE 38:6, (−)-riboflavin, and sulfite. In addition, six metabolites primarily considered as potential plasma biomarkers caused by 1/40 LD50 fentanyl, including plasmenyl-PE 38:4, SM 34:1, Cer[NS] 34:1, sulfite, taurine and L-tryptophan. Moreover, four metabolites primarily considered as potential plasma biomarkers caused by 1/100 LD50 carfentanyl, including L-histidine, plasmenyl-PE 38:4, LysoPA 20:4 and (−)-riboflavin. Finally, seven metabolites primarily considered as potential plasma biomarkers caused by 1/40 LD50 carfentanyl, including citric acid, D-glucuronic acid, succinic acid, plasmenyl-PE 38:4, lysoPA 20:4, L-histidine, and (−)-riboflavin.

After comparative analysis of plasma and urine metabolomics results, it was found that the two metabolites, succinic acid and taurine, showed significant changes in plasma and urine samples of fentanyl- and carfentanyl-contaminated rats, which were hypothesized to be the common biomarkers of the two biological substrates. As a key metabolite in organisms, succinic acid occupies a central position in energy metabolism, dominating the tricarboxylic acid cycle and supplying energy for cellular respiration, supporting physiological activities such as muscle contraction and nerve conduction, and participating in the synthesis of biomolecules as a metabolic precursor to maintain the metabolic network homeostasis (Liu et al., 2022). The dynamic changes of succinic acid in plasma decrease and urine increase suggest that there may be a functional inhibition of the tricarboxylic acid cycle pathway in the body. Furthermore, succinic acid is also associated with anesthetic awakening time. Lazarev (Lazarev et al., 2011) investigated the possibility of using 1.5 % succinic acid solution to activate anesthesia recovery in 91 patients aged 1–14 years. Based on clinical data and BIS index, it was established that the administration of 1.5 % succinic acid solution during exit from the anesthesia period shortens the awakening time, decreases the time of restoration of motor activity and adequate respiration, and accelerates brain function recovery. Fentanyl analogs may cause increased excretion of succinate via the kidneys after intracellular accumulation by inhibiting succinate dehydrogenase activity or interfering with the mitochondrial electron transport chain, while plasma-available succinate decreases because of its inability to participate effectively in circulating metabolism. This disturbance in energy metabolism directly affects the ATP supply to nerve cells, leading to neuronal energy deprivation, which in turn leads to decreased synaptic transmission efficiency and abnormal neuronal excitability, closely related to the neuroadaptive changes induced by fentanyl-type substance addiction. Taurine, on the other hand, plays an important role in immune regulation, enhancing immunity through enhancing macrophage and lymphocyte activity and balancing secretion of immune factors, and maintaining normal nerve impulse conduction by regulating neuronal osmotic pressure and ion concentration in the nervous system (Jong et al., 2021). Notably, studies have shown that acute administration of morphine induces a significant increase in GABA content in the ventral nucleus and parafascicular thalamic nucleus, as well as spinal taurine content (Kuriyama and Yoneda, 1978). The decrease in plasma level and increase in urine level suggests that there may be abnormal consumption of taurine in the central nervous system and renal reabsorption dysfunction. During the addiction process of fentanyl analogs, neurons are chronically in a state of calcium homeostasis imbalance after opioid receptor agonism, and taurine, as an osmotic pressure regulator and calcium channel antagonist, has its reserves depleted to buffer excitotoxicity, leading to a decrease in concentration in the peripheral blood. Secondly, renal reabsorption dysfunction further exacerbates the loss of taurine from the body, weakening its antagonistic effect on glutamate excitotoxicity and creating a vicious cycle of neurological damage (El Idrissi, 2008).

4.2. Metabolic pathway disruption and toxicological mechanisms

Abuse of fentanyl or carfentanyl could cause oxidative and immune system damage and nervous system disorders by altering glucose and energy metabolism. The urinary metabolomics data for both low and high dose treatments in our study revealed perturbations in taurine and hypotaurine metabolism in response to drug administration, as evidenced by altered levels of taurine and citric acid in the treatment cohort. Taurine is ubiquitously present at millimolar concentrations in various animal tissues, notably neural tissues, the retina, and neutrophils. Taurine, hypotaurine, and their metabolic precursors, including cysteic acid, cysteamine, and cysteinesulphinic acid, are hypothesized to function as antioxidants in vivo (Aruoma et al., 1988). We also noted disruption in the metabolism of glutathione, another important antioxidant, as characterized by elevated levels of glycine and L-glutamic acid in the treated groups. This further suggested that fentanyl-related drugs induce oxidative damage and disruptions in nutrient metabolism and cellular regulation, as reported in previously published studies (Dhummakupt et al., 2019, Gasmi et al., 2023, Anderson, 1998).

Disorders in cysteine and methionine metabolism (L-methionine and 2-ketobutyric acid) were also observed in our treatment groups, indicating drug-related triggering of disrupted energy metabolism, immune system damage, and oxidative stress similar to that reported in previous metabolomics studies (Li et al., 2022, Dhummakupt et al., 2019, Upadhyayula et al., 2023). We also found disorders in pentose and glucuronate interconversions similar to those seen in previous metabolomics analysis following sufentanyl administration, as our finding of the characteristic metabolite xylitol suggested a possible disturbance of glucose metabolism and energy metabolism (Li et al., 2022, Sun et al., 2018). We also observed disruptions in tyrosine metabolism (L-tyrosine, 5,6-dihydroxyindole and vanillylmandelic acid) in the treated groups, which are interesting in light of recent research that has suggested a significant interaction between tyrosine metabolism and the tumor immune microenvironment (Zhou et al., 2022, Schenck and Maeda, 2018). Our analysis of the rat metabolomic data also pinpointed a marked disruption in pyrimidine metabolism (orotic acid, cytidine, and uracil) in the treatment groups. These alterations are indicative of a broader metabolic imbalance and point to deficiencies in the hematological, nervous, or mitochondrial systems (Löffler et al., 2005).

Fentanyl and carfentanyl abuse can lead to immune dysfunction, cellular homeostasis disorders, oxidative damage, and energy metabolism abnormalities, as well as neurological disorders. Our plasma metabolomic analysis revealed metabolite changes consistent with these effects following administration of either drug at both low and high dosages. For instance, our metabolomics analysis showing modulation of histidine metabolism corroborates previous findings of disrupted histidine metabolism in sufentanyl-treated subjects. Histidine can be converted to histamine by histidine decarboxylase, a process that occurs in enterochromaffin-like cells in the stomach, in mast cells in the immune system, and in various brain regions where histamine functions as a neurotransmitter (Li et al., 2022, Brosnan and Brosnan, 2020).

We also found indications of a disrupted citrate cycle (TCA cycle), as evidenced by elevated levels of succinic acid and citric acid in the treatment groups, suggesting disruptions in the generation of cellular energy and precursors required for biosynthetic pathways. The TCA cycle is an aerobic metabolic pathway essential for cellular energy production and it provides key intermediates for many biosyntheses. Following the initial conversion of glucose to pyruvate in the cytoplasm, glucose metabolism is completed in the mitochondria via the TCA cycle and this process is essential for the maintenance of cellular homeostasis (Eniafe and Jiang, 2021, Martínez-Reyes and Chandel, 2020).

Our finding of disrupted riboflavin metabolism ((−)-Riboflavin) in the treated groups is also informative, as recent research suggests that alteration of the riboflavin network may be the cause of some severe metabolic disorders, such as multiple acyl-CoA dehydrogenase deficiency (MADD) or Brown-Vialetto-van Laere syndrome (Barile et al., 2016). Our metabolomic data also pinpointed a marked disruption in ether lipid metabolism (plasmenyl-PE 38:6) in the treatment groups. Ether lipids, known for their propensity to form nonlamellar inverted hexagonal structures, are instrumental in membrane fusion reactions. They also contribute to the organization and stability of lipid rafts, which are cholesterol-rich domains critical for cellular signaling. Beyond their structural functions, select ether lipids can also act as endogenous antioxidants and are implicated in cell differentiation and signaling mechanisms (Dean and Lodhi, 2018).

4.3. Dose- and time-dependent toxicokinetics

Analysis of the pathway enrichments suggested by the urinary metabolite peak areas revealed a progressive decline in 2-ketobutyric acid and L-tyrosine at 6 h, 1 d, and 3 d post-treatment in the low-dose fentanyl group. Notably, 2-ketobutyric acid exhibited a similar trend in the high-dose fentanyl group. Additionally, uracil displayed the same pattern at the same time points in the high-dose group. As an intermediate product of threonine and isoleucine catabolism, the continuous decline of 2-ketobutyric acid may reflect the inhibition of mitochondrial energy metabolism by fentanyl (Bui et al., 2019). The reduction of this metabolite might be associated with decreased tricarboxylic acid activity, thereby affecting cellular energy supply. The decrease in L-tyrosine could be linked to fentanyl-induced inhibition of hepatic tyrosine aminotransferase activity or enhanced conversion to neurotransmitters such as dopamine, indirectly reflecting the stress response of the central nervous system (Fernstrom and Fernstrom, 2007). The specific change of uracil in the high-dose group may imply additional damage to the pyrimidine metabolic pathway. As a precursor for RNA synthesis, the decrease in uracil levels may be associated with inhibited proliferation of renal tubular epithelial cells or dysfunction of DNA repair, suggesting that high-dose fentanyl may exert more significant toxic effects on the urinary system.

Further peak area analysis of pathway-enriched plasma metabolites indicated a sustained decrease in sulfite levels over 15 d in both the low and high-dose fentanyl groups, suggesting sulfite as a potential long-term plasma biomarker of fentanyl intake. Sulfite is primarily converted to sulfate by sulfite oxidase for excretion (Ulrike and Christiane, 2001). Fentanyl may hinder sulfite metabolism by inhibiting sulfite oxidase activity, while the sustained decline in sulfite levels may reflect chronic liver detoxification impairment, leading to reduced sulfite production or enhanced excretion. Additionally, sulfite exhibits antioxidant properties, and its decreased levels may be accompanied by disruption of the redox balance in vivo.

LysoPA 20:4 and (−)-riboflavin showed a decline over 15 d in both the low and high-dose carfentanyl groups, whereas plasmenyl-PE 38:6 exhibited an increase at 3 d in the low-dose group and at 15 d in the high-dose group. The decrease in LysoPA 20:4 may reflect impaired degradation of cell membrane phospholipids or be associated with inhibition of inflammation-related phospholipase A2 activity. The delayed elevation of plasmenyl-PE might represent a compensatory synthetic response of the body to membrane damage (Wallner et al., 2018). Enriched in polyunsaturated fatty acids, plasmenyl-PE may participate in antioxidant stress or membrane repair processes.

L-histidine levels rose at 6 h in the low-dose group but fell at 15 d in the high-dose group. Short-term low-dose exposure potentially activates histidine decarboxylase to enhance histamine biosynthesis as a adaptive response to drug-induced stress, whereas long-term high-dose exposure may compromise hepatic-renal function, leading to impaired histidine catabolism or augmented urinary excretion, both contributing to the observed decline in histidine levels. It is herein proposed that the early alterations of 2-ketobutyric acid, L-tyrosine, and uracil in urine serve as potential biomarkers for rapid screening of acute fentanyl exposure. Concurrently, the persistent decrease in plasma sulfite levels represents a novel biomarker for chronic fentanyl intake, warranting further validation in clinical and forensic settings.

4.4. Limitations

While this study identified toxicity-related metabolic perturbations of fentanyl and carfentanyl via urine and plasma metabolomics, it has three key limitations that should be acknowledged. The experimental design lacked organ-specific endpoints and hematotoxicological indices: only systemic biofluids were analyzed, with no inclusion of samples from key target organs and no measurements of critical hematotoxicological indices such as alanine transaminase, aspartate transaminase, creatinine, or coagulation function. Second, species extrapolation from the rat model to humans is limited: rats and humans differ in the metabolism of fentanyl analogs. These differences restrict direct extrapolation of findings to human exposure scenarios, limiting applicability to clinical toxicology or human risk assessment. Third, potential biomarkers lack targeted validation: candidate cross-matrix biomarkers like succinate and taurine were identified via untargeted metabolomics but not verified through subsequent targeted approaches.

4.5. Prospect

Building on the findings and limitations of this study, future research can expand in three directions to deepen mechanistic understanding of fentanyl analog toxicity and enhance translational impact. First, incorporating in vitro models and in silico simulations will refine toxicity assessment: human-derived cell models such as HepG2 hepatocytes and SH-SY5Y neurons can validate whether core metabolic pathway perturbations identified here are recapitulated in human cells, while cytotoxicity assays including CCK-8 and LDH release tests can establish causal links between metabolic dysregulation and cell damage. Meanwhile, in silico tools such as metabolic network simulations and quantitative structure–activity relationship models can predict how emerging fentanyl analogs perturb these pathways, enabling high-throughput risk screening while reducing reliance on animal models. Second, focusing on development and validation of Adverse Outcome Pathways (AOPs) will advance standardized risk assessment: using the identified metabolic perturbations as a core, a candidate AOP for fentanyl analog toxicity can be constructed. Subsequent studies can validate this AOP via in vitro experiments, such as immune cell assays and animal model supplements establishing a standardized framework for fentanyl analog risk assessment. Third, advancing targeted research and clinical translation of potential biomarkers will enhance practical utility: for candidates like succinate and taurine, stable quantitative methods should first be established using targeted metabolomics to verify their dynamic changes across different toxic doses and exposure durations in rats; next, human clinical samples can be collected to evaluate the detectability of these biomarkers and their correlation with clinical symptoms like altered consciousness or liver injury; finally, exploration of rapid detection kits based on these biomarkers can provide practical tools for early diagnosis and toxicity monitoring of fentanyl analog intoxication.

5. Conclusion

In this study, UHPLC-MS/MS-based metabolomics analysis of urine and plasma from fentanyl- and carfentanyl-exposed SD rats revealed that these compounds induced dose- and time-dependent toxic responses, with core mechanisms involving oxidative stress, energy metabolism dysregulation, and perturbation of taurine/glutathione metabolic pathways. Urine metabolomics identified potential biomarkers for fentanyl abuse (2-ketobutyric acid, L-methionine, uracil, L-glutamate) and carfentanyl abuse (2-ketobutyric acid, L-methionine, taurine). Notably, succinic acid and taurine exhibited significant alterations in both plasma and urine, serving as shared biomarkers across these biological matrices. Plasma metabolomics analysis demonstrated that fentanyl and carfentanyl disrupted key metabolic pathways, including riboflavin metabolism and sulfur metabolism. Potential plasma biomarkers for fentanyl abuse were identified as plasmenylethanolamine, riboflavin, sphingomyelin, and L-tryptophan, while those for carfentanyl abuse included L-histidine, plasmenylethanolamine, riboflavin, and lysophospholipids. This work provides critical insights into the toxicological mechanisms of fentanyl analogs and establishes a foundation for forensic detection, with future studies warranted to validate biomarker reliability and explore their practical applications in clinical and forensic contexts.

2.2. Animal treatment

Adult male Sprague-Dawley (SD) rats weighing 180 ± 20 g (Zhanyuan Biotechnology Co., Ltd., Chengdu, China) were selected for this study based on their stable metabolic profiles, low individual variability (consistent with FDA Guidance for Industry and OECD Test Guideline 407), and good comparability of hepatic metabolic enzyme activity to humans (Food and Administration, 2008, OECD, 1995, Wang et al., 2021). Male rats were used to avoid confounding from hormonal fluctuations, a common practice in preliminary toxicological metabolomics. While the metabolic and toxic effects of fentanyl analogs observed in this study could theoretically also occur in female rats, this was not verified here—future studies may include female rats to explore potential sex-related differences. Rats were housed individually in metabolic cages under controlled environmental conditions (temperature, 22 ± 2 °C; humidity, 55 ± 10 %) and provided free access to standard rodent chow and sterile water. After the experiment, the rats were killed by cervical dislocation. All experimental procedures were performed in accordance with the OECD Test Guideline 420 and approved by the Academy of Forensic Science Ethics Committee for research on animal subjects, in compliance with the ethical requirements for animal welfare and laboratory animal management (No. 2019-2-20). Additionally, this study was approved by the Institutional Animal Care and Use Committee (IACUC) of Shanghai Zhanyuan Biotechnology Co., Ltd. with the approval number zysw-2023081501.

After adaptation to the standard laboratory conditions for 2 weeks, the rats were randomly divided into four groups (n = 6): fentanyl low-dose (0.18 mg/kg·qd/day, 1/100 LD50), fentanyl high-dose (0.45 mg/kg·qd/day, 1/40 LD50), carfentanyl low-dose (0.034 mg/kg·qd/day, 1/100 LD50), and carfentanyl high-dose (0.085 mg/kg·qd/day, 1/100 LD50). As reported in the literature (Pfizer Pharmaceuticals Group, 2013, Van Bever et al., 1976), the half lethal doses (LD50) of fentanyl and carfentanyl administered to rats by acute intravenous injection were 18 mg/kg and 3.39 mg/kg, respectively. Intravenous injections were given daily for 5 consecutive days into the vein at the end of the tail. We adopted an intra-individual control instead of inter-group controls to reduce inherent inter-rat metabolic variability, ensuring reliable detection of drug-induced metabolite changes. To address potential procedural stress, the anesthetic effects of fentanyl analogs minimized stress during orbital bleeding (Van Herck et al., 1997), and sampling intervals avoided cumulative stress (Van Herck et al., 2001). During the administration period, we observed a reduction in water intake in the rats after drug injection—a phenomenon that may be attributed to the rats spending most of their time in an anesthetized state induced by fentanyl analogs.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

The authors are grateful to the National Key Research and Development Program of China (2022YFC3302003), National Natural Science Foundation of China (81971789), the Shanghai Key Laboratory of Forensic Medicine (21DZ2270800) and the Shanghai Forensic Service Platform (19DZ2292700) for their financial support of this study.

Footnotes

Appendix A

Supplementary data to this article can be found online at https://doi.org/10.1016/j.crtox.2025.100269.

Appendix A. Supplementary data

The following are the Supplementary data to this article:

Supplementary Data 1
mmc1.docx (183.7KB, docx)

References

  1. Amante E., Alladio E., Rizzo R., Di Corcia D., Negri P., Visintin L., Guglielmotto M., Tamagno E., Vincenti M., Salomone A. Untargeted metabolomics in forensic toxicology: a new approach for the detection of fentanyl intake in urine samples. Molecules. 2021;26(16) doi: 10.3390/molecules26164990. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Anderson M.E. Glutathione: an overview of biosynthesis and modulation. Chem. Biol. Interact. 1998;112:1–14. doi: 10.1016/s0009-2797(97)00146-4. [DOI] [PubMed] [Google Scholar]
  3. Armenian P., Vo K.T., Barr-Walker J., Lynch K.L. Fentanyl, fentanyl analogs and novel synthetic opioids: a comprehensive review. Neuropharmacology. 2018;134:121–132. doi: 10.1016/j.neuropharm.2017.10.016. [DOI] [PubMed] [Google Scholar]
  4. Aruoma O.I., Halliwell B., Hoey B.M., Butler J. The antioxidant action of taurine, hypotaurine and their metabolic precursors. Biochem. J. 1988;15(1):256. doi: 10.1042/bj2560251. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Barile M., Giancaspero T.A., Leone P., Galluccio M., Indiveri C. Riboflavin transport and metabolism in humans. J Inherit Metab Di. 2016;39(4):545–557. doi: 10.1007/s10545-016-9950-0. [DOI] [PubMed] [Google Scholar]
  6. Benyamin R., Trescot A.M., Datta S., Buenaventura R., Adlaka R., Sehgal N., Glaser S.E., Vallejo R. Opioid complications and side effects. Pain Physician. 2008;11(2):S105–S120. [PubMed] [Google Scholar]
  7. Brosnan M.E., Brosnan J.T. Histidine metabolism and function. J. Nutr. 2020;150:2570S–2575S. doi: 10.1093/jn/nxaa079. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bui D., Ravasz D., Chinopoulos C. The effect of 2-Ketobutyrate on mitochondrial substrate-level phosphorylation. Neurochem. Res. 2019;44(10):2301–2306. doi: 10.1007/s11064-019-02759-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Burns S.M., Cunningham C.W., Mercer S.L. DARK classics in chemical neuroscience: fentanyl. ACS Chem. Nerosci. 2018;9(10):2428–2437. doi: 10.1021/acschemneuro.8b00174. [DOI] [PubMed] [Google Scholar]
  10. Coopman V., Cordonnier J., De Leeuw M., Cirimele V. Ocfentanil overdose fatality in the recreational drug scene. Forensic Sci. Int. 2016;266:469–473. doi: 10.1016/j.forsciint.2016.07.005. [DOI] [PubMed] [Google Scholar]
  11. Courtwright D.T. The prepared mind: Marie Nyswander, methadone maintenance, and the metabolic theory of addiction. Addiction. 1997;92(3):257–265. doi: 10.1111/j.1360-0443.1997.tb03196.x. [DOI] [PubMed] [Google Scholar]
  12. Dayer L.E., Painter J.T., McCain K., King J., Cullen J., Foster H.R. A recent history of opioid use in the US: three decades of change. Subst. Use Misuse. 2019;54(2):331–339. doi: 10.1080/10826084.2018.1517175. [DOI] [PubMed] [Google Scholar]
  13. Dean J.M., Lodhi I.J. Structural and functional roles of ether lipids. Protein Cell. 2018;9(2):196–206. doi: 10.1007/s13238-017-0423-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Dhummakupt E.S., Rizzo G.M., Feasel M., Mach P.M., Tran B.Q., Carmany D.O., Demond P.S., McBride E.M., Maughan M., Sekowski J.W., Glaros T. Proteomic and metabolomic profiling identify plasma biomarkers for exposure to ultra-low levels of carfentanil. Toxicol. Sci. 2019;167(2):524–535. doi: 10.1093/toxsci/kfy259. [DOI] [PubMed] [Google Scholar]
  15. El Idrissi A. Taurine increases mitochondrial buffering of calcium: role in neuroprotection. Amino Acids. 2008;34(2):321–328. doi: 10.1007/s00726-006-0396-9. [DOI] [PubMed] [Google Scholar]
  16. Eniafe J., Jiang S. The functional roles of TCA cycle metabolites in cancer. Oncogene. 2021;40(19):3351–3363. doi: 10.1038/s41388-020-01639-8. [DOI] [PubMed] [Google Scholar]
  17. Fernstrom J.D., Fernstrom M.H. Tyrosine, phenylalanine, and catecholamine synthesis and function in the brain. J. Nutr. 2007;137(6):1539S–S1547. doi: 10.1093/jn/137.6.1539S. [DOI] [PubMed] [Google Scholar]
  18. U.S. Food and Drug Administration (FDA). (2008). Guidance for Industry: Safety Testing of Drug Metabolites (Final Guidance). [Document No. FDA-2008-D-0490].
  19. Frank R.G., Pollack H.A. Addressing the fentanyl threat to public health. N. Engl. J. Med. 2017;376(7):605–607. doi: 10.1056/NEJMp1615145. [DOI] [PubMed] [Google Scholar]
  20. Fu S.L., Luong S.S., Pham A., Charlton N., Kuzhiumparambil U. Bioanalysis of urine samples after manipulation by oxidizing chemicals: technical considerations. Bioanalysis. 2014;6(11):1543–1561. doi: 10.4155/bio.14.102. [DOI] [PubMed] [Google Scholar]
  21. Gasmi A., Nasreen A., Lenchyk L., Lysiuk R., Peana M., Shapovalova N., Piscopo S., Komissarenko M., Shanaida M., Smetanina K., Antonyak H., Fira L., Lykhatskyi P., Fira D., Bjorklund G. An update on glutathione's biosynthesis, metabolism, functions, and medicinal purposes. Curr. Med. Chem. 2023 doi: 10.2174/0109298673251025230919105818. [DOI] [PubMed] [Google Scholar]
  22. Goggin M.M., Nguyen A., Janis G.C. Identification of unique metabolites of the designer opioid furanyl fentanyl. J. Anal. Toxicol. 2017;41(5):367–375. doi: 10.1093/jat/bkx022. [DOI] [PubMed] [Google Scholar]
  23. Jeal W., Benfield P. Transdermal fentanyl- a review of its pharmacological properties and therapeutic efficacy in pain control. Drugs. 1997;53(1):109–138. doi: 10.2165/00003495-199753010-00011. [DOI] [PubMed] [Google Scholar]
  24. Jong C.J., Sandal P., Schaffer S.W. The role of taurine in mitochondria health: more than just an antioxidant. Molecules. 2021;26(16) doi: 10.3390/molecules26164913. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Kuhl C., Tautenhahn R., Böttcher C., Larson T.R., Neumann S. CAMERA: an integrated strategy for compound spectra extraction and annotation of liquid chromatography/mass spectrometry data sets. Anal. Chem. 2012;84(1):283–289. doi: 10.1021/ac202450g. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Kuriyama K., Yoneda Y. Morphine-induced alterations of gamma-aminobutyric acid and taurine contents and l-glutamate decarboxylase activity in rat spinal-cord and thalamus - possible correlates with analgesic action of morphine. Brain Res. 1978;148(1):163–179. doi: 10.1016/0006-8993(78)90386-4. [DOI] [PubMed] [Google Scholar]
  27. Lazarev V.V., Khelimskaia I.A., Tsypin L.E., Mikhel'son V.A. Reamberin administration for earlypostnarcosis recovery in children. Eksp. Klin. Farmakol. 2011;74(6):10–13. [PubMed] [Google Scholar]
  28. Li J.N., Li S., Yu L., Wei J., Sun H., Yang C., Tan H. Identification of preoperative serum metabolites associated with postoperative opioid consumption in gastric cancer patients by extreme phenotype sampling. Pain Physician. 2022;25(2):E385–E396. [PubMed] [Google Scholar]
  29. Lindahl A., Sääf S., Lehtiö J., Nordström A. Tuning metabolome coverage in reversed phase LC-MS metabolomics of MeOH extracted samples using the reconstitution solvent composition. Anal. Chem. 2017;89(14):7356–7364. doi: 10.1021/acs.analchem.7b00475. [DOI] [PubMed] [Google Scholar]
  30. Liu X.Y., Hoene M., Wang X.L., Yin P.Y., Häring H.U., Xu G.W., Lehmann R. Serum or plasma, what is the difference? Investigations to facilitate the sample material selection decision making process for metabolomics studies and beyond. Anal. Chim. Acta. 2018;1037:293–300. doi: 10.1016/j.aca.2018.03.009. [DOI] [PubMed] [Google Scholar]
  31. Liu X.T., Zhao G., Sun S.J., Fan C.L., Feng X.J., Xiong P. Biosynthetic pathway and metabolic engineering of succinic acid. Front. Bioeng. Biotechnol. 2022;10 doi: 10.3389/fbioe.2022.843887. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Löffler M., Fairbanks L.D., Zameitat E., Marinaki A.M., Simmonds H.A. Pyrimidine pathways in health and disease. Trends Mol. Med. 2005;11(9):430–437. doi: 10.1016/j.molmed.2005.07.003. [DOI] [PubMed] [Google Scholar]
  33. Manier S.K., Meyer M.R. Impact of the used solvent on the reconstitution efficiency of evaporated biosamples for untargeted metabolomics studies. Metabolomics. 2020;16(3):34. doi: 10.1007/s11306-019-1631-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Martínez-Reyes I., Chandel N.S. Mitochondrial TCA cycle metabolites control physiology and disease. Nat. Commun. 2020;11(1) doi: 10.1038/s41467-019-13668-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Montesano C., Vincenti F., Fanti F., Marti M., Bilel S., Togna A.R., Gregori A., Di Rosa F., Sergi M. Untargeted metabolic Profiling of 4-Fluoro-Furanylfentanyl and Isobutyrylfentanyl in Mouse Hepatocytes and Urine by Means of LC-HRMS. Metabolites. 2021;11(2) doi: 10.3390/metabo11020097. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. OECD . OECD Publishing; Paris: 1995. Test Guideline 407: Repeated Dose 28-Day Oral Toxicity Study in Rodents. OECD Guidelines for the Testing of Chemicals, Section 4. [Google Scholar]
  37. Pfizer Pharmaceuticals Group (2013) Material safety data sheet: Fentanyl Citrate Injectable Solution.
  38. Schenck C.A., Maeda H.A. Tyrosine biosynthesis, metabolism, and catabolism in plants. Phytochemistry. 2018;149:82–102. doi: 10.1016/j.phytochem.2018.02.003. [DOI] [PubMed] [Google Scholar]
  39. Shafi A., Berry A.J., Sumnall H., Wood D.M., Tracy D.K. Synthetic opioids: a review and clinical update. Ther. Adv. Psychopharmacol. 2022;12 doi: 10.1177/20451253221139616. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Sindelar M., Patti G.J. Chemical discovery in the era of metabolomics. J. Am. Chem. Soc. 2020;142(20):9097–9105. doi: 10.1021/jacs.9b13198. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Smith C.A., Want E.J., O'Maille G., Abagyan R., Siuzdak G. XCMS: processing mass spectrometry data for metabolite profiling using nonlinear peak alignment, matching, and identification. Anal. Chem. 2006;78(3):779–787. doi: 10.1021/ac051437y. [DOI] [PubMed] [Google Scholar]
  42. Stanley T.H. The history and development of the fentanyl series. J. Pain Symptom Manage. 1992;7(3):S3–S7. doi: 10.1016/0885-3924(92)90047-l. [DOI] [PubMed] [Google Scholar]
  43. Steuer A.E., Brockbals L., Kraemer T. Untargeted metabolomics approaches to improve casework in clinical and forensic toxicology-“where are we standing and where are we heading?”. Wiley Interdiscip. Rev. Foren. Sci. 2022;4(4) doi: 10.1002/wfs2.1449. [DOI] [Google Scholar]
  44. Sun H., Zhang A.H., Song Q., Fang H., Liu X.Y., Su J., Yang L., Yu M.D., Wang X.J. Functional metabolomics discover pentose and glucuronate interconversion pathways as promising targets for Yang Huang syndrome treatment with Yinchenhao Tang. RSC Adv. 2018;8(64):36831–36839. doi: 10.1039/c8ra06553e. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Suzuki J., El-Haddad S. A review: Fentanyl and non-pharmaceutical fentanyls. Drug Alcohol Depend. 2017;171:107–116. doi: 10.1016/j.drugalcdep.2016.11.033. [DOI] [PubMed] [Google Scholar]
  46. Szeremeta M., Pietrowska K., Niemcunowicz-Janica A., Kretowski A., Ciborowski M. Applications of metabolomics in forensic toxicology and forensic medicine. Int. J. Mol. Sci. 2021;22(6) doi: 10.3390/ijms22063010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Ulrike K., Christiane D. Enzymology and molecular biology of prokaryotic sulfite oxidation. FEMS Microbiol. Lett. 2001;1:1–9. doi: 10.1111/j.1574-6968.2001.tb10813.x. [DOI] [PubMed] [Google Scholar]
  48. Upadhyayula P.S., Higgins D.M., Mela A., Banu M., Dovas A., Zandkarimi F., Patel P., Mahajan A., Humala N., Nguyen T.T.T., Chaudhary K.R., Liao L., Argenziano M., Sudhakar T., Sperring C.P., Shapiro B.L., Ahmed E.R., Kinslow C., Ye L.F., Siegelin M.D., Cheng S., Soni R., Bruce J.N., Stockwell B.R., Canoll P. Dietary restriction of cysteine and methionine sensitizes gliomas to ferroptosis and induces alterations in energetic metabolism. Nat. Commun. 2023;14(1) doi: 10.1038/s41467-023-36630-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Van Bever W.F., Niemegeers C.J., Schellekens K.H., Janssen P.A. N-4-Substituted 1-(2-arylethyl)-4-piperidinyl-N-phenylpropanamides, a novel series of extremely potent analgesics with unusually high safety margin. Arzneimittelforschung. 1976;26(8):1548–1551. [PubMed] [Google Scholar]
  50. Van Herck H., De Boer S.F., Hesp A.P., Van Lith H.A., Baumans V., Beynen A.C. Orbital bleeding in rats while under diethylether anaesthesia does not influence telemetrically determined heart rate, body temperature, locomotor and eating activity when compared with anaesthesia alone. Lab Anim. 1997;31(3):271–278. doi: 10.1258/002367797780596284. [DOI] [PubMed] [Google Scholar]
  51. Van Herck H., Baumans V., Brandt C.J., Boere H.A., Hesp A.P., van Lith H.A., Schurink M., Beynen A.C. Blood sampling from the retro-orbital plexus, the saphenous vein and the tail vein in rats: comparative effects on selected behavioural and blood variables. Lab Anim. 2001;35(2):131–139. doi: 10.1258/0023677011911499. [DOI] [PubMed] [Google Scholar]
  52. Wallner S., Orsó E., Grandl M., Konovalova T., Liebisch G., Schmitz G. Phosphatidylcholine and phosphatidylethanolamine plasmalogens in lipid loaded human macrophages. PLoS One. 2018;13(10) doi: 10.1371/journal.pone.0205706. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Wang W., Teresa M., Cai J.W., Zhang C.H., Wong S., Yan Z.Y., Khojasteh S.C., Zhang S.L. Comparative assessment for rat strain differences in metabolic profiles of 14 drugs in Wistar Han and Sprague Dawley hepatocytes. Xenobiotica. 2021;51(1):15–23. doi: 10.1080/00498254.2020.1795949. [DOI] [PubMed] [Google Scholar]
  54. Wen B., Mei Z., Zeng C., Liu S. metaX: a flexible and comprehensive software for processing metabolomics data. BMC Bioinf. 2017;18:183. doi: 10.1186/s12859-017-1579-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Zhang X.W., Li Q.H., Xu Z.D., Dou J.J. Mass spectrometry-based metabolomics in health and medical science: a systematic review. RSC Adv. 2020;10(6):3092–3104. doi: 10.1039/c9ra08985c. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Zhou Y.Y., Li X.X., Long G., Tao Y.G., Zhou L.D., Tang J.N. Identification and validation of a tyrosine metabolism-related prognostic prediction model and characterization of the tumor microenvironment infiltration in hepatocellular carcinoma. Front. Immunol. 2022;13 doi: 10.3389/fimmu.2022.994259. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Data 1
mmc1.docx (183.7KB, docx)

Articles from Current Research in Toxicology are provided here courtesy of Elsevier

RESOURCES