Role of activation of lipid peroxidation in the mechanisms of acute methanol poisoning^*

Abstract

Context: The role of activation of lipid peroxidation in the mechanisms of acute methanol poisoning has not been studied.

Objective: We measured the concentrations of lipid peroxidation markers in acutely intoxicated patients with known serum concentrations of methanol and leukotrienes.

Methods: Blood serum samples were collected from 28 patients hospitalized with acute intoxication and from 36 survivors 2 years after discharge. In these samples, concentrations of 4-hydroxy-trans-2-hexenal (HHE), 4-hydroxynonenal (HNE), and malondialdehyde (MDA) were measured using the method of liquid chromatography-electrospray ionization-tandem mass spectrometry.

Results: The maximum acute serum concentrations of all three lipid oxidative damage markers were higher than the follow-up serum concentrations: HNE 71.7 ± 8.0 ng/mL versus 35.4 ± 2.3 ng/mL; p < .001; HHE 40.1 ± 6.7 ng/mL versus 17.7 ± 4.1 ng/mL; p < .001; MDA 80.0 ± 7.2 ng/mL versus 40.9 ± 1.9 ng/mL; p < .001. The survivors without methanol poisoning sequelae demonstrated higher acute serum concentrations of the markers than the patients with sequelae. A correlation between measured markers and serum leukotrienes was present: HNE correlated with LTC4 (r = 0.663), LTD4 (r = 0.608), LTE4 (r = 0.771), LTB4 (r = 0.717), HHE correlated with LTC4 (r = 0.713), LTD4 (r = 0.676), LTE4 (r = 0.819), LTB4 (r = 0.746), MDA correlated with LTC4 (r = 0.785), LTD4 (r = 0.735), LTE4 (r = 0.814), LTB4 (r = 0.674); all p < .001. Lipid peroxidation markers correlated with anion gap (r= −0.428, −0.388, −0.334; p = .026, .045, .080 for HNE, HHE, MDA, respectively). The follow-up serum concentrations of lipid oxidation markers measured in survivors with and without visual/neurological sequelae 2 years after discharge did not differ.

Conclusion: Our results demonstrate that lipid peroxidation plays a significant role in the mechanisms of acute methanol poisoning. The acute concentrations of three measured biomarkers were elevated in comparison with the follow-up concentrations. Neuronal membrane lipid peroxidation seems to activate leukotriene-mediated inflammation as a part of the neuroprotective mechanisms. No cases of persistent elevation were registered among the survivors 2 years after discharge.

Keywords:

• Methanol poisoning
• lipid peroxidation
• neuroinflammation
• lipid oxidative damage

Introduction

Methanol is one of the most widely used toxic alcohols worldwide, and it is often misused to produce adulterated alcoholic drinks. Poisonings can occur as isolated events, due to unintentional or intentional ingestion, or as cluster events, due to mass outbreaks or “epidemics” [1,2]. During the recent Czech Republic mass methanol poisoning outbreak in 2012, there were 121 affected persons; 41 affected individuals died, while the remaining survived with either short- or long-term sequelae [3,4].

Methanol is oxidized in humans by alcohol dehydrogenase (ADH) to neurotoxic formaldehyde, and further by aldehyde dehydrogenases, to produce formic acid [5]. Undissociated formic acid crosses the blood-brain barrier, inhibits cytochrome c oxidase in mitochondria, resulting in depletion of ATP, accumulation of lactate, and development of severe metabolic acidosis with a high anion gap [6]. Disruption of cellular oxygen utilization leads to a decrease in adenosine triphosphate (ATP) production and accumulation of lactate [7]. The accumulation of formic and lactic acids leads to severe metabolic acidosis, histotoxic hypoxia, damage to the retina, optic nerve and brain basal ganglia, and death [8].

Bilateral necrosis of the basal ganglia and necrotic lesions in subcortical white matter, both hemorrhagic and non-hemorrhagic types, are the typical magnetic resonance imaging (MRI) findings in brains of severely poisoned patients [9–11]. Brain neurons, retinal ganglion cells, and their axons, are particularly vulnerable to histotoxic hypoxia due to their high energy demands [12,13]. The mechanisms behind toxic brain damage and methanol-induced acute optic neuropathy are not fully understood. Mitochondrial dysfunction leads to production of reactive oxygen species and toxic aldehydes, with oxidative damage of cellular structures; this is one of the key pathogenetic elements in the processes of acute and chronic neurodegeneration [14]. Leukotriene-mediated neuroinflammation, as a response to acute neuronal cell injury, may have both negative and protective roles in methanol-induced toxic brain damage [1].

Oxidative neuronal damage is a result of an imbalance in the reduction/oxidation (redox) state, toward an oxidative status, which may lead to functional and structural changes in macromolecules such as proteins, lipids and nucleic acids; this results in cell apoptosis and death [15,16]. However, minor to moderate grades of oxidative stress have certain positive effects as well. For example, moderate membrane lipid peroxidation induces cellular antioxidant mechanisms, toxic aldehyde scavenging enzymes, and increases cell survival [17,18].

Neuronal cells are more vulnerable to oxidative stress due to their relatively poor antioxidant protection and auto-oxidation properties of the neurotransmitters [19]. Further, neuronal membranes contain highly unsaturated fatty acids, which are extremely prone to lipid peroxidation with the production of isoprostanes and reactive unsaturated aldehydes such as malondialdehyde (MDA), 4-hydroxy-trans-2-hexenal (HHE) and 4-hydroxynonenal (HNE) [20].

The process of membrane lipid peroxidation is closely related to the synthesis of eicosanoids (prostaglandins, thromboxanes, leukotrienes, lipoxins, and others), which are the key signaling molecules produced by enzymatic or non-enzymatic oxidation of arachidonic acid or other polyunsaturated fatty acids from neuronal cell membranes [21]. In experimental animal studies, it was demonstrated that the biomarkers of brain injury and the mediators of neuroinflammation reach the peripheral circulation either through blood brain barrier disruption or via the glymphatic system in the brain with an intact blood brain barrier and can be measured in the blood serum [22]. In our previous study, we demonstrated that the patients with acute methanol poisoning had elevated serum concentrations of leukotrienes LTB4, LTC4, LTD4, and LTE4, suggesting activation of the mechanisms of leukotriene-mediated neuroinflammation [1].

The role of membrane lipid peroxidation in this activation of leukotrienes production in methanol-induced brain damage has not been studied. However, the balance between the processes of lipid oxidative damage and neuroinflammatory reactions mediated by leukotrienes may determine the outcome of poisoning, and the health sequelae in survivors. Despite effective treatment, the incidence of CNS sequelae in patients with methanol poisoning remains high. Understanding the mechanisms involved in toxic brain damage is therefore crucial for the prevention of unfavorable outcome of treatment.

Aim of the study

We measured the acute concentrations and the dynamics of HHE, HNE, and MDA in peripheral blood serum of patients with acute methanol poisoning, and the follow-up concentrations of survivors. We investigated if the process of lipid peroxidation was related to the activation of leukotriene-mediated neuroinflammation and the concentrations of measured biomarkers of peroxidation were associated with acute serum concentrations of leukotrienes (LTB4, LTC4, LTD4 and LTE4) measured in our previous study and with the outcome [1]. Within the follow-up study of long-term health sequelae of poisoning, we measured the concentrations of the same markers in the survivors, 2 years after discharge, to test their associations with long-term outcomes of poisoning.

Materials and methods

Study design and setting

We performed a longitudinal cohort study of patients poisoned during the Czech Republic mass methanol poisoning in 2012. The intoxicated patients were hospitalized in 30 hospitals across the country. The admission clinical and laboratory data were collected prospectively using standardized data collection forms. Data regarding treatment and outcomes were obtained from discharge reports. Survivors of poisoning were examined 3–8 months and 2 years after discharge from hospital.

Lipid oxidative damage markers in peripheral blood serum were measured by the Laboratory of Medicinal Diagnostics, University of Chemistry and Technology, Prague. The study was conducted with the approval of the General University Hospital Ethics Committee in Prague, Czech Republic.

Selection of participants and treatment

The mandatory monitoring of all cases of acute methanol poisoning throughout the country began 3 days after the admission of the first methanol-poisoned patient. New cases of poisoning were reported to the Czech Ministry of Health and Toxicological Information Centre (TIC). Of all the reported cases of acute poisoning, only hospitalized patients were included in the study. Patients eligible for the study had to have had their serum samples collected upon admission and during hospitalization, at least four times in total.

The treatment methods complied with the European Association of Poisons Centers and Clinical Toxicologists (EAPCCT) and the American Academy of Clinical Toxicology (AACT) practice guidelines for managing methanol intoxication [6]. Ethanol and fomepizole were administered as antidotes, effectively blocking ADH [23]. The treatment included administration of folate [24,25]. Severe acidosis was managed with bicarbonate solution. Enhanced elimination techniques, including intermittent hemodialysis and continuous veno-venous hemodialysis/hemodiafiltration, were applied [26–29].

Clinical examinations

The diagnosis of methanol intoxication was established if the patient had a history of recent ingestion of an alcoholic beverage and if the patient’s serum methanol concentration exceeded 200 mg/L. Information regarding the intoxication circumstances and immediate complications was obtained either directly from the patients themselves, or from their relatives, if the patient’s condition did not allow for self-report.

The examination protocol included cerebral computed tomography (CT) and MRI of the brain. The patients underwent a neurological examination of reflexes, and motor, cerebellar, cranial and sensory nerves, together with a complete ocular examination that included visual acuity, visual fields, contrast sensitivity and color vision tests. A fundus examination was also performed. The patients were considered to have CNS sequelae of acute methanol intoxication if MRI or CT brain scans revealed symmetrical necrosis and hemorrhages of the basal ganglia and subcortical white matter [11].

Further, the patients were considered to have visual sequelae if signs of optic nerve toxic neuropathy were observed on admission and/or during hospitalization. These symptoms included pathologic findings on contrast sensitivity, visual acuity, perimeter, color vision and/or persisting lesions in fundoscopy on discharge [30].

Laboratory

To measure the concentrations of lipid oxidative damage markers, peripheral venous serum was taken from the intoxicated patients. Blood samples were first collected upon admission, then every 2–4 h during the first 48 h of hospitalization, and every 6–12 h during the following days of hospitalization. Additional samples were collected during follow-up examinations. Samples were spun to separate the serum, and were frozen at −80 °C before analysis.

The concentrations of lipid oxidative damage markers (HHE, HNE and MDA) were measured by solid-phase extraction, using the liquid chromatography-electrospray ionization-tandem mass spectrometry (LC-ESI-MS/MS) detection method, in order to quickly and effectively isolate biomarkers from blood serum.

Follow-up clinical examination protocol

Upon being released from hospital, patients were re-examined 3–8 months and 2 years after discharge. The examination protocol included a complete assessment of visual function, including a complete eye examination, routine ophthalmic tests and optical coherence tomography (OCT) of the retinal nervous fibers layer (RNFL). Further, MRI brain examinations, neurological and psychological examinations, Neuroprotection and Natural History in Parkinson’s Plus Syndromes scale (NNIPPS), and neurological examinations of motor and sensory functions, reflexes, cerebellar function and cranial nerves, were performed. Biochemical tests were also performed: glucose, glycohemoglobin, albumin, prealbumin, liver and renal function, lipids, thyroid stimulating hormone, vitamins B₁ and B₁₂, ethyl glucuronide screening in urine and carbohydrate-deficient transferrin.

Calculations and data analysis

For all variables, basic descriptive statistics (mean, CI, SD, skewness and kurtosis) were calculated and were tested for normality using the Kolmogorov–Smirnov test. The χ² (Chi²) test was used to compare the frequencies of categorical demographic and clinical variables between the group of patients with acute methanol intoxication and the follow-up group. To compare the concentration of lipid oxidative damage markers in peripheral blood serum, a one-way analysis of variance was used with the least significant difference test. Bivariate relationships were evaluated using Pearson’s correlation coefficient. Statistical significance was designated as p < .05. Statistical analyses were performed in Excel (Microsoft, Redmond, WA) and IBM SPSS version 20.0 (Chicago, IL).

Results

Demographic characteristics of the patients and admission data

Of the 121 subjects with acute methanol poisoning, 101 were hospitalized and 20 died before admission. From the group of hospitalized patients, 28 had blood serum samples taken at admission and during hospitalization (Group I). The samples were stored for later analysis. Of the 28 patients, six patients died during hospitalization. In the patients who survived methanol poisoning with sequelae, in five cases, the visual damage was diagnosed, in six cases, the necrosis of the basal ganglia of the brain, and in 13 patients both visual and CNS sequelae were found. There were four cases of complete blindness. The other patients with visual sequelae had color vision and visual field defects, diminished contrast sensitivity, and fundus lesions, with abnormal visual evoked potentials of optic nerves and abnormal retinal nerve fibers layer thickness. The average age of the patients in Group I was 54.2 ± 5.2 years. Each patient in Group I had, on average, 12 ± 2 blood samples drawn, and the average observation time was 88 ± 20 hours. Sixteen patients from Group I participated in the follow-up clinical examination of long-term health sequelae of poisoning. In the remaining six survivors, information on the outcome of treatment was taken from their discharge reports.

Measurement of the follow-up concentrations of lipid peroxidation markers was performed in 36 survivors of methanol poisoning, 2 years after discharge (Group II, or the Follow-up group). For 12 of these patients, the results of measurement of acute concentrations of MDA, HNE and HHE were also available.

The demographic and admission laboratory data for the patients from both groups are presented in Table 1. The groups did not differ in terms of time of admission to hospital, and serum methanol and ethanol concentrations; however, Group I had more severe acidosis due to the presence of patients who did not survive poisoning and died in hospital.

Table 1. Demographic and admission laboratory data in the patients with acute serum lipid oxidative damage markers measurements (Group I, acute) and in the follow-up group of survivors 2 years after discharge from hospital (Group II, follow-up).

	Group I (28)	Group II (36)	PI/II
Age	54.2 ± 5.2	47.6 ± 4.5	0.015
Gender F/M	9/19	6/30	0.150
Time [h]	38.4 ± 8.3	33.5 ± 8.0	0.408
MetOH [mg/L]	1220.0 ± 420.0	1340.0 ± 440.0	0.706
EtOH [mg/L]	160.0 ± 130.0	490.0 ± 340.0	0.078
pH	7.06 ± 0.10	7.19 ± 0.27	0.029
pCO2[kPa]	3.80 ± 0.59	3.72 ± 0.43	0.976
HCO3^− [mmol/L]	9.1 ± 2.7	12.4 ± 2.6	0.088
AG [mmol/L]	29.7 ± 4.0	27.1 ± 3.9	0.346
OG [mmol/L]	70.0 ± 17.0	51.0 ± 16.0	0.087
Formic acid [mg/L]	590.0 ± 140.0	500.0 ± 180.0	0.419
Glucose [mmol/L]	9.7 ± 2.0	7.6 ± 1.2	0.079
Creatinine [µmol/L]	95.0 ± 17.0	93.0 ± 13.0	0.837

Gender F/M: female/male; Time: time span between methanol ingestion and hospital admission; MetOH: serum methanol concentration on admission; EtOH: serum ethanol concentration on admission; Formic acid: serum formic acid concentration on admission; AG: anion gap; OG: osmolar gap; pH: arterial blood pH; HCO₃⁻: bicarbonate concentration; Glucose: serum glucose concentration on admission; Creatinine: serum creatinine concentration on admission; pCO₂: partial pressure of carbon dioxide on admission; p < .05 (bold numbers) was considered as significant.

Acute serum markers of lipid peroxidation and their dynamics

The acute concentrations of all three measured markers of lipid peroxidation, HHE, HNE and MDA, were significantly higher than the concentrations measured 2 years after hospital discharge in the Follow-up group (Figure 1).

Figure 1. Box-and-whisker plots of concentrations of serum lipid oxidative damage markers measured during hospitalization and at follow-up. Acute Cmax; maximum serum concentrations measured during hospitalization. Mean, standard error of the mean (SEM, box) and 95%CI of the mean (whiskers) are presented. All p < .001.

The acute serum concentrations of MDA, HHE and HNE did not differ significantly between the survivors and the patients who died, while the acute concentrations of cysteinyl-leukotrienes were significantly higher in survivors (Figure 2(A,B)). The observation time in the group of survivors did not differ from the observation time in the group of patients who died. Acute concentrations of all three measured markers of lipid peroxidation were higher in the patients who survived without sequelae than in the patients who survived with visual and CNS sequelae of poisoning (Figure 2(C)).

Figure 2. Serum lipid oxidative damage markers and leukotrienes concentrations measured during hospitalization and in the follow up group of survivors 2 years after discharge (A), in the survivors versus the patients who died during hospitalization (B), and in the survivors with sequelae versus survivors without sequelae (C), means ± 95%CI. LTB4, LTC4, LTD4, LTE4: cysteinyl-leukotrienes LTB4, BLTC4, LTD4, LTE4; HHE: 4-hydroxy-trans-2-hexenal; MDA: malondialdehyde; HNE: 4-hydroxynonenal; No sequelae – the patients in Group I who survived without health sequelae of acute methanol poisoning; Sequelae – the patients in Group I who survived with long-term visual and/or CNS sequelae of acute methanol poisoning; ***p < .001, **p < .01, *p < .05.

The dynamics of acute serum MDA, HNE and HHE concentration changes during hospitalization is shown in Figure 3. For all three measured markers, there was a gradual elevation of acute concentrations, without any peaks or decreases registered during the observation time. The rate of increase in acute serum concentration of markers of lipid oxidative damage was approximately 0.49 ng/mL/h.

Figure 3. Dynamics of concentration changes in acute serum lipid oxidative damage markers during the observation period, in patients hospitalized with acute methanol intoxication.

The acute concentrations of markers of lipid oxidation damage were negatively correlated with anion gap (AG), the laboratory parameter of severity of metabolic acidosis, and positively correlated with known acute serum concentrations of leukotrienes (Table 2). No association was observed between MDA, HHE and HNE acute concentrations and age, gender, time between ingestion and admission to hospital, or serum methanol, ethanol, formic acid, glucose and creatinine concentrations on admission (all p > .05). The patients with pre-hospital ethanol administration had higher MDA concentration than those without pre-hospital ethanol (r = 0.382; p = .045).

Table 2. Correlations of acute serum lipid peroxidation markers concentrations and acute serum LTs concentrations.

		HHE [ng/mL]	HNE [ng/mL]	MDA [ng/mL]	LTC4 [pg/mL]	LTD4 [pg/mL]	LTE4 [pg/mL]	LTB4 [pg/mL]	AG [mmol/L]
HHE [ng/mL]	Pearson’s Correlation
	Sig. (2-tailed)
	N
HNE [ng/mL]	Pearson’s Correlation	0.955**
	Sig. (2-tailed)	0.000
	N	28
MDA [ng/mL]	Pearson’s Correlation	0.927**	0.928**
	Sig. (2-tailed)	0.000	0.000
	N	28	28
LTC4 [pg/mL]	Pearson’s Correlation	0.713**	0.663**	0.785**
	Sig. (2-tailed)	0.000	0.000	0.000
	N	28	28	28
LTD4 [pg/mL]	Pearson’s Correlation	0.676**	0.608**	0.735**	0.747**
	Sig. (2-tailed)	0.000	0.001	0.000	0.000
	N	28	28	28	28
LTE4 [pg/mL]	Pearson’s Correlation	0.819**	0.771**	0.814**	0.856**	0.794**
	Sig. (2-tailed)	0.000	0.000	0.000	0.000	0.000
	N	28	28	28	28	28
LTB4 [pg/mL]	Pearson’s Correlation	0.746**	0.717**	0.674**	0.685**	0.395*	1.000
	Sig. (2-tailed)	0.000	0.000	0.000	0.000	0.038	0.000
	N	28	28	28	28	28	28
AG [mmol/L]	Pearson’s Correlation	−0.428**	−0.388**	−0.334**	−0.507**	−0.501*	−0.511*	−0.478
	Sig. (2-tailed)	0.026	0.045	0.026	0.006	0.007	0.005	0.010
	N	28	28	28	28	28	28	28

HHE: 4-hydroxy-trans-2-hexenal; HNE: 4-hydroxynonenal; MDA: malondialdehyde; LTB4, LTC4, LTD4, LTE4: cysteinyl-leukotrienes LTB4, BLTC4, LTD4, LTE4; AG: anion gap; p < .05 (bold numbers) was considered as significant.

In the patients with both acute and follow-up measurements of lipid peroxidation markers, there were no observed cases of persistent elevation (Figure 4). The concentrations of MDA, HHE and HNE decreased in all cases by 50–55%, and returned to the reference limits.

Figure 4. Decreases in concentrations of serum lipid oxidative damage markers measured during hospitalization (“Acute”) versus concentrations measured 2 years after discharge, in the same patients (“Follow-up”).

The measured follow-up levels of markers of lipid oxidative damage did not correlate with acute concentrations, or with serum concentrations of previously measured leukotrienes (both acute and follow-up ones); no association was observed between the follow-up concentrations and the long-term outcome (Figure 5).

Figure 5. Follow-up serum concentrations of markers of lipid peroxidation and leukotrienes measured 2 years after discharge in survivors without sequelae versus survivors with long-term visual and/or CNS sequelae of poisoning. HHE: 4-hydroxy-trans-2-hexenal; MDA: malondialdehyde; HNE: 4-hydroxynonenal; LTB4, LTC4, LTD4, LTE4: cysteinyl-leukotrienes LTB4, BLTC4, LTD4, LTE4; No sequelae – the patients in the follow-up Group II survived without health sequelae of acute methanol poisoning; Sequelae – the patients in the follow-up Group II survived with long-term visual and/or CNS sequelae of acute methanol poisoning; p < .05 was considered as significant.

Discussion

The results of our study indicate that the activation of membrane lipid peroxidation and its interaction with leukotriene-mediated neuroinflammation plays an important role in the mechanisms associated with toxic brain damage in acute methanol poisoning. The acute serum concentrations of markers of lipid oxidative damage (HNE, HHE, and MDA) were significantly elevated in patients hospitalized with methanol poisoning, as compared to the follow-up concentrations of these markers measured 2 years after discharge. The acute concentrations of all three markers demonstrated gradual increases during the first 3–4 days after admission to hospital. These acute concentrations were positively correlated with acute serum levels of leukotrienes, and negatively correlated with the severity of metabolic acidosis on admission. The patients with a better clinical outcome had relatively higher concentrations of measured lipid peroxidation markers, than those with a poor outcome. The elevation of serum markers of lipid oxidative damage was transient, and did not result in chronic changes. Finally, the serum concentrations of markers of lipid oxidative damage measured during the follow-up period were not associated with acute laboratory parameters of poisoning, follow-up concentrations of MDA, HHE and HNE, or the long-term outcome of poisoning.

The interaction between the direct cytotoxic effect of formic acid as the main toxic metabolite of methanol, activation of membrane lipid peroxidation, and the neuroinflammation involves complex mechanisms that are not yet fully understood [31]. Acidemia due to the accumulation of formic and lactic acids leads to the disruption of the blood-brain barrier integrity, dilatation of small cerebral vessels and brain edema [32–34]. On the cellular level, inhibition of cytochrome c oxidase and decrease in ATP production lead to the influx of sodium and calcium ions into neurons, neuronal overexcitation and swelling, excessive production of reactive oxygen species and toxic aldehydes, and membrane lipids peroxidation; this in turn leads to neuronal degeneration, apoptosis and death [7].

The lipid peroxidation is one of early non-specific protective mechanisms contributing to the activation of microglia and astrocytes, and triggering the neuroinflammation in acute traumatic and non-traumatic brain injury such as stroke, hypoxia, or ischemia [35]. Therefore, the products of lipid peroxidation are not specific to methanol poisoning. In our study, we found the association between the elevated markers of lipid peroxidation and leukotriene-mediated neuroinflammation, severity of poisoning, certain treatment modalities (pre-hospital ethanol administration), and the outcome. Neuronal membrane lipid peroxidation may play both negative and protective role at the cellular level; the outcome depends on its intensity and interaction with other processes, including synthesis of eicosanoids, activation of microglia, and neuroinflammation [36,37]. In severe oxidative stress, reactive toxic aldehydes produced massively as a result of lipid peroxidation, including MDA, HHE, and HNE, saturate the antioxidant systems, affect cellular proteins synthesis, inhibit several enzymatic systems, and directly damage intracellular micromolecules. On the other hand, the products of lipid peroxidation have signal effects, induce cellular antioxidant systems, act as intra- and intercellular secondary messengers, affect the expression of several genes, modify the structure and functions of certain cellular proteins and change their immunogenic properties [38]. HNE and MDA stimulate neutrophils activity and induce astrocytes proliferation necessary for the demarcation of the damaged regions by several signaling pathways [36,37,39]. The final effect of the production of reactive aldehydes depends on the severity of the acute injury, magnitude and duration of oxidative stress, character of the immune response, and other factors; the interaction between these factors determines if the protective or the destructive mechanisms will be prevailing.

One of the main toxic aldehydes scavengers in the brain neurons is the mitochondrial ALDH2. The activity of the enzyme is inducible by alcohol oxidation products and overexpression of the ALDH2 transgene effectively attenuates cell damage caused by oxidative stress [40]. Further molecules with antioxidant properties in the brain include superoxide dismutase, glutathione, glutathione peroxidase, metal binding proteins, alfa-tocopherol, ubiquinol, and carnosine [19,41]. Finally, compared to other tissues, the brain tissue contains relatively higher concentration of ascorbate with known antioxidant properties [42].

As demonstrated in the current study, acute methanol poisoning leads to increased production of the reactive toxic aldehydes, MDA, HHE and HNE, due to membrane lipid peroxidation. This process is associated with increased production of leukotrienes, suggesting that the mechanisms of neuroinflammation are triggered by the damage to neuronal membranes and activated by the products of lipid peroxidation. Leukotrienes are mediators of neuroinflammation, normally absent in the brain; they are produced by activated microglia and astrocytes in various crisis situations, such as brain hypoxia, hypoglycemia, head injury, epileptic fit, and stroke. Previous studies demonstrated that patients with stroke had elevated serum concentrations of MDA, and this elevation lasted for at least 7 days after admission. MDA concentrations measured 48 hours after admission in patients with stroke were approximately two times higher than in controls [43]. Concentrations of MDA were associated with the size of necrotic lesions, clinical severity of the stroke, CNS scores and the patient’s outcome [43,44].

With regard to the interaction between the pathophysiological mechanisms of membrane lipid peroxidation and neuroinflammation mediated by leukotrienes, certain biochemical enzymatic and non-enzymatic mechanisms may interconnect these two processes. The main precursors of HNE in humans are 13-hydroperoxyoctadecadienoic acid, produced by the oxidation of linoleic acid by 15-lipoxygenase-1, and 15-hydroperoxyeicosatetraenoic acid, produced by the oxidation of arachidonic acid by 15-lipoxygenase-2 [45,46]. Both these compounds are short-lived, and are further catabolized into various families of more stable compounds, including leukotrienes [47]. These biochemical mechanisms of transformation of common precursors may explain the strong positive associations between acute concentrations of markers of lipid peroxidation and leukotrienes, as observed in our study.

We found that the dynamics associated with elevated serum concentrations of MDA, HHE and HNE were gradual; there were no peaks observed during the period of observation which lasted more than 80 hours, although methanol and formic acid were typically eliminated during the first 24–48 hours of hospitalization by enhanced elimination methods. Formic acid may act as a trigger, activating slower mechanisms of non-traumatic brain injury, including neuroinflammation, which typically culminates on days 3–4 after admission, as it was demonstrated in our previously published study [ref. anonymized]. In animal models of acute poisoning, the enzymatic antioxidant systems typically increase their activity immediately after exposure to the toxic agent, as the first reaction to oxidative stress. However, several days later, the activity of antioxidant systems decrease by one-third to half of the original activity, mainly due to ATP deficiency, together with a 3- to 4-fold increase in the concentration of markers of lipid peroxidation [48]. Thus, our present results are in agreement with experimental toxicological data.

In our study, acute concentrations of lipid peroxidation markers correlated with anion gap, an indicator of severity of metabolic acidosis, on admission to hospital (see Table 2). It is known that severity of metabolic acidosis is one of the main prognostic factors associated with the outcome of acute methanol intoxication [49]. Decrease in arterial blood pH leads to the transformation of MDA to beta-hydroxyacrolein and increase of the reactivity [39].

Differences in acute serum concentrations of lipid peroxidation markers between survivors and those who died were not significant, unlike the concentrations of leukotrienes, which were higher in survivors, as we demonstrated previously [1]. The patients who survived with sequelae had similar acute serum concentrations of MDA, HHE, and HNE as the patients who died. Lower lipid peroxidation level could be insufficient for induction of protective neuroinflammation in the patients with poor outcome. Other factors, as brain ischemia, ischemia reperfusion injury, and overexcitation of hypoxic neurons could play a role in the outcome in these patients. On the other hand, in survivors, the acute levels of MDA, HHE and HNE were significantly higher in those who survived without visual and CNS damage (see Figure 2). This finding may demonstrate the protective role of mild to moderate lipid oxidative stress in the mechanisms involved in toxic brain damage caused by acute methanol exposure.

It is known that mild to moderate oxidative damage triggers the transcription of nuclear factor (erythroid-derived 2)-like 2 (Nrf2) [17,18]. Transcription of Nrf2 results in the activation of several cellular antioxidant systems, such as superoxide dismutase and glutathione. Further, mild membrane lipid peroxidation protects neurons from oxidative damage to proteins and nucleic acids, triggers the leukotriene-mediated inflammatory response, and prevents chronic neurodegenerative changes. The protective and curative effects of mild to moderate oxidative stress are employed in ozone therapy for the treatment of different diseases, such as backache, orthopedic diseases, neurodegenerative diseases, and diabetes mellitus [18].

The patients who received pre-hospital ethanol had higher MDA concentration. These patients had better outcome of poisoning than those without pre-hospital treatment, as was demonstrated in our earlier studies [50,51]. Early administration of ethanol has several protective effects besides ADH blocking: ethanol reduces ischemia reperfusion injury, prevents post-ischemic adhesive interactions between leukocytes and endothelial cells, and prevents overexcitation of brain neurons [52–54].

In acute methanol-induced neuronal damage of optic nerve and retina, the residual peripapillary edema persists for up to 2 months after discharge and the process of chronic remyelination of axons lasts for up to 2 years after acute optic neuritis [55,56]. The same process might take place in the brain neurons (optic nerve is part of CNS), what could have an impact on the biomarkers of neuroinflammation and lipid peroxidation during the first months after acute poisoning. On the other hand, chronic neuroinflammation and neurodegeneration with reactive microglia after a traumatic brain injury may persist after more than 1 year post-trauma [57]. In our study, we observed no cases of persistent elevation of markers of lipid oxidative damage in the population of survivors of methanol poisoning 2 years after discharge from hospital. The follow-up concentrations of MDA, HHE and HNE were within normal reference limits. This fact demonstrates that the processes of lipid peroxidation and leukotriene-mediated neuroinflammation were transient, adaptive responses to the acute impact without the shift to the chronic neurodegeneration. There were no differences in the follow-up concentrations of these markers between the groups of survivors without health sequelae and the patients with long-term visual and brain damage. The follow-up concentrations of the studied markers did not correlate with acute laboratory parameters, the results of longitudinal clinical examination, and the outcome of poisoning.

Limitations of the study

There were several limitations of the current study. The relatively small group of patients examined may have had an effect on the power of the study. Further, during hospitalization with acute poisoning, the effect of the patients’ comorbidities on their outcome was not considered. Finally, the study was not controlled with respect to the time from the intake of methanol to the beginning of hospital treatment, the blood sampling time for biochemical and toxicological laboratory analysis (laboratory parameters measured from the first blood sample collected at the hospital were considered as measured at admission), or the treatment modalities (choice of antidotes, methods of extracorporeal renal replacement therapy, alkalization and folate substitution).

Conclusions

In acute methanol poisoning, minor to moderate lipid peroxidation may play an important protective role in the mechanisms of methanol-induced brain damage by activating the leukotriene-mediated neuroinflammation. The patients with acute methanol poisoning had significantly higher concentrations of measured biomarkers of lipid peroxidation in peripheral blood serum than the survivors of poisoning examined 2 years after discharge. In survivors, the patients with poor outcome (visual and CNS sequelae) had lower acute serum concentrations of MDA, HHE, and HNE than those who survived without health sequelae. The absence of association between acute and follow-up concentrations of biomarkers of oxidative stress, along with the absence of difference in the follow-up concentrations between the patients who survived with and without visual/CNS sequelae, indicated that acute lipid peroxidation was transitory and was not followed by chronic neurodegeneration in survivors.

Disclosure statement

The authors acknowledge that they have no conflict of interests.

Additional information

Funding

This work was supported by the Ministry of Health of the Czech Republic (AZV) under Grant [16-27075A]; and under Grant [44/18/D]; and by the First Faculty of Medicine, Charles University in Prague under Grant [PROGRES Q25]; and under Grant [PROGRES Q29].