Lipocalin 2 (LCN2; OMIM:600181), also known as neutrophil gelatinase-associated lipocalin (NGAL), siderocalin (Scn), oncogenic lipocalin 24p3, super inducible protein 24 (SIP24), uterocalin, α₂-microglobulin-related protein, and neu-related lipocalin (NRL) is a 25-kDa protein involved with inflammatory response in multiple diseases. Historically, the first amino acid sequence of a LCN2 orthologue was deduced from a cDNA sequence of rat, which was isolated fortuitously whilst screening a library with a synthetic oligonucleotides designed to a sequence located in the ribosomal protein P1 (1). Based on its high degree of similarity with α₂ microglobulin, rat LCN2 was initially termed α₂-microglobulin-related protein (1). The identification of mouse LCN2 was reported one year later by Hraba-Renevey and colleagues (2). In their study they isolated mouse Lcn2 mRNA by differential screening as a gene that is induced by SV40 in G_o-arrested primary mouse kidney cell cultures. The name 24p3 of this gene was given in a very pragmatic way and simply referred to the plate (no. 24) and the sequential assigned number of the cDNA sequence (no. 3) isolated from respective plate (Suzanne Hraba-Renevey, personal communication).

Subsequently, human LCN2 was identified another four years later and purified to apparent homogeneity from exocytosed material of phorbol myristate acetate-stimulated neutrophils and suggested to be a new member of the lipocalin family (3). This family includes small secreted proteins that share a characteristic three-dimensional fold comprised of a single eight-stranded continuously hydrogen-bonded antiparallel β-barrel ( Figure 1 ). The calyx of this lipocalin fold is open at one end allowing uptake and binding of several small lipophilic molecules. This was the reason why LCN2 was initially thought to have transport functions. However, there is increasing evidence that many lipocalins exhibit great functional diversity and take over specialized and pleiotropic activities. This is especially documented in the numerous functions that have been reported for LCN2. Kjeldsen and coworkers already demonstrated that human LCN2 can be secreted in three different forms, namely its 25-kDa monomer, a ~46-kDa homodimer, and a ~135-kDa heterodimer composed out of LCN2 and metalloproteinase-9 (MMP-9) (3). This prompted to the assumption that LCN2 might modulate neutrophil gelatinase activity (3). The murine counterpart was identified as a superinducible protein with a molecular weight of 24-kDa (SIP24) in growth arrested, murine Balb/c 3T3 fibroblasts that become massively induced in response to mitogens, fibroblast growth factor (FGF), epidermal growth factor (EGF), and serum (5, 6). Similarly, hepatic expression of mouse LCN2/24p3 dramatically increased during turpentine-induced acute phase response and the production was significantly stimulated by tumor necrosis factor-α (TNF-α) (7). These findings suggest LCN2/SIP24/24p3 as a key systemic reactant to local or systemic inflammatory disturbances that is modulated in its expression by many factors. In line with its function as an acute-phase protein, hepatocytes were identified as a major source of LCN2 production mediating hepatoprotective effects in acute liver injury and showing positive correlation to hepatic inflammation (8, 9).

Nowadays, there is additional ample evidence that LCN2 is a key molecule that regulates iron homeostasis. Although LCN2 itself does not have an intrinsic binding ability for iron, it can bind to siderophores, which are a group of diverse, small, high-affinity iron-chelating compounds that are secreted by microorganisms in response to iron limitation in their environment to increase iron uptake (10–12). Devireddy and coworkers proposed distinct iron-dependent activities for murine LCN2. They suggested that iron-laden LCN2 (holo-24p3) binds to a cell surface receptor termed 24p3R, is internalized, and releases its bound iron, thereby increasing the intracellular iron content (13). On the contrary, iron-free LCN2 (apo-24p3) internalized by binding to its receptor can associate with an intracellular iron chelating siderophore, take over bound iron and transfer it outside the cell, thereby reducing the overall intracellular iron concentration and inducing the BCL2-interacting protein BIM/BCL2L11 expression and stimulating cell apoptosis (13).

Studies in two independent generated gene-targeted Lcn2-deficient mice confirmed the essential iron-dependent role of LCN2 in the early stages of innate immune response (14, 15). Consistent with the function of LCN2 in regulating iron homeostasis, respective mouse models exhibited an increased sensitivity to bacterial infection under iron-limiting conditions (14, 15). It was suggested that neutrophils from Lcn2 null mice were less able to inhibit bacterial growth compared with wild type neutrophils, confirming LCN2 as a bacteriostatic factor preventing bacterial siderophore-mediated iron acquisition (15). Comparative bacterial profiling of Lcn2 null mice and wild type mice further demonstrated that the lack of LCN2 protein provokes expansion of siderophore-dependent bacterial species, resulting in significant changes in the intestinal microbiome composition and increased inflammatory activity in the gastrointestinal tract (16). Importantly, mice lacking Lcn2 displayed a persistent colonization with segmented filamentous bacteria in the ileum that were associated with signs of autolysis in apical villi and detachment of epithelial cells from the basal membrane. Moreover, we have previously speculated that the expression of LCN2 is an intrinsic “help-me” sensor that, upon injury, develops a biological activity promoting production of neutrophil-attracting chemokines (17). Since neutrophils are regarded as the first line of defense in the innate arm of the immune system, loss of LCN2 would consequently prevent the effective clearance of respective microorganisms.

Interestingly, LCN2 also plays a physiological role in reproductive biology in both sexes. Originally termed as uterocalin, LCN2 is involved in the tissue remodelling processes in the uterus during pregnancy (18). Additionally, the expression of Lcn2 varies according to hormonal fluctuations in the different estrous stages of mice (19, 20). Berger and coworkers reported that Lcn2 null females have a significantly lower pregnancy rate compared with wild type animals (15). Investigating the underlying cause, the authors found that LCN2 is an essential component in modulating the membrane properties of the sperm in the fertilization process in male mice (21). Furthermore, in mouse models, infertility of both male and female animals was shown to be associated with increased expression of LCN2 in testis and ovaries (20, 22). However, to date, the molecular pathways of LCN2 and a possible interaction with its receptor in the reproductive tract are poorly understood.

LCN2 also has critical functions in the regulation of various aspects of energy metabolism and expenditure (23). LCN2 modulates expression of genes involved in β-oxidation in adipocytes, promotes β-cell function, and counteracts obesity-induced glucose intolerance suggesting LCN2 as an endogenous compensatory signal to counteract metabolic dysregulation in obesity through its anorexigenic activity (23, 24). In the central nervous system, LCN2 is a pronounced direct appetite suppressor and satiety signal in humans and mice modulating caloric intake as well as fat and lean mass content (25, 26). In mice, LCN2 impacts mitochondrial and peroxisomal function and integrity that directly impacts hepatic triglyceride balance, oxidative stress, and apoptosis (27). Finally, LCN2 was shown to be a key modulator of hepatic lipid homeostasis that in vitro and in vivo controls the formation of intracellular lipid droplets by regulating expression of Perilipin 5 that maintains the balance between lipogenesis and lipolysis (28). All these findings suggest that LCN2, besides its immunomodulatory functions, has strong impact on many aspects in shaping the energy metabolism.

Unfortunately, all these insights into LCN2 function in innate immunity, iron homeostasis, and energy metabolism and expenditure are still descriptive. Precise mechanisms how LCN2 signals are transferred from outside into the cells are widely unknown and, although some biochemical activities and affinities were nicely demonstrated, there a numerous questions unresolved yet. In particular, the precise receptor activities and responsible pathways relevant in transmitting LCN2 inputs to intercellular targets are unknown. Actually, there are six putative LCN2 receptors, namely neutrophil gelatinase-associated lipocalin receptor (NGALR), low density lipoprotein-related protein 2 (LRP2), LRP6, melanocortin 4 receptor (MC4R), MC1R, and MC3R. The size and extent of their external ligand-binding domain, hydrophobic membrane-spanning regions, and their intracellular domains inside the cell are markedly different ( Figure 2 ). Moreover, they are assigned to different chromosomes, reveal a marked different gene structure, and encode for mRNAs and proteins different in size ( Table 1 ).

NGALR (OMIM: 611461) is a gene that received a number of different names historically deduced from their proposed function or from its structure. The names NGALR and LCN2R were derived from the receptor’s affinity for LCN2. NGALR2 (or NgalR-2) and NGALR3 (or NgalR-3) refer to splice variants of NGALR that share overlapping biochemical characteristics (50). The name solute carrier family 22 member 17 (SLC22A17) indicates that the receptor belongs to the solute carrier (SLC) group of membrane transport proteins that commonly contain a number of hydrophobic transmembrane α-helixes. This family currently includes about 458 members that are classified into 65 families by similarities in their characteristics, homology, functions, and structures (51, 52). The SLC22 family contains about 30 distinct multi-membrane spanning proteins, 13 of which have been localized to plasma membrane (52, 53). Beside its affinity for LCN2, SLC22A17 binds to and mediates the endocytosis of filtered protein in the kidney and assists in iron uptake but does not drive the transport of other usual substrates of the SLC22 family (52). The names BOCT, BOCT1, and BOIT are abbreviations for brain-type organic ion transporter. The BOCT sequence (acc. no. NM_021551) popped up in the initial study reporting the expression cloning of murine NGALR (i.e., 24p3R) after searching the GenBank database with two identified overlapping NGALR sequences as queries (13). Finally, the introduced acronym 24p3R indicates that this receptor has affinity for murine LCN2 termed 24p3 (13, 41). 24p3R, which is now better referenced to as NGALR, was first identified as a LCN2 receptor after three rounds of expression cloning in COS-7 cells using a cDNA library prepared from murine FL5.12 cells, which are highly sensitive to LCN2/24p3-medicated apoptosis (13).

Homozygote Ngalr mutant mice are embryonic lethal (Yukio Nakamura, personal communication), suggesting that this receptor has already important activities in the developing embryo (54, 55). It is a transmembrane signaling receptor involved in insulin receptor recycling and transport of inorganic cations/anions, amino acids, and oligopeptides. Although NGALR has been shown to be involved in endocytosis, SLC22A17 is still considered as an orphan transporter for which no specific substrates or transport mechanisms have been identified (53). Nevertheless, potential functions of NGALR have been identified in various malignancies, including cancerogenesis, kidney disease, obesity, and many other diseases (53). In particular, it has been recently shown that the LCN2/24p3R axis is highly relevant in promoting nonalcoholic steatohepatitis in high-fat diet-fed Ob/Ob mice by activating hepatic stellate cells, representing the key pro-fibrogenic cell type within the liver (56).

Binding of mouse LCN2 to murine NGALR/24p3R overexpressed in human HeLa cells was first demonstrated by ligand-cell binding experiments, in which ³²P-labeled 24p3 was bound to FL5.12 cells. Based on titration curves and Scatchard plot analysis, the dissociation constant of 24p3 to NGALR was determined to ~ 92 pM (41). Direct binding of apo-LCN2 to the N-terminal part containing the first 105 residues of human NGALR was further demonstrated by solution-state biomolecular nuclear magnetic resonance (NMR) in conjunction with other biophysical methods (30). However, in this study, the binding constant of both molecules as determined by isothermal titration calorimetry and microscale thermophoresis was determined to be in the low micromolar range (~7-10 µM), while the affinity of LCN2 complexed with ferric-enterobactin was even about threefold lower (30). This suggests that the N-terminus of NGALR exhibits some specificity toward apo-LCN2, while NGALR does not or only poorly bind to holo-LCN2 (30). Finally, it should be noted that the affinity determined by NMR-based techniques is about six orders of magnitude lower than that observed in the ligand-cell binding experiments. In addition, contradictory findings showed that exogenous LCN2 had no effect on iron release or uptake in NGALR-expressing HeLa cells and relevant NGALR subdomains failed to bind LCN2, questioning the overall assumption that NGALR and LCN2 form effective receptor-ligand complexes (29). This contributes to the general discussion about problems and current standards with reproducibility and rigor in conducting measuring and evaluating binding affinities with common high-throughput technologies (57). Additionally, the interaction of a ligand with its receptor in cell-ligand binding experiments is strongly dependent on the cell system used (58), suggesting that binding characteristics of ligands and potential receptors might differ between cells analyzed. Finally, biomolecule associations and affinities are further influenced by equilibration time, temperature, half-life of complex, salt concentrations, pH, amount of active (not total) protein, and many other factors (57).

LRP2 (OMIM: 600073) is a large multi-ligand, type-1 transmembrane receptor that was first identified in rats by screening a rat kidney λgt 11 cDNA expression library with antiserum raised against large, negatively charged antigens isolated from rat glomeruli (35). The initially isolated cDNA clone (C1B) was then used as a probe to isolated longer λgt 11 cDNA clones. Subsequently, a 15.4 kb cDNA was assembled by several rounds of screening of a random-primed rat kidney λZAP II cDNA library that contained an uninterrupted open reading frame encoding 4660 amino acids (35). LRP2 is an endocytic receptor sharing structural similarity with the LDL receptor, which in vertebrates is the main receptor for lipoprotein uptake. Ligands for LRP2 are plasma carriers transporting vitamins and steroid hormones, which upon binding, become internalized from the apical cell surface and sorted into different cellular pathways (31). LRP2 is structurally composed of a large extracellular domain consisting of (i) complement-type repeats that have affinity for different ligands, (ii) eight β-propeller folds built by six sheets composing four antiparallel β-strands and containing up to six Tyr-Trp-Thr-Asp (YWTD) or YWTD-like repeats, (iii) flanking epidermal growth factor (EGF)-like domains, (iv) a short single membrane-spanning region, followed by (v) a small C-terminus enriched in Asn-Pro-X-Tyr (NPXY) and NPXY-like motifs required for efficient ligand-mediated receptor internalization and biological signaling (59, 60).

As its alternative name megalin indicates, full length LRP2 with its 4665-4660 amino acids in mice and humans is a giant protein with a molecular weight of ~ 600 kDa. LRP2 can physically interact with the cellular communication network factor 2 formerly known as connective tissue growth factor (CCN2/CTGF), bone morphogenetic factor-4 (BMP-4), and sonic hedgehog signaling (SHH) molecule (60). Affinity of human iron-free LCN2 (apo-LCN2) and siderophore-bound LCN2 to human LRP2 was first demonstrated by surface plasmon resonance (SPR) analysis with matrix bound LRP2. In the respective experimental setup, the dissociation constant of human LCN2 to human LRP2 was determined to ~60 nM (42), which is quite similar to the K _d determined for the interaction of murine LCN2 and MC4R which was determined to 51.39 ± 4.78 nM (43).

It was supposed that megalin functionally blocks the passage of elevated plasma LCN2 occurring in sepsis, ischemia or nephrotoxic injury into the urine (61). In line, with this suggestion the deletion of megalin in proximal tubules of murine kidney resulted in proteinurea containing large quantities of LCN2 (61). Mutated forms, in which positively charged amino acids were mutated on the surface of LCN2 without disrupting neither the typical lipocalin fold nor the ligand-binding site of LCN2, lost their affinity for LRP2, showing that the ligand-receptor binding is majorly formed by electrostatic forces and the net positive charge displayed by LCN2 (61). LRP2 has also affinity for the retinol-binding protein (RBP), α₁-microglublin (α₁M), odorant-binding protein IA (OBPIA), major urinary protein 6 (MUP6), and apolipoprotein M (APOM) that all belong to the lipocalin family (62, 63). Based on these findings, it might be possible that individual domains within LRP2 act as unspecific scavengers that recognize a broad range of ligands by their three-dimensional fold such as the symmetrical β-barrel fold that is a unifying characteristic of lipocalins. Nevertheless, the binding of other lipocalins with K _d values in the µM range such as RBP (K _d = 1.8 µM) and α₁M (K _d = 0.42 µM) to megalin have significantly lower affinities than that of LCN2 (K _d = 60 nM) (42, 63, 64). Moreover, the affinity of purified human or rat LRP2 to other proteins such as the lipoprotein lipase is 10 to 30 fold lower (K _d = 6.1 and 2.7 nM, respectively) than that of LCN2 (65). Nevertheless, since LRP2 is involved in the endocytosis of a wide range of molecules (LCN2, siderophore-bound iron, diverse lipoproteins, hemoglobin, albumin, hormones, toxins, drugs, vitamin-binding proteins, stress-related proteins and many others), LCN2 when overexpressed during malignancies might change the affinity for other LRP2 binding partners by blocking respective binding sides (66).

Recently, mouse LRP6 composed of the same basic structural motifs as LRP2 acting as a co-receptor for Wnt was shown to specifically bind to mouse LCN2 (49). In the respective study, the interaction of LCN2 and LRP6 was shown in co-immunoprecipitation assay and it was suggested that binding of LCN2 to LRP6 effectively inhibits Wnt/β-catenin signaling. Interestingly, LRP5 which is about 70% homologous to LRP6 and also active as a co-receptor in Wnt signaling failed to bind LCN2 (49, 67). However, information about the strength of the binding of LCN2 and LRP6 and independent confirmation of observed interaction in other species than mouse is still pending.

MC4R (OMIM: 155541) belongs to a family of G protein-coupled 7-transmembrane receptors that contains five family members (termed MC1R to MC5R). The individual members of that family share a sequence homology of 38 to 60% (68). Each receptor has different specificity for melanocortins that consists of adrenocorticotropin hormone (ACTH), α-melanocyte-stimulating hormone (α-MSH), β-MSH, and γ-MSH (68). Receptor activation results in formation of cyclic AMP (cAMP) and activation of protein kinase C (PKC) that in turn provokes influx of extracellular calcium resulting in inositol triphosphate (IP3) production and downstream activation of MAPK and JAK-STAT pathways (68).

In regard to LCN2, it was further demonstrated that osteoblast-derived LCN2 can cross the blood-brain-barrier and bind to MCR4 in paraventricular nucleus neurons of the hypothalamus as assessed by application of biotinylated Lcn2 in Lcn2 null mice (43). In addition, the overexpression of MC4R in HEK293T cells dose-dependently stimulated cAMP activity with an EC₅₀ of 1.41 ± 0.25 nM, while silencing or pharmacological inhibition of MC4R with the MC3R/MC4R antagonist SHU9119 abrogated Lcn2-activated cAMP production and signaling in the mouse hypothalamic, gonadotrophin-releasing hormone neuronal cell line GT1-7. This inhibition was independent from the two other putative Lcn2 receptors (i.e. MC1R and MC3R) because LRP2 is not expressed in GT1-7 cells and silencing of NGALR did not affected LCN2 signaling (43). Saturation and competition binding experiments in MCR4 transfected HEK293T showed that LCN2 binds to MC4R with a dissociation constant K _d of 51.39 ± 4.78 nM. This affinity is about 4.5-fold lower than that of the classical MCR ligand α-MSH.

In the same study, the authors showed that LCN2 also has affinity for MC1R (K _d = 86.96 ± 9.72 nM) and MCR3 (K _d = 82.13 ± 12.14 nM) and can stimulate cAMP production through these receptors. However, the authors suggested that MC4R is the most relevant MCR for LCN2 signaling in paraventricular and ventromedial neurons because MC1R and MC3R are not able to regulate appetite (43). Similar to MC4R, MC3R is a neural MCR that is expressed primarily in the central nervous system having important functions in regulating feed efficiencies by binding to identical ligands with similar binding affinities (69). The third putative MC receptor for LCN2, MC1R, is expressed in neurons, astrocytes, and microglia and also sensitive to the endogenous nonselective agonist α-MSH (70). The activation of MC1R can prevent neutrophil infiltration, thereby reducing the overall LCN2 content that is available within the brain (71). Furthermore, because LCN2 participates in the pathogenesis of blood-brain barrier disruption, MC1R with its lower affinity might be a kind of LCN2 sensor capable to prevent overshooting LCN2 concentrations within the brain by blocking neutrophil infiltration (72).

In addition to its activity in control of appetite, MC4R can directly impact glucose and lipid homeostasis, pain perception and even sexual behaviors by modulating of luteinizing hormone and prolactin surges in female and erectile activity in male rodents (73, 74). However, if these MC4R-driven activities can be modulated by LCN2 is not known yet.

As discussed, there are six putative receptors supposed to have (more or less affinity) for LCN2. The receptors either belong to the group of transmembrane transporter/signaling receptors (NGALR), multi-ligand receptors (LRP2, LRP6), or G-protein coupled seven transmembrane receptors (MC4R, MC1R, MC3R). If these putative receptors are indeed bona fide LCN2 receptors, LCN2 would have many possibilities to modulate cellular signaling via binding to one of these receptors or by binding combinations of these receptors ( Figure 3 ).

NGALR belongs to the SLC superfamily of proteins that can transport a wide array of molecules, including sugars, amino acids, oligopeptides, vitamins, nucleotides, metals, inorganic ions, organic anions, and drugs (52). Within the SLC group, NGALR belongs to the major facilitator superfamily (MFS) group that in particular transport organic ions including urea, amines, bile acids, and α-ketoglutarate (52). Importantly, NGALR assists in the uptake of iron and shows a more restricted substrate profile than other SLC members, suggesting that NGALR has more specialized physiological functions (75). It was suggested that NGALR contributes to adaptive osmotolerance by controlling the uptake of amino acids as osmolytes and that LCN2 downregulation counteracts increased proliferation and permanent damage of osmotically stressed cells (76). Similarly, it was proposed that murine LCN2 alters cellular iron trafficking through iron efflux via murine NGALR, thereby driving apoptosis (41). In line with this assumption, exogenous LCN2 sensitized primary cortical neuronal cells and neuroblastoma cell lines to cell death (77). Likewise, the addition of LCN2 increased intracellular reactive oxygen species formation, DNA damage, cell cycle arrest and cell death in mouse fibroblast cell line L929 (78). Additionally, in both acute and chronic white matter diseases in the brain, it could be shown that LCN2 has an inhibitory effect on the oligodendrocyte differentiation necessary for remyelination, which was mediated via SCL22A17/EGR1 signaling (79).

We showed that the lack of LCN2 is associated with endoplasmic stress, unfolded protein response, and modulation of the cellular labile chelatable iron pool (27, 80). However, there are also findings showing that LCN2 has protective and survival activity (9, 81), while other studies have shown that exogenously added LCN2 in HeLa cells that overexpress NGALR did not affect iron efflux or uptake (29). Similarly, there are numerous other studies that show conflicting data in the understanding of LCN2’ function in controlling apoptosis.

LRP2 is a multi-ligand binding receptor that has important roles during embryogenesis by supplying the fetus with nutrients and controlling sonic hedgehog signaling (31). It further serves as a receptor for thyroglobulin (82) and Vitamin D (83), has affinity for BMP-4 (83), forms specific complexes with cubulin and Dab2 that mediate the uptake of filtered proteins (e.g., vitamin-binding proteins, retinol-binding proteins, albumin, insulin, α₁-microglobulin, β₂-microglobulin, EGF, and lysozyme) and escape the glomerular filtration barrier and driving the endocytic uptake of clathrin-coated vesicles (84). It further has affinity for several ligands involved in cholesterol metabolism (e.g., APOA1, APOE, APOB100, APOJ/clusterin, and APOM) and is important for the endocytosis of high density lipoprotein particles (83). So there is large variety of activities that can be modulated by interacting of LCN2 with specific modules within LRP2.

As discussed, the MCRs (i.e., MC4R, MC1R, and MC3R) supposed to have affinity for LCN2 are belonging to the group of 7-transmembrane G-protein linked receptors. From these receptors MC4R has the highest affinity for LCN2. Its activation leads to increased intracellular cAMP (36, 39). Moreover, more recent studies have shown that MC4R is able to activate the extracellular regulated kinase (ERK) and modulate the activity of other G proteins receptors, most likely by sharing affinity for the same G-proteins that mediate signaling via these receptors (85, 86). This is in line with previous findings showing that the G protein G_sα is required for controlling MC4R-regulated energy expenditure and glucose metabolisms, while effects on food intake, linear growth, cholesterol metabolism, and expression of specific transcription factors in the paraventricular nucleus of the hypothalamus (i.e., Sim1, Crh) are primarily mediated by the G_q/11α subunit (87, 88). In addition to G_sα and G_q/11α, MC4R can also couple to other G proteins such as Gαi, which inhibits the adenylate cyclase activity, thereby decreasing intracellular cAMP levels, and Gαq, which activates phospholipase C (PLC) and downstream PKC (89). Presently, it is unknown if LCN2 can modulate the affinity of MC4R for G proteins. Nevertheless, LCN2 was shown to increase the formation of cAMP as efficiently as 0.5 nM α-MSH in mouse GT1-7 hypothalamic cells, but failed to activate phosphorylation of AMPK, ERK1, ERK2, or tyrosine kinase in these cells, further suggesting that cAMP formation and kinase phosphorylation might not be necessarily linked to each other (43).

So in sum, LCN2 seems to have many possibilities to interfere with cellular signaling through its putative receptors. This notion is underpinned by the finding that the overexpression of LCN2 in the esophageal squamous cell carcinoma cell line EC109 resulted in the upregulation of 167 genes and downregulation of 96 genes using a 2-fold threshold encoding genes that have further capacity to interact with thousands of other proteins (90). This study that was based on protein-protein interaction network showed that LCN2 is a highly pleiotropic protein having affinity for many other proteins, thereby causing a wide range of expression alterations in immunity-related terms, pathway-related terms, cellular response to molecules of bacterial origin, extracellular matrix organization, and cell cycle-related terms. However, there is an urgent need to clarify if these effects of LCN2 are mediated by acting as a signal-inducing ligand at a specific receptor, masking binding epitopes for other proteins on proposed LCN2 receptors, binding to other biomolecules (e.g., proteins, hormones, vitamins etc.) preventing their interaction with other proteins/receptors, or alternatively by directly or indirectly changing general cellular features (osmolarity, iron content, mitochondrial activity). It should be noted that the outcome of the induced biological effect is the sum of the different interaction levels with all receptors expressed at a specific time in a specific environment and the accessibility of possible target structures at the time of LCN2’ presence. This should results in an extreme high combinatorial diversity of LCN2 activities.

LCN2 is a pleiotropic molecule involved in many biological processes including control of iron homeostasis, cell differentiation, energy expenditure, bacterial defense, cell death, chemotaxis, cell migration, and many others. Most of the biological functions of LCN2 were established in work conducted in mice. Nowadays, six potential LCN2 receptors (NGALR, LRP2, LRP6, MCR4, MCR1, and MCR3) were identified that markedly differ in regard to structure and affinity. In humans, concentrations of LCN2 in urine and serum have become increasingly relevant as a prognostic and predictive biomarker for many diseases (150, 151). Nevertheless, changes in LCN2 or LCN2 receptor expression in the course of a specific disease was mostly only reported in association or correlation studies. Actually, there is a significant gap in the understanding of the underlying mechanisms by which LCN2 and its putative receptors contribute to the initiation or progression of a disease. It could be that some findings found in experimental mouse models cannot be translated 1:1 to the human situation. Murine and human LCN2 only share a sequence identity of 62% and there are significant biochemical and functional differences between both orthologs. This might result in different affinities for putative receptors and other biomolecules supposed to interact with LCN2, resulting in modified biological effects in both species. Contradictory findings with regard to biological effects of LCN2 show that the effects of LCN2 are complex and influenced by many factors, including species, disease, sex, cell type, posttranslational modifications, receptors, and binding partners. Possibly, some of the discrepancies and inconsistencies are raised by the fact that many findings on LCN2 and its receptors were established in in vitro models and have no relevance for the in vitro situation. Finally, the usage of antibodies with unclear sensitivity and specificity may lead to data that cannot be interpreted. It cannot be excluded that incorrect expression patterns of the LCN2 receptors can also be found in the literature because the antibodies used have not been sufficiently validated. We hope that our article will help to sensitize scientists for this critical issue that may provoke the production of misleading “scientific” data.

RW conceived and planned the drafting of this review. All other authors have written part of this review and helped in design and drawing of images. All authors contributed to the article and approved the submitted version.