Evolution of appetite regulating systems

All unapproved Citizendium articles may contain errors of fact, bias, grammar etc. A version of an article is unapproved unless it is marked as citable with a dedicated green template at the top of the page, as in this version of the 'Biology' article. Citable articles are intended to be of reasonably high quality. The participants in the Citizendium project make no representations about the reliability of Citizendium articles or, generally, their suitability for any purpose. This article is currently being developed as part of an Eduzendium student project in the framework of a course entitled Appetite and Obesity at University of Edinburgh. The course homepage can be found at CZ:(U00984) Appetite and Obesity, University of Edinburgh 2010. For the course duration, the article is closed to outside editing. Of course you can always leave comments on the discussion page. The anticipated date of course completion is 01 February 2011. One month after that date at the latest, this notice shall be removed. Besides, many other Citizendium articles welcome your collaboration!

Recently, there has been extensive research into the neuroendocrine mechanisms controlling appetite. The pro-opiomelanocortin (POMC) gene has an important role in these mechanisms, particularly through production of alpha melanocyte stimulating hormone (α-MSH). POMC and its end-products have not only been identified in humans, but also in many other vertebrates. This has led to further research into the origins of the POMC gene and the evolution of appetite regulating systems. This article details the structure and function of the POMC gene. It highlights variations between species, allowing a potential evolutionary route, originating at a common ancestral gene, to be mapped out.

Human POMC

Fig. 1 POMC and its post-translational processing, adapted from Millington, and Raffin-Sanson et al.

The human POMC gene encodes a hormone precursor protein, which is then cleaved by prohormone convertase enzymes into a number of different peptides. These include the melanocyte-stimulating hormones (alpha-, beta-, gamma- MSH), adrenocorticotropic hormone (ACTH), the lipotropins, and beta-endorphin . ACTH and the MSHs are referred to as the melanocortins, and all have the same core amino acid sequence, HFRW .

The human POMC gene is on chromosome 2p23, and has three exons and two “large” introns. Only exons two and three are translated; exon two codes for the signal peptide and the initial N-terminal amino acids, while exon three codes for “most of the translated mRNA” .

After excision of the introns to form a “parent” POMC, this molecule is then cleaved into its various peptides, as mentioned above, by PCs, specifically PC1 and PC2. These enzymes act at cleavage sites consisting of paired basic residues, arginine and lysine, and end-products of their action depend on which sites are used. As PC1 and PC2 act on different sites, and their expression varies in different tissues, processing of POMC peptides is tissue-specific . For example, the anterior pituitary corticotroph cells only express PC1, which results in the cleavage of POMC into the NH2-terminal peptide (N-term), joining peptide (JP), ACTH, β-LPH, and some γ-LPH and β-endorphin (β-end)(see Fig. 2). The latter two peptides are produced because the “last cleavage site is only partially used”. However, in melanotroph cells, found in the intermediate lobe of the rodent pituitary, and the human hypothalamus, and placenta, both PC1 and PC2 are expressed. This means that all the cleavage sites are used and smaller peptides are produced ). N-term is therefore cleaved to the γ-MSHs, ACTH gives rise to α-MSH and CLIP (corticotropin-like intermediate lobe peptide) and γ-LPH to β-MSH.

Fig. 2 The cleavage of POMC by PC1, adapted from Raffin-Sanson et al.

The POMC gene is expressed in number of tissues in man and animals. These include the anterior, and intermediate (only in rodents) lobes of the pituitary gland. It is also present in the nucleus tractus solitarii of the caudal medulla, and the hypothalamus, specifically the arcuate nucleus. Its expression has also been noted in the skin and the immune system,as well as other peripheral tissues.

The final products from the cleavage of POMC have a variety of functions. The melanocortins act on melanocortin receptors (MCRs) in different tissues. There are five types of MCR, and different melanocortin peptides bind to these with different affinities , for example, ACTH binds mainly to MC2R (see Fig. 3).

Fig. 3 The Melanocortin Receptors, adapted from Yang et al.

β-endorphin, on the other hand, is involved in pain processing as an inhibitory neurotransmitter in the central nervous system. Here it binds to opioid receptors, leading to an analgesic effect

CLIP is found in many areas of the brain, especially in nerve fibres, and has an important role in REM sleep, which in turn helps with the consolidation of memories .

The functions of β- and γ-LPH are still uncertain, although lipotropins are known for their role in lipolysis, mobilizing lipids for energy production, and they are important in haematopoiesis.

POMC, the Hypothalamus and Appetite Regulation

In mammals, alpha MSH is generally assumed to be the main POMC product involved in appetite regulation. Alpha MSH is a potent inhibitor of appetite (anorexigenic), acting on MC3-R and MC4-R in the arcuate nucleus. In POMC null mice, alpha-MSH was found to be the most potent anorexigenic signaller.

Rodents do not express beta-MSH, but evidence has been found indicating an important role for beta-MSH in human appetite regulation. A study screening for mutations in the beta-MSH region of POMC found an increased incidence of the beta-MSH variant Try221Cys in obese subjects. The variant was shown to have altered binding and signalling through MC4-R.

Desacetyl-alpha-MSH is a precursor for alpha-MSH and is widely distributed in the brain. Although desacetyl-alpha-MSH displays a similar potency for receptor binding to alpha-MSH, it does not affect food intake except at extremely high doses.

Fig. 4 Hypothalamic Pathways involved in Appetite Regulation; adapted from Millington GWM, Cowley et al., Cone RD, Cowley et al., Schwartz et al

The arcuate nucleus (ARC) is found at the bottom of the hypothalamus and consists of several groupings of specific neurons (see Fig. 4). There are two main populations of neurons regulating appetite; those co-expressing POMC and cocaine and amphetamine regulated transcript (CART) and those co-expressing neuropeptide Y (NPY) and agouti related peptide (AgRP). The POMC/CART neurons form part of the satiety pathway in the arcuate nucleus and neurons co-expressing NPY and AgRP are part of orexigenic signalling. AgRP acts as an inverse agonist at MC3-R and MC4-R, to help stimulate feeding.

POMC/CART and NPY/AgRP neuron populations both express leptin receptors (LepRb). Leptin has opposing effects on each neuron population, activating the anorexigenic POMC neurons and inhibiting the orexigenic NPY/AgRP neurons, in the arcuate nucleus. Using transgenic mice expressing green fluorescent protein (GFP), two CNS regions expressing POMC in response to leptin were found: the arcuate nucleus and the nucleus of the solitary tract (NTS). Another study used transgenic mice with POMC-GFP expression and NYP-GFP expression, measuring the number of inhibitory and excitatory postsynaptic currents. Leptin deficient (ob/ob) mice were shown to have an increased excitation of NPY neurons and increased inhibition of POMC neurons compared to their wild-type littermates. Treatment with leptin lead to normalisation of the synaptic signals to wild-type levels.

The full extent of the signalling between the neuron populations within the hypothalamus is not fully understood. It is thought that NPY neurons project to inhibit POMC expression, mediated through GABA. Using electron microscopy, co-expression of GABA and NPY was confirmed at these nerve terminals. Leptin also acts at the GABA nerve terminals to reduce the inhibition of POMC neurons.

Ghrelin mRNA has been shown to be expressed in the hyothalamus, using RT-PCR, but it is at present not clear whether the brain is a significant source of ghrelin. There is some immunocytochemical evidence for ghrelin-containing axons in several hypothalamic nuclei, but at present it is thought that circulating ghrelin, secreted from the empty stomach, is the key appetite-regulating signal. Ghrelin was shown using fluorescent protein tagged NPY neurons, to increase release of NPY and AgRP, as well as increasing the GABA-mediated inhibition of POMC neurons. Ghrelin is the endogenous ligand for the GHS-Receptor, which is dendely expressed in both the arcuate nucleus and the ventromedial nucleus.

There are five NPY receptors (Y1-Y5), which are also activated by the closely related peptide PYY. Using specific receptor agonists and measuring the effect on POMC mRNA levels, it was found that NPY down-regulates POMC expression through the Y2 receptor. Interestingly, activation of a presynaptic Y2 autoreceptor suppresses NPY release. PYY3-36, a satiety signal released from the gut after a meal, acts as an agonist at the presynaptic Y2 receptors on the nerve terminals of NPY/AgRP/GABA neurons. This leads to suppression of NPY release and increased POMC release in the arcuate nucleus.

POMC neurons involved in regulation of appetite are also found in the nucleus tractus solitarius (NTS), with projections to the paraventricular nucleus and elsewhere. Cholecystokinin (CCK) is another satiety-inducing peptide from the gut. Its effects are mediated through CCK1 receptors and vagal afferents to the NTS. Both feeding and intraperitoneal injection of CCK8 resulted in increased c-fos expression in NTS POMC neurons. The PVN contains oxytocin neurons that project to the NTS, and which are part of a feedback mechanism to stomach, closing off the gastric sphincter. In addition, magnocellular oxytocin neurons in both the supraoptic nucleus and the PVN have MC4 receptors, and in response to alpha MSH can release large amounts of oxytocin from their dendrites within the hypothalamus; the magnocellular neurones are also excited by gastric distension, and by systemic administration of CCK. The sites of action of dendritically released oxytocin are not clear, but the VMH is a likely target, as it has a very high density of oxytocin receptors.

Species Variation in the POMC Gene

Fig. 5 Evolutionary Tree for Vertebrates, adapted from Dores

The first vertebrates to evolve, about 530 million years ago, were jawless. Members of the agnatha superclass, the oldest extant class of vertebrates, are the jaw-less descendants of these early vertebrates. Modern day agnathans include lampreys and hagfish. The Gnathostomes (jawed vertebrates), appeared during the Middle Devonian era (~380 millions years ago) and bifurcated into the cartilaginous fish (Chondrichthyes) and the bony fish (Osteichthyes).

The Chondrichthyes diverged from the other major class of gnathostomes, the Osteichthyes, between 440 and 420 million years ago; and by 410 million years ago the two extant subclasses had been formed, Elasmobranchi and Holecephali. Osteichthyes is the largest class of vertebrates and can be divided into the Actinopterygians (ray-finned fish) and the Sarcopterygians (lobe-finned fish). The earliest tetrapods evolved from a common ancestor from the Sarcopterygii class during the Devonian era.

Vertebrates

Agnatha

Two POMC-related DNA sequences were found in lamprey using cDNA cloning. The final products, after post-translational modification, of the two genes are the same as in gnathostomes. (Kawauchi and Sower, 2006)One gene was found to encode 2 MSH sequences (labelled MSH-A and MSH-B) and beta-endorphin, but not ACTH, and so was named proopimelanotropin (POM). The other gene encoded ACTH, a different beta-endorphin and nasohypophyseal factor, and was named proopiocortin (POC). The gene structure of POM and POC was consistent with gnathostomes POMC (3 exons and 2 introns) (takahashi et al, 2005b)

Northern blot analysis showed tissue specific expression of POM and POC in the pars intermedia and pars distalis respectively, and similar distribution of MSH and ACTH to other vertebrates. (takahashi et al 1995)It is thought that the POC and POM genes originated from a single ancestral lamprey POMC gene which was duplicated. The 2 genes then evolved and diverged according to their tissue specific functions.

Gnathostomes

Osteichthyes

Sarcoptergii

POMC organisation among vertebrates in the Sarcoptergii class, consisting of the lobe-finned fish and the tetrapods, is very similar (Dores 2005). POMC in tetrapods, as described above, has three melanocortin sequences (γ-MSH, ACTH/α-MSH, and β-MSH) and a β-endorphin sequence. Lobe-finned fish, consisting of the lungfish and coelacanths (evolution of melanocortin systems in fish), also have three MSHs in their POMC, as well as one β-END. There are also many similarities between the POMC amino acid sequences of species within the Sarcoptergii class (Dores 2005). This highlights the fact that tetrapods and lobe-finned fish share a common POMC ancestor. However, mutations have been found in the core sequence of the γ-MSH region in POMC of one species of lobe-finned fish (the Australian lungfish). Similar mutations have also been found in some ray-finned fish, but not in other lobe-finned fish, or tetrapods (Dores at al 1999). This suggests that in the Australian lungfish, the γ-MSH sequence may be degenerating. This is similar to what has occurred in ray-finned fish (see below).

Actinopterygii

Like the Sarcoptergians, some ray-finned fish, such as the paddlefish and the sturgeon, have three MSH regions in their POMC sequences. However, mutations have been found in the core sequence of the γ-MSH sequence in these species, and at the proposed cleavage sites marking this region (Dores 1999, Kawauchi 2006), which probably means that it is non-functional and γ-MSH is not produced (Amemiya 1997) (Dores 2005). These mutations are similar to those found in the Australian lungfish, described above. All teleosts, a type of ray-finned fish, do not have the γ-MSH region at all. By sequencing the nucleotides of POMC mRNA, Kitahara et al noted its absence in salmon, and the same has been found in other teleosts (Dores 2005). This suggests the theory that the POMC gene in ray-finned fish accumulated mutations in the γ-MSH region over time, leading to its eventual deletion in teleosts (Dores 1999, Kawauchi 2006, Danielson 1999, Dores 2005).

Some ray-finned fish also express two or more POMC genes. For example, in the paddlefish, two POMC cDNA clones have been found (Danielson 1999). This is the same in sockeye salmon, rainbow trout, and sturgeon (takahashi 2006). However, in the flounder, three have been noted (Takahashi 2005). It has therefore been suggested that there has been duplication of the POMC gene in ray-finned fish (Danielson). Unlike in the lamprey, these genes are still all co-expressed in the pituitary (Dores 2005) and encode the same hormones as the tetrapod POMC. There are some differences between the two/ three POMC genes expressed however. For example, the β-MSH region is more highly conserved than the β-endorphin region in the paddlefish (Danielson) and POMC-C in the flounder has mutations in the β-endorphin sequence (Takahashi 2006). It has been suggested that in ray-finned fish, the degeneration of the γ-MSH sequence occurred after the duplication of the POMC gene (Danielson).

Chondrichthyes

The major distinguishing feature of Chondrithyan POMC is the presence of a 4th MSH sequence, δ-MSH, between the ACTH/α-MSH and β-MSH domains. This has been extensively researched in several extant members of both subclasses. The presence of δ-MSH exclusively in all Chondrichthyes suggests that it emerged after the divergence of Chondrichthyans, but before the radiation into subclasses. (kawauchi and sower)

The four MSHs can be classified into 2 groups, based on similarities in amino acid sequence and domain length: α-MSH/γ-MSH and β-MSH/δ-MSH. A region between δ- and β-MSH, known as the C-terminal extension of δ-MSH (CTED), was found to be very similar to the β-endorphin domain. (takahashi et al, 2004). It has therefore been hypothesised that internal gene duplication of the β-MSH and β-endorphin domains, and subsequent mutations in the duplicated region, led to the formation of δ-MSH and CTED.

Fig. 6 Hypothetical outline for the evolution and inheritance of the POMC gene, adapted from Dores and Lecaude, and Kawauchi and Sower

Invertebrates

POMC peptides, similar to those in vertebrates, have been identified in invertebrates as well. The POMC genes were found to contain the same domains in the same sequential order. Sequence identity of the leech POMC in its entirety was low when compared to vertebrate POMC, however ACTH and MSH displayed very high (over 80%) degrees of identity (salzet at al, 1997). It was also shown, through bioassays, that leech α-MSH displayed the same inhibitory action on human leukocytes as vertebrate α-MSH. So despite hundreds of millions of years of separate evolution and variations in peptide sequence, the leech MSH is still able to act on a specific human MSH receptor (salzet at al, 1997).

Molluscs have been found to contain a POMC-like peptide, similar to those found in mammals. Peptides formed from the precursor included γ-MSH, α-MSH, ACTH and a β-endorphin-like peptide. (Stefano et al, 1999)

When taken together, these findings support the theory that POMC and its end-products originated very early in chordate evolution. The common existence of α-, β- and γ-MSH in the same order suggests that they appeared before the bifurcation of invertebrates and vertebrates. At some point, an ancestral POMC gene may have existed containing a single melanocortin domain and a β-endorphin domain. The ACTH/α-MSH sequence may have been the original melanocortin, as it has been very highly conserved throughout evolution, probably due to some functional selection pressure. (Dores and Lecaude)