In Vivo Silencing of MicroRNA-132 Reduces Blood Glucose and Improves Insulin Secretion

Abstract

Dysfunctional insulin secretion is a hallmark of type 2 diabetes (T2D). Interestingly, several islet microRNAs (miRNAs) are upregulated in T2D, including miR-132. We aimed to investigate whether in vivo treatment with antagomir-132 lowers expression of miR-132 in islets thereby improving insulin secretion and lowering blood glucose. Mice injected with antagomir-132 for 24 h, had reduced expression of miR-132 expression in islets, decreased blood glucose, and increased insulin secretion. In isolated human islets treated with antagomir-132, insulin secretion from four of six donors increased. Target prediction coupled with analysis of miRNA–messenger RNA expression in human islets revealed DESI2, ARIH1, SLC25A28, DIAPH1, and FOXA1 to be targets of miR-132 that are conserved in both species. Increased expression of these targets was validated in mouse islets after antagomir-132 treatment. In conclusion, we identified a post-transcriptional role for miR-132 in insulin secretion, and demonstrated that systemic antagomir-132 treatment in mice can be used to improve insulin secretion and reduce blood glucose in vivo. Our study is a first step towards utilizing antagomirs as therapeutic agents to modulate islet miRNA levels to improve beta cell function.

Introduction

MicroRNAs (miRNAs) are small noncoding RNAs suggested to have crucial functions in the regulation of insulin secretion [1] allowing for fine tuning of beta cell gene expression to optimize insulin secretion. As such, miRNAs may well play a critical role in the development of dysregulated insulin secretion and type 2 diabetes (T2D).

miRNAs negatively regulate target gene expression through the inhibition of messenger RNA (mRNA) translation or by promoting mRNA destabilization [2]. Differential expression of islet miRNAs has been associated with development of T2D [1,3]. For instance, in islets from the diabetic Goto-Kakizaki (GK) rat, we detected 24 upregulated miRNAs compared with levels in islets from control Wistar rats, among them miR-132-3p (miR-132) [4]. In line with this observation, miR-132-3p is consistently upregulated in islets from other diabetic animal models [3] and after long-term glucose incubation [4]. Overall, in islets, miR-132 is one of the most differentially expressed miRNAs in rodent models of diabetes [3,4] with its expression being second highest in islets compared with other tissues [5].

One of the key features of miRNAs is their ability to modulate the expression of multiple targets thereby coordinating functionally-related cellular pathways. This makes them attractive targets of RNA-based therapeutics. Indeed, in vivo application of anti-miRNA oligonucleotides, such as antagomirs or locked-nucleic acid (LNA) antimiRs, was shown to efficiently reduce expression of miRNAs in different tissues [6]. Our group recently showed that intravenous administration of antagomir-132 efficiently silences gene targets of miRNA-132 in the kidney thereby reducing renal fibrosis [7].

In this study we aimed to investigate the potential of using antagomir-132 to lower blood glucose and increase insulin secretion in vivo. We also investigated how miR-132 modulation affects potential targets in the islets.

Materials and Methods

Animals and in vivo experiments

Eight-week-old male BALB/c mice were used for intravenous (i.v.) antagomir-132 administration (Charles River Nederland, Maastricht, The Netherlands). Twenty-four hours before i.v. tail injection of 40 mg/kg bodyweight antagomir-132 or control scramblemir (5 mg/mL solution, ∼200 μL per mice), mice were individually housed in metabolic cages. Standard chow diet and drinking water were provided ad libitum. We used the I-Stat system (Abbott, CHEM8+ cartridges) to determine blood glucose levels. Blood plasma was obtained by centrifugation (15 min at 3,000 rpm at room temperature). Insulin levels determined by enzyme-linked immunosorbent assay (ELISA) (Mercodia, Uppsala, Sweden) according to the manufacturer supplied protocols.

The animal welfare committee of the veterinary authorities of the Leiden University Medical Center approved all animal experiments and protocols.

Human islets

In Leiden, human islets were obtained from six nondiabetic cadaveric human organ donors (average age, 52 years; average body mass index, 29.8) and isolations were performed in the Good Manufacturing Practice facility of our institute as described previously [8]. Purified islets used in these experiments were unsuitable for clinical islet transplantation, and only used if research consent had been obtained, according to national laws. In Lund, human islets from 45 cadaver donors (Table 1) were derived from the Human Tissue Laboratory (HTL), Lund University Diabetes Centre (LUDC)/EXODIAB (Lund, Sweden) through the Nordic Network of Islet Transplantation (Uppsala, Sweden; www.nordicislets.org). For experiments involving human pancreatic islets, all procedures were approved by Uppsala and Lund ethics committees. The glycemic status of donors was based on previous glycated hemoglobin range defined in Fadista et al. [9].

Table 1.

Characteristics of Donors and Derived Islets from Lund University Diabetes Centre

Glycemia	Sex: M/F	Age	BMI	Purity (%)	Days culture
NGT (HbA_1c < 6%)	12/9	55.8	26.2	71.7	3.7
IGT (6% ≤ HbA_1c < 6.5%)	4/8	59.1	26.9	70.0	2.8
T2D (HbA_1c ≥ 6.5%)	3/6	55.7	28.3	72.4	2.6

BMI, body mass index; F, female; HbA_1c, glycated hemoglobin; IGT, impaired glucose tolerance; M, male; NGT, normal glucose tolerance; T2D, type 2 diabetes.

Islet isolation and in vitro insulin secretion

Pancreases were digested using 3 mg/mL collagenase (Sigma-Aldrich, St Louis, CA) in RPMI-1640 medium (Invitrogen, Carlsbad, CA) supplemented with 2 μg/mL DNAse I (Pulmozyme; Roche) and shaken at 37°C for 15–18 min until a homogeneous digest was obtained that was then washed three times with cold RPMI medium supplemented with 10% (v/v) heat-inactivated FCS (Bodinco, Alkmaar, The Netherlands) and penicillin/streptomycin (100 U/mL and 100 μg/mL, respectively; Invitrogen). Islets were purified from exocrine tissue by manual selection picking under a dissecting microscope.

The human islets procured by the HTL of LUDC/EXODIAB from Nordic Network of Islet Transplantation (www.nordicislets.org) were isolated and processed as previously described [9].

Isolated human and mice islets were treated for 24 h with miR-132 LNA inhibitor or control inhibitor (Exiqon, Vedbaek, Denmark) before a glucose-induced insulin secretion test (Krebs–Ringer bicarbonate buffer; preincubation/basal secretion in 2 mM glucose and stimulation in 20 mM glucose). Supernatant fractions were kept for determination of insulin concentration by ELISA.

Antagomir design

Cholesterol-conjugated RNA analogs, “antagomirs,” (Thermo Scientific, Waltham, MA) were synthesized as previously described [6]. For antagomir-132 the following sequence was used: 5′-c_sg_saccauggcuguagacug_su_su_sa_s-Chol-3′. As a control a “scramblemir” was used, constructed from a randomized nucleotide sequence that does not bind to any known miRNAs: 5′-a_su_sgacuaucgcuauucgc_sa_su_sg_s-Chol-3′. Lower-case letters represent 2′-OMe-modified nucleotides; subscript “s” represents phosphorothioate linkage; “Chol” represents a cholesterol group linked through a hydroxyprolinol linkage.

RNA isolation and reverse transcription-quantitative PCR analysis

Total RNA was isolated using Trizol reagent (Invitrogen). miR-132 knockdown was validated using Taqman^® miR assays (Applied Biosystems, Foster City, CA). RNU6B was utilized for normalization. To quantify mRNA levels, 250 ng total RNA was reverse transcribed using oligo(dT) primers and M-MLV First-Strand Synthesis system (Invitrogen). Quantitative polymerase chain reaction of target genes was carried out using SYBR Green Master Mix (Applied Biosystems). Used primer sequences of target genes can be found in the Supplementary Table S1. Levels of expression were normalized to Gapdh and quantified using the delta Ct method.

For the human islets isolated in Lund, total RNA was extracted using the miRNeasy Kit (Qiagen). Quantification of mature hsa-miR-132-3p (miRBase access no.: MIMAT0000426) in human islets was performed using Taqman miR assays (Applied Biosystems) (Assay ID: 000457) and using the human RNU48 (Assay ID: 001006) as normalizer.

Target prediction analysis and statistical analysis

Target prediction was performed using the miRNA-target interaction web server TargetScan 7.2 (www.targetscan.org) [10]. The results were filtered using RNAseq mRNA data [9] from the same islet donors from which we measured miR-132 expression. Pearson correlation analysis was used to find negative correlations between miRNA-132 and target mRNA expressions. Top 14 targets with the lowest P value where chosen for further validation.

Significant differences between two groups were determined using Student's t-test. Data are presented as mean ± standard error of the mean.

Results

Treatment with antagomir-132 reduces blood glucose and increase insulin secretion

Injection of antagomir-132 in mice 24 h before measurements reduced islet miR-132 expression, reduced blood glucose, and increased plasma insulin levels (Fig. 1A–E). Moreover, blood glucose levels stayed low in mice treated with antagomir-132 three days after the injection (Supplementary Fig. S1).

FIG. 1.

Silencing miR-132 increases insulin levels and decreases glucose levels. (A) Experimental setup, mice were housed in metabolic cages for 24 h before receiving i.v. antagomir-132 or control scramblemir injections. Mice were killed 24 h later. (B) Islets were isolated from an additional seven mice (three scramblemir control and four antagomir-132 treated), in which we demonstrate knockdown of miR-132. (C) Systemic silencing of miR-132 resulted in decreased blood glucose levels after 24 h compared with scramblemir controls (n = 9–10 mice). (D) Insulin levels (measured in plasma of selection of animals) were increased after 24 h as a result of miR-132 inhibition (n = 4–5 mice). (E) Insulin/glucose ratios were concordantly increased. (F) Treatment of mouse islets (isolated from untreated animals) with antagomir-132 increased glucose-stimulated insulin secretion (n = 5 mice) measured as stimulation index meaning the fold increase in insulin secretion at 20 mM glucose compared with 2 mM glucose. (G) Treatment of human islets with antagomir-132 showed a trend toward increased glucose-stimulated insulin secretion. (H) Data from (G) illustrated per individual donor shows antagomir-132 treatment resulted four times in an increase and two times in a decrease of glucose-stimulated insulin secretion (n = 2–3 per donor). miR, microRNAs.

Likewise, in vitro treatment of mice islets with antagomir-132 increased glucose-stimulated insulin secretion (Fig. 1F) measured as stimulation index, that is, the fold increase in insulin release at 20 mM glucose compared with 2 mM glucose. To investigate if treatment affected cell survival, we used fluorescein diacetate (FDA) and propidium iodide (PI) staining (Supplementary Fig. S2). We observed a fluorescent signal after FDA staining that was not different between scramble control and antagomir-132-treated cells and no staining of PI. Thus, we observed no changes in cell viability, the reason why we believe the main effect of the antagomir-132 treatment is on beta cell function.

We next investigated if the antagomir-132 had similar effects in human islets. The insulin secretion response in human islets was more variable after antagomir-132 treatment (Fig. 1G, H), and insulin secretion was increased in four of six donors investigated.

Antagomir-132 increases expression of potential miR-132 targets

We observed no differential miR-132 expression in islets from donors diagnosed with T2D compared with normal glucose tolerant donors (Fig. 2A). However, it is possible that there may be differential expression within a specific subgroup of T2D donors [11].

FIG. 2.

miR-132 levels in human T2D islets significantly correlate with miR-132 targets. (A) miR-132 levels in human islets derived from NGT, IGT, or T2D patients. (B) Significant negative correlations between miR-132 levels and miR-132 targets in human islets. (C) Validation of increased levels of identified miR-132 targets in islets isolated from mice and treated with antagomir-132 or scramblemir control (n = 3–4 mice). IGT, impaired glucose tolerance; NGT, normal glucose tolerance; T2D, type 2 diabetes.

By interrogating RNAseq data from 45 human islet preparations, we evaluated miR-132 targets predicted by TargetScan and identified 473 conserved targets. We correlated the expression of these genes with miR-132 expression in islets from the same donors. As miRNAs negatively regulate gene expression, we sorted the correlations according to the largest negative correlation and lowest P value and found 14 significant correlations (Table 2). Among top potential targets were DESI2 (desumoylating isopeptidase 2), ARIH1 (UBCH7-binding protein), SLC25A28 (Solute Carrier family 25, mitochondrial carrier OR mitoferrin), and DIAPH1 (diaphanous, drosophila homolog of 1) (Fig. 2). Interestingly, FOXA1 (Forkhead box protein A1) was also a suggested miR-132 target (Table 2).

Table 2.

miR-132 Targets That Significantly Negative Correlate with miR-132 in Human Islets

miR-132 targets	P	r
TAF4B	0.008383	−0.39
DAZAP2	0.00906	−0.38
DESI2	0.00996	−0.38
ZNF207	0.010803	−0.38
PIK3IP1	0.012102	−0.37
FBXO42	0.014532	−0.36
ARIH1	0.022558	−0.34
RND3	0.022678	−0.34
SLC25A28	0.025215	−0.33
POLE3	0.026699	−0.33
RTF1	0.029472	−0.32
LSM11	0.030938	−0.32
DIAPH1	0.032736	−0.32
FOXA1	0.044089	−0.30

r = Pearson correlation coefficient.

miR, microRNAs.

Finally, we investigated if antagomir-132 has an impact on target gene expression in mouse islets. Indeed, treatment of islets with antagomir-132 significantly increased the expression of four of the investigated targets (Fig. 2C; Supplementary Fig. S3).

Discussion

In this study, we demonstrate that antagomir-132 can efficiently increase insulin secretion and decrease blood glucose in mice. Furthermore, despite its systemic injection, the antagomir can decrease the expression of miR-132 in the islets.

Our work provide basis for a novel approach to modulate insulin secretion in diabetic conditions and future studies need to be directed to assess its effects on glucose control in diabetic animal models. For instance, miR-132 is ubiquitous in other tissues and thorough investigations are necessary to evaluate the system-wide effect of modulating this miRNA. Nevertheless, our data support the existence of a miR-132-dependent posttranscriptional network that could provide novel therapeutic targets capable of promoting insulin secretion and reducing blood glucose levels. Furthermore, we previously showed that antagomir-132 reduces renal fibrosis [7] supporting the idea that modulation of the relevant network could ultimately benefit not only glycemic control but also, although adding complexity in the therapeutic utility, kidney function.

The results regarding antagomir-132 in human islets are promising but not straightforward. There was an overall tendency of increased insulin secretion by antagomir-132 but only in islets from four donors, whereas the effect was minor in the islets from two donors. Individual genetic variations and differences in the environmental milieu influencing the isolated islets could be an explanation. Thus, treatment with antagomir-132 may prove to work only in specific individuals, an approach that can contribute in personalized diabetes medicine.

Comparing our current data with previous studies on miR-132 expression and function in insulin secretion suggests a complex scenario. Earlier works have shown increased miR-132 expression in the diabetic GK rat [4], in high-fat diet (HFD) mice [12], and after long-term (24–72 h) treatment of either glucose or palmitate in rat islets [4,12], or glucagon-like peptide 1 in rodent and human islets [13]. Meanwhile, overexpression of miR-132 increased insulin secretion in rat islets, and protected beta cells from palmitate and cytokine-induced apoptosis in rodent and human islets [12]. Thus, the upregulation of miR-132 in many diabetes models seem to be a part of a beta cell compensatory program, and that a certain level of overexpression of miR-132 can improve beta cell function and offer protection from beta cell failure during hyperglycemia. Previous data on the effect of antagomir-132 on insulin secretion in cell lines show no effect or a tendency to reduced insulin secretion [13,14]. That modulating miR-132 expression can lead to different effects on insulin secretion reflects an intrinsic cellular requirement for optimal miR-132 level. Therefore, depending on the prevailing expression level of miR-132 before antagomir-132 treatment, the outcome on insulin secretion will be different. The role of many miRNAs is to keep protein expression and physiological outputs at optimum levels [15]. This is, for example, true for miR-375 where both knockdown and overexpression can result in reduced insulin secretion [16 –18]. We believe this might also be the case with miR-132 expression. Under certain conditions, overexpression of miR-132 will improve insulin secretion, whereas under other conditions/individuals miR-132 knockdown is a prerequisite for increased secretion.

Our integrative analysis of expression data also provides potential miR-132 targets in human islets. Among the top targets, antagomir-132 treatment increased the expression of DESI2, ARIH1, SLC25A28, and DIAPH1. DESI2 is a cysteine protease that removes small ubiquitin-like modifier (SUMO) from SUMO-modified proteins. Hence, increased level of this protein may increase insulin secretion, as desumoylation improves insulin secretion [19]. ARIH1 is involved in ubiquitination, SLC25A28 codes for a mitochondrial iron transporter, and DIAPH1 is involved in actin polymerization. The transcription factor FOXA1 is also one of the top miR-132 targets in human islets. This transcription factor has been shown to play an important role for functional beta cells [20]. Altogether, miR-132 seems to influence a network of beta cell genes involved in diverse essential cellular functions such as ubiquitination, mitochondrial ion transport, and granular mobilization. These targets were identified by both computational approach and integrative analysis of mRNA–miRNA data and may not be direct targets of miR-132. Nonetheless, these results give valuable clues for future analysis of miR-132-mediated posttranscriptional expression network in the human islets.

In conclusion, we showed that in vivo inhibition of miR-132 levels in mouse islets resulted in reduced blood glucose and increased insulin secretion. We further identified components of a mir-132-dependent molecular pathway that may provide novel therapeutic targets that could improve beta cell function and glucose regulation in T2D.

Footnotes

Acknowledgments

The authors thank the Human Tissue Laboratory (HTL) at Lund University Diabetes Centre (LUDC) and the Nordic Network for Clinical Islet Transplantation (Uppsala, Sweden) for providing human donor islets. Th authors also thank Britt-Marie Nilsson and Anna-Maria Veljanovska-Ramsay for technical assistance.

L.E. and J.L.S.E. are supported by a grant from Swedish Foundation for Strategic Research (IRC-LUDC) and Albert Påhlsson Foundation. L.E. is supported by grants from the Swedish Research Council (project grant to L.E., SFO-EXODIAB), Region Skåne-ALF, the Swedish Diabetes Foundation, Novo Nordisk Foundation, and the Diabetes Wellness Network Sweden. J.L.S.E. is supported by grants from Syskonen Svenssons Fond För Medicinsk Forskning and from the European Union's Horizon 2020 Research and Innovation Programme under grant agreement no. 667191 (T2D Systems). J.H.E. was supported by a Junior Diabetes Fonds Fellowship from Diabetes Fonds (2013.81.1675). R.B. is supported by a grant from the Dutch Kidney Foundation (16OKG16) and R.B. and A.J.v.Z. are supported by a grant from the European Foundation for the Study of Diabetes (EFSD).

Author Disclosure Statement

No competing financial interests exist.

Supplementary Material

References

Esguerra

, Nagao

, Ofori

, Wendt

and Eliasson

. (2018). MicroRNAs in islet hormone secretion. Diabetes Obes Metab, 20(Suppl. 2):11–19.

Guo

, Ingolia

, Weissman

and Bartel

. (2010). Mammalian microRNAs predominantly act to decrease target mRNA levels. Nature, 466:835–840.

LaPierre

and Stoffel

. (2017). MicroRNAs as stress regulators in pancreatic beta cells and diabetes. Mol Metab, 6:1010–1023.

Esguerra

JLS

, Bolmeson

, Cilio

and Eliasson

. (2011). Differential glucose-regulation of microRNAs in pancreatic islets of non-obese type 2 diabetes model Goto-Kakizaki rat. PLoS One, 6:e18613.

Landgraf

, Rusu

, Sheridan

, Sewer

, Iovino

, Aravin

, Pfeffer

, Rice

, Kamphorst

, et al. (2007). A mammalian microRNA expression atlas based on small RNA library sequencing. Cell, 129:1401–1414.

Krutzfeldt

, Rajewsky

, Braich

, Rajeev

, Tuschl

, Manoharan

and Stoffel

. (2005). Silencing of microRNAs in vivo with “antagomirs.” Nature, 438:685–689.

Bijkerk

, de Bruin

, van Solingen

, van Gils

, Duijs

JMGJ

, van der Veer

, Rabelink

, Humphreys

and van Zonneveld

. (2016). Silencing of microRNA-132 reduces renal fibrosis by selectively inhibiting myofibroblast proliferation. Kidney Int, 89:1268–1280.

Zaldumbide

, Alkemade

, Carlotti

, Nikolic

, Abreu

, Engelse

, Skowera

, de Koning

, Peakman

, et al. (2013). Genetically engineered human islets protected from CD8-mediated autoimmune destruction in vivo. Mol Ther, 21:1592–1601.

Fadista

, Vikman

, Laakso

, Mollet

, Esguerra

, Taneera

, Storm

, Osmark

, Ladenvall

, et al. (2014). Global genomic and transcriptomic analysis of human pancreatic islets reveals novel genes influencing glucose metabolism. Proc Natl Acad Sci USA, 111:13924–13929.

10.

Agarwal

, Bell

, Nam

and Bartel

. (2015). Predicting effective microRNA target sites in mammalian mRNAs. Elife. 4: e05005.

11.

Ahlqvist

, Storm

, Käräjämäki

, Martinell

, Dorkhan

, Carlsson

, Vikman

, Prasad

, Aly

, et al. (2018). Novel subgroups of adult-onset diabetes and their association with outcomes: a data-driven cluster analysis of six variables. Lancet Diabetes Endocrinol, 6:361–369.

12.

Nesca

, Guay

, Jacovetti

, Menoud

, Peyot

, Laybutt

, Prentki

and Regazzi

. (2013). Identification of particular groups of microRNAs that positively or negatively impact on beta cell function in obese models of type 2 diabetes. Diabetologia, 56:2203–2212.

13.

Shang

, Li

, Keller

, Hohmeier

, Wang

, Feng

, Zhou

, Shen

, Rabaglia

, et al. (2015). Induction of miR-132 and miR-212 expression by glucagon-like peptide 1 (GLP-1) in rodent and human pancreatic beta-cells. Mol Endocrinol, 29:1243–1253.

14.

Malm

, Mollet

, Berggreen

, Orho-Melander

, Esguerra

JLS

, Göransson

and Eliasson

. (2016). Transcriptional regulation of the miR-212/miR-132 cluster in insulin-secreting beta-cells by cAMP-regulated transcriptional co-activator 1 and salt-inducible kinases. Mol Cell Endocrinol, 424:23–33.

15.

Bartel

. (2009). MicroRNAs: target recognition and regulatory functions. Cell, 136:215–233.

16.

Eliasson

. (2017). The small RNA miR-375—a pancreatic islet abundant miRNA with multiple roles in endocrine beta cell function. Mol Cell Endocrinol, 456:95–101.

17.

Poy

, Eliasson

, Krutzfeldt

, Kuwajima

, Ma

, Macdonald

, Pfeffer

, Tuschl

, Rajewsky

, Rorsman

and Stoffel

. (2004). A pancreatic islet-specific microRNA regulates insulin secretion. Nature, 432:226–230.

18.

Poy

, Hausser

, Trajkovski

, Braun

, Collins

, Rorsman

, Zavolan

and Stoffel

. (2009). miR-375 maintains normal pancreatic alpha- and beta-cell mass. Proc Natl Acad Sci USA, 106:5813–5818.

19.

Dai

, Spigelman

, Khan

, Braun

, Manning Fox

and MacDonald

. (2014). SUMO1 enhances cAMP-dependent exocytosis and glucagon secretion from pancreatic alpha-cells. J Physiol, 592:3715–3726.

20.

Vatamaniuk

, Gupta

, Lantz

, Doliba

, Matschinsky

and Kaestner

. (2006). Foxa1-deficient mice exhibit impaired insulin secretion due to uncoupled oxidative phosphorylation. Diabetes, 55:2730–2736.

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

0.10 MB

1.23 MB

0.16 MB

0.12 MB