Hypothalamic Growth Hormone-Releasing Hormone (GHRH ...

0888-8809/05/$15.00/0 Printed in U.S.A.

Molecular Endocrinology 19(5):1251–1262 Copyright © 2005 by The Endocrine Society doi: 10.1210/me.2004-0223

Hypothalamic Growth Hormone-Releasing Hormone (GHRH) Deficiency: Targeted Ablation of GHRH Neurons in Mice Using a Viral Ion Channel Transgene Paul R. Le Tissier, Danielle F. Carmignac, Sarah Lilley, Abdul K. Sesay, Carol J. Phelps, Pamela Houston, Kathleen Mathers, Charalambos Magoulas, David Ogden, and Iain C. A. F. Robinson National Institute for Medical Research, Divisions of Molecular Neuroendocrinology (P.L.T., D.F.C., P.H., C.M., I.C.A.F.), Neurophysiology (S.L., A.K.S., D.O.), and Biological Services (K.M.), The Ridgeway, London NW7 1AA, United Kingdom; and Department of Structural and Cellular Biology (C.J.P.), Tulane University School of Medicine, New Orleans, Louisiana 70112-2699 Animal and clinical models of GHRH excess suggest that GHRH provides an important trophic drive to pituitary somatotrophs. We have adopted a novel approach to silence or ablate GHRH neurons, using a modified H37A variant of the influenza virus M2 protein (H37AM2). In mammalian cells, H37AM2 forms a high conductance monovalent cation channel that can be blocked by the antiviral drug rimantadine. Transgenic mice with H37AM2 expression targeted to GHRH neurons developed postweaning dwarfism with hypothalamic GHRH transcripts detectable by RT-PCR but not by in situ hybridization and immunocytochemistry, suggesting that expression of H37AM2 had silenced or ab-

lated virtually all the GHRH cells. GHRH-M2 mice showed marked anterior pituitary hypoplasia with GH deficiency, although GH cells were still present. GHRH-M2 mice were also deficient in prolactin but not TSH. Acute iv injections of GHRH in GHRH-M2 mice elicited a significant GH response, whereas injections of GHRP-6 did not. Twice daily injections of GHRH (100 ␮g/d) for 7 d in GHRH-M2 mice doubled their pituitary GH but not PRL contents. Rimantadine treatment failed to restore growth or pituitary GH contents. Our results show the importance of GHRH neurons for GH and prolactin production and normal growth. (Molecular Endocrinology 19: 1251–1262, 2005)

G

to GHRH neurons (7, 8). Like all central nervous system neurons, the activity and survival of GHRH cells are dependent on the maintenance of resting membrane potential and ion fluxes through ion channels, and this can be manipulated by overexpression of heterologous ion channels to silence or ablate cells (9–11). In this report, we describe the first application of such a transgenic ion-channel strategy with the aim of silencing or ablating neuroendocrine cells. The strategy we chose is based on H37AM2, a variant of the influenza M2 viral protein that forms a homotetrameric monovalent cation channel in mammalian cells that can be blocked by the antiinfluenza drug rimantadine (12–14). The 37His residue is important for ion specificity; mutation of this residue to 37 Ala broadens the specificity of the channel to other monovalent cations and reduces the pH sensitivity of the channel but retains its sensitivity to blockade with rimantadine (Ogden, D., unpublished). When expressed in mammalian cells in vitro, H37A M2 is conditionally lethal; it kills cells unless they are cultured in the presence of rimantadine (11). When expressed in transgenic mice from a T-cellspecific p56Lck promoter, H37AM2 irreversibly ablated a cell lineage in the developing immune system (11). We have now made a modified version of this

HRH IS THE PRIMARY positive hypothalamic regulator of the GH axis (1). In addition to stimulating GH release from pituitary somatotrophs, it has important trophic effects to stimulate GH synthesis (2) and somatotroph proliferation (3). The effects of excess GHRH are somatotroph hyperplasia and chronically increased GH secretion, evident both from transgenic animal models (4) and from humans with ectopic GHRH production (5). A very recent report of a mouse with a partial disruption of the GHRH gene (6) shows that the effects of GHRH deficiency are similar those seen in mice or humans with inactivating mutations in the GHRH receptor (6–8), namely somatotroph hypoplasia, GH deficiency, and dwarfism. We have previously engineered a rat GHRH cosmid that efficiently targets a variety of transgene products First Published Online January 20, 2005 Abbreviations: ARC, Arcuate nucleus; CMV, cytomegalovirus; eGFP, enhanced green fluorescence protein; GHD, GH deficiency; GSH1, GS homeobox 1; hGH, human GH; ICC, immuncytochemistry; n.s., not significant; NT, nontransgenic; PRL, prolactin; rGHRH, rat GHRH; UTR, untranslated region. Molecular Endocrinology is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.

1251

1252 Mol Endocrinol, May 2005, 19(5):1251–1262

H37A

M2 channel construct and confirmed, using whole cell patch clamp techniques, that it generated a reversible rimantadine-sensitive monovalent cation conductance when expressed in an endocrine cell line in vitro. This channel construct was then cloned into a 38-kb rat GHRH cosmid, which was used to generate transgenic mice expressing H37A M2 in GHRH neurons. The resulting GHRH-M2 mice provide the first genetic model of specific hypothalamic GHRH deficiency caused by ionic ablation of GHRH neurons. They exhibit the expected severe secondary GH deficiency and dwarfism but also unexpected defects in prolactin (PRL) production. Some of these results have recently been reported in preliminary form (15).

RESULTS H37A

M2 Forms a Monovalent Cation Channel in GC Cells

GC cells were transiently transfected with the cytomegalovirus (CMV)-H37AM2 plasmid (Fig. 1A) and/or a CMV eGFP plasmid, and cultured in the presence of rimantadine. After 48 h, cells were fixed and immunostained for H37AM2 protein, and/or examined for eGFP fluorescence. The H37AM2 protein was expressed diffusely across the cell surface membrane (Fig. 1C). Unpermeabilized H37AM2-transfected cells showed a sparse and punctate pattern of labeling, and cells

Fig. 1. GC Cells Cotransfected with H37AM2 and eGFP A, The plasmid used to transfect GC cells contained a CMV promoter driving expression of an MluI cassette based on the hGH gene in which sequences between exons 2–5 were deleted and replaced with those corresponding to residues 2–97 of the H37AM2 channel protein. This provided an ATG, intron, splice donor and acceptor sites and the hGH polyA addition site. Numbered gray bars correspond to portions of hGH exons, not drawn to scale. B, The predicted protein product is H37AM2 with its N-terminal methionine replaced by the heptapeptide MATGSRT. C, GC cells were cotransfected with H37AM2 eGFP; plasmids. Left panel, H37AM2 immunofluorescence; right panel, eGFP fluorescence.

Le Tissier et al. • Ablation of GHRH Neurons in Transgenic Mice

transfected with eGFP only or exposed to the secondary antibody alone showed no H37AM2 staining (data not shown). Because M2 staining and eGFP fluorescence coincided in most individual cells, eGFP fluorescence guided the selection of cells for patch-clamp analysis in subsequent experiments. Whole-cell currents from eGFP/H37AM2 cotransfected cells were recorded in response to the voltage protocol shown in Fig. 2A. In the absence of extracellular Na⫹ and K⫹, and with no pH gradient across the cell membrane, this series of voltage steps induced a large, outwardly rectifying current at potentials positive to ⫺100 mV (Fig. 2B), carried purely by Cs⫹ ions from the internal solution. When the extracellular solution was exchanged for one that contained Na⫹, the current-voltage relationship became almost linear, reversing at ⫺17.8 ⫾ 1.1 mV (n ⫽ 21; Fig. 2, B and C). These current characteristics suggest that the transfected cells were now expressing an additional ion channel with nonselective permeability to monovalent cations. The prepulse inactivation protocol also shows that the channel does not inactivate because the level of hyper- or depolarization during the variable first step does not affect the size of the current evoked by the constant second step. This nonselective, noninactivating monovalent cation conductance was not observed in eGFP-only transfections or in untransfected GC cells (data not shown) and could thus be attributed to the H37AM2 channel protein. The identity of this current was confirmed using the specific M2 channel blocking drug, rimantadine. Figure 2, B and C, shows the effect of applying 23 ␮M rimantadine to H37AM2-GC cells. Rimantadine blockade of the H37AM2 channel current reached a steady state within 1 min, and it was fully reversed after a drug-free washout (Fig. 2, B and C). This effect of rimantadine was concentration dependent. Figure 2D shows the effects of increasing concentrations of rimantadine to suppress H37AM2 currents recorded in the presence of extracellular Na⫹ at a command potential of ⫺70 mV (below the level of activation of any voltage-dependent currents). The Hill coefficient of the concentration-inhibition curve is 0.8 with an IC50 of 1.5 ␮M. We conclude that addition of the N-terminal peptide tag to H37AM2 (see Materials and Methods) did not alter the ability of H37AM2 to form a nonselective monovalent cation channel, with a constitutively active conductance at physiological pH, that could be reversibly inhibited by rimantadine. Generation of GHRH-M2 Transgenic Mice This H37AM2 construct was then cloned into a 38-kb rat GHRH promoter construct (Fig. 3A) and used to generate transgenic mice. Three founders were identified; one was infertile, but from two others, stable lines were established (lines I and J). Both lines showed a similar phenotype, and all the results


Mol Endocrinol, May 2005, 19(5):1251–1262 1253

Fig. 2. Ion Channel Activity in GC Cells Cotransfected with H37AM2 and eGFP A, From a resting potential of ⫺50 mV, cells were whole-cell voltage-clamped using the voltage protocol shown (dashed line indicates region from which average current was taken). B, i–iv: Example currents recorded from the same cell in the absence of extracellular Na⫹ (i), (ii) in the presence of extracellular Na⫹(ii), in the presence of Na⫹ ⫹ 23 ␮M rimantadine (Rim) (iii) and after a 6-min washout of rimantadine, again in the presence of Na⫹(iv). C, Mean (⫾SEM) steady-state current-voltage relationships expressed in terms of current density. Numbers in brackets refer to numbers of measurements; Rim ⫽ rimantadine 23 ␮M. D, Concentration-inhibition curve for rimantadine showing suppression of currents recorded at ⫺70 mV in the presence of extracellular Na⫹. Data are individual (solid symbols) or mean ⫾SEM of 10 observations at 23 ␮M (open symbol), fitted with a Hill plot, giving the following parameters: slope 0.8, IC50 1.5 ␮M, maximum 93%.

presented will be from line I, comparisons being made between GHRH-M2 and nontransgenic (NT) littermates, unless otherwise stated. Hemizygous males and females were fertile, with normal litter sizes, but their transgenic progeny were severely dwarfed compared with NT littermates (Fig. 3B). For the first 2 postnatal weeks, there was no difference in size (Fig. 3C), but from 3 wk onwards, GHRH-M2 animals grew more slowly than their NT littermates, attaining only 60% of their weight by 6 wk of age (Fig. 3C), and remaining dwarfed in adulthood (Table 1). Measurements of pituitary GH contents showed that adult GHRH-M2 mice had severe GH deficiency (GHD) compared with their NT littermates (Table 1).

GHRH-M2 Mice Show Profound GHRH Deficiency and Anterior Pituitary Hypoplasia In situ hybridization and immuncytochemistry (ICC) performed on hypothalamic sections from NT mice showed the expected distribution of specific GHRH mRNA expression in the cells of the arcuate nucleus (ARC; Fig. 4A) and GHRH peptide in both cell bodies and in strongly stained terminals in the median eminence (Fig. 4E). In contrast, in GHRH-M2 mice, both GHRH mRNA expression (Fig. 4D), and GHRH peptide immunoreactivity (Fig. 4F) were virtually absent. ICC with antibodies for H37AM2 protein also failed to detect any positive cells in the ARC (data not shown). The


Fig. 3. GHRH-M2 Transgenic (T) Mice Are Dwarfed A, The tagged H37AM2 expression cassette flanked by hGH exon sequences (gray bars) and Mlu1 sites (M) was cloned into a specific Mlu1 site (M) of the first hypothalamic exon (H1) of a cosmid containing the entire rat GHRH gene (exons 2–5), and 16 kb upstream and 14 kb downstream flanking sequences. B, Both male and female GHRH-M2 transgenic mice showed dwarfism compared with NT littermates. C, The reduction in body weight in GHRH-M2 mice developed after weaning. Data shown are mean (⫾SEM) weights of agematched groups of female (n ⫽ 6–14) transgenic and NT littermates. ***, P ⬍ 0.001 vs. age-matched NT group. Wt, Weight.


the arcuate GHRH cell bodies and terminals in the median eminence of singly transgenic GHRH-eGFP mice (Fig. 4G). Compared with their NT littermates, GHRH-M2 mice had severe selective anterior pituitary hypoplasia, (Fig. 4, I and J) with the neurointermediate lobes being unaffected. ICC showed that GH⫹ve cells were present in GHRH-M2 mice, but the numbers were reduced in the hypoplastic GHRH-M2 anterior lobe (Fig. 4J). In one experiment, GHRH-M2 mice were crossed with GH-eGFP transgenic mice (17) so that the distribution and density of eGFP-tagged GH cells could be compared. In the double transgenic (GHRH-M2 ⫻ GHeGFP) progeny, eGFP⫹ve, GH cells were readily apparent (Fig. 4L), but their density was markedly reduced throughout the anterior pituitary with fewer, smaller clusters of GH cells, compared with the normal GH-eGFP mice (Fig. 4K). H37A M2 transcripts could be detected by RT-PCR in hypothalamic, testicular, and renal RNA from transgenic but not NT animals (Fig. 5). Although GHRH expression was virtually eliminated, as assessed by ICC or in situ hybridization, residual GHRH transcripts could still be amplified by RT-PCR, using specific primers for transcripts from both the hypothalamic and placental promoters (Fig. 5). Expression of both classes of transcript was evident in hypothalamus and placenta from both GHRH-M2 and NT mice, whereas placental but not hypothalamic transcripts were detected in testis. No GHRH or M2 expression was observed in spleen. Pituitary GH, PRL, and TSH contents were measured in two groups of female GHRH-M2 and NT mice at 14 and 42 d of age (Fig. 6). Specific GH deficiency was already evident at 14 d. PRL content was unaffected at 14 d but was significantly lower in the GHRH-M2 transgenic mice by 42 d. The effect of lower pituitary PRL content on relative lactational performance was not determined, but milk production was sufficient for hemizygous transgenic females to raise litters of a normal size. ICC showed a normal distribution of a reduced number of PRL⫹ve cells in the hypoplastic GHRH-M2 anterior lobe (data not shown). TSH content in the same extracts was unaffected at either age (Fig. 6). Measurements in other groups of mice showed that pituitary GH contents were already significantly lower in GHRH-M2 mice by 7 d of age (1.06 ⫾ 0.12 ␮g vs. 2.45 ⫹ 0.33 ␮g in NT littermates, P ⬍ 0.01), well before their dwarfism was apparent. In Vivo Studies in GHRH-M2 Mice

lack of GHRH immunoreactivity was not simply secondary to GHD or dwarfism because another transgenic line with dwarfism and pituitary GHD caused by a primary somatotroph defect (16) showed the expected increase in GHRH expression (Fig. 4C). GHRH-M2 mice were crossed with GHRH-eGFP transgenic mice (8), allowing visualization of GHRH neurons. Doubly transgenic GHRH-eGFP/GHRH-M2 mice (Fig. 4H) lacked any of the bright fluorescence in

We tested whether the remaining somatotrophs in GHRH-M2 mice would still respond to GH secretagogues. The mice were anesthetized, equipped with iv catheters, and given iv bolus doses of GHRP-6 (500 ng) and GHRH (100 ng), and their GH responses measured (Fig. 7). Plasma GH levels were undetectable before treatments and were barely detectable after GHRP-6 injections, whereas all mice showed a small but significant GH release in re-



Table 1. GH Deficiency and Dwarfism in Adult GHRH-M2 Mice Female GHRH-M2 Female NT Male GHRH-M2 Male NT a

n

Weight (g)

Nose-Anus Length (mm)

Pituitary GH (␮g)

6 5 6 6

13.8 ⫾ 0.5a 23.1 ⫾ 0.7 16.2 ⫾ 0.7a 29.4 ⫾ 0.7

66.8 ⫾ 1.1a 85.7 ⫾ 1.8 67.3 ⫾ 0.8a 91.0 ⫾ 0.7

0.8 ⫾ 0.3a 62.3 ⫾ 8.7 3.9 ⫾ 1.0a 125 ⫾ 9.3

P ⬍ 0.001 vs. NT group.

sponse to GHRH. Other groups of GHRH-M2 mice were given injections of GHRH (50 ␮g, sc) or saline, twice daily for 7 d. This GHRH treatment more than doubled pituitary GH contents but had no effect on PRL contents (Fig. 8), whereas weight gain was only marginally affected [⌬weight; 8.8 ⫹ 0.3 g in GHRH injected mice vs. 8.3 ⫹ 0.4 g in saline controls, not significant (n.s.)]. We tested whether treatment with rimantadine could prevent or reduce the ablation or silencing of GHRH neurons and resulting dwarfism in GHRH-M2 mice. Rimantadine was added to the drinking water of a group of female GHRH-M2 (transgenic, n ⫽ 3) and NT littermates (n ⫽ 4) at weaning. At the start of treatment, the weights were 20.5 ⫾ 0.6 g and 12.2 ⫾ 0.7 g (NT vs. transgenic, P ⬍ 0.001). Mice treated with rimantadine did not show any of the catch-up growth in the GHRH-M2 mice expected if GHRH deficiency was reversed (after 5 wk, weights were 23.2 ⫾ 1.0 g and 14.7 ⫾ 0.7 g, NT vs. transgenic, P ⬍ 0.001). In another group of animals, pituitary GH content was measured in groups of GHRH-M2 mice with and without rimantadine added to the drinking water from weaning. Treatment with rimantadine for 5 wk had no significant effect on pituitary GH content (2.9 ⫾ 1.3 ␮g with rimantadine treatment vs. 3.2 ⫾ 1.2 ␮g without, n ⫽ 4 in both groups, n.s.). Finally, we tested whether rimantadine could block or reduce ablation or silencing of GHRH neurons in GHRH-M2 mice if animals were treated from an age before GHRH (and therefore the transgene) is expressed. Treatment of transgenic and NT mice from 3 d before birth and continuing this treatment for a further 5 wk, by addition of rimantadine to the drinking water of lactating mothers and for 2 wk after weaning, failed to prevent the reduction in pituitary GH content in GHRH-M2 transgenic animals (3.2 ⫾ 0.2 ␮g vs. 58.0 ⫾ 8.7 ␮g in NT littermates, P ⬍ 0.001). We also treated pregnant mothers from 2 d after mating by addition of rimantadine to the drinking water, to deliver the drug in utero to transgenic and NT mice from an earlier age, and measured pituitary GH content of 1 d old offspring. At this age, no reduction in pituitary GH content was found in transgenic animals with or without rimantadine treatment compared with NT littermates (transgenic with rimantadine, 414.3 ⫾ 58.1 ng (n ⫽ 6); transgenic without rimantadine, 537.1 ⫾ 86.2 ng (n ⫽ 12); NT, 441.3 ⫾ 57.5 ng (n ⫽ 12), no significant differences).

DISCUSSION There is much circumstantial evidence that a normal GHRH axis is essential for the postnatal development, production, and secretion of GH. Observations in lit/lit mice (18, 19) and in humans with mutations in the GHRH receptor (20) show that defective GHRH receptor signaling results in profound, selective GH deficiency and dwarfism. Experimental ablation or inhibition of GHRH by chemical or immunological means (21, 22) provides strong circumstantial support for the notion that both acute and chronic GH release is strongly dependent on the proper functioning of arcuate GHRH neurons. No inactivating mutations or deletions in the GHRH gene have yet been reported in human subjects, but GHRH deficiency has been described in rodents as part of more complex phenotypes resulting from deletion of other genes, such as the transcription factor GS homeobox 1 (GSH1) (23) or the convertase PC1/3 (24), or after the expression of human GH (hGH) transgenes in central nervous system to inhibit GHRH expression (7, 25). Recently, a report describing a targeted disruption of the GHRH gene has appeared, confirming directly the requirement for GHRH for normal growth and GH production (6). Instead of an irreversible GHRH gene knockout approach, we chose a transgenic strategy using ion channels with the potential to reversibly silence disrupt, or ablate neuroendocrine neurons. Transgenic expression of homologous or heterologous K⫹ ion channels has been used to alter the activity of several types of excitable cells (26–28), but drugs cannot be used to control the conductance of these channels selectively because the same or related channels are present in many other cells. However, by using ion channels derived from nonmammalian systems, one can take advantage of drugs that act on these transgene channels specifically at doses that minimally affect mammalian channels (10, 11), in attempts to regulate the conductance of the transgene ion channel selectively. The ion channel we chose is based on the influenza virus M2 protein (13, 29), which has several advantages for use as a transgene. M2 is a simple single chain 97-residue protein with a single transmembrane domain that, when expressed in mammalian cells, assembles to form a pH-sensitive homotetrameric proton channel in the plasma membrane that can be blocked by the antiinfluenza drugs, amantidine, and



rimantadine (30–32). The ionic specificity of M2 could be broaden to monovalent cations by a single amino acid mutation, and expression of this H37AM2 variant

Fig. 5. RT-PCR Analysis of Gene Expression in GHRH-M2 Mice RT-PCR performed on RNA extracts from GHRH-M2 transgenic (T) and NT littermates shows M2 expression in the hypothalamus, testis, and kidney in GHRH-M2 transgenic mice. GHRH transcripts from the hypothalamic exon (Hypo-GHRH) were detectable in the hypothalamus and placenta of transgenic animals and from the placental exon (Plac-GHRH) in hypothalamus, testis, and placenta. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was amplified as positive control for cDNA; Blank, water negative control.

Fig. 4. Hypothalamic GHRH and M2 Expression in GHRH-M2 Transgenic (T) Mice A–D, In situ hybridization with a GHRH riboprobe with hypothalamic sections showing (A) specific GHRH mRNA expression in the ARC (arrows) in NT normal mouse. B, Riboprobe sense control shows no signal. C, GHRH expression is enhanced in the ARC of a different transgenic mouse(⌬exon3hGH) with primary GH deficiency and dwarfism, whereas in GHRH-M2 transgenic mice (D), GHRH expression is undetectable. E and F, Immunocytochemistry for GHRH in hypothalamic sections shows (E) GHRH staining (brown) in individual ARC cell bodies and a dense staining of their projections to the median eminence in NT mice, but little or no staining in (F) GHRH-M2 transgenic mice. G and H, eGFP in hypothalamic sections from (G) a GHRH-eGFP transgenic mouse and (H) a double transgenic GHRH-eGFP/GHRH-M2 mouse. I and J, Immunocytochemistry for GH was performed on sections of pituitaries from (I) NT and (J) GHRH-M2 mice. GH-containing cells are brown; nuclei are stained with hematoxylin (blue). Note the marked selective anterior pituitary hypoplasia in GHRH-M2 mice. K and L, eGFP in pituitary sections obtained from (K) a GH-eGFP transgenic mouse, and (L) a double transgenic GH-eGFP/ GHRH-M2 mouse.

was known to kill mammalian cells unless they are cultured in the presence of rimantadine; furthermore, expression in transgenic animals leads to ablation of cells in which it is expressed (11). For our transgene construct, H37AM2 was inserted into the backbone of an hGH expression cassette to provide transcriptional intron splicing and translational signals used previously to express other products in both GHRH and GH cells (8, 16, 17). It was also flanked with an MluI linker, so it could be inserted into the unique MluI cloning site previously introduced (7) into the hypothalamic promoter of the rat GHRH (rGHRH) cosmid. The N-terminal sequences of the GH signal peptide and H37AM2 are fortuitously similar, so it was possible to introduce H37AM2 in such a way that the first intron in the GH expression cassette could be retained, while adding only a small heptapeptide Nterminal tag to H37AM2 in the final splice product. Other mutational studies with M2 (Hay, A., National Institute for Medical Research; personal communication) suggested that this would likely be tolerated well without compromising channel assembly, conductance, or rimantadine sensitivity, but we sought to confirm this by expressing the modified H37AM2 construct in cell lines in vitro before making transgenic animals. When transfected into GC cells, H37AM2 immunoreactivity was readily detected in the cell membranes by immunofluorescence and Western blotting. Using cotransfection with eGFP, transfected cells could be identified and subjected to patch-clamp analysis for measurements of H37AM2 channel activity. These confirmed the presence of high conductance ion channels in cells cotransfected with H37AM2, but not in untransfected GC cells or cells transfected with eGFP alone. The channel properties conformed closely with those known for unmodified H37AM2 channels expressed in other cells (Ogden, D., unpublished), providing a broad specificity noninactivating monovalent cation conductance at physiological pH, which could be blocked in a dose-dependent and reversible fashion by rimantadine (32).



Fig. 6. Pituitary GH, PRL, and TSH Contents in GHRH-M2 Transgenic (T) Mice Pituitaries from groups of female GHRH-M2 transgenic and NT littermates at 14 and 42 d of age were assayed for GH, PRL, and TSH contents. Data shown are mean ⫾ SEM of six to 14 animals. **, P ⬍ 0.01; ***, P ⬍ 0.001 vs. NT littermates.

We then generated transgenic mice expressing this heptapeptide-tagged H37AM2 channel protein in GHRH neurons. Two lines of mice were established, both of which showed severe GH deficiency and dwarfism. When coronal sections of brains were examined for GHRH mRNA or peptide expression by in situ hybridization or immunocytochemistry, GHRH-M2 mice were essentially devoid of the GHRH expression readily detectable in WT hypothalamic ARC. This was clearly not secondary to GH deficiency, which would cause an increase of GHRH expression (33), as was evident when GHRH-M2 sections were compared with those from a different transgenic mouse line with primary GH deficiency (16). Any increased drive to GHRH gene expression caused by GH deficiency in

Fig. 7. Release of GH in Response to GH Secretagogues in GHRH-M2 Transgenic Mice Jugular catheters were inserted in anesthetized GHRH-M2 transgenic mice and blood samples obtained before, and 5 min after, iv injection of 500 ng GHRP-6 (hatched bar) and 90 min later, before, and 5 min after iv injection of 100 ng GHRH (solid bar). Blood samples were centrifuged and the plasma assayed for GH by RIA. Data shown are mean ⫾ SEM of five animals aged between 7 and 8 wk. ***, P ⬍ 0.001, vs. preGHRH. l.d., Minimum level of detection of the assay.

GHRH-M2 mice, would also drive more GHRHM2 expression, increasing the disruption of GHRH neurons. Although no GHRH mRNA or peptide could be detected in GHRH-M2 mice by in situ hybridization or immunocytochemistry, both GHRH and H37AM2 transcripts could be detected in hypothalamic extracts when RT-PCR was used. It is likely that the RT-PCR is detecting transcripts found in other regions of the hypothalamus (see below) and/or transcripts that have not yet been inactivated or ablated. H37A M2 transcripts could also be detected by RTPCR in the testis and kidney of GHRH-M2 mice. Although GHRH expression has been reported in both of these tissues (34, 35), no transcripts could be found from the hypothalamic promoter in testes (34) and, in this study, hypothalamic transcripts were not detected by RT-PCR in either tissue. In contrast, placental expression of GHRH from the hypothalamic promoter was detected by RT-PCR, but H37AM2 expression was not. Ectopic kidney and testis H37AM2 expression and failure of placental expression shows that even this H37A

Fig. 8. Effect of Repeated GHRH Treatment on Pituitary GH and PRL Contents in GHRH-M2 Transgenic Mice Groups of 5-wk-old female GHRH-M2 mice received injections of 50 ␮g of GHRH or saline sc twice daily for 7 d, after which their pituitaries were removed and assayed for GH and PRL. Data shown are mean ⫾ SEM of six animals. ***, P ⬍ 0.001, vs. saline-injected controls.


38kB promoter cosmid must lack of all the locus control sequences necessary for complete position-independent, tissue-specific GHRH expression in our construct. Similar kidney ectopic expression was also found in some other transgenic lines generated with this rGHRH cosmid (8). Relative to hypothalamic expression, however, testicular or kidney expression of M2 from the hypothalamic promoter is likely to have been at very low levels or in a small subset of cells because only low levels of hypothalamic expression have been found in these tissues (35, 36). Because we were able to routinely breed from male GHRH-M2 transgenic mice, the testicular expression of M2 does not appear to compromise fertility. Crossing the GHRH-M2 mice with GHRH-eGFP mice (8), in which eGFP was driven from the same GHRH transgene promoter, confirmed the results from in situ hybridization and ICC. No GHRH-eGFP cells were visible throughout the entire ARC in the double transgenic progeny compared with the hundreds of cells visible in the GHRH-eGFP mice. Taken together with the GH deficiency and dwarfism, we believe that essentially all of the arcuate hypothalamic GHRH population has been functionally silenced, and most likely ablated (see below), by the expression of the H37AM2 ion channel. In the course of these experiments, a population of fluorescent eGFP-positive cells was seen in singly transgenic GHRH-eGFP mice in the dorsal paraventricular nucleus at the top of the third ventricle (data not shown). These cells persist in doubly transgenic GHRH-eGFP/GHRH-M2 mice, presumably because they are less sensitive to the effects of the M2 transgene expression or M2 protein levels are lower in these cells. The GH cell hypoplasia and pituitary GH deficiency in GHRH-M2 mice confirms the physiological importance of GHRH in proliferation of somatotrophs, and in stimulation of GH gene transcription and release (3). Ghrelin (37), another potential endogenous physiological GH secretagogue, clearly cannot functionally compensate for the loss of GHRH in GHRH-M2 mice. GHRH injections were effective in stimulating GH release in GHRH-M2 mice showing that functional GHRH receptors (38) were maintained in the residual somatotrophs of GHRH-M2 mice despite the chronic lack of GHRH. That GHRP-6 (39) injections were less effective was not surprising because the in vivo responses to GH secretagogues and ghrelin are known to synergize with GHRH (40), and blockade of GHRH by immunoneutralization (41) greatly diminishes the in vivo GH response to GH secretagogues. Continuous short-term GHRH infusions are not very effective in stimulating growth in rodents (42), but do stimulate GH synthesis (2). Similar results were observed with a 7-d continuous sc GHRH treatment of GHRH-M2 mice, which showed no significant effect on growth over this period but doubled pituitary GH content. Clearly, the residual pituitary GH cell population could still respond to the trophic effects of exogenous GHRH. It may be worth noting that if humans


with true GHRH deficiency exist, our results would predict that they might be detected by a poorer response to GH secretagogues than to GHRH itself, unlike GHRH-receptor deficient subjects (43, 44). The GH deficiency in GHRH-M2 mice was not apparent in 1-d-old animals, consistent with the GHRHindependent somatotroph development described in lit/lit mice (18). However, GH deficiency was evident as early as 7 d, well before the growth reduction was apparent, and became progressively more severe. One surprise was the reduction in pituitary PRL in GHRH-M2 mice, which appeared with a delay relative to GH deficiency. A modest reduction would perhaps be expected in view of the marked hypoplasia of GH cells which share a common lineage with PRL cells, and some (7, 16) though not all (45, 46) rodent models with GH deficiency and somatotroph hypoplasia also exhibit some PRL deficiency. Many effects of GHRH may be mediated via the activation of Pit1 (18, 47), but reductions in this transcription factor are unlikely to explain the reduced PRL contents in GHRH-M2 mice per se, because there were no reductions in TSH, which is also Pit1 dependent (47). Because GHRH has little or no direct effect on PRL synthesis or release (48) and normal pituitary PRL content was found in mice with a disruption of the GHRH gene (6), the degree of PRL deficiency in GHRH-M2 mice was surprising. If GHRH is the only hypophysiotropic product of the GHRH neuron and is the exclusive ligand for the GHRH receptor, one would expect the phenotypes of deficiencies in GHRH ligand (GHRH-M2 mice) and its receptor (lit/lit mice) to be similar. However, the effect on PRL in GHRH-M2 mice appears more severe than would be predicted from the relative transcript abundance of GH and PRL in lit/lit mice (49), although good comparative assay data on pituitary PRL and GH protein levels in lit/lit mice are still lacking. GHRH-overexpressing transgenic animals have mild lactotroph hyperplasia but no increase in total pituitary PRL content (50). Perhaps the most telling comparison is the recently reported data from a mouse with a targeted disruption of the GHRH locus that eliminated GHRH expression but not other products of the GHRH gene (6), causing a specific reduction in GH, but not PRL. Because our approach ablates the GHRH neuron, rather than disrupting the GHRH peptide per se, and because GHRH infusions increased GH but not PRL contents (albeit over the short term), we have to consider the possibility that the GHRH peptide is not the sole hypophysiotropic product of the GHRH neuron. In line with this, targeted disruption of the homeobox gene GSH-1, which is required for normal GHRH neuron development, also leads to a similar phenotype of both GH and PRL cell hypoplasia (23). GSH-1 has been shown to be required for GHRH gene expression in the hypothalamus (51), but in this model it is unclear whether the reduction in pituitary PRL content was caused by a disruption of normal GHRH neuron development, or through effects on the many other cell


types normally expressing GSH-1. Our results, from specific ablation of GHRH neurons, are consistent with abnormal GHRH neuron development causing the broader pituitary effects described in animals with disrupted GSH-1. Many other active peptides have been colocalized with GHRH in the ARC, but the more obvious candidates for additional hypophysiotropic activities from this neuron are peptides (GHRH-RP and p75–92NH2) that are coproduced with GHRH as part of the same polypeptide precursor, transported to the median eminence and probably cosecreted with GHRH (52). GHRH-RP has GHRH-independent effects in other tissues (53), and both GHRH-RP and p75–92NH2 activate MAPK (54) in GH3 cells that produce both PRL and GH. This activity would be missing after GHRH neuronal ablation in GHRH-M2 mice, whereas it would be increased in lit/lit mice, because their GHRH gene expression is enhanced in the absence of GH feedback. In this context, it is interesting that PRL levels are unchanged in GHRH-knockout mice recently described by Alba and Salvatori (6) because these mice are still able to produce some GHRH-RP and p75– 92NH2-related products, at least in testis. It will be interesting to know what effect the targeted deletion has had on hypothalamic levels of these GHRH-precursor-derived peptides. Apart from the pituitary phenotype and dwarfism, the GHRH-M2 mice were otherwise apparently healthy, as were the GHRH knockout mice (6). It is therefore curious that no human GHRH mutation has yet been reported because it might be expected to generate a nonlethal but readily apparent autosomal recessive familial form of fertile dwarfism. Complete GH deficiency is not embryonic-lethal in animals or humans, but because the GHRH gene encodes other biologically active peptides (52) and is expressed from different promoters in tissues other than hypothalamus (34), it is conceivable that a global GHRH gene deletion might be embryonic lethal from other causes, whereas a hypothalamic ablation of GHRH cells, only occurring once the cells have differentiated to express GHRH from the hypothalamic promoter, is clearly nonlethal. Because other biologically active peptide products of the GHRH gene may still be normally expressed in GHRH-KO animal, it is possible that GHRH mutations leading to a premature stop codon, which would result in the absence of these peptides, could be embryonic-lethal in humans. The advantage of this ion channel approach in vitro is that the increased conductance that depolarizes and silences excitable cells, can be reversed by exposure to rimantadine which blocks the H37AM2 channels. The ablation strategy using M2 channels was first tested in the developing immune system (11) in which is was an efficient ablator, but the conditional reversibility in vitro did not extend to a successful conditional strategy in vivo. In this study, H37AM2was expressed in rapidly dividing cells early in embryonic development, and treatment with rimantadine in vivo was not suc-


cessful in preventing or reversing the immune cell ablation (11). However, we hoped that driving H37AM2 from a GHRH promoter expressed late in development (55) and in nondividing differentiated neurons, rimantadine treatment might provide some control over cell ablation and excitability in surviving GHRH neurons. This did not prove to be the case: under in vivo conditions, H37AM2 expression irreversibly ablated GHRH cells, and no recovery was seen after rimantadine treatment. The cell ablation by H37AM2 could not be blocked with rimantadine treatment of transgenic animals in utero and through neonatal development, possibly because of inadequate delivery of the drug across placental and mammary barriers.

MATERIALS AND METHODS Constructs The plasmid pEV3/H37A containing H37AM2 was kindly provided by Dr. Alan Hay. A plasmid containing the hGH gene was as previously described (7, 17). From these, a construct was generated in which the sequences encoding all but the first methionine residue of H37AM2 were fused at the 5⬘ end with sequences corresponding to the first exon, intron and initial portion of the second exon of the hGH gene, and at the 3⬘ end with sequences corresponding to the 3⬘-untranslated region (UTR) of hGH (Fig. 1A). This was achieved by ligation of overlapping PCR fragments, generated with primers containing hybrid sequences of hGH and H37AM2 at the fusion sites, to form a construct extending from the BamHI site 61 nucleotides upstream of the ATG to the SspI site 138 nucleotides downstream of the stop codon, the whole flanked with additional MluI sites. The full sequence of this construct is available on request. The resulting MluI cassette provides H37A M2 with 3⬘- and 5⬘-UTRs, an upstream ATG, an intron with splice donor/acceptor sequences, and a downstream polyA addition site, all derived from the hGH sequences we have previously shown to function in the context of transgenes expressed in GHRH and GH cells (7, 8, 16, 17). This cassette would be predicted to generate a protein product of H37A M2 (2–97) with its initial methionine replaced with an additional N-terminal heptapeptide [MATGSRT H37AM2 (2– 97); Fig. 1B]. For simplicity, this will be referred to as H37AM2 in this paper. For in vitro transfection experiments, the H37AM2 cassette was inserted into a modified version of the mammalian CMV expression vector pcDNA 3.1 (Invitrogen, Paisley, UK) to give pcDNA 3.1-M2 (Fig. 1A). In some experiments, this was cotransfected with a CMV vector expressing eGFP (Fig. 1C) (pEGFP-N2, BD Biosciences, Oxford UK) (17). For transgenesis (Fig. 3A), the H37AM2 MluI cassette was cloned into a unique MluI site in the 5⬘ of the first hypothalamic exon of GHRH, in a 38-kb rat cosmid, containing 16 kb 5⬘ and 14-kb 3⬘ flanking sequences, as previously described (7, 8). Cell Culture and Transfection GC cells (17) were cultured in DMEM supplemented with 15% horse serum, 5% fetal calf serum, 1% penicillin-streptomycin, and 1% l-glutamine (Invitrogen) at 37 C under 5% CO2. They were maintained at 30–40% confluency for no more than 18 passages. One day before transfection, cells were plated on poly-SD-lysine-coated glass coverslips in 60-mm cell culture dishes at a density of 3 ⫻ 105 per dish. Cells were transfected with Superfect (QIAGEN, Crawley, UK)

1260

Mol Endocrinol, May 2005, 19(5):1251–1262

according to manufacturer’s instructions, using 1.1 ␮g pcDNA 3.1-M2 plasmid DNA alone or as a mixture with the eGFP plasmid (0.6 ␮g pEGFP-N2 ⫹ 0.5 ␮g pcDNA 3.1-M2). After 2.5 h, the DNA suspension was removed and the cells rinsed with complete growth medium containing 23–100 ␮M rimantadine (Sigma-Aldrich, Gillingham, UK), and maintained in rimantadine-containing medium for 48 h before experimentation. Stably transfected cell lines were selected in G-418 (250 ␮g/ml; Invitrogen) and rimantadine (23 ␮M). Electrophysiology After cotransfection with H37AM2 and eGFP, GC cells were examined under light and fluorescence microscopy. Cells with as few other cell contacts as possible were subjected to whole-cell voltage-clamp via a single patch electrode using an Axopatch 1d amplifier, Digidata 1200 interface and pClamp6 software (Axon Instruments, Union City, CA). Data were low-pass filtered at 2 kHz and sampled at 5 kHz. Borosilicate glass pipettes with tip resistances of 3–6 M⍀ were filled with: Cs gluconate (130 mM), CsCl (15 mM) HEPES (15 mM), EGTA (5 mM), MgCl2 (2 mM), Na-ATP (2 mM), and phosphocreatine (2 mM); adjusted to pH 7.3 using CsOH, with a final solution osmolarity around 315 mOsm. Cells were bathed first in Na-free extracellular recording solution: N-methyl-D-glucamine (NMDG) (138 mM), HCl (130 mM), glucose (20 mM), HEPES (18 mM), CaCl2 (2 mM) and MgCl2 (1 mM), pH 7.3, and 290 mOsm. A fixed prepulse inactivation protocol was applied from a holding potential of ⫺50 mV, and steady-state currents were recorded from the end of the first step (Fig. 2A). Recordings were then repeated after exchanging the extracellular solution for one containing Na⫹: NaCl (130 mM), glucose (20 mM), HEPES (10 mM), NaHEPES (10 mM), CaCl2 (2 mM) and MgCl2 (1 mM), pH 7.3, and 300 mOsm. Single concentrations of rimantadine were applied to each cell via this solution and remaining steady-state currents remeasured as a percentage of control currents. Potentials given are not corrected for a liquid junction potential calculated as ⫺19 mV bath pipette. Generation and Analysis of GHRH-M2 Transgenic Mice All animal experiments were carried out in accordance with our Institutional and National guidelines. The rGHRH-H37AM2 DNA insert was released from the cosmid by digestion with NotI, purified by ultracentrifugation in a 5–20% salt gradient, and brought to a concentration of 1–5 ng/␮l in 10 mM Tris-HCl (pH 7.5), 0.1 mM EDTA (pH 8.0). Transgenic mice were generated by pronuclear microinjection of this construct into fertilized oocytes of superovulated (CBA/Ca ⫻ C57BL/10)F1 mice followed by oviductal transfer into pseudopregnant recipients. Genomic DNA from tail biopsies was amplified by PCR using primers 5⬘-AACCACTCAGGGTCCTGTGGACAG-3⬘ and 5⬘-ATGATGCAACTTAATTTTATTAGGACAA-3⬘, for the hGH 5⬘- and 3⬘-UTR sequences flanking the H37AM2 transgene. All lines were maintained as hemizygous, with NT littermates serving as controls for the transgenic animals. Some GHRH-M2 animals were crossed with animals from another transgenic line expressing eGFP in GH cells (17) or a line expressing eGFP in GHRH neurons (8). From the resulting progeny, pituitaries and brains were fixed, 12-␮m sections were cut and examined for eGFP fluorescence.


gen) in 1⫻ first-strand buffer supplemented with 10 pmol oligo(deoxythymidine)17 0.5 mM deoxynucleotide triphosphates (Amersham Pharmacia Biotech, Chalfont St. Giles, UK), 40 U ribonuclease Inhibitor (Promega) and 5 mM dithiothreitol. The mixture was incubated at 50 C for 45 min, then 55 C for 15 min and the cDNAs amplified by PCR. For the transgene product, the primers were those used for genotyping, to amplify from cDNA, a predicted fragment size of 464 bp. For GHRH, two forward primers were used: 5⬘-GGTCAGTGGGACCTGAGCAG-3⬘ for hypothalamic promoter transcripts and 5⬘-CGCAGGTCTCTCCTGGTTGC-3⬘ for placental promoter transcripts; in both cases the reverse primer was: 5⬘-CTGTCCACATGCTGTCTTCC-3⬘. These would generate predicted fragment sizes of 317 and 316 bp for hypothalamic and placental transcripts, respectively. Mouse GAPDH (glyceraldehyde-3-phosphate dehydrogenase) transcripts were amplified as internal controls. In Situ Hybridization Antisense and sense riboprobes corresponding to mouse GHRH cDNA (Image clone 1496474, see (8) and to H37AM2 transgene transcripts were labeled with either [␣35S]-uridine triphosphate or digoxigenin and in situ hybridizations were performed on cryostat sections prepared and developed as previously described (8, 33). Sections were also obtained from a transgenic mouse with somatotroph-specific expression of an exon-3-deleted isoform of hGH that causes primary pituitary GH deficiency and results in increased GHRH expression (16). ICC GC cells were fixed in 2% paraformaldehyde, permeabilized using Triton X-100 (0.1%; Sigma) and incubated with an antibody (R229/95) that recognizes the C terminus of M2 (courtesy of Dr. Alan Hay). After washing, antibody labeling was visualized with goat antirabbit IgG conjugated to tetramethylrhodamine isothiocyanate (Sigma). The same antibody was used for immunodetection of extracts from both stable and transiently transfected GC cell lines on Western blots, and confirmed the presence of a major protein band corresponding to the H37AM2 protein. ICC was performed for mouse GH and PRL on pituitary sections as previously described (16). For mouse GHRH on hypothalamic sections brains of WT and GHRH-M2 transgenic mice, fixed by perfusion with buffered 4% paraformaldehyde/0.25% glutaraldehyde, were sectioned frozen in the coronal plane at 30 ␮m. Sections from each 180-␮m interval were immunostained using a rabbit polyclonal antiserum directed against mouse GHRH specifically (gift from Dr. F. Talamantes) diluted 1:20,000. Further processing used biotinylated secondary antiserum and avidin-biotin/peroxidase reagents (Vector Laboratories, Burlingame, CA) with reduced diaminobenzidene as brown chromogen (56). RIAs Pituitary tissues were homogenized and assayed for GH, PRL, and TSH using mouse-specific RIA reagents kindly provided by A. L. Parlow and the National Hormone and Pituitary Program, as previously described (16). In Vivo Experiments

RT-PCR RNA was extracted from hypothalamus, testis, placenta, kidney, and spleen using Trizol reagent (Invitrogen), treated with RQ1 ribonuclease-free deoxyribonuclease (Promega, Southampton, UK) for 60 min at 37 C and repurified using Trizol reagent. RNA (1–5 ␮g) was transcribed in a 20-␮l reaction volume with 200 U reverse transcriptase (SuperScriptIII, Invitro-

Body weights and lengths were recorded in age-matched littermates, housed in groups with ad libitum access to food and water. To test pituitary responses to GH secretagogues, groups of 5 transgenic and NT H37AM2 mice were anesthetized with sodium pentobarbital (25 mg/kg ip) a jugular vein catheterized and 50 ␮l blood samples withdrawn before, 5 and 15 min after iv injection of 500 ng GHRP-6 (Ferring AB,


Malmo, Sweden) After 90 min, further blood samples were withdrawn before and after iv injection of 100 ng GHRH (human GHRH 27Nle(1–29)NH2; Bachem, Merseyside, UK). Plasma was obtained by centrifugation and assayed for mouse GH. In another experiment, groups of GHRH-M2 transgenic and NT mice were injected twice daily with 50 ␮g GHRH sc for 7 d, their weights recorded and their pituitary GH and PRL contents measured. Rimantadine (1 mg/ml) was added to the drinking water of other groups of GHRH-M2 mice or their mothers at various ages.


8.

9.

10. Data Analysis Unless otherwise stated, data are shown as mean ⫾ SEM. Differences between groups were analyzed by Student’s t test, with P ⬍ 0.05 taken as significant.

11.

Acknowledgments

12.

We are grateful to Dr. A. F. Parlow and the NIDDK for supply of RIA reagents, to Ferring AB for GHRP-6, to Dr. Teddy Fauquier for imaging of GHRH-eGFP cells, and to Drs. Brian Thomas and Alan Hay for M2 reagents, advice and discussion. We are also grateful to the staff of Biological Services, National Institute for Medical Research, for their excellent assistance in animal maintenance.

13.

Received June 1, 2004. Accepted January 11, 2005. Address all correspondence and requests for reprints to: Professor Iain C. A. F. Robinson, Division of Molecular Neuroendocrinology, National Institute for Medical Research, The Ridgeway, Mill Hill, London NW7 1AA, United Kingdom. Email: [email protected]. Current addresses for C.M.: Department of Neurosurgery, Barts and The London School of Medicine and Dentistry, Queen Mary, University of London, London E14NS, United Kingdom. Current addresses for P.H.: Molecular and Cellular Neuroscience, Imperial College, Hammersmith Hospital, Du Cane Road, London W6 0NN, United Kingdom.

REFERENCES 1. Mayo KE, Godfrey PA, Suhr ST, Kulik DJ, Rahal JO 1995 Growth hormone-releasing hormone—synthesis and signaling. Recent Prog Horm Res 50:35–73 2. Barinaga M, Bilezikjian LM, Vale WW, Rosenfeld MG, Evans RM 1985 Independent effects of growth-hormone releasing-factor on hormone release and gene-transcription. Nature 314:279–281 3. Mayo KE, Miller T, DeAlmeida V, Godfrey P, Zheng J, Cunha SR 2000 Regulation of the pituitary somatotroph cell by GHRH and its receptor. Recent Prog Horm Res 55:237–266 4. Hammer RE, Brinster RL, Rosenfeld MG, Evans RM, Mayo KE 1985 Expression of human growth hormonereleasing factor in transgenic mice results in increased somatic growth. Nature 315:413–416 5. Rivier J, Spiess J, Thorner M, Vale W 1982 Characterization of a growth hormone-releasing factor from a human pancreatic islet tumour. Nature 300:276–278 6. Alba M, Salvatori R 2004 A mouse with targeted ablation of the GHRH gene: a new model of isolated GH deficiency. Endocrinology 145:4134–4143 7. Flavell DM, Wells T, Wells SE, Carmignac DF, Thomas GB, Robinson ICAF 1996 Dominant dwarfism in transgenic rats by targeting human growth hormone (GH)

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

expression to hypothalamic GH-releasing factor neurons. EMBO J 15:3871–3879 Balthasar N, Mery PF, Magoulas CB, Mathers KE, Martin A, Mollard P, Robinson IC 2003 Growth hormone-releasing hormone (GHRH) neurons in GHRH-enhanced green fluorescent protein transgenic mice: a ventral hypothalamic network. Endocrinology 144:2728–2740 Johns DC, Marx R, Mains RE, O’Rourke B, Marban E 1999 Inducible genetic suppression of neuronal excitability. J Neurosci 19:1691–1697 Slimko EM, McKinney S, Anderson DJ, Davidson N, Lester HA 2002 Selective electrical silencing of mammalian neurons in vitro by the use of invertebrate ligandgated chloride channels. J Neurosci 22:7373–7379 Smith CA, Graham CM, Mathers K, Skinner A, Hay AJ, Schroeder C, Thomas DB 2002 Conditional ablation of T-cell development by a novel viral ion channel transgene. Immunology 105:306–313 Chizhmakov IV, Geraghty FM, Ogden DC, Hayhurst A, Antoniou M, Hay AJ 1996 Selective proton permeability and pH regulation of the influenza virus M2 channel expressed in mouse erythroleukaemia cells. J Physiol 494:329–336 Ogden D, Chizhmakov IV, Geraghty FM, Hay AJ 1999 Virus ion channels. Methods Enzymol 294:490–506 Tang Y, Zaitseva F, Lamb RA, Pinto LH 2002 The gate of the influenza virus M2 proton channel is formed by a single tryptophan residue. J Biol Chem 277:39880– 39886 Le Tissier P, Magoulas B, Mathers K, McGuinness L, Sesay AK, Carmignac DF, Robinson ICAF. Ablating GRF neurones in transgenic mice using a novel viral ion channel transgene. Program of the 84th Annual Meeting of The Endocrine Society, San Francisco, CA, 2002, p 106 (Abstract OR31-1). McGuinness L, Magoulas C, Sesay AK, Mathers K, Carmignac D, Manneville JB, Christian H, Phillips 3rd JA, Robinson IC 2003 Autosomal dominant growth hormone deficiency disrupts secretory vesicles in vitro and in vivo in transgenic mice. Endocrinology 144:720–731 Magoulas C, McGuinness L, Balthasar N, Carmignac DF, Sesay AK, Mathers KE, Christian H, Candeil L, Bonnefont X, Mollard P, Robinson IC 2000 A secreted fluorescent reporter targeted to pituitary growth hormone cells in transgenic mice. Endocrinology 141:4681–4689 Lin SC, Lin C, Gukovsky I, Lusis AJ, Sawchenko PE, Rosenfeld MG 1993 Molecular-basis of the little mouse phenotype and implications for cell-type-specific growth. Nature 364:208–213 Godfrey P, Rahal J, Beamer W, Copeland N, Jenkins N, Mayo K 1993 GHRH receptor of little mice contains a missense mutation in the extracellular domain that disrupts receptor function. Nature Genet 4:227–232 Maheshwari H, Silverman B, JD, GB 1998 Phenotype and genetic analysis of a syndrome caused by an inactivating mutation in the growth hormone-releasing hormone receptor: dwarfism of Sindh. J Clin Endocrinol Metab 83:4065–4074 Wehrenberg WB, Brazeau P, Luben R, Ling N, Guillemin R 1983 A noninvasive functional lesion of the hypothalamo-pituitary axis for the study of growth hormonereleasing factor. Neuroendocrinology 36:489–491 Bloch B, Ling N, Benoit R, Wehrenberg WB, Guillemin R 1984 Specific depletion of immunoreactive growth hormone-releasing factor by monosodium glutamate in rat median eminence. Nature 307:272–273 Li H, Zeitler PS, Valerius MT, Small K, Potter SS 1996 Gsh-1, an orphan Hox gene, is required for normal pituitary development. EMBO J 15:714–724 Zhu X, Zhou A, Dey A, Norrbom C, Carroll R, Zhang C, Laurent V, Lindberg I, Ugleholdt R, Holst JJ, Steiner DF 2002 Disruption of PC1/3 expression in mice causes


25.

26.

27.

28.

29. 30. 31. 32.

33.

34.

35.

36. 37. 38.

39.

40.

41.

42.

dwarfism and multiple neuroendocrine peptide processing defects. Proc Natl Acad Sci USA 99:10293–10298 Hollingshead PG, Martin L, Pitts SL, Stewart TA 1989 A dominant phenocopy of hypopituitarism in transgenic mice resulting from central nervous system synthesis of human growth hormone. Endocrinology 125:1556–1564 White BH, Osterwalder TP, Yoon KS, Joiner WJ, Whim MD, Kaczmarek LK, Keshishian H 2001 Targeted attenuation of electrical activity in Drosophila using a genetically modified K(⫹) channel. Neuron 31:699–711 Koster JC, Remedi MS, Flagg TP, Johnson JD, Markova KP, Marshall BA, Nichols CG 2002 Hyperinsulinism induced by targeted suppression of ␤ cell KATP channels. Proc Natl Acad Sci USA 99:16992–16997 Taylor MS, Bonev AD, Gross TP, Eckman DM, Brayden JE, Bond CT, Adelman JP, Nelson MT 2003 Altered expression of small-conductance Ca2⫹-activated K⫹ (SK3) channels modulates arterial tone and blood pressure. Circ Res 93:124–131 Pinto LH, Holsinger LJ, Lamb RA 1992 Influenza virus M2 protein has ion channel activity. Cell 69:517–528 Sansom MS, Kerr ID 1993 Influenza virus M2 protein: a molecular modelling study of the ion channel. Protein Eng 6:65–74 Wang C, Lamb RA, Pinto LH 1994 Direct measurement of the influenza A virus M2 protein ion channel activity in mammalian cells. Virology 205:133–140 Chizhmakov IV, Ogden DC, Geraghty FM, Hayhurst A, Skinner A, Betakova T, Hay AJ 2003 Differences in conductance of M2 proton channels of two influenza viruses at low and high pH. J Physiol 546:427–438 Bennett PA, Levy A, Sophokleous S, Robinson IC, Lightman SL 1995 Hypothalamic GH receptor gene expression in the rat: effects of altered GH status. J Endocrinol 147:225–234 Srivastava CH, Monts BS, Rothrock JK, Peredo MJ, Pescovitz OH 1995 Presence of a spermatogenic-specific promoter in the rat growth hormone-releasing hormone gene. Endocrinology 136:1502–1508 Matsubara S, Sato M, Mizobuchi M, Niimi M, Takahara J 1995 Differential gene expression of growth hormone (GH)-releasing hormone (GRH) and GRH receptor in various rat tissues. Endocrinology 136:4147–4150 Berry SA, Pescovitz OH 1988 Identification of a rat GHRH-like substance and its messenger RNA in rat testis. Endocrinology 123:661–663 Kojima M, Hosoda H, Date Y, Nakazato M, Matsuo H, Kangawa K 1999 Ghrelin is a growth-hormone-releasing acylated peptide from stomach. Nature 402:656–660 Mayo K, Miller T, DeAlmeida V, Zheng J, Godfrey P 1996 The growth-hormone-releasing hormone receptor: signal transduction, gene expression, and physiological function in growth regulation. Ann NY Acad Sci 26:184–203 Bowers CY, Momany FA, Reynolds GA, Hong A 1984 On the in vitro and in vivo activity of a new synthetic peptide that acts on the pituitary to specifically release growth hormone. Endocrinology 114:1537–1545 Bowers CY, Reynolds GA, Durham D, Barrera CM, Pezzoli SS, Thorner MO 1990 Growth hormone (GH)-releasing peptide stimulates GH release in normal men and acts synergistically with GH-releasing hormone. J Clin Endocrinol Metab 70:975–982 Clark RG, Carlsson LMS, Trojnar J, Robinson ICAF 1989 The effects of a growth hormone-releasing peptide and growth hormone releasing factor in conscious and anaesthetized rats. J Neuroendocrinol 1:249–255 Clark RG, Robinson IC 1985 Growth induced by pulsatile infusion of an amidated fragment of human growth hor-


43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

mone releasing factor in normal and GHRF-deficient rats. Nature 314:281–283 Gondo RG, Aguiar-Oliveira MH, Hayashida CY, Toledo SP, Abelin N, Levine MA, Bowers CY, Souza AH, Pereira RM, Santos NL, Salvatori R 2001 Growth hormone-releasing peptide-2 stimulates GH secretion in GH-deficient patients with mutated GH-releasing hormone receptor. J Clin Endocrinol Metab 86:3279–3283 Roelfsema F, Biermasz NR, Veldman RG, Veldhuis JD, Frolich M, Stokvis-Brantsma WH, Wit JM 2001 Growth hormone (GH) secretion in patients with an inactivating defect of the GH-releasing hormone (GHRH) receptor is pulsatile: evidence for a role for non-GHRH inputs into the generation of GH pulses. J Clin Endocrinol Metab 86:2459–2464 Struthers RS, Vale WW, Arias C, Sawchenko PE, Montminy MR 1991 Somatotroph hypoplasia and dwarfism in transgenic mice expressing a non-phosphorylatable CREB mutant. Nature 350:622–624 Tierney T, Robinson IC 2002 Increased lactotrophs despite decreased somatotrophs in the dwarf (dw/dw) rat: a defect in the regulation of lactotroph/somatotroph cell fate? J Endocrinol 175:435–446 Li S, Crenshaw 3rd EB, Rawson EJ, Simmons DM, Swanson LW, Rosenfeld MG 1990 Dwarf locus mutants lacking three pituitary cell types result from mutations in the POU-domain gene pit-1. Nature 347:528–533 Moore JR JP, Cai A, Hostettler ME, Arbogast LA, Voogt JL, Hyde JF 2000 Pituitary hormone gene expression and secretion in human growth hormone-releasing hormone transgenic mice: focus on lactotroph function. Endocrinology 141:81–90 Wojtkiewicz PW, Phelps CJ, Hurley DL 2002 Transcript abundance in mouse pituitaries with altered growth hormone expression quantified by reverse transcriptase polymerase chain reaction implicates transcription factor Zn-16 in gene regulation in vivo. Endocrine 18:67–74 Moore Jr JP, Cai A, Maley BE, Jennes L, Hyde JF 1999 Galanin within the normal and hyperplastic anterior pituitary gland: localization, secretion, and functional analysis in normal and human growth hormone-releasing hormone transgenic mice. Endocrinology 140:1789–1799 Mutsuga N, Iwasaki Y, Morishita M, Nomura A, Yamamori E, Yoshida M, Asai M, Ozaki N, Kambe F, Seo H, Oiso Y, Saito H 2001 Homeobox protein Gsh-1-dependent regulation of the rat GHRH gene promoter. Mol Endocrinol 15:2149–2156 Nillni EA, Steinmetz R, Pescovitz OH 1999 Posttranslational processing of progrowth hormone-releasing hormone. Endocrinology 140:5817–5827 Steinmetz R, Lazzaro N, Rothrock JK, Pescovitz OH 2000 Effects of growth hormone-releasing hormone-related peptide on stem cell factor expression in cultured rat Sertoli cells. Endocrine 12:323–327 Steinmetz R, Zeng P, King DW, Walvoord E, Pescovitz OH 2001 Peptides derived from pro-growth hormonereleasing hormone activate p38 mitogen-activated protein kinase in GH3 pituitary cells. Endocrine 15:119–127 Ishikawa K, Katakami H, Jansson JO, Frohman LA 1986 Ontogenesis of growth hormone-releasing hormone neurons in the rat hypothalamus. Neuroendocrinology 43: 537–542 Phelps CJ, Romero MI, Hurley DL 2003 Prolactin replacement must be continuous and initiated prior to 21 d of age to maintain hypothalamic dopaminergic neurons in hypopituitary mice. Endocrine 20:139–148

Molecular Endocrinology is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.