Plasma Distribution and Signaling Activities of IGF-II Precursors

Alicia G Marks; Julie M Carroll; Jonathan Q Purnell; Charles T Roberts, Jr

doi:10.1210/en.2010-0784

. 2011 Feb 1;152(3):922–930. doi: 10.1210/en.2010-0784

Plasma Distribution and Signaling Activities of IGF-II Precursors

Alicia G Marks ¹, Julie M Carroll ¹, Jonathan Q Purnell ¹, Charles T Roberts Jr ^1,^✉

PMCID: PMC3040053 PMID: 21285309

“Big” and pro-IGF-II are routinely present in human plasma in addition to mature IGF-II, and both activate signaling through the insulin receptor.

Abstract

IGF-II is thought to function through activation of the IGF-I receptor (IGF-IR) and the A isoform of the IR, with the IGF-IR being relevant to tumorigenesis and the IR to both tumorigenesis and metabolic control. In the paraneoplastic syndrome of nonislet cell tumor hypoglycemia, tumor-derived IGF-II has been proposed to exert both proliferative and metabolic effects, exemplifying this dual mode of action. Increased levels of IGF-II precursors (“big” and pro–IGF-II) have been reported in the circulation of nonislet cell tumor patients and have been proposed to exert greater or different effects than mature IGF-II. However, most studies have not defined which version is being investigated, and the relative activation of the IR and IGF-IR by IGF-II precursors has not been delineated. In this study, we determined the distribution of IGF-II isoforms in normal human plasma and their ability to activate the alternative versions of the IR. The majority (71%) of total IGF-II in human plasma was the mature form, while “big” and pro–IGF-II comprised 16% and 13%, respectively, with more variation seen in the levels of mature IGF-II. In IGF-IR–deficient cells expressing similar levels of human IR-A or IR-B, mature and “big” IGF-II exhibited similar activation of IR signaling, while pro–IGF-II exhibited significantly less activation. Downstream activation of Akt by mature and “big” IGF-II was greater in IR-A cells, consistent with previous reports of the greater affinity of IR-A for IGF-II. Thus, both IGF-II precursor forms are present in human plasma but do not preferentially activate the IR.

IGF-II is known to be important in murine prenatal growth. In humans, it is present postnatally at concentrations several-fold higher than IGF-I, but the postnatal role of IGF-II is poorly understood. In 1988, larger forms of IGF-II were first reported in the plasma of patients with nonislet cell tumor hypoglycemia (NICTH) (1), a paraneoplastic syndrome first described in 1929 (2). These larger forms of IGF-II were thought to be the cause of the hypoglycemia, because tumor cells from patients with NICTH contain large amounts of IGF-II mRNA, the proportion of the larger forms of IGF-II relative to total IGF-II in the plasma is elevated, and the levels of the larger forms of IGF-II decrease to normal with resolution of hypoglycemia after tumor removal (1, 3). The larger forms of IGF-II were subsequently defined as pro–IGF-II and “big” IGF-II and found to undergo glycosylation in vivo, while mature IGF-II does not (4, 5).

IGF-II is encoded by the IGF2 gene located at the human chromosomal locus 11p15.5. Pre–pro–IGF-II comprises the N-terminal 24–amino acid signal sequence, the 67–amino acid mature IGF-II peptide, and the 89–amino acid E domain at the C terminus (6). After signal peptide cleavage, pro–IGF-II enters the secretory pathway and is subsequently processed to the 67–amino acid mature IGF-II. Proteolytic cleavage at alternative sites in the E-domain sequence, at positions 104 or 87, produces intermediate forms designated as “big” IGF-II. Both pro–IGF-II and “big” IGF-II have been found in rodent and human plasma. In humans, “big” IGF-II and pro–IGF-II together make up 10–20% of the total IGF-II (7). However, the individual levels of each in human plasma have never been clearly determined.

Because of its homology to IGF-I (8), IGF-II binds to and activates the IFGI receptor (IGF-IR), although with a lower affinity than IGF-I (9). IGF-II also activates both isoforms (A and B) of the related IR that are the result of the alternative splicing of exon 11 that encodes the C terminus of the extracellular α subunit (10). IGF-II binding to the IR exhibits a slightly higher affinity for IR-A (exon 11), but with an overall intermediate affinity for both isoforms between that of IGF-I (less affinity) and insulin (greater affinity) (9, 11). Although the form being studied was not clearly identified, one study showed that the larger forms of IGF-II bind the IGF-IR (12). There have been no studies of the binding of “big” or pro–IGF-II to the IR, although one report described increased glucose metabolism in rat fat cells after exposure to “big” IGF-II, suggesting possible IR activation (3). Therefore, the relative level of receptor activation by “big” and pro–IGF-II is unknown.

Binding of insulin or IGF-II to the IR results in IR tyrosine autophosphorylation (13, 14). After autophosphorylation, the IR tyrosine kinase domain phosphorylates insulin receptor substrates-1 (IRS-1) and 2 (IRS-2) (15, 16). Phosphorylation of IRS-1 and IRS-2 initiates a signaling cascade that leads to serine/threonine phosphorylation of Akt/protein kinase B (17).

Previous IGF-II precursor analyses used column chromatography and RIA with a polyclonal antibody against IGF-II (18) or various immunoassays using antibodies against mature IGF-II and the E domain. Because the E domain antibodies were directed against the region corresponding to amino acids 68–104, these studies did not distinguish between “big” IGF-II and pro–IGF-II. A recent study of IGF-II levels in rat plasma by Western blot showed the specific detection and measurement of levels of all three forms of IGF-II at the same time using a single antibody against mature IGF-II (19). This same study also showed that this technique could be used on human plasma to achieve the same separation, suggesting that Western blotting could be an alternative to immunoassays that would distinguish between “big” IGF-II and pro–IGF-II. Another report from the same year also demonstrated successful separation of the forms of IGF-II in a single human sample, although the separation was not sufficient to individually distinguish “big” and pro–IGF-II (20).

One proposed theory for the cause of the hypoglycemia in NICTH patients is that “big” and/or pro–IGF-II have altered interaction with IGF binding proteins (IGFBPs), which could lead to more free IGF to activate the IR, resulting in hypoglycemia (21). Reduced levels of the 150-kDa ternary complex involving IGFBP-3 have been reported in NICTH patients (18, 22), and Valenzano et al. reported that the larger forms of IGF-II have “similar or slightly higher affinity” for IGFBP-3 when compared with mature IGF-II (12). These changes in IGFBP interaction appear to be small, however, and do not necessarily explain NICTH-associated hypoglycemia. Thus, we proposed an alternative hypothesis that “big” and/or pro–IGF-II cause greater activation of the IR than mature IGF-II, which would provide a mechanism for the hypoglycemia characteristic of the IGF-II–overexpressing tumors that result in NICTH. The purpose of this study was to address this concept by examining the distribution of IGF-II isoforms in human plasma and to evaluate the ability of “big” IGF-II and pro–IGF-II to activate the alternatively spliced isoforms of the IR, as well as to determine whether differential activation of the IR isoforms resulted in differential biological activity.

Materials and Methods

Reagents

Antibodies were obtained from the following sources: IGF-II mouse antirat monoclonal antibody (clone S1F2) from Upstate (Lake Placid, NY); IR rabbit polyclonal antibody, IRS-1 rabbit polyclonal antibody, IRS-2 rabbit polyclonal antibody, and PY20 mouse monoclonal antibody from Santa Cruz Biotechnology (Santa Cruz, CA); Akt mouse monoclonal antibody from Cell Signaling Technologies (Beverly, MA); and phospho-Akt [pS⁴⁷³] from Invitrogen (Camarillo, CA). Precast 10% and 16.5% PAGE gels were from Bio-Rad (Hercules, CA). ECL antimouse IgG horseradish peroxidase-linked whole antibody and ECL antirabbit IgG horseradish-linked whole antibody were used as secondary antibodies and were obtained from GE Healthcare UK Limited (Little Chalfont, Buckinghamshire, UK). Full-Range Rainbow molecular weight markers were from GE Healthcare (Piscataway, NJ) and were used with the 10% gels. The protein marker used with the 16.5% gels was the Bio-Rad Precision Plus protein standards. The Western Lightning Plus-ECL kit was from PerkinElmer (Waltham, MA). Recombinant human insulin was from Sigma Life Science (St. Louis, MO). Recombinant mature IGF-II, 1–104 “big” IGF-II, and pro–IGF-II were from Novozymes Biopharma AU Limited (Adelaide, SA, Australia). The WST-1 Cell Proliferation Reagent was from Roche Diagnostics (Indianapolis, IN).

Determination of IGF-II distribution in human, rat, and monkey plasma by Western immunoblotting

Forty-three deidentified human plasma samples from a previously conducted clinical study were used (23). Samples were from healthy male and female subjects with an age range of 19–64 yr and a body mass index range of 18–64 kg/m² (Table 1). These subjects were recruited from a local population through flyers and print advertisements, were free of medical conditions, not taking prescription drugs, nonsmokers, at their lifetime maximal weight, and consented to study before enrollment. In addition, single samples of adult rat, 10-d-old rat, and rhesus macaque plasma were analyzed. Rat and monkey plasma was obtained as excess material from unrelated studies previously approved by the Oregon Health and Science University Institutional Animal Care and Use Committee. Each sample was diluted 1:10 in sodium dodecyl sulfate (SDS) sample buffer before loading in a Bio-Rad Criterion 16.5% Tricine SDS-PAGE gel in the absence of reducing agent as described in Qiu et al. (19). Addition of a reducing agent prevented efficient separation of the various forms of IGF-II (data not shown). The gel was run in Bio-Rad Tris/Tricine/SDS buffer at a constant voltage of 100 V for 3.5 h, which was the time needed for sufficient separation of the three forms of IGF-II. In addition to the plasma samples, three mixtures of recombinant mature IGF-II, “big” IGF-II, and pro–IGF-II were run on each gel. The first mixture contained 5 nm mature IGF-II, 1 nm“big” IGF-II, and 1 nm pro–IGF-II, the second mixture contained 10 nm mature IGF-II, 5 nm“big” IGF-II, and 5 nm pro–IGF-II, and the third mixture contained 25 nm mature IGF-II, 10 nm“big” IGF-II, and 10 nm pro–IGF-II. After transfer to a nitrocellulose membrane, the blot was blocked with 5% BSA for 1 h before applying IGF-II mouse antirat monoclonal antibody diluted to 1:1000 in 5% BSA to the blot overnight. The following day, after washing the blot, ECL antimouse IgG horseradish peroxidase-linked whole antibody diluted to 1:5000 in 5% BSA was applied to the blot for a minimum of 1 h. Bands were visualized using a Western Lightning Plus-ECL kit and an α Inotech FluoroChem Q MultiImage III imaging system (α Inotech, Santa Clara, CA), and intensity was measured using α Inotech FluoroChem Q software. For “big” IGF-II in human plasma, glycosylated and unglycosylated bands were quantified together.

Table 1.

Characteristics of the 43 human subjects

Number	43
Age, yr	43 (19–64)
BMI, kg/m²	32.99 (18.99–64.34)
No. with BMI <30 kg/m^2b	19 (44%)
No. with BMI 30–39 kg/m^2b	17 (40%)
No. with BMI >40 kg/m^2b	7 (16%)
% Body fat^a	36.22 (4.83–53.19)
No. with NL % body fat^b,c	8 (19%)
No. with greater than NL % body fat^b,c	35 (81%)
Insulin sensitivity	2.60 (0.50–10.98)

Open in a new tab

Except where noted otherwise, data are presented as mean (range).

% body fat measured by underwater weighing.

Data are mean (% total).

Reference values for normal % body fat: women <32%, men <25% (American Council on Exercise).

Cell treatments and lysis

IGF-IR–deficient (R⁻) mouse fibroblasts expressing equivalent levels of human IR-A (R⁻ IR-A) or IR-B (R⁻ IR-B) have been previously described (24). Before treatment, cells were serum-starved overnight. For dose-response experiments, cells were treated with recombinant insulin, mature IGF-II, 1–104 “big” IGF-II, or pro–IGF-II at 0.5, 1, and 10 nm. For time-course experiments, cells were treated with 1 nm recombinant mature IGF-II, 1–104 “big” IGF-II, or pro–IGF-II for 1, 5, 15, 30, and 60 min. Cells were then lysed in 1× SDS sample buffer for Akt phosphorylation or in NP40 complete lysis buffer for IR, IRS-1, or IRS-2 immunoprecipitation. Samples in SDS buffer were immediately boiled for 5 min, followed by centrifugation in a microfuge to pellet cell debris. After collection, lysates in NP40 buffer were immediately placed on ice for 20 min with occasional agitation, followed by high-speed centrifugation for 10 min.

Immunoprecipitation and Western immunoblotting

Immunoprecipitation of IR, IRS-1, and IRS-2, Western immunoblotting, and chemiluminescence detection were performed as previously described (25). PY20 antibody was used for detection of phospho-IR, phospho–IRS-1, and phospho–IRS-2 in IR, IRS-1, or IRS-2 immunoprecipitates. Phospho-Akt [pS⁴⁷³] antibody was used for detection of phospho-Akt on blots of total cell lysates. After imaging, all blots were stripped in 100 mm Tris-HCl (pH 6.8), 10% SDS, and 100 mm β-mercaptoethanol for 40 min at 60 C followed by probing with IR, IRS-1, IRS-2, or Akt primary antibodies and reimaging.

Determination of cell viability by WST assay

R⁻ IR-A and R⁻ IR-B cells were seeded into a 96-well plate at 1000 cells per well. After overnight serum starvation, wells were treated with media or 1 or 10 nm recombinant insulin, mature IGF-II, 1–104 “big” IGF-II, or pro–IGF-II. After 72 h, 10 μl of WST-1 Cell Proliferation Reagent was added to each well and absorbance was measured 1 h later.

Statistical analysis

All values for phospho-proteins from the analysis of cell lysate immunoblots were normalized for the amount of the corresponding total protein (i.e., IR, IRS-1/2, Akt) in each sample. All values for plasma levels were normalized for the corresponding recombinant form of IGF-II run on the same gel. All bands were corrected for the background of each blot. INSTAT from Graphpad Software, Inc. (San Diego, CA) was used for statistical analysis. Unless otherwise stated, all analyses were one-way ANOVA with a Tukey-Kramer Multiple Comparisons Test post test. Microsoft Office Excel and Prism were used for graphing.

Results

IGF-II distribution in human plasma

“Big” and pro–IGF-II were present in all 43 human plasma samples (Fig. 1A). The mean percentage of total IGF-II was 71.04% for mature IGF-II (range 30.18-95.41%), 15.78% for “big” IGF-II (range 1.75–39.12%), and 13.18% for pro–IGF-II (range 1.33–40.47%). Levels of mature IGF-II exhibited more individual variability than levels of either “big” or pro–IGF-II (Fig. 1B). For comparison, plasma from an adult rat, a 10-d-old rat, and an adult rhesus macaque were also analyzed (Fig. 1C). Adult rat plasma contained only pro–IGF-II, confirming the findings of Qiu et al. (19), while 10-day-old rat plasma had equivalent levels of all three forms of IGF-II. Rhesus macaque plasma was similar to human plasma in that it contained all three forms of IGF-II, with the levels of mature IGF-II being the highest. Analysis of the human IGF-II levels did not show any correlation between mature IGF-II and “big” or pro–IGF-II. In addition, analysis of the human plasma samples showed no significant associations between any of the forms of IGF-II and gender, age, weight, body mass index, percent body fat, or insulin sensitivity (data not shown).

Dose dependence of IR pathway activation

To determine the levels of signaling through the IR pathway caused by each of the forms of IGF-II, we exposed cells to varying concentrations of recombinant insulin, mature, “big,” and pro–IGF-II and analyzed tyrosine phosphorylation of IR, IRS-1, IRS-2, and Akt. Insulin stimulated both IR-A and IR-B autophosphorylation in a dose-dependent fashion (Fig. 2). None of the forms of IGF-II resulted in significant IR-A autophosphorylation (Fig. 2A), although low levels of IR-B autophosphorylation were seen with mature IGF-II at 1 nm and with all three forms of IGF-II at 10 nm (Fig. 2B). Thus, in both IR-A and IR-B cells, autophosphorylation mostly occurred in response to insulin, with only a small response to any form of IGF-II in IR-B cells. The lack of significant IR-A autophosphorylation induced by IGF-II compared with IR-B is consistent with the relative activation seen in our previous studies with these cell lines (26) and may reflect the fact that the IR is not overexpressed to the extent seen in cell lines used in earlier studies that also described the increased affinity of IR-A for IGF-II (11, 14).

Fig. 2. — Dose–response of IR activation. After overnight serum starvation, cells were treated with 0.5, 1, and 10 nm recombinant insulin, mature, “big,” and pro–IGF-II for 5 min. After immunoprecipitation, lysates were evaluated by Western immunoblotting with PY20 antibody for phosphorylated IR levels followed by stripping and reprobing with IR antibody. Levels of phosphorylated IR were normalized for the levels of total IR in each sample. A, Phospho-IR levels in IR-A cells and representative immunoblot. B, Phospho-IR levels in IR-B cells and representative immunoblot.

Insulin treatment stimulated dose-dependent IRS-1 phosphorylation in both IR-A and IR-B cell lines (Fig. 3). In IR-A cells, 10 nm mature and “big” IGF-II stimulated a small amount of IRS-1 phosphorylation (Fig. 3A). In IR-B cells, mature IGF-II caused a low level of IRS-1 activation at 0.5 and 1 nm, while mature and “big” IGF-II caused a slightly higher level of activation at 10 nm (Fig. 3B). Thus, insulin treatment resulted in the highest levels of IRS-1 activation, while, within each cell line, mature and “big” IGF-II resulted in similar, lower levels.

Stimulation of IR-A with insulin resulted in high levels of IRS-2 phosphorylation at 0.5 and 1 nm, and a lower level of phosphorylation at 10 nm (Fig. 4). Mature and “big” IGF-II exposure resulted in low levels of IR-A activation at 0.5 and 1 nm and a higher level of activation at 10 nm, while pro–IGF-II resulted in a low level of activation at all concentrations (Fig. 4A). In IR-B cells, insulin treatment resulted in more activation of IRS-2 at 0.5 nm and less activation at 1 and 10 nm. Treatment with mature and “big” IGF-II resulted in lower levels of IR-B activation at 0.5 and 1 nm and a higher level of activation at 10 nm, while pro–IGF-II stimulated low levels of activation at all concentrations (Fig. 4B). Within each cell line, activation by insulin, mature, and “big” IGF-II at 10 nm was equivalent. Also, in both cell lines, IRS-2 was more activated by insulin at low concentrations, while mature and “big” IGF-II activation of IRS-2 was highest at 10 nm.

Akt phosphorylation occurred in dose-dependent manner in both cell lines after treatment with insulin, mature IGF-II, and “big” IGF-II, with the highest level of activation for all three occurring at 10 nm. Insulin stimulated higher levels of Akt phosphorylation in IR-A and IR-B than mature and “big” IGF-II. There was no significant phosphorylation caused by pro–IGF-II at any concentration. As a percentage of Akt activation by insulin, activation by mature and “big” IGF-II was greater in IR-A than in IR-B (Fig. 5).

Fig. 5. — Dose response of Akt activation. After overnight serum starvation, cells were treated with 0.5, 1, and 10 nm recombinant insulin, mature, “big,” and pro–IGF-II for 15 min. Lysates were evaluated by Western immunoblotting with phospho-Akt antibody, followed by stripping and reprobing with Akt antibody. Levels of phophorylated Akt were normalized for the levels of total Akt in each sample. A, Phospho-Akt levels in IR-A cells and representative immunoblot. B, Phospho-Akt levels in IR-B cells and representative immunoblot.

In IR-A cells, when compared at each level of the signaling cascade, there was no significant difference between activation by mature and “big” IGF-II. Likewise, in IR-B cells, there was no significant difference between activation by mature and “big” IGF-II at any level of the cascade.

Time course of Akt activation

To determine whether IGF-II isoforms differed in their time course of signaling, time courses of Akt activation were performed. In both IR-A and IR-B cells, maximal Akt activation by both mature and “big” IGF-II occurred at 15 min (Fig. 6). This was followed by continued but declining activation at 30 and 60 min in both cell lines. Throughout the time course, within each cell line, responses to both mature and “big” IGF-II were similar, with no significant difference between the two at any time point. In contrast, peak activation by pro–IGF-II was approximately half that of mature and “big” IGF-II and was significant at 60 min in both IR-A and IR-B cell lines.

Fig. 6. — Time course of Akt activation. After overnight serum starvation, the cells were treated with 1 nm recombinant mature, “big,” and pro–IGF-II for 1, 5, 15, 30, and 60 min. After treatment, lysates were evaluated by Western immunoblotting with phospho-Akt antibody for phosphorylated Akt levels, followed by stripping and reprobing with Akt antibody. Levels of phophorylated Akt were normalized for the levels of total Akt in each sample. A, Phospho-Akt levels in IR-A cells and representative immunoblot. B, Phospho-Akt levels in IR-B cells and representative immunoblot.

Cell viability as determined by WST assay

Figure 7 shows the results of WST assays in IR-A and IR-B cells. When compared with cells that were exposed to serum-free media alone for 72 h, cells that were exposed to insulin, mature IGF-II, and “big” IGF-II were more viable. The greatest effect was seen in IR-A cells treated with 10 nm insulin (1.8-fold vs. vehicle). With the exception of 1 nm“big” IGF-II in IR-A cells, responses to mature and “big” IGF-II were similar at every concentration and were all around 1.4-fold vs. vehicle. There was no improvement in viability for cells treated with any concentration of pro–IGF-II.

Fig. 7. — IGF-II isoform effects on cell viability. After overnight serum starvation, IR-A and IR-B cells were seeded into 96-well plates and then treated with 1 nm and 10 nm recombinant insulin, mature IGF-II, “big” IGF-II, and pro–IGF-II. Seventy-two hours after treatment, 10 μl of WST-1 Cell Proliferation Reagent was added to each well and absorbance was measured 1 h later.

Discussion

This study shows that all forms of IGF-II are present in human plasma and that “big” IGF-II makes up about 16% of the total IGF-II, while pro–IGF-II makes up about 13%. As a result, the combination of the two is 29% of total IGF-II, which is higher than the 10–20% reported previously (7). The finding of both pro–IGF-II and “big” IGF-II in every human sample is of interest, because the previous literature uses the terms pro–IGF-II and “big” IGF-II interchangeably, which has made it difficult to determine which of the two was present in previously analyzed human samples. Also, in the single human sample run for IGF-II on a Western blot by Qiu et al., “big” IGF-II was present but pro–IGF-II was not seen (19). In addition, the single human sample run by Zachariah et al. showed bands that did not appear to be separated enough to determine whether both pro–IGF-II and “big” IGF-II were present (20). Thus, it had been speculated that pro–IGF-II was not present in human plasma. Our study shows that pro–IGF-II is indeed routinely present in human plasma.

Our study confirms the results of Qiu et al. (19), which demonstrated that mature IGF-II and both precursor forms are found in neonatal rats, while pro–IGF-II is present in adult rats, but that mature IGF-II is absent in adult rat serum, which has also previously been demonstrated by a number of studies. Our data also demonstrate that all forms of IGF-II are present in rhesus macaque serum in a pattern that is similar to that in humans.

During the initial review of this manuscript, we became aware of the subsequent study by Qiu et al. (27) on mature and “big” IGF-II binding to IGFBP-2 and -3, which used the Western blot analysis approach described in this group's previous paper (19). Their analysis of a series of human serum samples showed levels of mature IGF-II and “big” IGF-II similar to the levels reported in this study, but they did not detect pro–IGF-II (26). This difference is possibly attributable to differences in the study population (Chinese vs. U.S. populations) or the fact that our study analyzed plasma rather than serum as done in Qiu et al., but we do not have a specific explanation for the potential differential detection of pro–IGF-II in human plasma vs. serum.

Overall analysis of our Western immunoblotting data shows that activation of Akt by mature and “big” IGF-II was maximal by 15 min but that maximal activation by pro–IGF-II did not occur until 60 min. Mature and “big” IGF-II treatment also resulted in greater Akt phosphorylation in the IR-A cell line than in the IR-B cell line. This is consistent with the higher affinity of IGF-II for IR-A (11). On the other hand, mature and “big” IGF-II–stimulated IR-B autophosphorylation and IRS-2 activation was greater than that seen in IR-A cells, which is somewhat counter-intuitive. While we do not have a molecular explanation for the lack of a consistent relationship between activation of all of the IR signaling pathway components we have assayed between IR-A and IR-B-expressing cells, the data suggest that each step of the cascade may be subject to independent modulation that precludes the prediction of the activation at a given stage from the level of activation of the previous level. With respect to signaling through IR-A or IR-B specifically, mature and “big” IGF-II treatment resulted in similar levels of activation, while pro–IGF-II resulted in less activation. For Akt, pro–IGF-II resulted in significantly less, or at least delayed, activation. In addition, the results of the WST assay, which was done to evaluate the biological activity resulting from activation of the IR, showed biological activity that correlated with the signaling results. Therefore, both the signaling and the biological activity studies suggest that activation of the IR by IGF-II is predominantly a property of mature and “big” IGF-II, not of pro–IGF-II, and that overall IR-mediated signaling by mature and “big” IGF-II are generally equivalent.

Our analysis also showed increasing levels of activation by insulin, mature IGF-II, and “big” IGF-II with progression through the signaling cascade. Levels of phosphorylated Akt were the highest, followed by levels of phosphorylated IRS-2, and then levels of phosphorylated IRS-I. Levels of phosphorylated IR were the lowest, with IR-A activation by IGF-II isoforms being undetectable, and activation of IR-B being minimal. This observation is consistent with progressive amplification of signaling at the more distal steps in the insulin signaling pathway and with the minimal IR activation in conjunction with more robust IRS and Akt activation by mature IGF-II and IGF-I, which we have reported previously (26).

Because the hypoglycemia seen in NICTH has been proposed to be the result of elevated levels of “big” and/or pro–IGF-II, and the current theory of altered interaction with IGFPBs did not seem to fully explain the hypoglycemia, we hypothesized that at least one of the larger forms of IGF-II would cause greater activation of the IR than mature IGF-II. This study shows that neither of the larger forms of IGF-II causes greater activation. Because signaling and biologic activity caused by “big” IGF-II was equivalent to mature IGF-II, and signaling caused by pro–IGF-II was less than that of mature and “big” IGF-II, differences in IR signaling do not necessarily explain the hypoglycemia in NICTH. The hypoglycemia could be attributable to a combination of binding protein interaction and other mechanisms, or it could simply be attributable to glucose consumption by the tumor, because the tumors tend to be large. Suppression of growth hormone production has also been proposed as a potential mechanism (28, 29). Despite this, the difference between IR signaling caused by “big” IGF-II and pro–IGF-II makes distinguishing between the two important in any situation in which either is suspected of being part of the pathophysiology of a disease.

Our study does have several limitations. First, as has been shown previously (4, 5), and consistent with the data obtained in this study, “big” and pro–IGF-II are glycosylated in vivo. If glycosylation alters the activation of the IR by “big” or pro–IGF-II, the results of this study, which used unglycosylated recombinant forms of IGF-II, could differ from the actual activation in vivo. In addition, if glycosylation interferes with the binding of the IGF-II antibody used in the Western immunoblotting, the levels of “big” and pro–IGF-II could be underestimated. However, this is unlikely, because the plasma proteins were denatured before Western immunoblotting. Second, while assessing cell viability is a test of biological activity because the IR can exert effects on cell growth and proliferation, it is not the most appropriate test of biological activity downstream of the IR. A more appropriate test of biological activity would be glut4-mediated glucose uptake, because this is a primary effect of IR activation. However, existing cell lines in which glut4-mediated glucose uptake could be done also contain IGF-IR, which would confound the results of such testing. Finally, this study did not include any evaluation of IGF-IR activation by the larger forms of IGF-II. While signaling studies of the IR can be done in R⁻ IR-A and R⁻ IR-B cells because there is not interference from IGF-IR, there are no readily available cell lines expressing IGF-IR that are IR-deficient. Thus any study of IGF-IR signaling would be confounded by IR signaling.

In summary, we routinely detected mature, “big,” and pro–IGF-II in human plasma. Mature and “big” IGF-II resulted in equivalent levels of signaling through both isoforms of the IR that, at the level of Akt, was more apparent with IR-A, whereas pro–IGF-II-stimulated signaling was significantly less, although delayed Akt activation was discernable. These data suggest that preferential activation of IR by IGF-II precursors does not play a significant role in the hypoglycemia of NICTH.

Acknowledgments

This work was supported in part by Grant R000153 from the National Center for Research Resources and Grant T32HD007497 from the National Institute of Child Health and Human Development (to A.G.M.), National Institutes of Health.

Disclosure Summary: The authors have nothing to declare.

Footnotes

Abbreviations:

IGF-IR: IGFI receptor
IGFBP: IGF binding protein
IRS: insulin receptor substrate
NICTH: nonislet cell tumor hypoglycemia
SDS: sodium dodecyl sulfate.

References

1. Daughaday WH, Emanuele MA, Brooks MH, Barbato AL, Kapadia M, Rotwein P. 1988. Synthesis and secretion of insulin-like growth factor II by a leiomyosarcoma with associated hypoglycemia. N Engl J Med 319:1434–1440 [DOI] [PubMed] [Google Scholar]
2. de Groot JW, Rikhof B, van Doorn J, Bilo HJ, Alleman MA, Honkoop AH, van der Graaf WT. 2007. Non-islet cell tumour-induced hypoglycaemia: a review of the literature including two new cases. Endocr Relat Cancer 14:979–993 [DOI] [PubMed] [Google Scholar]
3. Zapf J, Futo E, Peter M, Froesch ER. 1992. Can “big” insulin-like growth factor II in serum of tumor patients account for the development of extrapancreatic tumor hypoglycemia? J Clin Invest 90:2574–2584 [DOI] [PMC free article] [PubMed] [Google Scholar]
4. Hudgins WR, Hampton B, Burgess WH, Perdue JF. 1992. The identification of O-glycosylated precursors of insulin-like growth factor II. J Biol Chem 267:8153–8160 [PubMed] [Google Scholar]
5. Duguay SJ, Jin Y, Stein J, Duguay AN, Gardner P, Steiner DF. 1998. Post-translational processing of the insulin-like growth factor-2 precursor. Analysis of O-glycosylation and endoproteolysis. J Biol Chem 273:18443–18451 [DOI] [PubMed] [Google Scholar]
6. van Doorn J, Hoogerbrugge CM, Koster JG, Bloemen RJ, Hoekman K, Mudde AH, van Buul-Offers SC. 2002. Antibodies directed against the E region of pro-insulin-like growth factor-II used to evaluate non-islet cell tumor-induced hypoglycemia. Clin Chem 48:1739–1750 [PubMed] [Google Scholar]
7. Daughaday WH, Trivedi B. 1992. Measurement of derivatives of proinsulin-like growth factor-II in serum by a radioimmunoassay directed against the E-domain in normal subjects and patients with nonislet cell tumor hypoglycemia. J Clin Endocrinol Metab 75:110–115 [DOI] [PubMed] [Google Scholar]
8. Rinderknecht E, Humbel RE. 1978. Primary structure of human insulin-like growth factor II. FEBS Lett 89:283–286 [DOI] [PubMed] [Google Scholar]
9. Denley A, Bonython ER, Booker GW, Cosgrove LJ, Forbes BE, Ward CW, Wallace JC. 2004. Structural determinants for high-affinity binding of insulin-like growth factor II to insulin receptor (IR)-A, the exon 11 minus isoform of the IR. Mol Endocrinol 18:2502–2512 [DOI] [PubMed] [Google Scholar]
10. De Meyts P, Whittaker J. 2002. Structural biology of insulin and IGF1 receptors: implications for drug design. Nat Rev Drug Discov 1:769–783 [DOI] [PubMed] [Google Scholar]
11. Frasca F, Pandini G, Scalia P, Sciacca L, Mineo R, Costantino A, Goldfine ID, Belfiore A, Vigneri R. 1999. Insulin receptor isoform A, a newly recognized, high-affinity insulin-like growth factor II receptor in fetal and cancer cells. Mol Cell Biol 19:3278–3288 [DOI] [PMC free article] [PubMed] [Google Scholar]
12. Valenzano KJ, Heath-Monnig E, Tollefsen SE, Lake M, Lobel P. 1997. Biophysical and biological properties of naturally occurring high molecular weight insulin-like growth factor II variants. J Biol Chem 272:4804–4813 [DOI] [PubMed] [Google Scholar]
13. White MF, Shoelson SE, Keutmann H, Kahn CR. 1988. A cascade of tyrosine autophosphorylation in the beta-subunit activates the phosphotransferase of the insulin receptor. J Biol Chem 263:2969–2980 [PubMed] [Google Scholar]
14. Belfiore A, Frasca F, Pandini G, Sciacca L, Vigneri R. 2009. Insulin receptor isoforms and insulin receptor/insulin-like growth factor receptor hybrids in physiology and disease. Endocr Rev 30:586–623 [DOI] [PubMed] [Google Scholar]
15. Araki E, Lipes MA, Patti ME, Bruning JC, Haag B, 3rd, Johnson RS, Kahn CR. 1994. Alternative pathway of insulin signalling in mice with targeted disruption of the IRS-1 gene. Nature 372:186–190 [DOI] [PubMed] [Google Scholar]
16. Sun XJ, Wang LM, Zhang Y, Yenush L, Myers MG, Jr, Glasheen E, Lane WS, Pierce JH, White MF. 1995. Role of IRS-2 in insulin and cytokine signalling. Nature 377:173–177 [DOI] [PubMed] [Google Scholar]
17. Alessi DR, Andjelkovic M, Caudwell B, Cron P, Morrice N, Cohen P, Hemmings BA. 1996. Mechanism of activation of protein kinase B by insulin and IGF-1. EMBO J 15:6541–6551 [PMC free article] [PubMed] [Google Scholar]
18. Daughaday WH, Kapadia M. 1989. Significance of abnormal serum binding of insulin-like growth factor II in the development of hypoglycemia in patients with non-islet-cell tumors. Proc Natl Acad Sci USA 86:6778–6782 [DOI] [PMC free article] [PubMed] [Google Scholar]
19. Qiu Q, Jiang JY, Bell M, Tsang BK, Gruslin A. 2007. Activation of endoproteolytic processing of insulin-like growth factor-II in fetal, early postnatal, and pregnant rats and persistence of circulating levels in postnatal life. Endocrinology 148:4803–4811 [DOI] [PubMed] [Google Scholar]
20. Zachariah S, Brackenridge A, Shojaee-Moradie F, Camuncho-Hubner C, Umpleby AM, Russell-Jones D. 2007. The mechanism of non-islet cell hypoglycaemia caused by tumour-produced IGF-II. Clin Endocrinol (Oxf) 67:637–638 [DOI] [PubMed] [Google Scholar]
21. Baxter RC, Holman SR, Corbould A, Stranks S, Ho PJ, Braund W. 1995. Regulation of the insulin-like growth factors and their binding proteins by glucocorticoid and growth hormone in nonislet cell tumor hypoglycemia. J Clin Endocrinol Metab 80:2700–2708 [DOI] [PubMed] [Google Scholar]
22. Bond JJ, Meka S, Baxter RC. 2000. Binding characteristics of pro-insulin-like growth factor-II from cancer patients: binary and ternary complex formation with IGF binding proteins-1 to -6. J Endocrinol 165:253–260 [DOI] [PubMed] [Google Scholar]
23. Purnell JQ, Brandon DD, Isabelle LM, Loriaux DL, Samuels MH. 2004. Association of 24-hour cortisol production rates, cortisol-binding globulin, and plasma-free cortisol levels with body composition, leptin levels, and aging in adult men and women. J Clin Endocrinol Metab 89:281–287 [DOI] [PubMed] [Google Scholar]
24. Pandini G, Frasca F, Mineo R, Sciacca L, Vigneri R, Belfiore A. 2002. Insulin/insulin-like growth factor I hybrid receptors have different biological characteristics depending on the insulin receptor isoform involved. J Biol Chem 277:39684–39695 [DOI] [PubMed] [Google Scholar]
25. Denley A, Brierley GV, Carroll JM, Lindenberg A, Booker GW, Cosgrove LJ, Wallace JC, Forbes BE, Roberts CT., Jr 2006. Differential activation of insulin receptor isoforms by insulin-like growth factors is determined by the C domain. Endocrinology 147:1029–1036 [DOI] [PubMed] [Google Scholar]
26. Denley A, Carroll JM, Brierley GV, Cosgrove L, Wallace J, Forbes B, Roberts CT., Jr 2007. Differential activation of insulin receptor substrates 1 and 2 by insulin-like growth factor-activated insulin receptors. Mol Cell Biol 27:3569–3577 [DOI] [PMC free article] [PubMed] [Google Scholar]
27. Qiu Q, Yan X, Bell M, Di J, Tsang BK, Gruslin A. 2010. Mature IGF-II prevents the formation of “big” IGF-II/IGFBP-2 complex in the human circulation. Growth Horm IGF Res 20:110–117 [DOI] [PubMed] [Google Scholar]
28. Agus MS, Katz LE, Satin-Smith M, Meadows AT, Hintz RL, Cohen P. 1995. Non-islet-cell tumor associated with hypoglycemia in a child: successful long-term therapy with growth hormone. J Pediatr 127:403–407 [DOI] [PubMed] [Google Scholar]
29. Katz LE, Liu F, Baker B, Agus MS, Nunn SE, Hintz RL, Cohen P. 1996. The effect of growth hormone treatment on the insulin-like growth factor axis in a child with nonislet cell tumor hypoglycemia. J Clin Endocrinol Metab 81:1141–1146 [DOI] [PubMed] [Google Scholar]

[B1] 1. Daughaday WH, Emanuele MA, Brooks MH, Barbato AL, Kapadia M, Rotwein P. 1988. Synthesis and secretion of insulin-like growth factor II by a leiomyosarcoma with associated hypoglycemia. N Engl J Med 319:1434–1440 [DOI] [PubMed] [Google Scholar]

[B2] 2. de Groot JW, Rikhof B, van Doorn J, Bilo HJ, Alleman MA, Honkoop AH, van der Graaf WT. 2007. Non-islet cell tumour-induced hypoglycaemia: a review of the literature including two new cases. Endocr Relat Cancer 14:979–993 [DOI] [PubMed] [Google Scholar]

[B3] 3. Zapf J, Futo E, Peter M, Froesch ER. 1992. Can “big” insulin-like growth factor II in serum of tumor patients account for the development of extrapancreatic tumor hypoglycemia? J Clin Invest 90:2574–2584 [DOI] [PMC free article] [PubMed] [Google Scholar]

[B4] 4. Hudgins WR, Hampton B, Burgess WH, Perdue JF. 1992. The identification of O-glycosylated precursors of insulin-like growth factor II. J Biol Chem 267:8153–8160 [PubMed] [Google Scholar]

[B5] 5. Duguay SJ, Jin Y, Stein J, Duguay AN, Gardner P, Steiner DF. 1998. Post-translational processing of the insulin-like growth factor-2 precursor. Analysis of O-glycosylation and endoproteolysis. J Biol Chem 273:18443–18451 [DOI] [PubMed] [Google Scholar]

[B6] 6. van Doorn J, Hoogerbrugge CM, Koster JG, Bloemen RJ, Hoekman K, Mudde AH, van Buul-Offers SC. 2002. Antibodies directed against the E region of pro-insulin-like growth factor-II used to evaluate non-islet cell tumor-induced hypoglycemia. Clin Chem 48:1739–1750 [PubMed] [Google Scholar]

[B7] 7. Daughaday WH, Trivedi B. 1992. Measurement of derivatives of proinsulin-like growth factor-II in serum by a radioimmunoassay directed against the E-domain in normal subjects and patients with nonislet cell tumor hypoglycemia. J Clin Endocrinol Metab 75:110–115 [DOI] [PubMed] [Google Scholar]

[B8] 8. Rinderknecht E, Humbel RE. 1978. Primary structure of human insulin-like growth factor II. FEBS Lett 89:283–286 [DOI] [PubMed] [Google Scholar]

[B9] 9. Denley A, Bonython ER, Booker GW, Cosgrove LJ, Forbes BE, Ward CW, Wallace JC. 2004. Structural determinants for high-affinity binding of insulin-like growth factor II to insulin receptor (IR)-A, the exon 11 minus isoform of the IR. Mol Endocrinol 18:2502–2512 [DOI] [PubMed] [Google Scholar]

[B10] 10. De Meyts P, Whittaker J. 2002. Structural biology of insulin and IGF1 receptors: implications for drug design. Nat Rev Drug Discov 1:769–783 [DOI] [PubMed] [Google Scholar]

[B11] 11. Frasca F, Pandini G, Scalia P, Sciacca L, Mineo R, Costantino A, Goldfine ID, Belfiore A, Vigneri R. 1999. Insulin receptor isoform A, a newly recognized, high-affinity insulin-like growth factor II receptor in fetal and cancer cells. Mol Cell Biol 19:3278–3288 [DOI] [PMC free article] [PubMed] [Google Scholar]

[B12] 12. Valenzano KJ, Heath-Monnig E, Tollefsen SE, Lake M, Lobel P. 1997. Biophysical and biological properties of naturally occurring high molecular weight insulin-like growth factor II variants. J Biol Chem 272:4804–4813 [DOI] [PubMed] [Google Scholar]

[B13] 13. White MF, Shoelson SE, Keutmann H, Kahn CR. 1988. A cascade of tyrosine autophosphorylation in the beta-subunit activates the phosphotransferase of the insulin receptor. J Biol Chem 263:2969–2980 [PubMed] [Google Scholar]

[B14] 14. Belfiore A, Frasca F, Pandini G, Sciacca L, Vigneri R. 2009. Insulin receptor isoforms and insulin receptor/insulin-like growth factor receptor hybrids in physiology and disease. Endocr Rev 30:586–623 [DOI] [PubMed] [Google Scholar]

[B15] 15. Araki E, Lipes MA, Patti ME, Bruning JC, Haag B, 3rd, Johnson RS, Kahn CR. 1994. Alternative pathway of insulin signalling in mice with targeted disruption of the IRS-1 gene. Nature 372:186–190 [DOI] [PubMed] [Google Scholar]

[B16] 16. Sun XJ, Wang LM, Zhang Y, Yenush L, Myers MG, Jr, Glasheen E, Lane WS, Pierce JH, White MF. 1995. Role of IRS-2 in insulin and cytokine signalling. Nature 377:173–177 [DOI] [PubMed] [Google Scholar]

[B17] 17. Alessi DR, Andjelkovic M, Caudwell B, Cron P, Morrice N, Cohen P, Hemmings BA. 1996. Mechanism of activation of protein kinase B by insulin and IGF-1. EMBO J 15:6541–6551 [PMC free article] [PubMed] [Google Scholar]

[B18] 18. Daughaday WH, Kapadia M. 1989. Significance of abnormal serum binding of insulin-like growth factor II in the development of hypoglycemia in patients with non-islet-cell tumors. Proc Natl Acad Sci USA 86:6778–6782 [DOI] [PMC free article] [PubMed] [Google Scholar]

[B19] 19. Qiu Q, Jiang JY, Bell M, Tsang BK, Gruslin A. 2007. Activation of endoproteolytic processing of insulin-like growth factor-II in fetal, early postnatal, and pregnant rats and persistence of circulating levels in postnatal life. Endocrinology 148:4803–4811 [DOI] [PubMed] [Google Scholar]

[B20] 20. Zachariah S, Brackenridge A, Shojaee-Moradie F, Camuncho-Hubner C, Umpleby AM, Russell-Jones D. 2007. The mechanism of non-islet cell hypoglycaemia caused by tumour-produced IGF-II. Clin Endocrinol (Oxf) 67:637–638 [DOI] [PubMed] [Google Scholar]

[B21] 21. Baxter RC, Holman SR, Corbould A, Stranks S, Ho PJ, Braund W. 1995. Regulation of the insulin-like growth factors and their binding proteins by glucocorticoid and growth hormone in nonislet cell tumor hypoglycemia. J Clin Endocrinol Metab 80:2700–2708 [DOI] [PubMed] [Google Scholar]

[B22] 22. Bond JJ, Meka S, Baxter RC. 2000. Binding characteristics of pro-insulin-like growth factor-II from cancer patients: binary and ternary complex formation with IGF binding proteins-1 to -6. J Endocrinol 165:253–260 [DOI] [PubMed] [Google Scholar]

[B23] 23. Purnell JQ, Brandon DD, Isabelle LM, Loriaux DL, Samuels MH. 2004. Association of 24-hour cortisol production rates, cortisol-binding globulin, and plasma-free cortisol levels with body composition, leptin levels, and aging in adult men and women. J Clin Endocrinol Metab 89:281–287 [DOI] [PubMed] [Google Scholar]

[B24] 24. Pandini G, Frasca F, Mineo R, Sciacca L, Vigneri R, Belfiore A. 2002. Insulin/insulin-like growth factor I hybrid receptors have different biological characteristics depending on the insulin receptor isoform involved. J Biol Chem 277:39684–39695 [DOI] [PubMed] [Google Scholar]

[B25] 25. Denley A, Brierley GV, Carroll JM, Lindenberg A, Booker GW, Cosgrove LJ, Wallace JC, Forbes BE, Roberts CT., Jr 2006. Differential activation of insulin receptor isoforms by insulin-like growth factors is determined by the C domain. Endocrinology 147:1029–1036 [DOI] [PubMed] [Google Scholar]

[B26] 26. Denley A, Carroll JM, Brierley GV, Cosgrove L, Wallace J, Forbes B, Roberts CT., Jr 2007. Differential activation of insulin receptor substrates 1 and 2 by insulin-like growth factor-activated insulin receptors. Mol Cell Biol 27:3569–3577 [DOI] [PMC free article] [PubMed] [Google Scholar]

[B27] 27. Qiu Q, Yan X, Bell M, Di J, Tsang BK, Gruslin A. 2010. Mature IGF-II prevents the formation of “big” IGF-II/IGFBP-2 complex in the human circulation. Growth Horm IGF Res 20:110–117 [DOI] [PubMed] [Google Scholar]

[B28] 28. Agus MS, Katz LE, Satin-Smith M, Meadows AT, Hintz RL, Cohen P. 1995. Non-islet-cell tumor associated with hypoglycemia in a child: successful long-term therapy with growth hormone. J Pediatr 127:403–407 [DOI] [PubMed] [Google Scholar]

[B29] 29. Katz LE, Liu F, Baker B, Agus MS, Nunn SE, Hintz RL, Cohen P. 1996. The effect of growth hormone treatment on the insulin-like growth factor axis in a child with nonislet cell tumor hypoglycemia. J Clin Endocrinol Metab 81:1141–1146 [DOI] [PubMed] [Google Scholar]

PERMALINK

Plasma Distribution and Signaling Activities of IGF-II Precursors

Alicia G Marks

Julie M Carroll

Jonathan Q Purnell

Charles T Roberts Jr

Abstract