Measurements of protein-conjugated acrolein (PC-Acro), IL-6, and C-reactive protein (CRP) in plasma were useful for identifying silent brain infarction with high sensitivity and specificity. The aim of this study was to determine whether acrolein causes increased production of IL-6 and CRP in thrombosis model mice and cultured cells. In mice with photochemically induced thrombosis, acrolein produced at the locus of infarction increased the level of IL-6 and then CRP in plasma. This was confirmed in cell culture systems – acrolein stimulated the production of IL-6 in mouse neuroblastoma Neuro-2a cells, mouse macrophage-like J774.1 cells, and human umbilical vein endothelial cells (HUVEC), and IL-6 in turn stimulated the production of CRP in human hepatocarcinoma cells. The level of IL-6 mRNA was increased by acrolein through an increase in phosphorylation of the transcription factors, c-Jun, and NF-κB p65. Furthermore, CRP stimulated IL-6 production in mouse macrophage-like J774.1 cells and HUVEC. IL-6 functioned as a protective factor against acrolein toxicity in Neuro-2a cells and HUVEC. These results show that acrolein stimulates the synthesis of IL-6 and CRP, which function as protecting factors against acrolein toxicity, and that the combined measurement of PC-Acro, IL-6, and CRP is effective for identification of silent brain infarction.
The combined measurements of protein-conjugated acrolein (PC-Acro), IL-6, and C-reactive protein (CRP) in plasma were useful for identifying silent brain infarction. The aim of this study was to determine whether acrolein causes increased production of IL-6 and CRP, and indeed acrolein increased IL-6 synthesis and IL-6 in turn increased CRP synthesis. Furthermore, IL-6 decreased acrolein toxicity in several cell lines.
Polyamines (putrescine, spermidine, and spermine) exist at millimolar concentrations in cells and are essential for normal cell growth, mainly through stimulation of specific kinds of protein synthesis by changing the structure of bulged-out region of double-stranded RNA (Higashi et al. 2008; Igarashi and Kashiwagi 2010). However, when cells are damaged, polyamines unbind from RNA and the toxic compounds acrolein (CH2=CHCHO) and H2O2 are produced from polyamines, in particular from spermine, by polyamine oxidases (PAO, spermine oxidase, and acetylpolyamine oxidase) (Igarashi and Kashiwagi 2011). When the toxicities of acrolein and H2O2 were compared in a cell culture system, acrolein was more toxic than H2O2 (Sharmin et al. 2001). We found that levels of PAO and protein-conjugated acrolein (PC-Acro) are well correlated with the severity of stroke (Tomitori et al. 2005) and chronic renal failure (Igarashi et al. 2006). It was also shown that the level of PC-Acro in saliva is well correlated with severity of primary Sjögren's syndrome (Higashi et al. 2010), and that the level of PC-Acro in plasma is increased in Alzheimer's disease (Waragai et al. 2012).
There are reports that silent brain infarction (SBI) increases the risk of subsequent stroke and dementia (Kobayashi et al. 1997; Vermeer et al. 2007). It is, therefore, valuable to be able to estimate SBI at an early period by biochemical markers. We have reported that measurement of PC-Acro together with interleukin-6 (IL-6) and C-reactive protein (CRP) in plasma makes it possible to identify SBI with 84.1% sensitivity and 83.5% specificity (Yoshida et al. 2010). These biochemical markers also increased in the plasma of subjects with carotid atherosclerosis and white matter hyperintensity (Yoshida et al. 2010), which are risk factors of stroke (Bokura et al. 2006; Inoue et al. 2007). In this communication, we tried to determine whether acrolein produced during cell damage like brain infarction causes the increase of IL-6 and CRP in plasma of thrombosis model mice and also in cell culture systems.
Materials and methods
Acrolein and human CRP were purchased from Tokyo Chemical Industry Co. Ltd. (Tokyo, Japan) and Sigma-Aldrich (Tokyo, Japan), respectively. Human and mouse IL-6 and antibody against human IL-6 were obtained from R&D Systems (Minneapolis, MN, USA) and antibody against mouse IL-6 was from ENDOGEN (Woburn, MA, USA). Antibodies against c-Jun and NF-κB p65 were obtained from Santa Cruz Biotechnology, Inc. (Santa Cruz, CA, USA), and those against phospho-c-Jun and phosho-NF-κB p65 were from Cell Signaling Technology, Inc. (Beverly, MA, USA).
Photochemically induced thrombosis (PIT) model mice
All animal experiments were approved by the Institutional Animal Care and Use Committee of Chiba University and carried out according to the Guidelines for Animal Research of Chiba University. Male C57B/L mice (7 weeks old) were purchased from Japan SLC Inc (Hamamatsu, Japan). The thrombotic occlusion of the middle cerebral artery was induced by the photochemical reaction using 8-week-old mice weighting 22–26 g (Tanaka et al. 2007). Immediately after intravenous injection of photosensitizer, Rose Bengal (20 mg/kg), green light (wavelength: 540 nm) was illuminated the middle cerebral artery for 10 min. At indicated times after the induction of PIT, 2 mm thick coronal slices were incubated with 5% triphenyltetrazolium chloride solution at 37°C for 30 min. The volume of the infarction was analyzed on Macintosh computer using the National Institutes of Health Image program. Experiments were performed using seven mice in each group.
Mouse neuroblastoma Neuro-2a cells (Cheng et al. 2009) and mouse macrophage-like J774.1 cells (Moreau et al. 2007) were cultured in Dulbecco's modified Eagle's medium (Invitrogen Co., Carlsbad, CA, USA) containing 10% heat-inactivated fetal bovine serum, 50 U/mL penicillin G and 50 U/mL streptomycin, in a humidified atmosphere with 5% CO2 at 37°C. Human umbilical vein endothelial cells (HUVEC) (Tanabe et al. 2004) were grown in EGM-2 Bullet Kit (Lonza Co., Allendale, NJ, USA). The viable cell number was counted by microscope in the presence of 0.05% trypan blue. Human hepatocarcinoma FLC-4 cells were cultured as described previously (Kobayashi et al. 2012).
Measurement of PC-Acro, IL-6, CRP, and glutathione
Brain tissue (approximately 20 mg) was homogenized as reported previously (Saiki et al. 2009). The level of PC-Acro [Nε-3-formyl-3,4-dehydropiperidino lysine (FDP-lysine) in protein] was measured by western blotting (Nielsen et al. 1982) using 20 μg protein and polyclonal antibody against bovine serum albumin-conjugated acrolein (MoBiTec, Gottingen, Germany). Protein was determined by the method of Lowry et al. (1951). The level of PC-Acro in plasma was determined by the method of Uchida et al. (1998) using ACR-Lysine Adduct ELISA System (NOF Corporation, Tokyo, Japan). The level of IL-6 in plasma or in the culture medium obtained by centrifugation at 500 g for 5 min was determined using Endogen Human IL-6 ELISA kit (Pierce Biotechnology, Inc., Rockford, IL, USA) or Mouse IL-6 ELISA kit (Bender MedSystems, Inc., Vienna, Austria) according to the manufacturer's protocol. The level of CRP in plasma or in the culture medium was determined by western blotting using monoclonal antibody against human and mouse CRP (R&D Systems). The level of glutathione in cells was determined using NWLSS Glutathione Assay Kit (Northwest Life Science Specialties, Vancouver, WA, USA) according to the manufacturer's protocol.
Measurement of IL-6 mRNA and its transcription factors c-Jun and p65 subunit of NF-κB
Total RNA was isolated from 5 × 105 Neuro-2a or macrophage-like J774.1 cells using the TRIzol reagent (Invitrogen), and cDNA was synthesized using the iScript cDNA synthesis kit (Bio-Rad, Hercules, CA, USA) according to the manufacturer's protocol. Real-time PCR and data analysis were performed in a total volume of 50 μL using 96-well microwell plates and an ABI PRISM 7700 sequence detector (Applied Biosynthesis, Foster City, CA, USA) with standardization of levels to the house keeping gene hypoxanthine phosphoribosyltransferase (HPRT). The primers used were as follows. IL-6; forward primer (5′-GTCACAGAAGGAGTGGCTA-3′) and reverse primer (5′-AGAGAACAACATAAGTCAGATACC-3′), and HPRT; forward primer (5′-CCTAAGATGAGCGCAAGTTGAA-3′) and reverse primer (5′-CCACAGGACTAGAACACCTGCTAA-3′). SYBR green real-time PCR was performed according to the manufacturer's protocol. Neuro-2a or J774.1 cells (2 × 106) cultured as described above were suspended in 0.1 mL of a buffer containing 20 mM Tris/HCl (pH 7.5), 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, 50 mM NaF, and 0.05 mM FUT-175, and lysed by repeated (three times) freezing and thawing with intermitted mechanical mixing. The supernatant was obtained by centrifugation at 17 000 g for 15 min and used as cell lysate. Levels of c-Jun and p65 subunit of NF-κB and their phosphorylated forms were determined using western blotting using 20 μg cell lysate protein.
Measurement of CRP mRNA and protein
Total RNA was isolated from 5 × 105 human FLC4 hepatocarcinoma cells, and synthesis of cDNA, real-time PCR, and data analysis were performed as described above. The primers used were as follows. CRP; forward primer (5′-CACCCAGAAAGGAGAAATGATG-3′) and reverse primer (5′-TGAGAAAGTGGAGGGACTGC-3′), and glyceraldehyde 3-phosphate dehydrogenase (GAPDH); forward primer (5′-GGTATCGTGGAAGGACTCATGAC-3′) and reverse primer (5′-ATGCCAGTGAGCTTCCCGTTCAGC-3′). The level of CRP in medium was measured by western blotting using 10 μL of the medium after 24 h culture of FLC4 cells.
Values are indicated as mean ± SE. Data were analyzed by Student's t-test, and a statistical difference was shown by probability values.
Increase in IL-6 and CRP in plasma together with increase in PC-Acro in PIT model mice
It was determined whether IL-6 and CRP in plasma increased together with PC-Acro after brain infarction using the PIT model mice. As shown in Fig. 1a, the size of infarction and the level of PC-Acro at the locus of the infarction increased gradually from 6 to 24 h after the onset of infarction in the PIT model mice. Then, the level of PC-Acro, IL-6 and CRP in plasma was measured (Fig. 1b). The level of PC-Acro and CRP in plasma was increased gradually at 6 to 24 h after the onset of brain infarction. However, the level of IL-6 in plasma was the highest at 6 h and gradually decreased until 24 h. The level of IL-6 at 24 h was still much higher compared to the level of IL-6 in plasma of control mice. These results suggest that acrolein may first induce IL-6 production, and then IL-6 induces CRP production.
Induction of IL-6 production by acrolein, and that of CRP production by IL-6 in cell culture systems
It has been reported that IL-6 is produced in astrocytes, monocytes, and endothelial cells (Kishimoto 1989). Thus, it was tested whether acrolein induces IL-6 production using mouse Neuro-2a cells, mouse macrophage-like J774.1 cells, and HUVEC. Acrolein (25 to 50 μM) inhibited cell growth of Neuro-2a cells, J774.1 cells, and HUVEC, and the degree of inhibition was slightly different in the order HUVEC ≈ J774.1 cells > Neuro-2a cells (Fig. 2). Then, the degree of stimulation of IL-6 production by acrolein was examined. As shown in Fig. 2, basal levels of IL-6 production (in the absence of acrolein) were different dependent on cell lines, and the ability to produce IL-6 was more than 15 times higher in Neuro-2a cells than J774.1 cells and HUVEC. Acrolein (25 and 50 μM) increased IL-6 production in all three cell lines, and the degree of stimulation by 50 μM acrolein at 24 h culture was in the order J774.1 (2.7-fold) > HUVEC (2.3-fold) > Neuro-2a (twofold).
It has been reported that CRP is mainly synthesized in liver and excreted into blood (Arnaud et al. 2005). Thus, the effects of acrolein and IL-6 on CRP production were examined using human hepatocarcinoma FLC4 cells. It was confirmed that 5 and 10 ng/mL IL-6 induced CRP production at the level of transcription (Arnaud et al. 2005), but 5–20 μM acrolein did not induce CRP production in the absence or presence of IL-6, which was measured by the CRP level in the medium (Fig. 3). The addition of more than 30 μM acrolein to the medium caused severe damage in human hepatocarcinoma FLC4 cells (data not shown). These findings support the idea that acrolein induces IL-6 production, and that IL-6, in turn, induces CRP production.
Mechanism of IL-6 induction by acrolein
To clarify how acrolein increases IL-6 production, the level of IL-6 mRNA was estimated using real-time PCR. As shown in Fig. 4b, IL-6 mRNA increased at the early period (3 to 6 h) after the addition of acrolein in both macrophage-like J774.1 cells and HUVEC, and it increased at 24 h in Neuro-2a cells in the presence of 25 and 50 μM acrolein. It has been reported that transcription of IL-6 mRNA is stimulated by AP-1 (c-Fos and phosphorylated c-Jun) and NF-κB (p50 and phosphorylated p65) (Ndlovu et al. 2009). As shown in Fig. 4a, there are two AP-1 and one NF-κB recognized sequences at the upstream of the transcriptional starting site of IL-6 gene. Thus, the level of c-Jun, phosphorylated c-Jun, NF-κB p65, and phosphorylated NF-κB p65 was measured using western blotting, at the period in which the maximal increase in IL-6 mRNA was observed in three types of cells. It was found that phosphorylation level of both c-Jun and NF-κB p65 was stimulated significantly by 50 μM acrolein in 3 h-treated J774.1 cells, 6 h-treated HUVEC and in 24 h-treated Neuro-2a cells (Fig. 4c). The results indicate that transcription of IL-6 mRNA is enhanced by acrolein through acrolein stimulation of phosphorylation of c-Jun and NF-κB p65.
Protection of acrolein toxicity by IL-6
It was then determined how IL-6 and CRP influence acrolein toxicity in cultured cells. It has been reported that CRP stimulates IL-6 production in neutrophils (Jones et al. 1999) and whole blood samples (Asanuma et al. 2008). Thus, we tested whether CRP stimulates IL-6 production in the presence and absence of acrolein. As shown in Fig. 5, CRP led to increased IL-6 production in macrophage-like J774.1 cells and HUVEC, an effect that was enhanced in the presence of 40 μM acrolein. The effect of CRP was more clearly obtained in macrophage-like J774.1 cells than HUVEC. CRP did not stimulate IL-6 production of Neuro-2a cells in the presence and absence of 40 μM acrolein (data not shown), probably due to the high expression of IL-6 in Neuro-2a cells.
It was then determined whether IL-6 protects acrolein toxicity or not. For this purpose, an antibody against IL-6 was added to the culture medium of Neuro-2a cells and HUVEC. Addition of antibody against IL-6 did not influence the cell growth of Neuro-2a cells and HUVEC without acrolein, but it inhibited the cell growth of Neuro-2a cells and HUVEC cultured in the presence of 20 μM acrolein (Fig. 6a). Increase in cell toxicity by an antibody against IL-6 was parallel with the decrease in glutathione content in cells (Fig. 6b), which is a major detoxicating compound for acrolein in cells (Yoshida et al. 2012). Similar results were obtained with human neuroblastoma SH-SY5Y cells and rat pheochromocytoma PC12 cells (data not shown). Antibody against IL-6 did not inhibit the cell growth of J774.1 cells in the presence of acrolein (data not shown), suggesting that expression of IL-6 receptor in J774.1 cells may be somehow reduced. The results indicate that IL-6 protects acrolein toxicity of several types of cells including Neuro-2a cells and HUVEC, and CRP indirectly protects acrolein toxicity against Neuro-2a cells through stimulation of IL-6 production from macrophage-like or endothelial cells such as J774.1 cells and HUVEC.
In this study, it was examined whether acrolein produced from spermine, which is released from damaged cells during brain infarction (Saiki et al. 2011), triggers the production of IL-6 and CRP. Using the PIT model mice and cultured cells, it became clear that acrolein induces IL-6 production, and then IL-6 induces CRP production. The results support the idea that detection of SBI with high sensitivity and specificity (Yoshida et al. 2010) is possible by measuring PC-Acro, IL-6, and CRP in plasma of SBI patients. The data also support a previous report that IL-6 and CRP increase in plasma of SBI patients (Hoshi et al. 2005). At present, the combined measurements of these three markers, that is, PC-Acro, IL-6, and CRP, are the only reliable biochemical markers for SBI. Therefore, these biochemical markers contribute greatly to identify brain infarction at the early period and for application of suitable therapies.
Our data confirmed previous reports that IL-6 protects brain tissues (or cells) against infarction (Fujita et al. 2009). This was actually because of the decrease in acrolein toxicity by IL-6 (see Fig. 6). With regard to CRP, it was previously noted that there is a link between plasma CRP and the degree of atherosclerosis (Sun et al. 2005). The results also suggest that decrease in plasma CRP may represent a therapeutic modality for the treatment of atherosclerosis (Sun et al. 2005). Our results also indicate that CRP increases IL-6 production in endothelial and macrophage-like cells especially in the presence of 40 μM acrolein, which is a new aspect of CRP function. Increased levels of IL-6, in turn, decreased the toxicity of acrolein, which is also a new finding in this study.
In the thrombosis model mice, the increase in IL-6 was observed at the early period of brain infarction. Thus, the increase in IL-6 production by macrophages and vein endothelial cells which occurred at the early period during cell culture in the presence of acrolein may be more strongly involved in the protection of tissue damage during brain infarction than the increase in IL-6 production by neurons which occurred at the relatively late period during cell culture in the presence of acrolein.
It is of interest to know how acrolein increases IL-6 production. Although acrolein is toxic to cells, it increased phosphorylation of c-Jun and NF-κB p65, which are involved in the transcription of IL-6 mRNA. Acrolein may increase an activity of a protein kinase which catalyzes the phosphorylation of c-Jun and NF-κB p65 through modification of cysteine or lysine residue. In this respect, we have recently reported that phosphorylation of JNKs (c-jun N-terminal kinases), which catalyze phosphorylation of c-jun and NF-κB p65, was increased in acrolein toxicity-decreasing Neuro-2a cells (Tomitori et al. 2012). Experiments are in progress to determine which kinase is actually activated by acrolein.
Decrease in acrolein toxicity by IL-6 was observed in several types of cells including Neuro-2a cells and HUVEC. Furthermore, the level of IL-6 production in Neuro-2a cells was much greater than that in macrophage-like J774.1 cells and HUVEC. These results suggest that IL-6 is strongly involved in detoxication of acrolein in neurons.
Measurement of PC-Acro together with IL-6 and CRP made it possible to identify SBI with high sensitivity and specificity. In this study, we tried to clarify how acrolein induces the synthesis of IL-6 and CRP. It was found that acrolein increased IL-6 synthesis and then IL-6 increased CRP synthesis at the level of transcription. It also became clear that IL-6 decreases acrolein toxicity in several types of cells, and CRP enhances IL-6 production in macrophage-like J774.1 cells and HUVEC (Fig. 7).
Acrolein was thought of as one of the toxic compounds produced from unsaturated fatty acids by reactive oxygen species (ROS) such as superoxide anion radical (O2̇−), hydrogen peroxide (H2O2), and hydroxyl radical (̇OH) (Uchida et al. 1998). However, we found that it was more effectively produced from polyamines (spermine and spermidine) (Tomitori et al. 2005). The proposed mechanism of inefficient formation of acrolein from arachidonic acid (Esterbauer et al. 1991) and experimental results on inefficient production of acrolein from arachidonic acid (Bradley et al. 2010) also support the idea that acrolein is preferentially produced from polyamines, rather than from unsaturated fatty acids by lipid peroxidation.
We thank Drs. A. J. Michael and K. Williams for their help in preparing the manuscript. This study was supported by a Grant for Chiba Serum Institute Memorial Foundation, Chiba, Japan, and a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology, Japan. The authors have no conflict of interest.