Inhibition of Pro-Inflammatory Cytokines by Metabolites of Streptomycetes—A Potential Alternative to Current Anti-Inflammatory Drugs?

Current treatment of chronic diseases includes, among others, application of cytokines, monoclonal antibodies, cellular therapies, and immunostimulants. As all the underlying mechanisms of a particular diseases are not always fully clarified, treatment can be inefficient and associated with various, sometimes serious, side effects. Small secondary metabolites produced by various microbes represent an attractive alternative as future anti-inflammatory drug leads. Compared to current drugs, they are cheaper, can often be administered orally, but still can keep a high target-specificity. Some compounds produced by actinomycetes or fungi have already been used as immunomodulators—tacrolimus, sirolimus, and cyclosporine. This work documents strong anti-inflammatory features of another secondary metabolite of streptomycetes—manumycin-type polyketides. We compared the effect of four related compounds: manumycin A, manumycin B, asukamycin, and colabomycin E on activation and survival of human monocyte/macrophage cell line THP-1. The anti-cancer effect of manucycine A has been demonstrated; the immunomodulatory capacities of manumycin A are obvious when using micromolar concentrations. The application of all four compounds in 0.25–5 μM concentrations leads to efficient, concentration-dependent inhibition of IL-1β and TNF expression in THP-1 upon LPS stimulation, while the three latter compounds show a significantly lower pro-apoptotic effect than manumycin A. We have demonstrated the anti-inflammatory capacity of selected manumycin-type polyketides.


Introduction
Modern human society suffers from an alarming increase in immune system diseases associated with chronic inflammation. These vary in the target organ or organ system and also in severity of symptoms, ranging from mild disorders to serious life-threatening diseases [1]. Many of these, such as rheumatic arthritis, psoriasis, inflammatory bowel diseases, asthma, diabetes, etc., substantially decrease the quality of life and life-span and cause important economic losses. Inflammatory processes have recently been linked to many other diseases such as cancer [2,3] and atherosclerosis [4,5].
Though the effort to find a reliable treatment of inflammatory diseases is immense, breaking news on a universal drug with only mild side-effects are still missing. Currently, most of the clinically

Bacterial Strains and Active Compounds
Manumycin A was purchased from Sigma-Aldrich (St. Louis, MO, USA), manumycin B from Abcam (Cambridge, UK). Asukamycin A was isolated from the culture of Streptomyces nodosus ssp. asukaensis and colabomycin E from the culture of Streptomyces aureus SOK1/5-04 by a procedure described by Petrickova [34]. Strain SOK1/5-04 was isolated from colliery spoil heaps and is deposited in the Biology Centre Collection of Organisms (BCCO, No. BCCO 10_0005, www.actinomycetes.cz). The isolation quality check was performed using LC-MS described therein. All the compounds were dissolved in DMSO (Sigma-Aldrich (St. Louis, MO, USA), tissue culture grade) in 1 mM concentrations.

Cell Culture Conditions
Human monocytic leukemia cells THP-1 (purchased from American Type Culture Collection, ATCC) were used for the evaluation of the anti-inflammatory effect of manumycin-type metabolites. As the best studied, and until recently the only commercially available, manumycin A has been assayed for its anti-cancer activity in numerous cancer cell lines [27][28][29]. However, its anti-inflammatory activities have long been neglected. We have shown that manumycin A downregulates the transcription of the following pro-inflammatory genes: Il6, Tlr8, Il1b, Il10, and Egr1, and inhibits IL-1β, IL-6, and IL-8 production in the TNF-α-stimulated THP-1 monocyte cell line and human peripheral blood monocytes in 0.25-1 mM concentrations. Additionally, IL-18 release was inhibited in THP-1 cells under these conditions. However, in concentrations over 2 mM, manumycin A negatively affected the cell viability in a dose-dependent manner, perhaps due to the induction of apoptosis [24].
The aim of this work was to assess and compare anti-inflammatory and cytotoxic activities of four metabolites from the manumycin family, the type compound, manumycin A [30], and three related compounds-newly commercially available smaller derivative manumycin B [31], asukamycin [32,33], and our newly discovered compound, colabomycin E [34]. The last two were isolated in our laboratory from the cultures of producer streptomycete strains. Based on our previous observation, cytokines IL-1β and TNF-α are the most sensitive markers reflecting downregulation of inflammation using THP-1 cell line in our experimental setup [15]. Therefore, the capacity of manumycin-type metabolites to suppress both gene expression and secretion of IL-1β and TNF-α has been tested. The particular enzyme-inhibitory activities are presumably driven by different metabolite's molecule moieties. As they all together determine the overall effect of each compound on the human immune cells, we aimed to select a compound with the best anti-inflammatory, but weak pro-apoptotic features. Such compounds may represent an attractive candidate as a future anti-inflammatory drug lead.

Bacterial Strains and Active Compounds
Manumycin A was purchased from Sigma-Aldrich (St. Louis, MO, USA), manumycin B from Abcam (Cambridge, UK). Asukamycin A was isolated from the culture of Streptomyces nodosus ssp. asukaensis and colabomycin E from the culture of Streptomyces aureus SOK1/5-04 by a procedure described by Petrickova [34]. Strain SOK1/5-04 was isolated from colliery spoil heaps and is deposited in the Biology Centre Collection of Organisms (BCCO, No. BCCO 10_0005, www.actinomycetes.cz). The isolation quality check was performed using LC-MS described therein. All the compounds were dissolved in DMSO (Sigma-Aldrich (St. Louis, MO, USA), tissue culture grade) in 1 mM concentrations.

Cell Culture Conditions
Human monocytic leukemia cells THP-1 (purchased from American Type Culture Collection, ATCC) were used for the evaluation of the anti-inflammatory effect of manumycin-type metabolites. THP-1 cells were cultured as described previously [24]. Briefly, THP-1 cells were cultured in RPMI-1640 medium (Sigma-Aldrich, St. Louis, MO, USA) supplemented with 10% fetal calf serum, 2 mM L-glutamine, penicillin, and streptomycin (Sigma-Aldrich, St. Louis, MO, USA) in cell culture flasks passaged every third day. To evaluate the immunomodulatory effect of streptomycetes metabolites, cells were transferred to a 12 well plate and seeded at concentration 10 6 cells/1 mL. The total volume of cell suspension was 2 mL/well. Cells were cultured under a 5.5% CO 2 atmosphere at 37 • C. Cells were stimulated with LPS (1 µg/mL) together with manumycin-type metabolites (manumycin A, manumycin B, colabomycin E, and asukamycin) at three different concentrations (5 µM, 1 µM, and 0.25 µM). Only LPS-stimulated cells (1 µg/mL, cat. no. L-2654, Sigma Aldrich, St. Louis, MO, USA) were used as a positive control to evaluate the immunosuppressive capacity of manumycin-type metabolites. The impact of streptomycin on THP-1 viability was assessed by Trypan blue staining.

RNA Extraction
Total RNA was extracted as described previously [35]. Briefly, cells were transferred to 5 mL tubes, spun down, and the supernatant was discarded. Cell pellet was lysed, and RNA extracted using RNeasyMini kit (Qiagen, Hilden, Germany) according to the manufacturer recommendation. Concentration and purity of isolated RNA were checked by Nanodrop.

Quantitative RT-PCR
RNA was reverse transcribed as described previously [35] using a High Capacity cDNA Reverse Transcription Kit (4368813, ThermoFisher Scientific, Waltham, MA, USA). Gene expression of Il1b and Tnfa was determined using TaqMan Assays (Il1b Hs00174128_m1, Tnfa Hs00174128_m1). Peptidyl prolyl isomerase A (Ppia, Hs99999904_m1) was used as endogenous control. Relative quantification of gene expression was calculated as described previously [35]. Briefly, the Ct (number of the cycles needed to exceed the fluorescence threshold) of the housekeeping gene (Pppia) was subtracted from the Ct of the gene of interest (either I11b or Tnfa). This difference (∆Ct) was calculated for all samples. Next, ∆∆Ct was calculated as the difference between ∆Ct of the particular sample minus the average value of ∆Cts of control samples (unstimulated THP1 cells). Finally, the results of gene expression were expressed as 2 -∆∆Ct .

Cytokine Detection by ELISA
The effect of manumycin-type metabolites on the capacity of THP-1 cells to release cytokines was measured by ELISA. After 24 h stimulation of THP-1 cells with manumycin-type metabolites, cell culture supernatants were collected and stored in −20 • C until processed. Concentration of cytokines was detected using kits for IL-1β (88-7261-88) and TNF-α (88-7346-88) both purchased from ThermoFisher Scientific (Waltham, MA, USA). Absorbance was read at 492 nm using ELISA reader (Tecan, Schoeller Instruments, Prague, Czech Republic). Concentration was subtracted from standard calibration curve using software KIM (Schoeller Instruments, Prague, Czech Republic).

Statistics
Results were statistically evaluated and graphically processed using Graphpad Prism (San Diego, CA, USA). Data were statistically analyzed using a t-test with a Bonferonni post-hoc test for multiple comparisons. Statistical significance was set at p ≤ 0.05. Results are expressed as a mean with the standard error from three independent experiments.

The Effect of Manumycin-Type Metabolites on THP-1 Viability
We have determined the possible toxicity of novel streptomycetes metabolites (manumycin A-ManA; manumycin B-ManB; colabomycin E-Col; and asukamycin-Asu) on the viability of THP-1 cells at different time points. None of the metabolites had a substantial effect on the viability of  Figure 2A,B. However, after 24 h, the cytotoxicity of ManA at 5 µM (p = 0.00097) and 1 µM (p = 0.0419) concentrations and ManB at a 5 µM concentration (p = 0.0279) was observed, Figure 2C. Both compounds isolated in our laboratory (Asu, Col) did not show the cytotoxic effects in the assayed concentration range.
Microorganisms 2020, 8, x FOR PEER REVIEW 5 of 12 THP-1 cells at different time points. None of the metabolites had a substantial effect on the viability of the cell line in shorter exposure times (1 and 4 h) and in the assayed concentration range, Figure  2A and 2B. However, after 24 h, the cytotoxicity of ManA at 5 μM (p = 0.00097) and 1 μM (p = 0.0419) concentrations and ManB at a 5 μM concentration (p = 0.0279) was observed, Figure 2C. Both compounds isolated in our laboratory (Asu, Col) did not show the cytotoxic effects in the assayed concentration range.

The Capacity of Metabolites to Change the Gene Expression of Pro-Inflammatory Markers in THP-1
The anti-inflammatory effect of the model compound, manumycin A, has already been published [24]. The current study compared the anti-inflammatory effect of another three manumycintype metabolites at different time points. The ability of ManA, ManB, Asu, and Col to influence the gene expression of pro-inflammatory markers (Il1b, Tnfa) was determined after 1 and 4 h of stimulation.   Figure 3B.
In order to get the pilot data on the kinetics of manumycin metabolite inhibitory action, the Il1b and Tnfa genes transcription profiles were assayed also after 4h of exposition to manumycin-type compounds. Indeed, the presence of LPS significantly promoted gene expression of Il1b

Effect of Manumycin-Type Metabolites on Cytokine Production
The suppressing effect of manumycin-type metabolites on the secretion of pro-inflammatory cytokines was measured by ELISA in cell culture supernatants after 24 h incubation. The concentration of IL-1β was increased in LPS stimulated sample compared to non-stimulated control (p = 0.0022). A profound effect of ManA on the decrease of IL-1β production was observed only when 5 µM ManA was used (p = 0.0028). Concentrations of 1 µM and 0.25 µM ManA were not able to lower IL-1β secretion in comparison to LPS-stimulated cells being still increased in comparison to 5 µM ManA (p = 0.0191 and p = 0.0482, respectively). A concentration of 5 µM ManB decreased secretion of IL-1β (p = 0.0027); however, 1 µM and 0.25 µM ManB were not able to inhibit IL-1β secretion as efficiently as 5 µM ManB (p = 0.0484 and p = 0.0485, respectively). Only a marginal suppressing effect on the pro-inflammatory marker IL-1β secretion by THP-1 cells supplemented with various concentrations of Asu was observed, but the highest concentration of Col limited release of IL-1β (p = 0.0358), Figure 4A.

Discussion
Manumycin-type metabolites show quite complex inhibitory effects on key human enzymes  Figure 2C) in comparison to ManA, Col seems to be a good candidate to be further tested for its anti-inflammatory properties.

Discussion
Manumycin-type metabolites show quite complex inhibitory effects on key human enzymes [20,21,[24][25][26]36]. Many of these can be directly linked either to the regulation of cell proliferation and apoptosis or to the control of inflammatory response. Our previous study has shown that the most studied representative of this family of streptomycete secondary metabolites, manumycin A, exhibits both activities in similar active concentration in in vitro experiments using both human THP-1 monocyte/macrophage cell line and human peripheral blood monocytes. The pro-apoptotic and anti-inflammatory activities can be measured in as low as 1 µM concentrations of ManA and are concentration-dependent. This makes Man A an excellent candidate for a cancerostatic agent, which has been proven by numerous published studies before, e.g., the work of Di Paolo [28]. Additionally, it has been shown to work synergistically in combination with another cancerostatic agents, such as paclitaxel [37]. However, manumycin A-due to the high pro-apoptotic activity-cannot be considered a good candidate for a potential anti-inflammatory drug.
Our present study documents that other manumycin-type compounds show more suitable activity profiles. The cytotoxicity of all three other manumycins (ManB, Asu and Col) is substantially reduced, though their anti-inflammatory features remain the same or just slightly lower than in ManA. Though the set of structurally-related compounds is too small for a SAR (structure-activity relationship) assay, we can already relate some substructure-related activity changes. As the assayed compounds vary most exclusively in the structure of the upper polyketide chain, we expect that this part of the molecules is responsible for their variable cytotoxicity or pro-apoptotic features. The cytotoxicity should be, based on the current knowledge, directed by inhibition of two enzymes: the Ras-specific farnesyltransferase [20] and the TrxR-1cytosolic thioredoxin reductase [21]. The first uncouples the Ras G-protein from the cytosolic membrane and breaks up its signaling pathway. It can be speculated that the farnesyl transferase active site may bind the upper chain of manumycins due to its similarity to the farnesyl pyrophosphate, the donor of the farnesyl residue, as it is seen in some other FT inhibitors [38]. The strength of the competitive enzyme inhibition then depends on the structure of the upper chain. The second, TrxR-1-directed, inhibitory activity leads to an increased production of reactive oxygen species, one of the main elicitors of apoptosis [21,39], and is based on an irreversible adduct formation followed by conformational and activity changes. Similar mechanism has been shown also in algal brevetoxins [39]. This activity is supposed to rely on the central cyclic mC 7 N unit of manumycins; thus, we could expect that it will be less dependent on the structure of the variable upper polyketide chain and thus may be responsible for the basal pro-apoptotic activity found in all assayed manumycins. The same central part of the molecule is supposed to cause a covalent homodimerization of IKKβ leading to the inhibition of constitutive and TNF-α induced NF-κB activity with anti-inflammatory effects. The molecular mechanism of the second inflammatory response-inhibiting activity, the caspase 1 inhibition, has not been described yet.
In conclusion, our data suggest that manumycins may serve as promising drug for the suppression of excessive inflammation. Though the type molecule, manumycin A, shows too strong cytotoxicity, variations in its upper polyketide chain can change the ratio between pro-apoptotic and anti-inflammatory activities towards the second one. From this point of view, colabomycin E seems to be a better candidate for more detailed and mechanistic analysis of its anti-inflammatory potential. In general, the physicochemical properties of manumycins resemble those of plasma membrane phospholipids, it is generally expected that the compounds can cross the cytoplasmic membranes without any requirements for a transport system. The mouse model experiments also suggest their low toxicity and side-effects even in a long-time treatment [40]. We believe that streptomycetes and related actinomycetes produce even more immune cells-targeting active compounds. Our pilot experiments with selected, mostly human body-associated, streptomycetes showed that numerous strains produce various secondary metabolites with strong immunomodulatory effects (data not shown). This makes us believe that the immunomodulatory activities of streptomycete secondary metabolites have long been understudied and underestimated, compared to antibiotic, antifungal, and anti-cancer drugs. They should be considered as a valuable source for novel immunomodulatory drug leads discovery.