Expression and purification of biologically active IGF-binding proteins using the LCR/Mel expression system.

The anabolic effects and bioavailability of insulin-like growth factors I and II (IGF-I, IGF-II) are regulated in part by a family of IGF-binding proteins (IGFBPs). There are six known members of the IGFBP family, which share distinct structural characteristics and functional activities. To study the binding properties of these proteins, we have expressed recombinant IGFBP-3 and IGFBP-4 using the LCR/Mel expression system. Using this system, we found that recombinant IGFBP-3 was secreted by Mel cells and had a glycosylation pattern similar to that of native IGFBP-3. Recombinant IGFBP-4 secreted from Mel cells had a molecular size identical to that of non-glycosylated native IGFBP-4. The binding kinetics of recombinant IGFBPs was measured using a solid-phase ligand-binding assay, an in vitro solution-binding assay, and a cellular proliferation assay. IGF-I bound with high affinity to recombinant IGFBP-3 and IGFBP-4 with K(D)s of <0.25 nmol. As reported for native IGFBPs, IGF-II bound with affinity higher than IGF-I to recombinant IGFBP-3 and IGFBP-4 (K(D) of <0.05 nmol). Recombinant IGFBP-3 and IGFBP-4 were found to inhibit the IGF-induced proliferation of an NIH3T3 cell line engineered to overexpress the IGF-I receptor. We have compared the binding kinetics of Mel cell-expressed IGFBPs with that of recombinant protein expressed in Escherichia coli and found them to be equivalent. Here, we show that the LCR/Mel expression system represents an effective route for expression of biologically active IGFBPs.

Intact IGF-binding protein-4 and -5 and their respective fragments isolated from chronic renal failure serum differentially modulate IGF-I actions in cultured growth plate chondrocytes.

Impairment of longitudinal growth among children with chronic renal failure (CRF) may be partly attributable to the inhibition of insulin-like growth factor (IGF) activity by an excess amount of high-affinity IGF-binding proteins (IGFBP). Elevated levels of immunoreactive IGFBP-4 in CRF serum are inversely correlated with the standardized heights of these children, whereas levels of IGFBP-5, which circulates mainly as proteolyzed fragments, are positively correlated with growth parameters. To delineate the respective effects of these IGFBP on growth cartilage, the biologic effects of intact and fragmented forms of IGFBP-4 and IGFBP-5 on rat growth plate chondrocytes in primary cultures were characterized. Intact IGFBP-4 and IGFBP-5 and the amino-terminal fragment IGFBP-5(1-169) were recombinant proteins; the carboxy-terminal fragments IGFBP-5(144-252) and IGFBP-4(136-237) and the amino-terminal fragment IGFBP-4(1-122) were purified to homogeneity from CRF hemofiltrates. Intact IGFBP-4 and, to a lesser extent, IGFBP-4(1-122) inhibited IGF-I-induced cell proliferation. In contrast, intact IGFBP-5 was stimulatory in the absence or presence of exogenous IGF-I, whereas the amino-terminal fragment IGFBP-5(1-169) was inhibitory. Studies on the mechanism by which IGFBP-4 and IGFBP-5 exert opposite effects on chondrocyte proliferation demonstrated that intact IGFBP-4 prevented the binding of (125)I-IGF-I to chondrocytes, whereas intact IGFBP-5 enhanced ligand binding and was able to bind specifically to the cell membrane. These data suggest that intact IGFBP-4 and, to a lesser extent, IGFBP-4(1-122) act exclusively as growth-inhibitory binding proteins in the growth cartilage. IGFBP-5, however, can either stimulate (if it remains intact) or inhibit (if amino-terminal forms predominate) IGF-I-stimulated chondrocyte proliferation.

Expression, purification and characterization of the structure and disulfide linkages of insulin-like growth factor binding protein-4.

Insulin-like growth factor binding protein-4 (IGFBP-4), like the other five IGFBPs, is a critical regulator of the activity of insulin-like growth factor (IGF)-I and IGF-II. However IGFBP-4 seems to be the only IGFBP with no potential to enhance the mitogenic actions of the IGFs. IGFBP-1 to -3 and -5 each contain 18 conserved cysteine residues, IGFBP-6 lacks two of the twelve N-terminal cysteines, while IGFBP-4 has two additional cysteines in the central region. A plasmid was constructed to express rat IGFBP-4 as a thioredoxin fusion protein that included a hexahistidine sequence to permit affinity purification. The fusion protein was expressed in E.coli, purified using nickel-chelate affinity chromatography and cleaved by tobacco etch virus (TEV) protease to produce mature rat IGFBP-4 with an additional glycine residue at the N-terminus. Final purification was achieved by further nickel affinity chromatography and reverse phase HPLC. The isoelectric points of the recombinant IGFBP-4 were the same as those of the non-glycosylated isoforms of IGFBP-4 in rat serum. The binding affinities of the recombinant protein and IGFBP-4 secreted by rat cells to IGF-I were compared using a newly developed binding assay. No significant difference could be detected, consistent with proper folding of the recombinant protein. This indicates that glycosylation of IGFBP-4 does not affect its binding to IGF-I. Using mass spectrometry and tandem mass spectrometry no differences between authentic and recombinant IGFBP-4 could be detected. Eight of the ten disulfide linkages have been determined, including linkages of conserved cysteine residues not previously identified in other IGFBPs. Numbering the cysteine residues sequentially from the N-terminus only the disulfide connectivity of C1, C2, C5 and C6 could not be determined. However, C1 is not linked to C1 and C5 is not linked to C6. The established linkages were C3 to C8, C4 to C7, C9 to C 11, and C10 to C12. The two cysteines in the non-conserved mid-region unique to IGFBP-4 (C13 and C14) are linked together. Linkage of the C-terminal cysteine residues is identical to that of IGFBP-2, -5 and -6 (C15 to C16, C17 to C18 and C19 to C20). The central flexible core of IGFBP-4, containing two additional cysteines may contribute to its unique biological action.

Characterization and hormonal regulation of granulosa cell-derived insulin-like growth factor binding protein-4.

OBJECTIVE: Because of the potential importance of insulin-like growth factor binding protein-4 (IGFBP-4) to ovarian physiology and the obvious limitations imposed by in vivo-exclusive experimental paradigms, we set out to delineate the characteristics and hormonal regulation of granulosa cell-derived IGFBP-4 under in vitro circumstances. METHODS: Granulosa cells obtained by follicular puncture of the ovaries from diethylstilbestrol-primed intact immature rats were subjected to culture for up to 72 hours. Insulin-like growth factor binding protein-4 mRNA extracted from culture was subjected to Northern blot hybridization. Data normalization was assured by reprobing with the hamster Chinese hamster ovary B (CHOB) cDNA, and the IGFBP-4/CHOB ratio was calculated. Conditioned culture media were subjected to Western ligand blot before and after immunoprecipitation with a rat IGFBP-4-directed polyclonal antiserum (alpha-B104). RESULTS: Immunoprecipitation studies revealed granulosa cell-derived IGFBP-4 to be composed of a major 24-kDa species as well as a relatively minor 27-kDa moiety. Given cultures of untreated granulosa cells from immature estrogen-treated rats, transcripts corresponding to IGFBP-4 displayed an initial temporary decline culminating in a 6-hour nadir (a decrease of 67%; P < .05) followed by relatively prompt recovery (within 24 hours) to levels comparable to those noted at the outset of the culture (time 0). However, additional (albeit statistically insignificant) increments were noted at the 48-hour (but not 72-hour) time point. Treatment of granulosa cells with increasing concentration of FSH resulted in decrements of up to 30% (P < .05) in the steady-state levels of IGFBP-4 transcripts. A modest, biphasic, time-dependent response was noted for IGFBP-4 transcripts after treatment with high-dose FSH (100 ng/mL), an effect characterized by 24- and 48-hour increments (51% [P < .05] and 26% [P = .052] over untrated controls, respectively) and a 72-hour decrement (25%; P = .16). The concurrent provision of the C19 aromatase substrate androstenedione (10(-7) mol/L) to the culture medium from 72 hours enhanced the inhibitory effect of FSH (100 ng/mL) for a maximal decrement in IGFBP-4 transcripts of 49% (P < .05). Treatment with insulin-like growth factor (IGF)-I produced limited inhibition (up to 26%) of the steady-state levels of IGFBP-4 transcripts (P < .05). CONCLUSION: Findings indicate the existence of heterogeneously-sized IGFBP-4 species, of which the 27-kDa (as distinct from the 24-kDa) IGFBP-4 moiety constitutes a relatively minor component. The steady-state levels of granulosa cell-derived IGFBP-4 transcripts display relatively limited regulation in response to treatment with either FSH or IGF-I.

Insulin-like growth factor-binding protein-6 produced by human PC-3 prostate cancer cells: isolation, characterization and its biological action.

The PC-3 human prostatic carcinoma cell line has been extensively used as a model for studies on the regulation of prostate tumor cell proliferation. Because of the importance of IGF-binding proteins (IGFBPs) in the control of IGF activities that regulate cell proliferation in normal and malignant cell types, we undertook studies to characterize the IGFBPs produced by PC-3 prostate tumor cells in culture. We previously found, using an IGF-I affinity column for purification and a polyethylene glycol (PEG) precipitation assay for IGFBP detection, that PC-3 cells in culture produced a single predominant IGFBP, IGFBP-4, which inhibits IGF activities. We now present evidence that PC-3 cells also produce IGFBP-6 in abundant quantity; in the previous study this was apparently not detected in the IGF-I-bound fraction with the PEG precipitation and Western ligand blot assays. In the current study, IGF-II affinity purification of IGFBPs produced by PC-3 cells, followed by C8 HPLC reverse-phase chromatography using a shallow acetonitrile gradient, produced two major protein peaks. N-terminal amino acid sequence of peak 1 was identical to that of IGFBP-6 while that of peak 2 was identical to that of IGFBP-4. Characterization of purified IGFBP-6 from PC-3 cells revealed properties which are distinct from other IGFBPs. PEG did not precipitate the complex of 125I-IGF-II/IGFBP-6 while it precipitated the complexes between 125I-IGF-II and other IGFBPs. Indeed, IGFBP-6 decreased the amount of 125I-IGF-II tracer in the PEG precipitate in a dose-dependent manner. Also, the binding of IGFBP-6 with 125I-IGF-II was poor in Western ligand blots compared with other IGFBPs. In studies on IGFBP-6 actions, IGFBP-6 completely inhibited IGF-II-induced [3H]thymidine incorporation in MC3T3-E1 mouse osteoblast cells while it had only minimal inhibitory effects on IGF-I-induced [3H]thymidine incorporation. This differential effect is associated with the fact that IGFBP-6 has greater affinity for IGF-II than IGF-I. The results of this study indicated that (1) Western ligand blotting is not sensitive for identification of IGFBP-6, (2) the unique behavior of IGFBP-6 in the PEG assay system necessitates the use of charcoal adsorption procedure for IGFBP-6 activity detection and (3) PC-3 cells should provide a useful model system for studying regulation of IGFBP-6 expression and the role of IGFBP-6 in modulating IGF actions.