Institut National de la Santé et de la Recherche Médicale, Unité 515 Hôpital Saint Antoine Paris, France
Address correspondence to: Michel Binoux, M.D., INSERM Unité 515, Hôpital Saint Antoine, 75571 Paris, France.
To the editor:
A long-standing debate is revived by Robert Baxters letter to the editor concerning the physiological significance of IGF-binding protein-3 (IGFBP-3) proteolysis in the circulation. Although debate continues, the letter provides no new arguments. It is somewhat surprising, however, that it was sparked by a purely methodological publication (1)unless the author feels that the rationale for instigating such research was groundless until definitive experimental proof has been provided that IGFBP-3 proteolysis can increase the bioavailability of IGFs to its target tissues, although, in his words, "indeed, the definitive experiment is difficult to envisage."
The quotes selected from the introduction to our article as the basis for the criticism allude to earlier work that needs to be restored to its rightful scientific context. "The authors ... point out that in human pregnancy virtually all circulating IGFBP is degraded ..." The term "degraded" naturally refers to "limited proteolysis," with which the cited sentence begins. It has been clearly demonstrated that in pregnancy serum the characteristic 42-39-kDa doublet corresponding to IGFBP-3 is no longer detectable in Western ligand blotting, indicating functional alteration (2, 3). Western immunoblot analysis confirms this disappearance of the intact IGFBP-3 doublet, with concomitant increase of an approximately 30-kDa fragment that is consistently detected in serum, sometimes accompanied by 20- and 16-kDa fragments. Its proportional quantities relative to the 42-39-kDa form vary with physiopathological status, which strongly points toward physiological significance (4, 5, 6).
"The authors state that proteolysed IGFBP-3 loses affinity for IGFs which are then redistributed towards the 40-kDa binary IGFBP-IGF complexes." This quote omits the end of the sentence, which was "... and the free fraction of IGFs." The sentence summarized the conclusions of an extensive study previously undertaken (7), which merits reiteration, since Baxter chooses to cite publications with opposing conclusions. In that study, it was clearly shown that in pregnancy plasma, free IGF-I is enriched at the expense of the fraction in the 140-kDa complexes, whereas IGF-II is concentrated in the 140-kDa complexes at the expense of the 40-kDa complexes. The redistribution of the IGFs among the three circulating pools results from the loss of affinity of IGFBP-3 for IGFs, particularly IGF-I, and their accelerated kinetics of dissociation at 37 C. The results of a subsequent study by Lee and Rechler (8) supported this interpretation. In addition, Davenport et al. (9) demonstrated that 125I-IGF-I injected into pregnant rats was cleared from the serum five times faster than that injected into nonpregnant females. Urinary clearance was not significantly increased and 125I-IGF-I did not cross the placenta. They concluded that tissue uptake of IGF-I is increased in the mother in response to the metabolic needs of gestation. (9) The potential implications of IGFBP-3 proteolysis in IGF bioavailability have further been demonstrated by the more marked biological effects of pregnancy serum than normal serum in a chick embryo fibroblast assay (10).
The results of studies based on assays of the "free" fraction of IGF in serum are also germane. Relative to total IGF-I, this fraction is approximately doubled in pregnancy (11). Also, treatment of breast cancer patients with the anti-tumor drug suramin induces near-complete proteolysis of serum IGFBP-3 associated with an elevation in levels of "free" IGF-I (12). Interestingly, Baxter maintains that "the biological significance of these measurements is unclear."
It is possible that acid or SDS treatment provokes "further damage to the proteolysed protein" and amplifies the loss of affinity for IGF-I. Baxter and Skriver (13) have shown that this loss is even greater for 125I-IGF-I, possibly through alteration of tyrosine residues 24 and 60 during the iodination process (13). Even so, far from being a drawback, this is advantageous in our ligand immunofunctional assay where the aim is specifically to detect intact IGFBP-3 and exclude its proteolytic fragments.
We do not consider it speculative to state that "the LIFA for IGFBP-3 opens new perspectives in investigating the regulation of IGFBP-3 proteolysis and IGF-I bioavailability." First, the method furnishes a sensitivity, precision, and reproducibility of quantification of intact IGFBP-3 beyond the capabilities of the semiquantitative examination of immunoblots. Secondly, associated with the classical RIAs of IGFBP-3 and IGF-I, it allows of quantification of IGFBP-3 proteolysis and use of the IGF-I/intact IGFBP-3 ratio as an index of the exchangeable (bioavailable) fraction of IGFBP-3-bound IGF-I. The first practical applications of the technique were published in the May issue of this journal (14). Apart from classical indications in clinical investigation in cases of abnormal GH secretion and growth and nutritional disorders, determination of the proportions of intact and proteolysed IGFBP-3 as related to IGF-I levels will be useful, not least in view of epidemiological studies suggesting a link between high plasma IGF-I and low IGFBP-3 levels and the risk of prostate, breast, and colorectal cancers (15). Different approaches from ours are certainly valid for quantifying the intact and proteolysed forms of IGFBP-3. This is so in the novel ELISAs capable of differential determination of intact and fragmented IGFBP-3 variants developed by Diagnostic Systems Laboratories, Inc. (16).
Received May 24, 2001.
References
HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |