Abstract

   Somatostatin analogs, which are used to treat neuroendocrine tumors,
   inhibit hormone secretion or promote tumor shrinkage; however, their
   efficacy varies between patients, possibly because of differential
   expression of somatostatin receptors (SSTRs) in tumors. In this study,
   we evaluated the regulatory mechanism underlying the expression of
   SSTR2, the main octreotide target. Thirty miRNAs were found to be
   dysregulated in neuroendocrine cells (INS-1 cells) incubated with
   octreotide compared to that in placebo-treated cells. Among the
   upregulated miRNAs, miR-16-5p was elevated after short-term octreotide
   treatment. We conducted in vitro experiments to determine whether the
   expression of miR-16-5p was associated with the regulation of SSTR2
   expression and affected octreotide sensitivity in INS-1 cells.
   Overexpression of miR-16-5p by transfected mimics induced upregulation
   of SSTR2 expression. Additionally, the expression of miR-16-5p further
   enhanced octreotide-induced reduction in cell proliferation in both
   two- and three-dimensional culture of INS-1 cells. Thus, our results
   reveal the mechanism underlying SSTR2 expression regulation and may aid
   in developing therapeutic approaches for enhancing the response to
   octreotide, particularly in patients unresponsive to SSTR2-targeted
   somatostatin analog treatment.

Introduction

   Somatostatin (somatotropin release-inhibiting hormone, SST), a small
   polypeptide hormone produced in the hypothalamus [[40]1], regulates
   endocrine function mainly by suppressing the release of hormones and
   neurotransmitters and inhibits tumor growth by regulating tumor cell
   survival and angiogenesis [[41]2]. SST binds with high affinity to
   somatostatin receptors (SSTR1–5), which are specific G protein-coupled
   receptors present on cell surfaces [[42]3]. Neuroendocrine tumors
   (NETs) are a heterogeneous group of neoplasms with a low growth rate in
   most cases; however, they occasionally secrete hormones that cause
   specific clinical syndromes, resulting in significant disability and
   decreased quality of life [[43]4]. SSTR mRNA and protein expression
   patterns in NETs have been studied extensively, and SSTRs have been
   used as potential diagnostic and therapeutic targets because of their
   widespread expression and function in NETs.

   SST analogs (SSAs) have been developed and widely investigated [[44]3];
   for instance, octreotide and lanreotide are suitable for clinical
   applications because of their considerably longer half-life compared
   with that of natural SST. Similar to natural SST, SSAs bind to SSTRs;
   however, SSAs show different binding affinities for various receptors.
   Octreotide and lanreotide bind to SSTR2 with high affinity, followed by
   SSTR5, and activate inhibitory signals [[45]5, [46]6]. Thus, a positive
   correlation between the anticancer efficacy of octreotide and SSTR2
   expression exists in NETs [[47]2]. However, few studies have
   investigated mechanisms for enhancing SSTR expression and thereby
   increasing the anticancer effect of SSAs. Although acute administration
   of SST induces the activation of various inhibitory signals, the
   initial response against SSA is diminished. Continuous exposure to SSA
   reduces the initial drug effects via processes such as degradation,
   internalization, and phosphorylation of SSTRs [[48]7].

   MicroRNAs (miRNAs) are a class of non-coding RNAs comprising 21–25
   nucleotides. miRNAs have been reported to be closely associated with
   the regulation of different signaling pathways and several biological
   functions, including cell proliferation, differentiation, survival, and
   metabolism. miRNAs function by binding to complementary target mRNAs,
   resulting in mRNA translational inhibition or degradation [[49]8].
   Thus, several miRNA-targeting therapeutics have been developed and are
   being tested via clinical trials. Recently, miRNAs differentially
   expressed between SSA responders or non-responders in growth
   hormone-secreting pituitary adenomas have been identified [[50]9].
   However, the effects of miRNA-mediated control of SSTR2 expression and
   the subsequent effect on SST drug sensitivity in NETs remain poorly
   understood.

   In this study, we identified novel miRNAs involved in the early
   response to octreotide treatment and investigated SSTR2 expression
   during this early response. This study provides insight into the
   mechanism of octreotide, which may be useful for enhancing its
   anticancer effects, particularly in octreotide non-responder patients
   with NETs.

Materials and methods

Cell culture

   The INS-1 rat insulinoma cell line and GH3 rat pituitary GH- and
   PRL-producing cell line were purchased from Addexbio (San Diego, CA,
   USA) and the Korean Cell Line Bank. Both cell lines expressed SSTR2
   ([51]S1 Fig). INS-1 cells and GH3 cells were cultured in RPMI-1640
   (Gibco, Grand Island, NY, USA) and Dulbecco’s Modified Eagle Medium
   supplemented with 10% fetal bovine serum, 0.05 mmol/L
   2-mercaptoethanol, 100 U/mL penicillin, and 100 μg/mL streptomycin and
   maintained at 5% CO[2] and 37°C. HeLa cells were purchased from the
   Korean Cell Line Bank and grown in Dulbecco’s Modified Eagle Medium
   supplemented with 10% (v/v) fetal bovine serum and 1% (v/v)
   penicillin–streptomycin. Octreotide was purchased from TOCRIS
   Bioscience (Bristol, UK) and dissolved in deionized water to prepare a
   1 mM stock solution. For 3-dimensioanl culture, Costar® Ultra-Low
   attachment multi-well plates were purchased from Sigma-Aldrich Korea
   (St. Louis, MO, USA) and used as described previously [[52]10].

RNA extraction and miRNA synthesis

   Total RNA was extracted using TRIzol reagent (Life Technologies,
   Carlsbad, CA, USA) according to the manufacturer’s instructions. RNA
   concentrations were measured using a NanoDrop spectrophotometer
   (DeNovix, Wilmington, DE, USA). miRNAs were synthesized from 500 ng of
   total RNA using an miScript II RT Kit (Qiagen, Hilden, Germany)
   according to the manufacturer’s instructions.

Library preparation, sequencing, and data analysis

   For control and test RNAs, a library was constructed using the NEBNext
   Multiplex Small RNA Library Prep Kit (New England Biolabs, Ipswich, MA,
   USA) according to the manufacturer’s instructions. Briefly, 1 μg of
   total RNA from each sample was used to ligate the adaptors, after which
   cDNA was synthesized using reverse-transcriptase with adaptor-specific
   primers. PCR was performed for library amplification, and then the
   libraries were purified using a QIAquick PCR Purification Kit (Qiagen)
   and AMPure XP beads (Beckman Coulter, Brea, CA, USA). The yield and
   size distribution of the small RNA libraries were assessed by
   high-sensitivity DNA analysis on an Agilent 2100 Bioanalyzer (Agilent
   Technologies, Santa Clara, CA, USA). High-throughput sequences were
   produced by single-end 75 sequencing using the NextSeq 500 system
   (Illumina, San Diego, CA, USA). Sequence reads were mapped using the
   Bowtie 2 software tool to obtain the BAM file (alignment file). A
   mature miRNA sequence was used as a reference for mapping. Read counts
   mapped onto the mature miRNA sequence were extracted from the alignment
   file using bedtools (v2.25.0) and Bioconductor which uses R (version
   3.2.2) statistical programming language (R development Core Team,
   2011). The read counts were then used to determine the expression
   levels of miRNAs. The quantile normalization method was used to compare
   samples. miRWalk 2.0 was used for miRNA target analysis. Functional
   gene classification was performed using DAVID
   ([53]http://david.abcc.ncifcrf.gov/).

Transfection and treatment

   HeLa cells (1 × 10^5cells/well) were plated in 6-well plates. After 24
   h, 1 μg hSSTR2 or control plasmid (Sino Biological, Inc., Eschborn,
   Germany) was transfected using RNAiMAX transfection reagent
   (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s
   instructions. After an additional 24 h of incubation, an miRNA
   inhibitor was used to treat the HeLa cells. INS1 cells (1 × 10^5
   cells/well) were cultured in 6-well plates. A commercial miR-16-5p
   mimic (synthesized by Genolution, Seoul, South Korea) with the
   following sequence, sense, 5′-UAGCAGCACGUAAAUAUUGGCG-3′ and antisense,
   5′-CGCCAAUAUUUACGUGCUGCUA-3′ or negative control siRNA (Genolution) was
   transfected into INS1 cells. Transfection was performed using RNAiMAX
   transfection reagent (Invitrogen) according to the manufacturer’s
   instructions. After 24–48 h incubation, further analyses were
   performed.

Real-time quantitative PCR (RT-qPCR)

   To analysis miRNA expression, RT-qPCR was conducted on a C1000™ Thermal
   Cycler (Bio-Rad Laboratories, Hercules, CA, USA) using an miScript SYBR
   Green PCR Kit with miScript miRNA PCR Arrays (Qiagen). Gene expression
   levels were normalized to miRTC expression levels for the corresponding
   miRNA samples. The following target miRNAs were quantified by miScript
   Primer assays (Qiagen), rno-miR-16-5p (cat no. MS00033229; mature miRNA
   with the sequence UAGCAGCACGUAAAUAUUGGCG), and miRTC (MS00000001). cDNA
   was synthesized from 500 ng of total RNA using ReverTra Ace qPCR RT
   Master Mix with gDNA Remover (Toyobo, Osaka, Japan). Quantitative
   RT-PCR was performed using SYBG Green Real-time PCR Master Mix
   (Toyobo). Gene expression levels were normalized with beta-2
   microglobulin (B2M) mRNA expression levels of corresponding cDNA
   samples. All PCR primers were purchased from Bioneer (Daejeon, Korea).
   The following primers were used: rARRB1 (Forward
   5′-ACCTGGATGTCTTGGGTCTG-3′ -Reverse 3’-TAGCCGAGTCAGTGGCTTCT-5′, hSSTR2
   (forward 5′-TGAGAAGAAGGTCACCCGAAT-3′, reverse
   3′-AGGACCACCACAAAGTCAAAC-′) and hB2M (forward
   ′-TTACTCACGTCATCCAGCAGA-3′, reverse 3’-AGAAAGACCAGTCCTTGCTGA-′).

Cell viability assay and flow cytometry analysis

   Cell viability was assessed using the CCK8 assay. INS1 cells were
   seeded at 1 × 10^5 cells/ml in a 96-well plate, incubated for 24 h, and
   treated under various conditions. To measure cell viability, 10 μL/well
   of CCK8 reagent (Dojindo Laboratories, Kumamoto, Japan) was added to
   each well and the plates were incubated in a humidified incubator at
   37°C for 30 min. Absorbance at 440 and 640 nm was measured using a
   Spectra Max190 microplate reader (Molecular Devices, Sunnyvale, CA,
   USA). Cell viability for 3D spheroids was assessed in a CellTiter-Glo®
   3D Cell Viability Assay (Promega, Madison, WI, USA). To quantify
   apoptosis, double staining was performed following the protocol as
   described in Annexin V-FITC Apoptosis Detection Kit (BD Pharmingen™,
   Franklin Lakes, NJ, USA). INS1 cells were collected after incubation
   with mimic miR-15-5p and octretodie and analyzed as described
   previously [[54]11].

Immunocytochemistry

   Immunocytochemistry was performed as previously described [[55]10].
   Cells fixed in 4% paraformaldehyde were blocked with 3% bovine serum
   albumin for 30 min at room temperature (25–27°C) and incubated with
   SSTR2 primary antibody (clone UMBI; 1:500; Abcam, Cambridge, UK) at 4°C
   overnight. The cells were washed, incubated with the appropriate
   fluorescence-conjugated secondary antibody (1:200; Invitrogen) for 1 h
   at room temperature, and counterstained with Hoechst (Invitrogen). The
   coverslips were then mounted, and the cells were observed under a light
   microscope. Images were recorded using an Olympus BX53 microscope with
   Olympus Cell Sens software (Tokyo, Japan).

Western blotting

   Western blotting was performed as previously described [[56]10]. The
   following primary antibodies were used: SSTR2 (clone UMBI; 1:500;
   Abcam) and β-actin (C4; 1:5000; Santa Cruz Biotechnology, Dallas, TX,
   USA). As secondary antibodies, horseradish peroxidase-labeled rabbit
   IgG and mouse IgG antibodies were used (GeneTex, Irvine, CA, USA).
   Protein levels were semi-quantified via densitometry by analyzing the
   blots using Image Lab software (Bio-Rad Laboratories).

Statistical analysis

   A paired two-tailed Student's t-test was used to detect significant
   differences between two sets of data. The differences were considered
   as statistically significant when p values were <0.05.

Results

Identification of differentially expressed miRNAs in INS-1 cells during the
early response to octreotide treatment

   To identify the miRNAs involved in the early response of SSTR2 to
   octreotide treatment, INS-1 cells were treated with 1 μM octreotide for
   various incubation times. As shown in [57]Fig 1A, SSTR2 expression was
   downregulated after 5 min of treatment with 1 μM octreotide.
   Immunocytochemistry revealed that membranous SSTR2 expression was
   reduced, and punctate perinuclear staining of SSTR2 was observed in the
   cytoplasm, suggesting that a 5-min incubation time is sufficient for
   identifying miRNAs associated with the early drug response of SSTR2
   ([58]Fig 1B). Thus, we analyzed the change in miRNA after 5 min
   octreotide treatment. miRNA array analysis showed that 21 miRNAs were
   downregulated and 9 were upregulated in INS-1 cells after 5-min
   treatment with octreotide compared with 0-min treatment (absolute
   fold-change >1.5; [59]Fig 2A and 2B). Furthermore, KEGG pathway
   enrichment analysis revealed that the targets of these miRNAs were
   associated with cancer, mitogen-activated protein kinase (MAPK)
   signaling pathway, and endocrine and other factor-regulated calcium
   reabsorption ([60]S1 Table). The expression of miR-16-5p, which showed
   the largest number of target genes among the 30 differentially
   expressed miRNAs, was verified by qPCR. Similar to the miRNA array
   results, miR-16-5p was upregulated after 5 min of octreotide treatment
   in INS-1 cells ([61]Fig 2C). To confirm these results, we employed
   other neuroendocrine cells (GH3 cells). As shown in [62]Fig 3D,
   miR-16-5p was upregulated in GH3 cells after 5 min incubation with
   octreotide. These results confirmed the effect of octreotide on the
   expression of miR-16-5p. Next, we investigated the function of
   miR-16-5p to SSTR2 expression after octreotide treatment.

Fig 1. Expression of SSTR2 after octreotide treatment in INS-1 cells.

   [63]Fig 1
   [64]Open in a new tab

   (A) INS-1 cells were incubated with 1 μmol/L octreotide. After the
   indicated time, SSTR2 and β-actin levels were analyzed by western
   blotting. The images shown here are cropped and the full-length
   original blots are shown in [65]S6 Fig. (B) INS-1 cells were plated on
   coverslips in 24-well plates. On the next day, the cells were incubated
   with 1 μmol/L octreotide for the indicated time. Cells were fixed and
   processed for immunofluorescence staining of SSTR2 protein. Scale bar:
   20 μm.

Fig 2. Microarray analysis of miRNA expression.

   [66]Fig 2
   [67]Open in a new tab

   miRNA profiling was performed to determine the (A) number and (B)
   profile of differentially expressed miRNAs. Red and green indicate
   significantly upregulated or downregulated miRNAs, respectively. INS-1
   cells (C) and GH3 cells (D) were incubated with 1 μmol/L octreotide.
   After the indicated incubation time, total RNA was isolated, after
   which miR-16-5p levels were measured by qRT-PCR and normalized for
   input RNA using control miRNA (U87) with the 2^−ΔΔCt method. Data
   represent the mean of three independent experiments ± SD; *p < 0.05,
   **p < 0.01.

Fig 3. Expression of SSTR2 after treatment with miR-16-5p mimic or inhibitor
in INS-1 cells.

   [68]Fig 3
   [69]Open in a new tab

   INS-1 cells were transfected with 50 or 100 pmol of mimic miR-16-5p or
   control miRNA. (A) miR-16-5p levels were examined by RT-qPCR at 48 h
   after transfection. miRNA expression of target genes was normalized for
   input RNA using endogenous control miRNA (U87) and calculated by the
   2^−ΔΔCt method. (B) SSTR2, phospho-ERK (Thr202/Tyr204), ERK, and
   β-actin levels were analyzed by western blotting. The images shown here
   are cropped and the full-length original blots are shown in [70]S7 Fig.
   (C) INS-1 cells were plated on coverslips in 24-well plates and
   transfected with 100 pmol of mimic miR-16-5p or control miRNA. After 48
   h, the cells were fixed and processed for immunofluorescent staining of
   SSTR2 protein. Scale bar: 20 μm (D) Fluorescence intensity of SSTR2 was
   measured using IMT i-Solution Industrial Image Analysis Software. (E)
   INS-1 cells were treated with 1 or 10 nmol/L of miR-16-5p inhibitor or
   negative control. miR-16-5p levels at 24 h after treatment were
   determined by qRT-PCR. (F) Western blotting of SSTR2, phospho-ERK
   (Thr202/Tyr204), ERK, and β-actin levels. The images shown here are
   cropped and the full-length original blots are shown in [71]S8 Fig.

miR-16-5p regulates SSTR2 expression

   To investigate the effect of miR-16-5p on SSTR2 expression, INS-1 cells
   were transfected with an miR-16-5p mimic to upregulate the expression
   of miR-16-5p; increased expression of miR-16-5p was confirmed by
   RT-qPCR ([72]Fig 3A). Interestingly, total SSTR2 expression was
   significantly upregulated following the increase in miR-16-5p levels
   ([73]Fig 3B). Moreover, the pERK/ERK ratio was upregulated in
   miR-16-5p–overexpressing INS-1 cells ([74]Fig 3B); phosphorylation of
   ERK1/2 is predominantly associated with SSTR2 activation [[75]12]. To
   examine the localization of increased SSTR2, immunocytochemistry was
   performed. Both membranous and cytoplasmic expression of SSTR2 was
   upregulated in miR-16-5p mimic-transfected cells ([76]Fig 3C and 3D).
   We further employed a miR-16-5p inhibitor to downregulate miR-16-5p
   expression in INS-1 cells ([77]Fig 3E) and found that INS-1 cells
   treated with the miR-16-5p inhibitor showed reduced miR-16-5p
   expression, resulting in downregulation of SSTR2 expression ([78]Fig
   3F). In addition, downregulated SSTR2 was observed in has-miR16-5p
   inhibitor-treated SSTR2-overexpressing HeLa cells by immunostaining
   ([79]S2 Fig). Moreover, the pERK/ERK ratio was reduced after treatment
   with the inhibitor, suggesting reduced SSTR2 activation ([80]Fig 3F).
   These results demonstrate that miR-16-5p is associated with the
   regulation of SSTR2 expression.

miR-16-5p upregulation increases sensitivity to octreotide

   Because SSTR2 expression was found to be regulated by miR-16-5p, we
   assessed the effect of miR-16-5p on SSTR2 expression after octreotide
   treatment. INS-1 cells transfected with a miR-16-5p mimic or control
   miRNA were treated with octreotide for a short incubation time.
   Compared with control miRNA-transfected INS-1 cells, INS-1 cells
   exhibiting upregulation of miR-16-5p showed higher SSTR2 levels before
   and after octreotide treatment ([81]Fig 4A and 4B). We next evaluated
   whether the increased SSTR2 levels induced by the miR-16-5p mimic
   affected the anti-cancer properties of octreotide. INS-1 cells
   transfected with the miR-16-5p mimic showed reduced cell proliferation
   after octreotide treatment ([82]Fig 5A). To confirm the synergistic
   anti-cancer effect of combined treatment, annexin V-PI staining was
   performed. Annexin V-PI staining demonstrated that combined treatment
   with the mimic miR-16-5p and octreotide increased early and late
   apoptosis ([83]S4 Fig). Spheroids from 3-dimensional (3D) culture
   systems have been used as models to evaluate drug sensitivity [[84]13,
   [85]14]. Thus, we used spheroids to examine the role of miR-16-5p in
   octreotide sensitivity. The proliferation of miR-16-5p
   mimic-transfected INS-1 cells was reduced compared to that of spheroids
   from control miRNA-transfected INS-1 cells after 48 h octreotide
   treatment, demonstrating an increased sensitivity to octreotide
   ([86]Fig 5B). In parallel, immunostaining for SSTR2 was performed.
   Similar to 2D culture, SSTR2 expression was upregulated and maintained
   in spheroids from miR-16-5p mimic-transfected INS-1 cells ([87]S5 Fig).
   These findings suggest a role for miR-16-5p expression in mediating the
   antiproliferative action of octreotide.

Fig 4. Upregulation of SSTR2 expression in INS-1 cells after octreotide
treatment with mimic miR-16-5p.

   [88]Fig 4
   [89]Open in a new tab

   (A) INS-1 cells were transfected with 100 pmol of mimic miR-16-5p or
   control miRNA for 24 h and then treated with 1 μmol/L octreotide for
   the indicated durations. The cells were then fixed and processed for
   immunofluorescence staining of SSTR2 protein. Scale bar: 20 μm. (B)
   Fluorescence intensity of SSTR2 was measured with IMT i-Solution
   Industrial Image Analysis Software.

Fig 5. Sensitivity of INS-1 cells to octreotide after treatment with mimic
miR-16-5p.

   [90]Fig 5
   [91]Open in a new tab

   (A) INS-1 cells were transfected with 2.5 pmol of mimic miR-16-5p or
   control miRNA for 24 h and treated with the indicated concentration of
   octreotide for 24 h. Cell proliferation was measured in a CCK-8 assay.
   (B) Spheroids from miR-16-5p mimic- or control-transfected INS-1 cells
   were treated with octreotide for 48 h. The proliferation of six
   spheroids were measured by CellTiter-Glo® 3D Cell Viability Assay.
   Values represent the mean of three independent experiments ± SD; *p <
   0.05, **p < 0.01.

Discussion

   The absence or low expression of SSTRs in NETs has been identified as
   the main cause of NET resistance to the current generation of SSAs.
   Therefore, to enhance NET cell sensitivity to octreotide, we used miRNA
   profiling and identified the miRNA associated with the initial response
   to octreotide treatment. Notably, our results showed that modulation of
   miR-16-5p levels affects SSTR2 expression and cell sensitivity to
   octreotide, suggesting that it can be used in combination with SSAs to
   improve their therapeutic effect via regulation of SSTR2 expression.

   Few studies have determined the expression patterns of SSA-related
   miRNAs. Recently, Døssing et al. identified distinct dysregulation of
   miRNAs, including miR-7 and miR-148a, in the human neuroendocrine tumor
   cell line NCI-H727 after treatment with SSA [[92]15]. Another study
   revealed that circulating miR-200a is involved in metastasis during
   SSA-treated small intestine NET progression [[93]16]. However, little
   is known about the dysregulation of miRNAs during the early response to
   SSA. Thus, our study focused on miRNA expression patterns following
   treatment of cells with SSA under a shorter incubation period than that
   used in the two previous studies cited above. Only a few dysregulated
   miRNAs overlapped between our study and previous studies. Although the
   incubation time with octreotide was short, upregulation of miRNAs
   associated with the anticancer effect was found by KEGG analysis in our
   study. Moreover, our results showed that the predicted target genes of
   the 30 differentially expressed miRNAs were enriched in nine pathways,
   of which “microRNAs in cancer” and “MAPK signaling pathway” were the
   most prominent. Inhibition of cell proliferation by SSA through SSTRs
   involves several signal transduction pathways including control of the
   ERK/MAPK pathway [[94]17]. Induction of ERK/MAPK pathway activation by
   the miR-16-5p-induced upregulation of SSTR2 was also observed in this
   study, suggesting that regulation of SSTR2 by miR-16-5p affects
   downstream signaling. Thus, we identified a miRNA involved in both
   early regulation of SSTR2 expression and the anticancer effect of
   octreotide.

   Internalization of SSTR2 and intracellular trafficking are involved in
   SSTR desensitization. Recently, Gatto et al. suggested that low
   expression of ARRB1/β-arrestin 1 in pituitary adenoma is associated
   with reduced SSTR2 desensitization, leading to an improved response to
   SSA [[95]18]. In our study, experiments focused on SSTR2
   internalization and intracellular trafficking have not been performed.
   However, according to immunostaining, miR-16-5p overexpression may not
   only induce an increase in basal SSTR2 expression but also the
   internalization of SSTR2 under exposure to octreotide compared with
   that in the control. Indeed, miR-16-5p regulation in INS-1 cells also
   influenced the levels of Arrb1 mRNA, suggesting that miR-16-5p is
   involved in SSTR2 internalization and desensitization ([96]S3 Fig).

   SSTR2 regulation by chemical compounds has not been widely examined.
   Previous studies showed that histone deacetylase inhibitors and DNA
   methyltransferase inhibitors upregulate Sstr2 in NET cells [[97]19].
   Our study revealed that upregulation of SSTR2 occurs partly via
   miR-16-5p in NET cells. Although we did not investigate the direct
   chemical compound targets of miR-16-5p, enrichment analysis suggested
   that some genes targeted by miR-16-5p are associated with the
   proteasome-ubiquitin system, including USP25. Thus, chemical compounds
   targeting the proteasome are potential candidates for modulating the
   expression of miR-16-5p. Therefore, further studies are needed to
   verify whether miR-16-5p directly binds to proteasome-related genes and
   investigate the effect of proteasome-related chemicals on miR-16-5p.

   Several preclinical studies have shown that proteasome inhibitors
   impair proliferation and induce apoptosis in other neuroendocrine
   tumors such as pituitary tumors [[98]20–[99]22]. Thus, our results may
   be easy to apply in clinical settings to inhibit the reduction in SSTR2
   expression by SSA. Moreover, SSTRs have been exploited as targets for
   NET diagnosis [[100]23, [101]24]. Positron emission tomography imaging
   targeting SSTRs, such as imaging using ^68Ga-labeled SSAs (DOTATOC,
   DOTATATE, and DOTANOC), was recently reported as a useful imaging
   modality for diagnosing NET and selecting patients for SSA treatment
   [[102]25]. Thus, upregulation of SSTR2 by miR-16-5p may be effective in
   patients with low SSTR2 levels by enhancing positron emission
   tomography imaging-based diagnosis.

   Collectively, our study highlights a novel role for miR-16-5p in
   regulating the expression of SSTR2 in NET cells. We also demonstrated
   that upregulation of mir16-5p enhances the sensitivity of octreotide in
   INS-1 cells, suggesting the potential of using octreotide with
   miR-16-5p as a novel combination therapy for NET treatment. Moreover,
   as desensitization and downregulation of SSTR2 after SSA treatment is
   an important clinical issue, the ability to modulate SSTR expression
   via miR-16-5p in patients with NET is clinically relevant, as it would
   improve the therapeutic response or detection sensitivity.

Supporting information

   S1 Fig. Expression of SSTR2 in INS1 and GH3 cells.

   (A) INS1 and GH3 cells were plated on the coverslip. After 24 h, the
   cells were then fixed and processed for immunofluorescence staining of
   SSTR2 protein and counterstaining with DAPI.

   (DOCX)
   [103]Click here for additional data file.^ (206.6KB, docx)
   S2 Fig. Expression of SSTR2 after treatment with has-miR-16-5p
   inhibitor in hSSTR2-transfected HeLa cells.

   HeLa cells were transfected with 1 μg of hSSTR2 or control plasmid for
   24 h and then treated with the indicated concentration of has-miR16-5p
   inhibitor. (A) Cells were fixed and processed for immunofluorescence
   staining of SSTR2 protein. Scale bar: 50 μm. (B) Expression levels of
   has-miR-16-5p were determined after 24 h treatment by qRT-PCR. Data
   represent the mean of three independent experiments ± SD.

   (DOCX)
   [104]Click here for additional data file.^ (214.4KB, docx)
   S3 Fig. Expression of ARRB1 after modulation of miR-16-5p in INS1
   cells.

   (A) Ins1 cells were transfected with mimic miR-16-5p or control. ARRB1
   expression levels were determined after 24 h treatment by qRT-PCR. (B)
   (A) Ins1 cells were treated with miR16-5p inhibitor or control. ARRB1
   expression levels were determined after 24 h treatment by qRT-PCR. Data
   represent the mean of three independent experiments ± SD.

   (DOCX)
   [105]Click here for additional data file.^ (84.2KB, docx)
   S4 Fig. Analysis of apoptosis by Annexin V-APC/propidium iodide (PI)
   double staining of INS1 cells after treatment under the indicated
   conditions.

   Two-color flow cytometry dot plots show the percentages of living cells
   as negative for both annexin V and PI; early-stage apoptotic cells were
   Annexin V-positive and PI-negative, and late-stage apoptotic/necrotic
   cells were double-positive cells.

   (DOCX)
   [106]Click here for additional data file.^ (182KB, docx)
   S5 Fig. INS-1 cells were transfected with 100 pmol of mimic miR-16-5p
   or control miRNA for 24 h and treated with 10 nmol/L octreotide for 24
   h.

   Spheroids were then fixed with 4% PFA overnight at 4°C.
   Immunofluorescence staining of SSTR2 protein was performed. Scale bar:
   100 μm.

   (DOCX)
   [107]Click here for additional data file.^ (997.8KB, docx)
   S6 Fig. Full-length original blots of [108]Fig 1A.

   We used ChemiDoc XRS (Biorad), which enables direct digital
   visualization of chemiluminescent western blots for the image of
   signals accumulated in the chemiluminescence reaction.

   (DOCX)
   [109]Click here for additional data file.^ (157.7KB, docx)
   S7 Fig. Full-length original blots of [110]Fig 3B.

   We used ChemiDoc XRS (Biorad), which enables direct digital
   visualization of chemiluminescent western blots for the image of
   signals accumulated in the chemiluminescence reaction.

   (DOCX)
   [111]Click here for additional data file.^ (447.2KB, docx)
   S8 Fig. Full-length original blots of [112]Fig 3F (Red box).

   We used ChemiDoc XRS (Biorad), which enables direct digital
   visualization of chemiluminescent western blots for the image of
   signals accumulated in the chemiluminescence reaction.

   (DOCX)
   [113]Click here for additional data file.^ (344.4KB, docx)
   S1 Table. KEGG pathway enrichment analysis of miRNA targets in INS-1
   cells after 5-min treatment with octreotide vs. 0-min treatment.

   (DOCX)
   [114]Click here for additional data file.^ (13.1KB, docx)

Data Availability

   All relevant data are within the manuscript and its Supporting
   Information files.

Funding Statement

   This research was supported by the National Research Foundation of
   Korea (grant numbers NRF-2015R1D1A1A01057737 and NRF-2018R1C1B6003894)
   and Incheon National University Research Grant in 2018.

References