Abstract
Background:
Specialized pro-resolving mediators (SPMs) are bioactive lipids derived from n-3 and n-6 polyunsaturated fatty acids. SPMs promote resolution of inflammation and are reduced in Alzheimer’s disease. It is unknown whether SPMs are associated with post-stroke cognitive impairment (PSCI).
Objective:
In the present report, we aimed to study the levels of SPMs in PSCI patients in the acute phase of ischemic stroke.
Methods:
Levels of SPMs in the plasma from 36 patients with PSCI and 33 patients with post-stroke non-cognitive impairment (PSNCI) were measured by enzyme immunoassay.
Results:
We found that levels of the SPM lipoxin A4 (LXA4) were significantly reduced in PSCI patients compared with PSNCI patients. Interestingly, the LXA4 levels were positively correlated with Mini-Mental State Examination scores, but not with the National Institutes of Health Stroke Scale scores. Such alteration and correlation were not found in any of the other SPMs analyzed, i.e., including resolvin D1, resolvin D2, and maresin 1.
Conclusion:
We conclude that the plasma levels of LXA4 were reduced in PSCI patents in the acute phase of ischemic stroke and were correlated to cognitive function.
INTRODUCTION
Inflammation has emerging roles in dementia and is proposed as the potential novel markers for dementia diagnosis and treatment efficacy [1]. Increased levels of pro-inflammatory markers have been found in the cerebrospinal fluid (CSF) [2], postmortem brain [3–5], and plasma/serum of patients with Alzheimer’s disease (AD) [6–8]. Activated microglia, overexpressed cytokines, and oxidative stress have been demonstrated to contribute to the pathological and synaptic progression in AD. Thus, the abnormal pro-inflammatory signaling is believed as an important etiology of AD.
Specialized pro-resolving mediators (SPMs) are endogenous bioactive lipids derived from polyunsaturated fatty acids [9]. There are several types of SPMs, including arachidonic acid derived lipoxins (LXs), docosahexaenoic acid derived resolvin Ds (RvDs), neuroprotectins, and maresins (MaRs), and eicosapentaenoic acid derived resolvin Es. SPMs play key roles in resolution of inflammation [10] and are able to cease the infiltration of inflammatory cells, increase production of anti-inflammatory cytokines, enhance non-phlogistic clearance of inflammatory stimuli, as well as promote tissue regeneration. The disturbance of SPM signaling may lead to uncontrolled inflammation, and is linked to various disease conditions, including several neuroinflammatory and neurodegenerative diseases [11–13], such as AD. The levels of LXA4 are reduced in the CSF of mild cognitive impairment and AD patients [14]. The same finding was also found in the hippocampus of postmortem AD brain. Restoring resolution signaling with exogenous LXA4 has been demonstrated to ameliorate AD pathology and improve behavior abnormality in AD mouse models [15, 16]. The decreased levels of other SPMs, such as MaR1 and neuroprotectin 1, have also been reported in AD [17].
Post-stroke cognitive impairment (PSCI) is a type of vascular cognitive impairment that occurs within 6 months after stroke. Certain infarctions in strategic area of the brain may directly cause cognitive impairment. However, patients with stroke lesions occurring in brain regions that are not traditionally cognition-involved may also develop PSCI. The intrinsic mechanism is still unclear. Disturbed blood flow in the stroke brain does not only cause corresponding parenchyma damage, but also triggers inflammation, which may contribute to PSCI development. In this study, we aimed to find out whether SPMs signaling is disturbed in PSCI patients with ischemic stroke, as well as the potential association between SPMs and cognitive function.
SUBJECTS AND METHODS
Patients
Patients with acute ischemic stroke were enrolled from Shanghai Jiao Tong University Affiliated Sixth People’s Hospital between June 2019 to December 2019. All patients have undergone MRI or CT scan to confirm the acute cerebral infarction. Cognitive evaluation was done by Mini-Mental State Examination (MMSE) within the first 7 days of stroke onset. Patients with a MMSE score equal to or lower than 26 were defined as the PSCI group, and the ones with MMSE equal to or higher than 27 were defined as the non-PSCI group (PSNCI) [18, 19]. Exclusion criteria includes: acute cerebral hemorrhage; severe stroke (National Institutes of Health Stroke Scale score higher than 5); pre-stroke diagnosis of dementia or mild cognitive impairment; complaint of pre-stroke cognitive problems by the patients or their caregivers; significant aphasia or dysarthria; severe impairment of muscle strength that affects writing ability; complication of acute infection; history of autoimmune diseases or cancer; illiteracy or education lower than 6 years.
All the patients and their caregivers were informed and gave the written consent before being enrolled in the study. The study was approved by the ethical committee of Shanghai Jiao Tong University Affiliated Sixth People’s Hospital.
Preparation of plasma samples
Peripheral venous fast blood was collected in ethylenediaminetetraacetic acid (EDTA)-coated tubes within 72 h after stroke onset. Plasma samples were prepared and analyzed according to assay specification. Briefly, plasma samples were first diluted with methanol and water, and then acidified to pH 3.5 by hydrochloric acid. The acidified samples were then purified through C18 columns, which were pre-conditioned by methanol. Purified samples in the C18 columns were washed by water and hexane successively, and finally eluted with methyl formate. The eluted lipids were resuspended in assay buffer after evaporation of methyl formate by nitrogen gas, and were stored at –80°C before assay analysis.
Analysis of lipid mediators
The prepared lipid samples were analyzed with the LXA4 enzyme immunoassay (EIA) kits (Oxford Biomedical Research, Oxford, MI, USA), RvD1, RvD2 and MaR1 ELISA kits (Cayman Chemical, Ann Arbor, MI, USA), as well as leukotriene B4 kits (LTB4; Cayman Chemical, Ann Arbor, MI, USA) as a reference pro-inflammatory lipid mediator. The assays were performed according to the manufacturers’ instructions.
Statistics
All the data were processed by SPSS software (version 24. IBM corporation, Armonk, NY, USA). Chi-square test was used to examine difference of categorical variables. Analyses of data with non-equal variance were performed by the non-parametric Mann-Whitney U test. Correlation analysis was performed with the non-parametric Spearman’s rho test. Binary logistic regression was applied to analyze the association of baseline demographics and lipid mediators with cognitive status. In all analyses, p < 0.05 was defined as statistically significant.
RESULTS
Baseline characterization of patients and PSCI grouping
We enrolled 69 patients in total, among which 36 patients were defined as PSCI (MMSE score equal to or lower than 26), and 33 patients had PSNCI (MMSE score equal to or higher than 27). There was no difference with regards to age, gender, education, diabetes mellitus, hypertension, fasting blood glucose, plasma cholesterol, triglyceride, LDL, and homocysteine. We also compared blood cell counts to evaluate the influence of peripheral inflammatory cells. There was no difference with regards to total white blood cells, neutrophils, monocytes, or lymphocytes. The baseline characterization is shown in Table 1.
Baseline characterization of patients and PSCI grouping
NIHSS, the National Institutes of Health Stroke Scale; HbA1C, hemoglobin A1c; LDL, low-density lipoprotein. Statistics: Chi-square test was used for gender comparison; Mann–Whitney U test was applied for the comparisons of all other variables, and data are presented as mean±SD.
Plasma levels of SPMs and LTB4
Among the SPMs measured in the current study, LXA4 is reduced in the plasma of PSCI patients compared with PSNCI patients (Fig. 1A). No difference with regard to plasma RvD1, RvD2, or MaR1was found between PSCI and PSNCI patients (Fig. 1D-F). The ratio of SPM/LTB4 has been considered as an index for evaluating the balance between pro-resolving and pro-inflammatory actions. We found no difference of plasma LTB4 levels between PSCI and PSNCI patients (Fig. 1B). LXA4/LTB4 ratio was significantly lower in PSCI patients compared with PSNCI patients (Fig. 1C), but no difference of RvD1/LTB4, RvD2/LTB4 or MaR1/LTB4 was found between the two groups of patients (data not shown).
Levels of plasma SPMs in PSCI and PSNCI patients. LXA4 level was reduced in PSCI compared to PSNCI patients (A), and a similar result was observed when LXA4 was normalized to LTB4 (C). Levels of LTB4 (B) and other SPMs (D-F) were not different between PSCI and PSNCI group. Group comparison was done by Mann-Whitney U test. N.S., no significance. Error bars show median±SE.
Correlation analysis
Next, we test the possible correlations between SPMs and cognitive function. The levels of LXA4 and the ratio of LXA4/LBT4 are both positively correlated with MMSE scores (Spearman’s rho test, r = 0.412, p < 0.01, and r = 0.315, p < 0.01, respectively) (Fig. 2A, B). None of other SPMs (RvD1, RvD2 and MaR1) or their ratios to LTB4 was correlated with MMSE score (data not shown).
Plasma levels of LXA4 and the LXA4/LTB4 ratio were positively correlated to MMSE scores in stroke patients. Correlation analysis was done by Spearman’s rho test.
We also analyzed the potential association between SPMs and other blood lipids including cholesterol, triglyceride and LDL, and found no correlation of SPMs with these lipids (data not shown). Homocystein is both implicated in stroke and AD. We found homocysteine levels were not different between PSCI and PSNCI groups, and there was no correlation between plasma levels of SPMs and homocysteine (data not shown).
Logistic regression analysis
We did logistic regression analysis to explore the risk factors of PSCI. Aging was associated with an increased risk of PSCI (HR 1.27, 95% CI 1.023–1.577, p = 0.03), and plasma levels of LXA4 was associated with a reduced risk of PSCI (HR 0.945, 95% CI 0.907–0.984, p = 0.006). None of other components in the regression model was associated with PSCI risk (Table 2).
Logistic regression analysis of clinical characteristics and lipid mediators of inflammation for PSCI risk
NIHSS, the National Institutes of Health Stroke Scale; HbA1C, hemoglobin A1c; LDL, low-density lipoprotein; LXA4, lipoxin A4; LTB4, leukotriene B4; RvD1, resolvin D1; RvD2, resolvin D2; MaR1, maresin 1.
DISCUSSION
In the present study, we report that plasma LXA4 levels and the ratio of LXA4/LTB4 are reduced in acute ischemic stroke patients with PSCI and correlated with MMSE scores. Meanwhile, we also included analysis of other SPMs with regard to RvD1, RvD2, and MaR1 in our study, but there is no change of these SPMs between PSCI and PSNCI patients. These results indicate impaired resolution of inflammation may contribute to the pathogenesis of PSCI, and a special role of LXA4 compared with other SPMs in PSCI.
In an earlier study on patients with AD, mild cognitive impairment, or subjective cognitive impairment, LXA4 levels of AD patients are significantly reduced in the CSF [14]. The CSF levels of LXA4 are also correlated MMSE scores in these patients. These findings were not seen in RvD1 analysis in that study. In the present study, we show LXA4 is reduced in PSCI, a type of heterogeneous cognitive impairment related to vascular dementia. Thus, LXA4 is related to cognitive status in both neurodegenerative and vascular cognitive impairment. It is reasonable to suppose that LXA4 may be a specific marker in dementia, including disease spectrums of both AD and PSCI. Indeed, LXA4 has been implicated in mechanistic studies of both AD and stroke. For example, it has been shown that LXA4 or its analog could reduce AD-like pathology and improve cognition in various experimental AD models [15, 20–22]. The mechanisms include promoting resolution of inflammation, regulation of microglia to M2 polarization, enhancing neuronal survival against Aβ toxicity, ameliorating Aβ secretion, reducing tau phosphorylation through GSK-3β and p38 MAPK pathway. In stroke studies, LXA4 has been shown to reduce infarct volume, protect post-stroke BBB integrity, decrease levels of pro-inflammatory cytokines, as well as promote long-term recovery [23–30]. The beneficial actions of LXA4 in stroke have been associated with activation of PPAR-γ, nuclear factor erythroid 2-related factor 2, and inhibition of NFκ-B. The mechanistic role of LXA4 in PSCI has not been explored but may involve the above beneficial molecular pathways shown in studies of AD and stroke.
Abnormal inflammation has been proposed as a key pathophysiological change in PSCI [31]. Elevated levels of inflammatory cytokines [32–34], activated inflammasomes [35, 36], and M1-polarized microglia [37, 38] all contribute to PSCI pathogenesis. In line with these findings of increased inflammatory signals in PSCI, we reported decreased levels of LXA4, a classical SPM that exerts potent pro-resolving function in inflammation. LXA4 and LTB4 are both derived from arachidonic acid but play opposing roles in inflammatory response. Reduced levels of SPM/LTB4 ratio has been used as an index of the imbalanced pro-resolving and pro-inflammatory signals for inflammation-related diseases [39–42]. In our present study, plasma LXA4/LTB4 is reduced in PSCI patients and correlated with MMSE scores, indicating the imbalanced pro-resolving and pro-inflammatory signals may play a role in the patho-genesis of PSCI.
Older age and low education have been associated with increased risks of dementia [43]. In the present study, we found age was associated with increased PSCI risk, but education years showed no relationship with PSCI risk. This may be explained by relatively low average education levels (9 years of education) in our study. Other potentially relevant parameters, including cholesterol, triglyceride, and LDL are not related to PSCI risk in our regression model, indicating these atherosclerosis factors are more likely to play a role in stroke development, but not contribute to PSCI after stroke. Moreover, plasma levels of LXA4 were inversely correlated with risks of PSCI. The mechanisms of how decreased LXA4 contributes to PSCI pathogenesis remain to be explored. The changes of LXA4 may happen before or after the stroke attack. Patients with lower circulating LXA4 levels in advance of stroke may be more susceptible to PSCI, due to impaired capability to resolve post-stroke inflammation. On the other hand, such a decrease of LXA4 may be a post-stroke pathophysiological change that leads to cognitive decline.
In conclusion, our study indicated plasma level of LXA4 is related to cognitive status in ischemic stroke patients, and decreased LXA4 may be a potential risk factor for PSCI. The limitation of our study includes small sample size and cross-sectional design. It is also of note that liquid chromatography with tandem mass spectrometry (LC/MS/MS) is the best quantitative method to measure lipid mediators. Using EIA to analyze SPMs and LTB4 is a limitation of our study, but it has been widely accepted for comparative purpose. Moreover, the mechanisms of how LXA4 contributes to PSCI shall be further studied in basic research and larger longitudinal clinical studies.
Footnotes
ACKNOWLEDGMENTS
The study was supported by the National Natural Science Foundation of China (Grant No. 81501089, 81974158, 31811530007), Shanghai Committee of Science and Technology (Grant No. 17411950103) and the Swedish Foundation for International Cooperation in Research and Higher Education (Grant No. CH2017-7308).
