Glucose regulates expression of pro-inflammatory genes, IL-1β and IL-12, through a mechanism involving hexosamine biosynthesis pathway-dependent regulation of α-E catenin

Macrophages are highly sensitive to environmental stimuli. To acquire distinct functional phenotype against host cell infection, macrophages can be polarised into either pro-inflammatory cytokine secreting-classically activated (M1) or anti-inflammatory cytokine secreting-alternatively activated (M2) phenotype. A sustained or even short-term exposure to high glucose levels is known to induce M1 macrophage polarisation [1,2]. This has the potential to be relevant in disease such where derangements in glucoregulatory mechanisms play an important role. One such disease is COVID-19 as elevated glucose levels favour SARS-CoV2 infection and effects of glucose levels on macrophage function have been shown to play an important role in this [3,4]. Another example is Type 2 diabetes where hyperglycaemia leads to the accumulation of macrophages and other innate immune cells with M1 phenotype [5]. Moreover, it has been shown that M1 macrophages highly depend on glycolysis as the energy source [6,7], while M2 macrophages rely on fatty acid oxidation [8,9]. These studies suggest that glucose metabolism is involved in determining the phenotype of macrophages and thereby regulating inflammatory responses. However, the underlying mechanism is not fully understood.

There are several pathways involved in glucose-sensing mechanisms such as ATP-sensitive K⁺ channels [10], flux through the hexosamine biosynthesis (HB) pathway [11], AMP-activated protein kinase [12] and activated PKC [13]. Previously, we have shown that exposure of macrophage cell lines RAW264.7 and J774.2 to high levels of glucose increases the β-catenin protein level via the HB pathway [14]. β-catenin is a major component of adherens junction, which connects cadherin to α-catenin protein. Wnt/β-catenin pathway is known to promote M2 macrophage polarisation [15] and IL4-induced multinucleated giant cell formation [16]. β-catenin also stabilises α-catenin by preventing the proteasomal degradation [17]. Our previous studies found that glucose stimulation increases the α-E catenin level in rat pancreatic β-cell models (INS-1E and INS-832/3) [18].

In the current study, we sought to determine whether changes in glucose levels regulate α-E catenin levels in RAW 264.7 macrophages and to assess impacts this could have. We found that changes in glucose levels in RAW 264.7 rapidly alter the level of α-E catenin via a mechanism that requires activity of the HB pathway. We show changes in α-E catenin are associated with changes in expression of pro-inflammatory marker genes. Taken together, these data suggest that α-E catenin protein is regulated by the HB pathway and this contributes to changes in M1/M2 macrophage polarisation mediated by changes in glucose levels.

RAW264.7 cells were maintained in RPMI 1640 medium supplemented with 10% (v/v) newborn calf serum, 100 units/ml penicillin and 100 μg/ml streptomycin (all from Life Technologies). Cells were cultured on 12-well format and used for experiment when they were >90% confluent. Cells were serum starved overnight in RPMI medium in the presence of 0.5 mM glucose and treated with glucose, glucosamine or other inhibitors as indicated. All the reagents were purchased from Sigma–Aldrich.

After treatments, cells were rinsed twice with 0.5 ml of ice-cold 1× PBS and lysates were collected in buffer containing 20 mM Tris/HCl (pH 7.5), 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% Triton X-100, 2.5 mM sodium pyrophosphate, 1 mM β-glycerol phosphate, 1 mM vanadate, 100 mM NaF, 1 mM 4-(2-aminoethyl) benzenesulfonyl fluoride hydrochloride (AESBF), 4 µg/ml leupeptin, and 30 µM N-[N-(N-Acetyl-l-leucyl)-l-leucyl]-l-norleucine (ALLN), 4 µg/ml aprotinin, 0.4 µg/ml pepstatin. Cell lysates were centrifuged at 16100×g for 10 min, and supernatants were subjected to polyacrylamide gel electrophoresis for Western blotting.

After protein transfer, nitrocellulose membranes were incubated with primary antibodies against α-E catenin (1:1000; Cell Signaling Technology, catalogue number #240) and α-tubulin (1:20,000; Sigma–Aldrich, catalogue number T6074) at 4°C. After overnight incubation with primary antibodies, membranes were washed and incubated with respective secondary antibodies anti-rabbit IgG HRP (1:10000 Santa Cruz Biotechnology) or anti-mouse IgG HRP (1:20,000; Sigma–Aldrich) for 1 h at room temperature and developed with Clarity™ Western ECL substrate (Bio-Rad Laboratories).

α-E catenin siRNA was transfected to RAW 264.7 macrophages using Nepagene electroporator as per instruction manual. Briefly, 250 nM of siRNA was mixed with 1 × 10⁷ cells/100 μl of RAW 264.7 cells in Opti-Mem and electroporated using below conditions. Pouring pulse conditions are voltage at 175 V, pulse length 5 ms, pulse interval 50 ms, 2 pulses, 10% decay rate and + polarity. Transfer pulse conditions are 20 V voltage, 20 ms pulse length, 50 ms pulse interval, 5 pulse, 40% decay rate. Cells were used for experiments 48–72 h after transfection.

Seventy-two hours after siRNA transfection, RAW 264.7 cells were treated with 100 ng/ml LPS for 8 h in the presence of either 0.5 or 20 mM glucose in serum-free media. After treatments, cells were washed with 1× PBS and used for RNA isolation.

RNeasy Mini Kit (Qiagen) was used to extract total RNA from cells. Samples were then treated with DNase I (Life Technologies) to eliminate genomic DNA. cDNA synthesis (reverse transcription) was performed by using the qScript cDNA Super Mix kit (Dnature) with the same amount of RNA added for all samples. The cDNA synthesis conditions were as follows: 5 min at 25°C > 30 min at 42°C > 5 min at 85°C > hold at 4°C. cDNA samples were loaded on to a 394-well PCR plate prior to real-time PCR. In addition, DNase-free water was included as a non-template control. Each reaction was consisted of 5 μl qScript cDNA Super Mix, 5 ng cDNA (equivalent to RNA added) and 400 nM primers and water (up to 10 μl). PowerUp SYBR Green Master Mix (Applied Biosystems) was used for qPCR. The qPCR conditions were as follows: 2 min at 50°C > 10 min at 95°C > 40 cycles (15 s at 95°C, 1 min at 60°C) > 15 s at 95°C > 1 min at 60°C > 15 s at 95°C. The double delta C_t analysis was performed to calculate the gene expression fold change in relation to the control. A single distinct peak in melt curve was checked to validate the qPCR in each well.

Results are presented as means ± S.E.M. with the number of experiments indicated in the legend. Statistical significance was assessed using Student’s t test or two-way ANOVA as indicated in figure legends. Statistical significance is displayed as *P<0.05 or **P<0.01. Statistical analyses were performed using statistical software package GraphPad Prism 6.0 (GraphPad Software Inc.).

We found that high levels of glucose cause an increase in α-E catenin protein levels in the RAW 264.7 macrophage model (Figure 1A,B), consistent with our previous findings in rat pancreatic β-cells [18]. As we have previously found that β-catenin levels increased with glucose stimulation in macrophage cell lines through the HB pathway [14], we speculated that the same pathway might account for the glucose-induced up-regulation of α-E catenin. To investigate this possibility, RAW 264.7 cells were treated with glucosamine. Glucosamine is converted into glucosamine-6-phosphate (GlucN-6-P), bypassing the first three steps required for glucose in the HB pathway (Figure 1C). Glucosamine increased α-E catenin levels in macrophages in a similar manner to that seen with high glucose (Figure 2A,B). The conversion of fructose-6-phosphate into GlucN-6-P (the rate limiting-step of the HB pathway) is regulated by glutamine:fructose-6-phosphate-amidotransferase (GFAT) enzyme, which requires glutamine as a co-substrate [19]. Therefore, to further confirm that the HB pathway is responsible for modulating the level of α-E catenin, we treated RAW 264.7 cells with high glucose in the absence of glutamine and observed no significant change in α-E catenin level (Figure 2C,D). In line with this finding, when GFAT enzyme was inhibited with azaserine, the glucose effect on α-E catenin levels were attenuated (Figure 3A,B). The HB pathway mediates the glycosylation of proteins, which is important for protein stability, protein activity, cell–cell communication and signal transduction [20,21]. Next, we treated cells with tunicamycin, which is an inhibitor of N-linked glycosylation. The glucose effect on α-E catenin was attenuated in the presence of tunicamycin (Figure 3C,D), indicating that the regulation of α-E catenin level relies on N-linked glycosylation. These findings suggest that α-E catenin level in RAW264.7 is regulated via the HB pathway involving N-linked glycosylation.

We next investigated the role of α-E catenin in the function of the RAW cells using siRNA to reduce α-E catenin. We electroporated RAW 264.7 cells with an α-E catenin siRNA, which led to at least 50% reduction in levels of α-E catenin (Figure 4A). We then treated cells with either low (0.5 mM glucose) or high glucose (20 mM glucose) in the presence or absence of LPS. We found that decreased expression of Ctnna1 blunts glucose-induced up-regulation of IL-1β under the LPS-stimulated condition (Figure 4B). A more profound effect was observed on IL-12 expression where we observed the decreased expression in the α-E catenin knockdown samples and a loss of the glucose-mediated reduction in LPS-induced IL-12 expression (Figure 4C). Under LPS-stimulated conditions, we did not observe any effect of α-E catenin knockdown on TNF-α at either low or high glucose conditions (Figure 4D).

Macrophages have phenotypic plasticity which is an important for their function in inflammation. Polarisation towards the M1 state can be induced by the TLR agonist LPS [22] and here we see this increase levels of M1 polarisation markers, namely IL-1β and IL-12 and TNF-α. High glucose levels favour macrophage polarisation towards the M1 phenotype, characterised by the secretion of pro-inflammatory cytokines which induces inflammation of surrounding environment. So we were interested in how changes in α-catenin could interact with glucoses effects on this. Therefore, most novel finding of the current study is identification of a previously unreported role of α-E catenin in glucose-mediated effects on macrophage polarisation. This shown by the impacts of reducing α-E-catenin on glucose modulatory effects on pro-inflammatory genes IL-1β and IL-12. This implicates α-E catenin as a new component of mechanisms involved in switching macrophages between M1 and M2 states. We find that that α-E catenin protein level is increased upon glucose stimulation using the RAW 264.7 cell line as a macrophage cell model via the HB pathway which could thus contribute to the changes in macrophage function known to be brought on by changes in glucose levels [1,2] .

One outcome of the HB pathway is protein glycosylation and a lack of proper protein glycosylation results in accumulation of misfolded proteins in the endoplasmic reticulum (ER) which results in ER stress [23]. Thus, the finding that the glycosylation inhibitor tunicamycin blocks glucose-induced increases in α-E catenin in our study suggests normal ER function would be required for this mechanism of α-E catenin regulation. It is notable that a separate proteomic approach has also shown that ER stress also reduces both α-catenin and β-catenin protein levels in HeLa cells and that ER stress was reducing catenin levels by increasing rates of proteasomal degradation [23]. Proteasomal turnover of α and β-catenin is associated with increased CK1 and GSK3-mediated phosphorylation of β-catenin at Ser³⁷/Thr⁴¹ [24,25]. Tunicamycin regulates GSK3 in the proteomics study [23] and studies in other cell types also show that tunicamycin activates GSK3 which would be consistent with reductions in α and β-catenin [26–29]. However, we do not see any changes on GSK3β levels or phosphorylation status despite seeing a reduction in β-catenin Ser³⁷/Thr⁴¹ phosphorylation [14]. CK1 activity has also been reported to a change in response to changes in glucose levels in some cell types [30] and tunicamycin may increase CK1 activity [31], although it is not known if this occurs in macrophages. Therefore, the mechanisms linking changes in glucose-induced glycosylation to changes in α-E catenin levels remain to be elucidated.

In a similar study, it has been shown that high glucose levels decrease the mRNA and protein levels of silent mating type information regulation 2 homologue 1 (SIRT1) and the modulation of SIRT1 by inhibitors or siRNA further increases the glucose-induced increase in pro-inflammatory markers IL-1β and TNF-α [1]. A similar effect of glucose on pro-inflammatory cytokines to what we have seen in RAW264.7 cell line has been observed in BMDM only after long-term (7 days) high glucose treatment [2]. Here they have shown that long-term high glucose treatment increases IL-1β, but decreases IL-12 expression, however short-term glucose treatment only affects IL-1β expression [2]. These glucose-induced changes in IL-1β could have impacts on metabolism as elevation in IL-1β have been shown to have detrimental impacts on regulation of metabolism [32]. These impacts on metabolism may be part of a physiologically relevant cycle involved in regulating glucose metabolism as acute increases in circulating levels of IL-1β have been observed postprandially [32,33]. Of relevance to our work is the evidence that these changes are due to glucose-induced production of IL-1β in macrophages in the gut.

In addition, our findings could also have implications for pathology associated with Type 2 diabetes as in our previous work, we also have demonstrated that α-catenin inhibits insulin secretion in pancreatic β-cells [18]. Type 2 diabetes is characterised by β-cell failure [34] in which chronic inflammation involving macrophages has been implicated [35,36]. Thus, it will be worth investigating how glucose-induced changes in α-E catenin in both β-cells and macrophages might combine in promoting progression of this disease.

No large datasets are associated with the present study. All raw data for Western blots and qPCR are available upon request.

The authors declare that there are no competing interests associated with the manuscript.

This work was supported by the Health Research Council of New Zealand-Te Kaunihera Rangahou Hauora of Aotearoa.

Open access for this article was enabled by the participation of University Of Auckland in an all-inclusive Read & Publish pilot with Portland Press and the Biochemical Society under a transformative agreement with CAUL.

Waruni C. Dissanayake: Conceptualization, Formal analysis, Investigation, Methodology, Writing—original draft, Writing—review and editing. Jin Kyo Oh: Formal analysis, Investigation, Methodology. Brie Sorrenson: Conceptualization, Supervision, Investigation, Writing—original draft. Peter R. Shepherd: Conceptualization, Resources, Supervision, Funding acquisition, Writing—original draft, Writing—review and editing.

BMDM

bone marrow derived macrophages

CK1

casein kinase 1

ER

endoplasmic reticulum

GFAT

glutamine:fructose-6-phosphate-amidotransferase

GlucN-6-P

glucosamine-6-phosphate

GSK3

glycogen synthase kinase 3

HB

hexosamine biosynthesis

IL

Interlukin

LPS

lipopolysaccharides

PKC

protein Kinase C