HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Full Text (PDF) All Versions of this Article: 19/12/1680 04-3434fjev1    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Bewick, G. A. Articles by Bloom, S. R. Search for Related Content PubMed PubMed Citation Articles by Bewick, G. A. Articles by Bloom, S. R.

Post-embryonic ablation of AgRP neurons in mice leads to a lean, hypophagic phenotype

Gavin A. Bewick^*, James V. Gardiner^*, Waljit S. Dhillo^*, Aysha S. Kent^*, Nicholas E. White^*, Zoe Webster, Mohammad A. Ghatei^* and Stephen R. Bloom^*,1

^* Department of Metabolic Medicine, Faculty of Medicine, Imperial College London, London, UK; and
MRC Core Transgenic Facility, Clinical Sciences Centre, Hammersmith Hospital, London, UK

¹ Correspondence: Department of Metabolic Medicine, Faculty of Medicine, Imperial College London, 6th Floor, Commonwealth Building, Hammersmith Hospital, Du Cane Road, London W12 ONN, UK. E-mail: s.bloom{at}imperial.ac.uk

SPECIFIC AIMS

Both agouti-related protein (AgRP) and neuropeptide Y (NPY) stimulate food intake when administered into the third ventricle and are up-regulated in states of negative energy balance. However, mice with targeted deletion of either NPY or AgRP or both do not have major alterations in energy homeostasis. To address the lack of phenotype in the knockout models, we chose an approach aimed to selectively ablate the AgRP neuron. This was achieved using bacterial artificial chromosome (BAC) transgenesis to target expression of a neurotoxic CAG expanded form of ataxin-3 to AgRP expressing neurons in the arcuate. The resulting partial postnatal loss of AgRP expressing neurons was investigated.

PRINCIPAL FINDINGS

1. Transgene expression and AgRP neuronal loss
Brains from 3-wk-old mice, examined by dual in situ hybridization using a digoxigenin-labeled AgRP riboprobe and a radio-labeled ataxin-3 riboprobe, revealed that ataxin-3 expression in transgenic (Tg) mice was colocalized in AgRP-expressing neurons. This suggests that BAC-driven expression of the neurotoxic CAG- expanded ataxin-3 transgene was successfully directed to AgRP-expressing neurons. However, the ataxin-3 transgene did not appear to be expressed in every AgRP-containing neuron, indicating that transgene expression was either below the limit of detection in these cells or that transgene expression had been silenced in these neurons.

At 3 wk of age we found a significant 20% loss of AgRP-containing neurons in Tg mice compared with controls (473±20 cells WT vs. 379±29 cells Tg, n=4, P<0.05). By 16 wk of age a further (47%) loss of AgRP-containing neurons was observed in the arcuate of Tg mice compared with controls (699±66 cells WT vs. 368±78 cells Tg, n=5, P<0.05, Fig. 1 A, B, E). Ataxin-3 expression was undetectable in the arcuate of either 7- or 16-wk-old Tg mice (data not shown). These observations suggested the transgene was not expressed or silenced in half of the AgRP neurons while the remaining transgene positive AgRP neurons exhibited a progressive postnatal degeneration until 7 wk of age. Proopiomelanocortin (POMC) neuronal numbers were counted and found to be identical in WT and Tg mice (663±43 cells WT vs. 602±26 cells Tg, n=4, P=0.27, Fig. 1C-E ) suggesting cell loss was likely to be specific for AgRP neurons with no bystander effect.

View larger version (72K): [in this window] [in a new window]   Figure 1. AgRP neuronal loss at 16 wk of age. A, B) Representative photomicrographs of AgRP expression in the arcuate nucleus of WT (A) and Tg (B) mice illustrating ablation of 47% of AgRP neurons in Tg mice (699±66 cells WT vs. 368±78 cells Tg n=5, *P<0.05)). C, D) Representative photomicrographs of POMC expression in the arcuate nucleus of WT (C) and Tg (D) mice demonstrating similar POMC cell numbers in the arcuate of Tg and WT mice (663±43 cells WT vs. 602±26 cells Tg, n=4, P=0.27). E) Histograms of average total bilateral AgRP and POMC cell numbers from 16 hypothalamic-matched sections in WT (dark gray bars) and Tg mice (light gray bars).

2. Release of hypothalamic neuropeptides involved in appetite regulation
To investigate the function of the remaining AgRP/NPY neurons and the consequences of AgRP/NPY neuronal cell loss on -melanocyte-stimulating hormone (-MSH)/cocaine amphetamine regulated transcript (CART) neuronal function, release of AgRP-IR, NPY-IR, -MSH-IR and CART-IR from ex vivo hypothalamic slices from weight-matched WT and Tg mice was measured. Release of AgRP-IR was significantly reduced by 20% in hypothalamic explants from Tg mice (WT 1.72±0.13 fmol/explant (n=7) vs. Tg 1.37±0.08 fmol/explant (n=12), P<0.05). There was also a 33% reduction in NPY-IR release in Tg mice (WT 32.3±3.7 fmol/explant (n=8) vs. Tg 22.3±3.1 fmol/explant (n=12), P=0.054). Release of both -MSH-IR (WT 10.67±1.04 fmol/explant (n=7) vs. Tg 10.78±1.14 fmol/explant (n=12), P=0.9) and CART-IR (WT 428±53.5 fmol/explant (n=8) vs. Tg 392±66.7 fmol/explant (n=12), P=0.7,) was the same in WT and Tg mice.

3. Body weight and food intake in transgenic mice
Transgenic mice were significantly lighter than their age-matched wild-type controls at weaning (3 wk of age), WT 10.5 ± 0.5 g (n=9) vs. Tg 7.4 ± 0.4 g (n=14) P < 0.001. By 16 wk of age, wild-type mice were 34.7 ± 0.7 g while Tg mice were 28.6 ± 0.6 g (P<0.001, Fig. 2 A).

Food intake was significantly reduced in Tg mice. Tg mice between the ages of 4 and 6 wk ate on average 16% less per day than controls (WT 4.2±0.05 g (n=9) vs. Tg 3.5±0.1 g (n=14), P<0.0005). The difference in food intake was further increased at 14 to 16 wk of age when the difference in average daily food intake was 28% (5.0±0.24 g WT vs. 3.6±0.12 g Tg, P<0.0005, Fig. 2B ). To investigate whether the difference in food intake observed between age-matched control and Tg mice might be the result of differences in body weight between the 2 groups, average daily food intake was analyzed over a 2 wk period in mice aged 14–16 wk of age and adjusted for body weight. When food intake was adjusted for body weight, average daily food intake in Tg mice was significantly reduced by 16% compared with controls (WT 0.149±0.01 g/g of body weight vs. Tg 0.126±0.005 g/g of body weight, P<0.01, Fig. 2C ). Similar effects were seen in females and 2 other independent transgenic lines (data not shown).

View larger version (20K): [in this window] [in a new window]   Figure 2. Body weight and food intake (A), Tg mice (open triangles) have significantly lower body weight compared with WT (closed circle) controls (closed circle). B) Tg mice (light gray bars) have a significantly lower average daily food intake compared with WT mice (dark gray bars) at 14–16 wk of age. *P<0.0005. C) average daily food intake adjusted for body weight is significantly lower in Tg mice (light gray bars) compared with WT controls (dark gray bars) at 14–16 wk of age. *P<0.01.

4. Body composition
Total body fat was reduced by 53% in Tg mice compared with WT mice (WT 11.2±1.6 g (n=3) vs. Tg 5.28±0.36 g (n=4), P<0.05). However, lean mass was identical between Tg and WT mice (WT 20.9±0.99 g (n=3) vs. Tg 20.2±2.18 g (n=4), P=0.77).

5. Circulating leptin, insulin, and glucose levels
Circulating leptin was 25% lower in the Tg mice (WT 4.1±0.6ng/ml vs. Tg 3.1±0.6ng/ml, n=4, P=0.075). There was a significant reduction of 53% in circulating insulin levels in Tg mice compared with WT (WT 1.02±0.13ng/ml vs. Tg 0.48±0.14ng/ml, n=4, P<0.05) but no difference in circulating glucose between the 2 groups (WT 12.9±1.4mmol/l (n=11) vs. Tg 11.2±1.1mmol/l (n=9), P=0.39).

6. Uncoupling protein 1 mRNA levels in brown adipose tissue
UCP1 mRNA expression in BAT was increased in Tg mice to 146% compared with WT mice (n=7), P<0.05).

7. Effect of Ghrelin and PYY3-36 in transgenic mice
Ghrelin significantly increased food intake 2 h postinjection in ad libitum fed WT mice (0–2 h food intake in saline-treated WT mice 0.06±0.03 g vs. ghrelin-treated WT mice 0.39±0.06 g n=4, P<0.01). Ghrelin failed to stimulate food intake in ad libitum fed Tg mice (saline-treated Tg mice 0.08±0.02 g vs. ghrelin-treated Tg mice 0.10±0.02 g, P=0.46).

PYY 3–36 significantly reduced food intake 2 h postinjection in both WT and Tg mice fasted for 24 h. Food intake in saline-treated WT mice was 2.03 ± 0.15 g and when treated with PYY 3-36 was 1.55 ± 0.13 g (P<0.05), a 24% reduction in food intake. Food intake in Tg mice was 1.49±0.18 g when treated with saline and 1.09 ± 0.17 g when given PYY 3-36 (P<0.05), a 27% reduction in food intake.

CONCLUSIONS AND SIGNIFICANCE

A BAC clone was used to target expression of a neurotoxic CAG-expanded ataxin-3 transgene in arcuate AgRP/NPY neurons. Expression of the transgene in AgRP/NPY neurons resulted in the loss of 47% of these neurons by 16 wk of age while POMC neuron numbers were unaffected. Functionally, the loss of 47% of AgRP/NPY neurons in Tg mice resulted in a reduction in the release of AgRP-IR and NPY-IR from hypothalamic explants, by 20% and 33%, respectively.

Transgenic mice displayed a 53% (5.88 g) reduction in total body fat at 16 wk of age. The reduced adiposity in Tg mice is consistent with the reduction in food intake and an increase in energy expenditure indicated by the increase in UCP1. This reduced body fat accounts for the difference in body weight between WT and Tg mice, since Tg mice are 6.0 g lighter than WT mice but have identical lean mass at 16 wk of age. The reduction in adiposity in Tg mice was associated with the expected lowering of circulating plasma leptin levels and improved insulin sensitivity as shown by their reduced plasma insulin levels in the presence of normal circulating glucose levels.

The changes in food intake and body weight observed in the Tg mice are in contrast to the normal energy regulation in mice with targeted deletion of AgRP, NPY, or both and could be due to several reasons. Recent data suggest that postnatal days 4–12 are a critical time during which the hypothalamus is highly plastic, particularly to the trophic actions of leptin. This plasticity is lost at later stages of development. Our data suggest that at least half the ablated AgRP neurons are lost postweaning at 3 wk of age. Therefore, loss of at least half of the ablated AgRP neurons in our Tg mice occurs after the critical period during which compensatory changes in neuronal connections take place. Second, loss of AgRP/NPY neurons may also precipitate the loss of another unknown colocalized orexigenic factor that would still be present in mice with targeted deletion of AgRP, NPY or both.

Transgenic mice that have a 47% reduction in AgRP/NPY neurons failed to respond to peripherally administered ghrelin. The same effect was found in mice with targeted deletion of both AgRP and NPY but not those with deletion of either alone. However, Tg mice retain their sensitivity to PYY3-36. A possible explanation for these results is that the remaining AgRP/NPY neurons in Tg mice may be maximally active in an attempt to compensate for their reduced number. In this situation ghrelin, which stimulates these neurons, will be unable to increase their activity further, therefore failing to increase food intake. In contrast, PYY 3-36, which acts by inhibiting AgRP/NPY neuronal function, would still be able to inhibit activity of these neurons, resulting in decreased food intake.

In summary, a reduction in the number of arcuate AgRP/NPY neurons after embryogenesis in mice leads to a lean, hypophagic phenotype and suggests that the AgRP/NPY neuron has a critical role in regulating appetite and body weight.

View larger version (26K): [in this window] [in a new window]   Figure 3. Flow diagram outlining the main consequences of AgRP/NPY neuronal loss. Loss of AgRP/NPY neurons attenuates the orexigenic tone from this population of neurons, causing a reduction in food intake. Reduced food intake lowers body weight by reducing fat mass and in turn improves insulin sensitivity. Loss of AgRP/NPY neurons leads to a reduced inhibitory tone to brown adipose tissue (BAT) causing an increase in uncoupling protein 1 (UCP1) expression in BAT, a marker of increased energy expenditure. The remaining AgRP/NPY neurons are hypothesized to be maximally active to compensate for their reduced number. In this situation, ghrelin may not further increase AgRP/NPY neuronal activity, preventing an increase in food intake. However, PYY 3-36 would still be able to inhibit these neurons thus reducing food intake.

FOOTNOTES

To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.04-3434fje; doi: 10.1096/fj.04-3434fje

This article has been cited by other articles:

				J. A. Reed, S. C. Benoit, P. T. Pfluger, M. H. Tschop, D. A. D'Alessio, and R. J. Seeley Mice with chronically increased circulating ghrelin develop age-related glucose intolerance Am J Physiol Endocrinol Metab, April 1, 2008; 294(4): E752 - E760. [Abstract] [Full Text] [PDF]

				W. Fujimoto, T. Shiuchi, T. Miki, Y. Minokoshi, Y. Takahashi, A. Takeuchi, K. Kimura, M. Saito, T. Iwanaga, and S. Seino Dmbx1 is essential in agouti-related protein action PNAS, September 25, 2007; 104(39): 15514 - 15519. [Abstract] [Full Text] [PDF]

				A. Voss-Andreae, J. G. Murphy, K. L. J. Ellacott, R. C. Stuart, E. A. Nillni, R. D. Cone, and W. Fan Role of the Central Melanocortin Circuitry in Adaptive Thermogenesis of Brown Adipose Tissue Endocrinology, April 1, 2007; 148(4): 1550 - 1560. [Abstract] [Full Text] [PDF]

				H. Munzberg, E. E. Jobst, S. H. Bates, J. Jones, E. Villanueva, R. Leshan, M. Bjornholm, J. Elmquist, M. Sleeman, M. A. Cowley, et al. Appropriate Inhibition of Orexigenic Hypothalamic Arcuate Nucleus Neurons Independently of Leptin Receptor/STAT3 Signaling J. Neurosci., January 3, 2007; 27(1): 69 - 74. [Abstract] [Full Text] [PDF]

				K. G. Murphy, W. S. Dhillo, and S. R. Bloom Gut Peptides in the Regulation of Food Intake and Energy Homeostasis Endocr. Rev., December 1, 2006; 27(7): 719 - 727. [Abstract] [Full Text] [PDF]

				D. Titolo, F. Cai, and D. D. Belsham Coordinate Regulation of Neuropeptide Y and Agouti-Related Peptide Gene Expression by Estrogen Depends on the Ratio of Estrogen Receptor (ER) {alpha} to ER{beta} in Clonal Hypothalamic Neurons Mol. Endocrinol., September 1, 2006; 20(9): 2080 - 2092. [Abstract] [Full Text] [PDF]

				R. W. Gelling Diabetic Hyperphagia--Ghrelin in the Driver's Seat. Endocrinology, June 1, 2006; 147(6): 2631 - 2633. [Full Text] [PDF]

				G. M. Sutton, J. L. Trevaskis, M. W. Hulver, R. P. McMillan, N. J. Markward, M. J. Babin, E. A. Meyer, and A. A. Butler Diet-Genotype Interactions in the Development of the Obese, Insulin-Resistant Phenotype of C57BL/6J Mice Lacking Melanocortin-3 or -4 Receptors Endocrinology, May 1, 2006; 147(5): 2183 - 2196. [Abstract] [Full Text] [PDF]

				S. Luquet, F. A. Perez, T. S. Hnasko, and R. D. Palmiter NPY/AgRP Neurons Are Essential for Feeding in Adult Mice but Can Be Ablated in Neonates Science, October 28, 2005; 310(5748): 683 - 685. [Abstract] [Full Text] [PDF]

This Article Full Text (PDF) All Versions of this Article: 19/12/1680 04-3434fjev1    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Bewick, G. A. Articles by Bloom, S. R. Search for Related Content PubMed PubMed Citation Articles by Bewick, G. A. Articles by Bloom, S. R.