Free Insulin-Like Growth Factor (IGF) in Disorders of IGF Binding Protein 3 Complex Formation

William H. Daughaday

Division of Endocrinology, Department of Medicine, University of California, Irvine, Irvine, California 92697

Address all correspondence and requests for reprints to: William Daughaday, M.D., 530 South Bay Front, Box 157, Balboa Island, California 92662. E-mail: wdaymd{at}aol.com.


    Introduction
 Top
 Introduction
 References
 
The concentration of free IGF-I and -II in normal human serum is less than 0.5% of the total concentration of the IGF (1). Despite the fact that this component is critical in the feedback regulation of pituitary GH secretion and is important in the endocrine action of IGF-I on target tissues, its measurement has been little used in clinical practice. This is because the small concentrations of this entity are technically difficult to measure accurately and concentrations are subject to rapid change, depending on the fluctuations of the concentration of IGF binding protein 1 (IGFBP1) in fasting and metabolic disorders such as diabetes mellitus. For this reason, measurement of total serum IGF-I has been the accepted clinical method of evaluating IGF-I activity in patients. About 80% of what is being measured is IGF-I that is bound to the ternary complexes of IGF-I-IGFBP3-acid labile subunit (ALS) (2) and to a lesser extent to the ternary complex of IGF-IGFBP5-ALS. The remainder is in loose binary association with IGFBP1, -2, and -4. Measurement of serum IGFBP3 as an index of GH action, therefore, is a surrogate measure of the total IGF-I and IGF-II in serum. However, in conditions when the formation of the IGFBP3 ternary complex is impaired, total serum IGF no longer reflects free IGF. Two different examples by which this might occur are considered.

IGF-II binding in non-islet cell tumor hypoglycemia

It has long been recognized that certain large mesenchymal tumors and hepatomas can present with recurrent severe hypoglycemia without an elevation of serum insulin. A possible role for IGFs in non-islet cell tumor hypoglycemia (NICTH) was suspected early, but measurements of the concentrations of IGF-I in the sera of such patients were low and the concentrations of total IGF-II were usually normal (3). An explanation for this paradox was provided in 1988 by studies of extracts of a fibrosarcoma and sera from a patient with NICTH (4). The IGF-II radioimmunoactivity after acid gel filtration was of much larger size than recombinant IGF-II. Subsequently, it has been established that "big" IGF-II is a heterogeneous mixture of partially processed fragments of proIGF-II retaining the first 21 amino acids of the E domain of proIGF-II. Zapf et al. (5) found that IGF-II isolated from normal human serum and big IGF-II from NICTH sera equally displaced 125I-IGF-II bound to IGFBP2 and IGFBP3. This indicates that big IGF-II can form the binary complex with IGFBP3 normally. Another important abnormality in NICTH was observed when sera from normal individuals and patients with NICTH were passed through a Sephadex G-200 column in a neutral buffer (6). Nearly all the IGF-II radioimmunoactivity of normal sera eluted as the 150-kDa ternary IGF-IGFBP3-ALS complex. When NICTH sera were passed through this column, the IGF-II immunoactivity eluted with binary rather than with ternary complexes.

The sera of patients with NICTH has been found to contain 10 to 20 times the concentration of free IGF-II that is present in normal sera (7, 8). This was not recognized when total serum IGF-II was measured by RIA because it was masked by the larger decrease in IGF-II in the 150-kDa ternary complex. The decrease in the 150-kDa complex in sera of patients with NICTH cannot be attributed to a deficiency of ALS. The concentration of ALS was found to be 10.9 ± 5.4 mg/liter in NICTH sera compared with 24.2 ± 4.79 mg/liter in normal subjects (6), but in normal sera about half of the ALS is contained in the ternary complex and half is free. In view of the defective formation of the 150-kDa ternary complex in NICTH sera, the concentration of uncomplexed ALS in these sera was near normal. The ALS found in NICTH sera that did not form the ternary complex with the big IGF-II in NICTH sera was fully capable of forming the ternary complex when incubated with 125I-IGF-I covalently linked to IGFBP3 (6). When additional ALS was added to NICTH sera, there was an increase in ternary complex containing big IGF-II after separation through a neutral Sepharose 12 column (7). These results suggest that the binary complex of big IGF-II and IGFBP3 was present despite the suppression of GH secretion in NICTH but has reduced affinity for ALS.

Hypothesis. The big IGF-II present in NICTH binds normally to IGFBP3 to form the 50-kDa complex but by steric interference reduces the affinity of the binary complex for ALS.

Abnormal IGF-I secreted from extrahepatic sites

There has been a dispute about whether GH-stimulated IGF-I acts on cartilage as an endocrine factor or as a locally produced autocrine/paracrine growth factor (9, 10). Because the liver is normally the primary source of IGF-I in the circulation, it was reasoned that selective blocking of IGF-I synthesis in the liver would provide a definitive answer to this dispute. Because IGF-II ceases to be expressed soon after birth in rats and mice, it does not play a role in postnatal growth. Selective blocking of hepatic IGF-I synthesis was accomplished in mice by Sjogren et al. (11) by linking the Cre recombinase to the interferon-inducible Mx1 promoter and by Yakar et al. (12) by linking the Cre recombinase to the albumin promoter. In both cases, the fourth exon of the IGF-I gene was flanked by two loxP markers. The interferon promoter was activated by interferon injection of newborn mice, and the albumin promoter was physiologically activated early in postnatal development, thus inactivating the IGF-I gene selectively in the liver.

Remarkably, male liver IGF-I knockout (LID) mice grew normally, and the female mice exhibited only a slight retardation of growth. Measured free IGF-I levels in the sera of LID mice were actually slightly higher than in normal mice (Table 1Go), indicating that extrahepatic sources of IGF-I stimulated by the 4.5-fold increase in serum GH had replaced the liver as the source of the circulating IGF-I (13).


View this table:
[in this window]
[in a new window]
 
TABLE 1. IGF-I levels in sera

 
The near-normal concentrations of free IGF-I of LID mice resulted from increased extrahepatic secretion of IGF-I. This conclusion is not negated by the failure to find increased IGF-I gene transcription in kidney, muscle, spleen, lung, heart, and fat (13). Because of the great mass of these tissues, small changes in IGF-I mRNA might not be recognized, and a GH action in stimulating translation or secretion of IGF-I is possible. GH action on other tissues not examined, such as endothelium or connective tissue, could have contributed to the circulating IGF-I.

Studies of ALS knockout (ALSKO) and LID/ALSKO mice by Yakar et al. (13) provide new insights in the role of the ternary complex of IGF-I-IGFBP3-ALS in IGF-I clearance. The results of Yakar et al. (13) are abstracted in Table 1Go. The failure of ALSKO mice to form the ternary complex resulted in a 50% drop in total serum IGF-I but no change in serum free IGF-I. IGF-I turnover was not significantly increased, as evidenced by the unchanged serum GH and normal growth. In the LID mice, the serum total IGF-I concentration was reduced 44% yet the mean measured free IGF-I was actually slightly elevated, a result that is inconsistent with the 4.5-fold increase in serum GH, which indicates that the functional level of free IGF-I was below the pituitary set point. The LID/ALSKO mice were markedly growth-retarded, despite a 3.5-fold increase in free IGF-I and a 15-fold increase in GH. It is clear that despite the increase in GH the extrahepatic tissues were not able to make up for the increase in IGF-I clearance resulting from the ALS deficit.

I suggest that the measurements of free IGF-I in the sera of LID mutant mice did not accurately reflect functional activity. The method used in the studies of Yakar et al. (13) was developed by Frystyk et al. (2). This involves ultrafiltration of serum at physiological pH at 37 C and is widely accepted as the standard by which simpler methods are compared.

It is possible that immunologically reactive but functionally deficient forms of IGF-I were secreted by extrahepatic sites in LID and LID/ALSKO mice, accounting for the increased serum GH despite increased levels of free IGF-I. As discussed earlier, forms of incompletely processed proIGF-II are secreted by some mesenchymal tumors (4). It is possible that functionally impaired forms of proIGF-I were secreted by extrahepatic tissues in LID and LID/ALSKO mice. Another possibility is that partly protease cleaved IGF-I fragments that retain immunological activity but have impaired functional activity are released by extrahepatic sites. These possibilities deserve future attention.

Hypothesis. The extra hepatic synthesis of IGF-I in LID and LID/ALSKO mice may be associated with secretion of functionally impaired IGF-I molecules.

Abnormal IGFBP3 complex formation in LID and LID/ALSKO mice

An intriguing aspect of the LID mice is why the IGF-I synthesized extrahepatically formed the ternary complex with IGFBP3 and ALS incompletely as evidenced by a 50% drop in serum bound IGF-I. The reported normal liver IGFBP3 mRNA suggests that hepatic synthesis was normal. Although not measured, it is likely that hepatic ALS secretion was normal or increased because the expression of the ALS gene in rat liver is increased by GH (14).

A possible explanation for the low concentration of ternary bound IGF-I and the near normal concentration of serum free IGF-I in the LID mice can be proposed. When 123I-IGF-I is added to normal human serum, its appearance in the 150-kDa complex is exceedingly slow and incomplete after 8 h of incubation at room temperature (my unpublished observation). This must mean that there is little intact unbound IGFBP3 remaining in normal peripheral serum available to bind the added IGF-I and form the ternary complex with the readily available ALS. Just as formation of the IGF-I-IGFBP3 binary complex prolongs the half-life of disappearance of IGF-I from human plasma from 10 min to 30–90 min (15), I suggest that formation of the binary complex also prolongs the serum life of IGFBP3. Formation of the ternary complex with ALS further extends the half-life of disappearance to more than 12 h. In view of these findings, the absence of hepatic IGF-I in the LID mice might result in more rapid clearance of IGFBP3 from plasma.

There has been no comparable selective defect in hepatic IGF-I synthesis reported in human beings, but such a defect would not be expected to impair the formation of the IGF-II-IGFBP3-ALS complex.

In ALSKO mice, both IGFBP3 and IGF-I are synthesized in the liver and should form the binary complex normally. This protected IGF-I from degradation sufficiently so that there was no rise in serum GH and growth was normal. This would be comparable to the IGFBP3 potentiation of IGF-I action when both were injected simultaneously by Clark et al. (16). In the LID/ALSKO mice, the inefficient formation of binary complexes and the inability to form ternary complexes combined to increase IGF-I degradation greatly as evidenced by a 15-fold rise in serum GH and 30% decrease in body growth.

Hypothesis. Efficient formation of the IGF-I-IGBP3-ALS complex in mice occurs in the hepatic venous sinusoids where the concentrations of the reactants are optimal. When IGF-I is formed in extrahepatic tissues, the formation of the IGF-I-IGFBP-3 binary and ternary complexes are slowed, and the degradation of IGFBP3 is increased.


    Footnotes
 
Abbreviations: ALS, Acid labile subunit; ALSKO, ALS knockout; IGFBP, IGF binding protein; LID, liver IGF-I knockout; NICTH, non-islet cell tumor hypoglycemia.

Received May 20, 2003.

Accepted September 29, 2003.


    References
 Top
 Introduction
 References
 

  1. Baxter RC 1990 Circulating levels and molecular distribution of the acid-labile ({alpha}) subunit of the high molecular weight insulin-like growth factor-binding protein complex. J Clin Endocrinol Metab 70:1347–1353[Abstract]
  2. Frystyk J, Skjaerbaek C, Dinesen B, Orskov H 1994 Free insulin-like growth factors (IGF-I and IGF-II) in human serum. FEBS Lett 348:185–191[CrossRef][Medline]
  3. Widmer U, Zapf J, Froesch ER 1982 Is extra pancreatic tumor hypoglycemia associated with elevated levels of insulin-like growth factor II? J Clin Invest 55:833–839
  4. Daughaday WH, Emanuele MA, Brooks MH, Barbato AL, Kapadia M, Rotwein P 1988 Insulin-like growth factor II synthesis and secretion by a leiomyosarcoma with associated hypoglycemia. N Engl J Med 319:1434–1440[Abstract]
  5. Zapf J, Futo F, Peter M, Froesch ER 1992 Can "big" insulin-like growth factor II in serum of tumor patients account for the development of extra pancreatic tumor hypoglycemia? J Clin Invest 90:2574–2584[Medline]
  6. Baxter RC, Daughaday WH 1991 Impaired formation of the ternary insulin-like growth factor-binding protein complex in patients with hypoglycemia due to nonislet cell tumors. J Clin Endocrinol Metab 73:696–702[Abstract]
  7. Daughaday WH, Trivedi B, Baxter RC 1995 Abnormal serum IGF-II transport in non-islet cell tumor hypoglycemia results from abnormalities of both IGF binding protein-3 and acid labile subunit and leads to elevation of serum free IGF-II. Endocrine 3:425–428
  8. Frystyk J, Skjaerback C, Zapf J, Orskov H 1998 Increased levels of circulating free insulin-like growth factors in patients with non-islet cell tumour hypoglycemia. Diabetologia 41:589–594[CrossRef][Medline]
  9. Daughaday WH 2000 Growth hormone axis overview-somatomedin hypothesis. Pediatr Nephrology 14:537–540[CrossRef]
  10. Isaksson OGP, Lindahl A, Nilsson A, Isgaard J 1987 Mechanism of the stimulatory effect of growth hormone on longitudinal bone growth. Endocr Rev 8:426–438[Medline]
  11. Sjogren K, Liu J, Blad K, Skrtic S, Vidal O, Wallenius V, LeRoith D, Tornel J, Isaksson O, Jansson J, Ohlsson C 1999 Liver-derived IG principal source of IGF-I in blood but is not required for postnatal body growth in mice. Proc Natl Acad Sci USA 96:7088–7092[Abstract/Free Full Text]
  12. Yakar S, Liu JL, Stannard B, Butler A, Accili D, Saur B, LeRoith D 1999 Normal growth and development in the absence of hepatic insulin-like growth factor I. Proc Natl Acad Sci USA 96:7324–7329[Abstract/Free Full Text]
  13. Yakar S, Rosen CJ, Beamer WG, Ackert-Bieknell CL, Wu Y, Liu J-L, Ooi GT, Setser J, Frystyk J, Boiseclair YR, LeRoith D 2002 Circulating levels of IGF-1 directly regulate bone growth and density. J Clin Invest 110:771–781[Abstract/Free Full Text]
  14. Ooi GT, Cohen FJ, Tseng LY-H, Rechler MM, Glisclair VR 1997 Growth hormone stimulates transcription of the gene encoding the acid-labile subunit (ALS) of the circulating insulin-like growth factor binding protein complex and ALS promoter activity in rat liver. Mol Endocrinol 11:997–1007[Abstract/Free Full Text]
  15. Guler H-P, Zapf J, Schmid C, Froesch ER 1989 Insulin-like growth factor I and II in healthy man. Estimates of half-lives and production rates. Acta Endocrinol (Copenh) 121:753–758[Medline]
  16. Clark RG, Mortensen D, Reifsynder D, Mohler M, Etcheverry T, Mukk V 1993 Recombinant human insulin-like growth factor binding protein-(rhIGFBP-3): effects on the glycemic and growth promoting activities of rhIGF-I in the rat. Growth Regul 3:50–52[Medline]