Towards improved receptor targeting: anterograde transport, internalization and postendocytic trafficking of neuropeptide Y receptors

Y₁ receptor

The human Y₁ (hY₁) receptor consists of 384 aa and shows a sequence identity of more than 92% with its orthologs expressed in pig, guinea pig, mouse, rat and Bos taurus. Generally, the sequence identity among the Y receptor subtypes is rather low, but is highest for the Y₁ and the Y₄ receptor. Hence, hY₁ and hY₄ (and y₆ in other mammals) form one Y receptor superfamily, while hY₂ and hY₅ each form an autonomous group (Larhammar and Salaneck, 2004). Tissues in which the hY₁ receptor can be found are predominantly in the brain (dentate gyrus of the hippocampus) (Dumont et al., 1998) but also in the vascular smooth muscle cells, adipose tissue, kidney and the gastrointestinal tract (Michel et al., 1998). The most prominent role of this Y₁ receptor subtype is the induction of food intake and the regulation of energy homeostasis in synergy with Y₅ (Nguyen et al., 2012). Moreover, the Y₁ receptor modulates vasoconstriction (Hodges et al., 2009) and conveys neuroregenerative effects in vitro and in vivo (Howell et al., 2005; Thiriet et al., 2011), as well as antidepressant and anxiolytic effects in rodents (Wahlestedt et al., 1993; Verma et al., 2012). The hY₁ is overexpressed in more than 84% of primary human breast (Reubi et al., 2001) and Ewing sarcoma tumors (Korner et al., 2008), representing a promising diagnostic and therapeutic target for these types of cancer.

Y₂ receptor

The hY₂ receptor is a 381-aa-protein that is expressed predominately in neuronal tissue, but also in the spleen, liver, blood vessels, gastrointestinal tract, white and brown fat tissue. This receptor shares a sequence identity of more than 92% with its orthologs in mouse, Bos taurus, pig and guinea pig. In the brain, activation of presynaptically-expressed Y₂ receptors inhibits neurotransmitter release (Klapstein and Colmers, 1993), including NPY and glutamate, which makes this receptor an interesting target for antiepileptic drugs (Vezzani and Sperk, 2004). Furthermore, the Y₂ conveys anorexigenic effects in mice, rats and humans (Batterham et al., 2002) and is involved in memory retention, mood disorders (Verma et al., 2012), angiogenesis (Lee et al., 2003), and the reward system following alcohol consumption (Hayes et al., 2012). In addition, the Y₂ receptor is overexpressed in glioblastoma and neuroblastoma, where it induces tumor growth and vascularization (Kitlinska et al., 2005; Korner and Reubi, 2008; Lu et al., 2010). Interestingly, it was recently found that Y₂ receptors expressed in the central nervous system and in non-neuronal tissues might have different assignments. Only Y₂ receptors of the central nervous system regulate bone mass while peripherally-expressed receptors were not found to contribute to bone metabolism at all (Shi et al., 2011).

Y₄ receptor

With a sequence identity of <86%, the hY₄ receptor has the lowest proportion of identical sequence compared to its orthologs in relation to other Y receptor subtypes. The hY₄ receptor consists of 375 aa and is mainly expressed in the gastrointestinal tract and, to a much lesser extent, in the hippocampus, the hypothalamus and the area postrema, a region with an incomplete blood-brain barrier (Dumont et al., 1998; Lindner et al., 2008a; Bellmann-Sickert et al., 2011). There, the Y₄ receptor can receive signals from peripherally-circulating PP and NPY, modulating energy homeostasis and emotional behavior in synergy with the Y₂ receptor (Lin et al., 2009; Tasan et al., 2009). Moreover, the Y₄ receptor inhibits gastric emptying in humans, followed by a delayed postprandial increase in blood glucose levels and subsequent insulin secretion (Schmidt et al., 2005), thereby inducing anorexigenic effects. As the Y₄ receptor also has an inhibitory effect on gastrointestinal secretion, it might be a potent target in diarrheal disorders (Tough et al., 2006). Besides this, the Y₄ receptor was found to be expressed in several human colon adenocarcinoma cell lines (Cox et al., 2001), pointing towards this receptor having a role in the development of colon cancer in humans.

Y₅ receptor

The hY₅ receptor exists in two isoforms with similar pharmacological profiles (Rodriguez et al., 2003). The long isoform of the hY₅ receptor consists of 455 aa, while the short isoform constitutes a splice variant that lacks the first 10 aa. With respect to the other Y receptor subtypes, the Y₅ receptor possesses a relative large third intracellular loop (ICL3) with approximately 140 aa, while the C-terminus is rather short. The Y₅ receptor shares a sequence identity with its orthologs of over 85% and is abundantly expressed in the hypothalamus, thalamus, amygdala and temporal cortex (Durkin et al., 2000). It contributes to energy homeostasis and food intake together with Y₁ (Nguyen et al., 2012), and mediates anticonvulsant effects (Benmaamar et al., 2005). Moreover, recent findings suggest that the Y₅ receptor also contributes to anxiolysis in rats (Morales-Medina et al., 2012). Similar to the Y₁ receptor, the Y₅ receptor is expressed in distinct human breast cancer cell lines, where it promotes cell growth and migration (Sheriff et al., 2010).

Role of C-terminal sequences

With a focus on hY₂ receptor, the sequence motif ³³²Y(x)₃F(x)₃F³⁴⁰ located at the very proximal C-terminus was identified to be crucial for ER export (Walther et al., 2012). Complete or partial deletion of the motif as well as point mutations from tyrosine/phenylalanine to alanine led to dramatic ER retention of the receptor. Interestingly, transferring the motif to a more distal region of the hY₂ receptor C-terminus did not rescue the anterograde transport and the receptor still accumulated in the ER. As the very proximal C-terminus comprises the putative eighth helix, we concluded that the ³³²Y(x)₃F(x)₃F³⁴⁰ motif must be essential for its formation, and thus for the correct receptor conformation. This hypothesis was further confirmed by performing a low-temperature (28°C) experiment to rescue receptor folding. Thereby, only receptors with single mutations that affected the ER export motif (tyrosine/phenylalanine→alanine) or the formation of the eighth helix (³³⁸S→P) were partly rescued when cells were cultured at 28°C. In contrast, complete deletion as well as relocation of the motif to a more distal C-terminal region still prevented ER export despite lower temperature cultivation. These findings are in good agreement with the function and position of the triple phenylalanine motif of the dopamine D₁ receptor, which was found to bind to the ER chaperone DRiP78 (Bermak et al., 2001).

Like the hY₂ receptor, the hY₄ receptor comprises a comparable motif within the proximal C-terminus, namely ³²⁹F(x)₃I(x)₃V³³⁷. Interestingly, deletion of this motif did not result in impaired ER export but prevented transport of the receptor from the Golgi apparatus to the cell surface (Walther et al., 2012). Similarly, the motif of the hY₁ receptor ³²⁷F(x)₃L(x)₃F³³⁵ resembles the motif of the hY₂ and hY₄ receptors and is also located within the putative eighth helix. Nonetheless, deletion of this motif had no influence on the cell surface targeting of the hY₁ receptor (Walther et al., 2012).

As mentioned above, the hY₅ receptor exhibits a short C-terminus and a large ICL3. Thus, it seems reasonable that export motifs or structural determinants might be displayed by the ICL3 instead of the C-terminus. Indeed, deletion of the putative eighth helix of the hY₅ receptor C-terminus had no influence on cell surface expression of the receptor as it has been observed for the hY₁ receptor (Walther et al., 2012). Replacing the ICL3 of the hY₅ with the ICL3 of the hY₂ receptor, as well as deletion of large parts of Y5 ICL3 sequences (unpublished results), had no effect on membrane targeting either (Bohme et al., 2008). Thus, it can be suggested that the hY₅ receptor displays its export motif in some region other than the C-terminus/ICL3.

With respect to palmitoylation, which is also a common posttranslational modification of GPCR stabilizing the eighth helix, mutation of the Cys within the proximal C-terminus of the Y₁ and Y₂ receptor led to opposite results. While the palmitoylation motif is crucial for the G protein-coupling and desensitization of the Y₁ receptor, mutation of the Cys within the Y₂ receptor sequence had no effect on receptor signaling properties or cell surface targeting (Holliday and Cox, 2003; Walther et al., 2012).

Role of the N-terminus and its glycosylation motifs

All human Y receptors have an N-terminus of about 40–50 aa, with at least one putative N-glycosylation site. An additional glycosylation site is postulated within the ECL2 of the hY₄ receptor. Generally, N-glycosylation was found to be a prerequisite for the association of chaperons like calnexin and calreticulin, which facilitate GPCR folding in the ER (Dong and Wu, 2007). However, not all GPCR require their N-termini and receptor glycosylation for ER export. The Y₁, Y₂ and Y₄ receptors were reported to be glycosylated, however, the impact of this posttranslational modification on membrane targeting was not addressed (Hansen and Sheikh, 1992; Voisin et al., 2000; Holliday and Cox, 2003). To answer the question of whether the N-termini and their potential glycosylation sites of human Y receptors are important for membrane integration, N-terminally truncated mutants were investigated in our laboratory (Lindner et al., 2009). Interestingly, complete deletion of the hY₁ and hY₄ receptor N-termini prevented membrane integration. Receptors accumulated in intracellular compartments that resembled the Golgi apparatus (Figure 3B). However, N-terminal extension with arbitrary sequences of at least eight residues rescued membrane targeting, despite the absence of any glycosylation motifs. Here, the N-terminal extension seems to stabilize the overall receptor conformation of the hY₁ and hY₄ receptors. In contrast, complete deletion of the N-terminus was very well tolerated for the hY₂ and hY₅ receptors and no N-terminal extension was required for membrane integration. Consequently, it can be speculated that glycosylation does not play any role in the membrane targeting of Y receptors. Nonetheless, it cannot be excluded that the hY₄ displays a glycosylation within its ECL2 that might promote cell surface expression.

Receptor oligomerization

It is very likely that the N-terminus or the putative eighth helix contribute to the organization of the correct receptor conformation and thus are essential elements required for cell surface expression. Those structural elements may therefore favor receptor dimerization in the ER and promote transport to the membrane and/or enhance the membrane retention time, as has been shown for other GPCR (Salahpour et al., 2004). However, the role of class A GPCR oligomerization is discussed controversially, since its physiological relevance is still in question (Gurevich and Gurevich, 2008). Oligomerization of GPCR is generally thought to be achieved by three types of interaction:

• the most stable covalent cysteine-bridge like in metabotropic glutamate receptors;

• coiled-coil interactions favored by the putative eighth helix, such as in GABA_B receptors; and

• non-covalent contacts between transmembrane helices that might lead to the domain swapping of the helices (Breitwieser, 2004).

Non-covalent interactions were recently evaluated in more detail: the crystal structures of different GPCR revealed that receptor protomers contact each other via two types of interfaces. While some receptors, such as rhodopsin and the κ-opioid receptor, form less compact homodimers via their transmembrane helices I, II and their intracellular eighth helix, other GPCR such as CXCR4, tightly interact through its transmembrane helix V and VI also involving intracellular regions of helix III and IV (Filizola, 2013; Katrich, 2013). Moreover, some receptors seem to be capable of interacting via both interfaces, as shown for the μ-opioid receptor.

Although the crystal structures of Y receptors and the corresponding oligomerization interfaces have not been available up to now, Y receptor homodimerization at the cell membrane has been shown in vitro (Berglund et al., 2003a; Dinger et al., 2003) and in vivo (Estes et al., 2008) using ultracentrifugation, bioluminescence and fluorescence resonance energy transfer (BRET/FRET) as well as Western blot techniques. Moreover, Y receptor dimers were found to be pre-associated with heterotrimeric G proteins at the cell surface in a ratio of 2:1 or 2:2 (Estes et al., 2011; Kilpatrick et al., 2012). Pre-association with G protein was also observed for other GPCR, such as the GABA-B₂ receptor, and most likely occurs shortly after biosynthesis in the ER (David et al., 2006). Thus, Y receptor homodimerization and pre-formation of the G-protein-receptor complex might be in fact a prerequisite for anterograde transport and full functionality of the receptor at the cell surface. In addition to homodimers, heterodimerization was observed for the Y₁ and Y₅ receptors (Gehlert et al., 2007), presumably due to the fact that these receptors are often co-expressed in the same tissues, and are located on the same but opposite and overlapping chromosomal region. When stimulating the Y₁/Y₅ heterodimer, Gehlert et al. observed an increase in Y₅-receptor internalization with respect to its mono- and homodimer (Gehlert et al., 2007). These results were contradicted by the findings of later studies by Bohme et al, showing that co-expression of Y₁/Y₅ receptors did not influence Y₅ receptor endocytosis (Bohme et al., 2008). Vice versa, the Y₁ receptor was found to be internalized alone after stimulation despite being co-expressed with the Y₅ receptor (Bohme et al., 2008). Moreover, while some groups report on Y receptor dimer dissociation upon agonist stimulation (Berglund et al., 2003a; Estes et al., 2008), others have observed an increase in oligomerization following agonist-induced receptor activation (Gehlert et al., 2007). Clearly, the role of Y receptor dimerization, as for many other class A GPCR, is still a matter of debate. Thus, further studies will be necessary to identify additional sequences and interaction partners that promote Y receptor export and putative oligomerization to gain further insights into the exact trafficking mechanisms.

Y₁ receptor

The hY₁ receptor was found to be rapidly internalized upon agonist stimulation by different groups (Fabry et al., 2000; Gicquiaux et al., 2002; Bohme et al., 2008; Ouedraogo et al., 2008; Lindner et al., 2009; Kilpatrick et al., 2010; Lecat et al., 2011; Lundell et al., 2011), consistent with the results that were obtained for its orthologs in the rat (Holliday and Cox, 2003; Pheng et al., 2003; Holliday et al., 2005; Kilpatrick et al., 2012) or guinea pig (Parker et al., 2001, 2002b). Several studies on the internalization mechanism revealed that the Y₁ receptor undergoes clathrin-dependent endocytosis in an Arr3-dependent fashion and is transported back to the cell membrane, involving direct and indirect recycling mechanisms (Gicquiaux et al., 2002; Berglund et al., Pheng et al., 2003b; Holliday et al., 2005; Ouedraogo et al., 2008; Kilpatrick et al., 2010; Lecat et al., 2011). Beside the interaction with Arr3, the Y₁ receptor was also found to bind efficiently to Arr2, suggesting that it belongs to the group of ‘class B’ receptors (Ouedraogo et al., 2008; Kilpatrick et al., 2012). As mentioned before, interaction with Arr is usually a prerequisite for activated and phosphorylated receptors. For the identification of phosphorylation motifs, C-terminally truncated mutants of the rat and hY₁ receptor (whose C-terminal sequences only vary in three distal residues) were investigated with respect to G protein coupling, Arr binding and internalization (Holliday et al., 2005; Ouedraogo et al., 2008; Lecat et al., 2011). These studies showed that the truncation of the whole C-terminus is very well tolerated with respect to G protein signaling, supporting the idea that the formation of the correct Y₁ receptor structure does not require the presence of the C-terminus (see also anterograde transport). In contrast, truncation of the C-terminus had a dramatic impact on receptor desensitization, internalization and Arr recruitment. Detailed studies on point mutations clarified that individual serine/threonine residues within the C-terminal (S/T)-(S/T)-Φ-H-(S/T)-(E/D)-V-(S/T)-x-T motif (where x represents any aa and Φ a bulky hydrophobic residue) are phosphorylated by GRK2, and thus are required for receptor desensitization and Arr recruitment (Holliday et al., 2005; Ouedraogo et al., 2008; Kilpatrick et al., 2010, 2012) (Figure 4). Within this cluster, the number of serine/threonine residues rather than their actual position was found to be crucial for efficient Arr binding, suggesting that multiple phosphorylation events occur within this sequence (Kilpatrick et al., 2010).

Figure 4

Internalization motifs within human Y receptors.

(A) Position and sequence of identified and putative motifs are displaced. Motifs within boxes were identified as being responsible for either receptor internalization or postendocytic trafficking. Underlined letters represent key residues in postulated or identified consensus sequences. (B) Localization of human Y receptor fused to the enhanced yellow fluorescent protein at the C-termini in living HEK293 cells prior to (unstimulated) or after agonist stimulation. hY₁, hY₂ and hY₅ were stimulated with 1 μm pNPY, while hY₄ was stimulated with 100 nm hPP for 60 min. Scale bar: 10 μm; indep., independent; rec, recycling.

Another sequence within the C-terminus, namely the Y-x-x-Φ (YETI) motif, lays upstream of the serine/threonine cluster. This tyrosine-based sequence is known to promote direct interaction with the μ2 subunit of AP2, thereby mediating internalization (Pandey, 2009). In fact, the YETI motif was found to induce constitutive clathrin-dependent internalization of the Y₁ receptor once the downstream phosphorylation and Arr binding motif was absent (Holliday et al., 2005; Lecat et al., 2011). Interestingly, mutation of the tyrosine residue (AETI) in combination with the full length Y₁ receptor had no impact on internalization but impaired receptor reappearance at the plasma membrane after agonist-induced endocytosis (Lecat et al., 2011). Thus, the YETI motif within the C-terminus of the wild-type Y₁ receptor is believed to regulate receptor recycling rather than internalization. In addition, a second Y-x-x-Φ (YIRL) motif is present in the ICL3 of the Y₁ receptor and further adapter protein interaction sequences based on dileucine (D/E)-X_2-3-L-(L/I) or triple basic motifs (R/K)-(R/K)-(R/K) are predicted at the beginning of ICL2 (ERHQLI) and ICL3 (KRR), respectively (Pandey, 2009; Smith et al., 2012). However, the impact of these motifs on receptor internalization, recycling and Arr binding has not been estimated yet and is therefore speculative.

Beside the relevance of C-terminal sequences, a proline residue located in the ICL2 was identified as acting in concert with the DRY motif as an Arr anchor in the 5-HT_2c receptor (Marion et al., 2006). This proline residue can also be found in the ICL2 of the Y₁ receptor. However, none of the Y receptors displays a classical DRY motif, and substitution of the proline residue to an alanine only had a minor impact on Y₁ receptor internalization and Arr association (Ouedraogo et al., 2008). In conclusion, the C-terminus seems to be most crucial for the regulation of Y₁ receptor desensitization and trafficking, as summarized in Figure 4. However, it cannot be ruled out that further motifs within the three ICLs contribute synergistically to receptor internalization and trafficking properties, binding either to Arr or to proteins of the endocytic machinery.

Clearly, those protein-receptor interactions depend on the conformation of both binding partners and the corresponding accessibility of consensus motifs exposed by the cytoplasmic receptor site. According to the hypothesis of ‘functional selectivity or biased agonism’, different ligands can promote discrete receptor conformations and thus favor one downstream effect over another, as it has been observed for several GPCR, including adrenergic receptors and the vasopressin V2 receptor (Rahmeh et al., 2012). Interestingly, Pheng et al. reported that the peptidic ligand GR231118 induced sequestration of Y₁ receptor independent of G protein signaling (Pheng et al., 2003), which points towards a biased agonism of this compound. It is worth noting that GR231118 was initially identified to have agonistic effects only for Y₄, while displaying a classical antagonistic function towards Y₁ (Schober et al., 1998; Dumont and Quirion, 2000). In fact, Y₁ receptor internalization induced by GR231118 could not be confirmed in later experiments, even though the receptor species and cell types were consistent with the initial study (Tough et al., 2006). One possible explanation for this might be that internalization experiments by Pheng et al. were based on the internalization of the radio-labeled GR231118, while later studies focused on receptor redistribution subsequent to ligand stimulation. Thus, further investigations will be necessary to unravel how GR231118 is internalized and whether the Y₁ receptor can undergo biased signaling.

Y₂ receptor

For a long time, the internalization and postendocytic trafficking properties of the Y₂ receptor were controversial. Early radio-ligand binding studies on the guinea pig Y₂ receptor revealed that this receptor subtype does not undergo sequestration (Parker et al., 2001). Similar studies on the human and rhesus receptor, respectively, confirmed that Y₂ is resistant to endocytosis and shows only a weak interaction with Arr3 (Gicquiaux et al., 2002; Berglund et al., 2003b; Ouedraogo et al., 2008). In contrast, Bohme et al. clearly showed that the hY₂ receptor desensitizes and undergoes fast internalization in response to agonist stimulation comparable to rates observed for the hY₁ and hY₄ receptors (Bohme et al., 2008). Recent studies confirmed those results and revealed that internalization of the hY₂ receptor strongly depends on the agonist concentration used in the experiment (Lundell et al., 2011). In addition, Parker et al. reported that the internalization rate observed for hY₂ is much higher than for its ortholog in the guinea pig (Parker et al., 2008). It is worth noting that the C-terminal sequence of the guinea pig Y₂ receptor displays significant differences with respect to the serine/threonine cluster in comparison to other species, including humans. In a chimeric approach Bohme et al. showed that both the ICL3 and the C-terminus of the hY₂ receptor can individually promote endocytosis of the usually non-internalizing hY₅ receptor, thereby clearly demonstrating that these structural components determine agonist-induced internalization (Bohme et al., 2008). In addition, recent studies on Y₁/Y₂ chimeras revealed an enhanced internalization rate of the Y₁ receptor when its C-terminal sequence is exchanged with the Y₂ receptor C-terminus (Lundell et al., 2011).

Detailed investigation of the molecular mechanism underling Y₂ receptor internalization unraveled three key motifs within the C-terminus responsible for the regulation of hY₂ receptor trafficking (Walther et al., 2010). The first motif within the very distal C-terminus, namely D-S-x-T-E-x-T (DSFTEAT), was found to promote Arr3-dependent internalization subsequent to GRK2 phosphorylation. A second motif, D-x-Φ-H-(S/T)-(E/D)-V-(S/T)-x-T, that resembles the phosphorylation motif of the Y₁ receptor, was found in the more proximal C-terminus, comprising residues DAIHSEVSVT. This sequence also induces agonist-mediated internalization, but surprisingly in an Arr3-independent manner (Walther et al., 2010). The Arr3-independent internalization mechanism was only observed when all further distal C-terminal parts were truncated. Conversely, the presence of the intermediate basic sequence FKAKKNLEVRKN impeded Arr3-independent receptor internalization, presuming that this inhibitory motif masks the proximal internalization motif D-x-Φ-H-(S/T)-(E/D)-V-(S/T)-x-T in the full-length receptor (Walther et al., 2010). Moreover, the proximal motif was also found to promote Y₂ receptor recycling rather than internalization in the wild type receptor, similar to the function assigned to the YETI motif in the Y₁ receptor.

By comparing the Y₁ and Y₂ subtypes, it becomes clear that the regulation of receptor trafficking strongly depends on sequences within the C-terminus. Importantly, consensus sequences and exact mechanisms seem to vary between those two Y receptor subtypes. However, each receptor displays one primary phosphorylation motif within the C-tail that functions as a molecular switch, thereby regulating Arr recruitment and internalization. In addition, a second latent internalization motif is present that controls receptor recycling and, apart from that, becomes involved in receptor endocytosis under the condition that the primary motif is missing.

Interestingly, while all other Y receptors possess a proline residue within the ICL2 for the proposed Arr interaction, the Y₂ receptor displays a histidine at this position. Replacing this histidine with proline resulted in an enhanced Arr binding and internalization of the Y₂ receptor (Marion et al., 2006; Kilpatrick et al., 2010, 2012), suggesting that Arr also recognized motifs within the ICL. Clearly, the role of additional trafficking sequences within the cytoplasmic region of the Y₂ receptor requires further investigation.

Beside C-terminal residues, the N-terminus of the receptor has been found to contribute to receptor internalization properties. Even though hY₂ receptor export is not affected when the entire N-terminus is truncated, ligand binding, receptor activation, and consequently the internalization process were drastically impaired (Lindner et al., 2009). Importantly, elongation of the first transmembrane residues by any eight aa rescued receptor activation and internalization, suggesting a sequence-independent, stabilizing effect of the N-terminus towards the ligand-binding pocket and/or the active receptor conformation (Lindner et al., 2009). Similarly, Parker et al. confirmed that the Y₂ receptor N-terminus is not directly involved in agonist binding (Parker et al., 2008). However, this group discovered an acidic and proline-rich motif within the N-terminus contributing to receptor internalization. By mutating one aspartic acid residue within this motif to alanine, the receptor internalization was found to be accelerated. Since no other receptor characteristic was affected by this alanine substitution, it was suggested that the acidic motif within the N-terminus of the Y₂ receptor binds to components of the extracellular matrix. This, in turn, might affect Y₂ receptor internalization kinetics (Parker et al., 2012).

In conclusion, the hY₂ receptor is internalized in an Arr3-dependent manner, but higher agonist concentration seems to be required to promote receptor endocytosis (Walther et al., 2010; Lundell et al., 2011). Even though the serum concentration of NPY and PYY might be too low to drive receptor internalization in the periphery, local concentrations in the synaptic gap might be high enough to induce hY₂ endocytosis in the brain. That the Y₂ receptor desensitization might indeed have an important physiological relevance was impressively shown by Ortiz et al. (2007). Here, a stabilized PYY(13–36) analog was observed to reduce food intake in mice. In a long-term study, however, food intake returned to the control level after 3days of daily drug administration. This might be due to receptor desensitization or down-regulation.

Y₄ receptor

Similar to the Y₂ receptor, studies on Y₄ receptor desensitization and trafficking revealed opposing results. Initially, the hY₄ receptor was reported to be resistant to agonist-promoted desensitization and internalization when being expressed in Chinese hamster ovary cells (Voisin et al., 2000). In contrast, several other groups independently showed that the Y₄ receptor is rapidly internalized and subsequently transported back to the cell surface, most probably via the indirect recycling pathway (Parker et al., 2001, 2002a; Tough et al., 2006; Bohme et al., 2008). The determined internalization and recycling rates were either comparable or slightly slower than those observed for the Y₁ receptor. With respect to the exact internalization mechanism, Y₄ receptor endocytosis was found to be promoted by Arr3, suggesting a clathrin-dependent process (Berglund et al., 2003b). In these studies, however, Arr3 recruitment was less pronounced for the Y₄ than for the Y₁ receptor, supporting the idea that Y₄ receptor internalization and recycling rates might indeed be slightly prolonged.

In addition to these findings, the rate of Y₄ receptor sequestration was found to be highly sensitive to agonist affinity and efficacy, i.e., full agonists induced a higher internalization rate than partial agonists (Parker et al., 2002a; Tough et al., 2006). This is in agreement with the common mechanism of GPCR activation followed by sequestration, as described above, and underlines the fact that the endogenous Y₄ ligands generate two conformations, which are equally important for G protein and GRK/Arr binding. It is worth noting that the Y₄ receptor and its ligand PP emerged last and presumably quite quickly in evolution, leading to greater divergences in peptide and receptor sequences across mammalian species (Yulyaningsih et al., 2011). Thus, a combination of receptors and ligands originating from different species was observed to result in a decreased internalization rate due to reduced peptide efficacy (Tough et al., 2006).

Very little information is available about consensus sequences that drive internalization and recycling processes of the Y₄ receptor. As the Y₄ receptor is most closely related to the Y₁ receptor, it is not surprising that a variant of the serine/threonine cluster, which is known to promote Arr binding and internalization of the Y₁ receptor subtype, also occurs within the distal C-tail of the Y₄ receptor (STVHTEVSKG). Further sequences that might be involved in Y receptor trafficking include the proline-based Arr-binding motif and the tyrosine-based AP-interaction sequence within ICL2 and ICL3, respectively. Even though, all these motifs are present in the Y₄ receptor, their role in internalization and trafficking is still a matter of investigation.

Y₅ receptor

The regulation of Y₅ receptor trafficking is not well understood. Microscopic studies using EYFP-fused receptors and radio-ligand internalization experiments revealed an extremely slow internalization rate for both ligand and receptor (Parker et al., 2003; Bohme et al., 2008). Nonetheless, the receptor was still found to desensitize in inositol phosphate (IP)-accumulation assays (Bohme et al., 2008). The relatively small fraction of internalized receptor was even smaller in the presence of phenylarsine oxide, an inhibitor of clathrin-coated pit formation (Parker et al., 2003), suggesting a clathrin-dependent internalization mechanism. Studies on Arr recruitment are rather limited but we were unable to detect a pronounced Arr recruitment subsequent to Y₅ receptor stimulation in our lab (data unpublished). Contrasting with this, BRET studies showed that the Y₅ receptor can recruit Arr3 after stimulation, even though the signals were much lower than observed for the Y₁ receptor (Berglund et al., 2003b).

Comprehensive studies are necessary to unravel the structural determinants that modulate and inhibit Y₅ receptor internalization. It is reasonable that these structures being distinct to all other Y receptors, namely the long ICL3 and the short C-terminus, have an impact on Y₅ receptor trafficking. The Y₅ receptor chimera bearing either the ICL3 or C-terminus of the Y₂ receptor showed an accelerated internalization rate (Bohme et al., 2008). Still, it cannot be determined whether this effect is due to the deletion of inhibitory segments lying in the Y₅ receptor ICL3 and C-terminus or to the addition of internalization motif present in the Y₂ receptor sequence. Thus, further studies are required to estimate the mechanisms underlying Y₅ receptor trafficking.

Taking everything into account, it can be concluded that Y receptor subtypes internalize with rates in the following order: Y₁>Y₂≥Y₄>>Y₅. All receptors seem to undergo clathrin-mediated endocytosis, involving Arr in case of the Y₁, Y₂ and Y₄ receptors. The three latter receptors were also found to be recycled subsequent to endocytosis.