Differential Effects of Somatostatin, Octreotide, and Lanreotide on Neuroendocrine Differentiation and Proliferation in Established and Primary NET Cell Lines: Possible Crosstalk with TGF-β Signaling

Abstract

GEP-NETs are heterogeneous tumors originating from the pancreas (panNET) or the intestinal tract. Only a few patients with NETs are amenable to curative tumor resection, and for most patients, only palliative treatments to successfully control the disease or manage symptoms remain, such as with synthetic somatostatin (SST) analogs (SSAs), such as octreotide (OCT) or lanreotide (LAN). However, even cells expressing low levels of SST receptors (SSTRs) may exhibit significant responses to OCT, which suggests the possibility that SSAs signal through alternative mechanisms, e.g., transforming growth factor (TGF)-β. This signaling mode has been demonstrated in the established panNET line BON but not yet in other permanent (i.e., QGP) or primary (i.e., NT-3) panNET-derived cells. Here, we performed qPCR, immunoblot analyses, and cell counting assays to assess the effects of SST, OCT, LAN, and TGF-β1 on neuroendocrine marker expression and cell proliferation in NT-3, QGP, and BON cells. SST and SSAs were found to regulate a set of neuroendocrine genes in all three cell lines, with the effects of SST, mainly LAN, often differing from those of OCT. However, unlike NT-3 cells, BON cells failed to respond to OCT with growth arrest but paradoxically exhibited a growth-stimulatory effect after treatment with LAN. As previously shown for BON, NT-3 cells responded to TGF-β1 treatment with induction of expression of SST and SSTR2/5. Of note, the ability of NT-3 cells to respond to TGF-β1 with upregulation of the established TGF-β target gene SERPINE1 depended on cellular adherence to a collagen-coated matrix. Moreover, when applied to NT-3 cells for an extended period, i.e., 14 days, TGF-β1 induced growth suppression as shown earlier for BON cells. Finally, next-generation sequencing-based identification of microRNAs (miRNAs) in BON and NT-3 revealed that SST and OCT impact positively or negatively on the regulation of specific miRNAs. Our results suggest that primary panNET cells, such as NT-3, respond similarly as BON cells to SST, SSA, and TGF-β treatment and thus provide circumstantial evidence that crosstalk of SST and TGF-β signaling is not confined to BON cells but is a general feature of panNETs.

Keywords: BON-1 (BON); NT-3; QGP-1 (QGP); gastroenteropancreatic neuroendocrine tumor (GEP-NET); lanreotide (LAN); microRNA (miRNA); octreotide (OCT); somatostatin (SST); somatostatin analogues (SSAs).

Figure 1

Response of various NE markers to treatment with SST or SSAs in BON, QGP, and NT-3 cells. (A) BON, (B) QGP, or (C) NT-3 cells were treated with the indicated agents (BON, QGP: 10 nM or 1 µM; NT-3: 1 µM) for 24 h followed by RNA isolation and qPCR for the indicated NE markers. Data are displayed as mean ± SD after normalization with the housekeeping gene TBP relative to untreated controls (set arbitrarily at 1.0) and are representative of three assays. The asterisks (*) indicate a significant difference (p < 0.05, Student’s t-test). Please note that the scaling on the ordinates may vary among the graphs. (D) BON, QGP, or NT-3 cells were stimulated with the indicated agents for 48 h, followed by lysis and immunoblotting for CgA, SYP, or SSTR2, and either HSP90 or GAPDH as a loading control. The graphs below the blots show results from the quantification of densitometric readings from three blots, each derived from a separate experiment (mean ± SD, n = 3). The asterisks (*) indicate a significant difference (p < 0.05, Wilcoxon test) relative to untreated controls set arbitrarily at 1.0. The vertical lines between the lanes of some blots denote the removal of irrelevant lanes.

Figure 2

Effect of TGF-β1 treatment on expression of SERPINE1, SST, SSTR2, and SSTR5 in BON, QGP, and NT-3 cells. Cells were challenged with TGF-β1 (5 ng/mL) for the indicated times, followed by RNA isolation and qPCR analysis: (A) SERPINE1 in BON and QGP cells. (B) SERPINE1 in adherent (adh.) and non-adherent (non-adh.) NT-3 cells. Data in A-B represent the mean ± SD of four independent assays (p < 0.05, Wilcoxon test) relative to the untreated (0) control cell set at 1.0. (C) SST, SSTR2, and SSTR5 expression in NT-3 cells as measured by qPCR analysis. The data shown are from a representative assay out of three assays performed in total (mean ± SD after normalization to TBP and relative to untreated control cells set at 1.0). The asterisks (*) denote a significant difference (p < 0.05, Student’s t-test).

Figure 3

Effect of SST, SSAs, or TGF-β1 treatment on the proliferative activity of BON, QGP, and NT-3 cells. (A) BON and QGP cells were seeded at a density of 50.000 cells per 12-well on day 1 and treated on day 2 with either SST, OCT, or LAN (each at 1 µM) for 72 h in standard growth medium. Cells in parallel wells were incubated with LY294002 (LY, 50 µM) or U0126 (10 µM), both of which served as controls for growth-inhibitory agents. Data are plotted relative to vehicle-treated control cells set at 100% (not shown) and represent the mean ± SD of 3-5 independent assays (p < 0.05, Wilcoxon test) (B) NT-3 cells were challenged with TGF-β1 (5 ng/mL) for the indicated times (7 or 14 d), and cell counts were subsequently determined. Data shown are the mean ± SD relative to control cells (white-filled bar) set at 100% for both time points (n = 3). The asterisk (*) denotes a significant difference (p < 0.05, Wilcoxon test); ns, non-significant.

Figure 4

miRNA sequencing in BON (A), QGP (B), or NT-3 (C) cells treated with either vehicle, SST, OCT, or LAN. The heatmaps show log₂ fold changes as estimated by DESeq2. Values above 0 (orange/red colors) denote miRNAs upregulated in the vehicle (no treatment) group compared to the treatment group. In contrast, values below 0 (blue colors) represent those upregulated miRNAs in the treatment group. Grey cells indicate that the miRNA was not found (missing values). The left columns show p-values below 0.1 (dark grey, considered significant) or above 0.1 (light grey, considered non-significant).