Identification of ANXA3 as a biomarker associated with pyroptosis in ischemic stroke

BACKGROUND

Pyroptosis plays an important role in the pathological process of ischemic stroke (IS). However, the exact mechanism of pyroptosis remains unclear. This paper aims to reveal the key molecular markers associated with pyroptosis in IS.

METHODS

We used random forest learning, gene set variation analysis, and Pearson correlation analysis to screen for biomarkers associated with pyroptosis in IS. Middle cerebral artery occlusion/reperfusion (MCAO/R) and oxygen and glucose deprivation/reoxygenation (OGD/R) models were constructed in vitro and in vivo. Cells were transfected with an Annexin A3 silencing (si-ANXA3) plasmid to observe the effects of ANXA3 on OGD/R + lipopolysaccharides (LPS)-induced pyroptosis. qRT‒PCR and western blotting were used to detect the expression of potential biomarkers and pyroptotic pathways.

RESULTS

Samples from a total of 170 IS patients and 109 healthy individuals were obtained from 5 gene expression omnibus databases. Thirty important genes were analyzed by random forest learning from the differentially expressed genes. Then, we investigated the relationship between the above genes and the pyroptosis score, obtaining three potential biomarkers (ANXA3, ANKRD22, ADM). ANXA3 and ADM were upregulated in the MCAO/R model, and the fold difference in ANXA3 expression was greater. Pyroptosis-related factors (NLRP3, NLRC4, AIM2, GSDMD-N, caspase-8, pro-caspase-1, cleaved caspase-1, IL-1β, and IL-18) were upregulated in the MCAO/R model. Silencing ANXA3 alleviated the expression of pyroptosis-related factors (NLRC4, AIM2, GSDMD-N, caspase-8, pro-caspase-1, cleaved caspase-1, and IL-18) induced by OGD/R + LPS or MCAO/R.

CONCLUSION

This study identified ANXA3 as a possible pyroptosis-related gene marker in IS through bioinformatics and experiments. ANXA3 could inhibit pyroptosis through the NLRC4/AIM2 axis.

Differential effects of annexins I, II, III, and V on cytosolic phospholipase A2 activity: specific interaction model

Annexins (ANXs) are a family of proteins with calcium-dependent phospholipid binding properties. Although inhibition of phospholipase A2 (PLA2) by ANX-I has been reported, the mechanism is still controversial. Previously we proposed a 'specific interaction' model for the mechanism of cytosolic PLA2 (cPLA2) inhibition by ANX-I [Kim et al., FEBS Lett. 343 (1994) 251-255]. Here we have studied the cPLA2 inhibition mechanism using ANX-I, N-terminally deleted ANX-I (DeltaANX-I), ANX-II, ANX-II(2)P11(2), ANX-III, and ANX-V. Under the conditions for the specific interaction model, ANX-I, DeltaANX-I, and ANX-II(2)P11(2) inhibited cPLA2, whereas inhibition by ANX-II and ANX-III was negligible. Inhibition by ANX-V was much smaller than that by ANX-I. The protein-protein interactions between cPLA2 and ANX-I, DeltaANX-I, and ANX-II(2)P11(2) were verified by immunoprecipitation. We can therefore conclude that inhibition of cPLA2 by specific interaction is not a general function of all ANXs, and is rather a specific function of ANX-I. The results are consistent with the specific interaction model.

Calcium-dependent neutrophil secretion: characterization and regulation by annexins

To gain direct access to the secretory machinery and study the regulation, mechanisms, and effectors of Ca2+-dependent neutrophil secretion, we developed an efficient and reproducible method of plasma membrane permeabilization using streptolysin O. We confirmed previous studies that permeabilized neutrophils secrete in response to calcium alone, but we also found that the Ca2+ dose-response is biphasic. Secretion is detectable at <1.0 microM Ca2+ and reaches a plateau between 1.0 and 60 to 80 microM. When stimulated with >80 microM Ca2+, secretion is two- to threefold greater than at lower [Ca2+], suggesting that two distinct mechanisms of Ca2+-dependent secretion that differ in their affinity for Ca2+ exist in neutrophils. Although permeabilization allows 100% leak of lactate dehydrogenase, maximum secretion from permeabilized cells is 80% that of f-met-leu-phe-stimulated intact cells, indicating that the essential components of the Ca2+-dependent secretory apparatus are predominantly, if not entirely, membrane bound. Permeabilization causes leakage of 100% of annexins V and VI, but 41% of annexin I and 12% of annexin III are retained. Immunofluorescence microscopy revealed that retained annexins I and III are associated with granule membranes. Addition of soluble annexins I and III to permeabilized cells increased Ca2+-induced secretion up to 15% and 90%, respectively, implying that both annexins participate in this secretory pathway. While annexin V is not required for secretion, it inhibits the low Ca2+-affinity mechanism of secretion.

Annexins from Ehrlich ascites cells inhibit the calcium-activated chloride current in Xenopus laevis oocytes

The effect of annexins II, III and V, purified from different species, on the calcium-activated chloride current across the stage-V to stage-VI Xenopus laevis oocyte membrane was tested either directly, using calcium entry mediated by depolarization, by A23187 permeabilization of oocytes or indirectly by quisqualate stimulation of a metabotropic glutamate receptor in the membrane expressed by the oocyte after injection of mRNA. The annexins isolated from the Ehrlich ascites cell, which is a mouse tumor cell, were found to be potent inhibitors of the chloride current, showing half-maximal inhibition at 50 nM, whereas no block was found using bovine or porcine annexins isolated from lung tissue. Of the annexins tested, we found annexin III to be naturally occurring in the oocyte, while only trace amounts of annexins II and V could be demonstrated. The inhibition pattern varied somewhat according to the stimulus method, the inhibition being more complete when an indirect stimulus via the metabotropic receptor was applied compared to a direct calcium stimulus.

[Effect of coagulation inhibitor proteins (Calphobindins) on tissue factor expression of endothelial cells]

We established a method for measuring procoagulant action on human umbilical vein endothelial cells (HUVEC). HUVEC (2.5 x 10(4)/well) were stimulated with 1 microgram/ml endotoxin (lipopolysaccharide: LPS) for 6 hours at 37 degrees C in 5% CO2. After washing, the HUVEC were incubated with assay buffer containing Proplex ST 1 unit (factor VII)/ml, S2222 0.6 mg/ml and CaCl2 6.6 mM, for 30 minutes at 37 degrees C. The procoagulant activity was determined by measuring the supernatant at OD405. Calphobindin I, II and III (CPB I, CPB II and CPB III) are the calcium dependent phospholipid binding proteins that exhibit anticoagulant activity in vitro. In this study, we investigated the effects of CPB I, CPB II and CPB III on procoagulant activity (PCA) expressed on HUVEC. The results are as follows 1) CPBI inhibits the procoagulant activity on HUVEC in a dose-dependent manner (IC 50% less than 0.4 microM). The same doses (0.4 microM) of CPBII and CPBIII decreased the procoagulant activity to 28.1% (CPBII), and to 84.6% (CPB III). CPB anticoagulant activities were, CPBII greater than CPBI greater than CPBIII, in that order. 2) When 0.05% H2O2 was added to the cell culture medium wells, concentrations of CPBI in supernatants increased in a time-dependent manner, and they reached to the maximum after 8 hours. CPBI in supernatants after 24 hours were not detected without H2O2, but concentrations of 4.88 ng/ml/10(4) cells with 0.01% H2O2, and 9.60 ng/ml/10(4) cells with 0.05% H2O2 were detected.