Imbalanced Expression of IGF2 and PCSK4 Is Associated With Overproduction of Big IGF2 in SFT With NICTH: A Pilot Study

Imbalanced Expression of IGF2 and PCSK4 Is Associated With Overproduction of Big IGF2 in SFT With... Abstract Context Nonislet cell tumor hypoglycemia (NICTH) is a rare but serious paraneoplastic syndrome associated with large tumors. The high molecular weight IGF2, known as “big” IGF2, is produced by culprit tumors and leads to severe hypoglycemia. The detailed mechanism of its production in NICTH, however, remains unclear. Objective To clarify the mechanism of production of big IGF2 in light of the processing of pro-IGF2 in patients with solitary fibrous tumor (SFT) and NICTH. Design We enrolled 14 patients with SFT and divided them based on the presence or absence of hypoglycemia. In light of the processing of pro-IGF2 in SFT with hypoglycemia, we, retrospectively, compared the production levels of big IGF2 and the expression levels of IGF2 and proprotein convertase subtilisin/kexin type 4 (PCSK4), a proteolytic enzyme of pro-IGF2. Results In all patients with NICTH, big IGF2 was detected in serum by western immunoblotting analysis. Moreover, we showed that two patients without hypoglycemia also had a small amount of big IGF2 in their serum. By immunohistochemical analysis, the protein expression level of IGF2 was significantly higher in the NICTH group than in the non-NICTH group (P = 0.043). The IGF2/PCSK4 protein expression-level ratio in the NICTH group was significantly higher than that in the non-NICTH group (P = 0.021). Conclusion In patients with SFT and hypoglycemia, an imbalance of IGF2 and PCSK4 expression could lead to increased serum levels of big IGF2. Nonislet cell tumor hypoglycemia (NICTH) is a rare but serious paraneoplastic syndrome associated with large tumors (1). Although it is well known that a wide variety of tumor types can give rise to NICTH (2), tumors of mesenchymal origin are reported to be the most common cause of NICTH (3). High molecular weight IGF2, known as “big” IGF2, is produced by these tumors and leads to severe hypoglycemia through the activation of insulin receptors. The detailed mechanism of the production of big IGF2 in NICTH, however, remains unclear. IGF2 is a 7.5-kDa peptide, consisting of 67 amino acid residues, and is derived by post-transcriptional processing of the precursor molecule pro-IGF2 (4). Proprotein convertase subtilisin/kexin type 4 (PCSK4) is thought to be a critical pro-IGF2 convertase, and its transcripts are detectable in reproductive tissues, such as the testes, ovaries, and placenta (5). An in vitro study of HTR8/SVneo cells showed that specific inhibition of PCSK4 caused the accumulation of big IGF2 (6). Moreover, in a previous case report of NICTH, expression of PCSK4 mRNA in the tumor tissue was decreased compared with that in normal placental tissue (7). These reports indicate that even in most cases of NICTH, the post-transcriptional processing of pro-IGF2 is thought to be impaired, resulting in the production of a variety of pro-IGF2 peptides, namely big IGF2, that range in size from 11 to 18 kDa. However, there are no studies comparing pro-IGF2 processing with and without NICTH. Among mesenchymal tumors, solitary fibrous tumor (SFT) is one of the most frequently associated with NICTH (3). It is a rare soft tissue neoplasm, with an incidence of ∼0.2 per 100,000/year (8). Although SFT has been reported at almost every extrapleural anatomic site, the most common sites from which SFTs arise are the pleura and abdomen (9). Histologically, SFT characteristically appears as a circumscribed tumor comprising variably cellular and patternless distributions of bland spindle to ovoid cells within a prominent collagenous stroma, with diffuse expression of CD34 (10). Recent reports have shown that the pathogenesis of SFT is consistently associated with NAB2-STAT6 fusion genes. This gene fusion can convert a transcriptional repressor (NAB2) into a transcriptional activator (NAB2-STAT6) of mitogenic pathways that can lead to elevated expression levels of early growth response 1 (EGR1) target genes, including IGF2, H19, and RRAD (11). We recruited 14 patients with SFT and divided them based on the presence or absence of hypoglycemia. To elucidate the mechanisms of production of big IGF2, in light of the processing of pro-IGF2 in patients with SFT and NICTH, we, retrospectively, evaluated and compared the production levels of big IGF2 with the expression levels of IGF2 and PCSK4 between the two groups. Materials and Methods This retrospective study was approved by the Wakayama Medical University Ethics Committee, and all patients were enrolled after giving informed, written consent. Patients Fourteen patients with SFT who visited the Wakayama Medical University Hospital between May 2003 and February 2017 or the Kobe City Medical Center General Hospital or Tenri Hospital between May 2016 and February 2017 were recruited in this study. Their tumors were identified by CT, and all patients were histologically diagnosed with SFT by a preoperative biopsy. The pathological diagnosis of SFT was confirmed by two or more pathologists in each hospital, based on the tumor’s characteristic histological appearance, including the diffuse expression of CD34 and signal transducer and activator of transcription 6 (STAT6)-positive staining in tumor cell nuclei (10, 12). The clinical data of all 14 patients in this study are shown in Supplemental Table 1. Nine (#1 to #9) of the 14 patients had no episodes of hypoglycemia during their clinical course, whereas the five remaining patients (#10 to #14) exhibited symptoms of severe hypoglycemia. The former nine patients were defined as the non-NICTH group, and the latter five patients were defined as the NICTH group. Preoperative sera were available in eight patients (#1 to #5 and #10 to #12). Frozen, resected tumor tissues were available in six patients (#1 to #3, #5, #10, and #11), and formalin-fixed, paraffin-embedded (FFPE) tissues were available in 12 patients (#1 to #3, #5 to #11, #13, and #14). Clinical data Serum glucose (GA08 III; A&T Co., Kanagawa, Japan), HbA1c (HLC-723; Tosoh Co., Tokyo, Japan), immunoreactive insulin (IRI; AIA-CL2400; Tosoh Co.), GH (Daiichi Radioisotope Laboratory, Tokyo, Japan), and IGF1 (Daiichi Radioisotope Laboratory) were measured by commercially available assays before tumor resection. Serum IGF2 concentrations were measured using human IGF2 ELISA (RMEE30; Mediagnost, Reutlingen, Germany), according to the manufacturer’s instructions. Absorbance was measured at 450 nm in a microplate reader, and the average values were calculated for each set of triplicate samples. The tumor volume of each patient except patient #4 was calculated using CT data as follows: [tumor volume (cubic centimeters)] = [major axis (centimeters)] × [minor axis (centimeters)]2 × 0.5. The tumor volume of patient #4 could not be calculated because of multiple postoperative bone metastases and no primary lesion. Western immunoblotting analysis (the detection of serum big IGF2) In eight patients with SFT (#1 to #5 and #10 to #12), western immunoblotting (WIB) of IGF2 in preoperative serum was performed, according to the modified methods of Enjoh et al. (13). In brief, 87.5% acid–ethanol-extracted serum samples were electrophoresed on 15%/5% SDS-PAGE gel under nonreducing conditions. The size-fractioned proteins were electroblotted onto a nitrocellulose sheet. The sheet was blocked with 5% skim milk in Tris-buffered saline-Tween 20 and then incubated with anti-IGF2 antibody (05-166; MilliporeSigma, St. Louis, MO). After extensive washing, the sheet was incubated with horseradish peroxidase-conjugated anti-mouse IgG, and then IGF2–anti-IGF2 antibody complexes were detected with ImmunoStar LD (290-69904; Fujifilm Wako Pure Chemical, Chuo-Ku, Osaka, Japan). The band intensities of IGF2 were calculated using CS analyzer version 3.00.1023 (ATTO Co., Ltd., Tokyo, Japan). qRT-PCR, RT-PCR, and sequencing In six patients with SFT (#1 to #3, #5, #9, and #10), total RNA was extracted from the resected tumors by Trizol Reagent (15596026; Thermo Fisher Scientific, Waltham, MA), and cDNA was synthesized by TaqMan RT Reagents (N808-0234; Thermo Fisher Scientific), following the methods used by Uraki et al. (14). IGF2 and PCSK4 mRNA levels were quantified with real-time quantitative RT-PCR (qRT-PCR) using fluorescent SYBR Green technology (4309155; Thermo Fisher Scientific). PCR primers were synthesized (Japan Bio Services Co., Ltd., Saitama, Japan). Human IGF2 primers for RT-PCR were the following: sense strand 5′-TCCTCCCTGGACAATCAGAC-3′ and anti-sense strand 5′-AGAAGCACCAGCATCGACTT-3′. Human PCSK4 primers were the following: sense strand 5′-GACCTGGAGATCTCGCTCAC-3′ and anti-sense strand 5′-ACGTCCCCGTGTTGAAATAG-3′. The mRNA level of each target sequence was normalized by the GAPDH mRNA level, which was used as an endogenous, internal control. Human GAPDH primers were the following: sense strand 5′- GAAGGTGAAGGTCGGAGTCA-3′ and anti-sense strand 5′-GAAGATGGTGATGGGATTTC-3′. Furthermore, RT-PCR, for the detection of NAB2-STAT6 fusion transcripts, was performed. NAB2 exon 4 forward primer was the following: 5′-CCCTCCACTGAAGAAGCTGA-3′. STAT6 exon 3 reverse primer was the following: 5′-AGCCAGTCACCCAGAAGATG-3′. All of the NAB2-STAT6 fusion transcripts were sequenced using a BigDye Terminator v3.1/1.1 Cycle Sequencing Kit and an ABI Prism 3130 GeneticAnalyzer (Thermo Fisher Scientific). Immunohistochemistry Sections (4 μ) were prepared from FFPE tissue blocks of 12 patients with SFT (#1 to #3, #5 to #11, #13, and #14). For each patient, one section was stained with hematoxylin and eosin to confirm the presence of representative tumor cells in each core. Immunostaining was performed, according to the methods of Uraki et al. (15), using the peroxidase substrate kit 3,3′-diaminobenzidine (DAB; Vector Laboratories, Burlingame, CA). Each section from the 11 patients was stained with anti-STAT6 antibody (sc-621; Santa Cruz Biotechnology, Dallas, TX). Strong and diffuse nuclear expression of STAT6, a specific feature of SFT, was identified. Sections were also stained with an anti-IGF2 goat polyclonal antibody (AF-292-NA; R&D Systems, Minneapolis, MN) and an anti-PCSK4 rabbit polyclonal antibody (NBP1-88010; Novus Biologicals, Littleton, CO). The anti-IGF2 antibody recognizes both big IGF2 and mature IGF2. The mean intensity of DAB-stained cells in one field of view, selected at random, was measured by microscope (BZ-X700; Keyence, Tokyo, Japan). Statistical analysis All data were analyzed using JMP Pro 13 for Windows (SAS Institute Japan, Tokyo, Japan). Student’s t tests were used for comparing the means of two independent or paired samples. P < 0.05 was regarded as statistically significant. Results Age, sex, laboratory data, and tumor volume were compared between the NICTH and non-NICTH groups (Table 1). In the NICTH group, fasting plasma glucose (42 ± 19 mg/dL) was significantly lower than in the non-NICTH group (91 ± 12 mg/dL; P < 0.001). HbA1c was also significantly lower (4.8 ± 0.5% vs 5.8 ± 0.3%, respectively; P = 0.001), as was the serum IRI level (0.6 ± 0.4 µU/mL vs 5.6 ± 3.8 µU/mL, respectively; P = 0.020) and the IGF1 level (58 ± 45 ng/mL vs 130 ± 46 ng/mL, respectively; P = 0.047). The serum IGF2 concentration in the NICTH group was also significantly lower than in the non-NICTH group (487 ± 145 ng/mL vs 681 ± 72 ng/mL, respectively; P = 0.041), suggesting that this ELISA kit mainly measured mature IGF2. Tumor size was significantly larger in the NICTH group than in the non-NICTH group (2500 ± 1600 cm3 vs 180 ± 290 cm3, respectively; P = 0.002). Except patient #2, all patients with SFT showed a gene fusion between NAB2 exon 4 and STAT6 exon 3 (Fig. 1A and 1B). All of the transcript fusions were at defined exon boundaries (Fig. 1C). Table 1. Comparison of Clinical Data Between the Non-NICTH Group and NICTH Group Non-NICTH (n = 9) NICTH (n = 5) Means ± SD n Means ± SD n P Value Sex, male/female 7/2 8 2/3 5 0.217 Age, y 60 ± 17 8 68 ± 19 5 0.453 FPG, mg/dL 90 ± 13 7 42 ± 19 5 <0.001 A1C, % 5.7 ± 0.2 6 4.8 ± 0.5 5 0.003 IRI, µU/mL 4.1 ± 1.8 4 0.6 ± 0.4 5 0.004 GH, ng/mL 1.4 ± 1.3 4 0.2 ± 0.1 4 0.117 IGF1, ng/mL 130 ± 51 4 58 ± 45 5 0.047 IGF2, ng/mL 681 ± 72 5 487 ± 145 3 0.041 Tumor size, cm3 610 ± 1300 7 4300 ± 4000 4 0.002 Non-NICTH (n = 9) NICTH (n = 5) Means ± SD n Means ± SD n P Value Sex, male/female 7/2 8 2/3 5 0.217 Age, y 60 ± 17 8 68 ± 19 5 0.453 FPG, mg/dL 90 ± 13 7 42 ± 19 5 <0.001 A1C, % 5.7 ± 0.2 6 4.8 ± 0.5 5 0.003 IRI, µU/mL 4.1 ± 1.8 4 0.6 ± 0.4 5 0.004 GH, ng/mL 1.4 ± 1.3 4 0.2 ± 0.1 4 0.117 IGF1, ng/mL 130 ± 51 4 58 ± 45 5 0.047 IGF2, ng/mL 681 ± 72 5 487 ± 145 3 0.041 Tumor size, cm3 610 ± 1300 7 4300 ± 4000 4 0.002 P values were obtained using the t test; P < 0.05 was accepted as significant and is indicated in bold. Abbreviations: A1C, HbA1c; FPG, fasting plasma glucose. View Large Table 1. Comparison of Clinical Data Between the Non-NICTH Group and NICTH Group Non-NICTH (n = 9) NICTH (n = 5) Means ± SD n Means ± SD n P Value Sex, male/female 7/2 8 2/3 5 0.217 Age, y 60 ± 17 8 68 ± 19 5 0.453 FPG, mg/dL 90 ± 13 7 42 ± 19 5 <0.001 A1C, % 5.7 ± 0.2 6 4.8 ± 0.5 5 0.003 IRI, µU/mL 4.1 ± 1.8 4 0.6 ± 0.4 5 0.004 GH, ng/mL 1.4 ± 1.3 4 0.2 ± 0.1 4 0.117 IGF1, ng/mL 130 ± 51 4 58 ± 45 5 0.047 IGF2, ng/mL 681 ± 72 5 487 ± 145 3 0.041 Tumor size, cm3 610 ± 1300 7 4300 ± 4000 4 0.002 Non-NICTH (n = 9) NICTH (n = 5) Means ± SD n Means ± SD n P Value Sex, male/female 7/2 8 2/3 5 0.217 Age, y 60 ± 17 8 68 ± 19 5 0.453 FPG, mg/dL 90 ± 13 7 42 ± 19 5 <0.001 A1C, % 5.7 ± 0.2 6 4.8 ± 0.5 5 0.003 IRI, µU/mL 4.1 ± 1.8 4 0.6 ± 0.4 5 0.004 GH, ng/mL 1.4 ± 1.3 4 0.2 ± 0.1 4 0.117 IGF1, ng/mL 130 ± 51 4 58 ± 45 5 0.047 IGF2, ng/mL 681 ± 72 5 487 ± 145 3 0.041 Tumor size, cm3 610 ± 1300 7 4300 ± 4000 4 0.002 P values were obtained using the t test; P < 0.05 was accepted as significant and is indicated in bold. Abbreviations: A1C, HbA1c; FPG, fasting plasma glucose. View Large Figure 1. View largeDownload slide Validation of NAB2-STAT6 gene fusion in SFT. (A) RT-PCR using primers for NAB2 exon 4 and STAT6 exon 3 (RNA was extracted from frozen samples). (B) RT-PCR using primers for NAB2 exon 4 and STAT6 exon 3 (RNA was extracted from FFPE). (C) Representative sequence of the RT-PCR product (patient #10) shows the chimeric fusion between NAB2 exon 4 and STAT6 exon 3. Figure 1. View largeDownload slide Validation of NAB2-STAT6 gene fusion in SFT. (A) RT-PCR using primers for NAB2 exon 4 and STAT6 exon 3 (RNA was extracted from frozen samples). (B) RT-PCR using primers for NAB2 exon 4 and STAT6 exon 3 (RNA was extracted from FFPE). (C) Representative sequence of the RT-PCR product (patient #10) shows the chimeric fusion between NAB2 exon 4 and STAT6 exon 3. In eight patients with SFT (five without NICTH and three with NICTH), whose preoperative serum was available, WIB analysis was performed to investigate the serum levels of big IGF2 (Fig. 2A). As in previous reports, in all three patients with NICTH, most IGF2 was detected at 11 to 18 kDa (big IGF2), with a small amount observed at 7.5 kDa (mature IGF2). Interestingly, the non-NICTH group included two patients (#4 and #5) in whom serum big IGF2 was obviously detected. The band intensity of IGF2 was significantly greater in the NICTH group than the non-NICTH group (P = 0.004; Fig. 2B). The intensity of mature IGF2 in the NICTH group, however, was significantly suppressed (P = 0.042; Fig. 2C). In the NICTH group, the big IGF2/mature IGF2 intensity ratio was significantly higher than in the non-NICTH group (P = 0.012; Fig. 2D). These results indicate that the processing of pro-IGF2 in patients with NICTH is much more impaired than in patients without NICTH. Figure 2. View largeDownload slide Serum IGF2 isoform analysis. (A) WIB analysis of serum IGF2 in eight patients with SFT. Patients #1 to #5 are patients with SFT without hypoglycemia, and patients #10 to #12 are patients with SFT and hypoglycemia. Molecular size markers (in kilodaltons) are indicated by lines on the left. (B) Comparison of big IGF2 production levels between the non-NICTH group and the NICTH group. (C) Comparison of mature IGF2 production levels between the non-NICTH group and the NICTH group. (D) Comparison of the big IGF2/mature IGF2 production-level ratio between the non-NICTH group and the NICTH group. P values were obtained using the t test; P < 0.05 was accepted as significant in all panels. Figure 2. View largeDownload slide Serum IGF2 isoform analysis. (A) WIB analysis of serum IGF2 in eight patients with SFT. Patients #1 to #5 are patients with SFT without hypoglycemia, and patients #10 to #12 are patients with SFT and hypoglycemia. Molecular size markers (in kilodaltons) are indicated by lines on the left. (B) Comparison of big IGF2 production levels between the non-NICTH group and the NICTH group. (C) Comparison of mature IGF2 production levels between the non-NICTH group and the NICTH group. (D) Comparison of the big IGF2/mature IGF2 production-level ratio between the non-NICTH group and the NICTH group. P values were obtained using the t test; P < 0.05 was accepted as significant in all panels. Furthermore, in six patients with SFT (two with NICTH and four without NICTH), whose postoperative, frozen tumor tissues were available, qRT-PCR was performed to evaluate the gene-expression levels in tumor tissues of IGF2 and PCSK4, the latter of which is a proteolytic enzyme of pro-IGF2. The expression level of IGF2 mRNA in the NICTH group was 8.2-fold higher than in the non-NICTH group (P = 0.092; tentative value because of the limited number of samples; Fig. 3A). The expression level of PCSK4 mRNA in the NICTH group was 0.48-fold lower than in the non-NICTH group (P = 0.217; tentative value; Fig. 3B). The IGF2/PCSK4 expression ratio in the NICTH group was 10.4-fold higher than in the non-NICTH group (P = 0.006; tentative value; Fig. 3C). These results indicate that an imbalance of IGF2 and PCSK4 expression leads to impaired processing of pro-IGF2, resulting in the production of big IGF2 by tumor tissues. In the 12 patients with SFT whose FFPE tissues were available, immunohistochemistry for IGF2 and PCSK4 was performed, and protein expression levels were evaluated to measure the intensity of DAB-stained cells. The protein expression level of IGF2 was significantly higher in the NICTH group than in the non-NICTH group [51.4 (49.6 to 53.2) vs 38.4 (34.4 to 42.4), respectively; P = 0.043; Fig. 4C]. Although the protein expression level of PCSK4 in the NICTH group was lower than in the non-NICTH group [16.8 (14.6 to 19.0) vs 24.1 (20.0 to 28.2), respectively; P = 0.228; Fig. 4D], the expression levels were not significantly different between the two cohorts. As with the results of qRT-PCR, the IGF2/PCSK4 protein expression ratio in the NICTH group was significantly higher than in the non-NICTH group [3.3 (2.8 to 3.8) vs 1.8 (1.5 to 2.1); P = 0.021; Fig. 4E]. These results also support the hypothesis that an imbalance of IGF2 and PCSK4 expression impairs the processing of pro-IGF2, contributing to the production of big IGF2 by tumor tissues. Figure 3. View largeDownload slide Comparative analysis of gene-expression levels in SFT using qRT-PCR. (A) IGF2 mRNA expression levels in the non-NICTH group and the NICTH group. (B) PCSK4 mRNA expression levels in the non-NICTH group and the NICTH group. (C) The IGF2/PCSK4 mRNA expression-level ratio in the non-NICTH group and the NICTH group. *P = 0.093, **P = 0.217, ***P = 0.006. P values were obtained using the t test. These values are not statistically significant but tentative as a result of the limited number of samples. Figure 3. View largeDownload slide Comparative analysis of gene-expression levels in SFT using qRT-PCR. (A) IGF2 mRNA expression levels in the non-NICTH group and the NICTH group. (B) PCSK4 mRNA expression levels in the non-NICTH group and the NICTH group. (C) The IGF2/PCSK4 mRNA expression-level ratio in the non-NICTH group and the NICTH group. *P = 0.093, **P = 0.217, ***P = 0.006. P values were obtained using the t test. These values are not statistically significant but tentative as a result of the limited number of samples. Figure 4. View largeDownload slide Comparative analysis of protein expression levels in SFT by immunohistochemistry analysis. (A) Representative images of immunohistochemistry for IGF2. (B) Representative images of immunohistochemistry for PCSK4. (C) Comparison of IGF2 expression levels between the non-NICTH group and the NICTH group. (D) Comparison of PCSK4 expression levels between the non-NICTH group and the NICTH group. (E) Comparison of the IGF2/PCSK4 expression-level ratio between the non-NICTH group and the NICTH group. P values were obtained using the t test; (C and E) P < 0.05 was accepted as significant and is indicated in bold. Figure 4. View largeDownload slide Comparative analysis of protein expression levels in SFT by immunohistochemistry analysis. (A) Representative images of immunohistochemistry for IGF2. (B) Representative images of immunohistochemistry for PCSK4. (C) Comparison of IGF2 expression levels between the non-NICTH group and the NICTH group. (D) Comparison of PCSK4 expression levels between the non-NICTH group and the NICTH group. (E) Comparison of the IGF2/PCSK4 expression-level ratio between the non-NICTH group and the NICTH group. P values were obtained using the t test; (C and E) P < 0.05 was accepted as significant and is indicated in bold. Discussion Although most SFT cases demonstrate an overproduction of IGF2 (16), hypoglycemia has been reported to emerge in only 4% of patients with SFT (17). Many studies have demonstrated that the presence of big IGF2 in the serum is a crucial factor for developing hypoglycemia in cases of NICTH (18, 19). In this study, we sought to identify the underlying mechanism responsible for big IGF2 production in patients with SFT by analyzing these patients separately on the basis of the presence or absence of hypoglycemia. Consistent with previous reports, we detected big IGF2 in the serum of all patients with NICTH. Meanwhile, we initially expected that big IGF2 would not be identified in the serum of patients without NICTH. Contrary to our expectations, big IGF2 was observed in the serum of two patients with SFT without hypoglycemia. These patients may not have developed hypoglycemia, as the serum levels of big IGF2 were significantly lower than in patients with NICTH, as shown by WIB analysis. The precise mechanism by which big IGF2 appears in the serum of patients with NICTH remains unclear. At least three factors are thought to be necessary for the secretion of big IGF2: (1) overproduction of pro-IGF2; (2) decreased expression or activity of PCSK4, a proteolytic enzyme of pro-IGF2; and (3) decreased big IGF2 clearance from the circulation. Increased tumor production of pro-IGF2 contributes to elevated levels of big IGF2 in the serum. With regard to SFT specifically, a recent study showed that NAB2-STAT6 gene fusion, which is involved in the pathogenesis of SFT, induced the expression of EGR1 target genes, including IGF2, H19, and RRAD (11). We confirmed the NAB2 exon 4-STAT6 exon 3 fusion in 11 of 12 patients. Previous reports showed that this fusion variant corresponded to classic pleuropulmonary SFT, as we found in the current study (12, 20). IGF2 mRNA expression levels in patients with SFT and NICTH were markedly elevated, and a significant positive correlation between STAT6 and IGF2 mRNA expression levels was observed in our study (data not shown). These results suggest that NAB2-STAT6 gene fusion causes overproduction of pro-IGF2 in patients with SFT patients. Regarding NAB2-STAT6 gene fusion, different variants are recently described, and those are thought to be associated with the expression levels of IGF2 in SFT. In our study, however, as the tumors of all patients but patient #2 had the same fusion gene, we found no correlation between fusion variants and increased serum big IGF2. With respect to patients with NICTH who have tumors other than SFT, loss of imprinting of the IGF2 gene was reported to be associated with increased IGF2 expression (21, 22). Tumor overproduction of pro-IGF2 might exceed the capacity of the pro-IGF2 processing enzyme, resulting in the appearance of unprocessed IGF2 or big IGF2 in the serum. If big IGF2 were produced through this mechanism, then a certain amount of mature IGF2 should also be detected in the serum of patients with NICTH. In the current study, however, WIB analysis showed lower serum levels of mature IGF2 in patients with NICTH than in patients without NICTH, and serum IGF2 concentrations were also significantly lower. It seems that tumor overproduction of pro-IGF2 may not be the only cause of big IGF2 in the serum of patients with NICTH. In the current study, the big IGF2/mature IGF2 serum ratio in the NICTH group was significantly higher than in the non-NICTH group, indicating that the processing of pro-IGF2 is impaired in patients with NICTH. Decreased expression of PCSK4 mRNA in tumor cells might be associated with increased serum big IGF2 in patients with NICTH. A previous case report of NICTH showed lower expression of PCSK4 mRNA in SFT tissue than in normal placental tissue (7). Moreover, Qiu et al. (6) demonstrated that intracellular pro-IGF2 processing was blocked by a PCSK4-specific inhibitor. However, both mRNA and protein expression levels of PCSK4 in the NICTH group were similar to those in the non-NICTH group in the current study. Although overexpression of EGR1, as a result of the NAB2-STAT6 fusion gene, is part of the pathogenesis of SFT, we could not find the EGR1 binding site within at least 5 kb of the 5′-upstream sequence of PCSK4. A previous report identified 16 genes, including Rad, EF-1A2, NSE, PDGF-A, TGF-B1, as well as IGF2 as EGR1 target genes (23). However, to our best knowledge, there was no report that these molecules regulate PCSK4 expression. These findings suggest that increased serum big IGF2 is not a result of an absolute decrease of PCSK4 mRNA expression but a result of decreased activity of the processing enzyme PCSK4. We could not examine enzymatic activity using frozen tumor samples. Therefore, further studies are needed to ascertain whether PCSK4 activity is reduced in NICTH tumors. Type 2 IGF receptor plays a role in the clearance of IGF2 from the circulation. It is known that big IGF2 binds less readily to type 2 IGF receptor than mature IGF2 (24). Therefore, in patients with SFT and hypoglycemia, the clearance of big IGF2 might be decreased, which could lead to increased serum levels of big IGF2. The limitation of the current pilot study should also be noted; because of the retrospective single-center analysis, the sample size of this study was too small to show statistical significance adequately. Therefore, additional prospective studies with a large cohort are needed. In conclusion, we demonstrated an imbalance between IGF2 and PCSK4 expression in patients with SFT and hypoglycemia, which is associated with the increased levels of big IGF2 in their serum. Further studies are needed to evaluate PCSK4 activity in NICTH tumors. We also showed that some patients without hypoglycemia could nonetheless have a small amount of big IGF2 in their serum. Abbreviations: Abbreviations: DAB 3,3′-diaminobenzidine FFPE formalin-fixed, paraffin-embedded IRI immunoreactive insulin NICTH nonislet cell tumor hypoglycemia PCSK4 proprotein convertase subtilisin/kexin type 4 SFT solitary fibrous tumor STAT6 signal transducer and activator of transcription 4 WIB western immunoblotting Acknowledgments The authors thank Izumi Fukuda for the technique of WIB analysis. Disclosure Summary: The authors have nothing to disclose. References 1. de Groot JW , Rikhof B , van Doorn J , Bilo HJ , Alleman MA , Honkoop AH , van der Graaf WT . Non-islet cell tumour-induced hypoglycaemia: a review of the literature including two new cases . Endocr Relat Cancer . 2007 ; 14 ( 4 ): 979 – 993 . 2. Zapf J . Insulinlike growth factor binding proteins and tumor hypoglycemia . Trends Endocrinol Metab . 1995 ; 6 ( 2 ): 37 – 42 . 3. Bodnar TW , Acevedo MJ , Pietropaolo M . Management of non-islet-cell tumor hypoglycemia: a clinical review . J Clin Endocrinol Metab . 2014 ; 99 ( 3 ): 713 – 722 . 4. Livingstone C . IGF2 and cancer . Endocr Relat Cancer . 2013 ; 20 ( 6 ): R321 – R339 . 5. Gyamera-Acheampong C , Mbikay M . Proprotein convertase subtilisin/kexin type 4 in mammalian fertility: a review . Hum Reprod Update . 2009 ; 15 ( 2 ): 237 – 247 . 6. Qiu Q , Basak A , Mbikay M , Tsang BK , Gruslin A . Role of pro-IGF-II processing by proprotein convertase 4 in human placental development . Proc Natl Acad Sci USA . 2005 ; 102 ( 31 ): 11047 – 11052 . 7. Tani Y , Tateno T , Izumiyama H , Doi M , Yoshimoto T , Hirata Y . Defective expression of prohormone convertase 4 and enhanced expression of insulin-like growth factor II by pleural solitary fibrous tumor causing hypoglycemia . Endocr J . 2008 ; 55 ( 5 ): 905 – 911 . 8. Baldi GG , Stacchiotti S , Mauro V , Dei Tos AP , Gronchi A , Pastorino U , Duranti L , Provenzano S , Marrari A , Libertini M , Pilotti S , Casali PG . Solitary fibrous tumor of all sites: outcome of late recurrences in 14 patients . Clin Sarcoma Res . 2013 ; 3 ( 1 ): 4 . 9. Demicco EG , Park MS , Araujo DM , Fox PS , Bassett RL , Pollock RE , Lazar AJ , Wang WL . Solitary fibrous tumor: a clinicopathological study of 110 cases and proposed risk assessment model . Mod Pathol . 2012 ; 25 ( 9 ): 1298 – 1306 . 10. Thway K , Ng W , Noujaim J , Jones RL , Fisher C . The current status of solitary fibrous tumor: diagnostic features, variants and genetics . Int J Surg Pathol . 2016 ; 24 ( 4 ): 281 – 292 . 11. Robinson DR , Wu YM , Kalyana-Sundaram S , Cao X , Lonigro RJ , Sung YS , Chen CL , Zhang L , Wang R , Su F , Iyer MK , Roychowdhury S , Siddiqui J , Pienta KJ , Kunju LP , Talpaz M , Mosquera JM , Singer S , Schuetze SM , Antonescu CR , Chinnaiyan AM . Identification of recurrent NAB2-STAT6 gene fusions in solitary fibrous tumor by integrative sequencing . Nat Genet . 2013 ; 45 ( 2 ): 180 – 185 . 12. Vogels RJ , Vlenterie M , Versleijen-Jonkers YM , Ruijter E , Bekers EM , Verdijk MA , Link MM , Bonenkamp JJ , van der Graaf WT , Slootweg PJ , Suurmeijer AJ , Groenen PJ , Flucke U . Solitary fibrous tumor - clinicopathologic, immunohistochemical and molecular analysis of 28 cases . Diagn Pathol . 2014 ; 9 ( 1 ): 224 . 13. Enjoh T , Hizuka N , Perdue JF , Takano K , Fujiwara H , Higashihashi N , Marumoto Y , Fukuda I , Sakano K . Characterization of new monoclonal antibodies to human insulin-like growth factor-II and their application in western immunoblot analysis . J Clin Endocrinol Metab . 1993 ; 77 ( 2 ): 510 – 517 . 14. Uraki S , Ariyasu H , Doi A , Furuta H , Nishi M , Sugano K , Inoshita N , Nakao N , Yamada S , Akamizu T . Atypical pituitary adenoma with MEN1 somatic mutation associated with abnormalities of DNA mismatch repair genes; MLH1 germline mutation and MSH6 somatic mutation . Endocr J . 2017 ; 64 ( 9 ): 895 – 906 . 15. Uraki S , Ariyasu H , Doi A , Kawai S , Takeshima K , Morita S , Fukai J , Fujita K , Furuta H , Nishi M , Sugano K , Inoshita N , Nakao N , Yamada S , Akamizu T . Reduced expression of mismatch repair genes MSH6/MSH2 directly promotes pituitary tumor growth via the ATR-Chk1 pathway . J Clin Endocrinol Metab . 2018 ; 103 ( 3 ): 1171 – 1179 . 16. Steigen SE , Schaeffer DF , West RB , Nielsen TO . Expression of insulin-like growth factor 2 in mesenchymal neoplasms . Mod Pathol . 2009 ; 22 ( 7 ): 914 – 921 . 17. Briselli M , Mark EJ , Dickersin GR . Solitary fibrous tumors of the pleura: eight new cases and review of 360 cases in the literature . Cancer . 1981 ; 47 ( 11 ): 2678 – 2689 . 18. Hizuka N , Fukuda I , Takano K , Okubo Y , Asakawa-Yasumoto K , Demura H . Serum insulin-like growth factor II in 44 patients with non-islet cell tumor hypoglycemia . Endocr J . 1998 ; 45 ( Suppl ): S61 – S65 . 19. Famà F , Le Bouc Y , Barrande G , Villeneuve A , Berry MG , Pidoto RR , Saint Marc O . Solitary fibrous tumour of the liver with IGF-II-related hypoglycaemia. A case report . Langenbecks Arch Surg . 2008 ; 393 ( 4 ): 611 – 616 . 20. Barthelmeß S , Geddert H , Boltze C , Moskalev EA , Bieg M , Sirbu H , Brors B , Wiemann S , Hartmann A , Agaimy A , Haller F . Solitary fibrous tumors/hemangiopericytomas with different variants of the NAB2-STAT6 gene fusion are characterized by specific histomorphology and distinct clinicopathological features . Am J Pathol . 2014 ; 184 ( 4 ): 1209 – 1218 . 21. Reeve AE , Eccles MR , Wilkins RJ , Bell GI , Millow LJ . Expression of insulin-like growth factor-II transcripts in Wilms’ tumour . Nature . 1985 ; 317 ( 6034 ): 258 – 260 . 22. Ogawa O , Eccles MR , Szeto J , McNoe LA , Yun K , Maw MA , Smith PJ , Reeve AE . Relaxation of insulin-like growth factor II gene imprinting implicated in Wilms’ tumour . Nature . 1993 ; 362 ( 6422 ): 749 – 751 . 23. Svaren J , Ehrig T , Abdulkadir SA , Ehrengruber MU , Watson MA , Milbrandt J . EGR1 target genes in prostate carcinoma cells identified by microarray analysis . J Biol Chem . 2000 ; 275 ( 49 ): 38524 – 38531 . 24. Greenall SA , Bentley JD , Pearce LA , Scoble JA , Sparrow LG , Bartone NA , Xiao X , Baxter RC , Cosgrove LJ , Adams TE . Biochemical characterization of individual human glycosylated pro-insulin-like growth factor (IGF)-II and big-IGF-II isoforms associated with cancer . J Biol Chem . 2013 ; 288 ( 1 ): 59 – 68 . Copyright © 2018 Endocrine Society http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Clinical Endocrinology and Metabolism Oxford University Press

Imbalanced Expression of IGF2 and PCSK4 Is Associated With Overproduction of Big IGF2 in SFT With NICTH: A Pilot Study

Loading next page...
 
/lp/ou_press/imbalanced-expression-of-igf2-and-pcsk4-is-associated-with-hk4USUPFGf
Publisher
Oxford University Press
Copyright
Copyright © 2018 Endocrine Society
ISSN
0021-972X
eISSN
1945-7197
D.O.I.
10.1210/jc.2018-00593
Publisher site
See Article on Publisher Site

Abstract

Abstract Context Nonislet cell tumor hypoglycemia (NICTH) is a rare but serious paraneoplastic syndrome associated with large tumors. The high molecular weight IGF2, known as “big” IGF2, is produced by culprit tumors and leads to severe hypoglycemia. The detailed mechanism of its production in NICTH, however, remains unclear. Objective To clarify the mechanism of production of big IGF2 in light of the processing of pro-IGF2 in patients with solitary fibrous tumor (SFT) and NICTH. Design We enrolled 14 patients with SFT and divided them based on the presence or absence of hypoglycemia. In light of the processing of pro-IGF2 in SFT with hypoglycemia, we, retrospectively, compared the production levels of big IGF2 and the expression levels of IGF2 and proprotein convertase subtilisin/kexin type 4 (PCSK4), a proteolytic enzyme of pro-IGF2. Results In all patients with NICTH, big IGF2 was detected in serum by western immunoblotting analysis. Moreover, we showed that two patients without hypoglycemia also had a small amount of big IGF2 in their serum. By immunohistochemical analysis, the protein expression level of IGF2 was significantly higher in the NICTH group than in the non-NICTH group (P = 0.043). The IGF2/PCSK4 protein expression-level ratio in the NICTH group was significantly higher than that in the non-NICTH group (P = 0.021). Conclusion In patients with SFT and hypoglycemia, an imbalance of IGF2 and PCSK4 expression could lead to increased serum levels of big IGF2. Nonislet cell tumor hypoglycemia (NICTH) is a rare but serious paraneoplastic syndrome associated with large tumors (1). Although it is well known that a wide variety of tumor types can give rise to NICTH (2), tumors of mesenchymal origin are reported to be the most common cause of NICTH (3). High molecular weight IGF2, known as “big” IGF2, is produced by these tumors and leads to severe hypoglycemia through the activation of insulin receptors. The detailed mechanism of the production of big IGF2 in NICTH, however, remains unclear. IGF2 is a 7.5-kDa peptide, consisting of 67 amino acid residues, and is derived by post-transcriptional processing of the precursor molecule pro-IGF2 (4). Proprotein convertase subtilisin/kexin type 4 (PCSK4) is thought to be a critical pro-IGF2 convertase, and its transcripts are detectable in reproductive tissues, such as the testes, ovaries, and placenta (5). An in vitro study of HTR8/SVneo cells showed that specific inhibition of PCSK4 caused the accumulation of big IGF2 (6). Moreover, in a previous case report of NICTH, expression of PCSK4 mRNA in the tumor tissue was decreased compared with that in normal placental tissue (7). These reports indicate that even in most cases of NICTH, the post-transcriptional processing of pro-IGF2 is thought to be impaired, resulting in the production of a variety of pro-IGF2 peptides, namely big IGF2, that range in size from 11 to 18 kDa. However, there are no studies comparing pro-IGF2 processing with and without NICTH. Among mesenchymal tumors, solitary fibrous tumor (SFT) is one of the most frequently associated with NICTH (3). It is a rare soft tissue neoplasm, with an incidence of ∼0.2 per 100,000/year (8). Although SFT has been reported at almost every extrapleural anatomic site, the most common sites from which SFTs arise are the pleura and abdomen (9). Histologically, SFT characteristically appears as a circumscribed tumor comprising variably cellular and patternless distributions of bland spindle to ovoid cells within a prominent collagenous stroma, with diffuse expression of CD34 (10). Recent reports have shown that the pathogenesis of SFT is consistently associated with NAB2-STAT6 fusion genes. This gene fusion can convert a transcriptional repressor (NAB2) into a transcriptional activator (NAB2-STAT6) of mitogenic pathways that can lead to elevated expression levels of early growth response 1 (EGR1) target genes, including IGF2, H19, and RRAD (11). We recruited 14 patients with SFT and divided them based on the presence or absence of hypoglycemia. To elucidate the mechanisms of production of big IGF2, in light of the processing of pro-IGF2 in patients with SFT and NICTH, we, retrospectively, evaluated and compared the production levels of big IGF2 with the expression levels of IGF2 and PCSK4 between the two groups. Materials and Methods This retrospective study was approved by the Wakayama Medical University Ethics Committee, and all patients were enrolled after giving informed, written consent. Patients Fourteen patients with SFT who visited the Wakayama Medical University Hospital between May 2003 and February 2017 or the Kobe City Medical Center General Hospital or Tenri Hospital between May 2016 and February 2017 were recruited in this study. Their tumors were identified by CT, and all patients were histologically diagnosed with SFT by a preoperative biopsy. The pathological diagnosis of SFT was confirmed by two or more pathologists in each hospital, based on the tumor’s characteristic histological appearance, including the diffuse expression of CD34 and signal transducer and activator of transcription 6 (STAT6)-positive staining in tumor cell nuclei (10, 12). The clinical data of all 14 patients in this study are shown in Supplemental Table 1. Nine (#1 to #9) of the 14 patients had no episodes of hypoglycemia during their clinical course, whereas the five remaining patients (#10 to #14) exhibited symptoms of severe hypoglycemia. The former nine patients were defined as the non-NICTH group, and the latter five patients were defined as the NICTH group. Preoperative sera were available in eight patients (#1 to #5 and #10 to #12). Frozen, resected tumor tissues were available in six patients (#1 to #3, #5, #10, and #11), and formalin-fixed, paraffin-embedded (FFPE) tissues were available in 12 patients (#1 to #3, #5 to #11, #13, and #14). Clinical data Serum glucose (GA08 III; A&T Co., Kanagawa, Japan), HbA1c (HLC-723; Tosoh Co., Tokyo, Japan), immunoreactive insulin (IRI; AIA-CL2400; Tosoh Co.), GH (Daiichi Radioisotope Laboratory, Tokyo, Japan), and IGF1 (Daiichi Radioisotope Laboratory) were measured by commercially available assays before tumor resection. Serum IGF2 concentrations were measured using human IGF2 ELISA (RMEE30; Mediagnost, Reutlingen, Germany), according to the manufacturer’s instructions. Absorbance was measured at 450 nm in a microplate reader, and the average values were calculated for each set of triplicate samples. The tumor volume of each patient except patient #4 was calculated using CT data as follows: [tumor volume (cubic centimeters)] = [major axis (centimeters)] × [minor axis (centimeters)]2 × 0.5. The tumor volume of patient #4 could not be calculated because of multiple postoperative bone metastases and no primary lesion. Western immunoblotting analysis (the detection of serum big IGF2) In eight patients with SFT (#1 to #5 and #10 to #12), western immunoblotting (WIB) of IGF2 in preoperative serum was performed, according to the modified methods of Enjoh et al. (13). In brief, 87.5% acid–ethanol-extracted serum samples were electrophoresed on 15%/5% SDS-PAGE gel under nonreducing conditions. The size-fractioned proteins were electroblotted onto a nitrocellulose sheet. The sheet was blocked with 5% skim milk in Tris-buffered saline-Tween 20 and then incubated with anti-IGF2 antibody (05-166; MilliporeSigma, St. Louis, MO). After extensive washing, the sheet was incubated with horseradish peroxidase-conjugated anti-mouse IgG, and then IGF2–anti-IGF2 antibody complexes were detected with ImmunoStar LD (290-69904; Fujifilm Wako Pure Chemical, Chuo-Ku, Osaka, Japan). The band intensities of IGF2 were calculated using CS analyzer version 3.00.1023 (ATTO Co., Ltd., Tokyo, Japan). qRT-PCR, RT-PCR, and sequencing In six patients with SFT (#1 to #3, #5, #9, and #10), total RNA was extracted from the resected tumors by Trizol Reagent (15596026; Thermo Fisher Scientific, Waltham, MA), and cDNA was synthesized by TaqMan RT Reagents (N808-0234; Thermo Fisher Scientific), following the methods used by Uraki et al. (14). IGF2 and PCSK4 mRNA levels were quantified with real-time quantitative RT-PCR (qRT-PCR) using fluorescent SYBR Green technology (4309155; Thermo Fisher Scientific). PCR primers were synthesized (Japan Bio Services Co., Ltd., Saitama, Japan). Human IGF2 primers for RT-PCR were the following: sense strand 5′-TCCTCCCTGGACAATCAGAC-3′ and anti-sense strand 5′-AGAAGCACCAGCATCGACTT-3′. Human PCSK4 primers were the following: sense strand 5′-GACCTGGAGATCTCGCTCAC-3′ and anti-sense strand 5′-ACGTCCCCGTGTTGAAATAG-3′. The mRNA level of each target sequence was normalized by the GAPDH mRNA level, which was used as an endogenous, internal control. Human GAPDH primers were the following: sense strand 5′- GAAGGTGAAGGTCGGAGTCA-3′ and anti-sense strand 5′-GAAGATGGTGATGGGATTTC-3′. Furthermore, RT-PCR, for the detection of NAB2-STAT6 fusion transcripts, was performed. NAB2 exon 4 forward primer was the following: 5′-CCCTCCACTGAAGAAGCTGA-3′. STAT6 exon 3 reverse primer was the following: 5′-AGCCAGTCACCCAGAAGATG-3′. All of the NAB2-STAT6 fusion transcripts were sequenced using a BigDye Terminator v3.1/1.1 Cycle Sequencing Kit and an ABI Prism 3130 GeneticAnalyzer (Thermo Fisher Scientific). Immunohistochemistry Sections (4 μ) were prepared from FFPE tissue blocks of 12 patients with SFT (#1 to #3, #5 to #11, #13, and #14). For each patient, one section was stained with hematoxylin and eosin to confirm the presence of representative tumor cells in each core. Immunostaining was performed, according to the methods of Uraki et al. (15), using the peroxidase substrate kit 3,3′-diaminobenzidine (DAB; Vector Laboratories, Burlingame, CA). Each section from the 11 patients was stained with anti-STAT6 antibody (sc-621; Santa Cruz Biotechnology, Dallas, TX). Strong and diffuse nuclear expression of STAT6, a specific feature of SFT, was identified. Sections were also stained with an anti-IGF2 goat polyclonal antibody (AF-292-NA; R&D Systems, Minneapolis, MN) and an anti-PCSK4 rabbit polyclonal antibody (NBP1-88010; Novus Biologicals, Littleton, CO). The anti-IGF2 antibody recognizes both big IGF2 and mature IGF2. The mean intensity of DAB-stained cells in one field of view, selected at random, was measured by microscope (BZ-X700; Keyence, Tokyo, Japan). Statistical analysis All data were analyzed using JMP Pro 13 for Windows (SAS Institute Japan, Tokyo, Japan). Student’s t tests were used for comparing the means of two independent or paired samples. P < 0.05 was regarded as statistically significant. Results Age, sex, laboratory data, and tumor volume were compared between the NICTH and non-NICTH groups (Table 1). In the NICTH group, fasting plasma glucose (42 ± 19 mg/dL) was significantly lower than in the non-NICTH group (91 ± 12 mg/dL; P < 0.001). HbA1c was also significantly lower (4.8 ± 0.5% vs 5.8 ± 0.3%, respectively; P = 0.001), as was the serum IRI level (0.6 ± 0.4 µU/mL vs 5.6 ± 3.8 µU/mL, respectively; P = 0.020) and the IGF1 level (58 ± 45 ng/mL vs 130 ± 46 ng/mL, respectively; P = 0.047). The serum IGF2 concentration in the NICTH group was also significantly lower than in the non-NICTH group (487 ± 145 ng/mL vs 681 ± 72 ng/mL, respectively; P = 0.041), suggesting that this ELISA kit mainly measured mature IGF2. Tumor size was significantly larger in the NICTH group than in the non-NICTH group (2500 ± 1600 cm3 vs 180 ± 290 cm3, respectively; P = 0.002). Except patient #2, all patients with SFT showed a gene fusion between NAB2 exon 4 and STAT6 exon 3 (Fig. 1A and 1B). All of the transcript fusions were at defined exon boundaries (Fig. 1C). Table 1. Comparison of Clinical Data Between the Non-NICTH Group and NICTH Group Non-NICTH (n = 9) NICTH (n = 5) Means ± SD n Means ± SD n P Value Sex, male/female 7/2 8 2/3 5 0.217 Age, y 60 ± 17 8 68 ± 19 5 0.453 FPG, mg/dL 90 ± 13 7 42 ± 19 5 <0.001 A1C, % 5.7 ± 0.2 6 4.8 ± 0.5 5 0.003 IRI, µU/mL 4.1 ± 1.8 4 0.6 ± 0.4 5 0.004 GH, ng/mL 1.4 ± 1.3 4 0.2 ± 0.1 4 0.117 IGF1, ng/mL 130 ± 51 4 58 ± 45 5 0.047 IGF2, ng/mL 681 ± 72 5 487 ± 145 3 0.041 Tumor size, cm3 610 ± 1300 7 4300 ± 4000 4 0.002 Non-NICTH (n = 9) NICTH (n = 5) Means ± SD n Means ± SD n P Value Sex, male/female 7/2 8 2/3 5 0.217 Age, y 60 ± 17 8 68 ± 19 5 0.453 FPG, mg/dL 90 ± 13 7 42 ± 19 5 <0.001 A1C, % 5.7 ± 0.2 6 4.8 ± 0.5 5 0.003 IRI, µU/mL 4.1 ± 1.8 4 0.6 ± 0.4 5 0.004 GH, ng/mL 1.4 ± 1.3 4 0.2 ± 0.1 4 0.117 IGF1, ng/mL 130 ± 51 4 58 ± 45 5 0.047 IGF2, ng/mL 681 ± 72 5 487 ± 145 3 0.041 Tumor size, cm3 610 ± 1300 7 4300 ± 4000 4 0.002 P values were obtained using the t test; P < 0.05 was accepted as significant and is indicated in bold. Abbreviations: A1C, HbA1c; FPG, fasting plasma glucose. View Large Table 1. Comparison of Clinical Data Between the Non-NICTH Group and NICTH Group Non-NICTH (n = 9) NICTH (n = 5) Means ± SD n Means ± SD n P Value Sex, male/female 7/2 8 2/3 5 0.217 Age, y 60 ± 17 8 68 ± 19 5 0.453 FPG, mg/dL 90 ± 13 7 42 ± 19 5 <0.001 A1C, % 5.7 ± 0.2 6 4.8 ± 0.5 5 0.003 IRI, µU/mL 4.1 ± 1.8 4 0.6 ± 0.4 5 0.004 GH, ng/mL 1.4 ± 1.3 4 0.2 ± 0.1 4 0.117 IGF1, ng/mL 130 ± 51 4 58 ± 45 5 0.047 IGF2, ng/mL 681 ± 72 5 487 ± 145 3 0.041 Tumor size, cm3 610 ± 1300 7 4300 ± 4000 4 0.002 Non-NICTH (n = 9) NICTH (n = 5) Means ± SD n Means ± SD n P Value Sex, male/female 7/2 8 2/3 5 0.217 Age, y 60 ± 17 8 68 ± 19 5 0.453 FPG, mg/dL 90 ± 13 7 42 ± 19 5 <0.001 A1C, % 5.7 ± 0.2 6 4.8 ± 0.5 5 0.003 IRI, µU/mL 4.1 ± 1.8 4 0.6 ± 0.4 5 0.004 GH, ng/mL 1.4 ± 1.3 4 0.2 ± 0.1 4 0.117 IGF1, ng/mL 130 ± 51 4 58 ± 45 5 0.047 IGF2, ng/mL 681 ± 72 5 487 ± 145 3 0.041 Tumor size, cm3 610 ± 1300 7 4300 ± 4000 4 0.002 P values were obtained using the t test; P < 0.05 was accepted as significant and is indicated in bold. Abbreviations: A1C, HbA1c; FPG, fasting plasma glucose. View Large Figure 1. View largeDownload slide Validation of NAB2-STAT6 gene fusion in SFT. (A) RT-PCR using primers for NAB2 exon 4 and STAT6 exon 3 (RNA was extracted from frozen samples). (B) RT-PCR using primers for NAB2 exon 4 and STAT6 exon 3 (RNA was extracted from FFPE). (C) Representative sequence of the RT-PCR product (patient #10) shows the chimeric fusion between NAB2 exon 4 and STAT6 exon 3. Figure 1. View largeDownload slide Validation of NAB2-STAT6 gene fusion in SFT. (A) RT-PCR using primers for NAB2 exon 4 and STAT6 exon 3 (RNA was extracted from frozen samples). (B) RT-PCR using primers for NAB2 exon 4 and STAT6 exon 3 (RNA was extracted from FFPE). (C) Representative sequence of the RT-PCR product (patient #10) shows the chimeric fusion between NAB2 exon 4 and STAT6 exon 3. In eight patients with SFT (five without NICTH and three with NICTH), whose preoperative serum was available, WIB analysis was performed to investigate the serum levels of big IGF2 (Fig. 2A). As in previous reports, in all three patients with NICTH, most IGF2 was detected at 11 to 18 kDa (big IGF2), with a small amount observed at 7.5 kDa (mature IGF2). Interestingly, the non-NICTH group included two patients (#4 and #5) in whom serum big IGF2 was obviously detected. The band intensity of IGF2 was significantly greater in the NICTH group than the non-NICTH group (P = 0.004; Fig. 2B). The intensity of mature IGF2 in the NICTH group, however, was significantly suppressed (P = 0.042; Fig. 2C). In the NICTH group, the big IGF2/mature IGF2 intensity ratio was significantly higher than in the non-NICTH group (P = 0.012; Fig. 2D). These results indicate that the processing of pro-IGF2 in patients with NICTH is much more impaired than in patients without NICTH. Figure 2. View largeDownload slide Serum IGF2 isoform analysis. (A) WIB analysis of serum IGF2 in eight patients with SFT. Patients #1 to #5 are patients with SFT without hypoglycemia, and patients #10 to #12 are patients with SFT and hypoglycemia. Molecular size markers (in kilodaltons) are indicated by lines on the left. (B) Comparison of big IGF2 production levels between the non-NICTH group and the NICTH group. (C) Comparison of mature IGF2 production levels between the non-NICTH group and the NICTH group. (D) Comparison of the big IGF2/mature IGF2 production-level ratio between the non-NICTH group and the NICTH group. P values were obtained using the t test; P < 0.05 was accepted as significant in all panels. Figure 2. View largeDownload slide Serum IGF2 isoform analysis. (A) WIB analysis of serum IGF2 in eight patients with SFT. Patients #1 to #5 are patients with SFT without hypoglycemia, and patients #10 to #12 are patients with SFT and hypoglycemia. Molecular size markers (in kilodaltons) are indicated by lines on the left. (B) Comparison of big IGF2 production levels between the non-NICTH group and the NICTH group. (C) Comparison of mature IGF2 production levels between the non-NICTH group and the NICTH group. (D) Comparison of the big IGF2/mature IGF2 production-level ratio between the non-NICTH group and the NICTH group. P values were obtained using the t test; P < 0.05 was accepted as significant in all panels. Furthermore, in six patients with SFT (two with NICTH and four without NICTH), whose postoperative, frozen tumor tissues were available, qRT-PCR was performed to evaluate the gene-expression levels in tumor tissues of IGF2 and PCSK4, the latter of which is a proteolytic enzyme of pro-IGF2. The expression level of IGF2 mRNA in the NICTH group was 8.2-fold higher than in the non-NICTH group (P = 0.092; tentative value because of the limited number of samples; Fig. 3A). The expression level of PCSK4 mRNA in the NICTH group was 0.48-fold lower than in the non-NICTH group (P = 0.217; tentative value; Fig. 3B). The IGF2/PCSK4 expression ratio in the NICTH group was 10.4-fold higher than in the non-NICTH group (P = 0.006; tentative value; Fig. 3C). These results indicate that an imbalance of IGF2 and PCSK4 expression leads to impaired processing of pro-IGF2, resulting in the production of big IGF2 by tumor tissues. In the 12 patients with SFT whose FFPE tissues were available, immunohistochemistry for IGF2 and PCSK4 was performed, and protein expression levels were evaluated to measure the intensity of DAB-stained cells. The protein expression level of IGF2 was significantly higher in the NICTH group than in the non-NICTH group [51.4 (49.6 to 53.2) vs 38.4 (34.4 to 42.4), respectively; P = 0.043; Fig. 4C]. Although the protein expression level of PCSK4 in the NICTH group was lower than in the non-NICTH group [16.8 (14.6 to 19.0) vs 24.1 (20.0 to 28.2), respectively; P = 0.228; Fig. 4D], the expression levels were not significantly different between the two cohorts. As with the results of qRT-PCR, the IGF2/PCSK4 protein expression ratio in the NICTH group was significantly higher than in the non-NICTH group [3.3 (2.8 to 3.8) vs 1.8 (1.5 to 2.1); P = 0.021; Fig. 4E]. These results also support the hypothesis that an imbalance of IGF2 and PCSK4 expression impairs the processing of pro-IGF2, contributing to the production of big IGF2 by tumor tissues. Figure 3. View largeDownload slide Comparative analysis of gene-expression levels in SFT using qRT-PCR. (A) IGF2 mRNA expression levels in the non-NICTH group and the NICTH group. (B) PCSK4 mRNA expression levels in the non-NICTH group and the NICTH group. (C) The IGF2/PCSK4 mRNA expression-level ratio in the non-NICTH group and the NICTH group. *P = 0.093, **P = 0.217, ***P = 0.006. P values were obtained using the t test. These values are not statistically significant but tentative as a result of the limited number of samples. Figure 3. View largeDownload slide Comparative analysis of gene-expression levels in SFT using qRT-PCR. (A) IGF2 mRNA expression levels in the non-NICTH group and the NICTH group. (B) PCSK4 mRNA expression levels in the non-NICTH group and the NICTH group. (C) The IGF2/PCSK4 mRNA expression-level ratio in the non-NICTH group and the NICTH group. *P = 0.093, **P = 0.217, ***P = 0.006. P values were obtained using the t test. These values are not statistically significant but tentative as a result of the limited number of samples. Figure 4. View largeDownload slide Comparative analysis of protein expression levels in SFT by immunohistochemistry analysis. (A) Representative images of immunohistochemistry for IGF2. (B) Representative images of immunohistochemistry for PCSK4. (C) Comparison of IGF2 expression levels between the non-NICTH group and the NICTH group. (D) Comparison of PCSK4 expression levels between the non-NICTH group and the NICTH group. (E) Comparison of the IGF2/PCSK4 expression-level ratio between the non-NICTH group and the NICTH group. P values were obtained using the t test; (C and E) P < 0.05 was accepted as significant and is indicated in bold. Figure 4. View largeDownload slide Comparative analysis of protein expression levels in SFT by immunohistochemistry analysis. (A) Representative images of immunohistochemistry for IGF2. (B) Representative images of immunohistochemistry for PCSK4. (C) Comparison of IGF2 expression levels between the non-NICTH group and the NICTH group. (D) Comparison of PCSK4 expression levels between the non-NICTH group and the NICTH group. (E) Comparison of the IGF2/PCSK4 expression-level ratio between the non-NICTH group and the NICTH group. P values were obtained using the t test; (C and E) P < 0.05 was accepted as significant and is indicated in bold. Discussion Although most SFT cases demonstrate an overproduction of IGF2 (16), hypoglycemia has been reported to emerge in only 4% of patients with SFT (17). Many studies have demonstrated that the presence of big IGF2 in the serum is a crucial factor for developing hypoglycemia in cases of NICTH (18, 19). In this study, we sought to identify the underlying mechanism responsible for big IGF2 production in patients with SFT by analyzing these patients separately on the basis of the presence or absence of hypoglycemia. Consistent with previous reports, we detected big IGF2 in the serum of all patients with NICTH. Meanwhile, we initially expected that big IGF2 would not be identified in the serum of patients without NICTH. Contrary to our expectations, big IGF2 was observed in the serum of two patients with SFT without hypoglycemia. These patients may not have developed hypoglycemia, as the serum levels of big IGF2 were significantly lower than in patients with NICTH, as shown by WIB analysis. The precise mechanism by which big IGF2 appears in the serum of patients with NICTH remains unclear. At least three factors are thought to be necessary for the secretion of big IGF2: (1) overproduction of pro-IGF2; (2) decreased expression or activity of PCSK4, a proteolytic enzyme of pro-IGF2; and (3) decreased big IGF2 clearance from the circulation. Increased tumor production of pro-IGF2 contributes to elevated levels of big IGF2 in the serum. With regard to SFT specifically, a recent study showed that NAB2-STAT6 gene fusion, which is involved in the pathogenesis of SFT, induced the expression of EGR1 target genes, including IGF2, H19, and RRAD (11). We confirmed the NAB2 exon 4-STAT6 exon 3 fusion in 11 of 12 patients. Previous reports showed that this fusion variant corresponded to classic pleuropulmonary SFT, as we found in the current study (12, 20). IGF2 mRNA expression levels in patients with SFT and NICTH were markedly elevated, and a significant positive correlation between STAT6 and IGF2 mRNA expression levels was observed in our study (data not shown). These results suggest that NAB2-STAT6 gene fusion causes overproduction of pro-IGF2 in patients with SFT patients. Regarding NAB2-STAT6 gene fusion, different variants are recently described, and those are thought to be associated with the expression levels of IGF2 in SFT. In our study, however, as the tumors of all patients but patient #2 had the same fusion gene, we found no correlation between fusion variants and increased serum big IGF2. With respect to patients with NICTH who have tumors other than SFT, loss of imprinting of the IGF2 gene was reported to be associated with increased IGF2 expression (21, 22). Tumor overproduction of pro-IGF2 might exceed the capacity of the pro-IGF2 processing enzyme, resulting in the appearance of unprocessed IGF2 or big IGF2 in the serum. If big IGF2 were produced through this mechanism, then a certain amount of mature IGF2 should also be detected in the serum of patients with NICTH. In the current study, however, WIB analysis showed lower serum levels of mature IGF2 in patients with NICTH than in patients without NICTH, and serum IGF2 concentrations were also significantly lower. It seems that tumor overproduction of pro-IGF2 may not be the only cause of big IGF2 in the serum of patients with NICTH. In the current study, the big IGF2/mature IGF2 serum ratio in the NICTH group was significantly higher than in the non-NICTH group, indicating that the processing of pro-IGF2 is impaired in patients with NICTH. Decreased expression of PCSK4 mRNA in tumor cells might be associated with increased serum big IGF2 in patients with NICTH. A previous case report of NICTH showed lower expression of PCSK4 mRNA in SFT tissue than in normal placental tissue (7). Moreover, Qiu et al. (6) demonstrated that intracellular pro-IGF2 processing was blocked by a PCSK4-specific inhibitor. However, both mRNA and protein expression levels of PCSK4 in the NICTH group were similar to those in the non-NICTH group in the current study. Although overexpression of EGR1, as a result of the NAB2-STAT6 fusion gene, is part of the pathogenesis of SFT, we could not find the EGR1 binding site within at least 5 kb of the 5′-upstream sequence of PCSK4. A previous report identified 16 genes, including Rad, EF-1A2, NSE, PDGF-A, TGF-B1, as well as IGF2 as EGR1 target genes (23). However, to our best knowledge, there was no report that these molecules regulate PCSK4 expression. These findings suggest that increased serum big IGF2 is not a result of an absolute decrease of PCSK4 mRNA expression but a result of decreased activity of the processing enzyme PCSK4. We could not examine enzymatic activity using frozen tumor samples. Therefore, further studies are needed to ascertain whether PCSK4 activity is reduced in NICTH tumors. Type 2 IGF receptor plays a role in the clearance of IGF2 from the circulation. It is known that big IGF2 binds less readily to type 2 IGF receptor than mature IGF2 (24). Therefore, in patients with SFT and hypoglycemia, the clearance of big IGF2 might be decreased, which could lead to increased serum levels of big IGF2. The limitation of the current pilot study should also be noted; because of the retrospective single-center analysis, the sample size of this study was too small to show statistical significance adequately. Therefore, additional prospective studies with a large cohort are needed. In conclusion, we demonstrated an imbalance between IGF2 and PCSK4 expression in patients with SFT and hypoglycemia, which is associated with the increased levels of big IGF2 in their serum. Further studies are needed to evaluate PCSK4 activity in NICTH tumors. We also showed that some patients without hypoglycemia could nonetheless have a small amount of big IGF2 in their serum. Abbreviations: Abbreviations: DAB 3,3′-diaminobenzidine FFPE formalin-fixed, paraffin-embedded IRI immunoreactive insulin NICTH nonislet cell tumor hypoglycemia PCSK4 proprotein convertase subtilisin/kexin type 4 SFT solitary fibrous tumor STAT6 signal transducer and activator of transcription 4 WIB western immunoblotting Acknowledgments The authors thank Izumi Fukuda for the technique of WIB analysis. Disclosure Summary: The authors have nothing to disclose. References 1. de Groot JW , Rikhof B , van Doorn J , Bilo HJ , Alleman MA , Honkoop AH , van der Graaf WT . Non-islet cell tumour-induced hypoglycaemia: a review of the literature including two new cases . Endocr Relat Cancer . 2007 ; 14 ( 4 ): 979 – 993 . 2. Zapf J . Insulinlike growth factor binding proteins and tumor hypoglycemia . Trends Endocrinol Metab . 1995 ; 6 ( 2 ): 37 – 42 . 3. Bodnar TW , Acevedo MJ , Pietropaolo M . Management of non-islet-cell tumor hypoglycemia: a clinical review . J Clin Endocrinol Metab . 2014 ; 99 ( 3 ): 713 – 722 . 4. Livingstone C . IGF2 and cancer . Endocr Relat Cancer . 2013 ; 20 ( 6 ): R321 – R339 . 5. Gyamera-Acheampong C , Mbikay M . Proprotein convertase subtilisin/kexin type 4 in mammalian fertility: a review . Hum Reprod Update . 2009 ; 15 ( 2 ): 237 – 247 . 6. Qiu Q , Basak A , Mbikay M , Tsang BK , Gruslin A . Role of pro-IGF-II processing by proprotein convertase 4 in human placental development . Proc Natl Acad Sci USA . 2005 ; 102 ( 31 ): 11047 – 11052 . 7. Tani Y , Tateno T , Izumiyama H , Doi M , Yoshimoto T , Hirata Y . Defective expression of prohormone convertase 4 and enhanced expression of insulin-like growth factor II by pleural solitary fibrous tumor causing hypoglycemia . Endocr J . 2008 ; 55 ( 5 ): 905 – 911 . 8. Baldi GG , Stacchiotti S , Mauro V , Dei Tos AP , Gronchi A , Pastorino U , Duranti L , Provenzano S , Marrari A , Libertini M , Pilotti S , Casali PG . Solitary fibrous tumor of all sites: outcome of late recurrences in 14 patients . Clin Sarcoma Res . 2013 ; 3 ( 1 ): 4 . 9. Demicco EG , Park MS , Araujo DM , Fox PS , Bassett RL , Pollock RE , Lazar AJ , Wang WL . Solitary fibrous tumor: a clinicopathological study of 110 cases and proposed risk assessment model . Mod Pathol . 2012 ; 25 ( 9 ): 1298 – 1306 . 10. Thway K , Ng W , Noujaim J , Jones RL , Fisher C . The current status of solitary fibrous tumor: diagnostic features, variants and genetics . Int J Surg Pathol . 2016 ; 24 ( 4 ): 281 – 292 . 11. Robinson DR , Wu YM , Kalyana-Sundaram S , Cao X , Lonigro RJ , Sung YS , Chen CL , Zhang L , Wang R , Su F , Iyer MK , Roychowdhury S , Siddiqui J , Pienta KJ , Kunju LP , Talpaz M , Mosquera JM , Singer S , Schuetze SM , Antonescu CR , Chinnaiyan AM . Identification of recurrent NAB2-STAT6 gene fusions in solitary fibrous tumor by integrative sequencing . Nat Genet . 2013 ; 45 ( 2 ): 180 – 185 . 12. Vogels RJ , Vlenterie M , Versleijen-Jonkers YM , Ruijter E , Bekers EM , Verdijk MA , Link MM , Bonenkamp JJ , van der Graaf WT , Slootweg PJ , Suurmeijer AJ , Groenen PJ , Flucke U . Solitary fibrous tumor - clinicopathologic, immunohistochemical and molecular analysis of 28 cases . Diagn Pathol . 2014 ; 9 ( 1 ): 224 . 13. Enjoh T , Hizuka N , Perdue JF , Takano K , Fujiwara H , Higashihashi N , Marumoto Y , Fukuda I , Sakano K . Characterization of new monoclonal antibodies to human insulin-like growth factor-II and their application in western immunoblot analysis . J Clin Endocrinol Metab . 1993 ; 77 ( 2 ): 510 – 517 . 14. Uraki S , Ariyasu H , Doi A , Furuta H , Nishi M , Sugano K , Inoshita N , Nakao N , Yamada S , Akamizu T . Atypical pituitary adenoma with MEN1 somatic mutation associated with abnormalities of DNA mismatch repair genes; MLH1 germline mutation and MSH6 somatic mutation . Endocr J . 2017 ; 64 ( 9 ): 895 – 906 . 15. Uraki S , Ariyasu H , Doi A , Kawai S , Takeshima K , Morita S , Fukai J , Fujita K , Furuta H , Nishi M , Sugano K , Inoshita N , Nakao N , Yamada S , Akamizu T . Reduced expression of mismatch repair genes MSH6/MSH2 directly promotes pituitary tumor growth via the ATR-Chk1 pathway . J Clin Endocrinol Metab . 2018 ; 103 ( 3 ): 1171 – 1179 . 16. Steigen SE , Schaeffer DF , West RB , Nielsen TO . Expression of insulin-like growth factor 2 in mesenchymal neoplasms . Mod Pathol . 2009 ; 22 ( 7 ): 914 – 921 . 17. Briselli M , Mark EJ , Dickersin GR . Solitary fibrous tumors of the pleura: eight new cases and review of 360 cases in the literature . Cancer . 1981 ; 47 ( 11 ): 2678 – 2689 . 18. Hizuka N , Fukuda I , Takano K , Okubo Y , Asakawa-Yasumoto K , Demura H . Serum insulin-like growth factor II in 44 patients with non-islet cell tumor hypoglycemia . Endocr J . 1998 ; 45 ( Suppl ): S61 – S65 . 19. Famà F , Le Bouc Y , Barrande G , Villeneuve A , Berry MG , Pidoto RR , Saint Marc O . Solitary fibrous tumour of the liver with IGF-II-related hypoglycaemia. A case report . Langenbecks Arch Surg . 2008 ; 393 ( 4 ): 611 – 616 . 20. Barthelmeß S , Geddert H , Boltze C , Moskalev EA , Bieg M , Sirbu H , Brors B , Wiemann S , Hartmann A , Agaimy A , Haller F . Solitary fibrous tumors/hemangiopericytomas with different variants of the NAB2-STAT6 gene fusion are characterized by specific histomorphology and distinct clinicopathological features . Am J Pathol . 2014 ; 184 ( 4 ): 1209 – 1218 . 21. Reeve AE , Eccles MR , Wilkins RJ , Bell GI , Millow LJ . Expression of insulin-like growth factor-II transcripts in Wilms’ tumour . Nature . 1985 ; 317 ( 6034 ): 258 – 260 . 22. Ogawa O , Eccles MR , Szeto J , McNoe LA , Yun K , Maw MA , Smith PJ , Reeve AE . Relaxation of insulin-like growth factor II gene imprinting implicated in Wilms’ tumour . Nature . 1993 ; 362 ( 6422 ): 749 – 751 . 23. Svaren J , Ehrig T , Abdulkadir SA , Ehrengruber MU , Watson MA , Milbrandt J . EGR1 target genes in prostate carcinoma cells identified by microarray analysis . J Biol Chem . 2000 ; 275 ( 49 ): 38524 – 38531 . 24. Greenall SA , Bentley JD , Pearce LA , Scoble JA , Sparrow LG , Bartone NA , Xiao X , Baxter RC , Cosgrove LJ , Adams TE . Biochemical characterization of individual human glycosylated pro-insulin-like growth factor (IGF)-II and big-IGF-II isoforms associated with cancer . J Biol Chem . 2013 ; 288 ( 1 ): 59 – 68 . Copyright © 2018 Endocrine Society

Journal

Journal of Clinical Endocrinology and MetabolismOxford University Press

Published: Apr 20, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off