Phoenixin was first described as a reproductive peptide in 2013 by Yosten et al. (Yosten et al., 2013) who employed a bioinformatic approach in order to deduce possible conserved peptides from human genome data (Yosten et al., 2013). Its highly conserved nature is shown by the very small inter-species variability with no differences in the phoenixin-14 sequence in mammals and only one amino acid variation in the phoenixin-20 sequence in the initially studied tissues (Yosten et al., 2013; Jiang et al., 2015a). Since its discovery, a total of 69 original research papers (10/22) were published on Pubmed. The majority of these research papers have focused on reproduction as well as the neurophysiology of phoenixin, with several publications also focusing on food/water intake and digestion as well as its psychopathological properties and inflammation. Other publications have studied memory impairment and cardiovascular implications.

Several variants of phoenixin have been described, namely 5, 8, 14, 15, 17, 20, 26, 36, and 42 amino-acid long variants (Yosten et al., 2013; Private Communication; Cowan et al., 2015; Phoenix Pharmaceuticals, INC 2021; Pubmed, 2022). The physiologically relevant peptides and targets of most studies are phoenixin-14 and phoenixin-20 (Yosten et al., 2013; Pubmed, 2022). Both forms show a similar affinity to pituitary cell membranes in cell cultures (Yosten et al., 2013). Its precursor protein SMIM20 (also known as MITRAC7 and C4orf52) (Prinz et al., 2017; National Library of Medicine, 2022a) is scarcely researched so far. It is part of the MItochondrial Translation Regulation Assembly intermediate of Cytochrome c (MITRAC), which regulates parts of the respiratory chain, namely the abundance of cytochrome c oxidase (complex IV) through stabilization of COX1, a part of cytochrome c oxidase biogenesis (Dennerlein et al., 2015; Bourens and Barrientos, 2017). This is part of a feedback loop to avoid mitochondrial damage due to excessive precursor buildup (Dennerlein et al., 2015). Deviations from normal complex IV functioning can be the cause of encephalo-cardiomyopathies (Bourens and Barrientos, 2017). Although its absence does not seem to lead to a complete dysfunction of complex IV, early studies have shown a 50% reduction in complex IV activity (Bourens and Barrientos, 2017). Phoenixin-14 is MITRAC7’s most abundant cleavage product, potentially a result of carboxypeptidase cleavage (Lyu et al., 2013).

This review aims to summarize the currently available literature concerning phoenixin, its receptor and its involvement in physiology and pathology as well as outline potential pharmaceutical targets.

Phoenixin was mostly studied in rodents, where icv injection of phoenixin-14 showed a dose dependent anxiolytic effect (Jiang et al., 2015a). This was in part attenuated by a GnRH receptor antagonist while being unaffected by vasopressin/oxytocin receptor antagonists (Jiang et al., 2015a). Direct infusion into the amygdala did not show any anxiolytic effects (Jiang et al., 2015a), leading to a proposed GnRH dependent mode of action most likely located in the anterior hypothalamic area (AHA) (Jiang et al., 2015a). Therefore, a GnRH receptor dependent and oxytocin independent anxiolytic effect has been proposed (Jiang et al., 2015a). In humans, a negative association between plasma phoenixin levels and anxiety in obese men was observed (Hofmann et al., 2017).

In addition to anxiolytic effects, phoenixin injected icv into the lateral ventricles led to a dose dependent decrease in core body temperature, which was also in some form dependent on GnRH receptor activity, since its blockage decreased phoenixin-14’s effects (Jiang et al., 2015a). The reduction in body temperature peaked 20 min after injection and lasted for roughly 1 hour (Jiang et al., 2015a).

As shown previously by our research group, both restraint stress (Friedrich et al., 2020) and immunological LPS stress (Friedrich et al., 2022) altered central phoenixin immunoreactivity. Restraint stress simulated emotional stress and caused an increase in activity and phoenixin density in several nuclei, namely the RPa, mNTS and DMN (Friedrich et al., 2020). An effect of restraint stress on phoenixin was also shown in another restraint stress study, which showed a decrease in phoenixin plasma concentration 15 min after restraint stress was administered (Schalla et al., 2020). These levels correlated with serum nesfatin-1 levels (Schalla et al., 2020).

Immunological stress exerted through an ip injection of lipopolysaccharide (LPS) resulted in a similar pattern in brain activity, namely activating the SON, CeM, RPa, mNTS and DMN and increasing phoenixin immunoreactivity (Friedrich et al., 2022). Phoenixin-20 was shown to have protective properties against LPS induced inflammation in dental pulp cells, where it reduced the release of LDH and attenuated the activity of the TLR-4, Myd88 and NF-κB pathway (Sun et al., 2020). This anti-inflammatory response employed GPR173 since silencing of GPR173 with shRNA abolished the effects (Sun et al., 2020). In another study, the authors showed an attenuation of the stress response of the endoplasmatic reticulum (ER) as a protective mechanism of phoenixin-14 against LPS-induced immunological stress in astrocytes (Wang et al., 2020). It also reduced the activation of the NLRP3 inflammasome and the abundance of ROS as well as decreased IL-1β and IL-18 production (Wang et al., 2020). The effects of phoenixin-14 on the NLRP3 expression and the reduced interleukin production were GPR173 dependent as shown by siRNA knockdown (Wang et al., 2020). Phoenixin-14 also reduced the increase of the high mobility group box 1 (HMGB1) protein (Wang et al., 2020), an inductor of NLRP3 expression (Frank et al., 2016) under conditions of inflammatory stress. This protective mechanism was also present in the OGD/R model (Zhang and Li, 2020). The same attenuation of the NLRP3 inflammasome was observed in microglial cells in vitro when treated with phoenixin-20 (Zeng et al., 2020), where the expression of thioredoxin-interacting protein (TxNIP), necessary for NLRP3 inflammasome activation (Nasoohi et al., 2018), was reduced by phoenixin, with phoenixin-14 and -20 showing a similar mode of action. The effects of phoenixin-20 were dependent on the sirtuin-1 (SIRT-1) pathway, since the inhibition of this pathway by nicotinamide abolished the effect (Zeng et al., 2020). Apart from neuroprotective properties, phoenixin-14 was also shown to ameliorate indomethacin-induced duodenal ulcers (Zandeh-Rahimi et al., 2022). Although in this study the model was used as a surrogate for inflammatory bowel disease, its results might hint towards effectiveness in the treatment of stress-induced ulcers as well, since its effectiveness surpassed that of a histamine-2-receptor antagonist, which is considered a treatment option in stress ulcers (Young et al., 2020; Zandeh-Rahimi et al., 2022).

The effects of phoenixin on neuron excitability in the NTS, a nucleus that both contains phoenixin and is activated by it (Prinz et al., 2017; Friedrich et al., 2022), has also been shown by a distinctive decrease in spike frequency variability when animals were – by accident – subjected to construction noise while also receiving phoenixin treatment (Grover et al., 2020). The construction noise had a similar effect as a pretreatment with corticosterone (CORT) for 22 days (Grover et al., 2020). Without the stressor present, phoenixin increased the firing frequency as well as the depolarization by 32% and 50%, respectively (Grover et al., 2020). These results suggest that phoenixin also plays a role in the reaction to external sensory stressors – noise in this case – and not only internal stressors such as endotoxin presence or emotional restraint stress.

As described above, nesfatin-1 is seemingly interconnected with phoenixin, exerting an antagonistic effect in most regards. Apart from food intake and anxiety, nesfatin-1 is also involved in various stressors such as surgical stress (Stengel et al., 2010), metabolic stress (Sun et al., 2021) and restraint stress (Sun et al., 2021). The importance of nesfatin-1 in stress is further underlined by its capability to activate the hypothalamus-pituitary-adrenal (HPA) axis and increase corticosterone levels in plasma after it was applied icv (Konczol et al., 2010). As reported previously, corticosterone treatment for 22 days abolished the effects of phoenixin on NTS neuron excitability (Grover et al., 2020), thereby suggesting an indirect link between nesfatin-1 and phoenixin in chronic stress, where an increase in corticosterone induced by nesfatin-1 signaling reduced phoenixin’s effects on neuron excitability, thereby indirectly reducing neuron excitability in the NTS, which is the “gateway” nucleus for peripheral sensory signals (Ran et al., 2022). This could be interpreted as an adaptation or desensitization towards the stressor in order to avoid an excessive stress response. A previously proposed pathway could be an activation of nesfatin-1 receptors in the PVN, a highly relevant nucleus in stress reactions which contains many CRF neurons regulating ACTH release (Vermetten and Bremner, 2002). Phoenixin depolarized magnocellular neurons in the PVN and increased firing frequency in vitro (Gasparini et al., 2018). The PVN was also shown by autoradiography to contain nesfatin-1 receptors (Prinz et al., 2016), leading to the assumption of an activation of nesfatin-1-receptor expressing neurons and thereby increasing the activity of the HPA axis. Since nesfatin-1 is able to cross the blood-brain barrier (Pan et al., 2007) and also elicited an increase in CRF mRNA expression in the hypothalamus and increased plasma corticosterone levels after chronic application (Ge et al., 2015), interaction of peripheral and central signaling might also play a role. To summarize this theory, stressors relayed via the NTS could lead to an increase in nesfatin-1 signaling, which then activates the PVN and increases corticosterone levels, which in turn over a longer period of time reduce phoenixin effects in the NTS, thereby reducing the amount of peripheral sensory stress input relayed via the NTS and allowing for an adaptation to the stressor.

As of now, phoenixin is not yet in use for human pharmaceutical purposes, since research on phoenixin and its possible pharmacological properties are still in its infancy. The currently published studies do, however, provide us with some insight into its possible and probable clinical applications. The observed effects of phoenixin are mainly increased appetite (Schalla and Stengel, 2019), reduced anxiety (Jiang et al., 2015a), increased fertility (Basha et al., 2022) and cardioprotective properties (Wei et al., 2021). These fall perfectly in line with the pathologies observed in patients with anorexia nervosa, namely inhibited appetite and reduced body weight, increased parameters of anxiety and depressiveness and decreased fertility as a result of the low body weight (Figure 1) (Zipfel et al., 2015; Westmoreland et al., 2016). Anorexia nervosa is also one of the psychiatric diseases, for which both pharmacological as well as psychotherapeutic treatment still render unsatisfactory results with a high risk of relapse. Its mortality rate is the highest of all psychiatric disorders (Edakubo and Fushimi, 2020). New therapies are therefore of the utmost importance for patients suffering from anorexia nervosa.

The phoenixin pathway seems a possible target for future drug development. Its anxiolytic and orexigenic properties could be helpful in both the initial phase of treatment in order to stabilize body weight and reduce the malnutrition as well as in long-term treatment, possibly by both exerting an anxiolytic as well as an antidepressant effect. As previously reported, phoenixin serum levels were significantly lower in patients suffering from anorexia nervosa compared to healthy controls (Palasz et al., 2019). The observed normalization of phoenixin serum levels when patients gained weight and the resulting positive correlation between phoenixin and body weight normalization (Palasz et al., 2019) deliver an initial hint towards a possible role of phoenixin in the pathogenesis of anorexia nervosa.

Apart from its known effects on food intake and anxiety, a relevant information for the practicality of phoenixin as a pharmaceutical target that remains to be elucidated in order to evaluate usefulness is whether or not it can cross the blood-brain barrier. This question is not yet answered, hints in either direction exist (Schalla et al., 2017; Rocca et al., 2018). In the absence of specific active transport processes, size and polarity i.e. lipophilicity of molecules can determine its ability to cross the blood-brain barrier (Mangas-Sanjuan et al., 2010). While smaller molecules have a much better chance of passing it, larger molecules could potentially shield dipolar parts of a molecule and thereby aide its ability to cross (Warren, 2018). Threshold values for size and polarity are believed to be around <400 Da with less than eight possible hydrogen-bonding spots exposed by the protein (Pardridge, 2012). Although smaller molecules without any further modification have a better chance of passing the blood-brain barrier, neither the small phoenixin-14 molecules nor the larger SMIM20 molecules should be ruled out as possible pharmacological targets. It should be noted that the differentiation between phoenixin’s ability to penetrate into the cerebrospinal fluid and its ability to cross the blood-brain barrier is very important (Pardridge, 2012). Interestingly, icv administration of phoenixin led to a different effect than peripheral injection as described above. Besides reaching adjacent nuclei after icv injection (Yan et al., 1994), phoenixin could also be applied intranasally. Other peptides such as orexin A (Dhuria et al., 2009) as well as the GLP-1 antagonist exendin_(9–39) (Banks et al., 2004) were shown to be able to reach the brain this way in early animal models. The ability to avoid the blood-brain barrier this way is commonly discussed (Hanson and Frey, 2008) but not universally agreed upon (Pardridge, 2019). All of these speculations warrant further research, specifically the distribution of phoenixin after icv administration, elucidation of its ability to permeate the blood-brain barrier as well as its effects once it has crossed the blood-brain barrier.

Another aspect of phoenixin in regard to its pharmacological potential is the observed sex specific nature of its effects (McIlwraith et al., 2018; McIlwraith et al., 2019; Haddock et al., 2020). In addition, the receptor and its binding sequence have not yet been positively identified and proven, hindering the research into possible ligands and agonists as well as antagonists. Since GPR173 has recently been doubted as phoenixin’s receptor (Yanez-Guerra et al., 2022), the identification of phoenixin’s receptor is of utmost importance. Expression and coexpression studies as well as further research on the affinity of phoenixin to GPR173 as well as other GPRs could also hint towards possible side effects and interactions with other pathways and should be explored. Aside from phoenixin itself, a further elucidation of its pathway and proposed antagonist nesfatin-1 will be of importance. As of now it is unclear if phoenixin and nesfatin-1 interact directly, through their respective receptors or possibly even share binding capabilities for GPR173. Therefore, an indirect influence on the phoenixin pathway through nesfatin-1 modulation might be an option further down the line.

Importantly, it should be mentioned that all of phoenixin’s possible drug targets and pharmaceutical use cases remain highly speculative at the moment and the hypotheses discussed above should be interpreted as such. Current scientific data is by far premature for a use of phoenixin or its pathway in human medical drug treatment within the coming decade. Nonetheless, as outlined in this review, the possibilities are plentiful and from the limited research that is available, the properties of phoenixin seem promising. There is, however, much to be studied in order to gain more insight into phoenixin and develop use cases.