Honey polyphenolic fraction inhibits cyclooxygenase-2 expression via upregulation of microRNA-26a-5p expression in pancreatic islets

Abstract

Objectives

Honey total polyphenolic fraction (HTPF) is reported to have anti-disease potential, however the role of HTPF in the regulation of microRNAs (miRNAs) has never been investigated. This study was undertaken to investigate the potential of HTPF against inflammation via regulation of miRNAs in pancreatic islets of Langerhans.

Methods

Pancreatic islets were isolated from C57BL/6 mice and HTPF was purified from honey. Bioinformatics algorithms were used to determine miRNA target genes. Expression of miRNA and mRNA was determined using their specific taqman assays. Pairing between miRNA and 3^′ untranslated region (3^′UTR) of mRNA was confirmed using luciferase reporter clone containing the 3^′UTR of mRNA sequences and results were verified by transfection of mouse pancreatic β-cell line Min6 with miRNA inhibitors.

Results

The data showed that mmu-miR-26a-5p is a direct regulator of cyclooxygenase-2 (COX-2) expression and HTPF inhibits COX-2 expression or prostaglandin E₂ (PGE₂) production via up-regulating mmu-miR-26a-5p expression. Transfection of islets with anti-miR-26a-5p significantly enhanced COX-2 expression and PGE₂ production (p < .01), while HTPF treatment significantly inhibited anti-miR-26a-5p transfection-induced COX-2 expression or PGE₂ production (p < .05). These findings were further verified in pancreatic β-cells Min6. Moreover, the data also determined that HTPF also inhibits glucose-induced nuclear transcription factor (NF)-κB activity.

Conclusion

HTPF suppresses glucose-induced PGE₂ production and activation of NF-κB via negative regulation of COX-2 and mmu-miR26a-5p. These novel pharmacological actions of HTPF on glucose-stimulated pancreatic islets provide new suggestions that HTPF or HTPF-derived compounds inhibit glucose induced inflammation in pancreas by up-regulating the expression of microRNAs.

Keywords

Pancreatic islets Min6 cells honey polyphenols mm-miR-26a-5p COX-2

Introduction

Type 1 diabetes (T1D) is an organ-specific disorder caused by autoimmune destruction insulin-producing beta cells in Langerhans of the pancreatic islets.¹ Due to ethical and practical issues, the current understanding of destructiveness of pancreatic islets and the process to understand diabetic pancreatic pathology is limited. Small number of studies conducted on the autopsies and biopsies from T1D patients, indicate that beta-cell destruction and islet infiltration is performed through lymphocytes, macrophages, natural killers and dendritic cells and by the overexpression of a few immune system associated genes in T1D patients.^1,2 To explain the pathology more clearly and move forward in our understanding of the process leading to islets destruction is to have detailed information occurring at the molecular level. However, it is believed that the underlying the destruction process of pancreatic islets is multifactorial and its etiology still needs to be fully studied.³ Molecular evidences clearly show that the onset of T1D and the destruction process of islets is related to the overproduction of several pro-inflammatory cytokines and inflammatory mediators including cyclooxygenase-2 (COX-2, also known as PTGS2).^4,5 It is important to know that usually COX-2 is expressed at low levels in most of the cells and tissues, but it is well known to induce a wide range of inflammatory stimuli such as cytokines, chemokine and other inflammatory mediators.^4–6 In diabetic pancreatic islets, COX-2 is expressed constitutively and dominantly and its excessive induction is well known to increase the production of prostaglandin E2 (PGE2),^7,8 therefore, along with COX-2, PGE2 is considered as one of a possible therapeutic targets for treatment of T1D. MicroRNAs (miRNAs) are small non-coding RNAs known to regulate mRNA genes by pairing with their complementary sequences at 3’ untranslated regions (3’UTR) at the post-transcriptional level and it is speculated that miRNAs regulate more than 70% of all human protein-coding genes and participate in the regulation of almost all cellular events studied so far.^9–11 Now it is well established that miRNAs play an important role in the pathogenesis of T1D,^12–14 but their role in the destruction of diabetic islets remains largely unexplored.

Honey has been used as a traditional sweetener for many decades but its potential as a therapeutics has also been tested in recent years.^15,16 Numerous studies demonstrated its therapeutic values as it is a rich source of antioxidants, polyphenols and flavonoids.¹⁷ Studies showed that it has potential to improve lipid profile and reduces the risk of coronary heart diseases.^17–19 It has potential also as a tropical medicine as it shows healing of wounds, burns, diabetic foot, ulcers and also healing of psoriatic and herpes lesions.^19–21 Interestingly, honey and its polyphenolic contents have been reported to possess antioxidant, anti-inflammatory, antimicrobial and anti-cancer activities.^22–27 Most importantly, it is also now well documented that the glucose induced induction of COX-2 or PGE₂ mediated inflammatory activity in human islets suggested one of the major routes through which hyperglycemia contributes to the destruction of pancreatic beta cells.²⁸ In view of these, we hypothesized that the polyphenolic fraction of honey inhibits the glucose-induced COX-2 or PGE₂ mediated inflammatory activities by regulating the expression of microRNAs in pancreas, the main organ affected by hyperglycemia. To test this hypothesis, pancreatic islets were isolated from normal C57BL/6 black mice and treated with honey total polyphenolic fraction (HTPF) and glucose. Bioinformatics algorithms were used to screen 3’UTR of COX-2 mRNA for the identification of seed matched sequences complementary to miRNAs sequences. The role of specific miRNA in regulation of inflammatory activity was determined by transfection of islets with specific miRNA inhibitor. Our novel data showed HTPF inhibits the glucose-induced COX-2 expression and PGE₂ production via upregulating the expression of microRNA mmu-miR26a-5p in mouse pancreatic islets. These investigations further verified in pancreatic beta cell lines Min-6 using transfection with reporter constructs and anti-miRNAs. These are novel findings have never been investigated before and may be of value in the design of novel therapies for the reducing inflammation or treatment of hyperglycemic patients.

Methods

Extraction of honey polyphenols

Natural honey was collected from Egypt and total polyphenols were extracted as described previously.²⁹ Briefly, the solid-phase extraction of polyphenols was performed using Amberlite XAD-2 column (catalog # 10357, Sigma-Aldrich Inc., St. Louis, MO., USA). The sugar contents from honey were removed by an acidification of the column using hydrochloric acid solution and the polyphenolic fraction from the column was eluted using methanol solution and the collected fraction of polyphenols was then dried using a Rotavapor and re-suspected in minimum volumes of ethanol and extensively dialyzed against PBS (10 mm, pH 7.4) and filtered with 0.2 μm PES membrane (catalog # 720-1320, Thermo Scientific™ Nalgene™ syringe filters) and sterilized fraction was used for culture. The purified phenolic content was further reconfirmed by Folin-Ciocalteau method as described previously.³⁰

Pancreatic islets isolation and pancreatic β-cell line Min6 culture

With the ‘Institutional Review Board’ Ethical approval (IRB approval #19-09-02) from the Deanship of Scientific Research, Qassim University, KSA, mouse pancreatic islets were isolated in accordance of the World Medical Association Guidelines for use of laboratory Animals. Islets from 3-6 months old C57BL/6 mice were isolated as described by the published method.¹⁴ Briefly, the mice were euthanized, and collagenase P solution, 1 mg/mL (catalog # 11215809103, Roche, Indianapolis, IN, USA) was injected into the bile duct to inflate the pancreas, followed by removal and incubation at 37°C for digestion until the pancreas formed a milky solution with only a few clumps. After digestion, islets were purified in density gradient Histopaque solution (Sigma-Aldrich). Islets were manually selected, washed in HEPES buffer, and cultured overnight in RPMI 1640 complete medium supplemented with 15% heat-inactivated fetal bovine serum (FBS). Whereas, the mouse pancreatic beta cell line Min6³¹ was cultured in DMEM (catalog # SLM-120-B, Millipore, Darmstadt, Germany) supplemented with 15% FBS as described previously.³²

Treatment of pancreatic islets and pancreatic β-cell line Min6 with honey polyphenolic fraction and glucose

Purified islets were plated in 24-well plates (30-40 islets per well), allowed 24 h recovery after purification, washed twice and transferred to complete RPMI 1640 complete medium to stabilize basal gene expression and serum-starved for 12 h as described previously.¹⁴ Whereas pancreatic Min6 cells (70–80% confluent) were also serum-starved for 12 h in DMEM culture medium as described previously.^32,33 Both serum-starved islets and Min6 cells were washed twice with PBS and pretreated with honey polyphenolic fraction (HTPF, 20–40 μg/mL) for 2 h and stimulated with glucose (0–25 mm) for 8–24 h in serum free RPMI 1640 medium and DMEM culture medium, respectively.

Transfection of islets and Min6 cells with miRNA inhibitors

Freshly isolated mouse pancreatic islets or Min6 cells were transfected with anti-miRNAs (100 nM; Qiagen, Hilden, Germany or Ambion, Foster City, CA, USA) in replicates using HiPerfect Transfection Reagent (Qiagen) without disrupting of islets structure in accordance to the manufacturers’ instructions with some modifications.^14,34 Briefly, freshly isolated intact islets (∼50) or Min6 cells (70–80% confluent) were coated in culture plates and allowed to recover for 24–72 h and were transfected with the miRNA inhibitors, or miRNA controls. After 72 h post-transfection, islets or Min6 cells were pretreated with honey polyphenolic fraction (20–40 μg/mL) for 2 h or stimulated with glucose (0–25 mm) for 8–24 h to analyze the expression of miRNA, mRNA or protein.

Luciferase reporter assays

Luciferase reporter assays were performed using luciferase assay kit (Promega, Madison, Wisconsin, USA) as described previously.¹⁰ Briefly, a luciferase reporter vector containing the entire 3′untranslated region of PTGS2 (or COX-2) mRNA (# ENST00000367468) and empty vector containing only luciferase gene and its constitutively active promoter (SwitchGear Genomics, Menlo Park, California, USA) were used in the reporter assays. Pancreatic beta cell line Min6 were co-transfected with 100 ng reporter plasmid, 50 and 100 nM selected anti-miRNAs (Qiagen) or negative control miRNAs (Qiagen) using HiPerfect Transfection Reagent (Qiagen) in a 96-well plate. Min6 cells were cultured 24 h after transfection and treated with HTPF (20–40 μg/mL) for 2 h and dual luciferase activity was measured. Each experiment was performed in triplicate and repeated three times using different sets of Min6 cells.

MicroRNA isolation, first strand cDNA synthesis and TaqMan assays

Total RNA containing miRNA fraction was isolated from non-transfected or transfected pancreatic islets and Min6 cells using mirVana miRNA isolation kit (Ambion, Foster City, CA, USA) as described by the manufacturers and published procedure.^35,36 Total RNA (0.5–1.0 μg) was reverse-transcribed using Super Script First Strand cDNA synthesis kit (Applied Biosystems) and desired mRNAs and miRNAs expressions were quantified by TaqMan Gene Expression Assays (Applied Biosystems) using the Step One Real Time PCR System (Applied Biosystems). GAPDH/snoRNA202 expression was used as an endogenous control. Relative expression levels were analyzed using double delta CT method as described previously.³⁷

Western immunoblotting for protein expression

Protein expression in non-transfected and transfected pancreatic islets or Min6 cells was determined by western blotting as described previously.^38,39 Briefly, the total cell lysates were prepared and were loaded on 10% resolving SDS-PAGE with 4% stacking and blots were probed with primary antibodies specific for the COX-2 (catalog # D5H5; Cell Signaling Technology Beverley, MA, USA) and β-actin (catalog # 8457; Cell Signaling Technology).

Prostaglandin E₂ ELISA

Release of prostaglandin E2 (PGE₂) in the culture of non-transfected or transfected pancreatic islets and Min6 cells was determined using PGE₂-specifc ELISA kit (Cayman Chemicals, Ann Arbor, MI, USA) according to the instructions of the manufacturer.

Nuclear factor-kappa B activity assays

Nuclear levels of NF-κBp65 and NF-κBp50 in treated or untreated Min6 cells were determined in the nuclear extracts using a highly sensitive Transcription Factor ELISA Kit according to the instructions of the manufacturer (Abcam, MA, USA).

Statements

The study was performed in accordance of ARRIVE guidelines (http://www.nc3rs.org.uk/arrive-guidelines). Moreover, we also confirm that all methods were conducted in accordance with our institutional guidelines and regulations for the use of animals.

Statistical analysis

The data are expressed as the mean±SD unless stated otherwise and were collected from at least three independent experiments. One-way ANOVA of variance followed by Tukey-Kramer multiple comparisons test, and two-way ANOVA of variance followed by Bonferroni post hoc comparisons test. p < .05 was considered significant. All statistical analysis was carried out using Graph Pad Prism version 5.0 (Graph Pad Software Inc., San Diego, CA, USA) and Origin software package version 6.1 (Origin Lab Corporation, Northampton, MA, USA).

Results

Bioinformatics algorithms for the detection of conserved seed matched sequences of microRNAs in 3^′UTR of COX-2 (or Ptgs2) mRNA

First we scanned 3’-untranslated (3’UTR) of mouse COX-2 mRNA for the determination of conserved seed matched sequences of microRNAs using TargetScan bioinformatics algorithm. TargetScan algorithm showed that the 3’UTR of mouse COX-2 (ENST00000367468.4) contains 2487 nucleotide bases and has seven conserved sites for different microRNAs namely miR-26a-5p, miR-103-3p, miR-107-3p, miR-101-3p, miR-143-3p and miR-144-3p (Figure 1(a)). The details of the predicted consequential pairing of target region of 3^′UTR of COX-2 with these highly conserved miRNAs are given in Figure 1(b). Interestingly, results also analyzed that the microRNAs miR-26-5p, miR-103-3p/miR-107-3p, miR-101a-3p, miR-143-3p and miR-144-3p seed sequence conserved for mouse 3’UTR of COX-2 mRNA is also conserved for other species such as human, chimp, rhesus, squirrel, rat, rabbit, pig, cow, cat, dog and brown bat (Figure 1(c)). In order to validate these results, we further analyzed the whole 3’UTR of COX-2 mRNA for the sequence complementary to mmu-miR26a-5p by another computer based bioinformatics approach miRBase, which uses different algorithm methods. Using miRBase algorithm, the results showed that the stem-loop sequence of mmu-miR26a-5p showed similar complementary sequences in the same two positions of 3’UTR of mouse COX-2 mRNA (Supplemental data Figure 2). Furthermore, another bioinformatics approach miRanda database (miRDB) also confirmed the occurrence of complementary sequences of mmu-miR26a-5p in the 3’UTR of COX-2 mRNA (Supplemental data Figure 3). All three algorithms TargetScan, miRBase and miRanda databases identified sequences conserved in the 3^′UTR of COX-2 are complementary to microRNA-26a-5p seed sequence.

Figure 1.

TargetScan bioinformatics algorithm for the prediction of seed matched sequence of conserved mouse microRNAs in 3^′UTR of COX-2 (or PTGS2) mRNA. (a) 3^′UTR of mouse PTGS2 mRNA shows conserved sites for miR-26a-5p, miR-103-3p, miR-107-3p, miR-101-3p, miR-143-3p and miR-144-3p. (b) Duplex prediction of mmu-miR-26a-5p, mmu-miR-103-3p, mmu-miR-107-3p, mmu-miR-101a-3p.1, mmu-miR-143-3p or mmu-miR-144-3p with the seed sequence in the 3^′UTR of mouse PTGS2. (c) Cross-species conservation of the miR-26a-5p, miR-103-3p, miR-107-3p, miR-101-3p, miR-143-3p or miR-144-3p seed sequence in the 3^′UTR of PTGS2 mRNA. The sequences showing in red boarders are the locations of potential seed-matched sequence for PTGS2 mRNA.

Expression of COX-2, mmu-miR-26a-5p, mmu-miR-103-3p, mmu-miR-107-3p, mmu-miR-101a-3p, mmu-miR-143-3p and mmu-miR-144-3p in un-stimulated and glucose-stimulated mouse pancreatic islets

Treatment of primary islets with 25 mm glucose for 24 h resulted significant increased expression of COX-2 mRNA and protein and PGE₂ production (Figure 2(a)–(c); p < .01), whereas treatment with 2.0 mm glucose was ineffective (p > .05). Interestingly, primary islets treated with 25 mm glucose for 24 h also showed a significant down-regulation of mmu-miR-26a-5p expression (4.4 folds; p < .001), whereas treatment of islets with 2.0 mm glucose for 24 h caused only 0.87 folds change in the expression of mmu-miR-26a-5p (p > .05). These results indicate an inverse correlation between the expression of COX-2 or PGE₂ production with mmu-miR-26a-5p, when islets were treated with 25 mm glucose (Figure 2). In addition, we also determined the expression of other miRNAs conserved in 3’UTR of COX-2 such as mmu-miR-103-3p, mmu-miR-107-3p, mmu-miR-101a-3p, mmu-miR-143-3p and mmu-miR-144-3p in un-stimulated and glucose-stimulated mouse pancreatic islets but no significant change in their expressions was observed after stimulation glucose (2.0 or 25 mm) (Figure 2(d); p > .05).

Figure 2.

Expression of COX-2 and expression of microRNAs conserved in 3^′UTR of COX-2 mRNA in glucose stimulated mouse pancreatic islets. (a) COX-2 mRNA expression in primary mouse pancreatic islets stimulated with glucose (2.0 or 25 mm) for 24 h. Expression of COX-2 was determined by TaqMan assay using COX-2 specific primer sequence. ^#p < .0001 versus untreated or 2.0 mm glucose treated islets. (b) Protein expression of COX-2 in primary mouse pancreatic islets stimulated with glucose (2.0 or 25 mm) for 24 h. COX-2 protein in pancreatic islets was determined by COX-2 specific primary antibodies using western immunoblotting. Band images were digitally captured and the band intensities (pixels/band) were obtained using the Un-Scan-It software and are expressed in average pixels. ^*p < .0001 versus untreated or 2.0 mm glucose treated islets. (c) Levels of PGE₂ in the culture medium of mouse pancreatic islets treated with glucose (2.0 or 25 mm) for 24 h ^*p < .0001 versus untreated or 2.0 mm glucose treated islets. (d) Expression levels of mmu-miR-26a-5p, mmu-miR-103-3p, mmu-miR-107-3p, mmu-miR-101a-3p, mmu-miR-143-3p and mmu-miR-144-3p in glucose-stimulated mouse pancreatic islets were determined by TaqMan assays. ^#p < .001. (e) Regulation of COX-2 expression by mmu-miR-26a-5p via 3^′UTR of COX-2 mRNA in pancreatic beta cell line Min6. Dual luciferase assays in Min6 cells co-transfected with the reporter vector and anti-miR-26a-5p. ^#p < .0001 versus Min6 cells transfected with Empty 3’UTR vector containing only luciferase gene and its constitutively active promoter; ^*p < .05 versus #. Unstimulated pancreatic islets or Min6 cells were used as controls and expression of snoRNA202/GAPDH was used as an endogenous control. Each experiment was performed in triplicate and repeated five times using different sets of mouse pancreatic islets or Min6 cells.

Validation of mmu-miR-26a-5p binding in the 3^′UTR of COX-2 mRNA

To test whether the 3^′UTR of COX-2 is a functional target of the bioinformatics predicted mmu-miRNA-26a-5p, we used a luciferase reporter clone containing entire 3^′UTR of COX-2 mRNA including the predicted the seed-matched binding sites and luciferase assay was performed in mouse pancreatic beta cell line Min6. As shown in Figure 2(d) and (a) significant concentration-dependent increase in luciferase activity in Min6 pancreatic beta cells co-transfected with COX-2 reporter and anti-miR-26a-5p (p < .05). Transfected Min6 cells with 3^′UTR of COX-2 mRNA luciferase reporter vector alone and transfected Min6 cells with empty vector containing only luciferase active promoter gene were used as positive and negative controls of the reporter assays, respectively. These results clearly demonstrated that mmu-miR-26a-5p binds the seed matched sequence present in the 3^′UTR of mouse COX-2 mRNA.

Honey TPF up-regulates microRNA-26a-5p expression and down-regulates COX-2 expressions in pancreatic islets

Purified islets after 72 h culture in RPMI 1640 complete medium were pretreated with honey total phenolic fraction (Honey TPF, 20–40 μg/mL) for 2 h and stimulated with glucose (25 mm) for 8–24 h and cytotoxicity of islets was examined using the CytoTox-Glo™, Cytotoxicity Assay Kit (Promega, Madison, WI, USA). No cytotoxic effect of honey TPF and glucose was noticed on pancreatic primary islets (data not given). The expression level of mature mmu-miR-26a-5p was quantified by TaqMan Assays using quantitative RT-PCR method, and values were normalized with endogenous control snoRNA202. Stimulation of pancreatic islets with glucose (25 mm) alone showed a significant down-regulation of mmu-miR-26a-5p expression (p < .001). Importantly, treatment of islets with honey TPF markedly up-regulates glucose-induced mmu-miR-26a-5p expression in a dose-dependent manner (Figure 3(a); p < .05). However, treatment of pancreatic islets with honey TPF alone had no effect on mmu-miR-26a-5p expression (p > .05). Interestingly these results were reversed when COX-2 mRNA expression was determined in the same pancreatic islets treated with honey TPF and glucose. As shown in Figure 3(b), treatment of purified islets with glucose protein significantly enhanced the mRNA expression of COX-2 (p < .001). However, pretreatment of islets with honey TPF inhibited glucose-stimulated COX-2 mRNA in a dose-dependent manner (p < .05). To investigate whether suppression of COX-2 mRNA expression also affected at protein level, pancreatic islets lysates were analyzed by Western immunoblotting using COX-2-specific primary antibodies (catalog # D5H5; Cell Signaling Technology, MA, USA) and the protein loading of cell lysates on SDS-PAGE were analyzed using β-actin-specific antibodies (catalog # 8457; Cell Signaling Technology). Results showed that pretreatment of islets with honey TPF significantly inhibited the glucose-induced protein expression of COX-2 in a dose-dependent manner (Figure 3(c)). These results were further validated quantifying the band intensities using the UN-SCAN-IT software (p < .05; Figure 3(c)).

Figure 3.

Honey polyphenolic fraction up-regulates mmu-miR-26a-5p expression and down-regulates COX-2 expression in glucose-stimulated mouse pancreatic islets. (a) Effect of honey total polyphenolic fraction (Honey TRF) on glucose-induced down-regulation of mmu-miR-26a-5p in mouse pancreatic islets quantified by TaqMan assays using real-time PCR. *p < .001 versus untreated islets; #p < .01 versus islets stimulated with glucose alone; @p > .05 versus untreated islets (b) Effect of Honey TRF on glucose-induced up-regulation of COX-2 mRNA expression determined by TaqMan assay using COX-2 specific primer sequence. *p < .0001 versus untreated islets; #p < .05 versus glucose-stimualted islets; @p > .05 versus untreated islets. (c) Effect of Honey TPF on glucose-induced up-regulation of COX-2 protein in pancreatic islets lysates determined by COX-2 specific primary antibodies using western immunoblotting. *p < .001 versus untreated islets; #p < .05 versus glucose-stimualted islets. (d) Effect of Honey TPF on glucose-induced PGE2 production in the culture medium of mouse pancreatic islets. *p < .0001 versus untreated islets; #p < .05 versus islets stimulated with glucose alone; @p < .01 versus islets stimulated with glucose alone; $p > .05 versus untreated islets. Unstimulated pancreatic islets were used as controls and expression of snoRNA202/GAPDH was used as an endogenous control. Each experiment was performed in triplicate and repeated five times using different sets of mouse pancreatic islets.

Involvement of Honey TPF in the inverse co-regulation of mmu-miR-26a-5p and COX-2 was further verified by transfection of mouse pancreatic islets with anti-miR-26a-5p used as miRNA inhibitor. Mouse pancreatic islets were transfected with 100 nM anti-miR-26a-5p or anti-miR-control and were then treated with Honey TPF (20–40 μg/ml). As shown in Figure 4(a), transfection of islets with anti-miR-26a-5p significantly suppressed the mmu-miR-26a-5p expression compared with non-transfected islets (p < .01). Importantly, treatment of these transfected islets with Honey TPF significantly up-regulated expression of mmu-miR-26a-5p with the increase of doses of honey TPF (p < .05). In the same experimental settings, mouse pancreatic islets were also transfected with anti-miR-control and followed with the treatment of honey TPF (20–40 μg/ml). As shown in Figure 4(b), transfection of islets with anti-miR-control showed no alterations in the expression of mmu-miR-26a-5p as compared with non-transfected pancreatic islets (p > .05). Moreover, treatment of these anti-miR-control-transfected islets with honey TPF also showed no effect in the alteration of mmu-miR-26a-5p expression (Figure 4(b); p > .05). In order to validate the negative correlation between mmu-miR-26a-5p and COX-2, the COX-2 mRNA and protein levels were determined in the same sets of pancreatic islets transfected with anti-miR-26a-5p or anti-miR-control. As expected, anti-miR-26a-5p-transfected islets showed significant up-regulation of COX-2 mRNA as compared with non-transfected islets (p < .01). Interestingly, treatment of the same anti-miR-26a-5p-transfected islets with honey TPF significantly reduced the effect of anti-miR-26a-5p transfection on COX-2 mRNA expression (Figure 3(c); p < .05) and protein expression (Figure 4(e)). Whereas, transfection of islets with anti-miR-control followed by honey TPF showed no effect on the expression of COX-2 mRNA (Figure 4(d); p > .05) and COX-2 protein (Figure 4(f); p > .05).

Figure 4.

Honey polyphenolic fraction up-regulates the mmu-miR-26a-5p expression and down-regulates COX-2 expression in anti-miR-26a-5p-transfected mouse pancreatic islets. (a) Effect of honey total polyphenolic fraction (Honey TRF) on mmu-miR-26a-5p expression in pancreatic islets transfected with anti-miR-26a-5p. ^*p < .001 versus untransfected treated islets; ^#p < .01 versus islets transfected with anti-miR-26a-5p alone; ^@p > .05 versus untransfected islets (b) Effect of Honey TRF on mmu-miR-26a-5p expression in pancreatic islets transfected with anti-miR-control. (c) Effect of Honey TRF on COX-2 mRNA expression in pancreatic islets transfected with anti-miR-26a-5p. ^*p < .0001 versus untransfected treated islets; ^#p < .05 versus islets transfected with anti-miR-26a-5p alone; ^$p < .01 versus islets transfected with anti-miR-26a-5p alone; ^@p > .05 versus untransfected islets. (d) Effect of Honey TRF on COX-2 mRNA expression in pancreatic islets transfected with anti-miR-control. (e) Effect of Honey TRF on COX-2 protein expression in pancreatic islets transfected with anti-miR-26a-5p. Band images were digitally captured and the band intensities (pixels/band) were obtained using the Un-Scan-It software and are expressed in average pixels. ^*p < .001 versus untransfected treated islets; ^#p < .05 versus islets transfected with anti-miR-26a-5p alone; ^@p < .001 versus islets transfected with anti-miR-26a-5p alone; ^$p > .05 versus untransfected islets. (f) Effect of Honey TRF on COX-2 protein expression in pancreatic islets transfected with anti-miR-control. (g) Effect of Honey TRF on PGE2 production in the culture medium of pancreatic islets transfected with anti-miR-26a-5p. ^*p < .001 versus untransfected treated islets; ^#p < .05 versus islets transfected with anti-miR-26a-5p alone; ^@p < .001 versus islets untransfected. (h) Effect of Honey TRF on PGE2 production in pancreatic islets transfected with anti-miR-control. Untransfected islets were used as controls and expression of snoRNA202/GAPDH/β-actin were used as endogenous controls in their respective experiments. Each experiment was performed in triplicate and repeated five times using different sets of mouse pancreatic islets.

Honey TPF up-regulates microRNA-26a-5p expression and down-regulates PGE₂ production in pancreatic islets

We also determine whether inhibition of COX-2 expression also inhibits PGE₂ production in the same sets of mouse pancreatic islets transfected with anti-miR26a-5p or anti-miR-control, the culture medium was quantified for PGE₂ secretion. Transfection of islets with anti-miR-26a-5p markedly enhanced the PGE2 production (p < .001). This transfection induced PGE₂ production was significantly reduced by honey TPF in a dose-dependent manner (Figure 4(g); p < .05). As expected, transfection with anti-miR-control, followed by honey TPF treatment induces no alterations in the production of PGE2 (Figure 4(h); p > .05). These results not only validated that mmu-miR26a-5p regulates COX-2 expression and PGE₂ production in pancreatic islets but also confirmed the potential of polyphenols of honey in the inhibition of potent COX-2 via microRNA-26a-5p.

Effect of honey TPF on negative co-regulation of mmu-miR-26a-5p and COX-2 and the involvement of 3^′UTR of COX-2 mRNA

Involvement of honey TPF in the negative co-regulation of mmu-miR-26a-5p and COX-2 was confirmed by a luciferase reporter assays using a reporter clone containing entire 3′UTR of COX-2 mRNA in pancreatic beta cell line Min6. The results showed a significant increase in relative luciferase activity (RLA) in Min6 beta cells co-transfected with 3^′UTR COX-2 reporter and anti-miR-26a-5p (p < .05). This concentration dependent increase of RLA was reversed when these co-transfected Min6 cells were treated with honey TPF (Figure 5(a)). This marked concentration-dependent reduction in luciferase activity, may indicate that honey TPF reduces luciferase activity of 3^′UTR COX-2 reporter via modulation of mmu-miR-26a-5p expression. Involvement of honey TPF in the negative co-regulation of mmu-miR-26a-5p and COX-2 was further evaluated by transfection of pancreatic beta cells Min6 with anti-miR-26a-5p and then co-treated with honey TPF and glucose. As shown in Figure 5(b), treatment with glucose alone or transfection with anti-miR-26a-5p significantly reduced the mmu-miR-26a-5p expression in Min6 cells (p < .01). Interestingly, transfection with anti-miR-26a-5p synergizes with glucose noticeably reduced mmu-miR-26a-5p expression (p < .001; Figure 5(b), bar 5). Whereas, treatment of anti-miR-26a-5p-transfected Min6 cells with honey TPF consistently up-regulated the glucose-inhibited mmu-miR-26a-5p expression in a dose-dependent manner (Figure 5(b); p < .05). As expected transfection of Min6 cells with anti-miR-control without glucose or honey TPF had no effect on mmu-miR-26a-5p expression (Figure 5(b), bar 4; p > .05). In same sets of transfected Min6 cells, COX-2 mRNA was quantified. Figure 5(c) shows that Min6 cells treated with glucose alone significantly upregulated COX-2 mRNA expression compared with untreated Min6 cells (p < .0001). Transfection of these pancreatic beta cells with anti-miR-26a-5p also significantly increased COX-2 expression as compared with those Min6 cells transfected with anti-miR-control (p < .05). But a sharp increase of COX-2 mRNA expression was observed in anti-miR-26a-5p-transfected Min6 cells stimulated with glucose for 24 h (Figure 5(c); p < .001). Importantly, honey TPF significantly inhibited glucose-induced COX-2 mRNA in a dose-dependent manner in Min6 cells transfected with anti-miR-26a-5p (Figure 5(c); p < .05). To determine whether this effect of honey TPF on inhibition of COX-2 mRNA also affected the protein level, lysates of these transfected Min6 cells were prepared for COX-2 protein estimation. Western blot of COX-2 protein analysis showed that transfection of Min6 cells with anti-miR-26a-5p effectively further enhanced the glucose-induced COX-2 protein (p < .05). This enhanced expression of COX-2 protein was significantly inhibited by honey TPF in a dose-dependent manner (Figure 5(d); p < .05). Taken together, the data clearly indicate mmu-miR-26a-5p is a direct regulator of COX-2 expression in mouse pancreatic beta cells and honey TPF induced inhibition of COX-2 mRNA/protein expression is mediated via up-regulation of mmu-miR-26a-5p expression. We also determine whether inhibition of COX-2 expression also inhibits PGE₂ production, same sets of Min6 cells culture medium were analyzed. The data showed that the transfection of Min6 cells with anti-miR-26a-5p transfection effectively further increased the glucose-induced PGE₂ (p < .05) pancreatic beta cells. This enhanced expression of PGE₂ production was consistently inhibited by honey TPF in a dose-dependent manner (Figure 5(e); p < .01). These novel results clearly indicate that honey TPF inhibits COX-2 expression and PGE₂ production via up-regulation of mmu-miR-26a-5p expression.

Figure 5.

Honey polyphenolic fraction directly up-regulates the mmu-miR-26a-5p expression and down-regulates COX-2 expression via 3^′UTR of COX-2 mRNA sequence in mouse pancreatic beta cell line Min6. (a) Effect of honey total polyphenolic fraction (Honey TRF) on the luciferase activity in Min6 cells co-transfected with the reporter vector and anti-miR-26a-5p. ^@p < .0001 versus Min6 cells transfected with Empty 3’UTR vector containing only luciferase gene and its constitutively active promoter; ^#p < .05 versus @; ^*p < .05 versus #. Empty 3’UTR vector containing only luciferase gene and its constitutively active promoter was used as control. (b) Effect of Honey TRF on glucose-reduced mmu-miR-26a-5p expression in Min6 cells transfected with anti-miR-26a-5p. ^*p < .01 versus untreated Min6 cells or Min6 cells transfected with anti-miR-control alone; ^*p < .01 versus #. (c) Effect of Honey TRF on glucose-induced COX-2 mRNA expression in Min6 cells transfected with anti-miR-26a-5p. ^@p < .001 versus untreated Min6 cells or Min6 cells transfected with anti-miR-control alone; ^*p < .05 versus untreated Min6 cells or Min6 cells transfected with anti-miR-control alone; ^#p < .05 versus @ or *; ^##p < .05 versus #; ^###p < .05 versus # or ##. (d) Effect of Honey TRF on glucose-stimulated COX-2 protein expression in Min6 cells transfected with anti-miR-26a-5p. Band images were digitally captured and the band intensities (pixels/band) were obtained using the Un-Scan-It software and are expressed in average pixels. ^*p < .01 versus untreated Min6 cells; ^*p < .05 versus untreated Min6 cells or Min6 cells transfected with anti-miR-control alone; ^#p < .05 versus ## or *; ^**p < .05 versus # or ##. (e) Effect of Honey TRF on glucose-induced PGE₂ production in the culture medium of Min6 cells transfected with anti-miR-26a-5p. ^*p < .0001 versus untreated Min6 cells; ^#p < .001 versus untransfected Min6 cells or Min6 cells treated with glucose alone; ^#p < .01 versus ## or **. Untransfected or untreated Min6 cells were used as controls and expression of snoRNA202/GAPDH/β-actin were used as endogenous controls in their respective experiments. Each experiment was performed in triplicate and repeated at least three times using different sets of Min6 cells.

Honey TPF inhibits glucose-induced NF-κB signaling via mmu-miR-26a-5p in mouse pancreatic β-cells

In order to investigate whether honey TPF inhibits the glucose-induced downstream signaling via microRNA modulation, we transfected pancreatic beta cells Min6 with anti-miR-26a-5p and then treated with honey TPF for 2 h and stimulated with glucose for 30 min and the levels of nuclear NF-κBp65 and NF-κBp50 were determined by the highly sensitive NF-κB p65/p50 Transcription Factor Binding Assay kit (Abcam). Our data showed that the transfection of MIN6 cells with anti-miR-26a-5p or MIN6 cells treated with glucose alone significantly increased the nuclear levels of NF-κBp65 (p < .05) and this level was markedly increased when anti-miR-26a-5p transfected Min6 cells were treated with glucose (p < .01; Figure 6(a)). Whereas transfection with anti-miR-control showed no effect on the nuclear levels of NF-κBp65 (p > .05). Interestingly, treatment of transfected Min6 cells with honey TPF consistently inhibited the glucose-stimulated nuclear levels of NF-κBp65 in a dose-dependent manner (p < .05; Figure 6(a), bar 6 and 7). Furthermore, same sets of anti-miR-26a-5p transfected Min6 cells also significantly increased the nuclear levels of NF-κBp50 upon glucose treatment or non-treatment (p < .05). As expected, transfection with anti-miR-control showed no effect on the nuclear levels of NF-κBp50 (p > .05). Again, treatment of anti-miR26a-5p transfected Min6 cells with honey TPF consistently inhibited the glucose-stimulated nuclear levels of NF-κBp50 in a dose-dependent manner (p < .05; Figure 6(b), bar 6 and 7). Taken together, these findings proved that honey TPF inhibits glucose-induced NF-κB activation through modulation of microRNA mmu-miR26a-5p in pancreatic beta cells.

Figure 6.

Honey polyphenolic fraction inhibits glucose induced NF-κB signaling events through mmu-miR-26a-5p in mouse pancreatic beta cell line Min6 cells. (a) Effect of Honey TRF inhibits glucose stimulated NF-κB p65 activation in Min6 cells transfected with anti-miR-26a-5p. ^#p < .01 versus untreated Min6 cells or Min6 cells transfected anti-miR-control; ^*p < .05 versus #; ^@p < .05 versus *. (b) Effect of Honey TRF inhibits glucose stimulated NF-κB p50 activation in Min6 cells transfected with anti-miR-26a-5p. ^#p < .01 versus untreated Min6 cells or Min6 cells transfected anti-miR-control; ^*p < .05 versus #; ^@p < .05 versus *. Activated nuclear factor-kappa B p65 or p50 in the nucleus was determined by highly specific transcription factor ELISA kit (Abcam). The positive control nuclear extracts supplied with the kit were used. Data are representative of three experiments and all experiments were repeated three times using different sets of Min6 cells.

Discussion

This is the first study that shows that the polyphenolic fraction of natural honey inhibits COX-2 mRNA/protein expression and PGE₂ production via upregulating microRNA mmu-miR-26a-5p expression in pancreatic islets. In recent years, scientists from all over the globe actively investigating the role of microRNAs in diabetes and several novel miRNAs were discovered and their roles in regulation of genes relevant to the pathogenesis of diabetes were identified.^40,41 In the last two decades, lots of research on prostaglandin pharmacology and physiology was conducted but the most important was based on the identification of COX-2, as it expresses predominantly in numerous pathological conditions and now it’s well known for inducing inflammation in almost all major tissues including pancreatic islets.^4,42 It is well known that PGE2 is generated via activation of COX-2, which has now been considered as a key player in the induction of pro-inflammatory pathways not only in diabetes but also in several metabolic disorders.⁴³ Importantly, developed pharmacological inhibitors of COX-2 show promising therapeutic effects against the onset and progression of a number of inflammatory diseases.⁴⁴ Although diabetes is predominant in every population but the treatment is still limited to a few classes of drugs and none of the treatment reverts this chronic disorder.^45,46 In view of these, there is an urgent need for the development/identification of new compounds that inhibit the metabolic process of diabetes. Previous studies reported honey comprises about 180 chemical substances including amino acids and protein; polyphenols; vitamins such as ascorbic acid, thiamine, riboflavin, niacin, pantothenic acid, and pyridoxine; and minerals such as zinc, potassium, chlorine, calcium, sulfur, sodium, phosphorous, magnesium, iron, copper.⁴⁷ The polyphenolic fraction of honey comprises of heterogeneous chemical compounds that can be further characterized into flavonoids such as flavonols, flavones, flavanols, flavanones, anthocyanidin, chalcones, and isoflavones) and non-flavonoids such as phenolic acids.^17,47 Most importantly, polyphenols of natural honey performed range of therapeutics activities for numerous human disorders,^15–21 therefore this study was designed to investigate the therapeutic potential of polyphenolic fraction of honey against the glucose induced COX-2 mRNA/protein expression and PGE₂ production via regulation of microRNAs in mouse pancreatic islets. To prove these, we first isolated total polyphenolic fraction from natural honey (honey TPF) using the Amberlite XAD-2 column, whereas the pancreatic islets were isolated from black C57BL/6 mice. Our data determined the role of honey TPF for the first time in suppressing the pro-inflammatory genes through regulation of miRNA in mouse pancreatic islets.

Bioinformatics has now become a powerful approach for the identification of novel microRNA target genes relevant to the disease pathogenesis, numerous miRNAs seed matched have been identified and experimentally tested successfully and now this approach has become a method of choice^9,48,49 and the role of miRNAs has now well reported in numerous human disorders.⁵⁰ Here we used three different bioinformatics algorithms namely TargetScan, miRBase and miRanda for the scanning of 3’UTR of mouse COX-2 (or Ptgs2) mRNA for the determination of seed matched sequences of miRNAs. TargetScan bioinformatics algorithm detected seven conserved sites in the 3’UTR of mouse COX-2 for different microRNAs such as miR-26a-5p, miR-103-3p, miR-107-3p, miR-101-3p, miR-143-3p and miR-144-3p. Moreover, this algorithm also detected that the these miRNA seed sequences conserved in mouse 3’UTR of COX-2 mRNA are also conserved in other species such as human, chimp, rhesus, squirrel, rat, rabbit, pig, cow, cat, dog and brown bat, indicating that the seed matched sequence of these miRNAs in 3^′UTR of COX-2 mRNA is highly conserved. In order to confirm these findings, we further analyzed the whole 3’UTR of COX-2 mRNA for the sequences complementary to miRNAs by another computer based bioinformatics approach miRBase and miRanda database. All three algorithms TargetScan, miRBase and miRanda databases identified sequences conserved in the 3^′UTR of COX-2 are complementary to miR-26a-5p seed sequence (Supplemental data). In order to validate these computer based predictions, experimental studies were performed with all seven highly conserved miRNAs predicted in 3^′UTR of mouse COX-2 mRNA using their specific taqman assays. Our experiments on purified mouse pancreatic islets showed that treatment with 25 mm glucose for 24 h significantly upregulated the COX-2 mRNA/protein or PGE₂ production and significantly downregulated the expression of mmu-miR-26a-5p. Whereas expressions of other miRNAs such as mmu-miR-103-3p, mmu-miR-107-3p, mmu-miR-101a-3p, mmu-miR-143-3p and mmu-miR-144-3p were not significantly affected upon glucose treatment of islets under identical experimental conditions. These findings may indicate an inverse correlation between the expression of COX-2 or PGE₂ with mmu-miR-26a-5p only (not with mmu-miR-103-3p, mmu-miR-107-3p, mmu-miR-101a-3p, mmu-miR-143-3p or mmu-miR-144-3p), when pancreatic islets were treated with 25 mm glucose. In order to validate that the binding of mmu-miR-26a-5p in the 3^′UTR of COX-2 mRNA, we used a luciferase reporter clone containing entire 3^′UTR of COX-2 mRNA including the predicted the seed-matched binding sites and luciferase assay was performed in mouse pancreatic beta cell line Min6. The characteristics of Min6 cell line are very much identical to those directly purified from pancreatic islets, clearly suggesting that this cell line is a best model for studying the glucose-induced mechanisms in pancreatic beta cells.⁵¹ Our data showed a significant concentration-dependent increase in luciferase activity in Min6 beta cells co-transfected with COX-2 reporter and anti-miR-26a-5p. These findings clearly demonstrated that mmu-miR-26a-5p binds the seed matched sequence present in the 3′UTR of mouse COX-2 mRNA. Therefore, all further investigations in this study were undertaken with mmu-miR-26a-5p only, not with mmu-miR-103-3p, mmu-miR-107-3p, mmu-miR-101a-3p, mmu-miR-143-3p or mmu-miR-144-3p.

After validation of inverse co-relation of COX-2 and mmu-miR-26a-5p, the efficacy of honey TPF was tested on modulation of COX-2 and mmu-miR26a-5p expressions in purified pancreatic islets or Min6 cells. To prove the therapeutic potential of honey TPF, mouse pancreatic islets were transfected with anti-miR-26a-5p or anti-miR-control and then treated with honey TPF. The data showed that transfection of islets with anti-miR-26a-5p inhibited the mmu-miR-26a-5p expression compared with non-transfected islets. Importantly, treatment of these transfected islets with honey TPF enhanced mmu-miR-26a-5p expression. These results were completely reversed, when we estimated mRNA expression of COX-2. These findings not only confirmed the negative correlation of mmu-miR26a-5p and COX-2 but also showed that honey TPF potentially inhibited COX-2 mRNA expression via upregulating the mmu-miR26a-5p expression in pancreatic islets. These results were further re-validated at the levels of COX-2 protein and PGE₂ production. Transfection of islets with anti-miR-26a-5p markedly enhanced the COX-2 protein and PGE₂ production and this induced COX-2 protein or PGE₂ production was significantly inhibited upon honey TPF treatment. Therapeutic potential of honey TPF in the negative co-regulation of mmu-miR-26a-5p and COX-2 was confirmed in pancreatic beta cell line Min6. Treatment of honey TPF with co-transfected Min6 cells of commercially purchased luciferase reporter clone of entire 3^′UTR of COX-2 mRNA and anti-miR-26a-5p, showed dose-dependent decrease in luciferase activity. This concentration-dependent reduction in luciferase activity, may indicate that honey TPF reduces luciferase activity of 3^′UTR COX-2 reporter via modulation of mmu-miR-26a-5p expression. Involvement of honey TPF in the negative co-regulation of mmu-miR-26a-5p and COX-2 was further reevaluated by transfection of Min6 with anti-miR-26a-5p and then co-treated with honey TPF and glucose. The data showed that the treatment with glucose alone or transfection with anti-miR-26a-5p significantly reduced the mmu-miR-26a-5p expression in Min6 cells. Interestingly, transfection with anti-miR-26a-5p synergizes with glucose noticeably reduced mmu-miR-26a-5p expression. Whereas, treatment of anti-miR-26a-5p-transfected Min6 cells with honey TPF consistently upregulated the glucose-inhibited mmu-miR-26a-5p expression in a dose-dependent manner. In same sets of transfected Min6 cells, COX-2 mRNA was quantified and found that the Min6 cells treated with glucose alone significantly upregulated COX-2 mRNA expression. Transfection with anti-miR-26a-5p also significantly increased COX-2 expression as compared with those Min6 cells transfected with anti-miR-control, but a sharp increase of COX-2 mRNA expression was observed in anti-miR-26a-5p-transfected Min6 cells stimulated with glucose for 24 h. Importantly, honey TPF significantly inhibited glucose-induced COX-2 mRNA in a dose-dependent manner in Min6 cells transfected with anti-miR-26a-5p. To determine whether this effect of honey TPF on inhibition of COX-2 mRNA also affected the protein level, lysates of these transfected Min6 cells were prepared for COX-2 protein estimation. Western blot of COX-2 protein analysis showed that transfection of Min6 cells with anti-miR-26a-5p effectively further enhanced the glucose-induced COX-2 protein. This enhanced expression of COX-2 protein was significantly inhibited by honey TPF in a dose-dependent manner. Taken together, the data clearly indicate mmu-miR-26a-5p is a direct regulator of COX-2 expression in mouse pancreatic beta cells and honey TPF caused inhibition of COX-2 mRNA/protein expression is mediated via up-regulation of mmu-miR-26a-5p expression. We also determine whether inhibition of COX-2 expression also inhibits PGE₂ production, same sets of Min6 cells culture medium were analyzed. The data showed that the transfection of Min6 cells with anti-miR-26a-5p transfection effectively further increased the glucose-induced PGE₂ pancreatic beta cells. This enhanced expression of PGE₂ production was consistently inhibited by honey TPF in a dose-dependent manner. These novel results clearly indicate that honey TPF inhibits COX-2 expression and PGE₂ production via up-regulation of mmu-miR-26a-5p expression.

The nuclear transcription factor (NF)-κB regulates numerous aspects of immune system and is known as a pivotal mediator of proinflammatory responses.⁵² Activation of NF-κB signaling events induces the expression of multiple proinflammatory genes, including those encoding cytokines, chemokines, and numerous other proinflammatory mediators including COX-2 and PGE₂.^53,54 Analysis of 5^′-flanking sequence of COX-2 gene found two sites for NF-κB.⁵⁴ In vitro suppression of NF-κB activity has been shown to inhibit COX-2 expression or PGE₂ production in several cell types and tissues, indicating that NF-κB play an important role in COX-2 induction and PGE₂ production.^55,56 Furthermore, the role of NF-κB activity in miRNAs mediated abnormal expressions of COX-2 and PGE₂ has also been reported.⁴⁹ The NF-κB complex is composed of five structurally related proteins, including NF-κBp65 (also called RelA), RelB, c-Rel, NF-κBp50 (also called NF-κB1) and NF-κB2. The nuclear levels of NF-κBp65 or NF-κB50 are the best and widely studied NF-κB proteins to determine the transcription activity in the induction of inflammation.^55–57 Earlier, we have shown that up-regulation of inflammatory genes is accompanied by down-regulation of miRNAs and increased expression of nuclear NF-κBp65 and NF-κB50.^10,11 In this study, we investigated whether honey TPF inhibits the glucose-induced NF-κB activation via modulation of mmu-miR-26a-5p and COX-2 expression. For these studies, we transfected Min6 with anti-miR-26a-5p and then treated with honey TPF for 2 h and stimulated with glucose for 30 min and the levels of nuclear NF-κBp65 and NF-κBp50. The data showed that the transfection of Min6 cells with anti-miR-26a-5p or Min6 cells treated with glucose alone significantly increased the nuclear levels of NF-κBp65 and NF-κBp50. These nuclear levels of NF-κB proteins were markedly enhanced when anti-miR-26a-5p transfected Min6 cells were treated with glucose. Whereas, treatment of transfected Min6 cells with honey TPF consistently inhibited the glucose-stimulated nuclear levels of NF-κBp65 and NF-κBp50. These findings indicating that honey TPF inhibits glucose-induced NF-κB activation through modulation of mmu-miR26a-5p in pancreatic beta cells. The novel findings (shown in Figure 4) not only validated that mmu-miR26a-5p regulates COX-2 expression and PGE₂ production in pancreatic beta cells but also confirmed the potential of honey TPF in the inhibition of potent COX-2 via microRNA-26a-5p. Taken all together, the findings clearly suggest that honey TPF suppresses glucose induced activation of NF-κB signaling and PGE₂ production via negative regulation of COX-2 and mmu-miR26a-5p. This study has a few limitations; the most obvious limitation of the study is the study was performed only on natural honey. It would have been better to test the other honey types on the same system. In addition, the analysis of specific polyphenols present in the honey on the same pancreatic beta cells would further strengthen the investigations.

Conclusions

This is the first report that shows the purified fraction of polyphenols from natural honey inhibits glucose-induced COX-2 expression and PGE₂ production via up-regulation of the expression of mmu-miR26a-5p in pancreatic islets. These findings were further verified in mouse pancreatic β-cell line Min6. Moreover, the data also determined that honey TPF inhibits glucose-induced NF-κB activity. These novel pharmacological actions of honey TPF on microRNA regulation provide new suggestions that honey TPF or honey-derived compounds may be of value for the treatment of diabetic patients where inflammation and microRNAs play an active role.

Supplemental Material

sj-pdf-1-eji-10.1177_20587392221076473 – Supplemental Material for Honey polyphenolic fraction inhibits cyclooxygenase-2 expression via upregulation of microRNA-26a-5p expression in pancreatic islets

Supplemental Material, sj-pdf-1-eji-10.1177_20587392221076473 for Honey polyphenolic fraction inhibits cyclooxygenase-2 expression via upregulation of microRNA-26a-5p expression in pancreatic islets by Syed S Ahmed, Sultan AlNohair, Waleed A Abdulmonem, Homaidan T Alhomaidan, Naila Rasheed, Mohamed SM Ismail, Manal A Albatanony and Zafar Rasheed in European Journal of Inflammation

Footnotes

Acknowledgements

The authors thank Mr Casimero A. Victoria for helping in transfection experiments.

Authors’ contributions

Syed Suhail Ahmed, Sultan AlNohair, Waleed Al Abdulmonem, Homaidan T. Alhomaidan, Manal A. Albatanony carried out experimentation and data interpretation. Naila Rasheed and Zafar Rasheed performed data interpretation and manuscript drafting. All authors have read and approved the final manuscript.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by the Deanship of Scientific Research (Grant # med-2019-2-2-I-6736), Qassim University, KSA.

Availability of data and materials

All data and materials used in this current study are available from the corresponding author on reasonable request.

Ethical approval

The study was approved by the ‘Institutional Review Board’ Ethical Committee from the Deanship of Scientific Research, Qassim University, KSA (IRB approval #19-09-02).

ORCID iD

Zafar Rasheed

Supplemental material

Supplemental material for this article is available online.

Nomenclature

References

Planas

Carrillo

Sanchez

, et al. Gene expression profiles for the human pancreas and purified islets in type 1 diabetes: new findings at clinical onset and in long-standing diabetes. Clin Exp Immunol 2009; 159(1): 23–44. DOI: 10.1111/j.1365-2249.2009.04053.x. Epub 2009 Nov 11. PMID: 19912253; PMCID: PMC2802692.

Yip

Fuhlbrigge

Alkhataybeh

, et al. Gene expression analysis of the pre-diabetic pancreas to identify pathogenic mechanisms and biomarkers of type 1 diabetes. Front Endocrinol 2020; 11: 609271. DOI: 10.3389/fendo.2020.609271. PMID: 33424774; PMCID: PMC7793767.

Katsarou

Gudbjörnsdottir

Rawshani

, et al. Type 1 diabetes mellitus. Nat Rev Dis Primers 2017; 3: 17016. DOI: 10.1038/nrdp.2017.16. PMID: 28358037.

Robertson

. Dominance of cyclooxygenase-2 in the regulation of pancreatic islet prostaglandin synthesis. Diabetes 1998; 47(9): 1379–1383. DOI: 10.2337/diabetes.47.9.1379. PMID: 9726224.

Shanmugam

Todorov

Nair

, et al. Increased expression of cyclooxygenase-2 in human pancreatic islets treated with high glucose or ligands of the advanced glycation endproduct-specific receptor (AGER), and in islets from diabetic mice. Diabetologia 2006; 49(1): 100–107. DOI: 10.1007/s00125-005-0065-7. Epub 2005 Dec 10. PMID: 16341840.

Turini

DuBois

. Cyclooxygenase-2: a therapeutic target. Annu Rev Med 2002; 53: 35–57. DOI: 10.1146/annurev.med.53.082901.103952. PMID: 11818462.

Pop-Busui

Kellogg

Cheng

. Cyclooxygenase-2 pathway as a potential therapeutic target in diabetic peripheral neuropathy. Curr Drug Targets 2008; 9(1): 68–76. DOI: 10.2174/138945008783431691. PMID: 18220714.

Schoenberger

Kim

Sheng

, et al. Increased prostaglandin E2 (PGE2) levels in proliferative diabetic retinopathy, and correlation with VEGF and inflammatory cytokines. Invest Opthalmology Vis Sci 2012; 53(9): 5906–5911. DOI: 10.1167/iovs.12-10410. PMID: 22871833.

Rasheed

. Bioinformatics approach: a powerful tool for microRNA research. Int J Health Sci 2017; 11(3): 1–3. PMID: 28936142; PMCID: PMC5604275.

10.

Rasheed

Al-Shobaili

Rasheed

, et al. MicroRNA-26a-5p regulates the expression of inducible nitric oxide synthase via activation of NF-κB pathway in human osteoarthritis chondrocytes. Arch Biochem Biophys 2016; 594: 61–67. DOI: 10.1016/j.abb.2016.02.003. Epub 2016 Feb 27. PMID: 26854724.

11.

Rasheed

Abdulmonem

, et al. MicroRNA-125b-5p regulates IL-1β induced inflammatory genes via targeting TRAF6-mediated MAPKs and NF-κB signaling in human osteoarthritic chondrocytes. Scientific Rep 2019; 9(1): 6882. DOI: 10.1038/s41598-019-42601-3.

12.

Poy

Eliasson

Krutzfeldt

, et al. A pancreatic islet-specific microRNA regulates insulin secretion. Nature 2004; 432(7014): 226–230. DOI: 10.1038/nature03076. PMID: 15538371.

13.

Tian

Ouyang

, et al. Cross-talks between microRNAs and mRNAs in pancreatic tissues of streptozotocin-induced type 1 diabetic mice. Biomed Rep 2015; 3(3): 333–342. DOI: 10.3892/br.2015.426. Epub 2015 Feb 12. PMID: 26137232; PMCID: PMC4467227.

14.

Abuhatzira

Tahhan

, et al. Multiple microRNAs within the 14q32 cluster target the mRNAs of major type 1 diabetes autoantigens IA‐2, IA‐2β, and GAD65. FASEB J 2015; 29(10): 4374–4383. DOI: 10.1096/fj.15-273649. Epub 2015 Jul 6. PMID: 26148972; PMCID: PMC4566937.

15.

Samarghandian

Farkhondeh

Samini

. Honey and health: a review of recent clinical research. Pharmacogno Res 2017; 9(2): 121–127. DOI: 10.4103/0974-8490.204647. PMID: 28539734; PMCID: PMC5424551.

16.

Ediriweera

ERHSS

Premarathna

NYS

. Medicinal and cosmetic uses of bee’s honey - a review. AYU (An International Quarterly Journal of Research in Ayurveda) 2012; 33(2): 178–182. DOI: 10.4103/0974-8520.105233. PMID: 23559786; PMCID: PMC3611628.

17.

Cianciosi

Forbes-Hernández

Afrin

, et al. Phenolic compounds in honey and their associated health benefits: a review. Molecules 2018; 23(9): 2322. DOI: 10.3390/molecules23092322. PMID: 30208664; PMCID: PMC6225430.

18.

Jaganathan

Mandal

. Antiproliferative effects of honey and of its polyphenols: a review. J Biomed Biotechnol 2009; 2009: 1–13. DOI: 10.1155/2009/830616. Epub 2009 Jul 19. PMID: 19636435; PMCID: PMC2712839.

19.

Khalil

Sulaiman

. The potential role of honey and its polyphenols in preventing heart disease: a review. Afr J Traditional, Complement Altern Medicines 2010; 7(4): 315–321. DOI: 10.4314/ajtcam.v7i4.56693. Epub 2010 Jul 3. PMID: 21731163; PMCID: PMC3005390.

20.

Al-Waili

Salom

Al-Ghamdi

. Honey for wound healing, ulcers, and burns; data supporting its use in clinical practice. The Scientific World JOURNAL 2011; 11: 766–787. DOI: 10.1100/tsw.2011.78. PMID: 21479349; PMCID: PMC5720113.

21.

Estevinho

Pereira

Moreira

, et al. Antioxidant and antimicrobial effects of phenolic compounds extracts of Northeast Portugal honey. Food Chem Toxicol 2008; 46(12): 3774–3779. DOI: 10.1016/j.fct.2008.09.062. Epub 2008 Oct 4. PMID: 18940227.

22.

Alevia

Rasines

Cantero

, et al. Chemical extraction and gastrointestinal digestion of honey: influence on its antioxidant, antimicrobial and anti-inflammatory activities. Foods 2021; 10(6): 1412. DOI: 10.3390/foods10061412. PMID: 34207097; PMCID: PMC8235604.

23.

Masad

Haneefa

Mohamed

, et al. The immunomodulatory effects of honey and associated flavonoids in cancer. Nutrients 2021; 13(4): 1269. DOI: 10.3390/nu13041269. PMID: 33924384; PMCID: PMC8069364.

24.

Afrin

Haneefa

Fernandez-Cabezudo

, et al. Therapeutic and preventive properties of honey and its bioactive compounds in cancer: an evidence-based review. Nutr Res Rev 2020; 33(1): 50–76. DOI: 10.1017/S0954422419000192. Epub 2019 Dec 3. PMID: 31791437.

25.

Battino

Giampieri

Cianciosi

, et al. The roles of strawberry and honey phytochemicals on human health: a possible clue on the molecular mechanisms involved in the prevention of oxidative stress and inflammation. Phytomedicine 2021; 86: 153170. DOI: 10.1016/j.phymed.2020.153170. Epub 2020 Jan 11. PMID: 31980299.

26.

Gasparrini

Afrin

Forbes-Hernández

, et al. Protective effects of Manuka honey on LPS-treated RAW 264.7 macrophages. Part 2: control of oxidative stress induced damage, increase of antioxidant enzyme activities and attenuation of inflammation. Food Chem Toxicol 2018; 120: 578–587. DOI: 10.1016/j.fct.2018.08.001. Epub 2018 Aug 2. PMID: 30077706.

27.

Alvarez-Suarez

Giampieri

Cordero

, et al. Activation of AMPK/Nrf2 signalling by Manuka honey protects human dermal fibroblasts against oxidative damage by improving antioxidant response and mitochondrial function promoting wound healing. J Funct Foods 2016; 25: 38–49.

28.

Persaud

Burns

Belin

, et al. Glucose-induced regulation of COX-2 expression in human islets of Langerhans. Diabetes 2004; 53(Suppl 1): S190–S192. DOI: 10.2337/diabetes.53.2007.s190. PMID: 14749287.

29.

Istasse

Jacquet

Berchem

, et al. Extraction of honey polyphenols: method development and evidence of Cis isomerization Ubertas academica. Anal Chem Insights 2016; 11: ACI.S39739–57. DOI: 10.4137/ACI.S39739. PMID: 27547032; PMCID: PMC4981221.

30.

Jaganathan

Mandal

. Honey constituents and their apoptotic effect in colon cancer cells. J ApiProduct ApiMedical Sci 2009; 1(2): 29–36. DOI: 10.3896/IBRA.4.01.2.02.

31.

Miyazaki

J-I

Araki

Yamato

, et al. Establishment of a pancreatic β cell line that retains glucose-inducible insulin secretion: special reference to expression of glucose transporter isoforms*. Endocrinology 1990; 127(1): 126–132. DOI: 10.1210/endo-127-1-126. PMID: 2163307.

32.

Amior

Srivastava

Nano

, et al. The role of Cox‐2 and prostaglandin E 2 receptor EP3 in pancreatic β‐cell death. FASEB J 2019; 33(4): 4975–4986. DOI: 10.1096/fj.201801823R. Epub 2019 Jan 10. PMID: 30629897.

33.

Cheng

Delghingaro-Augusto

Nolan

, et al. High passage MIN6 cells have impaired insulin secretion with impaired glucose and lipid oxidation. PLoS One 2012; 7(7): e40868. DOI: 10.1371/journal.pone.0040868. Epub 2012 Jul 13. PMID: 22808281; PMCID: PMC3396628.

34.

Lakey

JRT

Young

ATL

Pardue

, et al. Nonviral transfection of intact pancreatic islets. Cel Transplant 2001; 10(8): 697–708. PMID: 11814112.

35.

Rasheed

Al‐Shobaili

. Epigallocatechin‐3‐ O ‐gallate up‐regulates microRNA‐199a‐3p expression by down‐regulating the expression of cyclooxygenase‐2 in stimulated human osteoarthritis chondrocytes. J Cell Mol Med 2016; 20(12): 2241–2248. DOI: 10.1111/jcmm.12897. Epub 2016 Aug 12. PMID: 27515563; PMCID: PMC5134388.

36.

Rasheed

Al-Shaya

. Epigallocatechin-3-O-gallate modulates global microRNA expression in interleukin-1β-stimulated human osteoarthritis chondrocytes: potential role of EGCG on negative co-regulation of microRNA-140-3p and ADAMTS5. Eur J Nutr 2018; 57(3): 917–928. DOI: 10.1007/s00394-016-1375-x. Epub 2017 Jan 21. PMID: 28110479.

37.

Pfaffl

. A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res 2001; 29(9): 45e–45. DOI: 10.1093/nar/29.9.e45. PMID: 11328886; PMCID: PMC55695.

38.

Al Abdulmonem

Rasheed

Aljohani

ASM

, et al. Absence of CD74 isoform at 41kDa prevents the heterotypic associations between CD74 and CD44 in human lung adenocarcinoma-derived cells. Immunological Invest 2020; 50(8): 891–905. DOI: 10.1080/08820139.2020.1790594. Epub ahead of print. PMID: 32646312.

39.

Al Abdulmonem

Rasheed

Al Ssadh

, et al. Bacterial lipopolysaccharide induces the intracellular expression of trophoblastic specific CD74 isoform in human first trimester trophoblast cells: correlation with unsuccessful early pregnancy. J Reprod Immunol 2020; 141: 103152. DOI: 10.1016/j.jri.2020.103152. Epub 2020 May 28. PMID: 32521377.

40.

Vasu

Kumano

Darden

, et al. MicroRNA signatures as future biomarkers for diagnosis of diabetes states. Cells 2019; 8(12): 1533. DOI: 10.3390/cells8121533. PMID: 31795194; PMCID: PMC6953078.

41.

Hathaway

Pinti

Durr

, et al. Regulating microRNA expression: at the heart of diabetes mellitus and the mitochondrion. Am J Physiology-Heart Circulatory Physiol 2018; 314(2): H293–H310. DOI: 10.1152/ajpheart.00520.2017. Epub 2017 Oct 6. PMID: 28986361; PMCID: PMC5867655.

42.

Kimple

Keller

Rabaglia

, et al. Prostaglandin E2 receptor, EP3, is induced in diabetic islets and negatively regulates glucose- and hormone-stimulated insulin secretion. Diabetes 2013; 62(6): 1904–1912. DOI: 10.2337/db12-0769. Epub 2013 Jan 24. PMID: 23349487; PMCID: PMC3661627.

43.

Nørregaard

Kwon

T-H

Frøkiær

. Physiology and pathophysiology of cyclooxygenase-2 and prostaglandin E2 in the kidney. Kidney Res Clin Pract 2015; 34(4): 194–200. DOI: 10.1016/j.krcp.2015.10.004. Epub 2015 Nov 12. PMID: 26779421; PMCID: PMC4688592.

44.

Blain

Jouzeau

Netter

, et al. Les anti-inflammatoires non stéroïdiens inhibiteurs sélectifs de la cyclooxygénase 2.Intérêt et perspectives. La Revue de Médecine Interne 2000; 21(11): 978–988, French. DOI: 10.1016/s0248-8663(00)00254-x. PMID: 11109595.

45.

Scheen

. Drug interactions of clinical importance with antihyperglycaemic agents. Drug Saf 2005; 28(7): 601–631. DOI: 10.2165/00002018-200528070-00004. PMID: 15963007.

46.

Gerich

. Insulin-dependent diabetes mellitus: pathophysiology. Mayo Clinic Proc 1986; 61(10): 787–791. DOI: 10.1016/s0025-6196(12)64818-6. PMID: 3528691.

47.

Ashagrie Tafere

. Chemical composition and uses of honey: a Review. J Food Sci Nutr Res 2021; 04(3): 194–201. DOI: 10.26502/jfsnr.2642-11000072.

48.

Rasheed

Al-Shobaili

Rasheed

, et al. Integrated study of globally expressed microRNAs in IL-1β-stimulated human osteoarthritis chondrocytes and osteoarthritis relevant genes: a microarray and bioinformatics analysis. Nucleos Nucleot NuclAcids 2016; 35(7): 335–355. DOI: 10.1080/15257770.2016.1163380. Epub 2016 May 6. PMID: 27152662.

49.

Quillet

Saad

Ferry

, et al. Improving bioinformatics prediction of microRNA targets by ranks aggregation. Front Genet 2020; 10: 1330. DOI: 10.3389/fgene.2019.01330. PMID: 32047509; PMCID: PMC6997536.

50.

Hassan

Majd

Rahnam

, et al. Masoodi AMCross-talk between oxidative stress signaling and microRNA regulatory systems in carcinogenesis: Focused on gastrointestinal cancers. Biomed Pharmacother 2020; 131: 110729.

51.

Ishihara

Asano

Tsukuda

, et al. Pancreatic beta cell line MIN6 exhibits characteristics of glucose metabolism and glucose-stimulated insulin secretion similar to those of normal islets. Diabetologia 1993; 36(11): 1139–1145. DOI: 10.1007/BF00401058. PMID: 8270128.

52.

Liu

Zhang

Joo

, et al. NF-κB signaling in inflammation. Signal TransducTarget Targeted Ther 2017; 2: 17023. DOI: 10.1038/sigtrans.2017.23. Epub 2017 Jul 14. PMID: 29158945; PMCID: PMC5661633.

53.

Poligone

Baldwin

. Positive and negative regulation of NF-κB by COX-2. J Biol Chem 2001; 276(42): 38658–38664. DOI: 10.1074/jbc.M106599200. Epub 2001 Aug 16. PMID: 11509575.

54.

Tazawa

, et al. Characterization of the genomic structure, chromosomal location and promoter of human prostaglandin H synthase-2 gene. Biochem Biophysical Res Commun 1994; 203(1): 190–199. DOI: 10.1006/bbrc.1994.2167. PMID: 8074655.

55.

Rasheed

Alghasham

. Lactoferrin from camelus dromedarius inhibits nuclear transcription factor-kappa B activation, cyclooxygenase-2 expression and prostaglandin e2 production in stimulated human chondrocytes. Pharmacognosy Res 2016; 8(2): 135–141. DOI: 10.4103/0974-8490.175612. PMID: 27034605; PMCID: PMC4780140.

56.

Charalambous

Lightfoot

Speirs

, et al. Expression of COX-2, NF-κB-p65, NF-κB-p50 and IKKα in malignant and adjacent normal human colorectal tissue. Br J Cancer 2009; 101(1): 106–115. DOI: 10.1038/sj.bjc.6605120. Epub 2009 Jun 9. PMID: 19513071; PMCID: PMC2713702.

57.

Zago

Rico de Souza

Hecht

, et al. The NF-κB family member RelB regulates microRNA miR-146a to suppress cigarette smoke-induced COX-2 protein expression in lung fibroblasts. Toxicol Lett 2014; 226(2): 107–116. DOI: 10.1016/j.toxlet.2014.01.020. Epub 2014 Jan 25. PMID: 24472607.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.72 MB

Honey polyphenolic fraction inhibits cyclooxygenase-2 expression via upregulation of microRNA-26a-5p expression in pancreatic islets

Abstract

Objectives

Methods

Results

Conclusion

Keywords

Introduction

Methods

Extraction of honey polyphenols

Pancreatic islets isolation and pancreatic β-cell line Min6 culture

Treatment of pancreatic islets and pancreatic β-cell line Min6 with honey polyphenolic fraction and glucose

Transfection of islets and Min6 cells with miRNA inhibitors

Luciferase reporter assays

MicroRNA isolation, first strand cDNA synthesis and TaqMan assays

Western immunoblotting for protein expression

Prostaglandin E2 ELISA

Nuclear factor-kappa B activity assays

Statements

Statistical analysis

Results

Bioinformatics algorithms for the detection of conserved seed matched sequences of microRNAs in 3′UTR of COX-2 (or Ptgs2) mRNA

Validation of mmu-miR-26a-5p binding in the 3′UTR of COX-2 mRNA

Honey TPF up-regulates microRNA-26a-5p expression and down-regulates COX-2 expressions in pancreatic islets

Honey TPF up-regulates microRNA-26a-5p expression and down-regulates PGE2 production in pancreatic islets

Effect of honey TPF on negative co-regulation of mmu-miR-26a-5p and COX-2 and the involvement of 3′UTR of COX-2 mRNA

Honey TPF inhibits glucose-induced NF-κB signaling via mmu-miR-26a-5p in mouse pancreatic β-cells

Discussion

Conclusions

Supplemental Material

sj-pdf-1-eji-10.1177_20587392221076473 – Supplemental Material for Honey polyphenolic fraction inhibits cyclooxygenase-2 expression via upregulation of microRNA-26a-5p expression in pancreatic islets

Footnotes

Acknowledgements

Authors’ contributions

Declaration of conflicting interests

Funding

Availability of data and materials

Ethical approval

ORCID iD

Supplemental material

Nomenclature

References

Supplementary Material

Prostaglandin E₂ ELISA

Bioinformatics algorithms for the detection of conserved seed matched sequences of microRNAs in 3^′UTR of COX-2 (or Ptgs2) mRNA

Validation of mmu-miR-26a-5p binding in the 3^′UTR of COX-2 mRNA

Honey TPF up-regulates microRNA-26a-5p expression and down-regulates PGE₂ production in pancreatic islets

Effect of honey TPF on negative co-regulation of mmu-miR-26a-5p and COX-2 and the involvement of 3^′UTR of COX-2 mRNA