Therapeutic Potential of Targeting Prokineticin Receptors in Diseases

A. Prokineticins: Structure, Distribution, and Regulation

The genomic structure and chromosomal localization of human and murine PKs are determined. The pk1 gene, located on human chromosome 1p13.1 and murine chromosome 3, consists of 3 exons (Kaser et al., 2003): the first encodes 19 residues of the signal peptide and an alanine, valine, isoleucine, threonine, glycine (AVITG) sequence, the second 42 residues, and the third the last 39 residues (LeCouter et al., 2003a) (Fig. 1A). To date, only one PK1 transcript (86 amino acids) has been identified without any alternative splicing variants. In contrast, the pk2 gene is located on human chromosome 3p21.1 and murine chromosome 6 and consists in 4 exons (Jilek et al., 2000). Its transcription can result in four different splicing- variants due to the presence of the additional exon (exon 3) (Jilek et al., 2000): the canonical PK2 (81 amino acids) encoded by exons 1, 2, and 4 (LeCouter et al., 2003b); the inactive long form of PK2 (PK2L – 102 amino acids) encoded by all four exons and activated in vivo by proteolytic cleavage (PK2β) (Chen et al., 2005); the truncated PK2 (72 amino acids) encoded by exons 1 and 2 and containing part of introns 2 and 3′ (Jilek et al., 2000); and the PK2C (65 amino acids) encoded by exons 1 and 4 (Lattanzi et al., 2022) (Fig. 1B).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 1

Schematical representation of prokineticin systems. (A, B) Prokineticin genes and their splicing isoforms. (C, D) Amino acid sequence of mammalian prokineticins, mamba intestinal toxin 1 and Bv8. Created with BioRender.com.

The amino acid sequence analysis of PK1 and PK2 revealed a high percentage of sequence conservation across the species (Fig. 1, C and D). They share characteristic, highly conserved structural motifs (Li et al., 2001; Lin et al., 2002a). The AVITG sequence at the amino-terminus (NH2)-terminus, the tryptophan at position 24 (Trp24), and the COOH-terminal domain are critical for receptor binding and activation (Kaser et al., 2003; Bullock et al., 2004). Substitutions (Bullock et al., 2004; Dodé and Rondard, 2013), insertions (Bullock et al., 2004), or deletions (Negri et al., 2005) in these portions result in products with antagonistic activity, with reduced receptor affinity and therefore lower efficacy, or without any biologic functionality (Bullock et al., 2004; Lattanzi et al., 2012).

Moreover, the presence of charged residues inside, hydrophobic residues on the surface, and ten evenly distributed Cys forming 5-disulphide bridges gives these molecules a compact structure noted as colipase folding, with NH2- and COOH-ends on the surface (Boisbouvier et al., 1998; Kaser et al., 2003; Morales et al., 2010).

PKs are ubiquitously expressed thoughout the mammalian species. In general, the expression profile of PKs is dynamic. Regulators of PKs have been found at both transcriptional and post-transcriptional levels. At the transcriptional stage, the expression of PK1 and PK2 is modulated by hypoxia via the hypoxia-induced factor one binding site in their promoter regions (LeCouter et al., 2001, 2003a). Hormones such as steroids, follicle-stimulating hormone, and human chorionic gonadotropin (Brouillet et al., 2012a), high insulin concentration (Ujvari et al., 2018), and peroxisome proliferator-activated receptor gamma (Garnier et al., 2015) regulate PK1 in the reproductive tract. Instead, proneural basic helix-loop-helix factors (Zhang et al., 2007), circadian rhythm genes (Cheng et al., 2005), homeobox transcription factors, and granulocyte colony-stimulating factor regulate PK2 expression (Hoffmann et al., 2006). At the post-transcriptional level, the major prokineticin regulators are microRNAs (Su et al., 2017; Meng et al., 2019; Wang et al., 2019).

B. Prokineticin Receptors: Structure, Distribution, Regulation, and Signaling

Prokineticin receptors were originally called as GPR73a and GPR73b, orphan receptors. Molecular cloning and characterization of prokineticin receptors (PKRs) were independently performed by three different research groups in 2002 (Lin et al., 2002a; Masuda et al., 2002; Soga et al., 2002). PKRs consist of seven transmembrane (TM) segments connected by alternative extracellular and intracellular loops (ECL and ICL, respectively). Due to the presence of two Cys forming a disulphide bond between ECL2 and ECL3, they belong to the “rhodopsin-like” or class A GPCRs, as the chemokine and opioid receptors (Vincenzi et al., 2022). Although PKRs are widely distributed and commonly co-expressed in various organs and tissues, analysis of expression pattern revealed that PKR1 is mainly localized in peripheral tissues such as the gastrointestinal tract, adipose tissue, endocrine glands, reproductive organs, lungs, heart, and hematopoietic cells, whereas PKR2 is mainly found in the central nervous system (CNS) (Masuda et al., 2002).

Human and mouse PKRs are 80% identical. PKR1 and PKR2 have approximatively 85% amino acid homology, with most differences localized in the NH2-terminus (Lin et al., 2002a). The genes encoding PKRs are organized in three exons and located on two different chromosomes: the pkr1 gene on human 2p.13.3 and mouse chromosome 6; the pkr2 gene on human 20p13 and mouse chromosome 2 (Parker et al., 2000; Lin et al., 2002a). Pkr2 gene is in a region with high mutational propensity. Indeed, numerous pkr2 pathogenic rare mutations and polymorphisms have been associated with various pathologies, including Kallmann syndrome, Hirschsprung syndrome, idiopathic pregnancy loss, precocious puberty, mood disorders, methamphetamine abuse and other diseases as documented in the supplemental data (Supplemental Table 1). In addition, a splicing variant of PKR2 containing only the last three TM domains (TM4-7) has been identified as a functional receptor that can form heterodimers with PKR2 and can be upregulated in pathologic states (Lattanzi et al., 2019a).

PKRs are activated by nanomolar concentrations of PK1 and PK2. PK2 has a higher affinity than PK1 for PKRs and slightly higher affinity for PKR1 than PKR2 (Lin et al., 2002a; Soga et al., 2002). PK2β acts as a PKR1-selective ligand (Chen et al., 2005; Lattanzi et al., 2018), whereas PK2C binds both PKR1 and PKR2 (Lattanzi et al., 2022).

All these endogenous prokineticins were first recognized by NH2-end and ECL2 (extracellular site 1) of the PKRs. The insertion of the peptide sequence AVITG of prokineticins into an orthosteric TM-binding site (site 2) of receptors leads to the conformational changes required for activation of signal transduction. The orthosteric TM-binding pocket also accommodates small non-peptide antagonists of PKRs and is extremely conserved in PKR1 and PKR2 (Levit et al., 2011). Allosteric small-molecule binding site is located among TMs 3, 4, 5, 6, and 7 and accommodates PKR1 agonist that interacts with Arg144, Asn141, Gly219, and Phe300 within the PKR1 binding pocket (Gasser et al., 2015). Interestingly, the only different residue in the ECL2 (Val207 in PKR1 corresponding to Phe198 in PKR2) could be important for designing subtype-specific ligands and targeting PKR-mediated different pathologic conditions (Levit et al., 2011). Intracellular loops (ICL2 and ICL3) of PKRs are essential for interaction with G proteins and mutations at these sites can reduce G-coupling, leading to loss of function (Peng et al., 2011). Activated PKRs trigger multiple intracellular signaling pathways via the coupling of several heterotrimeric G proteins (Fig. 2), such as Gq/11, Gi/o and Gs or selectively binding to β-arrestin 2 (Casella and Ambrosio, 2021). The genetic inactivation of PKRs in mice displayed developmental defects and organ dysfunction as summarized in Table 1.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 2

Intracellular signaling pathway triggered by PK/PKR binding. PKR1 exert its biologic activity by activating Gi, Gs, Gq signaling pathway, whereas PKR2 uses G12/13, Gs and Gi pathways dependent on cell type, expression levels, and pathologic condition. Created with BioRender.com.

The activation of PKRs and the interaction with Gq protein mediate the intracellular mobilization of calcium (Ca²⁺) through the activity of phospholipase-C and the formation of inositol triphosphate (Lin et al., 2002a; Masuda et al., 2002). Moreover, PK2 can induce translocation of protein kinase C (PKC)-ε to the plasma membrane in neurons via Gq (Vellani et al., 2006). Activation of PKRs/Gq sensitizes transient receptor potential cation channel subfamily V member 1 (TRPV1) and enhances the hyperalgesia mediated by PK2 (Negri et al., 2006a; Vellani et al., 2006; Maftei et al., 2020). PKR2 can also couple with G12 in endothelial cells (ECs) and reduce the expression of tight junction proteins, such as zona occludens-1, leading to fenestrations of these cells and increased vascular permeability (Urayama et al., 2007). PKRs can also signal through Gi-coupling. For example, Lin et al. (Lin et al., 2002b) reported that pertussis toxin pre-treatment inhibits the PK1/PKR1-induced p44/p42 mitogen-activated protein kinase (MAPK) cascade, suggesting Gi protein involvement. PK1, via Gi, can also trigger phosphorylation of Akt through activation of the phosphatidylinositol 3-kinase leading to EC differentiation, proliferation, and migration (Lin et al., 2002b). The ability of PK2 to stimulate phosphorylation of p44/p42 MAPK in both PKR1- and PKR2-transfected cells has also been described (Lin et al., 2002a). Moreover, the activation of both PKRs induces cAMP accumulation via the Gs protein and regulates, for example, contractility (Chen et al., 2005; Nguyen et al., 2013).

C. Exogenous Prokineticin Receptor Ligands: Antagonists

Several inhibitors of PKRs have been patented by pharmaceutical companies, but, details of their design have not been published. The first hits for such series were likely found as results of screening large high-throughput screening libraries, which were then optimized to increase activity and improve their ADMETox properties.

Scientists of Takeda Pharmaceutical Company identified two series of PKR1 antagonists, illustrated by compounds 1 (Goldby et al., 2015) and 2 (Mitchell and Teall, 2016) (Fig. 3), for the treatment of gastrointestinal, psychiatric, and neurologic disorders. Thompson et al. (Thompson and Melamed, 2007) of Merck & Co. patented a series of compounds, such as PKR-A, which acts as a PKR2 antagonist with IC50 < 10 μM. Cheng et al. showed that PKR-A inhibits PKR2-induced Ca²⁺ release with an IC50 of 48nM. In vivo, it reduces infarct volume and central inflammation while improving functional outcome in a rat model of cerebral ischemia (Cheng et al., 2012). Its analog PKRA7 non-selectively inhibits both PKR1 and PKR2 in low nanomolar range. In vivo, PKRA7 inhibits angiogenesis in gliomas and blocks myeloid cell infiltration in pancreatic cancer (Curtis et al., 2013).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 3

Representative small molecules as PKR1 and PKR2 ligands. PC1, PC7, and PC27 are the PKR1 preferring antagonist. PKR-A and PKRA7 are PKR2 selective and preferring antagonist, respectively. IS20 is a PKR1 selective agonist. Created with BioRender.com.

Balboni et al. (Balboni et al., 2008) utilized patented nonpeptidic PKR antagonists deposited by Janssen and Merck as the basis for developing a series of triazinediones that selectively inhibit PKR1 and, to a lesser extent, PKR2. This group created homology models of the human PKR1 and PKR2 (universal protein codes: Q8TCW9 and Q8NFJ6, respectively) using the crystal structures of the human kappa opioid (Protein Data Bank code: 4DJH) and neurotensin-1 (Protein Data Bank code: 4GRV) receptors as templates. In this way, docking studies allowed the authors to design some triazinediones as new PKR1 antagonists, such as PC1 (Lattanzi et al., 2015a) (Fig. 3). PC1 was designed to mimic the N-terminal AVITG sequence, while the methoxybenzyl moiety acts as an isosterol of the indole of the Trp24. PC1 inhibits intracellular Ca²⁺ mobilization induced by both PKR1 and PKR2. In vivo, it alleviates various disorders triggered by overactivation of PK2 signaling, such as inflammatory (Giannini et al., 2009) and neuropathic pain (Maftei et al., 2014; Guida et al., 2015; Lattanzi et al., 2015a,b; Castelli et al., 2016; Moschetti et al., 2019a) and Alzheimer’s disease (Maftei et al., 2019). The analog PC7 shows enhanced analgesic effects and also attenuates experimental autoimmune encephalomyelitis and preeclampsia in mice (Abou-Hamdan et al., 2015; Reynaud et al., 2021). Further improvements led to PC27, which exhibits analgesic activity with an EC₅₀ ten times lower than that of PC7 (Congiu et al., 2014).

1. Prokineticin Receptor Ligands: Agonists

A similar approach of Balboni et al. (Balboni et al. 2008) was used to develop a series of dehydroamides as agonists of PKR1 (Gasser et al., 2015). PKR1 sequence alignment and protein folding using “CPHmodels” software tools were used to identify the best template crystal structure, which was the turkey β1 adrenergic receptor (Protein Data Bank ID: 2VT4). Whereas previous homology studies examined traditional structure-based design, a deep-learning method based on recurrent neural network and Monte Carlo Tree Search was used to generate a new set of putative PKR1 agonists (Karpov, 2022) (Fig. 4). These studies were based on recurrent neural networks, which were pre-trained on the ChEMBL database (Mendez et al., 2019) as described elsewhere (Xia et al., 2020). Among the putative hits, a series of dehydroamides, such as IS20 (Fig. 3), were found to act as selective, biased PKR1 agonists that bind the Arg144, Asn141, Gly219, and Phe300 in the allosteric pocket of PKR1 (Gasser et al., 2015). IS20 shows potent cardioprotective effects in mouse models of heart failure and doxorubicin (DOX)-induced cardiotoxicity (Gasser et al., 2015, 2019). It also shows potent neuroprotective effects in a model of Parkinson’s disease (PD) (Neal et al., 2018).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 4

Illustration of the workflow used to generate new PKR1 agonists. These studies were based on recurrent neural networks, which were pre-trained on the ChEMBL database. A powerful machine learning algorithm have been developed which was the first trained on the SMILES canonicalization task using public ChEMBL molecules with the following fine-tuning on PKR1 data. ADMETox filters developed Transformer CNN (https://jcheminf.biomedcentral.com/articles/10.1186/s13321-020-00423-w) as well as molecular docking scores were used as reward to select compounds and fine-tune the generator. Molecular dynamic simulations were used to prioritize the generated compounds for experimental testing. Created with BioRender.com.

A. Role of Prokineticin in Heart Development

Congenital heart defects are the principal birth defects that can lead to a variety of congenital diseases in adulthood. The epicardium, composed of heterogenous epithelial cells, covers the heart and plays a key role in cardiac development and regeneration. It has been implicated in potential repair strategies for the heart. The epicardium undergoes epicardial mesenchymal transformation to form EPDCs. EPDCs can differentiate into many cardiac cells (ECs, pericytes, smooth muscle cells) and play an important role in myocardium maturation. Previous studies have shown that epicardial-PKR1 signaling plays a role in cardiac development, and that a defect in epicardial signaling due to a deficiency of PKR1 in the epicardium leads to forms of congenital heart disease in the adult heart (Boulberdaa et al., 2011). More specifically, genetic inactivation of PKR1 specifically in the epicardium of mice, using epicardial specific such as Wilms tumor 1 (WT1 _GFPcre) and Gata4cre mice, impairs epicardial mesenchymal transformation in the heart (Arora et al., 2016a). These mice showed inadequate development of coronary vascular networks and impaired interactions between EPDCs-cardiomyocytes, resulting in impaired cardiomyocyte proliferation and cardiac arrhythmia. All these abnormalities in cardiac developmenl lead to partial embryonic and postnatal mortality (Arora et al., 2016a) as seen in the total PKR1-knockout (PKR1(−/−)) (Boulberdaa et al., 2011). The impairment of the coronary vascular network in epicardial PKR1-deficient mice (PKR1−/−) is due to impaired differentiation of EPDCs into vasculogenic cell types. However, the reduced cardiomyocyte proliferation and contractile deficits result from abnormal release of epicardial paracrine factors (miRNAs) in the mutant hearts (unpublished observations). Lipid deposition in the mutant hearts is due to the lack of control of EPDCs’ differentiation into adipocytes as observed in vitro (Qureshi et al., 2018). Survival rates of adult mice were reduced by 80% after coronary ligation as a myocardial mouse model, suggesting that the mortality in these mice after myocardial infarction was due to congenital cardiac dysfunction.

B. Role of Prokineticin in Kidney Development

During kidney development, nephrogenesis and glomerulogenesis are important events. The mesenchymal cells form pretubular aggregates and undergo mesenchymal-epithelial transition (MET) to form nephrons during nephrogenesis. Nephrons are the filtering units of the kidney. However, glomerulogenesis occurs via angiogenesis and metanephric vasculogenesis through the differentiation of progenitor cells. The number of nephrons plays an important role in renal function, and any defect in nephron development can lead to hypertension problems later in life.

Studies in mice have shown that the PK2/PKR1 signaling pathway is essential for kidney development. Using two cre transgenic lines (Gata5 and WT1), ablation of PKR1 specifically in nephron progenitors caused partial embryonic and postnatal lethality because of a lack of MET in WT1⁺ renal mesenchymal cells. The defective MET leads to impaired proliferation and increased apoptosis in WT1⁺ renal mesenchymal cells, resulting in hypoplastic kidneys with premature glomeruli and necrotic nephrons. Moreover, PKR1 activation in cultured WT1⁺ embryonic kidney progenitor cells promote MET and accumulation of NFATc3 in the nucleus. Both events were alleviated by an NFATc3 inhibitor and siRNA for PKR1-mediated MET process in cultured WT1⁺ embryonic kidney cells, suggesting that PKR1 promotes the MET process via NFATc3 signaling. Similarly, PKR1(−/−) mice exhibited neonatal kidney disorders (Boulberdaa et al., 2011). Disruption of capillary angiogenesis and severe tubular defect were observed in endothelial-specific PKR1-knockout (ec-PKR1(−/−)) mice (Dormishian et al., 2013).

C. Prokineticin in Brain Development and Kallmann Syndrome

Neurogenesis begins when neuronal progenitor and stem cells begin differentiating division, resulting in the formation of neurons and glia in the cortical layers. Neurogenesis occurs in the embryonic to early postnatal stages. In the adult mammalian brain, neurogenesis also occurs in two neural niches, the subgranular layer of the dentate gyrus (DG) and the OB throughout life and after stroke during the recovery process. Indeed, PKR1 expression was observed in the granular and periglomerular layers of the OB. PKR2, on the other hand, is expressed in the subventricular zone (SVZ), the entire rostral migratory stream (RMS), and the ependymal and subependymal layers of the olfactory ventricle, where adult neurogenesis occurs (Cheng et al., 2006). In vertebrates, the hypothalamic-pituitary-gonadal axis controls reproduction through a pulsatile release of gonadotropin-releasing hormone (GnRH) from the hypothalamus. GnRH-expressing neurons are formed in the olfactory placode which directs their migration to their final destination, the hypothalamus. GnRH reaches the anterior pituitary, binds to GnRH receptor 1, and induces secretion of the two pituitary gonadotropins, luteinizing hormone and follicle-stimulating hormone, which ultimately stimulate steroid production and gametogenesis in both males and females (Plant, 2015).

Idiopathic hypogonadotropic hypogonadism (IHH) is a rare congenital disorder of the hypothalamic-pituitary-gonadal axis characterized by failure of gonadotropin secretion leading to delayed puberty and infertility. When congenital IHH is associated with anosmia, it is referred to as Kallmann syndrome (KS). Kallmann syndrome and IHH are genetically heterogeneous (Vezzoli et al., 2016). Like several GPCRs, PKR2 is also involved in GnRH neurons development and migration (Ng et al., 2005; Martin et al., 2011). Pkr2 and pk2 were identified as Kallmann genes by analyzing the phenotype of PKR2(−/−) or PK2(−/−) mice, which show many KS-like features. PKR2(−/−) or PK2(−/−) mice exhibit an OB that is reduced in size, has an altered architecture, and has an accumulation of neural progenitor cells (Ng et al., 2005), resulting in a reduced number of GnRH neurons in the hypothalamus. Lack of GnRH secretion is associated with low plasma levels of testosterone and follicle-stimulating hormone and impaired sexual development. Male PKR2(−/−) or PK2(−/−) mice show small testicular tubules with no lumen, Leydig cells with reduced interstitial space, and absent haploid spermatocytes and spermatids. Similarly, PKR2(−/−) or PK2(−/−) mice exhibit impaired estrous cycles due to the absence of mature follicles and corpora lutea (Matsumoto et al., 2006; Pitteloud et al., 2007).

The IHH phenotype is observed in mice only in the presence of homozygous pk2 mutations, whereas the KS phenotype also occurs in the presence of a heterozygous pk2 or pkr2 mutation, suggesting that additional mutations in disease-related genes may be involved in this case (Pitteloud et al., 2007; Cox et al., 2018). Pk2 and pkr2 mutations have also been observed in human patients with IHH and KS (Abreu et al., 2010). The severity of the phenotype of individuals with KS carrying pkr2 mutations in the heterozygous state is in most cases due to the synergistic effect of the pathogenic mutations. The pkr2 is involved in several digenic and trigenic associations such as PK2/PKR2, FGFR1/PKR2, PK2/GNRHR, and PKR2/CHD7/FEZF1 (Cole et al., 2008; Canto et al., 2009; Sarfati et al., 2010a,b; Méndez et al., 2015; Zhang et al., 2020).

Most Kallmann-pkr2 mutations localized in the highly conserved transmembrane domains alter the folding of receptor proteins retained by quality control system in the endoplasmic reticulum (Araki and Nagata, 2011). Some cell-permeable small antagonists/agonists have been described as molecules that can rescue the phenotype by interacting with intracellularly retained receptors to assist their folding and transport to the plasma membrane (Chen et al., 2014). Other mutations in the cytoplasmic domain of PKR2, which includes the first, second, and third ICL and the carboxyl tail, likely result in impaired G-protein coupling. Finally, pkr2-mutations in the extracellular domain impair ligand binding (Martin et al., 2011; Libri et al., 2014; Sbai et al., 2014).

The homozygous pkr2 founder mutation L173R is responsible for the infertility characteristic of KS. In contrast, patients with the heterozygous condition do not show infertility associated with GnRH deficiency, but exhibit selectively increased protection against Trypanosoma cruzi (T. cruzi) infection, which may explain the high occurrence of the heterozygous pkr2 L173R mutation in the human genome (Avbelj Stefanija et al., 2012; Lattanzi et al., 2021).

A paradoxical gain-of-function mutant of pkr2 was found in a patient with early puberty and designated TM1-5 because it lacks the last two transmembrane domains and the carboxyl-terminal tail. Analysis in cell culture showed that the TM1-5 mutant lacked signal transduction activity, but co-transfection of TM1-5 markedly increased ligand-induced signaling responses of cells expressing wild-type PKR2 (Sposini et al., 2015; Fukami et al., 2017). In 192 KS patients, Dodé et al. (Dodé et al., 2006) identified ten and four different point mutations in the genes encoding PKR2 and PK2, respectively. Interestingly, one patient carrying the p.R73C mutation in the pk2 gene suffered from severe sleep disturbance and marked obesity. Abreu et al. (Abreu et al., 2008) reported the case of three patients with KS who had mutations in the pkr2 gene (p.L173R or p.R268C mutations in the heterozygous state) and developed obesity without or with (in 1 of 3 patients) type 2 diabetes, chronic arterial hypertension, and hypertriglyceridemia. Recent studies have shown a loss of ∼75% of GABAergic interneurons in OB of both PK2 and PKR2 mutant mice, indicating that PK2/PKR2 signaling is an important player of survival and migration of GnRH neurons and is critical for tangential and radial migration of OB interneurons rather than affecting OB neurogenesis (Wen et al., 2019).

A. Prokineticin Receptor 1 Signaling in Myocardial Infarct-Induced Heart Failure

Myocardial infarction (MI) is the most common cause of HF leading to morbidity and mortality worldwide, despite the development of numerous therapeutic agents to treat HF. Expression of the entire PK system is increased three days after coronary ligation in a mouse model of MI (Nguyen et al., 2013). Only PKR1 levels declined rapidly after the compensatory remodeling phase of MI. Moreover, intracardiac pkr1 gene transfer after coronary ligation in a mouse model of MI reduces mortality and improves cardiovascular function by promoting angiogenesis and cardiomyocyte survival and increasing proliferation and mobilization of resident EPDCs toward the injured area (Urayama et al., 2009; Nguyen et al., 2013). Intraperitoneal administration of a PKR1 agonist, IS20, activates the AKT survival pathway in the heart. Moreover, IS20 treatment of mice with MI reduces mortality and improves cardiac function, which has a similar effect to intracardiac pkr1 gene transfer. Indeed, IS20 reduces apoptosis in cardiac cells, provokes proliferation of EPDCs, and increases capillary formation (Gasser et al., 2015).

B. Prokineticin Receptor 1 Signaling in Anticancer Drug-Mediated Heart Failure

Cardiotoxicity induced by anticancer drugs such as anthracyclines, targeted therapies, and immune checkpoint inhibitors causes ischemia, arrhythmia, hypertension, myocarditis, and cardiac dysfunction leading to HF. In mouse models of cardiotoxicity induced by the anthracycline doxorubicin (DOX), IS20 attenuates apoptosis and fibrosis and improves survival and cardiac function. Importantly, IS20 does not interfere with the antitumor effect of DOX in a mouse model of breast cancer (Gasser et al., 2019). In vitro activation or overexpression of PKR1 in cardiomyocytes activates AKT through phosphorylation to protect cardiomyocytes, EPDCs, and ECs from hypoxic insult (Guilini et al., 2010; Arora et al., 2016a) as well as DOX-mediated apoptosis (Gasser et al., 2019). DOX at high concentrations accumulates reactive oxygen species in cardiomyocytes, which are attenuated by IS20. The mechanism by which the detoxification pathway is activated is translocation of nuclear factor erythroid 2-related factor 2 (NRF2) to the nucleus, which increases expression of the detoxification gene.

In cardiac ECs, activation or overexpression of PKR1 induces ECs proliferation, migration, and branching to promote angiogenesis via Gα11-mediated regulation of both MAPK and AKT activity. Consistent with these findings, endothelial-specific PKR1(−/−) mice exhibit capillary rarefaction, apoptosis, and interstitial fibrosis, leading to cardiac and renal dysfunction. Indeed, these mice showed abnormal cardiac and renal insulin signaling, leading to ectopic lipid deposition (Dormishian et al., 2013).

C. Prokineticins/Prokineticin Receptor 2 Signaling in Development of Pathologic Hypertrophic Cardiomyopathy

Chronic overload of the heart caused by hypertension leads to pathologic hypertrophic growth of the myocardium and endotheliopathies such as vasoconstriction. It is not fully understood why there is a mismatch between the excessive energy demand of the myocardium during hypertrophic growth and angiogenesis in these pathologic events. Pathologic hypertrophy induced by transaortic constriction in mice increases the expression of PKR2 in their cardiomyocytes (Demir et al., 2021). Sustained activation of PKR2 activates the extracellular signal-regulated kinase 5 (ERK5) pathway to induce hypertrophy in cardiomyocytes. However, activation of the PKR2/Gα12/13/matrix metalloprotease pathway in cardiomyocytes leads to the release of activin A and soluble endoglin, which act as paracrine factors and induce endotheliopathies (vascular rarefaction) (Demir et al., 2021). Similarly, transgenic mice overexpressing PKR2 in their cardiomyocytes (TG-PKR2) showed hypertrophy and impaired endothelial integrity related to paracrine regulation (Urayama et al., 2009). Overall, pressure overload-mediated maintenance of PKR2 signaling in cardiomyocytes contributes to cardiac hypertrophy via autocrine signaling and to vascular rarefaction via cardiac cytokine-mediated communication between cardiomyocytes and ECs (Guilini et al., 2010; Alfaidy et al., 2019). In ECs cultured in vitro, activation of PKR2, unlike PKR1, does not induce angiogenesis but couples to Gα12-13-mediated degradation of zonula occludens one, a cell-cell adhesion molecule, at tight junctions and forms a fenestration of ECs. PKR1 expression is dominant in the physiologic state, whereas PKR2 becomes dominant upon pathologic stimuli (Guilini et al., 2010; Nebigil, 2016).

D. Prokineticin Receptor 1 Signaling Controls Fate of Adult Cardiac Transcription Factor 21-Positive Cardiac Fibroblast Progenitor Cells (Tcf21⁺CFP)

Transcription factor 21 (Tcf21) is expressed in mesenchymal tissues, including the epicardium, and plays a key role in cellular differentiation. Tcf 21⁺CFPs originate in the embryonic epicardium and continue to be expressed in quiescent adult CPFs with the promising potential for repairing injured heart. They localize to the epicardium, perivascular or interstitial areas, depending on the type of cardiac injury (Acharya et al., 2012). PK2 regulates the fate of adult tcf21⁺CFPs. It promotes vasculogenic transformation of tcf21⁺CFPs and inhibits their differentiation into adipocytes (Qureshi et al., 2017). Moreover, tcf21_TM^icre CFPs-restricted inducible knockout mice (tcf21-PKR1(−/−)) on a high-fat diet exhibit high levels of fat deposition in the pericardium, atrioventricular groove, and perivascular area, along with disrupted vascular networks, leading to impaired cardiac function (Qureshi et al., 2017).

PKR1 signaling also regulates EPDC activity in a paracrine manner. Overexpression of the pkr1 gene in transgenic mice heart does not cause structural and functional abnormalities in cardiomyocytes. In fact, increased PK2 production in their cardiomyocytes acts as a paracrine factor that promotes EPDC differentiation into endothelial and smooth muscle cells for neovascularization (Urayama et al., 2008).

A. Prokineticin 2 Levels in Patients with Metabolic Syndrome

To date, two studies have been conducted in adult patients to investigate the association between PK2 and metabolic syndrome. Wang et al. (Wang et al., 2016) measured serum PK2 levels in 162 middle-aged and elderly Chinese patients with cardiovascular risk factors. They found a positive correlation with several cardiometabolic risk factors, including blood lipids, fasting plasma glucose, HbA1c, blood pressure, body mass index (BMI), and uric acid. However, multiple logistic regression analysis showed that PK2 was independently associated with metabolic syndrome.

Mortreux et al. (Mortreux et al., 2019) demonstrated that plasma PK2 was lower in participants with diabetes mellitus type 2 compared with nondiabetics in the D.E.S.I.R. cohort (Data from an Epidemiologic Study on the Insulin-Resistance syndrome), but this association disappeared after adjustment for BMI and/or caloric intake. In univariate regression studies, they showed that PK2 was significantly inversely associated with BMI, waist circumference, fasting blood glucose, HbA1c, and low-density lipoprotein cholesterol. In a multivariable model, BMI, energy intake, and plasma low-density lipoprotein cholesterol remained associated with PK2 levels.

B. Prokineticin Receptor 1 in Insulin Resistance

In blood vessels, the transcapillary passage of insulin from ECs to skeletal muscle is rate-limiting for the control of insulin-stimulated glucose uptake (Kubota et al., 2011). The impaired process of insulin delivery from ECs is a crucial step for the development of insulin resistance. Overexpression of PKR1 in ECs promotes not only angiogenesis but also transendothelial uptake of insulin, suggesting that PKR1 is a positive regulator of insulin uptake (Von Hunolstein and Nebigil, 2015). Interestingly, EC-specific PKR1 knockout mice (EC-PKR1(−/−)) show impaired capillary formation, low transcapillary insulin uptake, glucose and insulin sensitivity, resulting in polyphagia, polydipsia, and polyuria. EC-PKR1(−/−) mice have loss of adipose tissue with macrophage infiltration and fibrosis, resulting in severe lipodystrophy (Dormishian et al., 2013). This insulin resistance in EC-PKR1(−/−) could be rescued by pkr1 gene transfection with an adenovirus carrying PKR1 cDNA. The EC-PKR1(−/−) phenotype is reminiscent of peripheral insulin resistance, as human patients with type 2 diabetes exhibit impaired insulin secretion and endothelial dysfunction (Kolka and Bergman, 2013).

C. Prokineticin 2/Prokineticin Receptor 1 Pathway in Diabetes-Induced Cardiomyopathy

Diabetic cardiomyopathy is characterized by the structural, functional, and metabolic changes in the heart associated with diabetes, which leads to HF. Recently, studies in vitro and in vivo have shown that a drug used to treat type 2 diabetes, metformin, exerts a cardioprotective effect via the PK2/PKR1 pathway (Yang et al., 2020). Interestingly, the expression of PK2, PKR1, and PKR2 is reduced in diabetic mice. Accordingly, the low phosphorylated active form of Akt and glycogen synthase kinase-3 beta was attenuated by metformin treatment in these mice. Moreover, high glucose-mediated cardiomyocyte injury is protected by metformin or PK2 treatment, which can be reversed by a PKR1 antagonist (PC7) or an AKT inhibitor, indicating that metformin exerts a cardioprotective effect via activation of PKR1 pathway. Whether PKR1 agonists protect the heart from diabetes-associated cardiac remodeling remains to be investigated.

D. Role of Prokineticin Receptor 1 in Diabetes-Mediated Skeletal Muscle Dysfunction

Diabetes mellitus type 2 can also be caused by insulin resistance in skeletal muscle. The PKR1 pathway is involved in the regulation of metabolic function in murine myoblasts, satellite cells, and their differentiated myotubes via Gq-mediated phosphatidylinositol 3-kinase/AKT and MAPK/ERK signaling pathways. PKR1 promotes the translocation of glucose transporter 4 to the plasma membrane of skeletal muscle cells. In the model of palmitate-induced insulin resistance in myotubes, PKR1 increases insulin-stimulated glucose uptake and glucose transporter 4 translocation. Low levels of PKR1 were detected in obese mice induced to become obese by a high-fat diet and in human skeletal muscle cell-derived myotubes under conditions of insulin resistance. Taken together, these results suggest that PKR1 plays a key role in insulin sensitivity and may be a potential therapeutic target to improve skeletal muscle function in insulin resistance conditions (Mok et al., 2021).

E. Role of Prokineticin Receptor 1 in Diabetes-Mediated Renal, Neuronal, and Testicular Dysfunction

Diabetic nephropathy is characterized by severe glomerular and tubular damage, fibrosis, increased glomerular mesangial matrix, thickened basement membrane, exfoliated renal tubule brush border, and massive loss of podocytes. Proinflammatory and pro-fibrotic signaling pathways in glomerular and renal tubular cells lead to diabetic nephropathy. In db/db mice, low expression of PK2 and PKR1 is associated with low levels of phosphorylated Akt signaling. However, levels of PKR2 and phosphorylated ERK did not change significantly. A recent study showed that geniposide, a bioactive compound found in a variety of medicinal herbs such as Gardenia jasminoides, ameliorated renal damage in db/db mice with diabetic nephropathy, similar to WT1-PKR1(−/−) mice renal defects such as high glomerular and tubular injury, decreased WT1 in glomerular podocytes, and massive loss of podocytes (Arora et al., 2016b). The PK2/PKR1 pathway is involved in the protective effects of geniposide. Geniposide increases the expression levels of PK2, PKR1, and active Akt. It has been shown to be effective in the clinical treatment of diabetic nephropathy (Dai et al., 2022).

Matrine (Mat) is an active antidiabetic, cardioprotective, and neuroprotective component of Sophora flavescens ait root extracts. Mat administration increased protein expressions of PK2, PKR1, and PKR2 in the hippocampus, which decreased significantly in diabetic mice. In addition, Mat could also improve the diabetes-induced impairment of spatial learning and memory by alleviating endoplasmic reticulum stress and, in part, modulating PK2/PKR signaling (Zhang et al., 2022a).

More than 85% of male patients with diabetes develop testicular dysfunction. In streptozotocin-mediated mouse models of type 1 diabetes, reproductive capacity was found to be significantly impaired due to testicular dysfunction. This testicular dysfunction was ameliorated by metformin administration via the PK2/PKR pathway. The reduction of p-Akt and p-glycogen synthase kinase-3 beta in diabetes-induced testicular damage was normalized by metformin via the PK2/PKR1 pathway by attenuating apoptosis and inducing autophagy (Liu et al., 2019).

A. Prokineticin Levels in Obese Patients

Recently, Wang et al. (Wang et al., 2021) investigated the association between PK2 and childhood obesity. They demonstrated that children with obesity had significantly elevated fasting serum PK2 levels compared with age-matched healthy children. This positive correlation between serum PK2 levels and BMI was demonstrated in obese children with and without nonalcoholic fatty liver disease, which is a common and serious complication of childhood obesity. Although obesity is generally associated with a chronic low-level inflammatory state and numerous studies in rodents have demonstrated the role of PK2 in the inflammatory process, there was no correlation between PK2 and IL-6, tumor necrosis factor alpha, and white blood cell and neutrophil counts in this cohort study. Finally, circulating PK2 levels were found to correlate positively with homeostatic model assessment for insulin resistance in the entire cohort.

Interestingly, the obese children did not exhibit significant abnormalities in sleep and circadian rhythms, suggesting that high circulating PK2 levels are not mainly caused by a disturbed circadian rhythm.

These data may be related to another study in patients with a complete loss-of-function mutation in the pk2 gene (Balasubramanian et al., 2014). In these patients, no abnormalities were observed in circadian phase markers (e.g., melatonin, cortisol, and core body temperature), at the expense of their psychomotor vigilance performance. Although limited by small sample size, this study suggests that the activity of the central circadian pacemaker is intact despite the inactivity of PK2, suggesting that the regulation of circadian phenotype is not matched in humans and mice.

All these human studies on the relationship between PK2 and obesity lead to quite heterogeneous results and conclusions. The discrepancies may be related to the relatively small sample size of the cohorts, the different clinical characteristics of the different study cohorts (including age and sex distribution), the medical complications of the participants, the differences in the measurement of PK2 levels, etc. This implies that future studies with larger sample sizes detecting PK2 under different conditions and in different populations are needed. Either way, all these data suggest that PK2 or its receptors may be a therapeutic target for the treatment of obesity and related diseases.

B. Regulation of Prokineticin Signaling in Neuronal Basis of Obesity

The etiology of obesity is a complex process of dysregulation between food intake, energy expenditure, and energy stores which can be observed at both central and peripheral levels (Hill et al., 2012). In the CNS, the nucleus arcuate of the hypothalamus regulates food intake by balancing two neuronal circuits that secrete peptides with two opposite functions: an anorexigenic one triggered by pro-opiomelanocortin and the cocaine- and amphetamine-regulated transcript, and an orexigenic one triggered by neuropeptide Y (NPY) and the Aguti-related peptide. It is dysregulation in this system that increases or decreases food intake (Sohn, 2015). Food intake is affected not only by changes at the level of the hypothalamus, but also by changes at the level of adipose tissue (Würfel et al., 2022).

Negri et al. (Negri et al., 2004) were the first to study how injection of Bv8/PK2 into different brain regions can affect intake behaviors. They found that intracerebroventricular injection of Bv8/PK2 in rats reduced food intake but stimulated drinking. However, injection of Bv8/PK2 into the arcuate of the hypothalamus selectively suppressed eating but not drinking. In contrast, injection of Bv8/PK2 into the subfornical organ stimulated drinking but not eating.

Gardiner et al. (2010) have reported that in rodents, intracerebroventricular administration of Bv8/PK2 leads to an anorectic effect that is abolished by PK2 antibodies. PK2, even when administered to rodents via the peripheral route, leads to a regulation of food intake that acts via the brainstem and depends on activation of PKR1 (Gardiner et al., 2010; Beale et al., 2013). Indeed, PK2 administered intraperitoneally increases immunoreactivity in the dorsal motor vagus nucleus of the brainstem and induces anorexia in wild-type and PKR2(−/−) mice, but not in PKR1(−/−) or in wild-type mice treated with PC1, a PKR1-preferring antagonist (Negri et al., 2004; Beale et al., 2013).

Moreover, the PK2-mediated anorexic effect is partially dependent on arcuate of the hypothalamus activation of the melanocortin system. Indeed, PK2 induces c-fos immunoreactivity in pro-opiomelanocortin -expressing neurons. In hypothalamic explants, PK2 stimulates the release of alpha-melanocyte-stimulating hormone, a pro-opiomelanocortin-derived anorexigenic peptide that binds to the melanocortin-4 receptor and plays an important role in recognizing the balance between orexigenic Aguti-related peptide and anorexigenic melanocortin signaling and regulating feeding behavior. In contrast, simultaneous intracerebroventricular administration of PK2 with Aguti-related peptide, an orexigenic peptide, significantly reduced the anorectic effect of PK2 (Gardiner et al., 2010). The anorectic effect of intracerebroventricular administration of PK2-mediated anorectic effect is even more pronounced in melanocortin-4 receptor(−/−) mice (Chaly et al., 2016).

Melanocortin receptor accessory protein 2 (MRAP2), a regulator of energy homeostasis, enhances melanocortin-4 receptor signaling such that loss-of-function mutations in the Mrap2 gene are associated with obesity and hyperglycemia (Asai et al., 2013; Baron et al., 2019). MRAP2 and PKR1 co-localize in neurons. Moreover, MRAP2 inhibits the PKR1 cell surface trafficking via its C-terminal region (Rouault et al., 2017), acts as a suppressor of PKR1 signaling, and promotes food intake and weight gain, which can be reduced by activation of PKR1 (Chaly et al., 2016). MRAP2 C-terminal region also binds to the N-terminal region of PKR2, preventing glycosilation and transport on the cell surface (Verdinez and Sebag, 2021; Fullone et al., 2022a,b). In ex vivo hypothalamic explants, PK2 reduces MRAP2 expression. In adipocytes of PKR1(−/−) mice, which serve as models of obesity, MRAP2 expression is markedly increased (Fullone et al., 2022b).

Chronic administration of PK2 leads not only to a decrease in food intake but also to a decrease in body weight in lean and obese mice (Beale et al., 2013), in part due to the release of alpha-melanocyte-stimulating hormone via activation of signal transducer and activator of transcription 3 (STAT3) and ERK (Gardiner et al., 2010; Beale et al., 2013; Maftei et al., 2021). Conversely, PK2β does not induce STAT3 and ERK phosphorylation and, when injected intraperitoneally into mice, does not reduce food intake, likely because it cannot activate STAT3, a transcription factor whose dysregulation leads to obesity (Jiang et al., 2013a; Maftei et al., 2021).

C. Regulation of Energy Expenditure in Olfactory Bulb by Prokineticin System

The OBs help coordinate food selection and intake, and because they express high levels of PK2, which is involved in OB neurogenesis (Ng et al., 2005), they represent another important central area involved in food intake. PK2 injected into the OB of mice has an anorectic effect, whereas PK2 short hairpin RNA injected into the OB causes dysregulation of feeding behavior. In addition, OB shows a decrease in PK2L and PKR1 levels in fed mice compared with fasting mice. PK2L is cleaved into PK2β leading to the hypothesis that, at least at OB, this is the major PK2 isoform that regulates food intake by binding to PKR1 (Mortreux et al., 2019).

D. Role of Prokineticin 2 in Torpor and Temperature Regulation

The neurons of the paraventricular nucleus (PVN) receive a variety of inputs from different areas of the brain, integrate them, and transmit various outputs through different neurotransmitters and pathways involved in controlling stress, metabolism, growth, reproduction, immune system, and autonomic functions in the gastrointestinal, renal, and cardiovascular systems. The PVN also controls the central regulation of food intake and heat production. PK2 and PKR2 are expressed primarily in the PVN of the hypothalamus. Fasting increases PK2 expression in the PVN, which is decreased to undetectable baseline levels upon resumption of food intake (Zhou et al., 2012). PK2(−/−) mice respond to fasting with torpor, a compensatory mechanism for negative energy balance. This transient hypometabolic state is characterized by lower energy expenditure and weight loss in PK2(−/−) mice than in wild-type littermates. Moreover, in contrast to wild-type mice, no appreciable increase in arousal is observed throughout the fasting period. While daily ingestion of a limited amount of food during fasting rescued the body weight loss and hypothermic phenotype in wild-type mice, this was not the case in PK2(−/−) mice, which appeared unable to use food to compensate for body weight loss. It is suggested that increased expression of PK2 in the PVN during fasting may be a way to trigger physiologic responses, such as activation of the sympathetic nervous system, stimulation of locomotor activity, prolonged periods of wakefulness, and utilization of the body's energy stores. However, when PK2 expression is measured throughout the hypothalamus, it is not an increase but a decrease that is triggered by fasting (Gardiner et al., 2010).

Jethwa et al. (Jethwa et al., 2008) showed that targeted genetic disruption of PKR2- mediated signaling also predisposed mice to torpor when exposed to acute food deprivation, but also when maintained at room temperature (21–22°C) and ad libitum food. It was characterized by a marked decrease in body temperature, locomotor activity and respiratory quotient.

However, PKR2(−/−) mice showed comparable levels of hyperphagia to their control littermates after food deprivation, but their respiratory quotient tended to increase more slowly after refeeding, suggesting that the period of food deprivation had resulted in greater reliance on catabolism of fat reserves in the mutant mice. Surprisingly, food intake was reduced in the PK2R(−/−) mice compared with their control littermates which was unexpected given the known anorectic role of PK2, but their body weight and abdominal fat depots were similar to their intact littermates. The author suggested that loss of PKR2 signaling primarily leads to reduced food intake and likely also to an attenuation of circadian rhythms, so that compensatory strategies such as torpor entry are used to maintain energy expenditure.

E. Role of Prokineticin on Controlling Excessive Fat Formation, Insulin Resistance, and Epigenetic Regulation

The PK signaling suppresses the growth of visceral and subcutaneous adipose tissue. PK2 is released from adipose tissue and binds to PKR1, the receptor subtype expressed mainly in adipocytes (Soga et al., 2002), reducing adipose tissue growth (Szatkowski et al., 2013; Qureshi et al., 2017). Moreover, low PKR1 transcript has been found in obese patients visceral and subcutaneous tissues, demonstration of a key role of PKR1 in adipose tissue growth.

In vitro, PK2 suppresses mouse preadipocyte proliferation and their differentiation into adipocytes via PKR1 signaling, as observed in the human preadipocyte cell strain Simpson-Golabi-Behmel syndrome 10 days after the adipogenic stimuli. Accordingly, adipose tissue-specific PKR1(−/−) (ad-PKR1(−/−)) mice exhibit abnormally excessive accumulation of abdominal fat without altering in the glucose or insulin tolerance tests (Szatkowski et al., 2013). Both PKR1(−/−) and ad-PKR1(−/−) mice suffer from obesity as a result of adipocyte hyperplasia and macrophage infiltration in their adipose tissue. Interestingly, only PKR1(−/−), but not ad-PKR1(−/−) mice, exhibit insulin resistance and type 2 diabetes, suggesting that nonadipocyte incidents may contribute to the occurrence of the diabetes-like syndrome in PKR1(−/−) mice. Despite the increased body weight, PKR1(−/−) mice did not have increased food intake at 40 weeks and their body weight and visceral adipose tissue mass did not increase further on the high fat diet. The increase in fat mass in PKR1(−/−) on the normal diet is associated with chronic low-level inflammation and a marked increase in the expression of adipokines, which in turn cause a shift in the polarization of adipose tissue macrophages from an anti-inflammatory M2 to a pro-inflammatory M1 state (Mantovani et al., 2004). PK2 promotes the mouse M1 type of macrophage (an inflammatory phenotype) (Martucci et al., 2006) and attenuates the production of anti-inflammatory cytokines such as interleukin (IL)-10 and IL-4 in splenocytes (Franchi et al., 2008; Lattanzi and Miele, 2021). Whether PKR1 signaling promotes a switch from the pro-inflammatory M1 phenotype to the anti-inflammatory M2 phenotype to preserve adipocyte function in obesity remains to be investigated.

One of the important events linking obesity and cardiovascular disease may be the excessive development of epicardial adipose tissue (EAT) between the myocardium and the visceral layer of the epicardium. EAT may differentiate into pericoronary EAT and infiltrate and surround the coronary arteries (coronary EAT), contributing to the development and progression of coronary artery disease and atrial fibrillation. Interestingly, human EPDCs derived from atrial appendages spontaneously undergo adipocyte differentiation to form EAT in the high-calorie intact model. However, the PK2/PKR1 pathway activates the demethylase lysine demethylase 6A, which suppresses peroxisome proliferator-activated receptor gamma expression and inhibits adipogenic signaling and epicardial progenitor cell differentiation into adipocytes (Qureshi et al., 2018). On the other hand, PK2/PKR1-mediated lysine demethylase 6A suppresses repressive marks on vascular gene promoters mediated by histone tri-methylation of lysine 27 on histone H3 protein. Activation of vascular and endothelial cell precursors leads to differentiation of epithelial cells into vascular smooth muscle and ECs, such that PK2/PKR1 signaling promotes vascular formation. Epigenetic changes in human EPDCs by PK2 and PKR1 orchestrate stem cell formation and differentiation of hEPDCs into vasculogenic and adipogenic cells.

A. The Prokineticin System in Traumatic Brain Injury

TBI is the most common cause of death in trauma patients. It is characterized by persistent blood-brain barrier dysfunction and neuroinflammation that ultimately leads to cell death. In the adult mammalian brain, two neurogenic regions modulate physiologic functions: the subgranular zone of the dentate gyrus in the hippocampus and the SVZ, which produces neuroblasts that migrate to OBs via the RMS. Indeed, in TBI, proliferating cells mobilize from the SVZ into the injured cortex. However, the extent of the lack of change in RMS migration depends on the type of brain injury. In the mouse model with traumatic cortical injury, high expression of PK2 was detected in microglia. However, no PK2 level was found in reactive astrocytes, immature neurons, and leukocytes (Mundim et al., 2019), confirming that intact cortex does not express PK2. High levels of PK2 were found in astrocytes and neurons in the glutamate- (Cheng et al., 2012) and amyloid-beta-induced toxicity models (Severini et al., 2015), in which a PKR antagonist has beneficial effects (Caioli et al., 2017; Maftei et al., 2019). Indeed, PKR antagonism was detrimental in TBI because it inhibited SVT-derived neuroblast migration (Mundim et al., 2019). In the other study, migration of cells from neurospheres or SVZ occurred when neurospheres were co-cultured with PK2-expressing cells. These experiments demonstrated that PK2 plays a key role in the recruitment of SVZ cells to injured areas, which is an important compensatory restructuring in repair. Interestingly, PK2 upregulation in an injury model of zebrafish is associated with increased proliferation and migration of neuroprogenitors toward the injury site, indicating it may have a key role to induce a potential regenerative program (Ayari et al., 2010).

Another mechanism involved in neuroprotective pathway of PK2 is the inhibition of ferroptosis, a cell death program. Ferroptosis has been linked to the pathogenesis of various acute neuronal injuries, such as stroke and TBI (Xie et al., 2019) and chronic neurodegenerative diseases such as PD (Guiney et al., 2017) and Alzheimer's disease (AD) (Ayton et al., 2015).

Ferroptosis is characterized by the accumulation of phospholipid hydroperoxides (e.g., HOO- arachidonoyl-PE) catalyzed by iron-dependent mechanisms. A key enzyme in the biosynthesis of arachidonic acid-PE is the acyl-CoA synthetase long-chain family member 4 (Acsl4), which contributes to the performance of ferroptosis (Stockwell et al., 2017). PK2 has been shown to increase levels of Fbxo10, a ubiquitin ligase that binds to Acsl4 and promotes ubiquitination and degradation of Acsl4. PK2 inhibits ferroptosis in TBI, preventing mitochondrial dysfunction and protecting neurons from TBI (Bao et al., 2021).

B. The Prokineticin System in Parkinson’s Disease

Parkinson’s disease (PD) is a progressive neurodegenerative disorder caused by loss of nigrostriatal dopaminergic innervation and the appearance of Lewy bodies with aggregated α-synuclein.

As a disulfide-rich secretory peptide highly expressed in OB and the suprachiasmatic nucleus (SCN) (Ng et al., 2005), a new paradigm of compensatory protective function of PK2 signaling in response to neurotoxic stress has been documented in nigral dopaminergic neurons in experimental PD models (Gordon et al., 2016; Désaubry et al., 2020). PK2 expression in the adult mouse brain has been found to be either absent or sparse in the ventral midbrain, including the nigra (Cheng et al., 2006; Zhang et al., 2009), but high levels of PK2 expression have been found in surviving nigral dopaminergic neurons from brains, olfactory neurons (ON) and serum from patients with PD (Schirinzi et al., 2021, 2022) and in mouse models of PD (Gordon et al., 2016). Treatment with PK2 or overexpression of recombinant PK2 protects dopaminergic neurons from Parkinsonian neurotoxin-induced oxidative stress, mitochondrial dysfunction, and cell death, whereas antagonism of PKRs exacerbates dopaminergic degeneration in experimental PD (Gordon et al., 2016; Luo et al., 2019). Mechanistic studies revealed that activation of the survival signaling pathways ERK and AKT, as well as enhanced mitochondrial biogenesis, are involved in the potent neuroprotective effects mediated by PK2 (Schirinzi et al., 2021).

Recently, PK2 was shown to regulate a novel neuron-astrocyte signaling mechanism by promoting an alternative A2 protective phenotype in astrocytes (Neal et al., 2018). Astrocytes expressing high numbers of PKRs play an important role in postnatal neuroblast migration (Gengatharan et al., 2016). However, their dysfunction leads to neuronal death or neuronal dysfunction (Liddelow and Barres, 2017; Rivetti di Val Cervo et al., 2017). PK2 treatment or overexpression in primary astrocyte cultures and mouse brain increases the A2 phenotype of astrocytes, which reduces excitotoxicity and promotes long-term neuronal survival by modulating mitochondrial energy metabolism and antioxidant pathway and reducing inflammatory factors (Becerra-Calixto and Cardona-Gómez, 2017; Neal et al., 2018).

In primary astrocyte cultures, treatment with IS20, a small molecule PKR1 agonist, mimics the neuroprotective effect of PK2 and favors the A2 astrocyte phenotype over the pro-inflammatory A1 phenotype in neurodegenerative diseases such as PD (Neal et al., 2018). These findings collectively suggest that PK2 is upregulated and secreted during neurotoxic insults in dopaminergic neurons. The released PK2 subsequently activates astrocyte cells to promote a compensatory neuroprotective response against inflammatory stress by providing trophic support. PK2 also mediates survival of dopaminergic neurons via an autocrine effect on PKRs by activating the ERK and AKT signaling pathways and promoting mitochondrial biogenesis (Fig. 7).

• Download figure
• Open in new tab
• Download powerpoint

Fig. 7

A proposed model for PK2 to attenuate dopaminergic neurotoxicity and promote an alternative A2 protective phenotype in astrocytes in Parkinson disease. Basal PK2 expression is low or undetectable in the nigral region, but its expression is highly induced during the early stages of neurotoxic insult in dopaminergic neurons. This small regulatory peptide is then secreted from dopaminergic neurons and function as a neuroprotective signal to counteract inflammatory injury via binding and activation of PKRs on astrocyte cells. PK2 also mediates dopaminergic neuronal survival via an autocrine fashion acting on PKRs by activating the ERK and AKT signaling pathways and promoting mitochondrial biogenesis. Created with BioRender.com.

C. The Prokineticin System in Alzheimer’s Disease

Alzheimer’s disease (AD), the most common cause of dementia, is a neurodegenerative disease, characterized by the extracellular deposition of misfolded amyloid beta (Aβ) plaques and by intraneuronal tangles composed by the hyperphosphorylated tau protein (Karran et al., 2011).

Analysis of PK2 levels in human brain tissue samples collected postmortem from clinically well-documented and neuropathologically confirmed cases of AD revealed that PK2 expression levels in the hippocampus of AD patients are significantly higher than in cognitively normal controls. In addition, mean serum PK2 levels in AD patients are significantly higher than in controls, suggesting that blood PK2 is a potential biomarker for this pathology (Lattanzi et al., 2019b).

In a nontransgenic animal model of AD induced by intracerebroventricular administration of Aβ_1-42, rats show consistent deficits in learning and long-term memory in parallel with a strong upregulation of the PK system in the cortex and hippocampus that occurs on the first day after Aβ infusion and persists for 35 days (Lattanzi et al., 2019b; Maftei et al., 2019). Notably, in the hippocampus, PK2 and PKR1 are overexpressed in both neurons and astrocytes, whereas PKR2 is overexpressed only in neurons (Maftei et al., 2019). Subchronic treatment with PC1 improves learning and memory performance of Aβ_1-42-infused rats serving as control subjects by reducing overexpression of PK2/PKRs, decreasing Aβ_1-42-induced activation of microglia and astrocytes, and restoring neurogenesis. In the hippocampus of Aβ_1-42-infused rats, overexpression of the novel PKR2 splice variant, TM 4-7, occurs, and the expression ratio between TM4-7 and PKR2 increases with the progression of days after the Aβ_1-42 insult (Lattanzi et al., 2019a).

The increase in PK2 was also observed in another AD animal model. In Tg2546 transgenic mice, PK2 levels are statistically higher than in controls at 6 and 20 months of age, a period when the disease is neuropathologically detectable. In the hippocampus of the same transgenic mice, PC1 prevents the reduction in long-term potentiation without altering basal synaptic transmission (Severini et al., 2015). Indeed, in vitro study has also demonstrated that Aβ_1-42 in primary cortical cultures (CNs) increases PK2/PKR levels, and PC1, the PKR antagonist, protects CNs from Aβ_1-42-induced neurotoxicity in a dose-dependent manner by reducing Aβ_1-42-induced overexpression of PK2 (Severini et al., 2015). Moreover, incubation of CNs with Bv8 as well as with Aβ_1-42 leads to an increase in neuronal α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) currents that is prevented by PC1, demonstrating the involvement of PK2 in altering glutamatergic transmission in this model (Caioli et al., 2017). PK2 acts as a mediator of brain damage and plays a crucial role in Aβ-induced neuronal death.

D. Role of Prokineticin 2 in Electroconvulsive Shock-Induced Memory Impairment

Although electroconvulsive shock (ECS) has become the most effective treatment of depression (McClintock et al., 2014), its role in learning and memory impairments has been controversial. Recently, Chen et al. (Chen et al., 2023) demonstrated that acute ECS promotes the activation of A1 type astrocytes and the release of pro-inflammatory factors associated with spatial learning and memory function in rats. Indeed, the recovery phase of learning and memory function after ECS was facilitated by an increase in A2 subtype astrocytes, which was associated with an increase in the level of PK2 in astrocytes. A PKR2 antagonist, PKRA7, inhibits PK2-mediated activation of A2 subtype astrocytes, and attenuates the enhancement of spatial learning and memory function by PK2 in rats. PKRA7 did not alter the antidepressant effect of ECS, but it further exacerbates the synaptic defects caused by ECS. Moreover, PKRA7 administration in rat hippocampus increases the expression of postsynaptic density protein-95 (PSD-95), indicating that PK2/PKR2 signaling promotes synaptic plasticity (Chen et al., 2023). This study indicates an acute compensatory role of PK2 in triggering synaptic plasticity, activating A2 astrocytes, and improving learning and memory function.

A. Role of Prokineticin 2 in Sleep-Awakes and Neurobehavioral Neuronal Networks

In mammals, the duration and intensity of sleep are regulated by homeostatic processes, whereas the timing of sleep depends on circadian processes. PK2 expression in the SCN is rhythmic, preferentially absent at night and high during the day (Burton et al., 2015; Morris et al., 2021). PK2 suppresses GABAergic function to regulate circadian clock 2 (Per2). Antagonism or complete absence of PKR2 in mice reduces Per2 expression and favors GABAergic tone, reducing neuronal firing rate and output, weakening circadian coordination, and making the circadian system more susceptible to environmental perturbation (Li et al., 2006; Colwell, 2011). PK2- and PKR2-positive cells trigger circadian oscillations as pacemakers of the SCN and also tune their period and amplitude (Li et al., 2006). The firing rate of SCN neurons fluctuates in unison with the PK2 oscillation, likely due to the ability of PK2 to regulate the membrane translocation of TRPV2, an ion channel of the TRP family (Burton et al., 2015). Moreover, PK2(−/−) mice have increased rapid eye movement sleep during both the light and dark periods, although total sleep time decreases during the light period due to a reduction in non-rapid eye movement sleep time. These mice also have decreased arousal triggered by novel environments and increased alertness (Hu et al., 2007).

Despite the direct link between PK2/PKR2 and circadian cycle alteration in mice, the role of the PK2 pathway in human circadian rhythmicity remains to be elucidated. In two PK2(−/−) subjects, Balusabramanian et al. (Balasubramanian et al., 2014) reported no changes in circadian phase markers but impaired psychomotor vigilance task, suggesting that PK2 plays a role only in relaying circadian timing information to some neurobehavioral neuronal networks without affecting the function of the central circadian pacemaker.

B. Major Role of Prokineticin 2 in Mood Regulation and Stress

Disruption of circadian rhythms affects mood regulation and stress response, which is associated with mood disorders. Kishi et al. (Kishi et al., 2009) found a significant association between PKR2 and major depression and bipolar disorder in the Japanese population. Following this study, they also discovered a significant association between PKR2 and methamphetamine dependence, suggesting that pkr2 is altered as common circadian clock gene in mood disorders and drug dependence (Kishi et al., 2010). Otherwise, PK2 does not appear to play a role in the pathophysiology of methamphetamine addiction (Kishi et al., 2011). Intracerebroventricular infusion of PK2 enhances anxiety- and depression-like behaviors that are completely absent in PK2(−/−) mice. PK2(−/−) mice also showed a lack of adaptation to the new environment with low locomotor activity and arousal, no changes in body temperature and food intake. In addition, these mice lapse into torpor after fasting (Zhou et al., 2012). Similarly, PKR2(−/−) mice show spontaneous torpor that is enhanced by fasting (Jethwa et al., 2008).

A. Nociceptive Pain

In rodents, local or systemic injections of very low doses of Bv8 lower the nociceptive threshold to thermal, mechanical, and chemical stimuli by activating both PKR1 and PKR2 in primary sensory neurons (Negri et al., 2002). Systemic Bv8-induced hyperalgesia shows a typical biphasic pattern consisting of a first phase due to direct activation of nociceptors and a second phase due to central sensitization by Bv8-induced increased expression of calcitonin gene-related peptide and substance P (De Felice et al., 2012).

The increased excitability of nociceptors is the result of a cooperative interaction between PKRs and TRP channels; in particular, PKR1 colocalizes with TRPV1 in small dorsal root ganglia (DRG) neurons, whereas PKR2 colocalizes mainly with transient receptor potential cation channel subfamily A member 1 (TRPA1) in medium/large DRG neurons (Negri et al., 2006b; Vellani et al., 2006; Maftei et al., 2020). This is a functional co-localization, because PKR1(−/−) mice show impaired nociception at temperatures between 46°C and 48°C, to capsaicin, and to protons (stimuli that activate TRPV1), whereas PKR2(−/−) mice show impaired nociception at cold temperatures and to mustard oil-induced inflammatory hyperalgesia (stimuli that activate transient receptor potential cation channel subfamily A member 1). The protein kinase C(PKC)-ε pathway is involved in this interaction between PKRs and TRP (Vellani et al., 2006). Also in skin preparations, Bv8 induces peripheral sensitization of nociceptors to heat and lowers their nociceptive threshold through TRPV1 activation (Hoffmann et al., 2016). Nevertheless, other mechanisms are also involved in Bv8/PK2-induced hyperalgesia: in rat DRG neurons, PK2 increases proton-gated channel activity and sensitizes ionotropic P2X receptors (Ren et al., 2015), and in rat primary sensory neurons, PK2 suppresses GABA-A-activated currents (Xiong et al., 2010).

B. Neuropathic Pain

Peripheral nerve injury induced by chronic constriction injury (CCI) and spared nerve injury (SNI) in mice results in thermal hyperalgesia and tactile allodynia. In the injured sciatic nerve, upregulation of PK2 occurs as early as 3 days after injury and then shifts toward the DRG and spinal cord, where it becomes significant at 10 days and persists up to 17 days after CCI and SNI (Maftei et al., 2014; Guida et al., 2015; Lattanzi et al., 2015b). In the injured sciatic nerve, PK2 is detectable in some axons, mainly in association with activated Schwann cells and infiltrating macrophages. PK2 levels are also markedly increased in DRG neurons and satellite cells and in activated astrocytes in the spinal cord, but not in microglia. Treatment with the PKR antagonists, PC1 or PC7, starting on the day of injury blocks the onset of thermal hyperalgesia and tactile allodynia, delays the recurrence of painful symptoms after interruption of treatment, and restores the balance between pro- and anti-inflammatory cytokines (Maftei et al., 2014; Guida et al., 2015; Lattanzi et al., 2015b).

Both diabetes and chemotherapy drugs are responsible for severe neuropathic pain symptoms. In a mouse model of diabetic neuropathy induced by streptozotocin administration and in two mouse models of chemotherapeutic neuropathy induced by bortezomib or vincristine injection, severe hyperalgesia and allodynia are observed in association with upregulation of PK2, PKRs, and proinflammatory cytokines in the sciatic nerve, DRG, and spinal cord. In these mice, subchronic administration of PC1 impairs pain and reduces PK2/PKRs and pro-inflammatory cytokines (Castelli et al., 2016; Moschetti et al., 2019a, 2019b). PC1 protects against known neuronal cell changes after chemotherapy (Livni et al., 2019): in DRG neurons from bortezomib- or vincristine-treated mice, PC1 impairs the reduction in neurite length and growth induced by both chemotherapeutic agents (Moschetti et al., 2020).

C. Inflammatory Pain

PK2 plays a central role in inflammatory pain. In an animal model of inflammatory pain induced by Freund's Complete Adjuvant (FCA) in the mouse paw, the development and duration of hyperalgesia in the inflamed paw is associated with PK2 and PK2L levels and is abolished by local or systemic injection of PKR antagonists such as A-24 (Lattanzi et al., 2012) or PC1 (Giannini et al., 2009). Significant PK2 overexpression by granulocytes is not only detectable locally in the paw but also systemically (Giannini et al., 2009). In rodents, inflammatory stimuli lead to an early and rapid increase in plasma levels of granulocyte colony-stimulating factor, which activates PK2 transcription in bone marrow-derived CD11b+/Gr1 cells (Shojaei et al., 2007), explaining the systemic increase in PK2 levels in splenic and paw granulocytes. PK2 released from inflamed tissue triggers macrophage recruitment and stimulates migration of proinflammatory macrophages. In vitro, PK2 activates the release of proinflammatory cytokines (IL-1β and IL-12) and inhibits anti-inflammatory cytokines (IL-10) from macrophages. Bv8 also alters the Th1/Th2 balance and promotes a Th1-like response. PKR1(−/−) mice show a significant decrease in inflammation-induced hyperalgesia and less upregulation of PK2 (Martucci et al., 2006; Franchi et al., 2008). PKR2(−/−) mice also show reduced inflammation-induced hyperalgesia, but deletion of the pkr2 gene does not affect inflammation-induced PK2 upregulation. These data suggest that both receptors are responsible for inflammatory pain, but only PKR1 is involved in regulating the increase in PK2 expression (Giannini et al., 2009). In two animal models of inflammatory visceral pain (intrarectal injection of TNBS and mustard oil), increased PK2 levels due to gastrointestinal inflammation also trigger visceral pain via direct activation of PKRs (Watson et al., 2012).

All these data suggest an underlying role of the PK system in inflammatory and neuropathic pain, at least in rodents (Fig. 8). Therefore, reducing the expression of PK2 or antagonizing PKRs may represent a promising target strategy for a new therapeutic approach.

• Download figure
• Open in new tab
• Download powerpoint

Fig. 8

Comparison of prokineticin system alterations in inflammatory and neuropathic pain. In inflammatory and neuropathic pain, the production of circulating factors shifts toward those with a pro- inflammatory phenotype. PK2 is highly upregulated and blocking its receptors inhibits hyperalgesia in both inflammatory and neuropathic pain. Interestingly, the largest gap in our knowledge is the role of PK2 splicing variants in neuropathic pain and the involvement of the PKR2 pathway in inflammatory pain. Created with BioRender.com.

D. Pleasure

Touch is one of the four modalities (touch, temperature, nociception-pain, and proprioception-body position) of somatic sensation distributed throughout the body. Touch as a tactile stimulus is perceived by mechanoreceptors and also provides the affiliative or emotional somatic pleasure (McGlone et al., 2014). Affective or pleasant touch such as stroking, caressing or hugging is important for the exchange of social information (Morris, 1946) and behavioral development (Bales et al., 2018). Any form of physical affection that evokes pleasant touch sensations (cuddling, hugging, hand holding, caressing, soothing pats, etc.) plays a critical role in early childhood development and provide feelings of security, care, and support at every stage of life. Pleasant touch provides emotional and psychologic support to alleviate social isolation and stress. Touch plays an important role in pain relief (Leknes and Tracey, 2008). In patients with chronic pain, gentle touch has been shown to lose its analgesic function and is perceived as abnormally low, contributing to pain-related disorders (Gossrau et al., 2021).

Recently, a study in mice has shown that PK2 encodes and transmits pleasant touch information from skin to brain via a hard-wired spinal pathway. PKR2 neurons are a unique population of spinal excitatory interneurons that are 85% positive to G Protein-Coupled Receptor 83 (GP83) and predominantly receive monosynaptic and polysynaptic C fiber inputs (Liu et al., 2022). Mice with ablation of spinal PKR2 neurons exhibited a profound loss of pleasant touch sensation (significant reduction of the heart rate or analgesic effect in response to gentle stroking) together with loss of sensitivity to acute thermal pain and itch transmission. However, they displayed no significant alteration in temperature pain or inflammatory pain induced by capsaicin. Mice with conditional knockout of pkr2 in small nociceptive sensory neurons of DRGs displayed normal pain and itch behaviors, but they failed to show significant reduction of the heart rate or analgesic effect in response to gentle stroking. DRG-PKR2(−/−) mice had deficit in social novelty recognition and the heightened stress/anxiety-like behaviors. Indeed, PK2 has been identified as a neuropeptide that encodes sensation of pleasant touch and transmits it to spinal PKR2 interneurons, which conveys a positive pleasant touch signal to the brain via gpr83 neurons (Liu et al., 2022). This finding on the role of PK2/PKR2 in pleasant touch identified PK2 as a potential analgesic factor, and represent a promising target for the treatment of certain type of pain.

A. Prokineticin 2 as a Novel Immunomodulatory Factor in Diagnosis and Treatment of Sepsis

Sepsis is a highly lethal disease, and unfortunately, the exact causes of lethality remain poorly understood. Several studies suggest macrophage involvement in sepsis, which may alter the prognosis of the disease. PK2 levels were found to decrease dramatically in patients with sepsis and sepsis shock and to be associated with mortality (Yu et al., 2022). The authors concluded that PK2 may be a marker for diagnosing sepsis and septic shock and a marker for predicting mortality in adult patients with sepsis and septic shock. Accordingly, the role of PK2 in sepsis-related survival, bacterial burden, organ damage, and inflammation was investigated in an animal model with different sepsis models, such as cecal ligation and puncture–induced polymicrobial sepsis. In the mouse model of sepsis, PK2 levels were decreased as sepsis developed. In an in vivo model of polymicrobial sepsis induced by ligation and puncture of the cecum, administration of recombinant PK2 reduced organ damage and increased survival in both PK2(−/−) and wild-type mice. More specifically, PK2 significantly increased the phagocytic and bactericidal function of macrophages by activating STAT3. Depletion of macrophages attenuated the protective effect of PK2 against polymicrobial sepsis. Downregulation of PKR1, but not PKR2, with siRNA completely abolished the PK2-mediated enhancement of macrophage phagocytosis activity. Thus, the PK2/PKR1/STAT3 pathway was shown to promote bacterial clearance thereby playing a pivotal role in mitigating sepsis progression. It may not only represent a new target for sepsis therapy, but also an important marker for early diagnosis of sepsis, and a potential survival predictor.

B. Prokineticin 2 is Associated with the Pathogenesis of Collagen-Induced Arthritis in Mice

Rheumatoid arthritis (RA) is characterized by an increase in synovial cell proliferation and inflammatory cell infiltration in the synovial joints. Long-term inflammation in RA patients leads to cartilage and bone loss and causes joint deformities associated with chronic inflammation that affect quality of life. The hormones, neuropeptides sympathetic and sensory nerves connect the central nervous system (CNS) and peripheral tissues, resulting in morning stiffness and pain. Synovial cells play a key role in the chronic inflammatory tissue of rheumatoid arthritis.

PK2 levels were found to be significantly lower in synovial fluid than in plasma in patients with OA and RA. Moreover, plasma PK2 concentration correlated significantly with that in plasma in synovial fluid in OA patients but not in RA patients (Noda et al., 2021). In in vitro studies, they demonstrated that PK2 suppressed the secretion of IL-6 and MMP-3 from synovial fluid of patients in a concentration-dependent manner, which was abolished by the PKR1-preferential antagonist PC7. PKR1 was upregulated in the synovial tissue of OA patients. In addition, an anti-inflammatory effect of PK2/PKR1 signaling was mediated by inhibiting phosphorylation of NF-κB signaling. However, PK2 also suppressed IL-6 secretion from synovial fluid of RA patients without altering NF-κB phosphorylation and these effects were not abolished by PC7. The lower anti-inflammatory effect of PK2 in the synovial tissue of RA patient is due to the low expression of PKR1 and may be related to the chronicity of inflammation. This different regulation of PKR1 expression between RA-synovial fluid and OA-synovial fluid may contribute to the differential effect of PK2 in OS- versus RA-synovial fluid after exposure to inflammatory cytokines including IL-1β and tumor necrosis factor alpha. Indeed, injection of PK2 into healthy knee joints of mice resulted in increased inflammation due to mobilization of granulocytes but not macrophages into the synovial fluid. The inflammatory effect of PK2 can be observed in acute arthritis models, such as collagen antibody-induced arthritis (Noda et al., 2021).

The same authors have previously shown that PKR1 protein is expressed in infiltrating neutrophils, whereas PKR2 protein is present in macrophage-like mononuclear cells in the collagen-induced rheumatoid arthritis mouse model. In contrast to PKR1, pkr2 gene expression was higher in inflamed joints. PKRA7, an antagonist of both PKR1 and PKR2, significantly suppressed the severity of arthritis and inflammatory cytokines. However, the mechanism of how PKRA7 reduces arthritis has not been investigated in this mouse model (Ito et al., 2016). Therefore, PK2-mediated anti-inflammatory or inflammatory effects in arthritis are due to the effector cell type and receptor expression, acute or chronic phase of the disease, microenvironment in the presence of other pro-inflammation cytokines. This study should be verified by the effect of PKR1(−/−) or PKR2(−/−) or overexpression in tumor necrosis factor alpha-primed-OA- or RA-synovial fluids.

Recently, in the same arthritis model (collagen-induced arthritis type II), arthritic mice were shown to develop severe and prolonged hyperalgesia that was reversed by repeated administration of the PKRs antagonist PC1 (Impellizzeri et al., 2023). The anti-hyperalgesic effect of PC1 is due to its direct action on nociceptors and its ability to reduce the synthesis and release of PK2, since inflammation-induced PK2 upregulation is absent in mice lacking PKR1 (Giannini et al., 2009). In addition, PC1 is able to prevent the arthritis-induced increase in malondialdehyde in plasma, thereby reducing oxidative stress (Impellizzeri et al., 2023).

C. Role of Prokineticin in Infection

PK2 was recently identified as a novel modulator of bacterial pneumonia. Serum PK2 levels not only are lower in patients than in healthy controls, but also correlate with the severity of infection, being significantly lower in patients with severe pneumonia than in patients with simple pneumonia. In the mouse model of bacterial pneumonia induced by intratracheal instillation with Pseudomonas aeruginosa, PK2 levels in the lungs of infected mice are significantly lower than in that of control mice, confirming clinical findings and demonstrating a protective role of PK2 in the disease. This is due to the capacity of PK2 to increase the host's ability to eliminate bacteria by enhancing the chemotaxis, phagocytosis, and killing functions of macrophages. Because macrophages represent the most important cells required for the protective effect of PK2 in the host, this effect is lost when macrophages are depleted. In addition, PK2 increases the expression of NO and the activity of NOS in macrophages, which play a critical role in killing bacteria. These data, combined with the observation that administration of recombinant PK2 significantly alleviates lung injury and improves survival in mice, lead us to identify in PK2 a protective mediator factor and a potential adjuvant therapeutic for bacterial pneumonia (Tu et al., 2022).

Trypanosoma cruzi (T. cruzi), the etiological agent of Chagas disease, invades and resides in mammalian cells via a group of Gp85/trans-sialidase (TS). In particular, the specific LamG domain of Tc85, a group II TS, interacts with PKR2 and mediates parasite penetration and infiltration in mammalian cells (Lattanzi et al., 2021). Indeed, T. cruzi invasion into mammalian cells is inhibited by PKR2 interference or by treatment with anti-PKR2 antibodies or with a synthetic peptide (Khusal et al., 2015). The LamG domain of Tg85 not only binds to PKR2 but also activates it, demonstrating a physiologic role following T. cruzi infection of the nervous system by triggering STAT3 and ERK activation in mouse DRG (Lattanzi et al., 2021).

Olfactory dysfunction (OD), comprehending loss (anosmia) or reduction of the sense of smell, is one of the long-term effects of COVID-19. The expression of PK2 in olfactory neurons (ONs) of patients with post-COVID-19 OD has been significantly increased and correlated with the data obtained by odor tests to their residual odor, suggesting a contribution of PK2 in OD recovery (Schirinzi et al., 2023). Since efficient mitochondria in ONs are critical for proper olfactory signaling (Fluegge et al., 2012) and PK2 enhances mitochondrial bioenergetics (Neal et al., 2018), the authors hypothesize that PK2 may improve olfaction by stimulating mitochondrial activity or directly or counteracting the deleterious effects of chronic inflammation, which impairs mitochondrial activity exacerbating oxidative stress. The potential neuroprotective role of PK2 has recently been demonstrated in PD, where patients in the early stages of the disease have elevated levels of PK2 in ONs (Schirinzi et al., 2022). These data suggest the possibility that PK2 represents an alternative pharmacological target for the period after COVID-19. PK2 restores mitochondrial function and attenuates inflammation, thereby leading to recovery and improvement of odor.

A. Prokineticin Function in Reproductive System

Although PK2 is undetectable in the female reproductive organs, expression of PK1 was found abundantly in the stroma of the ovaries, particularly in the cells involved in steroid production in the hilar region (Ferrara et al., 2004; Alfaidy et al., 2016, 2019). The main expression of PK1 was found in the granulosa of the primary and secondary cells controlled by the ovarian cycle and to a lesser extent in the tertiary follicle (Fraser et al., 2005; Alfaidy et al., 2016, 2019). However, the pk1 and pk2 genes are strongly expressed in mammalian testis (Ferrara et al., 2004). The PK1 is mainly present in Leydig cells, whereas PK2 is present in the testicular tubules in primary spermatocytes.

Circulating levels of PK1 are markedly increased during pregnancy and it is the only prokineticin expressed in the placenta (Hoffmann et al., 2006; Reynaud et al., 2021). The peak of PK1/PKR1 expression occurs in the first trimester of pregnancy (Hoffmann et al., 2006). PK1 plays a key role in placental development throughout pregnancy (Brouillet et al., 2012b) and contributes to the birth process (Dunand et al., 2014). PK1 promotes trophoblast cell adhesion to fibronectin and laminin matrices, thereby mediating fetal-maternal dialog during implantation (Brouillet et al., 2012b). It may regulate the expression of implantation-related genes (Haouzi et al., 2009; Alfaidy et al., 2016, 2019). During the first trimester of pregnancy, PK1 controls extravillous trophoblastic cell migration, invasion, and pseudovascular network formation (Brouillet et al., 2010). PK1 also acts on fetal ECs in the stroma, increasing their proliferation, migration, invasion, and branching (Brouillet et al., 2010).

In the myometrium, PK1 induces proinflammatory cascades and increases myometrial smooth muscle cell contractility (Catalano et al., 2010), promoting both preterm and term births (Gorowiec et al., 2011). In the uteroplacental unit, PK1 and its receptors have been shown to provide intrauterine quiescence during the last trimester of pregnancy (Dunand et al., 2014). PK1 has been associated with early pregnancy pathologies such as ectopic pregnancies (Shaw et al., 2010) and miscarriages (Salker et al., 2010). Recently, mutations in the pk1, pkr1, and pkr2 genes have been reported to be associated with recurrent miscarriage (Su et al., 2014), but some variants have a protective role in the early stages of pregnancy (Su et al., 2013). Thus, it is not yet known whether the increase in PK1 levels is a cause or consequence of ectopic pregnancy or miscarriage.

B. Prokineticin Signaling Linked to the Reproductive System Disorders Such as Gestational Hypertension, Preeclampsia and Choriocarcinoma

Pregnancy-induced hypertension leads to complications in 5–11% of all pregnancies (Burton, 2009). PE, a form of pregnancy-induced hypertension, is the most threatening pregnancy pathology as it brings serious consequences to both the mother and the fetus (Tjoa et al., 2004). Recent studies have shown that PE not only affects the maternal cardiovascular system during pregnancy, but may also have long-term adverse effects, including maternal cardiovascular disease, vascular dementia (RR=3.46), AD (RR=1.45), and unspecified dementia (RR=1.40). These adverse effects occur approximately 10–20 years after an PE episode (Basit et al., 2018). The incriminated organ in the pathology of PE is the placenta, the key organ that ensures the success of pregnancy.

Adverse factors released by the placenta in PE include PKs, particularly PK1, which is abundantly produced in the first trimester of human pregnancy, with a peak of expression just prior to the establishment of the feto-maternal circulation (Hoffmann et al., 2006; Brouillet et al., 2012a; Alfaidy, 2016; Alfaidy et al., 2016). PK1 is mainly expressed in the endocrine unit of the placenta (the syncytiotrophoblast layer) and its receptors, PKR1 and PKR2, are highly expressed in the trophoblasts and microvascular cells of the placenta (Hoffmann et al., 2006; Brouillet et al., 2010, 2012b; Bardin et al., 2015; Traboulsi et al., 2015). The expression of PK1 in the placenta has been reported to be upregulated by hypoxia and β-hCG (Brouillet et al., 2010, 2012b).

PK1 controls early invasion of extravillous trophoblast cells during the first trimester of pregnancy and increases proliferation, migration, and permeability of trophoblasts and ECs (Brouillet et al., 2010). Moreover, circulating PK1 levels in nonpregnant women are approximately 50 pg/ml and increase 5-fold (≈ 250 pg/ml) during the first trimester of pregnancy and even higher (400–500 pg/ml) in the context of PE (Hoffmann et al., 2007; Brouillet et al., 2013). These results suggest that sustained PK1 levels beyond the first trimester of pregnancy could cause the development of pregnancy pathologies, including PE.

Mimicking the increase in circulating PK1 levels beyond 11.5 days post coïtus (dpc) in gravid mice, similar to the levels observed beyond the first trimester of pregnancy in women with PE, caused the development of PE symptoms in mice (Sergent et al., 2016; Reynaud et al., 2018). These symptoms were manifested by i) a failure to establish normal feto-maternal circulation, ii) the development of hypoxic placenta, a hallmark of stressed placenta, iii) a decrease in the expression of key genes known to be involved in trophoblast differentiation toward an invasive phenotype, iv) an increase in local and circulating anti-angiogenic factors (i.e., soluble fms-like tyrosine kinase and soluble endoglin) that have been shown to have effects on maternal vasculature and renal function and are known to be elevated in PE (Burton, 2009; Brouillet et al., 2012b; Cotechini et al., 2014).

Importantly, the maintenance of PK1 in the first trimester causes stress to the placenta, whereas its increase in the third trimester is associated with a compensatory mechanism to ensure the continuation of the pregnancy (Hoffmann et al., 2007; Sergent et al., 2016). In the developed model, a significant increase in placental efficiency was observed in the PK1-treated group (at 18.5 dpc), suggesting an adaptation of the placenta in situ, which was supported by an increase in placental vascularization and proliferation (Hoffmann et al., 2007; Sergent et al., 2016). Recently, the use of antagonists for PKR1 and PKR2 (PC7, PKR-A) have been reported to reverse some of PK1 adverse effects during gestation, as they blocked excessive trophoblast proliferation, migration, and invasion (Nguyen et al., 2013; Severini et al., 2015; Landucci et al., 2016; Traboulsi et al., 2017; Reynaud et al., 2021). In addition, a recent study by Reynaud et al. (Reynaud et al., 2021) demonstrated that PC7 and PKRA that were tested independently or in combination in trophoblast cells and during early gestation in the gravid mouse did not cause any adverse pregnancy outcome, suggesting that their potential consideration to treat PE.

A group of pregnancy-related disorders that originated from trophoblastic cells is called gestational trophoblastic diseases, which include a rare, aggressive neoplasm called choriocarcinoma. The development of choriocarcinoma is associated with a previous gestational event (molar, physiologic, ectopic pregnancy, selective or spontaneous abortion) (Bruce and Sorosky, 2023). One study has shown that PK1 and its receptors may also be directly involved in the development of gestational choriocarcinoma (Traboulsi et al., 2017). Increased levels of PK1 were found in both placenta and bloodstream of gestational choriocarcinoma patients. Indeed, PK1 increased proliferation, migration, and invasion of the choriocarcinoma cell line (JEG3) in two-dimensional and three-dimensional tumor cell spheroid systems. The PKR2 antagonist (PKR-A) significantly reduced tumor development and progression and preserved pregnancy in animal models of gestational choriocarcinoma by JEG3 injection in the placenta (Traboulsi et al., 2017). Thus, the use of PKR2 antagonist can be a therapeutic option for the treatment of gestational choriocarcinoma.