Lipocalin-2: a novel link between the injured kidney and the bone

Abstract

Purpose of the review

Fibroblast growth factor 23 (FGF23) excess is associated with left ventricular hypertrophy (LVH) and early mortality in patients with chronic kidney disease (CKD) and in animal models. Elevated Lipocalin-2 (LCN2), produced by the injured kidneys, contributes to CKD progression and might aggravate cardiovascular outcomes. The current review aims to highlight the role of LCN2 in CKD, particularly its interactions with FGF23.

Recent findings

Inflammation, disordered iron homeostasis and altered metabolic activity are common complications of CKD, and are associated with elevated levels of kidney produced LCN2 and bone secreted FGF23. A recent study shows that elevated LCN2 increases FGF23 production, and contributes to cardiac injury in patients and animals with CKD, whereas LCN2 reduction in mice with CKD reduces FGF23, improves cardiovascular outcomes and prolongs lifespan.

Summary

In this manuscript, we discuss the potential pathophysiological functions of LCN2 as a major kidney-bone crosstalk molecule, linking the progressive decline in kidney function to excessive bone FGF23 production. We also review associations of LCN2 with kidney, cardiovascular and bone and mineral alterations. We conclude that the presented data support the design of novel therapeutic approaches to improve outcomes in CKD.

INTRODUCTION

Chronic kidney disease (CKD) is a public health epidemic affecting ~15% of Americans and millions of people worldwide (1). Kidney disease treatment costs are the largest yearly expenditure by Medicare for a single medical issue, totaling more than $35.9 billion annually and despite new discoveries, dialysis or transplantation are the only long-term treatments for kidney failure (2, 3). Progressive CKD is strongly associated with bone and mineral abnormalities and cardiovascular disease (CVD), which contributes to premature death in patients with kidney disease (4–6).

Disordered bone and mineral metabolism is a common complication of CKD that begins early and worsens progressively as kidney function declines (7–9). Skeletal abnormalities (referred to as renal osteodystrophy), hypovitaminosis D, the ensuing hyperparathyroidism and hyperphosphatemia are common consequences of CKD (8, 10, 11), and regrouped under the umbrella term “CKD-Mineral and Bone Disorder (MBD)”. Abnormal mineral metabolism is a risk factor for arterial and cardiac calcifications (12), and kidney damage (13) and faster progression to end-stage renal disease (14, 15).

Excess fibroblast growth factor 23 (FGF23) is among the earliest signs of disordered mineral metabolism in CKD (16–22). CKD is the most common cause of elevated FGF23 and the clinical setting with the highest circulating levels. FGF23 primarily targets the kidney to inhibit phosphate (Pi) reabsorption and suppress Vitamin D production resulting in increased urinary Pi output and reduced intestinal Pi absorption. While increased FGF23 levels help maintaining circulating Pi within normal levels in the early stages of the disease, FGF23 excess may contribute to kidney disease progression (23, 24), and CVD in patients with CKD (25–41). In parallel to its impact on mineral metabolism, FGF23 has emerged as a non-conventional risk factor for development of cardiovascular disease and overall mortality in CKD (27, 42, 43). Elevated FGF23 levels are powerfully associated with increased risk of mortality across all stages of CKD, in various CVD cohorts, and in the general population. Higher FGF23 is independently associated with increased risk of CVD events (27, 42–45). Among specific CVD event types, elevated FGF23 is most strongly and consistently associated with increased risk of heart failure events and prevalence of left ventricular hypertrophy (LVH). In experimental animal models, increased FGF23 and FGF23 signaling lead to LVH and premature death, while we and others have shown that reduction of FGF23 prevents LVH and prolongs survival in mice with CKD (46, 47).

Multiple circulating and skeletal factors may ultimately contribute to FGF23 excess in the later stages of CKD (8, 48). However, given that FGF23 rapidly rises following kidney injury, FGF23 may respond to molecules secreted by the injured kidney. Yet, this link is missing in our understanding of CKD pathophysiology. In addition, CVD remains the leading cause of death in patients with CKD, and the increased risk of CVD is only partially explained by the presence of traditional risk factors, such as hypertension and diabetes mellitus. In addition, patients with CKD and CVD have a higher mortality rate compared with patients with CVD and normal renal function. This underscores the need to expand our understanding of the pathophysiology of non-traditional CVD risk factors in patients with CKD to develop appropriate prevention and novel therapeutic approaches.

Recently, we have found that increased circulating levels of Lipocalin 2 (LCN2), in patients and mice with CKD, correlated with excess FGF23 (49). We further found that bone is a target of kidney-secreted LCN2 and that increased LCN2 stimulates bone production of FGF23 in mice with CKD. Deletion of LCN2 in mice prevented increases in bone and circulating levels of FGF23 and development of LVH, without altering the course of CKD progression (49). Here, we will discuss this novel regulatory pathway of FGF23 by LCN2, as well as the crucial role of LCN2 in the pathophysiological context of CKD.

Bone and kidney crosstalk: FGF23 regulation

Several hypotheses have been proposed to explain FGF23 elevation in CKD. The general idea is the existence of a mechanism acting like a “mineralostat” leading to increased FGF23 in response to the severity of kidney insult (50, 51). As FGF23 increases in response to high phosphate diets and states of impaired phosphate excretion, to maintain normal serum phosphate by suppressing phosphate reabsorption in the kidneys, it was suggested that FGF23 responds to elevated concentrations of phosphate in the blood (52). Although a dedicated mechanism sensing free phosphate remains elusive, it was proposed that FGF receptor 1 directly senses free phosphate and modulates FGF23 accordingly (53). It remains unclear whether this leads to increased FGF23 production or stabilization. Given that bone is the Pi reservoir and that variations in local Pi follow bone turnover, the sensing of extracellular phosphate by the bone would have to be minimalist, so as not to trigger an inappropriate increase in FGF23. Finally, in CKD, increase in FGF23 antedates phosphate (Pi) increase, which does not align with prior elevations in circulating Pi, further suggesting that the putative bone Pi sensing mechanism does not play a role in early CKD.

In the search for systemic regulators of FGF23 in CKD, it was hypothesized that the injured kidney could produce molecules which could regulate FGF23 bone production. Indeed, circulating and bone FGF23 levels are elevated in patients (44, 51, 54, 55) and experimental models of acute kidney injury (AKI) (51, 56, 57). Recent advances in the field showed that FGF23 increases in response to inflammatory signaling, altered iron homeostasis and perturbed metabolic changes (57–62). This further suggests that kidney-bone mediators at the crossroads of inflammation, iron, and metabolic activity might explain the early elevations in FGF23 in CKD.

LCN2 at the crossroads of inflammation, iron homeostasis and metabolic activity

Lipocalin-2 (LCN2), also named neutrophil gelatinase-associated lipocalin (NGAL), or oncogene 24p3 is a 21-kDa secreted glycoprotein, member of the lipocalin family, transporters of small hydrophobic/lipophilic molecules in mammalian organisms (63). Unlike other lipocalins, LCN2 has the unique ability to also bind hydrophilic molecules or macromolecules (64, 65). Circulating as a monomer, a homodimer or bound to MMP9, LCN2 is involved in multiple pathophysiological processes, and its function appears to be dependent on the tissue and cells secreting it as well as LCN2 monomer/dimer structure (65). LCN2 was identified over 20 years ago as a top responder to LPS in murine macrophages (66), and further isolated from neutrophils, which are one of the major sources of LCN2 (67). LCN2 is ubiquitous, and is expressed in low levels by multiple cells/organs. Lcn2 expression increases in response to infection and/or inflammation in a variety of tissues and organs including adipose tissue (68), bone (69), liver (70), lungs (71), brain (72–74), heart (75), eyes (76), uterine tract (77, 78), and kidneys (49, 79, 80).

The multiple and sometime divergent LCN2 functions can be explained by the relative abundance of its different receptors. A ubiquitous solute carrier type (SLC22A17) with wide tissue distribution, and the more restricted low density lipoprotein-related protein 2 (LRP2) or megalin appear to mediate endocytosis and exocytosis of iron-bound and iron-free LCN2 (81–83). More recently, G protein-coupled receptor melanocortin 4 receptor (MC4R), and to lesser extent MC1R and MC3R were also shown to bind LCN2 in a tissue specific manner (69, 84). Thus, tissue distribution of these receptors might explain the different outcomes associated with LCN2 excess or deficiency.

In inflammation, LCN2 serves primarily as a chemoattractant for neutrophils, leading to increased leucocyte infiltration and activation (85). LCN2 plays a major role in the expansion, activation or deactivation of macrophages and T-cells, and contributes to the fate of the inflammatory response, in particular TLR signaling (86–89). These contradictory findings highlight the pleiotropic roles of LCN2, which can act as a protective molecule, involved in the resolution of inflammation, or as a pathological signal, further aggravating its consequences.

Overall, secretion of LCN2 during inflammation and/or infection plays a major antibacterial role through iron-scavenging processes. LCN2 binds bacterial siderophores and sequesters iron, thus limiting iron availability necessary for bacterial growth (90). Subsequent studies have shown the existence of mammalian siderophores that LCN2 also binds, suggesting that LCN2 might be involved in the regulation of systemic and local iron metabolism. Consequently, several studies have shown strong association between iron metabolism, hematological indices and LCN2 concentration (91–93). LCN2 appears to regulate tissue specific iron balance (94–96), and might play a pivotal role in the protection of tissue against high free iron levels. Finally, given the relationship between inflammation and LCN2, elevated serum LCN2 appear to strongly correlate with prevalent anemia in patients with chronic inflammatory conditions, such as CKD (93).

Local and systemic iron imbalance dramatically affects downstream metabolic functions, and LCN2 has been shown to play a pivotal role in a wide range of metabolic processes. LCN2 regulates retinoic acid transport in adipocytes, and osteoblast-produced LCN2 crosses the blood-brain barrier and binds to MC4R in the paraventricular ventromedial neurons in the hypothalamus, which results in suppression of appetite (69, 97). Excess LCN2 is also associated with insulin resistance, making adipose-derived LCN2 a potent regulator of glucose and insulin metabolism (68, 69, 98, 99). Thus, LCN2 stands at the crossroads of inflammation, iron homeostasis and metabolism and has a central role in the regulation of physiological and pathological mechanisms in multiple organs and tissues.

The role of LCN2 in the kidney

Given LCN2 propensity to be rapidly induced in response to inflammation, and its crucial role in regulating metabolism, LCN2 has been proposed to be a reliable biomarker in the diagnosis of acute and chronic inflammatory diseases associated with metabolic complications, such as AKI and CKD. In the kidney, constitutive LCN2 expression is associated with the maturation of renal epithelia (100, 101). Following a kidney insult, kidneys become the main source of LCN2 (79, 80). Contrary to other tissues producing LCN2, a consequent fraction of the kidney-derived LCN2 is directly excreted in the urine, making LCN2 not only a fair biomarker for AKI but also a predictor of clinical outcomes (99, 102, 103). Several studies suggested that LCN2 might be more than a biomarker of kidney injury and could be actively involved in disease progression and associated outcomes in both patients and mice. Although LCN2 appears to have a beneficial impact on acute kidney injury, in CKD it may participate in mechanisms of its progression.

In AKI, kidney-produced LCN2 protects against kidney inflammation (104), and promotes tubular cell survival (105), but the effects of Lcn2 deletion appear controversial: it may either alleviate or worsen fibrosis and cell death following tissue injury (104, 106–108). A new study shows that renal macrophages are actively producing LCN2 in AKI. Lcn2 deletion in macrophages decreased CCL5 production and partly blocked the CCL5/IL4 axis, leading to improved renal fibrosis, and low T cells recruitment (109). Renal macrophages produced LCN2 can also help manage iron trafficking in an optimal gradient, which contributes to tissue repair and recovery from AKI (108).

In contrast to AKI, elevated serum and urine levels of LCN2 in CKD appear to be indicative of disease progression (110), but the role of LCN in CKD is poorly understood. Urinary LCN2 correlates with low glomerular filtration rate and tissue damage in kidney biopsies in CKD patients (111). In the model of 5/6 nephrectomy, Lcn2 deletion (112) and blockade of LCN2 (113) partially reduced tubuloinsterstitial fibrosis, suggesting that LCN2 excess in CKD is maladaptive and actively contributes to disease progression.

LCN2 regulates FGF23 production in inflammation and CKD

In patients with CKD, LCN2 is a fair indicator not only of disease progression (114), but also of nutrition, anemia and inflammation (115). We have recently found that LCN2 excess in patients with CKD positively correlates with FGF23 excess. Paralleling FGF23, kidney expression and circulating levels of LCN2 increase as kidney function declines in CKD. We further found that LCN2 administration to healthy mice, led to increase in bone Fgf23 mRNA and high circulating levels of FGF23, suggesting that LCN2 might mediate kidney-to-bone signal leading to FGF23 excess in CKD.

Multiple studies have shown that binding of LCN2 to its receptors, and in particular MC4R, activates the canonical Gα-mediated cyclic adenosine-monophosphate/protein kinase A (cAMP/PKA) signaling pathway (69, 84). In osteoblasts and osteocytes, activation of cAMP/PKA signaling by PTH is a known stimulator of FGF23 production (116, 117). Similar to PTH, activation of cAMP/PKA by LCN2 increases Fgf23 transcription in bone cells. Deletion of Lcn2 in mice with CKD leads to a dramatic reduction of FGF23 levels, without altering the course of CKD progression (49). Suppression of LCN2 also has a marked effect on FGF23 stimulation in acute inflammation, in response to administration of interleukin-1 beta (IL-1β) (49). In contrast, deletion of Lcn2 has no impact on FGF23 stimulation in diet-induced iron deficiency or hyperphosphatemia (49). This suggests that LCN2 contributes to FGF23 excess in CKD as a part of an inflammatory complex rather than in addition to reduced iron bioavailability or phosphate loading.

LCN2 deficient CKD mice, showed delayed development of LVH and consequently extended lifespan, whereas administration of LCN2 to the same mice suppressed these benefits (49). Whether these effects are solely mediated by FGF23 reduction in LCN2 deficient mice remains an open question (Figure 1). Indeed, plasma and serum LCN2 are good predictors of cardiometabolic risk in patients with CKD, whereas urine LCN2 is associated with renal dysfunction (118). LCN2 is also associated with LVH and diastolic dysfunction in CKD (119), and suppression of LCN2 reduces CVD in experimental models of cardiomyopathy (120). Finally, LCN2 also directly aggravates cardiomyocyte injury (121, 122). In aggregate, these results strongly suggest that excess LCN2 in CKD is not only a determinant of disease progression, but also is directly, or indirectly by regulating FGF23, a potent mediator of cardiac injury in CKD (Figure 2).

Figure 1: LCN2 excess and cardiac toxicity in CKD.

Progressive alterations in kidney morphology and function induce inflammation-dependent lipocalin secretion leading to increased circulating and urinary LCN2. Circulating LCN2 increases kidney and systemic inflammation, reduces circulating iron and might induce LVH. LCN2 also increases bone FGF23 production, which contributes to excess FGF23 in CKD. Excess FGF23 induces phosphaturia and leads to development of cardiac disease and early mortality.

Figure 2: LCN2 and FGF23 cross-talk.

LCN2 levels are rapidly elevated following kidney insult. LCN2 recruits circulating leucocytes, contribute to kidney ferroptosis and fibrosis, and excess FGF23. LCN2 also increases bone production of FGF23 and serum FGF23 levels. In turn, FGF23 reduces phosphate and vitamin D levels and contributes to cardiovascular disease. Concomitant reductions in LCN2 and FGF23 might slow CKD progression and alleviate cardiac outcomes.

LCN2: a new marker of disordered bone and mineral metabolism in CKD?

The recent causal association of circulating LCN2 regulating FGF23 in the bones of animals and patients with CKD adds a new string to the bow of controlling FGF23 in CKD. This also suggests that LCN2 is an important mediator of mineral metabolism abnormalities in CKD. LCN2 origins in the largest quantities by the injured kidney in CKD, and we found minimal alterations in bone Lcn2 expression (49). LCN2 is also produced by osteoblasts and osteocytes in bone at homeostasis, and LCN2, like FGF23 has the potential to act either as a paracrine factor or hormonal regulator (69). However, the role of circulating or bone produced LCN2 in the regulation of osseous activities appears controversial and remains largely unknown in CKD.

Excess LCN2 in mice reduces osteogenesis and increases osteoclastogenesis (123), whereas absence of Lcn2 leads to mild osteopenia, due to reduced bone formation (124). In contrast, deletion of Lcn2 in experimental models of disuse induced bone loss, such as Duchenne Muscular Dystrophy, leads to reductions in bone resorption and increases bone mass (125). In a randomized control trial, circulating LCN2 levels predicted future risk of osteoporotic fractures requiring hospitalization (126). Taken together these results suggest that LCN2 in CKD, in addition to its pro-inflammatory, iron regulatory and metabolic potential, could represent a novel bone and mineral metabolism factor, and that its role deserves further studies.

Conclusion

Several studies in experimental cancer models have shown that LCN2 might be a valid target and anti-LCN2 therapy leads to tumor reduction, with minimal off-target complications (127, 128). Different strategies, such as monoclonal anti-LCN2 antibodies (129) and small interfering RNA against Lcn2 (130), led to reduction in angiogenesis in breast tumors. Anti-LCN2 antibodies also reduced tissue damage and neutrophil infiltration in experimental aortic aneurysm (131). Further studies are needed to understand the multiple roles of LCN2 in the pathophysiology of progressive CKD.

Key points:

• In CKD, LCN2 is produced by the kidney and increases in the circulation in response to kidney injury.

• Deletion of Lcn2 in mice with CKD reduces Fgf23 bone mRNA expression and circulating FGF23 levels.

• Deletion of Lcn2 in mice with CKD results in marked improvement in cardiac function.