Regulation. PTN expression is regulated in a temporal and cell type-dependent manner (also see "Expression"). PTN expression has been reported to be regulated by: cytokines and growth factors, such as midkine in a compensatory manner (Herradon et al., 2005), platelet-derived growth factor (PDGF) (Li et al., 1992a; Antoine et al., 2005), epidermal growth factor (Pufe et al., 2003), fibroblast growth factor (FGF) 2 (Hatziapostolou et al., 2006), and tumour necrosis factor α (Pufe et al., 2003); signalling molecules, such as cAMP (Mourlevat et al., 2005), hydrogen peroxide (Polytarchou et al., 2005) and nitric oxide (Polytarchou et al., 2009); hormones (Tamura et al., 1995; Vacherot et al., 1995; Roger et al., 2006); transcription factors, such as AP-1 (Florin et al., 2005; Polytarchou et al., 2005; Poimenidi et al., 2009) and HOXA5 (Chen et al., 2005); tumour suppressors, such as menin (Gao et al., 2009; Feng et al., 2010) and PTEN (Li et al., 2006); miscellaneous conditions, such as hypoxia (Antoine et al., 2005), mechanical loading (Liedert et al., 2004), serum (Poimenidi et al., 2009) and X-rays (Polytarchou et al., 2004). Contradictory observations regarding retinoic acid suggest either induction of expression by retinoic acid (Kretschmer et al., 1991; Bloch et al., 1992; Azizan et al., 2000; Brunet-de Carvalho et al., 2003; Mitsiadis et al., 2008) and regulation by the retinoic acid receptors (Marzan et al., 2011), or lack of effect (Li et al., 1992a). PTN expression is also altered during disease (see "Implicated in").
PTN has been suggested to mediate FGF2 stimulatory effect on human prostate cancer cell proliferation and migration (Hatziapostolou et al., 2006), as well as hydrogen peroxide and nitric oxide stimulatory effects on human prostate cancer and endothelial cell migration (Polytarchou et al., 2005; Polytarchou et al., 2009). It is also suggested to regulate vascular endothelial growth factor (VEGF) receptor expression (Kokolakis et al., 2006) and VEGF-mediated proliferation and angiogenesis of endothelial cells (Héroult et al., 2004).

Signalling. PTN acts via various receptors (reviewed in Papadimitriou et al., 2009), like the receptor protein tyrosine phosphatase beta/zeta (RPTPβ/ζ) (Maeda et al., 1996; Meng et al., 2000; Ulbricht et al., 2003; Lu et al., 2005; Perez-Pinera et al., 2007b; Herradón and Ezquerra, 2009; Koutsioumpa et al., 2009; Mikelis et al., 2009; Polytarchou et al., 2009), ανβ3 integrin (Mikelis et al., 2009; Feng et al., 2010; Mikelis et al., 2011), nucleolin (Take et al., 1994; Said et al., 2005; Koutsioumpa et al., 2012), N-syndecan (Raulo et al., 1994; Nolo et al., 1995; Kinnunen et al., 1998a; Raulo et al., 2006; Landgraf et al., 2008) and anaplastic lymphoma kinase (ALK) (Stoica et al., 2001; Bowden et al., 2002; Powers et al., 2002; Lu et al., 2005; Perez-Pinera et al., 2007c; Yanagisawa et al., 2010).
PTN has been the first natural ligand identified for RPTPβ/ζ with high affinity (Maeda et al., 1996). It has been proposed that PTN binding to RPTPβ/ζ leads to dimerization of the receptor and loss of its tyrosine phosphatase activity. This subsequently increases the tyrosine phosphorylation of β-catenin, its dissociation from E-cadherin and its accumulation in the cytoplasm (Meng et al., 2000). Other downstream targets of the PTN/RPTPβ/ζ signalling are β-adducin (Pariser et al., 2005b) and a member of the Src family, Fyn (Pariser et al., 2005a), all affecting cytoskeletal integrity, adhesion and cell migration. On the other hand, it has been reported that PTN binding to RPTPβ/ζ leads to dephosphorylation and activation of c-Src kinase, focal adhesion kinase, phosphatidylinositol 3-kinase (PI3K), and Erk1/Erk2 (Souttou et al., 2001b; Polykratis et al., 2005; Hienola et al., 2006; Diamantopoulou et al., 2010; Himburg et al., 2010; Gao et al., 2011; Mikelis et al., 2011).
PTN has been also shown to directly bind ανβ3 integrin, which forms a functional complex with RPTPβ/ζ on the cell surface. In cell lines that express RPTPβ/ζ, the presence or absence of the ανβ3 integrin, determines whether PTN stimulates or inhibits cell migration (Mikelis et al., 2009). Through the RPTPβ/ζ/c-Src axis, PTN leads to β3 integrin Tyr773 phosphorylation, which is also required for PTN-induced cell migration (Mikelis et al., 2009).
PTN also binds nucleolin (Take et al., 1994; Said et al., 2005; Koutsioumpa et al., 2012). Through this binding it may inhibit human immunodeficiency virus type 1 (HIV-1) infection (Said et al., 2005). Moreover, cell surface nucleolin mediates PTN-induced endothelial cell migration (Koutsioumpa et al., 2012).
N-syndecan was the first PTN receptor identified (Raulo et al., 1994). PTN binds N-syndecan (Raulo et al., 1994; Kinnunen et al., 1998a; Raulo et al., 2005; Raulo et al., 2006) and induces neurite outgrowth (Raulo et al., 1994; Kinnunen et al., 1998b), other actions related to the nervous system (Nolo et al., 1995; Iseki et al., 2002; Landgraf et al., 2008) and development (Imai et al., 1998; Asahina et al., 2002; Tare et al., 2002). The PTN/N-syndecan pathway has been suggested to involve c-Src activation (Kinnunen et al., 1998b).
ALK is a transmembrane tyrosine kinase (Stoica et al., 2001) suggested to promote PTN-induced cell proliferation (Souttou et al., 2001a; Powers et al., 2002), survival (Bowden et al., 2002) and neuronal differentiation (Souttou et al., 2001a). The PTN/ALK pathway is supposed to activate the Ras-MAPK (Souttou et al., 2001a) or the PI3K-Akt (Bai et al., 2000; Slupianek et al., 2001) signalling pathways. However, there are also studies showing that PTN is not a ligand for ALK (Moog-Lutz et al., 2005; Mathivet et al., 2007). In line with this notion, it has been suggested that instead of PTN directly binding ALK, the latter is indirectly activated by PTN binding to RPTPβ/ζ (Perez-Pinera et al., 2007a).
Two forms of PTN have been suggested to differentially bind PTN receptors; PTN15 has been shown to bind ALK and promote proliferation of glioblastoma cells, whereas PTN18 has been shown to bind RPTPβ/ζ and promote haptotactic migration (Lu et al., 2005). In a later study, neither of the two PTN forms where able to activate ALK in neuroblastoma and glioblastoma cells (Mathivet et al., 2007). Moreover, based on the fact that PTN binding to ανβ integrin occurs through its C-terminal domain (Mikelis et al., 2011), PTN15 that lacks the C-terminal domain of the full length molecule (Lu et al., 2005) is not expected to bind ανβ3.
In fetal alveolar epithelial type II cells, PTN exerts its effects via cross-talk with Wnt/β-catenin signalling (Weng et al., 2009), although this has not been linked to any of the PTN receptors up to date.

Growth and maturation of brain
Neurite outgrowth promoting activity was the first to be acknowledged when PTN was first identified and is considered one of the characteristic PTN functions (Kovesdi et al., 1990; Kretschmer et al., 1991; Rauvala et al., 1994; Amet et al., 2001; Chang et al., 2004; Yanagisawa et al., 2010; Yao et al., 2011). Moreover, PTN is involved in the neurite outgrowth promoting actions of the Y-P30 polypeptide, produced by peripheral blood mononuclear cells of the maternal immune system in pregnancy, during brain development of the embryo (Landgraf et al., 2008), and mediates the neuritogenic activity of embryonic brain-derived chondroitin sulphate/dermatan sulphate hybrid chains (Bao et al., 2005). Levels of PTN expression are significantly regulated by amphetamine administration and PTN seems to have important roles in the modulation of synaptic plasticity (Le Grevès, 2005), the protection of nigrostriatal pathways against amphetamine insult (Gramage et al., 2010b), and limitation of amphetamine-induced neurotoxic and rewarding effects (Gramage et al., 2010a). In the same line, levels of the endogenous expression of PTN affect cognitive deficits and long-term alterations of hippocampal long-term potentiation after adolescent amphetamine treatment (Gramage et al., 2011).

Proliferation-mitogenesis
PTN has been suggested to promote proliferation of endothelial cells (Courty et al., 1991; Fang et al., 1992; Laaroubi et al., 1994), epithelial cells (Fang et al., 1992; Delbé et al., 1995; Souttou et al., 1997; Bernard-Pierrot et al., 2004), prostate cancer LNCaP cells (Hatziapostolou et al., 2005; Hatziapostolou et al., 2006), fibroblasts (Fang et al., 1992; Souttou et al., 1997), osteoblasts (Zhou et al., 1992; Yang et al., 2003), human peripheral blood mononuclear cells (Achour et al., 2001), fetal alveolar epithelial type II cells (Weng et al., 2009) and adult rat hepatocytes (Asahina et al., 2002). Moreover, it mediates the stimulatory effect of hydrogen peroxide (Polytarchou et al., 2005) and FGF2 (Hatziapostolou et al., 2006) on LNCaP cell proliferation. Immobilised PTN has been shown to stimulate proliferation of oligodendrocyte CG-4 and primary progenitor glial 0-2A cells (Rumsby et al., 1999), as well as human umbilical vein and bovine brain capillaries endothelial cells (Papadimitriou et al., 2000). Conversely, it has been proposed to inhibit the proliferation of C6 glioma cells (Parthymou et al., 2008) and the VEGF-induced proliferation of human umbilical vein endothelial cells (Heroult et al., 2004).

Cell migration
PTN has been shown to promote migration of endothelial (Papadimitriou et al., 2001; Souttou et al., 2001b; Ulbricht et al., 2003; Li et al., 2005b; Polykratis et al., 2005; Mikelis et al., 2009), glioblastoma (Ulbricht et al., 2003; Lu et al., 2005; Mikelis et al., 2009), human osteoprogenitor (Yang et al., 2003), human osteoblasts (Li et al., 2005b) and human prostate cancer LNCaP (Hatziapostolou et al., 2005) cells. PTN has been reported to mediate the stimulatory effect of hydrogen peroxide on human endothelial and prostate cancer LNCaP cell migration (Polytarchou et al., 2005; Polytarchou et al., 2009), of eNOS/NO on human endothelial and prostate cancer cell migration (Polytarchou et al., 2009), of aprotinin on human endothelial cell migration (Koutsioumpa et al., 2009) and of FGF2 on LNCaP cell migration (Hatziapostolou et al., 2006). Conversely, it has been suggested to inhibit C6 glioma cell migration (Parthymou et al., 2008) and the VEGF-induced migration of human umbilical vein endothelial cells (Heroult et al., 2004; Polykratis et al., 2004).

Differentiation
PTN has been reported to play a negative role during adipogenesis (Gu et al., 2007; Yi et al., 2011) and to inhibit fetal alveolar epithelial type II cell differentiation into type I cells (Weng et al., 2009).

Skeletal system
Initially, PTN mRNA was found during development in bone and cartilage progenitors and in dental pulp (Tezuka et al., 1990; Vanderwinden et al., 1992). One biological function that was early attributed to PTN is the promotion of osteoblast attachment to the extracellular bone matrix through its C-terminal domain (Gieffers et al., 1993). It was later shown that bone mass loss observed due to oestrogen deficiency is compensated in transgenic mice over-expressing PTN (Masuda et al., 1997). PTN is prominently expressed in the cell matrices that act as target substrates for bone formation and may play an important role in bone formation, probably by mediating recruitment and attachment of osteoblasts/osteoblast precursors to the appropriate substrates for deposition of new bone (Imai et al., 1998). PTN has the ability to promote adhesion, migration, expansion, and differentiation of human osteoprogenitor cells (Yang et al., 2003) and to regulate periosteal bone formation and resorption in response to four-point bending of right tibias in C57BL/6J mice (Xing et al., 2005), although it was more recently suggested by using PTN knockout mice that it is not a key upstream mediator of the anabolic effects of mechanical loading on the skeleton (Kesavan and Mohan, 2008). Interestingly, the PTN transgenic mice develop a phenotype characterized by higher bone mineral content and density (Imai et al., 1998) and increased bone growth (Tare et al., 2002); however, a more recent study with targeted PTN over-expression in mouse bone suggests that although PTN mice have advanced bone growth in length and maturation during early stages of bone development, the difference is diminished in later life and the bones become brittle (Li et al., 2005a). On the other hand, although PTN-deficient mice seem to have normal bone formation (Lehmann et al., 2004), they show growth retardation in the weight-bearing bones by two months of age and low bone formation and osteopenia, as well as resistance to immobilization-dependent bone remodelling, during adulthood (Imai et al., 2009).

Injury repair, survival and regeneration in several systems
PTN levels in human serum (Weiss et al., 2009) and rat bone (Petersen et al., 2004) are increased during fracture healing (Petersen et al., 2004; Li et al., 2005a; Weiss et al., 2009).
Treatment with PTN chimaeras after canine carotid artery balloon angioplasty injury resulted in endothelial healing (Brewster et al., 2006). Up-regulation of PTN in heart failure and cardiac ischemia may contribute to the revascularization of the injured heart (Christman et al., 2005; Li et al., 2007). PTN has been reported to facilitate wound healing of injured fetal alveolar epithelial type II cells (Weng et al., 2009).
PTN has been suggested to play a role in the survival of hematopoietic stem cells (Himburg et al., 2010), and the survival and regeneration of dopaminergic neurons (Hida et al., 2003; Jung et al., 2004; Hida et al., 2007). PTN is highly expressed within the injured nerve suggesting a role in peripheral nerve regeneration (Blondet et al., 2005; Jin et al., 2009), in macrophages, astrocytes and endothelial cells after neuronal injury (Takeda et al., 1995; Yeh et al., 1998), in neurons and glial cells after spinal cord injury in rats (Wang et al., 2004), and in denervated nerve and muscle suggestive of a role in axonal regeneration (Mi et al., 2007).

Bacterial growth
PTN has been shown to have bactericidal properties through an unknown, up to date, mechanism (Svensson et al., 2010).

Apoptosis
PTN prevents apoptosis of SW-13 epithelial cells (Bowden et al., 2002) and spermatocytes (Zhang et al., 1999a), and inhibits transforming growth factor β1-induced apoptosis in hepatoma cell lines (Park et al., 2008). Similarly, the functions of PTN/RPTPβ/ζ on human embryonic stem cells seem to depend mainly on its anti-apoptotic effect (Soh et al., 2007). Conversely, PTN potentiates cardiomyocyte apoptosis (Li et al., 2007).

Orthologs. PTN is a basic amino acid (24%) and cysteine rich protein. Its 10 cysteine residues are conserved in vertebrates and all form disulfide bonds (Hulmes et al., 1993). PTN exhibits high sequence conservation among species (reviewed in Rauvala et al., 2000 and references therein). The PTN C-terminal domain is the most conserved domain evolutionarily (Svensson et al., 2010) (dbSNP, HomoloGene).

Peptides/chimaeras and structure-function relationship
Several PTN peptides and truncated PTN constructs have been studied in an attempt to elucidate the structure/function relationship of PTN:
- The truncated construct P1-40 (dominant negative effector through its ability to form dimers with PTN) has been shown to prevent PTN-induced transformation of the mouse embryonic fibroblast NIH 3T3 cell line and the formation of tumours in the human breast cancer MDA-MB-231 cell line (Zhang et al., 1997).
- Peptides P1-21 and P121-139 have been demonstrated to stimulate endothelial tube formation, proliferation and in vivo angiogenesis (Papadimitriou et al., 2000; Papadimitriou et al., 2001).
- Peptides P9-59 and P60-110 have been reported to induce endothelial cell migration and tube formation, while P9-110 to inhibit migration. All three peptides caused an increase of endothelial adhesion (Polykratis et al., 2004). In a different study, the truncated constructs P9-59 and P60-110 (equivalents of the two out of the three fragments that are produced after MMP-2 cleavage) have been shown to promote or inhibit NIH 3T3 cell proliferation, respectively (Dean et al., 2007).
- The peptide P65-97 has been demonstrated to inhibit the mitogenic, tumourigenic and angiogenic activities of PTN (Hamma-Kourbali et al., 2008).
- The truncated construct P1-110 has been extensively studied. It has been reported to prevent proliferation and tumour formation in NIH 3T3 cells (Bernard-Pierrot et al., 2001), bind ALK, inhibit in vitro and in vivo PTN-induced angiogenesis of endothelial cells, in vitro and in vivo PTN-induced transforming activity of MDA-MB 231 cells (Bernard-Pierrot et al., 2002), to inhibit in vitro and in vivo proliferation of U87 MG glioblastoma cell line, in vivo angiogenesis, and growth and angiogenesis of U87 MG xenografts in nude mice (Dos Santos et al., 2010), and inhibit PTN-induced MDA-MB-231 breast tumour and endothelial cell proliferation and growth (Ducès et al., 2008).
- The peptide P111-136 has been shown to inhibit PTN-mediated PC-3 cell growth and angiogenesis (Hamma-Kourbali et al., 2011).
- The peptide P112-136 has been demonstrated to inhibit PTN binding to ανβ3, but not RPTPβ/ζ, inhibit in vivo angiogenesis and the PTN-induced migration and tube formation of human endothelial cells (Mikelis et al., 2011).
- The peptide P122-131 has been reported to inhibit adhesion, proliferation and migration of DU145 and LNCaP cells and in vivo angiogenesis (Diamantopoulou et al., 2010).
- Peptides generated after proteolysis with plasmin were shown to have stimulatory or inhibitory effects on endothelial migration and tube formation, and only stimulatory on cell adhesion (Polykratis et al., 2004).
- Chimaeras of PTN and FGF have been demonstrated to induce re-endothelialization after angioplasty injury (Brewster et al., 2006).

PTN knock-out and PTN-over-expressing mice
PTN-deficient mice are born without major anatomical defects and exhibit enhanced hippocampal long-term potentiation (Amet et al., 2001). However, mice deficient in both PTN and midkine have been shown to exhibit severe auditory deficit (Zou et al., 2006), have high mortality rates within a month of their birth (Muramatsu, 2010) and exhibit female infertility (Muramatsu et al., 2006).
PTN-over-expressing mice exhibit abnormalities in bone formation (Masuda et al., 1997; Hashimoto-Gotoh et al., 2004; Li et al., 2005a) and show decreased hippocampal long-term potentiation (Pavlov et al., 2002).