miR-135b-5p inhibits LPS-induced TNFα production via silencing AMPK phosphatase Ppm1e

AMPK activation in monocytes could suppress lipopolysaccharide (LPS)-induced tissue-damaging TNFa production. We are set to provoke AMPK activation via microRNA (“miRNA”) downregulating its phosphatase Ppm1e. In human U937 and THP-1 monocytes, forced expression of microRNA-135b-5p (“miR-135b-5p”) downregulated Ppm1e and activated AMPK signaling. Further, LPS-induced TNFα production in above cells was dramatically attenuated. Ppm1e shRNA knockdown in U937 cells also activated AMPK and inhibited TNFα production by LPS. AMPK activation is required for miR-135b-induced actions in monocytes, AMPKα shRNA knockdown or T172A dominant negative mutation almost abolished miR-135b-5p's suppression on LPS-induced TNFα production. Significantly, miR-135b-5p inhibited LPS-induced reactive oxygen species (ROS) production, NFκB activation and TNFα mRNA expression in human macrophages. AMPKα knockdown or mutation again abolished above actions by miR-135b-5p. We conclude that miR-135b-5p expression downregulates Ppm1e to activate AMPK signaling, which inhibits LPS-induced TNFα production via suppressing ROS production and NFκB activation.

AMP-activate protein kinase (AMPK) plays a pivotal role in maintaining cellular energy balance [10].

AMPK activation is required for miR-135b-5p's inhibition on LPS-induced TNFα production
If AMPK activation is the primary reason of miR-135b-5p-induced action against TNFα production by www.impactjournals.com/oncotarget . Human U937 or THP-1 macrophages were transfected with miR-135b-5p construct or non-sense control microRNA ("miR-C"), and stable cells were established via neomycin selection. Expression of miR-135b-5p B and D. and Ppm1e mRNA (C and E., left panels) were tested by quantitative real-time PCR ("RT-qPCR") assay; Ppm1e protein expression was examined by Western blot assay (C, right panels). Experiments in this figure were repeated three times, and similar results were obtained. Ppm1e protein expression (vs. β-actin) was quantified (C). "Ctrl" stands for non-transfected control cells (B-E). *p<0.05 vs. "miR-C" group (B-E).

DISCUSSION
LPS is sensed by CD14 and LPS-binding protein (LBP), and binds to its receptor Toll-like receptor 4 (TLR-4) on monocytes [31,32]. This will lead to the recruitment of several key adaptor proteins (MyD88, TRAF6 and others) to activate downstream NFκB signaling cascade [31,32]. ROS production also plays a pivotal role in the process. Sanlioglu et al., showed Rac1-dependent ROS production induced LPS-induced NFκB activation and TNFα production [27]. On the other hand, ROS scavengers could attenuate LPS-induced inflammatory response [27]. For instance, Shen et al., demonstrated that cordycepin inhibited LPS-induced ROS production and subsequent TNFα production [17]. In the present study, we showed that miR-135b-5p activated AMPK signaling to inhibit LPS-induced ROS production and subsequent NFκB activation. This could be one key reason of TNFα inhibition by miR-135b-5p.
Existing evidences have implied AMPK as an antioxidative signaling under a number of stress conditions [17,29,30]. AMPK activation by energy depletion could attenuate oxidative stress via increasing NADPH content [30]. In this regard, AMPK-ACC signaling activation inhibits ROS accumulation via increasing NADPH production [30]. A recent study by She et al., demonstrated that AMPK activation could decrease H 2 O 2 -induced oxidative damages [29]. Recently, Zhang's group showed that cordycepin suppressed LPS-induced ROS production and NFκB activation through activating AMPK-NADPH signaling [17]. Our recent unpublished work showed that GSK621, a novel AMPK activator [16], attenuated LPSinduced ROS production, NFκB activation and subsequent TNFα expression (Wu et al., unpublished studies). In line with these findings, we show that AMPK activation by miR-135b-5p decreased LPS-induced ROS production and NFκB activation. Such effects by miR-135b-5p were almost abolished with AMPK inhibition. Thus, we propose that miR-135b-5p activates AMPK to attenuate LPS-induced ROS production, and subsequent NFκB activation, which then inhibit TNFα mRNA expression and production. Figure 5: miR-135b-5p inhibits LPS-induced ROS production, NFκB activation and TNFα mRNA expression. miR-135b-5p expressing U937 cells were constructed with scramble control shRNA ("sh-C"), AMPKα shRNA ("shAMPKα", No.1) or dominant negative AMPKα (T172A, "dnAMPKα"), these cells or the control U937 cells were treated with LPS (100 ng/mL) or medium control ("C") for applied time, relative ROS intensity A., NFκB activation B and C. and TNFα mRNA expression D. were tested by listed assays. Experiments in this figure were repeated for three times, and similar results were obtained. IKKα/β phosphorylation was quantified (B). *p < 0.05 vs. "C" group (A, C and D). # p < 0.05 vs. LPS only group (A, C and D). ** p < 0.05 (A, C and D). www.impactjournals.com/oncotarget Although many AMPK activators have been developed thus far [33], there are few of them are being tested in clinical stages for various disease. The results of this study showing AMPK activation by miR-135b-5p via downregulating Ppm1e provide a new strategy to activate AMPK and to inhibit LPS inflammatory responses.

Chemicals and antibodies
LPS, puromycin and neomycin were purchased from Sigma Chemicals (Shanghai, China). All the antibodies utilized in this study were purchased from Cell Signaling Technology (Danvers, MA). The cell culture reagents were purchased from Hyclone of Thermo Fisher Scientific (Shanghai, China).

Cell culture
The human monocyte cell lines, U937 and THP-1, were purchased from the Cell Bank of Fudan University (Shanghai, China). Cells were cultured in RPMI 1640 supplemented with 10% FBS and 1% L-glutamine at 37 °C.

TNFα enzyme-linked immunosorbent assay (ELISA) assay
Following treatment of cells, TNFα content in the conditional medium was evaluated via the TNFα ELISA kit (R&D Systems, Abingdon, UK) as described [9].

Western blots
As described [9], the protein lysates (20 μg per sample) were separated by SDS-PAGE gel, and were transferred onto PVDF membranes, which were then probed with primary and secondary antibodies. Enhanced chemiluminescence (ECL, Amersham, Shanghai, China) regents were utilized to detect targeted bands. The total gray of each protein band was quantified by Bio-Rad Quantity One software, and was normalized to corresponding loadings [9].

AMPKα dominant negative mutation
The pSuper-puro construct with dominant negative (T172A) AMPKα and the empty vector were provided by Dr. Lu's group at Nanjing Medical University [37,40]. Lipofectamine 2000 was applied to transfect mutant AMPKα or the vector to miR-135b-expressing U937 cells. Stable cells were again selected by puromycin (1.0 μg/ mL). AMPKα mutation was verified by Western blot assay.

Reactive Oxygen Species (ROS) assay
ROS production was measured by dichlorofluorescin (DCF) oxidation assay as described [17]. Briefly, after applied treatment, cells were incubated with 10 μM of DCFH-DA (Invitrogen) for 30 min. Cells were then washed, trypsinized and resuspended in PBS. DCF fluorescence intensity was then tested using a FACS BD machine. The fluorescent intensity value of treatment group was expressed as fold changes of the control group. www.impactjournals.com/oncotarget

Measuring NFκB (p65) DNA-binding activity
The detailed protocol of this assay was described in our previous study [9]. Briefly, after treatment of cells, NFκB (p65) DNA-binding activity, analyzing from 1.0 μg of cell nuclear extracts, was examined using the TransAM™ ELISA kit (Active Motif, Carlsbad, CA) with the manufacturer's protocol. The OD value of treatment group was always normalized to that of control group.

Statistics analysis
The statistical analyses were performed via the SPSS software (18.0), with p < 0.05 taken as significant. Data were expressed as mean ± standard deviation (SD). For comparisons among multiple groups, two-way ANOVA with the Bonferroni post hoc testing was performed.

ACKNOWLEDGMENTS
This work is partly supported by the NSFC.

CONFLICTS OF INTEREST
The authors declare no conflicts of interest.

Author contributions
Ping Li and Jian-bo Fan contributed equally to this study. All authors carried out the experiments, participated in the design of the study and performed the statistical analysis, conceived of the study, and participated in its design and coordination and helped to draft the manuscript. All authors read and approved the final manuscript.