Pergamon Progress in Growth Factor Research, Vol. 6. Nos. 2~,, pp. 425~,32, 1995

Copyright © 1996 Elsevier Science Ltd. All rights reserved Printed in Great Britain.

0955-2235/95 S29.00 + .00

0955-2235(95)00026-7

PHENOTYPIC MANIFESTATIONS OF INSULIN- LIKE GROWTH FACTOR BINDING PROTEIN-1 (IGFBP-1) AND IGFBP-30VEREXPRESSION IN

TRANSGENIC MICE

Liam J. Murphy,* Kadaba Rajkumar and Peter Molnar

Departments of Internal Medicine & Physiology, University of Manitoba, Winnipeg, Manitoba R3E OW3, Canada

To provide further insight into the function of the IGFBPs, transgenic (Tg) mice which overexpressed IGFBP-1 and IGFBP-3 were generated. In this report we have compared the phenotypic manifestations observed in these Tg mice. The IGFBP-1 Tg mice were significantly smaller at birth, birth weight and gamed less weight in the post- natal period. Organ weight was proportionately reduced relative to body weight in most organs. However the brain was markedly smaller in IGFBP-1 Tg mice. Mean plasma levels of Tg-derived IGFBP-1 ranged from 8 to 80 ng m1-1 in the different groups of lGFBP-1 Tg mice. In addition homozygous mice also demonstrated fasting hyperglycemia, impaired glucose tolerance and reduced fecundity. Two of the seven IGFBP-3 founders had measurable levels of hlGFBP-3 in the circulation and were bred to homozygosity. Maximal plasma levels of transgene-derived IGFBP-3 were 72-198 ng m1-1. Transgene expression was detected in the kidney, small intestine and colon by Northern blot analysis. The birth weight, litter size and body weight of IGFBP-3 Tg mice were not significantly different from wild-type mice. However, the spleen, liver and heart of IGFBP-3 Tg mice derived from both founders were significantly heavier compared with organs from wild-type mice. The relative weight of other organs such as the brain, kidney and lungs were similar to wild-type mice. From these data, we conclude that over expression of IGFBP-I results in inhibition of IGF action and in profound impairment of brain development, modest inhibition of fetal and postnatal growth and inhibition of the metabolic effects of the IGFs. In contrast, modest over- expression of hlGFBP-3 has little effect other than some selective organomegaly.

Keywords: Hyperglycemia, glucose intolerance, metabolic effects of IGFs , growth, organomegaly.

*Correspondence to: Liam J. Murphy, M.D. Ph.D., Basic Medical Science Bldg., Room 435, 730 William Avenue, Winnipeg, MB, R3E OW3, Canada, Fax: (204) 774-7751.

Acknowledgements This work was supported by grants from the Medical Research Council and the Canadian Diabetes Association. L.J.M. is a recipient of a Medical Research Council Scientist award and an endowed Research Professorship in Metabolic Diseases.

425

426 L. J. Murphy et al.

INTRODUCTION

The insulin-like growth factor binding proteins (IGFBPs) are present in the serum, other biological fluids and tissue extracts. They bind the IGFs with affinities comparable with the IGF-I receptor and thus modulate the bioavailability of the IGFs. Both enhancement and inhibition of IGF action by various IGFBPs have been reported in different in vitro assays [1]. In addition to the other potential func- tions that have been proposed for the IGFBPs, one important function that the IGFBPs appear to serve, is to limit the hypoglycemic effects of circulating IGF-I and IGF-II [2, 3].

Six members of the IGFBP gene family have now been identified and the cognate cDNAs from a variety of species have been cloned [4]. Although they share approx- imately 50% homology, they are derived from separate genes and show different regulation and tissue-specific expression [5]. Of the IGFBPs which have been shown to be present in the circulation, IGFBP-3 appears to be responsible for the major- ity of the IGF binding capacity present in plasma [1]. However, virtually all the IGFBP-3 present in the circulation is saturated with IGF-I or IGF-II [1], and IGF- I or II bound to IGFBP-3 in the ternary complex has a long half-life compared with IGF bound to the other IGFBPs present in the circulation [6]. Although IGFBP-1 is present in much lower concentrations than IGFBP-3 and has an affinity for IGFs, is approximately one-fifth to one-tenth that of IGFBP-3 [2], it is largely unsaturated and is acutely regulated by a variety of hormones, most notably insulin and growth hormone [7, 8].

The physiological role of the IGFBPs remains unclear. There appears to be considerable overlap and possibly some redundancy of function of the IGFBPs. Indeed IGFBP-1 and -3 are located on the same chromosome, chromosome 7 in man, and may have arisen by gene duplication [9]. However, the differences in tissue specific expression and regulation of these two IGFBPs would argue for unique functions for each of these IGFBPs. As yet no mutant animal models have been described which could provide any insight into the action of these or other IGFBPs. An inherent difficulty in studying the physiology of the IGFBPs is the fact that perturbations which modulate expression of one binding protein, have effects on other IGFBPs. For example, hypophysectomy, food-deprivation and diabetes reduce IGFBP-3 expression but increase expression of IGFBP-1 and IGFBP-2 [4, 7]. Transgenic mouse technology, and more recently the generation of mutant mice by homologus recombination gene deletion, have proved to be useful techniques to study the function of various gene products. In an attempt to determine the physi- ological roles of the IGFBPs we have generated transgenic mice which over-express IGFBP-1 and IGFBP-3. Here we will review the phenotypic manifestations of IGFBP-1 and -3 over-expressions in transgenic mice.

GENERATION OF TRANSGENIC MICE

The methods used for the generation and characterization of the Tg mice have been described in detail elsewhere and will only be reviewed here briefly [10, 11]. The founders for the IGFBP-3 and IGFBP-1 Tg mice were derived from slightly different genetic backgrounds. However, it is important to note that non-transgenic

Overexpression of lGFBP-1 and-3 in Transgenic Mice 427

mice derived from the F1 cross of the founders with CD-1 mice were bred in a similar fashion to the TG mice to provide wild-type, non-transgenic mice of the same genetic background as the Tg animals.

For IGFBP-3 Tg mice, a full-length human IGFBP-3 cDNA was subcloned downstream of the mouse metallothionein (nMT-1) promoter to generate the trans- gene. The plasmid containing the mMT-1/IGFBP-3 fragment was linearized and micro-injected into fertilized embryos derived from CD-I mice which had been superovulated and mated with B6CBAF~/J F1 hybrid mice. The micro-injected embryos were transferred into CD-1 foster mice. For IGFBP-1 Tg mice the trans- gene was constructed using -5 kb rat genomic fragment containing the entire coding region of the IGFBP-1 gene including the 3' untranslated sequence, the 5' untranslated region and 78 bp of 5' flanking DNA. This fragment was inserted downstream of the mouse phosphoglycerate kinase (PGK) promoter. Tg mice were generated by pronuclear injection of the PGK-IGFBP-1 fragment devoid of plasmid sequences into fertilized BL57J/CDA F1 zygotes. Founders were bred to homozygosity with wild-type CD-1 mice.

Two IGFBP-3 Tg founders, B252 and 174 were breed to homozygosity whereas four IGFBP-1 founders, 57C, 195A, 277A and 205D were used to develop four IGFBP-1 Tg mouse strains. Southern blot analysis of D N A from the different mouse strains indicated that they were all unique in terms of the site of insertion of the transgene and also differed widely in transgene copy number (Table 1).

Serum levels of the transgene products were determined using species specific radioimmunoassays. The levels of human IGFBP-3 in founders were quite low in the absence of induction however levels as high as 200 ng m1-1 were achieved in homozygous mice supplemented with zinc chloride. Serum levels of rat IGFBP-1 were undetectable in offspring of the 205D founder but mean serum rat IGFBP-1 concentrations ranged between 8-80 ng ml -~ in the other IGFBP-1 Tg mouse strains (Table 1).

TABLE 1. Characterization of the IGFBP transgenic mouse strains

Transgene Copy number* Promoter Serum Levels #

IGFBP-3

174 12 mMT 198 B252 26 mMT 72

IGFBP-1

57C 15 PGK 24 277A 6 PGK 79 195C 10 PGK 31 205D 5 PGK Not detected

* Copy number of the transgene was determined in homozygous mice by comparison with human lymphocyte DNA for IGFBP-3 transgenic mice and the endogenous PGK promoter for IGFBP-1 trans- genic mice. # Maximal serum levels, (ng per ml) of transgene product observed in transgenic mice of the indicated strain. IGFBP-3 transgenic mice were supplement with drinking water containing 25 mM zinc chloride.

428 L. J. Murphy et al.

GROWTH CHARACTERISTICS OF TG MICE

IGFBP-3 transgenic mice were phenotypically normal at birth and grew propor- tionately postnatally (Fig. 1). There was no significant difference between the body weight of either 174 or B252 Tg mice and wild-type mice at any age. In contrast IGFBP-1 Tg mice of all founders, except 205D, were significantly smaller at birth. The birth weight of Tg offspring of founder 205D, although reduced, was not signif- icantly different from non-transgenic litter mates. This strain of IGFBP-1 Tg mice was considered to be a low expressing strain since in all tissues examined, transgene- derive IGFBP-1 mRNA was less abundant than in the other IGFBP-1 Tg strains and the transgene product was not detected in the circulation. After weaning, the growth retardation observed in the IGFBP-1 Tg mice persisted and by the end

i 0

120

100

80

60

.-U'--

L

[ ] IGFBP-3

im IGFBP-1

N e w b o r n M a l e s F e m a l e Males F e m a l e

1 m o n t h o l d A d u l t

FIGURE 1. Body weight of Tg mice expressed as a percentage of their non-transgenic litter mates. Data which represent the mean ~ E M for six to 13 mice per group are shown for the IGFBP-3 Tg strain 174 and the IGFBP-1 Tg strain 57C.** indicates P<0.01 for the difference between the Tg mice and the wild-type litter mates.

Overexpression of lGFBP-I and-3 in Transgenic Mice 429

of 1 month of age the Tg mice were ~5 grams lighter than the non-transgenic mice.

ALLOMETRY IN IGFBP-1 AND IGFBP-3 TG MICE

Organ weight relative to total body weight for the IGFBP-1 and IGFBP-3 Tg mice have been discussed in detail elsewhere [10, 11]. In IGFBP-3 Tg mice there was selective enlargement of the liver, spleen and heart. This was apparent in the two independent IGFBP-3 Tg mouse strains. Splenomegaly was also observed in two of the four IGFBP-1 Tg mouse strains; however, the most notable finding in the IGFBP-1 Tg mice was the very marked reduction in brain size. This reduction in brain weight was observed in all four IGFBP-I Tg stains including the 205D strain. Not only was the relative brain weight reduced, the mice were approximately 5 g lighter and thus the absolute brain weight was reduced to 60% of the wild type mice. There was no gross neurological manifestations of this apparent reduction in brain weight. The relative organ weights for adult male IGFBP-3 Tg mice of the 174 strain and IGFBP-1 transgenic mice of the 57C strain are shown in Fig. 2. For comparison, similar data from growth hormone [12] and IGF-1 transgenic mice [13] are also shown. It is of interest that splenomegaly has been noted in all of these Tg mice strains. IGFBP-3 Tg mice, like GH Tg mice demonstrated enlargement of both the liver and heart.

F)-I P-3

O Body Brain Heart Spleen Lung Kidney Liver

FIGURE 2. Relative organ weight in IGFBP-I and IGFBP-3 Tg mice. Data have been expressed in terms of that observed in wild-type mice arbitrary assigned a value of 1. The data shown represent the mean ~ E M for five mice and is for adult male mice of the 174 and 57C strains. For comparison data for GH and 1GF-I Tg mice are shown. The latter are derived from data published in refs. [12] and [13].

430 L. J. Murphy et al.

METABOLIC ABNORMALITIES IN TRANSGENIC MICE

IGFBP-3 transgenic mice of both strains were normoglycemic. In contrast, IGFBP-1 transgenic mice of the 57C, 195C and 277A strain were hyperglycemic and hyperinsulinemic [10]. On average, fasting blood glucose was increased 37- 68% above levels seen in wild-type mice depending upon the individual Tg mouse strains. Mice from the 205D strain did not demonstrate hyperglycemia. The hyperglycemic IGFBP-1 Tg mice were found to be equally responsive to insulin as wild-type mice. The nature of this disturbance in glucose homeostasis in IGFBP-1 transgenic mice is unclear but does not appear to be explained on the basis of insulin resistance.

REPRODUCTIVE FUNCTION IN TRANSGENIC MICE

IGFBP-3 Tg mice appeared to reproduce and suckle their young appropriately. Litter size was not significantly different in IGFBP-3 Tg mice to that observed in wild-type mice (Fig. 3). In contrast, litter size was reduced in IGFBP-1 Tg mice. The mean litter size in the strains 195C, 57C and 277A were 5.9, 7.1 and 7.9

l O . ]

8 .

27

6 •

L i t t e r size

TT T

T

25

4:

I

16 14

W t 174 B 2 5 2 195C 5 7 C 2 7 7 A

IGFBP-3 IGFBP-1

FIGURE 3. Litter size in IGFBP-1 and IGFBP-3 Tg mice. The mean :[:SEM number of pups per litter is shown. The number in each histogram indicates the number of litters examined. The Wt mice are derived from a non-transgenic offspring of an IGFBP-I Tg mouse crossed to a CD-I mouse and breed in an iden- tical fashion to the Tg mice. *indicates P < 0.1 for the differences between Tg and Wt mouse groups.

Overexpression of IGFBP-I and-3 in Transgenic Mice 431

compared with 9.8 for non-transgenic mice, derived from the same genetic back- ground and bred in an identical fashion (Fig. 3). Abundant expression of the trans- gene was apparent in various tissues of the female reproductive tract including the uterus and ovary [10]. When the ovaries of wild-type and IGFBP-1 Tg mice were compared following spontaneous ovulation, the Tg mice had significantly less corpus lutei compared with wild-type mice. In addition, the uterine weight gain and uterine DNA synthesis was significantly reduced in ovariectomized IGFBP-1 following estradiol treatment compared to wild-type mice (manuscript in preparation).

FUTURE DIRECTIONS

These preliminary studies in IGFBP-1 and IGFBP-3 Tg mice indicate that these binding proteins are likely to have important functions in a number of organ systems. In addition, our observations indicate that these two binding proteins serve quite different functions. Despite comparable levels of transgene product in the circulation, quite different phenotypic manifestations were apparent. For example hyperglycemia was observed in IGFBP-1 Tg animals but not the IGFBP-3 Tg mice. Cardiomegaly and hepatomegaly were prominent features of IGFBP-3 overexpres- sion but not apparent in the IGFBP-1 Tg mice. However, it is important to note that some of these difference in phenotype may be due to the nature of the trans- gene construct used. The PGK promoter resulted in widespread tissue expression probably from very early in embryonic life whereas the mMT promoter result in quite restricted tissue expression and was only apparent after induction with zinc chloride. Differences in transgene constructs may explain the phenotypic differences between the IGFBP-1 Tg mice reported here and those reported by Dai et al, [14]. Their Tg mice, which were generated using the mMT promoter to drive a human IGFBP-1 cDNA, were normoglycemic and of normal body weight. The only mani- festations of IGFBP-1 overexpression was a reduced brain weight. Thus, they resembled the 205D strain of IGFBP-I Tg mice generated in this laboratory where the only abnormality detected was a reduction in brain weight. It will be important for other laboratories to confirm the observations reported here using different promoters and transgene constructs. In addition there is a clear need for these studies of IGFBP overexpressing mice to be complemented with studies in IGFBP mutant mice generated by homologous recombination gene deletion.

REFERENCES

1~ Baxter RC. Circulating binding proteins for the insulinlike growth factors. Trends Endocrinol Metab. 1993; 4: 91-96.

2. Lewitt MS, Baxter RC. Insulin-like growth factor binding protein-l: a role in glucose counter- regulation? Mol Cell Endocrinol. 1991; 79: C147-152.

3. Meuli C, Zapf J, Froesch ER. NSILA-carrier protein abolishes the action of nonsuppressible insulin-like activity (NSILA-S) on perfused rat heart. Diabetologia. 1978; 14: 255-261.

4. Rechler MM. Insulin-like growth factor binding proteins. Vit Horm. 1993; 47: 1-114. 5. Shimasaki S, Ling N. Identification and molecular characterization of insulin-like growth factor

binding proteins (IGFBP-I, -2, -3, -4, 5 and -6). Prog Growth Factor Res. 1991; 3; 243-266. 6. Zapf J, Hauri C, Waldvogel M, Foresch ER. Acute metabolic effects and half-lives of intravenously

administered insulin-like growth factor I and II in normal and hypophysectomized rat.

432 L. J. Murphy et al.

J Clin Invest. 1986; 77: 1768-1775. 7. Murphy L J, Luo J, Seneviratne C. Hormonal regulation of insulin-like growth factor binding

protein-I expression in the rat. Adv Exp Med Biol. 1991; 293: 149-161. 8. Seneviratne C, Luo J, Murphy LJ. Transcriptional regulation of insulin-like growth factor binding

protein-I expression by growth hormone. Mol Endocrinol. 1990; 4: 1199-1204. 9. Lee PDK, Conover CA, Powell DR. Regulation and function of insulin-like growth factor-binding

protein-1. Proc Soc E.,cptl Biol Med 1993; 204: 4-29. 10. Rajkumar K, Barron D, Lewitt MS, Murphy LJ. Growth retardation and hyperglycemia in insulin-

like growth factor binding protein-1 transgenic mice. Endocrinology. 1995; 136: 4029-4034. I 1. Murphy L J, Molnar P, Lu X, Huang H. Expression of human insulin-like growth factor binding

protein-3 in transgenic mice. J Mol Endocrinol. 1996; in press. 12. Shea BT, Hammer RE, Brinster RL. Growth allometry of organs in giant transgenic mice.

Endocrinology. 1987; 121: 1924-1.930. 13. Mathews LS, Hammer RE, Behringer RR, D'Ercole AJ, Bell GI, Brinster RL, Palmiter RD.

Growth enhancement of transgenic mice expressing human insulin-like growth factor I. Endocrinology 1988; 123: 2827-2833.

14. Dai Z, Xing Y, Boney CM, Clemmons DR, D'Ercole AJ. Human insulin-like growth factor- binding protein-I (hlGFBP-1) in transgenic mice: characterization and insights into the regulation of IGFBP-1 expression. Endocrinology. 1994; 135: 1316-1327.