Brain Research, 579 (1992) 1-7 Elsevier BRES 17656

1

Research Reports

The effects of pituitary stalk transection, hypophysectomy and thyroid hormone status on insulin-like growth factor 2-, growth hormone releasing hormone-, and somatostatin mRNA prevalence in rat brain A. Levy 1, Martha C. Matovelle 2, Stafford L. Lightman 1 and W.S. Young III Laboratory of Cell Biology, National Institute of Mental Health, Bethesda, MD 20892 (USA)

(Accepted 10 December 1991) Key words: Insulin-like growth factor 2; Gene expression; Pituitary; Somatostatin; Growth hormone releasing factor; Choroid plexus; Arcuate; Growth factor

We have used in situ hybridization histochemistry to determine the effects of pituitary stalk transection, hypophysectomy and drug-induced changes in thyroid status on mRNA levels encoding insulin-like growth factor 2, somatostatin, and growth hormone-releasing factor in the choroid plexus, hypothalamic periventricular nucleus, and arcuate nucleus, respectively. Pituitary stalk transection and hypophysectomy in Sprague-Dawley rats decreased insulin-like growth factor 2 and somatostatin mRNA and increased growth hormone-releasing factor mRNA. In each case, the effect of hypophysectomy exceeded that of pituitary stalk transection. Treatment with propylthiouracil for 10 days decreased somatostatin mRNA, markedly increased growth hormone-releasing factor mRNA but had no significant effect on insulin-like growth factor 2 mRNA. Treatment with triiodothyronine had no effect on the mRNAs measured. These findings corroborate the clinical observation of abnormal somatic growth in disturbances of thyroid and growth hormone status and provide further evidence of the effects of these metabolic disturbances and of pituitary disconnection and hypophysectomy on insulin-like growth factor 2 mRNA prevalence. INTRODUCTION The place of IGF-1 (insulin-like growth factor 1) as an intermediate in growth h o r m o n e function and as an inhibitor of growth h o r m o n e release has long been established 2'21. Insulin-like growth factor 2 (IGF-2), despite being identified at the same time as IGF-1 and sharing with it 70% amino acid homology 19, has a less clear place in physiology. Recent work showing that IGF-2 stimulates tritiated thymidine uptake into foetal rat brain 3'16 implicates IGF-2 in the growth and development of the normal central nervous system, and it is of interest that in the h u m a n adult particularly high IGF-2 m R N A levels have been found in a variety of malignancies 1'6'11. Nevertheless, the presence of abundant IGF-2 transcripts in the choroid plexus of adult brain, contrasting as it does with more developmentally restricted expression in other tissues 2° has not yet been adequately explained, As disturbances of both thyroid and growth hormone

status are known to have major neuromodulatory effects during development, we sought to determine whether surgically and pharmacologically-induced changes in these axes significantly alter IGF-2 gene expression. The experimental paradigms we have chosen therefore combine partial and complete abolition of growth hormone feedback with secondary hypothyroidism (produced by hypothalamo-pituitary disconnection and hypophysectomy) and isolated changes in thyroid hormone status, induced by oral treatment with triiodothyronine or propylthiouracil, an antithyroid analogue of thiourea. Quantitative evaluation of IGF-2 transcripts was limited to the primary site of central nervous system IGF-2 gene expression, the choroid plexuses of the lateral ventricles. The estimation of changes in somatostatin (SRIH) transcripts was confined to the periventricular nucleus as direct and reciprocal neural connections have been demonstrated between arcuate growth hormone-releasing factor ( G R H ) neurons and the SRIH-containing neurons of periventricular nucleus 7 and as hybridization to this

Present address: Neuroendocrinology Unit, 5th Floor, Laboratory Block, Chafing Cross and Westminster Hospital Medical School, Fulham Palace Road, London W6 8RF, UK. 2 Present address: Department of Anatomy and Cell Biology, East Carolina University School of Medicine, Greenville, North Carolina, USA. Correspondence: W.S. Young, Laboratory of Cell Biology, NIMH, Building 36, Room 2DI0, Bethesda, MD 20892, USA.

Fig. 1. Dark field image of coronal sections of rat brain taken through the arcuate nucleus and hybridized in situ to a probe complementary to growth hormone-releasing factor. Sham-operated animal (a) and following hypophysectomy (b). site could be easily delineated. To facilitate comparison between brains, G R H transcripts were measured in the arcuate nucleus 1.3 m m caudal to the rostral end of the paraventricular nucleus. Hybridization histochemistry using 35S-labelled specific oligodeoxynucleotide probes was used throughout the study for in situ m R N A estimation,

MATERIALS AND METHODS Animals

Male Sprague-Dawley rats hypophysectomized or pituitary stalk transected by the trans-oral approach and sham-operated animals were purchased from Zivic-Miller (Zelienople, PA). Prior to decapitation, total fluid intake was recorded for 48 h and only those

Fig. 2. Dark field image of coronal section of lateral ventricle choriod plexus at the level of the paraventricular nucleus. Sham-operated animal (a) and following hypophysectomy (b). drinking in excess of 200 ml/kg/day were used for the study 13. Rats were rendered thyrotoxic or hypothyroid by feeding them a powdered diet containing either 60/~g/kg triiodothyronine (equivalent to 3 /zg/kg/day (0.54 /~g/rat/day)) or 0.5% propylthiouracil, and killed after 10 days treatment or 10 days post-operatively,

Hybridization histochemistry Tissue preparation and hybridization histochemistry were performed as previously described29'3°. Briefly, immediately after decapitation,

brains were frozen on powdered dry ice. Coronal 12/zm sections cut through the periventricular nucleus and arcuate nucleus were thaw mounted onto gelatin/chrome alum coated slides and stored at -7&C. Sections were pre-hybridized at room temperature by passing them sequentially through 4% formaldehyde (i.e. 10% formalin) in phosphate buffered saline (PBS) (5 min), PBS alone (2 rinses), 0.25% acetic anhydride in 0.1 M triethanolamine HCI, pH 8 for 10 min, 70% ethanol, 80% ethanol (1 min each), 95% ethanol (2 min), 100% ethanol (1 min) 100% chloroform (5 min), 100% ethanol and

95% ethanol (1 min each) before being allowed to dry upright. Hybridization buffer contained 50% formamide, 600 mM NaCI, 80 mM Tris-HCl (pH 7.5), 4 mM EDTA, 0.1% sodium pyrophosphate, 0.2% sodium dodecyl sulphate, 0.2 mg/ml sodium heparin, 10% dextran sulphate and 100 mM dithiothreitol. Pairs of sections were incubated overnight at 37°C under Parafitm coverslips in 50/zl hybridization buffer containing 106 dpm of 35S end-labelled probe per slide. Following hybridization, slides were dipped briefly in 2 × SSC containing 50% formamide and washed for 1 h in 4 changes of the same at 42°C. After a final 30 min wash in 1 x SSC at room temperature, the slides were rinsed in water followed by 70% ethanol and air dried,

by comparison of the mean optical density alone, as the area of the choroid plexus was found to depend on the cross sectional area of the lateral venticle rather than the extent of hybridization.

Northern analysis Oligodeoxynucleotides were end-labelled using [32p]dATP (NEN) and terminal deoxynucleotide transferase, and hybridized to formaldehyde-denatured, size fractionated total rat brain RNA bound to nylon filters (Nytran). In each case the oligonucleotides hybridized to a single mRNA species of appropriate size: approximately 4.6 kb for IGF-227, 0.7 kb for SRIH 23 and 0.7 kb for GRH 14. The rat TSHI3 subunit probe had been previously validated 12.

Probes Oligodeoxynucleotides were synthesized on an Applied Biosystems DNA synthesizer (courtesy of Dr. M.J. Brownstein) to the following sequences: IGF-222: 5' GGT CAC A G A CTG ATG GTA CTA CAT TGC A G A ATI" ACC ACA TAA T I T GGT 3'; GRHIS: 5' TTG CGG GCA TAT AAT TGG CCC AGG ATI" CTC CGG TAG CTG CTG GTG A A G 3'; SRIH26: 5' GAC ACC GCC CAA AGC CAG GAC GAT GCA GAG CGC GGC CAG CGC GCA CTG 3'; TSH32: 5' GTC TCT GTA TGT ACA GAC ATC CTG AGA GAG TGC GTA CTI" GGG A A G A A A 3' and purified by polyacrylamide gel electrophoresis. For in situ hybridization, the probes were labelled using [a5S]dATP (NEN, Boston MA) and terminal deoxynucleotidyltransferase (Boehringer Man-

analysis of S R I H m R N A probe hybridization was confined to the periventricular nucleus at the level of the most anterior extension of the paraventricular nucleus. IGF-2 transcripts w e r e identified only in the choroid plexus 8'25. The rostral end of the paraventricular nucleus

nheim),

was used as a l a n d m a r k to o b t a i n lateral ventricle cho-

Image analysis

roid plexus sections at similar levels in e a c h animal. A t the sensitivity used for these experiments, GRH mRNA transcripts appeared to be entirely c o n f i n e d to the arcu-

Slides bearing hybridized sections were apposed to X-ray film (X-Omat. Kodak, Rochester, NY) for 3-7 days and the resulting autoradiographs analysed densitometrically using a Macintosh II computer equipped with an image capture board (Scion, Walkersville, MD 21793) running the programme 'Image' by Wayne Rasband, NIH, Bethesda, MD. The mean optical density of autoradiographs was measured in dpm/mg by comparison with simultaneously exposed 35S brain paste standards. As an increase in hybridization to the arcuate and periventricular nuclei resulted in an increase in both mean optical density and area of autoradiographic images, changes in GRF and SRIH transcripts were estimated by comparing the products of the image area and optical density. IGF-2 hybridization to the choroid plexus was estimated

GRF mRNA

300

SRIH mRNA

RESULTS

SRIH transcripts were widely distributed throughout brain as previously demonstrated 4. Quantitative

the rat

nucleus 5. Pituitary stalk transection and hypophysectomy inc r e a s e d a r c u a t e nucleus G R H m R N A (Fig. la,b) and ate

decreased choroid plexus IGF-2 (Fig. 2a,b) and periventricular S R I H m R N A transcripts (Fig. 3) Propylthiouracil-induced hypothyroidism increased GRH mRNA but had no significant effects on IGF-2 and SRIH m R (Fig. 4). T r e a t m e n t with triiodothyronine signifi-

NAs

300-1

4 10

0

~.......

IGF-2 mRNA

12oq

Control

T3

PTU

[ ] Sham Surgery

80 40

*

[~[] Stalk Transection Hypophysectomy

o Fig. 3. Changes in arcuate nucleus growth hormone-releasing factor, periventricular nucleus somatostatin and choroid plexus insulin-like growth factor2 (IGF-2) transcripts in response to pituitary stalk transection and hypophysectomy, mRNA hybridization in arbitrary units is shown as percent of control values. Error bars are S.E.M. * indicates significance of P < 0.05 and ** of P < 0.01 using Duncan's rank order test. n --> 8.

[ ] SRIH

[ ] GRH l IGF2 Fig. 4. Changes in brain growth hormone-releasing factor, somatostatin and insulin-like growth factor 2 transcripts in response to hypothyroidism and thyrotoxicosis induced with oral propylthiouracil and triiodothyronine treatment, respectively, mRNA hybridization in arbitrary units is shown as percent of control values. Error bars are S.E.M. ** indicates significance of P < 0.01 using Duncan's rank order test. n >_ 8.

300'

by Northern analysis is shown in Fig. 6.

0 ~-~

DISCUSSION

,< 8 Z ~ ~, 100"~ "-" r,~ [--, 0 Blank PTU

IGF-2 gene expression varies markedly between and within tissues at different stages in development and under different nutritional circumstances24. The pattern of transcription of the IGF-2 gene locus is itself complex 6 and the finding of high IGF-2 m R N A levels in a variety of human malignancies perhaps supports the claim that IGF-2, like IGF-1, is an important regulatory protein 1.

~~

T3

Fig. 5. The effects of oral propylthiouracil and triiodothyronine on pituitary TSHB transcripts, mRNA hybridization in arbitrary units is shown as percent of control values. Error bars are S.E.M. ** indicates significance of P < 0.01 using Duncan's rank order test. 6. n

=

cantly increased periventricular SRIH m R N A and decreased choroid plexus IGF-2 m R N A levels but had no effect on G R H m R N A (Fig. 4). The efficacy of oral treatment with propylthiouracil and triiodothyronine was confirmed by in situ analysis of pituitary TSHB transcripts using a probe complementary to rat TSHB m R N A (Fig. 5). An example of oligonucleotide probe validation 1

9.49

2

3

4

5

- -

746 -

4.40 --

2.37

-

-

i~: {,,~:~ 0.24 -.............. Fig. 6. RNA blot hybridized to the IGF-2 oligonucleotide 3' endlabelled with [32p]dCTP. Total RNA was derived from liver (lane 1), kidney (lane 2), hypothalamus and cerebral cortex (both tissue blocks containing choroid plexus) (lanes 3 and 4) and testis (lane 5).

In the placenta, where the IGF-2 and IGF-2 receptor genes are transcribed and translated, IGF-2 is thought tO act as a human embryonic growth factor 17 and in muscles induced to hypertrophy in hypophysectomized animals IGF-2 m R N A increases, indicating a local, pituitary-independent effect of IGF-2 on skeletal muscle growth. As far as we are aware, however, in the adult brain no specific function has yet been assigned to IGF-2, although recent work indicates that brain IGF-2 binding protein m R N A is greatly increased TM and IGF-2 m R N A moderately decreased 28 following hypophysectomy. We have found a relatively modest reduction in IGF-2 m R N A hybridization in response to pituitary stalk transection which reached significance after hypophysectomy (falling to 79% of control values). This confirms and extends recent findings obtained using Northern analysis 2s and implies that the IGF-2 gene is being regulated at least in part by either a pituitary hormone or a hormone released by an end organ. The minimal decrease in choroid plexus IGF-2 m R N A after treatment with triiodothyronine and the lack of a response to hypothyroidism makes it unlikely that thyroid hormones play a significant role in IGF-2 gene regulation in the adult although clearly an effect of thyroid hormones on IGF-2 gene expression in the foetus and neonate cannot be excluded, particularly in view of the major effect of these hormones on neuronal development. Concomitant with the decrease in choroid plexus IGF-2 m R N A hybridization in response to pituitary disconnection and hypophysectomy was a decrease in S R I H m R N A hybridization (confirming previous findings 31) and an increase in G R H m R N A levels. In each case the change in m R N A prevalence was quantitatively greater after hypophysectomy than pituitary stalk transection. It therefore seems likely that the reason for the quantitative difference observed is that hypophysectomy removes a subpopulation of somatotropes that secrete growth hormone at basal levels in the complete absence of G R H ~°. The change induced by pituitary stalk transection alone represents the cessation of G H secretion by the somatotropes that remain completely silent when deprived of G R H , along with a reduction in growth hor-

mone secretion by the remainder. The marked increase

tuitary stalk transection and hypophysectomy on IGF-2

in G R H m R N A in response to P T U - i n d u c e d hypothyroidism, in keeping with previous reports 9, suggests that

and SRIH transcripts are predominantly related to changes in the G H axis rather than thyroid h o r m o n e de-

in stalk-sectioned and hypophysectomized animals, hypothyroidism contributes to enhanced G R H transcrip-

ficiency, although hypothyroidism may contribute to the marked increase in G R H m R N A . Although the changes

tion.

we observed were modest, our results demonstrate that

In conclusion, we have used in situ hybridization histochemistry to demonstrate an increase in G R H and decrease in S R I H and IGF-2 m R N A hybridization in response to pituitary stalk transection and hypophysectomy. By treating rats with triiodothyronine and P T U , we have been able to demonstrate that the effects of pi-

REFERENCES 1 Briner, J., Hassam, S., Schonle, E.J. and Rohrer, U., Increased expression of IGF-2 in tumor tissue of nephroblastoma, Verh. Dtsch. Ges. Pathol., 73 (1989) 436-439. 2 Celia, S.G., De-Gennaro-Collona, V., Locatelli, V., Moiraghi, V., Loche, S., Wehrenberg, W.B. and Muller, E.E., Growth hormone (GH) autofeedback action in the neonatal rat: involvement of GH-releasing hormone and somatostatin, J. Endocrinol., 124 (1990) 199-205. 3 Enberg, G., Tham, A. and Sara, V.R., The influence of purifled somatomedins and insulin on foetal rat brain DNA synthesis in vitro, Acta. Physiol. Scand., 125 (1985) 305-308. 4 Fitzpatrick-McElligott, S., Card, J.P., Lewis, M.E. and Baldino, F.J., Neuronal localization of prosomatostatin mRNA in the rat brain with in situ hybridization histochemistry, J. Comp. Neurol., 273 (1988) 558-572. 5 Frohman, L.A., Downs, T.R., Chomczynski, P., Brar, A. and Kashio, Y., Regulation of growth hormone-releasing hormone gene expression and biosynthesis, Yale J. Biol. Med., 62 (1989) 427-433. 6 Gray, A., Tam, A.W., Dull, T.J., Hayflick, J., Pintar, J., Cavenee, W.K., Koufos, A. and Ullrich, A., Tissue-specific and developmentally regulated transcription of the insulin-like g r o w t h factor 2 gene, DNA, 6 (1987) 283-295. 7 Horvath, S. and Palkovits, M., Synaptic interconnections among growth hormone-releasing hormone (GHRH)-containing neurons in the arcuate nucleus of the rat hypothalamus, Neuroendocrinology, 48 (1988) 471-476. 8 Hynes, M.A., Brooks, P.J., Van Wyk, J.J. and Lund, P.K., Insulin-like growth factor II messenger ribonucleic acids are synthesized in the choroid plexus of the rat brain, Mol. Endocrinol., 2 (1988) 47-54. 9 Jones, P.M., Burrin, J.M., Ghatei, M.A., O'Halloran, D.J., Legon, S. and Bloom, S.R., The influence of thyroid hormone status on the hypothalamo-hypophyseal growth hormone axis, Endocrinology, 126 (1990) 1374-1379. 10 Kineman, R.D., Faught, W.J. and Frawley, L.S., Bovine pituitary cells exhibit a unique form of somatotrope secreting heterogeneity, Endocrinology, 127 (1990) 2229-2235. 11 Lambert, S., Vivario, J., Boniver, J. and Gol-Winkler, R., Abnormal expression and structural modification of the insulin-like growth-factor-II gene in human colorectal tumors, Int. J. Cancer, 46 (1990) 405-410. 12 Levy, A. and Lightman, S.L., Bromocriptine reduces rat thyrotrophin B-subunit mRNA stability, J. Endocrinol. Invest., 13 (1989) 49-54. 13 Lightman, S.L., Ninkovic, M., Hunt, S.P. and Iversen, L.L., Evidence for opiate receptors on pituicytes, Nature, 305 (1983) 235-237. 14 Margioris, A.N., Broekmann, G., Bohler, H.C.J., Grino, M., Vamvakopoulos, N. and Chrousos, G.P., Expression and local-

IGF-2 gene expression continues to be regulated by changes in G H status in the adult rat. Acknowledgements. We would like to thank Dr. Michael Brownstein and Dr. Harvey Whitfield for supplying the oligonucleotide probes. A.L. was supported by a Wellcome Trust Research Training Fellowship.

ization of growth hormone-releasing hormone messenger ribonucleic acid in rat placenta: in vitro secretion and regulation of its peptide product, Endocrinology, 126 (1990) 151-158. 15 Mayo, K.E., Cerelli, G.M., Rosenfeld, J.G. and Evans, R.M., Characterization of cDNA and genomic clones encoding the precursor to rat hypothalamic growth hormone-releasing hormone, Nature, 314 (1985) 464-467. 16 Nielsen, F.C. and Gammeltoft, S., Mannose-6-phosphate stimulates proliferation of neuronal precursor cells, FEBS Lett., 262 (1990) 142-144. 17 Ohlsson, R., Holmgren, L., Glaser, A., Szpecht, A. and Pfeifer-Ohlsson, S., Insulin-like growth factor 2 and short-range stimulatory loops in control of human placental growth, EMBO J., 8 (1989) 1993-1999. 18 Ooi, G.T., Orlowski, C.C., Brown, A.L., Becker, R.E., Unterman, T.G. and Rechler, M.M., Different tissue distribution and hormonal regulation of messenger RNAs encoding rat insulin-like growth factor-binding proteins-1 and -2, Mol. Endocrinol., 4 (1990) 321-328. 19 Rinderknecht, E. and Humbel, R.E., Primary structure of human insulin-like growth factor II, FEBS Letts., 89 (1978) 283286. 20 Rotwein, P., Burgess, S.K., Milbrandt, J.D. and Krause, J.E., Differential expression of insulin-like growth factor genes in rat central nervous system, Proc. Natl. Acad. Sci. USA, 85 (1988) 265-269. 21 Salmon, W.D.J. and Daughaday, W.H., A hormonally controlled serum factor which stimulates sulfate incorporation by cartilage in vitro, J. Lab. Clin. Med., 49 (1957) 82-85. 22 Soares, M.B., Ishii, D.N. and Efstratiadis, A., Developmental and tissue-specific expression of a family of transcripts related to rat insulin-like growth factor II mRNA, Nucl. Acid. Res., 13 (1985) 1119-1134. 23 Sonntag, W.E., Boyd, R.L. and Booze, R.M., Somatostatin gene expression in hypothalamus and cortex of aging male rats, Neurobiol. Aging, 11 (1990) 409-416. 24 Straus, D.S. and Takemoto, C.D., Effect of dietary protein deprivation on insulin-like growth factor (IGF)-I and -II, IGF binding protein-2, and serum albumin gene expression in rat, Endocrinology, 127 (1990) 1849-1860. 25 Stylianopoulou, F., Herbert, J., Soares, M.B. and Efstratiadis, A., Expression of the insulin-like growth factor II gene in the choroid plexus and the leptomeninges of the adult rat central nervous system, Proc. Natl. Acad. Sci. USA, 85 (1988) 141-145. 26 Tavianini, M.A., Hayes, T.E., Magazin, M.D., Minth, C.D. and Dixon, J.E., Isolation, characterization, and DNA sequence of the rat somatostatin gene, J. Biol. Chem., 259 (1984) 1179811803. 27 Ueno, T., Takahashi, K., Matsuguchi, T., Ikejiri, K., Endo, H. and Yamamoto, M., Multiple polyadenylation sites in a large 3'-most exon of the rat insulin-like growth factor II gene, Biochim. Biophys. Acta, 1009 (1989) 27-34.

28 Wood, T.L., Berlowitz, M., Gelato, M.C., Roberts, C.T.J., Leroith, D. and Millard, W.J., Hormonal regulation of rat hypothalamic neuropeptide mRNAs: effect of hypophysectomy and hormone replacement on growth-hormone-releasing factor, somatostatin and the insulin-like growth factors, Neuroendocrinology, 53 (1991) 298-305. 29 Young III, W.S.., In situ hybridization histochemistry. In A. Bj6rklund, T. H6kfelt, EG. Wouterlood and A.N. van den Pol (Eds.), Handbook of Chemical Neuroanatomy, Vol. 8, Analysis of Neuronal Microcircuits and Synaptic Interactions, Elsevier, New York, 1990, pp. 481-512. 30 Young III, W.S.., Bonnet, T.I. and Brann, M.R., Mesenceph-

alic dopamine neurons regulate the expression of neuropeptide mRNAs in the rat forebrain, Proc. Natl. Acad. Sci. USA, 83 (1986) 9827-9831. 31 Zeitler, P., Vician, L., Chowen-Breed, J.A., Argente, J., Tannenbaum, G.S., Clifton, D.K. and Steiner, R.A., Regulation of somatostatin and growth hormone-releasing hormone gene expression in the rat brain, Metabolism, 39 (1990) 46-49. 32 Chin, W.W., Muccini Jr., J.A. and Shin, L., Evidence for a single rat thyrotropin-fl-subunit gene: thyroidectomy increases its mRNA, Biochem. Biophys. Res. Commun., 128 (1985) 1152-1160.