The proteasome is an enzymatic complex found in the nucleus and cytoplasm of eukaryotic cells, archaea and certain bacteria. This complex is responsible for the degradation of intracellular proteins that are damaged or misfolded. It works in collaboration with the ubiquitin system, which tags proteins for proteasome processing. The proteasome plays an important role in the regulation of many cellular processes, such as the cell cycle, the defense against oxidative stress and inflammatory responses. The proteasome is composed of two types of domains: a core particle and one or two regulatory domains. The core particle is formed by four stacking rings, each of them consisting of seven α or β subunits. Central rings have three catalytic subunits, namely β1, β2 and β5, which have caspase-like, trypsin-like and chymotrypsin-like (CTL) activity, respectively. An alternative form of the proteasome, called immunoproteasome, is present in most animal cells but it is abundantly expressed in immune cells, where its primary role is to process proteins for antigen presentation by major histocompatibility complex (MHC) class I molecules [1
]. Expression of the immunoproteasome is induced by interferon-γ (IFN-γ), tumor necrosis factor (TNF) and bacterial lipopolysaccharide (LPS) under inflammatory conditions, such as infections or autoimmune diseases [3
]. In the presence of such stimuli, catalytic subunits of the constitutive form are respectively substituted by inducible subunits β1i (LPM2), β2i (MECL-1) and β5i (LMP7) to form the immunoproteasome. Unlike its constitutive counterparts, which have caspase-like activity, the β1i subunit also has CTL activity [6
The proteasome has been implicated as a modulator of inflammatory responses by participating in the activation of nuclear factor κB (NFκB), a transcription factor that regulates the expression of many genes involved in inflammation [8
]. Five NFκB family members have been described, namely RelA (p65), RelB, cRel, p50 and p52, respectively encoded by genes rela, relb, crel, nfkb1
. After the stimulus, NFκB proteins form dimers, which bind to κB sites on target genes either as homodimers or heterodimers. In resting cells, NFκB is sequestered in the cytoplasm by inhibitor κB (IκB) proteins. Activation of NFκB is triggered by phosphorylation of IκB, followed by its ubiquitination and proteasomal degradation, thus releasing NFκB and promoting its translocation into the nucleus [9
]. It has been demonstrated that the immunoproteasome subunit β1i is involved in the proteolytic processing of NFκB precursor proteins (p100/p105), as well as in the degradation of inhibitor κB α (IκBα) [10
]. Later, it was observed that β1i-deficient retinal pigment epithelial cells exhibited diminished activation of NFκB in response to TNF [13
]. However, the role of the immunoproteasome in NFκB activation and in the degradation of IκB proteins is still under debate [10
]. Other studies have demonstrated that immunoproteasome subunits are not essential in the activation of NFκB either in cancer cell lines or in peritoneal macrophages stimulated with TNF [17
Due to the role of the proteasome in many physiological processes, it has become a major target for the design of new drugs as a therapeutic for several diseases. Many proteasome inhibitors have been identified from natural and synthetic sources. Two of them, bortezomib and carfilzomib, are currently approved for the treatment of multiple myeloma. Although a number of second-generation proteasome inhibitors are in clinical trials [19
], undesirable side effects have been associated to these molecules. The immunoproteasome has emerged as a therapeutic target and as a strategy to reduce the toxicity associated with the inhibition of the constitutive proteasome in cells [20
]. These molecules are not only valuable as potential therapeutics but would also allow a better understanding of the physiological roles attributed to the immunoproteasome. Several highly selective immunoproteasome inhibitors have been recently described, including both peptidic [22
] and nonpeptidic inhibitors [23
In previous studies, we have shown a marked anti-inflammatory activity for a pseudopterane diterpene (compound 1) isolated from the octocoral Pseudopterogorgia acerosa
]. Compound 1 inhibited the production and expression of proinflammatory mediators in macrophages stimulated with LPS, TNF and other toll-like receptor ligands. Our results showed that this anti-inflammatory effect is due to the inhibition of IκBα degradation and the subsequent activation of NFκB. We then analyzed if the effect of compound 1 might be influenced by a modulation of the ubiquitin-proteasome system, affecting the proteasomal degradation of phosphorylated IκBα. We show herein that compound 1 inhibits the CTL activity of the proteasome induced by LPS in vitro and reduces the expression of MHC class I in macrophages. This inhibitory effect might occur by a mechanism that involved the modulation of immunoproteasome activity, since a reduction in the CTL activity of the purified immunoproteasome was observed. In vivo, compound 1 reduces the production of proinflammatory mediators in the lung of animals treated by intranasal inoculation of LPS. Molecular docking simulations predicted that compound 1 preferentially interacts with the catalytic site of subunits β1i and β5i, suggesting that the effect of this compound might be dependent on immunoproteasome activity.
2. Materials and Methods
In vivo studies were carried out by using female C57Bl/6 mice with an age of eight weeks, obtained from Instituto de Investigaciones Científicas y Servicios de Alta Tecnología (INDICASAT)’s mouse facility. Mice were kept at 25 °C under a light/dark cycle of 12 h and had free access to food and water. All experiments were performed in accordance with guidelines from the Institutional Animal Welfare Committee and the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The protocol was also approved by the Institutional Animal Care and Use Committee of INDICASAT AIP (IACUC-15-004).
2.2. Acute Pulmonary Inflammation
C57BL/6 mice (n = 5) were anesthetized with Ketamine/Xylazine (93/6 mg/Kg) and then treated by intranasal inoculation with lipopolysaccharide (LPS) from Escherichia coli 0111:B4 (Sigma Aldrich, Saint Louis, MO, USA) (0.5 mg/Kg) or saline for control group. Compound 1 (5 mg/Kg) was administered by intraperitoneal (i.p.) injection 2 h before and 10 h after LPS administration. The control group was not treated with compound 1. Mice were euthanized 24 h after the challenge with LPS and the concentrations of tumor necrosis factor (TNF) and interleukin (IL)-6 were determined in lungs and in bronchoalveolar lavage (BAL) by the enzyme-linked immunosorbent assay (ELISA) method. The expression of proteasome and immunoproteasome subunits was determined by quantitative polymerase chain reaction (PCR) in whole lung homogenate of animals from LPS and control groups.
2.3. Cell Culture and Proteasome Activity Assay
Peritoneal macrophages from C57BL/6 mice were obtained five days after intraperitoneal instillation of 2 mL of thioglycollate 3%, by peritoneal washing with chilled Roswell Park Memorial Institute (RPMI) medium. Cells were seeded in RPMI with 10% fetal calf serum (FCS) at a density of 1 × 106
/well in 24-well plates and cultured for 2 h at 37 °C in an atmosphere of 5% CO2
. Non-adherent cells were removed by washing and adherent cells were pre-incubated with compound 1 (25 μM) or isogorgiacerodiol (25 or 50 μM) for 2 h at 37 °C in an atmosphere of 5% CO2
. Then, cells were stimulated with bacterial LPS from Escherichia coli
0111:B4 (InvivoGen, San Diego, CA, USA) (1 μg/mL) for different periods of time (2, 4 or 8 h). Supernatants were discarded and cells were incubated with the fluorogenic peptide Suc-Leu-Leu-Val-Tyr-AMC to evaluate the proteasome CTL activity or Z-Leu-Leu-Glu-AMC to evaluate caspase-like activity as previously described [24
]. After 2 h, supernatants were harvested and the fluorescence of free fluorophore 7-amino-4-methycoumarin (AMC) was measured by using FLx800 BioTek (Winooski, VT, USA) at wavelength/band pass 360/40 for excitation and 460/40 for emission.
2.4. Western Blot Analysis
Western blot analysis was performed as previously described by González et al. [16
]. Briefly, peritoneal macrophages were stimulated with 1 μg/mL of LPS with or without 25 μM of compound 1. Cells were further lysed and 20 μg of total extracts were diluted in loading buffer, boiled and applied to a sodium dodecyl sulfate (SDS) polyacrilamide gel (12%) under reducing conditions. Protein bands were transferred to a polyvinylidene difluoride (PVDF) membrane and further incubated overnight with a monoclonal antiubiquitin antibody specific for Lys48 [26
]. For Western blot image densitometry, ImageJ v. 1.50i [27
] was used as recommended by the software developer.
2.5. Major Histocompatibility Complex Class I Flow Cytometry Analysis
The experiments of cell surface quantification of MHC-I expression were performed in bone marrow-derived macrophages (BMDM), since these cells have lower levels of basal MHC-I expression than elicited peritoneal macrophages. The BMDM were extracted and cultured as previously described by González et al. [16
]. Cultured cells were stimulated with LPS at a concentration of 1 μg/mL in the presence or absence of compound 1 (25 μM). After 24 h, expression of MHC-I was evaluated by flow cytometry.
After 24 h of stimulus, cells were collected, washed with phosphate-buffered saline (PBS) and blocked for 15 min with 200 µL bovine serum albumin (BSA) 1% in PBS. Cells were then washed and incubated for 30 min at 4 °C with antimouse CD11b fluorescein isothiocyanate (FITC) (5 µg/mL) and/or phycoerythrin (PE) antimouse H-2Ld/H-2Db clone 28-14-8 (Biolegend, San Diego, CA, USA) diluted in PBS BSA 1%. After several washes, cells were resuspended in PBS and analyzed by flow cytometry. Event acquisition was performed with a Partec CyFlow® cytometer and the data were analyzed using FlowMax software (PARTEC, Münster, Germany) and FCS Express 4 Flow Cytometry (De Novo software, Los Angeles, CA, USA).
2.6. Quantitative Real Time Polymerase Chain Reaction
Elicited peritoneal macrophages were stimulated with LPS (10 ng/mL) for 2, 4 and 8 h or preincubated for 1 h with compound 1 (25 μM), PR-957 (200 nM) or Polymyxin B (15 μg/mL) and then stimulated with LPS (10 ng/mL) for 8 h. After the stimuli, total RNA was extracted using TRIzol (Life Technology Corporation: Invitrogen and Applied Biosystems, Foster City, CA, USA). The cDNA was obtained by using a High-Capacity cDNA Reverse Transcription Kit (Life Technology Corporation: Invitrogen and Applied Biosystems). The ABI 7500 (Applied Biosystems) was used to perform the quantitative real-time PCR analysis using SYBR Green master mix (Applied Biosystems). Amplification conditions were: 95 °C (10 min), 40 cycles of 95 °C (15 s), and 59 °C (60 s). The data were normalized to the hypoxanthine phosphoribosyltransferase (HPRT) expression and were represented as the difference relative to the control level. The 2-∆∆CT method was used to analyze the relative gene expression. The following forward/reverse primer pairs were used: 5’-TGACCAAGGACGAATGTCTG-3’/5’-GATTTGGTCTCCCAAAAGCA-3’ for β1; 5’-GTGAATCAGCACGGGTTTT-3’/5’-AATCCGCTGCAACAATGACT-3’ for β5; 5’-CATCATGGCAGTGGAGTTTGAC-3’/5’-ACCTGAGAGGGCACAGAAGATG-3’ for β1i; 5’-ACCACACTCGCCTTCAAGTTC-3’/5’-GCCAAGCAGGTAAGGGTTAATC-3’ for β5i and 5’-GCTGGTGAAAAGGACCTCT-3’/5’-CACAGGACTAGAACACCTGC-3’ for HPRT.
2.7. Purified Proteasome Activity Assay
The assay was performed by using the Proteasome-GloTM Assay Systems (Promega, Madiscon, WI, USA) following the manufacturer’s instructions. Briefly, 2 nM of mouse 20S proteasome or immunoproteasome (Boston Biochem, Cambridge, MA, USA) were added to a white 96-well plate in a volume of 50 μL. Then, 50 μL of compound 1 was added at different concentrations (50, 25 and 12.5 μM) and the plate was incubated for 1 h at 37 °C. Following, 50 μL of the Proteasome-GloTM reagent containing the luciferin detection reagent and the substrates (Suc-LLVY for Chymotrypsin-like or Z-nLPnLD for Caspase-like) were added. The plate was mixed at 300–500 rpm for 30 s and incubated at room temperature for 30 min. Luminescence was measured by using a plate reader Synergy HT from BioTek.
2.8. Cytokine Measurements
Animals were euthanized and the BAL was obtained by injecting 1 mL of PBS into the trachea and collecting again. The BAL was centrifuged and the supernatants were harvested. The whole lung homogenate was obtained by homogenization of the tissue in 1X PBS containing 0.1% of Triton100 and protease inhibitors. Homogenates were centrifuged and the supernatants were stored. The concentrations of TNF and IL-6 in lungs and BAL were determined by ELISA (DuoSet kit, RD System, Minneapolis, MN, USA), according to the manufacturer’s protocol.
2.9. Molecular Docking Simulation
ACD/ChemSketch v. 12 (ACD/Labs, Toronto, Canada) was used to draw the 2D structures of compound 1 and isogorgiacerodiol, and to convert them into 3D structural data. Avogadro v.1.1.1 [28
] was used to optimize the geometry of both molecules. Antechamber v.1.27 [29
] was used for an additional energy minimization step before docking. For preparation of the receptor protein, we isolated the dimer formed by chains K and L from the murine constitutive proteasome (Protein Data Bank (PDB) accession code 3UNB) and immunoproteasome (PDB accession code 3UNF), corresponding to subunits β5/β5i and β6, respectively. All ligands and water molecules were removed before docking. The co-crystallized ligand in both structures was used to define the position of the binding site of subunits β5/β5i and to set grid parameters for docking. Docking simulations per se were performed with Dock v. 6.7 [30
] by using grid score and Hawkings GB/SA as primary and secondary energy scoring functions, respectively. The program was set to output the best 10 docking poses and those with the lowest energy were selected for further analyses in each experiment. Prediction of hydrogen bonds and other noncovalent interactions were done with Chimera v.1.11 [31
] and LigPlot+ [32
2.10. Statistical Analysis
Data are presented as means + standard error of the mean (SEM) or mean + standard deviation (SD). Results were analyzed using a statistical software package (GraphPad Prism 6). Statistical analyses were performed by unpaired t test, Mann-Whitney test, Kruskal-Wallis multiple comparisons test. A difference between groups was considered to be significant if p 0.05 (* p 0.05; ** p 0.01; *** p 0.001). The half maximal inhibitory concentration (IC50) was calculated adjusting a sigmoidal dose−response curve following GraphPad Prism 6 procedure.
We have previously shown that compound 1 inhibits the inflammatory response induced in macrophages after LPS challenge by a mechanism involving the reduction of IκBα degradation [16
]. Since the proteasome is critical for IκBα degradation and activation of NFκB, we suspected that the compound might be interfering with the activity of the proteasome. Here, we found accumulation of polyubiquitinated proteins in murine macrophages stimulated with LPS and treated with compound 1, which correlates with a reduced proteasomal CTL and caspase-like activities. Analyses on purified proteasomes revealed that compound 1 inhibits the CTL activity of immunoproteasome. We also showed that compound 1 reduces the LPS-induced surface expression of MHC class I molecules in vitro and the production of proinflammatory mediators in vivo. Docking simulations have predicted a selective interaction of compound 1 with the β5i subunit. Thus, the anti-inflammatory effect of this compound might be dependent, at least in part, on the modulation of immunoproteasome activity.
It has been previously shown that LPS induces proteasomal activation in immune cells and that a proteasome inhibitor, lactacystin, blocks the expression of multiple genes involved in the response of macrophages to LPS [4
]. Hence, we determined the effect of compound 1 on the proteasome activity in LPS-stimulated macrophages. Considering the relevance of chymotrypsin-like activity of proteasomes in the inflammatory response of macrophages [37
], we evaluated the effect of compound 1 on β1i and β5i immunoproteasome subunits and their counterparts in the constitutive proteasome. We showed herein that compound 1 inhibits the CTL activity of the proteasome, thus resulting in the accumulation of polyubiquitinated proteins. This effect of compound 1 on CTL activity did not occur in the absence of LPS, suggesting that the compound might be modulating the immunoproteasome activity. Compound 1 also inhibited the caspase-like activity of the proteasome; however, this effect did not exclusively occur in the presence of LPS. The inhibition of constitutive proteasomal CTL activity has been previously reported for different plant-derived terpenoids, and this effect is associated with anti-inflammatory and/or anti-cancer properties of these compounds [39
The effect of compound 1 on purified immunoproteasome and constitutive proteasome was further evaluated. While compound 1 inhibited the CTL activity of immunoproteasome at the three concentrations analyzed, the inhibition of the constitutive proteasome only occurred at a higher compound concentration. Previous reports have shown that in RAW 264.7 cells early TNF production induced by LPS is not regulated by the immunoproteasome and that the inhibition of β2 and β5 constitutive proteasome subunits is required for a decrease in the production of this cytokine [5
]. However, we have previously demonstrated that compound 1 inhibits the expression and secretion of inflammatory mediators, including TNF, in peritoneal macrophages stimulated with LPS as early as 3 h of stimulus [16
]. Differences in the kinetic expression of immunoproteasome subunits induced by LPS in cell lines and primary macrophages might explain these discrepancies.
In immune cells, pro-inflammatory stimuli induce the replacement of the constitutive proteasome by the immunoproteasome, which increases MHC class I antigen processing and regulates inflammatory responses. Stimulation of cells with IFN-γ and TNF leads to the expression of immunoproteasome subunits [26
]. It has been reported that the expression of LPS-induced immunoproteasome subunits is implicated in the production of certain inflammatory mediators [5
]. We have shown that LPS preferentially increased mRNA expression of β1i and β5i compared to β1 and β5 in murine peritoneal macrophages that were not affected by treatment with compound 1
. No expression of mRNA for proteasome subunits was observed in control cells without stimulus. However, β5i protein has been detected in the cytoplasm of nonstimulated peritoneal macrophages [17
]. Although we have not found differences in mRNA expression between β1i and β5i in LPS-stimulated macrophages, other authors have reported higher protein levels of β1i than those of β5i after stimulation with IFN-γ [17
]. Previous reports of the occurrence of mixed-subunit proteasomes, involving one or two inducible subunits coupled with constitutive ones, may support these findings [12
]. Further studies are necessary to characterize the proteasome composition influenced by LPS stimulus.
Immunoproteasome has been largely implicated in the processing of MHC class I-restricted peptides [43
]. Activation of cytotoxic T lymphocytes depends on the recognition of peptides presented by MHC-I molecules. Immunoproteasome generates peptides for MHC-I presentation more efficiently than the constitutive proteasome, probably by means of the substitution of the caspase-like activity of β1 subunit by the CTL activity of β1i [44
]. Proinflammatory stimuli such as IFN-γ, TNF and LPS upregulate the cell surface expression of MHC-I molecules [45
]. Our results show that LPS induces an increment in the levels of MHC-I in cell surface, an effect that was avoided by compound 1. These results are congruent with the idea that compound 1 might be interfering with immunoproteasome activity, affecting MHC-I expression. Our data are consistent with previous findings in which the inhibition of immunoproteasome subunits by using PR-957 reduces cell surface expression of MHC-I in splenocytes and cytokine production in monocytes [47
]. Deficiency of β5i in mice generates a reduction in MHC-I cell surface expression levels compared to wild type mice [2
], which was not observed in the absence of β1i [48
]. These results point out a role of β5i in MHC class I expression.
In vivo, inhibition of immunoproteasome modulates immune responses and disease progression in several models. Treatment of animals with PR-957 reduces signs of experimental arthritis, which is associated with a reduction in joint expression of proinflammatory mediators [47
]. This inhibitor also attenuated the progression of experimental autoimmune encephalomyelitis and prevented the expression of pro-inflammatory mediators in the brain and spinal cord of animals [49
]. A mouse colitis model has revealed that the blockage of β5i subunit reduced the pathological sings of the disease and cytokine production in the colon [34
]. Immunoproteasome subunits are rapidly induced in lungs after viral infection of mice [50
]. Here we have shown that compound 1 inhibited the production of proinflammatory mediators in the lungs and in the bronchoalveolar lavage of mice treated with LPS. This effect could be at least partially dependent on the inhibition of immunoproteasome subunits, since LPS significantly upregulated the expression of β1i and β5i in lungs. Further studies are necessary to demonstrate the interaction of compound 1 with immunoproteasome in vivo.
Molecular docking simulations supported the notion that compound 1 inhibits CTL activity of the immunoproteasome. The compound was predicted to bind to the catalytic site of the β5i subunit, oriented towards its S1 specificity pocket. The compound forms at least two hydrogen bonds with residues from the subunit, one between a methoxyl group and residue Gly23, and another between a carbonyl group and the N-terminal Thr1 residue. Thr1 is actively involved in the catalytic mechanism of the subunit [51
] and was recently shown to participate in the activation of proteolytic activity during biogenesis of the proteasome [53
]. Furthermore, several well-studied proteasome inhibitors, such as PR-957, interact with this residue [36
Conversely, in the subunits of the constitutive proteasome, compound 1 was predicted to bind towards a small cavity of the neighboring β6 subunit, leaving the S1 pocket empty. The compound also appears to form a hydrogen bond with residue Gly23, but involves a carbonyl group at C-20 rather than a methoxyl group. Taken together with the experimental evidence, these predictions suggest that compound 1 might selectively inhibit the CTL activity of the immunoproteasome, by binding to the β5i subunit. Docking of the structurally-related compound isogorgiacerodiol, which presents a substitution of the methoxyl by a hydroxyl group at C-9, was predicted to bind outside the catalytic site of the β5i subunit, oriented towards the neighboring β6 subunit. These results point out the methoxyl group of compound 1 as critical for the orientation that facilitates its binding to the catalytic site of the β5i subunit.
Since it has been demonstrated that the β1i subunit of the immunoproteasome has CTL activity, we also evaluated the interaction of compound 1 with this subunit. The compound was also predicted to bind to the catalytic site of the β1i subunit, with binding energy estimates similar to those for the β5i subunit. Results suggest that the effect of the compound on the CTL activity of the immunoproteasome could be the consequence of binding to two subunits with similar activity. This finding is consistent with reported evidence that inhibition of multiple proteolytic sites is needed for a marked reduction of proteasome-mediated proteolysis [54
Different contributions to inflammatory responses and other functions related to the immune system have been attributed to different proteasome subunits. Deficiency in the β1i subunit induced a reduction of cytokine production by murine B cells stimulated with LPS, through a mechanism partially dependent on NFκB inhibition [12
]. Altered NFκB activity in nonobese diabetic mice has been attributed to a defect in proteasome function due to a lack of β1i subunit expression [10
]. Selective inhibition of the β5i subunit leads to partial reduction of TNF and IL-6 production induced by LPS in vitro, which is completely abrogated if β1i and β2i are also inhibited [47
]. The in vitro and in vivo identification of heterogeneous proteasome populations that differ in their enzymatic features [55
] may explain the role attributed to individual proteasome subunits in pathological conditions.
Proteasome inhibitors are promising candidates for the treatment of inflammatory diseases and cancer. Currently, there are three proteasome inhibitors approved by the United States Food and Drug Administration (FDA) for clinical use in humans, namely, bortezomib, carfilzomib and ixazomib. The immunoproteasome is overexpressed in malignant cells and in cells involved in autoimmune disorders [20
]. Selective inhibitors of the immunoproteasome have the advantage of being effective as a treatment for such conditions, while preventing the onset of undesirable side effects associated with the inhibition of the constitutive proteasome [40
]. Here we propose a natural compound as a potential specific inhibitor of the CTL activity of the immunoproteasome, opening a path for further studies to characterize this compound as a new agent for the treatment of inflammatory and autoimmune diseases.