Deficiency of Lipin2 Results in Enhanced NF-κB Signaling and Osteoclast Formation in RAW-D Murine Macrophages

Lipin2 is a phosphatidate phosphatase that plays critical roles in fat homeostasis. Alterations in Lpin2, which encodes lipin2, cause the autoinflammatory bone disorder Majeed syndrome. Lipin2 limits lipopolysaccharide (LPS)-induced inflammatory responses in macrophages. However, little is known about the precise molecular mechanisms underlying its anti-inflammatory function. In this study, we attempted to elucidate the molecular link between the loss of lipin2 function and autoinflammatory bone disorder. Using a Lpin2 knockout murine macrophage cell line, we showed that lipin2 deficiency enhances innate immune responses to LPS stimulation through excessive activation of the NF-κB signaling pathway, partly because of TAK1 signaling upregulation. Lipin2 depletion also enhanced RANKL-mediated osteoclastogenesis and osteoclastic resorption activity accompanied by NFATc1 dephosphorylation and increased nuclear accumulation. These results suggest that lipin2 suppresses the development of autoinflammatory bone disorder by fine-tuning proinflammatory responses and osteoclastogenesis in macrophages. Therefore, this study provides insights into the molecular pathogenesis of monogenic autoinflammatory bone disorders and presents a potential therapeutic intervention.


Introduction
Alterations in human Lpin2 cause Majeed syndrome, a rare inherited autosomal recessive autoinflammatory bone disorder that is characterized by early onset chronic multifocal osteomyelitis, neutrophilic skin inflammation, and dyserythropoietic anemia [1][2][3][4][5]. Patients with a monogenic form of this disorder present with poor prognosis; however, IL-1β blockade has showed promising clinical outcomes [6]. Lpin2 encodes lipin2, a member of the lipin1-3 protein family. Lipins are phosphatidic acid phosphatases (PAPs) that are critical for fine-tuning cellular lipid metabolism [7,8]. The mechanisms by which loss-offunction Lpin2 mutations contribute to autoinflammatory bone disorder development are largely unknown.
Toll-like receptors (TLRs) are a family of pattern recognition receptors (PRRs) that are critical for proper innate immune responses. Among the 10 TLRs in humans, TLR2/4 are activated by lipopolysaccharides (LPS) to transmit downstream signaling, such as the NF-κB and MAP kinase pathways, to elicit proinflammatory transcription [9,10]. Notably, recent findings link lipin2 function to innate immunity and inflammation and are consistent with the pathophysiological conditions in Majeed syndrome [8,11]. Lpin2 deficiency in murine and human macrophages augmented the proinflammatory cytokines IL-1β and TNFα [12,13]. Lipin2-deficient macrophages displayed enhanced MAP kinase and P2X7 signaling pathway activation and NLRP3 inflammasome formation in response to LPS and ATP stimulation [12]. Lpin2 knockout mice exhibited elevated serum IL-1β and TNFα levels and upregulated hepatic and splenic proinflammatory transcripts in response to a high intraperitoneal LPS dose [12]. These data suggest that lipin2 constrains proinflammatory responses in vitro and in vivo. However, the molecular mechanisms by which lipin2 modulates cellular inflammatory signaling have not been clarified.
Monogenic Lpin2 mutation causes familial chronic multifocal osteomyelitis and osteolytic foci that are characteristic of Majeed syndrome [14,15]. Notably, a very recent study has reported enhanced osteoclastogenesis in monocyte-derived M2-like macrophages from a patient with Majeed syndrome [16]. In this study, we aimed to determine the molecular link between loss of lipin2 function and autoinflammatory bone disorder pathophysiology. Our findings may advance the understanding of the molecular pathogenesis and therapeutic intervention for Lpin2-mutation and other monogenic autoinflammatory bone disorders.

Lipin2 Deficiency Enhances Proinflammatory Responses in Macrophages by Modulating Inflammatory Transcription
To determine the physiological role of lipin2 in regulating macrophage proinflammatory and osteoclastic signaling, we knocked out Lpin2 in murine macrophage RAW-D cells, a RAW264.7 subclone with a higher potential to differentiate into osteoclasts [17,18]. We treated the control and the Lpin2 knockout cells with LPS and compared their responsiveness to proinflammatory stimulation. As shown in a previous study [12], marked IL-1β protein induction was observed in the Lpin2 knockout cells following LPS stimulation ( Figure 1A). We then investigated the role of lipin2 in LPS-elicited proinflammatory transcription. We conducted a microarray expression analysis using LPS-treated or untreated Lpin2 knockout and control RAW-D cells. Scatterplots and heatmaps showed upregulated proinflammatory transcripts in response to LPS in Lpin2 knockout cells ( Figure 1B,C). Qiagen ingenuity pathway analysis (IPA) of differentially expressed genes (DEGs) (fold change > 2) in Lpin2 knockout cells identified PRRs, TLRs, NF-κB, and p38 MAP kinase signaling ( Figure 1D). These data indicate that lipin2 deficiency promotes macrophage proinflammatory responses.

Lpin2 Knockout Enhances the NF-κB Signaling Pathway in RAW-D Cells
To confirm whether lipin2 modulates inflammatory responses via TLR signaling, we assessed the activation status of the TLR4 downstream inflammatory signaling. We observed that Lpin2 knockout cells showed IκB phosphorylation and downregulation, and accelerated p50 processing, which are NF-κB activation read-outs, while no impact on Akt pathway was observed (Figure 2A). A luciferase reporter assay validated the LPS-mediated NF-κB overactivation in Lpin2 knockout cells ( Figure 2B). We investigated whether the NF-κB signaling pathway was altered in response to accumulation of phosphatidic acid (PA), a lipin2 PAP substrate functioning as a signaling molecule [19]. However, the gross PA level did not markedly differ between Lpin2 knockout and control cells ( Figure 2C).

Elevated IRAK Signaling Induces Downstream Pathways in Lpin2 Knockout Macrophages
Our IPA and western blotting results demonstrated that lipin2 is vital to the TLR4 downstream signaling pathways (Figure 1 andFigure 2). To explore the underlying mechanism of lipin2 deficiency-induced proinflammatory signaling overactivation, we evaluated TLRproximal adaptor protein and kinase, MyD88, and IRAK1, which comprise the Myddosome, a signaling complex formed in response to TLR activation [20,21] (Figure 3A). IRAK1 was remarkably upregulated, whereas the levels of TLR4 and MyD88 were unchanged in Lpin2 knockout cells stimulated by LPS within a short period of 60 min ( Figure 3B). IRAK1 accumulation might be regulated by pre-and post-transcriptional control because the induction of the protein level in Lpin2 knockout cells is greater when compared to two-fold induction of transcription level ( Figure 3C). IRAK1 transmits TLR4 signaling downstream by complexing with IRAK4 [22,23]. To test whether the IRAK complex activity results in the overactivation of downstream NF-κB signaling in Lpin2 knockout cells, we treated Lpin2 knockout cells with the IRAK inhibitor, IRAK1/4. The inhibitor treatment suppressed the activation of the downstream NF-κB signaling pathway as well as JNK signaling, the activity of which is reported to increase in lipin2-deficient macrophages [12,13] (Figure 3D).

Elevated TAK1 Activity Induces Downstream Signaling in Lpin2 Knockout Macrophages
It was reported that the founding family member lipin1 serves as a scaffold for several cytoplasmic and nuclear signaling molecules to regulate lipid and energy metabolism [24][25][26][27][28][29].
We hypothesized that lipin2 also has a similar ability for proinflammatory signaling control in macrophages. We screened lipin2 interacting proteins by Flag-lipin2 co-immunoprecipitation (IP)/mass spectrometry and identified TAK1 as a potential lipin2 interacting protein (data not shown). The TAK1-TAB complex transduces proinflammatory signaling by activating NF-κB and MAP kinase signaling via IKKβ and MAPKK phosphorylation [30,31] ( Figure 3A). We speculated that this signaling complex might be a potential lipin2 target. To confirm the interaction between lipin2 and TAK1, we ectopically co-expressed different HA-tagged protein kinases and Flag-lipin2 in 293T cells and performed a co-IP assay. Lipin2 only interacted with TAK1 ( Figure 4A). We also demonstrated endogenous interaction between TAK1 and lipin2 ( Figure 4B). TAK1 activity is positively controlled by phosphorylation and K63-linked polyubiquitination [32,33]. Thus, we examined whether lipin2 interaction affects TAK1 polyubiquitination. Lipin2 co-expression markedly impaired TAK1 polyubiquitination ( Figure 4C). We observed the phosphorylated active form of TAK1 in LPS-treated Lpin2 knockout cells ( Figure 4D). TAK1 inhibitor treatment suppressed NF-κB signaling as well as JNK activation ( Figure 4D).

Lipin2 Deficiency Promotes RANKL-Dependent Osteoclastogenesis and Osteoclastic Resorption Activity in RAW-D Macrophages
Lpin2 alterations cause multifocal osteomyelitis, and there might be a major link between lipin2 function and the intrinsic osteolytic pathophysiology. Thus, we explored the possibility that lipin2 deficiency promotes osteoclastogenesis in macrophages. Lpin2 knockout cells displayed higher potential than control cells for RANKL-mediated osteoclast-like multinuclear cell (MNC) formation ( Figure 5A,B). Lpin2 knockout cells showed increased osteoclastic resorption activity ( Figure 5C). The transcription levels of the osteoclast markers, Ctsk and Acp5, were markedly increased in Lpin2 knockout cells ( Figure 5D). These data demonstrate that lipin2 deficiency increases RANKL-dependent formation of osteoclastlike MNCs. Data are means ± SD (n = 3). *** P < 0.001, Student's t-test. (D) RT-PCR was performed to determine relative mRNA expression levels of the osteoclastic marker genes Ctsk and Acp5 in control and Lpin2 KO RAW-D cells treated with 50 ng/mL RANKL. The cells were harvested for RNA extraction at the indicated time points. Gapdh was used for normalization. Data are means ± SEM (n = 3). *** P < 0.001. One-way ANOVA followed by the multiple comparisons test.

Lipin2 Negatively Regulates NFATc1 Activity
Since MAP kinase and NF-κB signaling play critical roles in osteoclast differentiation [34], we conducted western blot analysis to evaluate their activation status. RANKL stimulation unexpectedly did not induce excessive MAP kinase and NF-κB pathway activation in Lpin2 knockout cells ( Figure 6A). However, we observed impaired NFATc1 phosphorylation in Lpin2 knockout cells ( Figure 6B). NFATc1 is a master transcription regulator of osteoclastogenesis. RANKL stimulation is known to promote calcineurin-mediated NFATc1 dephosphorylation, inducing its nuclear translocation and transactivation activity [35][36][37]. We observed an enhanced RANKL-induced NFATc1 accumulation in the nucleus of Lpin2 knockout cells ( Figure 6C). A previous study demonstrated the physical interaction between lipin1 and NFATc4, which resulted in the suppression of NFATc4 transcriptional activity and downstream proinflammatory cytokine expression [28]. For this reason, we tested whether lipin2 interacts with NFATc1. A co-IP assay with transfected 293T cells demonstrated that lipin2 binds to NFATc1 ( Figure 6D).

Discussion
Lpin2 alterations cause Majeed syndrome, which is characterized by chronic autoinflammatory multifocal osteomyelitis [3][4][5][6]. Therefore, lipin2 could prevent excessive proinflammatory signal activation. A previous study demonstrated that lipin2 plays a crucial role in limiting p38, ERK, and JNK activities in inflammatory macrophages. Overactivation of these kinases by lipin2 deficiency facilitates excessive Il1b mRNA expression and mature IL-1β overproduction [12]. In the present study, we showed that Lpin2 depletion in a murine macrophage cell line resulted in an elevated NF-κB signaling pathway. Therefore, lipin2 may modulate the TLR4 downstream inflammatory signaling axes, MAP kinase, and NF-κB pathways.
Besides PAP function, lipin1 acts as a scaffold for various signaling molecules, such as PPARα, NFATc4, SREBP1, and ERK1/2 [24,25,27,28]. Lipin2, like lipin1, interacts with PPARα and modifies its transcriptional activity [38]. These foregoing studies suggest that lipin family members are multifunctional in nature. In this study, we demonstrated that lipin2 negatively modulates TAK1 activity. We also found that IRAK1 is upregulated in Lpin2 knockout cells. Since IRAK1 is an unstable protein subjected to proteasomal degradation [22,39], lipin2 may possibly contribute to this process. However, further investigations are required to elucidate the precise molecular mechanisms through which lipin2 controls the related signaling pathways. In summary, our findings suggest that lipin2 may play a critical role in suppressing NF-κB signaling in part through the TAK1 pathway. Targeting the overactivated signaling might be an efficient therapeutic strategy, and the use of specific inhibitors should be explored for their effectiveness in the treatment of lipin2-deficient autoinflammatory disorder.
A recent study has identified that the disease-associated Lpin2 mutation is linked to elevated osteoclast formation. M2-like macrophages derived from the patient with Majeed syndrome display enhanced osteoclastogenesis. The macrophages show a proinflammatory phenotype with elevated NFATc1 and phosphorylated JNK levels [16]. Here, we revealed that Lpin2 knockout in RAW-D macrophages promotes osteoclast-like MNC formation by activating the NFATc1 pathway. These findings may provide evidence of the involvement of lipin2 in bone metabolism. However, one limitation is that Lpin2 knockout mice lack phenotypes resembling osteomyelitis. Hence, in-depth studies may be necessary to bridge the gap between human pathophysiology and the absence of osteomyelitis phenotype in mice [8]. In this work, we presented a model demonstrating that lipin2 regulates proinflammatory and osteoclastic signaling via upstream modulation of NF-κB and NFATc1 signaling. This research may lay the foundation for elucidating the pathogenesis of monogenic autoinflammatory bone disorders, which may help in the development of novel therapeutic strategies.

Immunoblots and Immunoprecipitation
The cells were lysed in an NP-40 cell lysis buffer (50 mM Tris-Cl [pH 7.5], 120 mM NaCl, and 0.5% (v/v) NP-40) supplemented with a cOmplete protease inhibitor cocktail (Sigma-Aldrich Corp., St. Louis, MO, USA) and phosphatase inhibitors (PhosSTOP; Roche Diagnostics, Mannheim, Germany). The lysate protein concentrations were measured with Bio-Rad protein assay dye (Bio-Rad Laboratories, Hercules, CA, USA). Forty micrograms of whole-cell lysate was dissolved in an SDS sample buffer and resolved by SDS-PAGE. For Phos-tag SDS-PAGE, each sample was resolved in an SDS-PAGE gel containing 50 µM Phos-tag (Fujifilm Wako) according to the manufacturer's instructions. The gels were transferred to polyvinylidene difluoride (PVDF) membranes (Bio-Rad Laboratories), which were then blocked with 5% (w/v) nonfat dry milk in Tris-buffered saline with 0.05% (v/v) Tween 20 (TBST; pH 8.0) and probed with primary antibodies, as indicated in the figures (1:1000-1:4000 with 5% (w/v) nonfat dry milk in TBST). For immunoprecipitation, the cells were harvested in an NP-40 cell lysis buffer containing protease and phosphatase inhibitors.

Osteoclast Formation and Osteoclastic Resorption Activity Assays
Murine macrophage RAW-D cells were cultured with 50 ng/mL RANKL for 3 d to induce osteoclastogenesis. The cells were fixed with 3.7% (v/v) formaldehyde, and osteoclast formation was detected by tartrate-resistant acid phosphatase (TRAP) staining. Images were acquired using Vert A1 microscopy and ZEN software (ZEISS, Jena, Germany). TRAPpositive MNCs containing ≥3 nuclei were scored as differentiated osteoclast-like cells. Osteoclastic resorption activity measurement was conducted with a bone resorption assay kit (PG Research, Kodaira, Japan) according to the manufacturer's instructions. Briefly, RAW-D cells were seeded at 1 × 10 4 /well on a 48-well plate coated with fluoresceinated calcium phosphate and cultured for 3 d in phenol red-free medium containing 100 ng/mL RANKL. Osteoclastic resorption activity was evaluated by measuring the fluorescence intensity of 100 µL of conditioned medium at excitation/emission (Ex/Em) wavelengths of 485/535 nm in a SpectraMax M2e microplate reader (Molecular Devices LLC, San Jose, CA, USA).

Cellular Fractionation
Cells were fractioned according to a previously described protocol [43]. Briefly, 10 7 RAW-D cells were collected in 1 mL of ice-cold phosphate-buffered saline (PBS), centrifuged at 1000× g for 3 min at 4 • C, and the cell pellets were resuspended in 900 µL of ice-cold PBS containing 0.1% (v/v) NP-40 plus protease inhibitor. Afterwards, 300 µL of whole-cell lysate fraction was withdrawn, and the remainder was centrifuged at 10,000× g for 10 s at 4 • C. After that, 400 µL of supernatant was collected and centrifuged at 10,000× g for 10 s at 4 • C, and 200 µL of supernatant (cytosolic fraction) was collected. The pellets (nuclear fraction) obtained after the first spin were washed twice with 1 mL of ice-cold 0.1% (v/v) NP40/PBS and dissolved in a 1×SDS sample buffer. The nuclear and whole-cell lysate fractions dissolved in the SDS sample buffer were sonicated with a VP-050 microprobe (TAITEC, Koshigaya, Japan).

Phosphatidic Acid (PA) Measurement
The PA content was measured with a PicoProbe phosphatidic acid assay kit (BioVision, Milpitas, CA, USA) according to the manufacturer's instructions. Briefly, 10 6 WT and 10 6 Lpin2 knockout cells were harvested and lysed with 1 mL of a PA assay buffer, and their protein concentrations were determined with Bio-Rad protein assay dye. The lipids were extracted by adding 3.75 mL of chloroform/methanol/12 N HCl (2:4:0.1), 1.25 mL of chloroform, and 1.25 mL of 1 M NaCl, vortexing, centrifuging at 3000× g for 10 min at RT, and collecting the organic layer. The chloroform was evaporated in the hood for overnight, and the PA was dissolved in 5% (w/v) Triton X-100, converted to a fluoresceinated PA intermediate, and quantified by measuring the latter at Ex/Em wavelengths of 535 nm/587 nm in a SpectraMax M2e microplate reader (Molecular Devices LLC).

Luciferase Reporter Assay
Luciferase assay was carried out using the Dual-Glo Luciferase Assay System (Promega, Madison, WI, USA) according to the manufacturer's instructions. Briefly, control or Lpin2 knockout RAW-D cells were transfected with pGL4.32[Luc2P/NF-kB-RE] firefly luciferase reporter and pRL-TK renilla luciferase expression plasmids (Promega) at a ratio of 50:1. At 24 h after transfection, cells were stimulated with 200 ng/mL LPS. After 4 h, cells were harvested for evaluating reporter activities. The firefly and renilla luciferase activities were measured in a multimode reader, TriStar2 LB942 (Berthold Technologies, Bad Wildbad, Germany). The values of firefly luciferase activities were normalized with that of renilla luciferase.

Mass Spectrometry
HeLa cells were transfected with pcDNA3-Flag-lipin2. After 36 h, the cells were treated with 10 µM MG132 for 12 h and harvested in an NP-40 cell lysis buffer containing protease and phosphatase inhibitors. Lysate from empty vector-transfected cells was prepared and used as the control IP sample. Five hundred microliters corresponding to 0.5 mg of cell lysate was incubated with 10 µL of anti-Flag antibody-conjugated bead slurry (Sigma-Aldrich Corp) at 4 • C, with gentle rocking for 2 h. The beads were washed twice with 1 mL of an NP-40 washing buffer and twice with TBST. Samples were purified by liquid chromatography/tandem mass spectrometry (LC-MS/MS) as previously described [44] at Kazusa Genome Technologies in the Kazusa DNA Research Institute (Kisarazu, Japan).

Quantification and Statistical Analysis
Data are means ± SD or SEM for ≥3 independent experiments or biological replicates. Between-group differences were analyzed by one-way ANOVA with multiple comparison tests or Student's t-test. Statistical analyses were performed in GraphPad Prism9 (GraphPad Software, San Diego, CA, USA). P < 0.05 was considered a statistically significant difference.