ELSEVIER
Expression of a high molecular weight form of insulin-like growth factor II in a Beckwith-Wiedemann syndrome associated adrenocortical adenoma
P.N. Schofielda,*, A. Nystromb, J. Smitha, L. Spitze, D. Grante, J. Zapfd
ªDepartment of Anatomy, University of Cambridge, Downing Street, Cambridge CB2 3DY, UK Department. of Experimental Drug Research, Karolinska Hospital, Stockholm, Sweden
“Departments of Surgery and Endocrinology, The Hospital for Sick Children, Great Ormond Street, London, UK dDepartment of Internal Medicine, Metabolic Unit, University Hospital, Zurich, Switzerland
Received 12 July 1994; revision received 19 April 1995; accepted 27 April 1995
Abstract
Beckwith-Wiedemann syndrome is a rare condition (1/13 700 live births) occurring in both inherited and sporadic forms in the population. It is manifest as a fetal overgrowth syndrome, in which hypertrophy dominates the clinical picture. An additional complication is that these children are predisposed to a specific subset of childhood neoplasms, amongst which are Wilms’ tumour and adrenocortical carcinoma. We report here the synthesis by an associated adrenal tumour of large quantities of a high molecular weight form of insulin-like growth factor Il (IGF-II), associated with profound suppression of circulating IGFs in the patient’s serum. As with other tumours of this type, the tumours showed loss of material on chromosome 11p.
Keywords: Beckwith-Wiedemann syndrome; Insulin-like growth factor II; Adrenocortical adenoma; Chromosome 11 tumour hypoglycaemia
1. Introduction
Beckwith-Wiedemann syndrome is a rare inher- ited condition which usually presents as a particular congenital pattern of organomegaly and generalised hyperplasia. This is more fully described in Ref. [10]; an additional complication is that these children have a 12% risk of developing a series of rare childhood tumours, mainly hepatoblastoma, adrenocortical car- cinoma and Wilms’ tumour. The disease is manifest
in both sporadic and inherited forms, although the inherited form is much more rare than the sporadic (for review see Ref. [17]). Previous evidence sug- gested that the lesion was inherited as a dominant mutation but partial penetrance effects noted early in the study of this syndrome [20] are probably due to the recently described phenomenon of parental im- printing of the disease locus or the degree to which the patient is mosaic for the lesion (often paternal disomy) in sporadic cases [5,13,15,21,38].
In the type of tumours found in these patients, mainly Wilms’ tumour (WT), but also adrenocortical carcinoma (ADCC), rhabdomyosarcoma and hepa-
* Corresponding author, Tel .: +44 1223 333754; Fax: +44 1223 333786.
toblastoma, expression of the insulin-like growth factor II (IGF-II) gene is usually substantially elevated, as demonstrated by RNA assay or in situ hybridisation [1,11,23,34]. Examination of expres- sion of the IGF-Il gene in hyperplastic areas of patients mosaic for the BWS lesion also demonstrates inappropriately high levels of expression of IGF-II [40]. These observations have been taken to impli- cate IGF-II in the BWS lesion. However, although the IGF-II gene is located close to the involved region of chromosome 11 (11p15.5) recombination analysis has in some cases shown [26] and in others failed to demonstrate linkage (14,22]. More recent data suggest two candidate loci, one at 1lp15.5 centromere proximal to IGF2 [19,30] and the other in the region of 11p13. This leads to the hypothesis that IGF-II expression may be controlled by the gene/region affected in BWS, either directly through a process such as imprinting, or rather more indi- rectly as a concomitant property of the state of cell differentiation. Examination of the imprinted status of IGF-II in hyperplastic tissues of BWS indi- viduals, however. has indicated heterogeneity, and while in some cases imprinting is affected in others it is not [24,40] Consequently dysregulation may result from either aberrant imprinting, breakdown of classical promoter control or perhaps changes in messenger stability. In some developmental tu- mours it is clear, however, that at least one factor leading to IGF-II overexpression is loss of its genomic imprint [23]. Such analysis has not been carried out on tumours from a BWS genetic back- ground.
Irrespective of the mechanisms leading to IGF-II overexpression it is clear that the pathology of BWS is consistent with overexpression of a growth pro- moter, whether this be the primary lesion or a secon- dary consequence. Previous measurements of the amount of IGF-II protein produced by Wilms’ tu- mour have indicated that there may be a level of translational control suppressing protein production from abundant mRNA [1,11], however other authors report high levels of protein expression from Wilms’ tumours [e.g. 27]. To implicate IGF-II in the ob- served hyperplasia protein production needs to be demonstrated directly.
We report here a benign adrenocortical tumour from a patient suffering from overt BWS with high
levels of expression of IGF-II mRNA which is trans- lated into a high molecular weight peptide. This is the first report of production of IGF-II peptide from a tumour derived from a BWS background. We also demonstrate that the tumour has undergone loss of material from chromosome 11.
2. Materials and methods
2.1. Patient
The patient was a 5-month-old boy who presented with the characteristic facial appearance of WBS and marked hemihypertrophy and macrosomia with an abdominal mass and elevated plasma cortisol. The constitutive karyotype indicated a possible small duplication deficiency anomaly at the tip of the short arm of chromosome 11 (46, XY, DUP. (11) (p15.2). Both parents had normal karyotypes and there was no evidence for a family history of the syndrome. Blood glucose was normal (3.4 mM) at examination before surgery, but the infant was obese with hemihypertro- phy of the right side of the tongue, right arm and leg, and had a moderately sized umbilical hernia. Cortisol secretion was mildly elevated (2400 h cortisol 778 nmol/l, 0900 h cortisol 860 nmol/l) Investigation revealed a 3 × 5 cm suprarenal mass, which was diagnosed on removal as a non-malignant adrenal adenoma.
2.2. Nucleic acid analysis
Genomic DNA was prepared from peripheral lymphocytes as described in [33], and three polymorphic loci examined. Catalase (CAT), Parathy- roid hormone (PTH) and p2.1 (D11S150). Probes and hybridisation conditions used are described in Ref. [21]. The 5’ VNTR p2.1 probe was the kind gift of Dr. Tony Brookes [4], and the human H19 probe from Dr. S. Tilghman. Alleles are numbered by the decreasing size of the polymorphic restriction enzyme fragments visualised on hybridisation. Tumour DNA was prepared according to [29], and restriction fragment polymorphism analysis carried out as above. Total RNA was prepar- ed by homogenising the tissue with a polytron homogeniser in guanidinium isothiocyanate as described in Ref. [31], and subjected to Northern blotting and hybridisation to an IGF-II cDNA [31] (hep 5).
2.3. Cell culture and IGF assay
Small samples of the tumour were separated into pieces for RNA, DNA, and IGF peptide analysis, together with material for cell culture. Tumour mate- rial was cut into fragments about 0.5 mm3, in 50/50 a-MEM/Ham’s F12/10% FCS. Fragments were al- lowed to settle under gravity and incubated for 1 h at room temperature in collagenase [25]. Collagenased samples were shaken vigorously to break into clumps of 50-100 cells and plate into 25 cm2 tissue culture flasks at a density of about 1 × 106 cells per flask in 5 ml. Parallel cultures were set up on irradiated mouse 10T1/2 feeder cell layers with and without 10% serum. Feeder cells did not affect outgrowth of cultures. After 1 day when cells had attached, cul- tures were overlaid with serum free medium (Alpha modified Eagles’ medium/Ham’s F12, 50/50 v/v) and successive samples collected after 24 and 48 h. IGF peptides were assayed as described in [43], as modi- fied in [42]. Tumour IGFs were measured after ho- mogenisation in 1 M acetic acid, 0.02% human se- rum albumin, 1 mM PMSF, aprotinin and pepstatin. Molecular weight standards were insulin (6 kDa), ribonuclease A (13.7 kDa) and ovalbumin (43 kDa). Elution profiles were compared with that of recombi- nant human IGF-II and serum from a normal individ- ual.
3. Results
3.1. Measurement of IGFs in tumour and serum
Serum levels of insulin-like growth factors were assayed as described in Section 2, and were found to be surprisingly low, with mean values of two deter- minations IGF-I 18 ng/ml, IGF-II, 136 ng/ml. (Usual levels for age group (0-8 years) are in the region of 93 ± 10 and 553 ± 60 ng/ml respectively [44].) In- sufficient serum was available to estimate the mo- lecular size of this material. Such a suppression of circulating IGF-I levels is also seen in cases of tu- mour hypoglycaemia usually associated with large sarcomas, which produce a high molecular weight form of IGF-II. However, IGF-II levels are not usu- ally affected so profoundly. It was consequently im- portant to assay production of IGF-II from the tu- mour cells. Over the first 24 h, 15 ng/ml/106 cells of IGF-II was produced by the tumour. Successive samples were taken from the remaining serum free
18
16
43.0
13.7
6.0
ng/ml IGF-II(tumour sample)
14
300
12
ng/ml IGF-II
10
200
8
6
4
100
2
Vi
0
0
20
40
60
80
100
120
140
FRACTION NUMBER
cultures on day 2 and day 3. During this time epithe- lioid cells grew out and dominated the culture, inter- spersed with small tightly clumped nests of morpho- logically similar cells. These clumps formed secon- darily and were not present in the initial culture. During this period levels of secreted IGF-II increased to 45 ng/ml per 106 cells per 24 h, and IGF-I to 11.25 ng/ml per 106 cells per 24 h. Assay of total IGF-II extracted from 30 mg of solid tumour and chromatographed in 1 M acetic acid over a Biogel P- 60 column (Fig. 1) indicated that all the IGF-II de- tected was of high molecular weight (>15 kDa), and yielded 32 ng of large IGF-II/30 mg tumour dry weight. Normal IGF-II was undetectable in the sam- ple and no measurable IGF-II of any form was found in a sample of normal tissue taken from nearby.
A second sample of tissue was homogenised and RNA prepared as described in Section 2. Ten micro- grams of total RNA was separated on a formaldehyde denaturing RNA gel together with parallel samples from fetal adrenal gland taken from two fetuses of 12 and 15 weeks, and control samples. Clearly levels of mRNA are substantially elevated to that usual in fetal tissue [3,34,39]. The predominant mRNA species was 6.0 kb, characteristic of the fetal gland and use of the IGF-II PIII promoter [31], 5.0, 4.8 and 1.9 kb transcripts were present at lower levels (see Fig. 2).
ABCDEFG
6.0
4.8
IGF-11
3.2
1.9
1.4
GAPDH
Lack of IGF-II RNA expression in mature adrenal glands has been previously reported [39].
3.2. Tumour genotype.
DNA was extracted from the tumour and periph- eral lymphocytes (assumed to represent the constitu- tive genotype), cut with the appropriate restriction enzymes and restriction fragment length polymor- phisms examined at the parathyroid hormone, cata- lase and DI 1S150 loci (see Fig. 3).
The patient was heterozygous for one of the loci studied (CAT) and had clearly lost one of the two alleles (allele 1, 3.9 kb) in the tumour. Lack of paren- tal samples meant that unfortunately, the parental origin of the alleles lost was not determined. PstI digestion of the DNAs followed by probing of filters with probe p2.1 (D11S150) indicated homozygosity in the constitutional genome and lack of a signal in the tumour. To check that DNA was present on the filter in equal quantities it was rehybridised with a
A
B
C
D
1-
2-
1-
1 -
1-
2-
2-
T C
TCN
TC
T C
probe to the H19 gene which lies close to IGF-II. Two fragments were visible on Pstl digestion which were of equal intensity in the two tracks indicating that lack of a signal from the tumour DNA was not due to a loading artefact. Longer exposure of the fil- ter suggests that two alleles are visible in both the constitutive and tumour samples, but at a much lower abundance. This suggests mosaicism in the patient and heterogeneity in the tumour. Unfortunately, in- ability to obtain blood samples from either parent makes this interpretation speculative.
4. Discussion
Developmental tumours from sporadic or geneti- cally predisposed backgrounds are often found to lack one of a pair of alleles on 11p compared to the constitutive genotype, this loss of heterozygosity may be due to deletion or deletion/duplication events. The individual studied here was constitutively chromo-
somally abnormal with a cytogenetically detectable deletion duplication event in the region of 11p15.2, however, he was heterozygous at the catalase locus, indicating that the constitutive deletion did not stretch to 11p13. Allele loss was found in the adrenal adenoma at D11S150 (11p15.5) and reduction to homozygosity at catalase, 11p13. PTH, centromere proximal to HBBC and TH [28] is uninformative. Chromosome material between 11p13 and the 11p15 area of chromosome 11 in the tumour from this pa- tient has therefore been lost.
Allele loss on chromosome 11 has previously been reported in hepatoblastomas and adrenocortical adenomas from BWS patients [12,18]. In general three distinct regions of loss of heterozygosity were demonstrated, dependent on tumour type. Of these, loss of material in adrenal adenomas lay between 11p13 and 11p15.5 (HRAS-1) in different individual tumours. In the adenoma described here, one copy of the CAT locus at 11p13 was shown to be lost and both alleles at D11S150 are lost completely. We can- not distinguish accurately between hemizygosity at this region in the constitutive genotype followed by loss of the remaining allele in the tumour, and an event in which both alleles are lost simultaneously, although the former would be more likely. We con- firm previous observations [6,7,12,18] of loss of het- erozygosity in a benign BWS neoplasm, which im- plies strongly that loss of heterozygosity, usually found in malignant tumours, may be an early event in tumorigenesis rather than being necessarily associ- ated with malignancy.
This study has demonstrated that IGF-II expres- sion is elevated at both messenger RNA and protein levels in a tumour from a patient suffering from Beckwith-Wiedemann syndrome. Interestingly this patient does not have constitutively elevated serum IGF-II which leaves open the possibility that hyper- plasia generally seen in BWS is due to overexpres- sion of the gene in the fetal phase of development which is switched off before birth [31]. Explanted cells from the tumour expressed IGF-II of predomi- nantly a 15 kDa form, but this was insufficient to raise circulating levels of IGF-II in serum, which were paradoxically suppressed.
This is the first reported investigation of insulin- like growth factor expression in a developmental tumour from a patient with Beckwith-Wiedemann
syndrome. Whilst lesions in the region of IGF2 are frequently reported in sporadic Wilms’ tumours, ad- renal cortical lesions and related childhood neo- plasms, these tumours and their predisposing syn- dromes are not associated with fetal overgrowth, though Wilms’ tumour is frequently associated with kidney dysplasia.
Although it has been suggested that overexpres- sion of IGF-II in BWS causes the fetal overgrowth and hypoglycaemia seen in these patients [10], meas- urements of IGF-II levels in patients do not always support this hypothesis [cf. 35,36,41]. The data pre- sented here indicate that, as expected, levels of IGF- II expression are markedly enhanced over normal tissue and the promoter usage is quantitatively and qualitatively characteristic of fetal adrenal cortex. This mRNA, unlike that seen in many Wilms’ tu- mours [1,11] is translated, but yields a 15-25 kDa form of IGF-II on acid gel filtration. This form of IGF-II is the predominant form extracted from first trimester fetal cortex [16] and supports the contention that this class of tumour represents a persistent and undifferentiated fetal stem cell population. However, normal serum contains significant amounts of this form of IGF-II. In serum of patients bearing WT without the BWS background this form is reduced and normal 7 kDa IGF-I and II levels maintained [44]. In this patient the total levels of both IGF-I and II in serum are dramatically suppressed. This latter phenomenon is strikingly reminiscent of extrapan- creatic tumour hypoglycaemia, where IGF-I concen- trations are reduced probably as a result of suppres- sion of insulin secretion [2,32,37]. Large molecular weight forms of IGF-II are reported in tumour hypo- glycaemia associated most frequently with large in- tra-abdominal sarcomas which occur with concomi- tant suppression of IGF-I levels [9,42 and refs. therein]. Recent data [42] have demonstrated that ‘big’ IGF-II is found with the IGFBP2 and IGFBP3 containing 50 kDa serum binding protein complex rather than the usual 150 kDa ALS containing com- plex, and consequently would be expected to have increased insulin-like bioactivity, resulting from ready extravasation of the small complex, interaction with peripheral tissue insulin receptors and conse- quent hypoglycaemia. Failure to properly glycosylate the IGF-II prepropeptide may be important in gen- erating ‘big IGF-II’, but it is clear that there may also
be a defect in binding to IGFBP3 [9] whose serum concentration is decreased [8].
Although hypoglycaemia has been reported with both Wilms’ tumour and adrenocortical carcinomas it is clear that in this case the concentration of all forms of IGF-II in serum was severely depressed, and the observed suppression of IGF-1 and II does not fit easily into the existing hypothetical framework. Ele- vation of plasma corticosteroids would not be ex- pected to reduce IGF-I and II (see Ref. [2] for dis- cussion) and consequently the phenomenon must be the result of a mechanism distinct from those postu- lated so far.
Acknowledgements
The authors would like to thank Dr. L. Wong of the MRC tissue bank for provision of normal adrenal samples. This work was funded by the Cancer Re- search Campaign of Great Britain, and Barncancer- fonden of Sweden.
References
Baccarini, P., Fiorenino, M.D., Manicini, A.M., Grigioni, W.F. et al. (1993) Detection of antisense transcripts of the insulin-like growth factor 2 gene in Wilms tumour. Am. J. Pathol., 143, 1535-1542.
[2] Bang, P. and Hall, K. (1992) Insulin-like growth factors as endocrine and paracrine hormones. In: The Insulin-Like Growth Factors; Structure and Biological Functions, pp. 151-177. Editor: P.N. Schofield. Oxford University Press. Oxford.
[3] Brice, A.L., Cheetham, J.E., Bolton, V.N., Hill, N.C.W. and Schofield, P.N. (1989) Temporal changes in the expression of the insulin-like growth factor II gene associated with maturation in the human fetus. Development, 106, 543-554.
[4] Brookes, A.J., Hedge, P.H. and Solomon, E. (1989) A highly polymorphic locus on chromosome 11 which has homology to a collagen triple helix coding sequence. Nu- cleic Acids Res., 4, 1792.
[5] Brown, K.W., Williams, J.C., Maitland, N.J. and Mott, M.G. (1990) Genomic Imprinting and Beckwith-Wiedemann syn- drome. Am. J. Hum. Genet., 46, 1000-1001.
[6] Byrne, J.A,, Little, M.H. and Smith, P.J. (1992) The M1 subunit of ribonucleotide reductase refines mapping of ge- netic rearrangements at chromosome 11p15. Cancer Genet. Cytogenet., 59, 206-209.
[7] Byrne, J.A., Sims, L.A., Little, M.H., Algar, E.M. and Smith, P.J. (1993) Three non overlapping regions of chro- mosome arm 11p allele loss identified in infantile tumours of adrenal and liver. Genes Chromosomes Cancer, 8, 104- 111.
[8] Cotterill, A.M., Holly, J.M.P., Davies, S.C. et al. (1991) The insulin like growth factors and their binding proteins in a case of non-islet cell tumour-associated hypoglycaemia. J. Endocrinol., 131, 303-311.
[9] -Daughaday, W.H., Trivedi, B. and Baxter, R.C. (1993) Se- rum “big insulin-like growth factor Il” from patients with tumour hypoglycaemia lacks normal E-domain O-linked glycosylation, a possible determinant of normal peptide processing. Proc. Natl. Acad. Sci. USA, 90, 5823-5827.
[10] Engstrom, W., Lindham, S. and Schofield, P.N. (1988) Wiedemann-Beckwith syndrome. Eur. J. Paediatr., 147, 450-457.
[11] Haselbacher, G.K., Irminger, J.C., Zapf, J., Ziegler, W.H. and Humbel, R.E. (1987) Insulin-like growth factor II in human adrenal pheochromocytomas and Wilms’ tumors: expression at the mRNA and protein level. Proc Natl Acad Sci USA, 84, 1104-1106.
[12] Hayward, N.K., Little, M.H., Mortimer, R.H., Clouston, W.M. and Smith, P.J. (1988) Generation of homozygosity at the c-Ha-ras-1 locus on chromosome 1lp in an adrenal ade- noma from an adult with Beckwith-Wiedemann syndrome. Cancer Genet. Cytogenet., 30, 127-132.
[13] Henry, I., Bonaiti Pellie, C., Chehensse, V., Beldjord, C., Scwartz, C., Utermann, G. and Junien, C. (1991) Unipa- rental paternal disomy in a genetic cancer predisposing syn- drome. Nature (London), 351, 665-667.
[14] Henry, 1., Jeanpierre, M., Barichard, F., Serre, J.L., Mallet, J., Turleau, C., De Grouchy, J. and Junien, C. (1988) Dupli- cation of HRAS1, INS and IGF2 is not a common event in Beckwith-Wiedemann syndrome. Ann. Genet., 31, 216-220.
[15] Henry, I., Puech, A., Rieswijk, Ahnine, L., Mannens, M., Beldjord, C., Bitoun, P., Tournade, M.F., Landrieu, P. and Junien, C. (1993) Somatic Mosaicism for partial paternal isodisomy in Weidemann-Beckwith Syndrome: a post fer- tilisation event. Eur. J. Hum. Genet., 1, 19-29.
[16] Hill, D.J. (1990) Relative abundance and molecular sizes of immunoreactive IGFs I and II in human fetal tissues. Early Human Dev., 21, 49-58.
[17] Junien, C., (1992) Genetics of Beckwith-Weidemann syn- drome. Current Opin. Genet. Dev., 2, 431-438.
[18] Little, M.H., Thomson, D.B., Hayward, N.K. and Smith, P.J. (1988) Loss of alleles from the short arm of chromosome 11 in a hepatoblastoma from a child with Beckwith-Wiedemann syndrome. Human Genet., 79, 186-18.
[19] Mannens, M., Hoovers, J.M.N., Redeker, E., Verjaal, M., Feinberg, A.P., Little, P. et. al. (1994) Parental imprinting of human chromosome region 11p15.3-pter involved in the Beckwith-Weidemann syndrome and various human neo- plasia. Eur. J. Hum. Genet., 2, 3-23.
[20] Nikawa, N., Ishikaryama, S., Takahashi, S., Inagawa, A., Tonoki, H., Ohta, Y., Hase, N., Tamei, T and Kaji, T. (1986) The Beckwith-Wiedemann syndrome. Pedigree studies on 5 families with evidence for autosomal dominant inheritance with varying expressivity. Am. J. Med. Genet., 24, 41-45.
[21] Nystrom, A., Cheetham, J.E., Engstrom, W. and Schofield, P.N. (1992) Molecular analysis of patients with Wiedemann- Beckwith Syndrome II; paternally derived disomies of chromosome 11: Eur J. Paediatr., 151, 511-514.
[22. Nystrom, A., Hedborg, F. and Ohlsson, R. (1994) Insulin like growth factor II cannot be linked to a familial form of Beckwith-Wiedemann syndrome. Eur. J. Pediatr., 153, 574- 580.
[23] Ogawa, O., Eccles, M.R., Szeto, J., McNoe, L.A., Yun, K. et al. (1993) Relaxation of Insulin-like growth factor II gene imprinting implicated in Wilms’ tumour. Nature (London), 362, 749-751.
[24] Ohlsson, R., Nystrom, A., Pfeifer-Ohlsson, S., Tohonen, V., Hedborg, F., Schofield, P.N., Flam, F. and Ekstrom, T.(1993) The insulin-like growth factor II gene is imprinted in human development and in Beckwith-Wiedemann syn- drome. Nature (Genetics), 4, 94-97.
[25] Pera, M.F., Blasco Lafita, M.J. and Mills, J. (1987) Cultured stem cells from human testicular teratomas; the nature of human embryonal carcinoma and its comparison with two types of yolk sac carcinoma. Int. J. Cancer, 40, 334-343.
[26] Ping, A.J., Reeve, A.E., Law, E.J. et. al. (1989) Genetic linkage of BWS to 11p15. Am. J. Hum. Genet. 44, 720-728.
[27] Renqiu, Q., Ruelicke, T., Hassam, Haselbacher, G.K. and Schoenle, E.J. (1993) Systemic effects of insulin-like growth factor II produced and released from Wilms tumour tissue. Eur. J. Paediatr., 152, 102-106.
[28] Richard, C.W., Boehnke, M., Berg, D.J., Lichy, J.H., Meeker, T.C., Hauser, E. et al. (1993) A radiation hybrid map of the distal short arm of chromosome 11 containing the Beckwith-Wiedemann and associated embryonal tumour disease loci. Am. J. Hum. Genet., 52, 915-921.
[29] Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecu- lar Cloning, A Laboratory Manual. Cold Spring Harbor Press, Cold Spring Harbor, NY.
[30] Sait, S.N.J., Nowak, N.J., Singh-Kahlon, P., Weksberg, R., Squire, J., Shows, T.B. and Higgins, M.J. (1994) Localisa- tion of Beckwith Weidemann and rhabdoid tumour chromo- some rearrangements to a defined interval in chromosome band 11p15.5. Genes Chromosomes Cancer, 11, 97-110.
[31] Schofield, P.N. and Tate, V. (1987) Regulation of human IGFII transcription in fetal and adult tissues. Development, 101, 743-750.
[32] Schofield, P.N., Conner, H., Turner, R.C. and Zapf, J. (1989) Tumour hypoglycaemia: raised tumour IGF-II mRNA associated with reduced plasma somatomedins. Br. J. Cancer, 60, 661-663.
[33] Schofield, P.N., Lindham, S. and Engstrom, W. (1989)
Analysis of gene dosage on chromosome 11 in children suf- fering from Beckwith-Weidemann syndrome. Eur. J. Pae- diatr., 148, 2320-2324.
[34] Scott, J., Cowell, J., Robertson, M.E., Priestly, L.M., Wadey, R., Hopkins, B., Pritchard, J., Bell, G.I., Rall, L.B., Graham, C.F. and Knott, T.J. (1985) Insulin-like growth factor II gene expression in Wilms’ tumour and embryonic tissues. Nature (London), 317, 260-262.
[35] Spencer, G.S.G. (1981) Somatomedin C in the Beckwith- Wiedemann syndrome. Arch. Dis. Child., 56, 77-78.
[36] Spencer, G.S.G., Schabel, F. and Frisch, H. (1980) Raised somatomedin associated with normal growth hormone. Arch. Dis. Child., 55, 151-153.
[37] Teale, J.D. and Marks, V. (1990) Inappropriately elevated plasma insulin-like growth factor II in relation to suppressed IGF-I in the diagnosis of non islet cell tumour hypoglycae- mia. Clin. Endocrinol., 33, 87-98.
[38] Viljoen, D. and Ramesar, R., (1992) Evidence for paternal imprinting in familial Beckwith-Wiedemann syndrome. J. Med. Genet., 29, 221-225.
[39] Voutilainen, R. and Miller, W.L. (1988) Developmental and hormonal regulation of mRNAs for IGF-II and steroidogenic enzymes in human fetal adrenals and gonads. DNA, 7, 9-15.
[40] Weksberg, R., Shen, D.R., Fei, Y .- L. et al. (1993) Disruption of Insulin-like growth factor 2 imprinting in Beckwith- Wiedemann syndrome. Nature Genet., 5, 143-150.
[41] Weninger, M., Lischka, A., Pollak, A., Vergesslich, C., Ogris, E. and Frisch, H. (1984) Somatomedin Aktivitat bei Beckwith-Weidemann syndrom. Monatsschr. Kinderheilkd., 32, 900-903.
[42] Zapf, J., Futo, E., Peter, M. and Froesch, E.R. (1992) Can “Big Insulin-like growth factor II” in serum of tumour pa- tients account for the development of extrapancreatic tu- mour hypoglycaemia? J. Clin. Invest., 90, 2574-2584.
[43] Zapf, J., Walter, H. and Froesch, E.R. (1981) Radioimmu- nological determination of insulin like growth factors I and II in normal subjects and in patients with growth disorders and extrapancreatic tumour hypoglycaemia. J. Clin. Invest., 68, 1321-1330.
[44] Zumkeller, W., Schwander, J., Mitchell, C.D., Morrell, D.J., Schofield, P.N. and Preece, M.A. (1993) Insulin-like growth factor I, II and IGF binding protein 2 in the plasma of chil- dren with Wilms tumour. Eur. J. Cancer, 29, 1973-1977.