¹Pediatric Oncology, Fred Hutchinson Cancer Research Center, 1100 Fairview Ave N, C1-169, Seattle, Washington 98105, USA

²Department of Biochemistry/Howard Hughes Medical Institute, University of Washington Medical Center, Box 357370, Seattle, Washington 98195, USA

³Department of Pathology, University of Washington Medical Center, Box 357470, Seattle, Washington 98195, USA

⁴Department of Medicine and Population Center for Research in Reproduction, University of Washington School of Medicine; and the Geriatric Research, Education and Clinical Center, VA Puget Sound Health Care System, Seattle, Washington 98108, USA

^aAuthors for correspondence

Multiple endocrine neoplasia type 2B (MEN2B) is an autosomal dominant syndrome characterized by the development of medullary thyroid carcinoma, pheochromocytomas, musculoskeletal anomalies and mucosal ganglioneuromas. MEN2B is caused by a specific mutation (Met918Thr) in the RET receptor tyrosine kinase. Different mutations of RET lead to other conditions including MEN2A, familial medullary thyroid carcinoma and intestinal aganglionosis (Hirschsprung disease). Transgenic mice were created using the dopamine -hydroxylase promoter to direct expression of RET^MEN2B in the developing sympathetic and enteric nervous systems and the adrenal medulla. DH-RET^MEN2B transgenic mice developed benign neuroglial tumors, histologically identical to human ganglioneuromas, in their sympathetic nervous systems and adrenal glands. The enteric nervous system was not affected. The neoplasms in DH-RET^MEN2B mice were similar to benign neuroglial tumors induced in transgenic mice by activated Ras expression under control of the same promoter. Levels of phoshorylated MAP kinase were not increased in the RET^MEN2B-induced neurolgial proliferations, suggesting that alternative pathways may play a role in the pathogenesis of these lesions. Transgenic mice with the highest levels of DH-RET^MEN2B expression, unexpectedly developed renal malformations analogous to those reported with loss of function mutations in the Ret gene.

ganglioneuroma; ret proto-oncogene; multiple endocrine neoplasia; renal agenesis; neuroblastoma; transgenic; autonomic nervous system

Introduction

Expression of the c-ret proto-oncogene (Ret) initiates in neural crest-derived neuronal precursors shortly after they migrate from the dorsal neural tube and persists in mature neurons of the peripheral nervous system (Pachnis et al., 1993; Tsuzuki et al., 1995). Mice lacking expression of either Ret or the co-ligand, glial cell line-derived neurotrophic factor (GDNF), are missing superior cervical and intestinal ganglion cells (Moore et al., 1996; Pichel et al., 1996; Schuchardt et al., 1994). Similarly, loss-of-function mutations of the RET gene are associated with intestinal aganglionosis in humans (Edery et al., 1994; Romeo et al., 1994). However, the precise role of RET in the migration, proliferation, survival and differentiation of neuroblasts is unclear.

Gain-of-function mutations in RET are responsible for both MEN2A and MEN2B, as well as familial medullary carcinoma of the thyroid (reviewed in Edery et al., 1997). In addition, rearrangements of the RET gene are found in many papillary thyroid carcinomas (Jhiang et al., 1992; Santoro et al., 1992; Wajjwlku et al., 1992). The MEN2B syndrome is characterized by medullary thyroid carcinoma, pheochromocytomas, musculoskeletal anomalies, sensory deficits and mucosal ganglioneuromas. More than 95% of individuals with MEN2B possess a single point mutation at codon 918 of the RET gene resulting in a Thr substitution for Met (Met918Thr) (Carlson et al., 1993; Eng et al., 1994; Hofstra et al., 1994). Studies in vitro indicate that the RET^MEN2B gene product does not constitutively dimerize (unlike the RET^MEN2A receptor), may be active as a monomer, responds to ligand-stimulation and phosphorylates different substrates than wild-type RET or RET^MEN2A (Borrello et al., 1995; Iwashita et al., 1996; Rizzo et al., 1996; Songyang et al., 1995). How these molecular properties underlie the cellular events that culminate in the MEN2B phenoytpe remains to be discovered.

We have developed a transgenic mouse model to further understand the role of RET in the development of the peripheral nervous system and the pathogenic effects of the MEN2B mutation. These mice possess a transgene that utilizes the human dopamine -hydroxylase (DH) promoter (Kapur et al., 1991; Mercer et al., 1991) to target expression of RET^MEN2B to sympathetic neurons, enteric neurons, adrenal chromaffin cells and their precursors in transgenic mice.

Results

Production and viability of DH-RET^MEN2B mice

An initial group of seven DH-RET^MEN2B transgenic mice, possessing 8 - 260 copies of the transgene (Figure 1), were produced by zygotic micronuclear injection (Table 1). Two of these `founder' mice died at 12 weeks and 18 weeks of life of unknown causes, another developed prominent vaginal prolapse (occasionally observed in non-transgenic mice as well) and was euthanized at 13 weeks and a fourth never transmitted the transgene. The three remaining founders displayed no external phenotypic abnormalities and were successfully mated to produce three lines, designated `low' (CH7220), `intermediate' (CH7194) and `high' (CH7214) in reference to the relative number of transgenic copies present in each, as determined by Southern blot analysis. Transgenic mice from the low copy line (four copies) were viable, fertile and bred to homozygosity. In contrast, only 26% (n=98) of transgenic pups from the intermediate line (approximately 60 copies) survived to weaning. The remainder appeared normal at birth, but either died in the first 24 h or developed as runts that died sometime later during the first 3 weeks of life. The founder for the high copy line was 8% mosaic for a single transgene insertion site that contained approximately 260 copies of the transgene. Only one of her transgenic offspring, a male, survived to weaning. He went on to sire many litters from which less than 5% of the transgenic pups survived more than 24 h and none survived 3 weeks.

Neuroglial hyperplasia in sympathoadrenal tissues of DH-RET^MEN2B transgenic mice

All of the transgenic lines displayed hyperplasia of sympathetic ganglia, although there was considerable variability, even within a line. Sympathetic hyperplasia was readily visualized by X-gal staining of whole mounts of tissue from DH-RET^MEN2B transgenic mice that also carried a DH-nlacZ transgene (Figure 2). The latter provided a histochemical marker for nuclei of catecholaminergic cells including sympathetic and enteric neurons (Kapur et al., 1991, 1992; Mercer et al., 1991). Neuroglial hyperplasia was most apparent in the preaortic sympathetic complex medial to the kidneys. Several interconnected ganglia, the largest of which is the celiac ganglion, normally exist in this region. In transgenic mice, enlargement of the entire complex formed a prominent midline mass which extended superiorly into the adrenal medulla on each side of the aorta. Similar changes distorted the gross anatomy of the paravertebral sympathetic chain such that the typical discrete ganglia were enlarged and often fused into contiguous cords of neuronal cell bodies. Hyperplasia of the paravertebral sympathetic chain extended superiorly into the thorax, but spared the superior cervical ganglia. In general, neuroglial overgrowth correlated with transgene copy number, with the most pronounced changes in mice derived from the high and intermediate lines.

Histologically, the accumulation of sympathoadrenal neuroglial tissue in DH-RET^MEN2B represented an increase in neurons, Schwann cells and nerve fibers. In some mice, the lesions were highly cellular with little or no mature neural stroma and relatively small neurons that varied in size and cytoplasmic basophilia (Figure 3b). Microscopically, the cellular lesions mimicked a normal phase of ganglion cell development normally restricted to the perinatal period, but which appeared to be prolonged into young adulthood in transgenic mice. In other transgenic mice, the sympathoadrenal proliferations resembled mature human ganglioneuromas. Neuronal cell bodies increased in diameter, overall cellularity decreased and the neuropil assumed a paucinuclear morphology comparable to peripheral nerve (Figure 3c). At all stages, neurons demonstrated strong RET, tyrosine hydroxylase (not shown) and peripherin immunoreactivity identical to that of normal sympathetic ganglion cells (Figures 4 and 5). Enlarged fiber tracts exited the preaortic region and distributed to the mesentery and gonads (not shown). Despite the resemblance of these tumors to different maturational stages of neuroblastoma, neither metastasis nor overt malignant neoplasms (i.e. neuroblastoma, pheochromocytoma) were observed.

The adrenal glands were enlarged and their cortices disrupted. The normal adrenal medulla consists primarily of a uniform collection of chromaffin cells arranged in cords and separated by thin-walled vessels (Figure 3a). In DH-RET^MEN2B mice, normal chromaffin cells were preserved histologically (Figure 3b and c) and their presence was confirmed by immunostaining with NPY and PNMT (not shown). However, abundant neuroglial tissue existed primarily in central portions of the adrenal medullae of transgenic mice (Figure 3b and c). The neurons displayed immuno-reactivity to Ret and peripherin (Figure 4a and c) and, in the most severely affected animals, represented massive contiguous extensions of the preaortic ganglia through the inferior medial pole of the adrenal cortex. In contrast, no neuronal cell bodies were evident histologically or immunohistochemically in control adrenal glands (Figure 3b), although peripherin-positive nerve fibers were evident in the interstices of chromaffin cell nests (Figure 4d).

The enteric nervous system was unaffected in DH-RET^MEN2B mice. In whole mount intestinal segments examined by acetylcholinesterase staining, enlarged extrinsic nerves were present in the intestinal mesentery, but no difference in the pattern of the neuronal fiber network within the colon was apparent (Figure 6). Similarly, no alterations were appreciated in histological sections of gut (not shown). The number of enteric neurons in the gastrointestinal tracts of DH-RET^MEN2B mice were normal, as revealed by prenatal estimates of X-gal positive nuclei in mice co-expressing the DH-nlacZ transgene and postnatal counts of cuprylinic blue-positive cell bodies. No significant differences in neuronal density were evident between transgenic mice (285±58 ganglion cells per standardized field) and their non-transgenic littermates (324±30 ganglion cells per standardized field).

Renal agenesis/dysplasia occurs in two of the DH-RET^MEN2B transgenic lines

Most of the transgenic offspring from the intermediate and high copy lines died in the first day of life or became runted and died before weaning. Autopsies of these pups revealed renal abnormalities. The anomalies, which were often asymmetric, ranged from small dysplastic kidneys to renal agenesis. In most cases, ureters were present, but in 3% of all cases no ureter was found. Occasionally, unilateral renal agenesis was seen in conjunction with a contralateral kidney of normal size and histology. The cause of death in such mice was unclear, but renal failure is suspected. Histological analysis of the hypoplastic kidneys revealed that most were dysplastic with microcyst formation, tubular atrophy, and few or absent glomeruli (Figure 7). More detailed studies of the renal malformations are in progress and will be reported separately. Besides the peripheral nervous system, adrenal glands, and the kidneys there were no gross or microscopic abnormalities evident in any of the other major organs examined including the brain, heart, thymus, lungs, liver, spleen, and gastrointestinal tract.

Phenotypic variability correlates with RET over-expression

Variability in the amount of neuroglial hyperplasia and renal malformations was observed between and within transgenic lines. In general, interline variability correlated with copy number such that renal malformations and the most dramatic neuroglial hyperplasia were observed in the intermediate and high copy lines. However, even in these lines some animals had relatively mild neuroglial overgrowth and/or normal kidneys. Analysis of RET expression in preaortic neuroglial tissue from individual newborn transgenic animals was performed to determine whether similar variability existed in Ret protein levels. Protein extracts were analysed for RET expression by Western blots probe with an antibody that did not distinguish between endogenous Ret and Ret^MEN2B. The greatest amount of RET protein was found in pups from the intermediate and high copy lines (Figure 8). However, variability in RET expression was present within litters. In general, those pups with the greatest amounts of RET receptor also exhibited the most dramatic neuroglial hyperplasia and most severe renal malformations (i.e. bilateral renal hypoplasia/dysgenesis). In part, the differences may refect genetic variability, since a mixed DBA´C57B1/6J background was present in all of the lines.

MAPK is not activated in sites of neuroglial hyperplasia

To evaluate whether expression of RET^MEN2B is associated with activation of mitogen-activated protein kinase (MAPK), total protein was isolated from pooled sympathetic ganglia, adrenals and segments of large intestine from newborn and adult transgenic mice and their wild-type siblings. Dramatic over-expression of Ret protein was evident in the adrenals and sympathetic ganglia of newborn transgenic pups (Figure 9, upper panel). A small increase in Ret was also detected in the gut of newborn mice. However, Ret over-expression was transient and did not persist into adulthood in any of these sites (data not shown). Total MAPK and activated MAPK were assessed in the same tissue samples with antibodies that recognized all forms of MAPK and dually phosphorylated MAPK, respectively. The analysis was performed on one blot, probed first for activated MAPK (Figure 9, middle panel), stripped of bound antibody and re-incubated with antisera recognizing total MAPK (Figure 9, lower panel). The study failed to demonstrate a consistent correlation between Ret over-expression and activation of MAPK. In fact, newborn sympathetic and adrenal tissues with the greatest amounts of Ret protein showed little-or-no activation of MAPK. The only tissue to show a slight increase in activated MAPK was colon from transgenic animals, in which RET expression was minimally elevated.

Sympathetic innervation of target organs in DH-RET^MEN2B mice

To determine whether the sympathetic hyperplasia seen in these mice was associated with an increase in the innervation of target tissues, we analysed the catecholamine content of plasma (not shown), adrenal glands and several tissues (submaxillary gland, brown fat, colon, spleen and heart) that receive substantial sympathetic innervation (Figure 10). Norepinephrine content was reduced in submaxillary gland and increased in colon, but unchanged in other organs from transgenic mice.

Discussion

Neuroglial tumors of the sympathoadrenal axis are the predominant finding in DH-RET^MEN2B transgenic mice

The RET receptor tyrosine kinase is widely expressed during development and adult life, but is found at highest levels in the developing nervous system and kidneys (Pachnis et al., 1993; Tsuzuki et al., 1995). Although the cellular effects mediated by RET-mediated signals during development and adult life are not clear, RET is essential for the development of the enteric nervous system and kidneys (Durbec et al., 1996; Schuchardt et al., 1994). Activating mutations in RET, as found in human patients with MEN2A and MEN2B, suggest that RET may influence the proliferation, survival and/or differentiation of certain neural crest-derivatives such as thyroid C-cells, adrenal chromaffin cells and neurons of the peripheral nervous system (Califano et al., 1996; Edery et al., 1997; Kusafuka and Puri, 1997). In addition, RET may influence neurite extension and perhaps innervation of target organs given the ganglioneuromas and muscular weakness that characterize patients with MEN2B.

To study the potential role of RET during development we created transgenic mice that express the RET^MEN2B cDNA under control of the human DH promoter. This promoter has been used to express other cDNAs and consistently targets expression to sympathetic ganglion cells, adrenal chromaffin cells, enteric neurons and their neural crest-derived precursors. In DH-RET^MEN2B newborn pups, RET protein levels were increased in all of these sites, and the consistent phenotype observed in three transgenic lines was neuroglial hyperplasia of sympathetic ganglia and the adrenal medulla. Although the severity of the phenotype varied, a general correlation existed between transgene dosage, RET protein levels in newborns and the amount of neuroglial overgrowth, suggesting that over-expression of RET^MEN2B promoted benign neuroglial tumor development. This indicates that RET^MEN2B is capable of triggering the proliferation of neural precursors in vivo. As judged by the presence of neurons within the adrenal medulla, expression of RET^MEN2B may also be capable of influencing the cell fate of sympathoadrenal precursors. It is, however, possible that the presence of these neurons is related to the survival of neurons otherwise slated for apoptosis, or perhaps the migration of neuronal precursors into the developing adrenal gland.

The anatomical distribution and histological features of the neuroglial proliferation in DH-RET^MEN2B transgenic mice most closely resemble human ganglioneuromas, benign neoplasms which often arise from the maturation of undifferentiated neuroblastomas. Neuroblastoma is the most common type of extracranial solid pediatric neoplasm. The majority of primary tumors are diagnosed before 4 years of age as primary lesions in the sympathoadrenal axis, frequently in the adrenal medulla (Brodeur and Castleberry, 1997). A subset of neuroblastomas matures spontaneously or in response to chemotherapy, even after distant metastasis. The final phases of the maturation process are represented by benign ganglioneuromas indistinguishable from the tumors found in newborn and adult DH-RET^MEN2B mice. Histologically, the transition from neuroblastoma to ganglioneuroma is identical to prenatal events that underlie the formation of normal sympathetic ganglia from neural crest-derived neuroblasts. In some humans and DH-RET^MEN2B mice, the neuroblast stage appears to be prolonged leading to an overgrowth of neuroglial tissue, extension into the adrenal medulla, and delay of maturation until the postnatal period.

RET is present normally in enteric and sympathetic neuroblasts where it is thought to influence their proliferation, survival and/or differentiation (Edery et al., 1997; Pachnis et al., 1993; Tsuzuki et al., 1995). Although RET^MEN2B appears to have increased activity upon ligand binding, it also has significant activity in the monomeric state and probably interacts with different substrates than endogenous RET (Borrello et al., 1995; Santoro et al., 1995; Songyang et al., 1995). Therefore, expression of RET^MEN2B under the control of the DH promoter probably increases baseline activity, and potentially alters targets, for the RET tyrosine kinase in these neuroblast populations. The net result appears to be expansion of the neuroblast population and delayed differentiation. Similar dual influences of RET^MEN2B on cell proliferation and differentiation have been observed in PC12 cells in vitro. Transfection of the latter with RET^MEN2B induces an `aberrant' differentiation program characterized by a flat morphology and expression of a variety of neural markers, with continued proliferation and an inability to form neurites in response to NGF (Califano et al., 1996). However, a different effect has been observed in neuroblastoma cell lines in which up-regulation of RET transcripts precedes, and activated forms of RET promote, neural differentiation (D'Alessio et al., 1995; Ikeda et al., 1990; Ikuno et al., 1995). Thus, specific responses of a given cell type to DH-RET^MEN2B may depend on other variables, emphasizing the value of a transgenic approach to study the function of activated forms of RET in vivo.

RET^MEN2B expression did not affect terminal innervation or induce intestinal ganglioneuromas or pheochromocytomas

Despite the abundant neurites and supportive glia elaborated by these masses and the increased size of nerve fibers extending to some target organs such as the intestinal mesentery, ovaries and kidneys, the amount of terminal innervation as evidence by catecholamine content was not significantly affected. The normal terminal innervation densities in DH-RET^MEN2B and DH-Ras mice presumably reflect end-organ regulation imposed by a finite number of cellular targets and limiting amounts of neurotrophins produced by these organs.

The phenotype of MEN2B includes adrenal pheochromocytomas and mucosal ganglioneuromas, but neither neuroblastomas nor ganglioneuromas of the sympathoadrenal axis. Expression of RET^MEN2B under control of the DH promoter failed to induce pheochromocytomas or mucosal ganglioneuromas. Mucosal ganglioneuromas are masses of neurons, Schwann cells and nerve fibers formed in the lips and throughout the gastrointestinal tract. The nodules can be seen in newborns but usually become more prominent with age. We anticipated the formation of these lesions in DH-RET^MEN2B mice because the DH promoter has been shown to direct the expression of transgenes to enteric neural precursors (Kapur et al., 1992). Instead, no differences in neuronal density, ganglion histology, or plexus architecture were evident in DH-RET^MEN2B intestines. Relative resistance of the murine enteric nervous system to RET^MEN2B-induced changes correlates with similar resistance observed in DH-Ras mice, despite evidence for expression of both transgenes in neonatal gut. In the gut, this apparent resistance may reflect, at least in part, a relatively low level of promoter activity in the gut that parallels low endogenous DH expression (Sweetser et al., 1997).

Pheochromocytomas in patients with MEN2B are neoplasms that usually arise in the adrenal medulla due to presumed oncogenic transformation in cells of the sympathoadrenal lineage. Pheochromocytomas share many of histological and biochemical properties of chromaffin cells and do not exhibit the extensive neural differentiation found in ganglioneuromas. The presence of ganglioneuromas in adrenal glands from DH-RET^MEN2B mice indicates an effect of the transgene on the sympathoadrenal lineage with a different outcome than expected based on the human syndrome. Possible explanations for this discrepancy include differences in the timing and levels of RET^MEN2B induced by the DH promoter and/or differences in the molecular and cellular programs that regulate the behavior of murine and human development. The latter phenomenon is difficult to anticipate, but has been observed in attempts to produce other murine models for hereditary human disorders, even when endogenous genes are mutated by homologous recombination (Brannan et al., 1994; Forss-Petter et al., 1997; Snouwaert et al., 1992). In contrast, expression of RET^MEN2A under control of the calcitonin gene related peptide promoter led to thyroid tumors `morphologically and biochemically similar to human medullary thyroid carcinoma' (Michiels et al., 1997).

One variable which was not anticipated in design of our transgene is the recent discovery that the long and short RET isoforms have different biological effects (Rossel et al., 1997). Two RET isoforms have been described which differ in the length of the carboxy terminus, by nine versus 51 amino acids (Tahira et al., 1990). PC12 cells transfected with a cDNA encoding the MEN2B-mutation in the long isoform respond with more prominent neurite outgrowth than cells transfected with a similar construct encoding the short isoform. As our transgene encoded only the short isoform, it is possible that some of the human MEN2B features are only induced by the longer isoform. In any case, the production of ganglioneuromas in DH-RET^MEN2B mice provides as an opportunity to investigate the molecular and cellular events underlying the pathogenesis of these tumors.

The RET^MEN2B receptor tyrosine kinase appears to influence cell behavior without activating MAPK

Sympathetic neurons, enteric neurons and adrenal chromaffin cells are believed to develop from a multipotent precursor whose fate is determined by a balance of microenvironmental cues, including glucocorticoids, neurotrophins and possibly ligands for RET (Anderson, 1993; Carnahan et al., 1991). Sympatho-adrenal precursors migrate into the developing adrenal gland and find themselves surrounded by glucocorticoids that repress expression of trkA, the NGF receptor and promote the differentiation of medullary chromaffin cells. As a result, only rare neurons are found in the normal adrenal medulla. By contrast, sympathoadrenal precursors exposed to nerve growth factor in glucocorticoid-poor environments develop neuronal characteristics in vivo and in vitro.

The endogenous signal for neuronal differentiation of sympathoadrenal precursors is believed to proceed through intracellular activation of the Ras/MAPK cascade and can be mimicked in vitro by various effectors of MAPK activation (Birren et al., 1993; Buchman and Davies, 1993; D'Arcangelo and Halegoua, 1993; Wood et al., 1992). We have previously shown that expression of activated Ras in transgenic mice under control of the DH promoter led to sympathoadrenal neuroglial hyperplasia and in some instances neuroblastoma. Here we show that the DH-RET^MEN2B transgene induces a phenotype indistinguishable from the benign lesions found in DH-Ras mice. This action apparently occurs independently of MAPK activation. Studies in vitro suggest a role for MAPK activation in determination of sympathoadrenal precursor cell fate. RET stimulation is associated with MAPK activation in neuroblastoma cells (Worby et al., 1996). However, RET-mediated signals can use alternative pathways involving Ras, but not Raf or MAPK activation (Santoro et al., 1994). The nature of the alternative intracellular signals remain to be clarified, but phosphatidyl-inositol 3-kinase and/or phosphorylation of the cytoplasmic tyrosine kinase, Snt, may be an important components (Rizzo et al., 1996; Santoro et al., 1994; van Weering and Bos, 1997). The induction of ganglioneuromas in DH-RET^MEN2B transgenic mice, independent of MAPK kinase activation, supports the existence of such alternative pathways for RET^MEN2B-signal transduction.

In contrast with DH-Ras transgenic mice, malignant neuroblastoma was never observed in DH-RET^MEN2B animals. The difference may, in part, reflect discordant pathways regulated by activated Ras and RET^MEN2B, or possibly differences in the `strength' of signals initiated by the two constructs. This could be related to differences in transgene expression with advancing age between these two lines of transgenic mice. As compared with newborn DH-RET<sup>MEN2B</sup> transgenic mice, adults have a markedly lower level of RET<sup>MEN2B</sup> as measured by Western analysis and verified by Northern analysis (results not shown). This contrasts with the age-related increase in transgene expression and Ras protein levels observed in some DH-Ras mice (Sweetser et al., 1997). This suggests that chronic Ras activation may be necessary for the malignant transformation, presumably via a second `hit' or mutation, to occur.

Impaired renal development two lines of DH-RET^MEN2B mice

Renal agenesis or hypodysplasia were unexpected findings in DH-RET^MEN2B mice from the intermediate and high copy lines. During development RET is expressed by epithelial cells of the mesonephric duct and ureteric bud (Pachnis et al., 1993; Tsuzuki et al., 1995). GDNF, produced by the metanephric blastema, promotes formation and branching of the ureteric bud (Pichel et al., 1996; Vega et al., 1996). Through a series of reciprocal inductive interactions, the branches of the ureteric bud eventually form the collecting system of the kidney, while glomeruli and proximal segments of nephrons are derived from the metanephric blastema (Lechner and Dressler, 1997). Disruption of RET expression through the use of specific antisense-oligodeoxynucleotide using an in vivo organ culture system affected the expression pattern of proteoglycans at the epithelial : mesenchymal interface and inhibited metanephric growth (Liu et al., 1996). In mice lacking RET or GDNF, the ureteric bud fails to develop or develops aberrantly, so that interactions with the metanephric blastema are impaired and the metanephric blastema undergoes apoptosis (Schuchardt et al., 1996). Thirty percent of mice heterozygous for null mutations in either Ret or Gdnf develop renal hypoplasia or aplasia, indicating that levels of Ret and Gdnf expression may be limiting for renal development.

The similarity between the renal anomalies in DH-RET^MEN2B mice and animals with loss-of-function mutations in the Ret or Gdnf genes suggest an effect of the transgene on RET signal transduction in the ureteric bud. It is conceivable that insertion of the DH-RET^MEN2B transgene may have disrupted a gene crucial for renal development, however, this is unlikely given two independent affected lines and a transgene comprised primarily of human cDNA unsuitable for homologous recombination with the murine Ret gene. Moreover, the phenotype correlated with the highest levels of the transgene expression and with the greatest neuroglial overgrowth. Interference with renal development could be an indirect consequence of neuroglial overgrowth. The abundant neuroglial tissue might compress renal vessels, mechanically interfere with ureteric bud/metanephric blastema interactions, or produce factors which compromise nephrogenesis. Expression of RET^MEN2B within, or adjacent to, the developing metanephric blastema or ureters could titrate out the amount of gdnf and/or neurturin (both co-ligands for Ret), and thereby mimic the gdnf-null phenotype. Although no X-gal staining is evident within the kidneys of DH-nlacZ transgenic embryos utilizing the same promoter sequences, there is significant hyperplasia of the paravertebral sympathetic ganglia in transgenic embryos just adjacent to the developing ureteric bud and metanephric blastema (not shown). Further investigation is required to determine how the DH-RET^MEN2B transgenic results in renal malformations. Although humans with MEN2B have not been reported to have renal malformations, it is possible that interference with normal Ret signaling in an analogous fashion could underlie some aspects of the developmental abnormalities seen in MEN2B.

Materials and methods

Production of DH-RET^MEN2B transgenic mice

The RET^MEN2B mutation (Met918Thr) was introduced by PCR into a cDNA encoding the short isoform (1072 amino acids) of human RET (Takahashi et al., 1988, 1989). The mutation and absence of polymerase errors were verified by sequencing. The transforming ability of the mutation was demonstrated in NIH3T3 cells by foci formation and growth in soft agar using the pSV₂neo expression vector as previously described (results not shown) (Takahashi et al., 1988). The RET^MEN2B cDNA was inserted downstream of the human DH promoter (5.8 kb) using the vector described previously (Figure 1) (Mercer et al., 1991). Between the transcriptional start site and the RET^MEN2B cDNA, the construct contains the first intron of the rat preproinsulin gene (+13 - +183 nt). A polyadenylation site was provided from the murine protamine 1 gene. The 10.8 kb transgene was injected into the male pronuclei of fertilized DBA´C57B1/6 F₁ eggs and seven founder mice which carried the transgene were identified by dot hybridization assay. Transgene copy number and number of insertion sites was determined by dot and Southern blots of transgenic mouse DNA using probes to both the DH promoter (a labeled 4.3 kb HindIII fragment) and the 5' half of human RET (a labeled 1.6 kb HindII/NdeI fragment), normalized to human DNA. The probes did not hybridize to non-transgenic mouse DNA under the conditions used (not shown). To normalize DNA loading blots were also probed to an 800 bp StuI/BglII fragment from the mouse metallothionein I promoter (Searle et al., 1984). Independent lines were established and maintained from three founders through crosses with C57B1/6J mice.

Immunohistochemistry

Tissues were fixed in 10% neutral buffered formalin (NBF) and embedded in paraffin. Five-micron sections were used for hematoxylin-and-eosin staining or indirect immunocytochemistry. Antibodies were used with the following specificities: Ras (pan-Ras(Ab-3) Oncogene Science, Uniondale, NY, USA; 1 : 1000), tyrosine hydroxylase (TE101 Eugene Tech International Inc., Ridgefield Park, NJ, USA; 1 : 1000), the short isoform of RET (C-19 Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA; 1 : 1000), neuropeptide Y (NPY, Peninsula Laboratories, Belmont CA, USA; 1 : 1000) and PNMT (Eugene Tech International Inc.; 1 : 500). Specimens for RET immunostaining were subjected to microwave pretreatment for 12 min while submerged in 10 mM citrate pH 6.0 for 12 min and cooled at room temperature for 20 min. Incubations with primary antisera were performed overnight at 4°C. Bound antibody was visualized after incubation with biotinylated goat anti-rabbit or anti-mouse antisera (Vector Laboratories Inc.; 1 : 250), followed by horseradish peroxidase-conjugated streptavidin (Zymed Laboratories Inc., So. San Francisco, CA, USA; 1 : 250) and reaction with 0.05% diaminobenzidine (Sigma, St. Louis, MO, USA) in phosphate buffered saline (PBS).

Acetylcholinesterase staining of the myenteric plexus

To examine the architecture of the myenteric plexus, 1-cm sections from the mid-colon were flushed with PBS, fixed in NBF for 90 min at 4°C and rinsed in 0.1 M phosphate buffer pH 7.4 three times for 30 min each. Intestinal segments were then stained for acetylcholinesterase as described (Baljet and Drukker, 1975). To better visualize the distribution and density of sympathetic and enteric neurons, DH-RET^MEN2B mice were bred to DH-nlacZ transgenic mice (Mercer et al., 1991). The double-transgenic pups were sacrificed at 2 weeks of age, dissected to remove unwanted visceral organs, and fixed for 1 h in 0.1 M phosphate buffer pH 7.3 containing 0.2% glutaraldehyde, 2% formaldehyde, 5 mM EGTA and 2 mM MgCl₂. They were then rinsed three times for 30-min in 0.1 M phosphate buffer pH 7.3, 2 mM MgCl₂, 0.1% sodium deoxycholate, and 0.2% NP40. The specimens were stained overnight at room temperature with 2.5 mM X-gal, 5 mM potassium ferricyanide and 5 mM potassium ferrocyanide in rinse buffer.

Counts of enteric ganglion cells in whole mount preparations stained with cuprolinic blue staining

Cuprolinic blue staining was performed using a modification of a described procedure (Heinicke et al., 1987). Segments of large and small intestine were removed, flushed with PBS, and fixed by immersion in modified Carnoy's fixative (60 ml ethanol, 30 ml chloroform, 5 ml acetic acid) at room temperature. The fixed tissues were stored in 70% ethanol at 4°C for up to 2 weeks and then opened longitudinally along the mesenteric border and pinned out in staining buffer (1 M MgCl₂ in 0.025 M sodium acetate, final pH=5.6). The mucosa and submucosa were removed with forceps. After 15 - 30 min, the tissue was incubated in freshly filtered 0.3% cuprolinic blue (quinolinic phthalocyanine; Polysciences Inc., Warrington, PA, USA) in staining buffer with gentle agitation. Finally, the segments were rinsed and differentiated for 30 min in staining buffer at 37°C. After dehydration, the muscular preparations were mounted, `serosa-up', in glycerol on glass slides. Neurons, which are grouped in `columns' that parallel the circular muscle layer, were identified based on their size and deep blue cytoplasm. Counts were performed with a compound microscope on fields four ganglion cell columns wide using a 20´ objective and a reticle that standardized the field height.

Western blots

Harvested tissues, frozen on dry ice and stored at -70°C, were homogenized on ice in lysis buffer, (10 mM sodium phosphate (pH 7), 100 mM NaCl, 1% Triton X-100, 5 mM EDTA, 1 mM sodium vanadate, 2 mM phenylmethylsulfonyl fluoride, 10 g/ml each of aprotinin, leupeptin and pepstatin). Lysates were centrifuged at 15 000 g for 15 min at 4°C. Ret protein was analysed in supernatants by electrophoresis in 8% polyacrylamide-SDS gels, transfer to nitrocellulose and blot incubation with Ret antiserum (C-19, Santa Cruz Biotechnology Inc., Santa Cruz, CA, USA; 1 : 1000) followed by horseradish peroxidase linked anti-rabbit IgG (NA9340, Amersham International, Bucking-hamshire, UK; 1 : 2000). Immunoreactive bands were visualized by chemiluminescence (Renaissance^Ò system, DuPont NEN, Boston, MA, USA). For analysis of MAPK levels supernatants were resolved on 10% polyacrylamide-SDS gels, transferred to PVDF membranes (Millipore Corp, Bedford MA, USA). The membranes were probed with anti-active MAPK Ab (Promega Corp, Madison, WI, USA; 25 ng/ml) and visualized by chemiluminescence (Phototope^TM-Star Western Blot Detection Kit, New England Biolabs, Inc, Beverly MA, USA). To detect RET protein in the same samples, membranes were stripped by agitating in 62.5 mM Tris pH 6.8, 100 mM -mercaptoethanol and 2% sodium dodecylsulfate for 1 h at 55°C, then treated with Ret antisera as above. For detection of total MAPK the same membrane was re-stripped and incubated with anti-MAPK antibody (# 7884, a gift from Jean Campbell; 1 : 10 000) and visualization by chemiluminescence.

Catecholamine assays

Tissue samples were frozen on dry ice and stored at -70°C prior to analysis. Samples were sonicated for 15 - 30 s (Branson Sonifer 450, microtip, constant duty, output 5.5) in 0.4 to 1.0 ml of 0.1 M perchloric acid, 0.1% cysteine buffer with 10 pg/l or 100 pg/l dihydroxybenzylamine (DHBA). Sonicates were centrifuged for 15 - 45 min at 18 700 g. A 200 l aliquot of supernatant was added to 20 mg of acid-washed alumina and 800 l of 0.5 M Tris, 2% EDTA buffer pH 8.0 and rotated overnight at 4°C. The alumina was washed twice with 1 ml of cold distilled water and catecholamines were extracted with 200 l 0.1 M perchloric acid, 0.1% cysteine buffer. Samples were processed using a Beckman Ultrasphere ODS 5 4.6 mm´25 cm column. The mobile phase was 0.1 M citric acid, 0.1 M sodium phosphate dibasic, 0.1 mM disodium EDTA, 0.012% 1-octanesulfonic acid and 7% methanol at 1 ml/min. The working potential of the electrochemical detector (BAS LC-4C) was +0.8 V and the full-scale sensitivity was 5 or 10 nA. The samples were quantified using an HP 3393A integrator with DHBA as the internal standard.

Acknowledgements

DAS is supported by NIH training grant in Pediatric Hematology and Oncology (T32CA09351). RPK is supported by NIH RO1DK52530. AMM received support from VA Medical Research funds and NIH grant HD12629. We thank Masahide Takahashi for the kind gift of a human c-ret cDNA clone.

Anderson DJ. (1993). Annu. Rev. Neurosci. 16, 129-158. MEDLINE

Baljet B and Drukker J. (1975). Stain Technol. 50, 31-36. MEDLINE

Birren SJ, Lo L and Anderson DJ. (1993). Development 119, 597-610. MEDLINE

Borrello MG, Smith DP, Pasini B, Bongarzone I, Greco A, Lorenzo MJ, Arighi E, Miranda C, Eng C, Alberti L, Bocciardi R, Mondellini P, Scopsi L, Romeo G, Ponder BAJ and Pierotti MA. (1995). Oncogene 11, 2419-2427. MEDLINE

Brannan CI, Perkins AS, Vogel KS, Ratner N, Nordlund ML, Reid SW, Buchberg AM, Jenkins NA, Parada LF and Copeland NG. (1994). Genes Dev. 8, 1019-1029. MEDLINE

Brodeur GM and Castleberry RP. (1997). Principles and Practice of Pediatric Oncology. Pizzo PA and Poplock DG (eds). Lippincott-Raven Publ.: Philadelphia.,

Buchman VL and Davies AM. (1993). Development 118, 989-1001. MEDLINE

Califano D, D'Alessio A, Colucci-D'Amato GL, De Vita G, Monaco C, Santelli G, Di Fiore PP, Vecchio G, Fusco A, Santoro M and de Franciscis V. (1996). Proc. Natl. Acad. Sci. USA 93, 7933-7937. MEDLINE

Carlson KM, Dou S, Chi D, Scavarda N, Toshima K, Jackson CE, Wells SAJ, Goodfellow PJ and Donis KH. (1993). Development 119, 1005-1017. MEDLINE

Carnahan JF, Anderson DJ and Patterson PH. (1991). Dev. Biol. 148, 552-561. MEDLINE

D'Alessio A, De Vita G, Cali G, Nitsch L, Fusco A, Vecchio G, Santelli G, Santoro M and de Franciscis V. (1995). Cell Growth Diff. 6, 1387-1394. MEDLINE

D'Arcangelo G and Halegoua S. (1993). Mol. Cell. Biol. 12, 3146-3155.

Durbec PL, Larsson-Blomberg LB, Schuchardt A, Costan-tini F and Pachnis V. (1996). Development 122, 349-358. MEDLINE

Edery P, Eng C, Munnich A and Lyonnet S. (1997). Bioessays 19, 389-395. MEDLINE

Edery P, Lyonnet S, Mulligan LM, Pelet A, Dow E, Abel L, Holder S, Nihoul FC, Ponder BA and Munnich A. (1994). Nature 367, 378-380. MEDLINE

Eng C, Smith D, Mulligan L, Nagai M, Healey C, Ponder M, Gardner E, Scheumann G, Jackson C and Tunnacliffe A. (1994). Hum. Mol. Genet. 3, 237-241. MEDLINE

Forss-Petter S, Werner H, Berger J, Lassmann H, Molzer B, Schwab MH, Bernheimer H, Zimmerman F and Nave KA. (1997). J. Neurosci. Res. 50, 829-843. MEDLINE

Heinicke EA, Kiernan JA and Wijsman J. (1987). J. Neurosci. Meth. 21, 45-54.

Hofstra RM, Landsvater RM, Ceccherini I, Stulp RP, stelwagen T, Luo Y, Pasini B, Hoppener JW, van AHK, Romeo G et al. (1994). Nature 367, 375-376. MEDLINE

Ikeda I, Ishizaka Y, Tahira T, Suzuki T, Onda M, Sugimura T and Nagao M. (1990). Oncogene 5, 1291-1296. MEDLINE

Ikuno N, Shimokawa I, Nakamura T, Ishizaka Y and Ideda T. (1995). Path. Res. Pract. 191, 92-99. MEDLINE

Iwashita T, Asai N, Murakami H, Matsuyama M and Takahashi M. (1996). Oncogene 12, 481-487. MEDLINE

Jhiang SM, Caruso DR, Gilmore E, Ishizaka Y, Tahira T, Nagao M, Chiu I-M and Massaferri EL. (1992). Oncogene 7, 1331-1337. MEDLINE

Kapur RP, Hoyle GW, Mercer EH, Brinster RL and Palmiter RD. (1991). Neuron 7, 717-727. MEDLINE

Kapur RP, Yost C and Palmiter RD. (1992). Development 116, 167-175. MEDLINE

Kusafuka T and Puri P. (1997). Pediatr. Surg. Int. 12, 11-18. MEDLINE

Lechner MS and Dressler GR. (1997). Mech. Dev. 62, 105-120. MEDLINE

Liu ZZ, Wada J, Kumar A, Carone FA, Takahashi M and Kanwar YS. (1996). Dev. Biol. 178, 133-148. MEDLINE

Mercer EH, Hoyle GW, Kapur RP, Brinster RL and Palmiter RD. (1991). Neuron 7, 703-716. MEDLINE

Michiels F-M, Chappuis S, Caillou B, Pasini A, Talbot M, Monier R, Lenoir GM, Geunteun J and Billaud M. (1997). Proc. Natl. Acad. Sci. USA 94, 3330-3335. MEDLINE

Moore MW, Klein RD, Farinas I, Sauer H, Armanini M, Phillips H, Reichardt LF, Ryan AM, Carver-Moore K and Rosenthal A. (1996). Nature 382, 76-79. MEDLINE

Pachnis V, Mankoo B and Costantini F. (1993). Development 119, 1005-1017. MEDLINE

Pichel JG, Shen L, Sheng HZ, Granholm A-C, Drago J, Grinberg A, Lee EJ, Huang SP, Saarma M, Hoffer BJ, Sariola H and Westphal H. (1996). Nature 382, 73-76. MEDLINE

Rizzo C, Califano D, Colucci-D'Amato GL, De Vita G, D'Alessio A, Dathan NA, Fusco A, Monaco C, Santelli G, Vecchio G, Santoro M and de Franciscis V. (1996). J. Biol. Chem. 271, 29497-29501. MEDLINE

Romeo G, Ronchetto P, Luo Y, Barone V, Seri M, Ceccherini I, Pasini B, Bocciardi R, Lerone M, Kaariainen H et al. (1994). Nature 367, 377-378. MEDLINE

Rossel M, Pasini A, Chappius S, Geneste O, Fournier L, Schuffenecker I, Takahashi M, van Grunsven LA, Urdiales JL, Rudkin BB, Lenoir GM and Billaud M. (1997). Oncogene 14, 265-275. MEDLINE

Santoro M, Carlomagno R, Hay ID, Herrman MA, Grieco M, Melillo R, Pierotti MA, Bongarzone I, della Porta G, Berger N, Peix JL, Paulin C, Fabien N, Vecchio G, Jenkins RB and Fusco A. (1992). J. Clin. Invest. 89, 1517-1522. MEDLINE

Santoro M, Carlomagno R, Romano A, Bottaro DP, Dathan NA, Grieco M, Fusco A, Vecchio G, Matoskova B, Kraus MH and Di Fiore PP. (1995). Science 267, 381-383. MEDLINE

Santoro M, Wong WT, Aroca P, Santos E, Matoskova B, Grieco M, Fusco A and Di Fiore PP. (1994). Mol. Cell. Biol. 14, 663-675. MEDLINE

Schuchardt A, D'Agati V, Larsson BL, Costantini F and Pachnis V. (1994). Nature 367, 380-383. MEDLINE

Schuchardt A, D'Agati V, Pachnis V and Costantini F. (1996). Development 122, 1919-1929. MEDLINE

Searle PF, Davison BL, Stuart GW, Wilkie TM, Norstedt G and Palmiter RD. (1984). Mol. Cell. Biol. 4, 1221-1230. MEDLINE

Snouwaert JN, Brigman KK, Latour AM, Malouf NN, Boucher RC, Smithies O and Koller BH. (1992). Science 257, 1083-1088. MEDLINE

Songyang Z, Carraway III KL, Eck MJ, Harrison SC, Feldman RA, Mohammadi M, Schlessinger J, Hubbard SR, Smith DP, Eng C, Lorenzo MJ, Ponder BAJ, Mayer BJ and Cantley LC. (1995). Nature 373, 536-539. MEDLINE

Sweetser DA, Kapur RP, Froelick GJ, Kafer KE and Palmiter RD. (1997). Oncogene 15, 2783-2794. MEDLINE

Tahira T, Ishizaka Y, Itoh F, Sugimura T and Nagao M. (1990). Oncogene 5, 97-102. MEDLINE

Takahashi M, Buma Y and Hiai H. (1989). Oncogene 4, 805-806. MEDLINE

Takahashi M, Buma Y, Iwamoto T, Inaguma Y, Ikeda H and Hiai H. (1988). Oncogene 3, 571-578. MEDLINE

Tsuzuki T, Takahashi M, Asai N, Iwashita T, Matsuyama M and Asai J. (1995). Oncogene 10, 191-198. MEDLINE

van Weering DHJ and Bos JL. (1997). J. Biol. Chem. 272, 249-254. MEDLINE

Vega QC, Worby CA, Lechner MS, Dixon JE and Dressler GR. (1996). Proc. Natl. Acad. Sci. USA 93, 10657-10661. Article MEDLINE

Wajjwlku W, Nakamura S, Hasegawa Y, Miyazaki K, Satoh Y, Funahashi H, Matsuyama M and Takahashi M. (1992). Jpn. J. Cancer Res. 83, 671-675. MEDLINE

Wood KW, Sarnecki C, Roberts TM and Blenis J. (1992). Cell 68, 1041-1050. MEDLINE

Worby CA, Vega QC, Zhao Y, Chao HH-J, Seasholtz AF and Dixon JE. (1996). J. Biol. Chem. 271, 23619-23622. MEDLINE

Figure 1 Structure of the DH-RET^MEN2B transgene. The RET^MEN2B cDNA is depicted by a black box with an asterisk to mark the site of the MetThr substitution at amino acid position 918. Intron 1 from the rat preproinsulin gene, inserted to improve expression, is shown as stippled boxes upstream of the cDNA. A DNA fragment containing the polyadenylation site from the murine protamine 1 gene is indicated by a hatched arrow downstream of the coding sequence

Figure 2 Whole mount tissues of 28-day old wild-type (a) and DH-RET^MEN2B transgenic (b) siblings from the low copy line, both carrying a DH-nlacZ and stained with X-gal to demonstrate sympathetic ganglia. The paravertebral sympathetic ganglia (p) are enlarged and often fused in DH-RET^MEN2B transgenic mice. The celiac ganglia (cg) are considerably bigger in many of the transgenic mice with extension of sympathetic tissue superiorly through the adrenal cortex into the adrenal medullae (a)

Figure 3 Histological comparison of adrenal glands from a wild-type mouse (A) and DH-RET^MEN2B transgenic mice at 11 (B, low copy line) and 13 (C, founder 7187) weeks. A low magnification view of each adrenal gland is shown as an insert with the area of enlargement indicated. The adrenals from transgenic mice displayed a marked expansion of the adrenal medulla with disruption of the surrounding cortex (c). Normal appearing medullary chromaffin cells (m) were preserved in foci adjacent to the cortex in transgenic mice. However, most of the medulla was replaced by pleomorphic neurons (n), nerve fibers and Schwann cells (the latter seen as small elongated nuclei). Spontaneous maturation, characterized by increased size and separation of neuronal cell bodies and decreased stromal cellularity, was evident within the adrenal medullae of some of the DH-RET^MEN2B transgenic lines. Bars: 50 m, 250 m (insets)

Figure 4 Immunohistochemical comparison of adrenal medullae from an 11-week old DH-RET^MEN2B mouse (A and C) and its non-transgenic littermate (B and D) stained for RET (A and B) or peripherin (C and D). RET-positive neuron cell bodies abound in the adrenal medulla of the transgenic mouse, but are not detected in the control tissue that is composed almost entirely of medullary chromaffin cells (m). A subset of normal chromaffin cells in transgenic mice and non-transgenic littermates express relatively low levels of RET protein. In addition to neurons and chromaffin cells, abundant nerve fibers (f) with associated small, condensed, elongated nuclei of Schwann cells were demonstrated by anti-peripherin immunohistochemistry in DH-RET^MEN2B adrenal glands (C). Far less numerous, peripherin-positive, nerve processes are present in wild-type mice (D). Adrenal cortex (c); Bars: (A and B), 25 m; (C and D), 50 m

Figure 5 A comparison of anti-RET immunohistochemical staining in celiac ganglia from a 11-week-old non-transgenic mouse (A) and its DH-RET^MEN2B transgenic sibling (B). The insets show low magnification views of the celiac ganglia. The transgenic ganglion exhibits greater cellularity, broader bands of neuropil and marked variability in the diameters of RET-positive ganglion cells. Bars: 50 m, 250 m (insets)

Figure 6 Whole-mount acetylcholinesterase staining of the rectal myenteric plexuses from an adult transgenic mouse (A, intermediate-copy line) and its non-transgenic littermate (B). No differences in the architecture of the plexus were identified by this technique

Figure 7 Transverse sections through the kidneys (k) and preaortic ganglion of a newborn non-transgenic mouse (A) and its DH-RET^MEN2B littermate (B) from the high-copy line. The celiac ganglion (cg) of the transgenic pup is massively enlarged. In addition, the transgenic individual has hypoplastic kidneys which contain dilated tubules, few glomeruli and an overall reduction of nephrons. Ureters (ur) are present bilaterally

Figure 8 Comparison of RET levels in adrenal glands and celiac ganglia from three DH-RET^MEN2B transgenic mouse lines. Fifty micrograms of total protein extracted from selected tissues were analysed for RET expression by Western blot. For the low copy line the adrenal glands and celiac ganglia were discrete, separate structures and protein was extracted from adrenal glands (A), or from preaortic ganglia (PG) pooled from three newborn mice. In the intermediate and high copy lines, the preaortic ganglia and adrenals were often fused together and not distinct so the combination was analysed for near term (E19) fetuses (individually numbered). The highest levels of RET were found in individuals with renal hypoplasia or agenesis (identified by asterisks)

Figure 9 Western blot analysis of RET, activated MAPK and total MAPK proteins in various tissues from transgenic (TG) newborn mice, as well as wild-type (WT) littermate controls. Celiac ganglia from adult wild-type and transgenic mice were also compared in the lanes indicated by asterisks. The small sizes of celiac and thoracic paravertebral sympathetic (TS) ganglia in newborn wild-type mice precluded their inclusion in the data set. RET over-expression is evident in sympathetic ganglia, adrenals, and, to a much lesser degree, colons from transgenic animals, but does not correlate with MAPK activation. Adr, adrenal gland; SCG, superior cervical ganglion

Figure 10 Analysis of tissue catecholamine levels from DH-RET^MEN2B (age 16±1 weeks) transgenic animals and their sex-matched wild-type littermates (wt). Levels of norepinephrine, epinephrine and dopamine were measured in pooled submaxillary ganglia (SMG), colon (Col), spleen (Spl), heart (Hrt), brown fat (BF) and adrenal glands (Adr). Equal numbers of each sex were used. In no case was a statisitically significant difference seen, except for in those tissues indicated by an asterisk

Table 1 Table 1