Impavido attenuates inflammation, reduces atherosclerosis, and alters gut microbiota in hyperlipidemic mice

Summary

Impavido (Miltefosine) is an FDA-approved drug for treating leishmaniasis and primary amebic meningoencephalitis. We have shown previously that Miltefosine increased cholesterol release and dampened Nlrp3 inflammasome assembly in macrophages. Here, we show that Miltefosine reduced LPS-induced choline uptake by macrophages, and attenuated Nlrp3 inflammasome assembly in mice. Miltefosine-fed mice showed reduced plasma IL-1β in a polymicrobial cecal slurry model of systemic inflammation. Miltefosine-fed mice showed increased reverse cholesterol transport to the plasma, liver, and feces. Hyperlipidemic apoE^−/− mice fed with WTD + Miltefosine showed significantly reduced weight gain and markedly reduced atherosclerotic lesions versus mice fed with WTD. The 16S rDNA sequencing and analysis of gut microbiota showed marked alterations in the microbiota profile of Miltefosine-fed hyperlipidemic apoE^−/− versus control, with the most notable changes in Romboutsia and Bacteriodesspecies. Taken together, these data indicate that Miltefosine causes pleiotropic effects on lipid metabolism, inflammasome activity, atherosclerosis, and the gut microbiota.

Graphical abstract

Highlights

• •Miltefosine modulated LPS-induced choline uptake
• •Miltefosine dampened Nlrp3 inflammasome assembly in mice
• •Miltefosine increased cholesterol efflux and reduced atherosclerosis in mice
• •Miltefosine altered gut microbiota in WTD-fed hyperlipidemic mice

ImmunologyMicrobiologyMicrobiome

Introduction

Leishmania parasite preferentially infects phagocytic cells in human hosts, such as macrophages and dendritic cells.^1,2 Macrophages induce NLRP3 inflammasome assembly during Leishmania infection in both mice and humans, leading to the processing and release of mature interleukin-1 beta (IL-1β) at the infection site.^3,4,5,6 IL-1β promotes disease progression, as the mice lacking NLRP3, ASC, or caspase 1 showed defective IL-1β production at the infection site and were resistant to cutaneous Leishmania infection.⁴ In addition, the IL-1β levels in patients with cutaneous leishmaniasis positively correlate with areas of necrosis.⁷ Miltefosine is an FDA-approved drug to treat visceral and cutaneous leishmaniasis.⁸ The mechanism of action of Miltefosine is not fully clear, but it can freely integrate into the cell membrane and redistribute in the ER, Golgi, and mitochondria.⁸ Studies from our lab and others have shown that Miltefosine causes cholesterol release from cells^9,10 and modulates inflammatory responses in a variety of immune cells including macrophages, mast cells, and eosinophils.^10,11,12 Accumulation and oxidation of excess cholesterol in the arterial intima is the major cause of coronary artery disease (CAD). Oxidized low-density lipoprotein (LDL)-cholesterol (LDL-C) in the artery wall promotes the recruitment of monocytes, which transform into arterial wall macrophages and uptake LDL-C to form lipid-laden foam cells.^13,14,15,16 The oxidized LDL acts as potent activators of the toll-like receptor (TLR) pathway and cholesterol crystals in plaques promote the assembly of NLRP3 inflammasome.^14,17 The inflammasome-mediated processing of IL-1β may lead to beneficial antimicrobial activity but can also result in amplification of an inflammatory cascade, worsening the pathogenesis of various chronic inflammatory diseases such as metabolic syndrome and atherosclerosis.^18,19,20 Inflammasome-mediated activation of caspase 1 and caspase 11 also results in cleavage of Gasdermin D (GsdmD), a downstream effector of inflammasome activity required for efficient release of IL-1β from cells.^{21,22,23,24,25,26} Studies from our lab and others have shown the involvement of GsdmD pathway in promoting atherosclerosis in mice and humans.^{27,28,29,30,31} In addition to chronic inflammation caused by a disruption in sterol homeostasis, metabolites of the dietary phosphatidylcholine (PC), such as choline and trimethylamine N-oxide (TMAO), can also promote chronic inflammatory pathologies such as atherosclerosis and cardiovascular disease (CVD) in a gut-microbiota dependent manner.^32,33,34

In contrast to activated atherogenic pathways, the atheroprotective pathways such as autophagy and reverse cholesterol transport (RCT) are known to become dysfunctional with aging and in advanced atherosclerosis.^17,35,36 We have shown before that Miltefosine promoted cholesterol efflux from macrophages, induced autophagy/mitophagy, and blunted NLRP3 inflammasome assembly. Here, we used a wild-type (WT) and a hyperlipidemic mouse model of atherosclerosis to test the effects of Miltefosine on inflammasome activity, reverse cholesterol transport, atherosclerosis, and gut microbiota.

Results

Miltefosine blunts endotoxin-induced choline uptake

Previous studies have shown that exposure to LPS induces expression of choline transporter CTL1 and increases choline uptake by macrophages.^37,38,39 The LPS-induced choline uptake promotes inflammasome assembly, whereas blocking choline uptake prevents NLRP3 inflammasome.^37,38 We have shown previously that Miltefosine ablates Nlrp3 inflammasome assembly in macrophages,¹⁰ thus we tested if one of the mechanisms of Miltefosine anti-inflammasome activity is via reduced choline uptake. Bone marrow-derived macrophages (BMDMs) from WT-C57BL6J mice were pretreated with ±7.5 μM Miltefosine for 16h. The control and Miltefosine treated BMDMs were primed with 1 μg/mL LPS for 4h, followed by incubation with radiolabeled choline at final concentration of 2.5 μCi/mL for 0, 20, or 40 min. The extracellular media containing radiolabeled choline was removed after the indicated time and the intracellular radioactivity was determined by liquid scintillation counting. As shown in Figure 1A, there was a significant (∼35%) reduction in choline uptake in Miltefosine treated versus control cells at 20 min. At 40 min, Miltefosine treated cells showed a significant (∼40%) reduction in choline uptake versus control cells. The macrophages treated with Miltefosine showed significantly reduced CTL1 induction (Figures 1B and 1C). These data indicate that Miltefosine may be blunting inflammasome assembly via dampening LPS-induced choline uptake by macrophages.

Figure 1.

Miltefosine dampened LPS-induced choline uptake and CTL1 expression

(A) Mouse BMDMs were plated in 6-well plate and treated with ±7.5 μM Miltefosine for 16h. The cells were primed by incubation with ±1 μg/mL LPS for 4h, followed by incubation with tritium labeled choline at final concentration of 2.5 μCi/mL ³H-choline. The radioactivity uptake assay was performed at 37°C and radioactive dpm counts were determined by using scintillation counter. Each sample in a group was compared with others by ANOVA using Tukey’s multiple comparisons test (N = 5, mean ± SD for all groups; for 0 min group, n.s. (nonsignificant), for 20 min group, ∗∗∗∗, p < 0.0001 for untreated versus LPS treated, ∗∗∗, p = 0.0002 for LPS versus LPS + Miltefosine, and ∗∗, p = 0.0024 for untreated versus LPS + Miltefosine, for 40 min group, ∗∗∗∗, p < 0.0001 for untreated versus LPS treated, ∗∗∗∗, p < 0.0001 for LPS versus LPS + Miltefosine, and ∗∗∗, p = 0.0006 for untreated versus LPS + Miltefosine).

(B) Western blot analysis of LPS induced expression of choline transporter CTL1 in control and Miltefosine treated cells, β-actin was used as loading control.

(C) Quantification of western blot bands with graph showing ratio of CTL1 and β-actin expression, with ∗∗∗∗p < 0.0001 by two-tailed t-test.

Miltefosine attenuates in vivo NLRP3 inflammasome activity

To determine the effect of Miltefosine on in vivo NLRP3 inflammasome, the WT mice were fed with a chow diet ±20 mg/kg/day Miltefosine for 3 weeks. The chow-fed and Miltefosine-fed mice were injected i.p. with 5 μg LPS, followed 4h later with i.p. injection of ATP. The mice were sacrificed 30 min later, and the peritoneal cavity lavage was collected and analyzed for IL-1β levels. As shown in Figure 2A, Miltefosine-fed mice showed significantly reduced IL-1β levels in peritoneal lavage fluid. To determine if Miltefosine can also reduce IL-1β levels in hyperlipidemic conditions, the apoE^−/− mice fed with a western type diet (WTD) ± 50mg/kg/day Miltefosine for 3 weeks were injected with LPS +ATP for inflammasome induction. As shown in Figure 2A, the WTD-fed apoE^−/− mice showed robust IL-1β levels in peritoneal lavage, whereas Miltefosine-fed mice showed ∼ 65% reduction. Control mice received either saline or LPS + saline injections and showed negligible levels of IL-1β in peritoneal lavage (Figure S1). To determine if Miltefosine affected the release of IL-1β in plasma, we tested plasma from control or Miltefosine fed mice. The mice fed on chow + Miltefosine or WTD +Miltefosine diet showed significantly reduced IL-1β in plasma (Figure 2B). These data indicated that Miltefosine dampened in vivo NLRP3 inflammasome activity and IL-1β release in mice.

Figure 2.

Miltefosine attenuated in vivo NLRP3 inflammasome activity

(A) Age-matched (10-week-old) male WTC57BL6J mice fed with chow ±20 mg/kg Miltefosine for 3 weeks or C57BL6J-apoE^−/− knockout mice fed with WTD ±50 mg/kg Miltefosine for 3 weeks were used. The mice were primed for inflammasome assembly by an I.P. injection of LPS (5μg/mouse). After 4h of LPS injection, the NLRP3 inflammasome assembly was induced by I.P. injection of ATP (0.5 mL of 30 mM, pH 7.0). The peritoneal cavity was lavaged with 5 mL sterile PBS, and IL-1β levels in peritoneal lavage were determined by ELISA (N = 6, mean ± SD for all groups, ∗∗∗∗p < 0.0001 for C57BL6J-control versus C57BL6J-Miltefosine, ∗∗∗∗p < 0.0001 for C57BL6J-control versus apoE^−/−-control, ∗∗∗∗p < 0.0001 for apoE^−/−-control versus apoE^−/−-Miltefosine by two-tailed t-test).

(B) IL-1β levels in plasma were determined by ELISA (N = 5, mean ± SD for all groups, ∗∗∗p < 0.001 for C57BL6J-control versus C57BL6J-Miltefosine, ∗∗∗∗p < 0.0001 for apoE^−/−-control versus apoE^−/−-Miltefosine by two-tailed t-test).

Miltefosine reduced inflammation-induced cytokine release in hyperlipidemic mice

Hyperlipidemia and sepsis-induced inflammation promote atherosclerosis and CVD in humans.^40,41,42 The pilot studies showed that WT C57BL6J mice fed with either chow +20 mg/kg/day Miltefosine diet or WTD +50 mg/kg/day Miltefosine diet did not lose weight (at week 6 versus body weight at week 0), and no visible adverse effect on mobility or appearance were observed. To determine the effect of Miltefosine treatment on inflammatory responses in hyperlipidemic animals, a polymicrobial cecal slurry injection model of peritonitis was used.⁴³ The cecal slurry injection dose of 4 μL/g body weight was selected as this dose did not induce mortality but still led to a transient decrease in body temperature to 35°C. The apoE^−/− mice fed with a WTD diet ±50 mg/kg/day Miltefosine for 3 weeks were i.p. injected with cecal slurry, and the plasma was collected at 2, 4, and 6h after injection. As shown in Figure 3, the Miltefosine-fed mice group had significantly reduced IL-1β release in plasma versus control mice. These data indicated that Miltefosine reduce inflammatory responses in hyperlipidemic mice.

Figure 3.

Miltefosine dampened IL-1β release in polymicrobial cecal slurry mouse model of systemic inflammation

The age and sex-matched apoE^−/− mice were fed with WTD or WTD +50 mg/kg Miltefosine for 3 weeks. The mice were injected i.p. with cecal slurry (from WT C57BL6J mice) with the dose of 4 μL/g of body weight. The blood was collected by tail bleed at indicated times after injection and plasma IL-1β levels were determined by ELISA (N = 6, mean ± SD for all groups; for 2 h males and females group, n.s. (nonsignificant), for 4h male group ∗∗∗, p = 0.0003, for 4 h female group ∗∗∗, p = 0.0005, for 6h male group ∗∗∗∗, p < 0.0001, for 6 h female group, ∗∗∗∗, p < 0.0001 with two-tailed t-test.

Miltefosine increases reverse cholesterol transport in mice

To determine if Miltefosine treatment in mice leads to increased reverse cholesterol transport (RCT), the RCT assay⁴⁴ was performed in WT mice, as apoE^−/− mice are not suitable for these studies because of inherent defects in lipoprotein packaging and excessive plasma cholesterol pool even with chow-diet feeding. Donor BMDMs from WT mice were loaded with 100 μg/mL acetylated-low-density lipoprotein (Ac-LDL) and radiolabeled ³H-cholesterol for 48 h to generate foam cells, which were implanted s.c. into the back flanks of the 9-week-old WT recipient mice that were fed on chow diet or chow diet containing 20 mg/kg/day Miltefosine. The radioactive cholesterol mobilized to plasma was determined by collection of blood samples at 24, 48, and 72 h. The liver and feces samples were collected post euthanasia and processed to determine the percent of RCT to these pools. As shown in Figure 4A, Miltefosine increased RCT to plasma in mice by ∼30% at 24 h, by 26% at 48 h, and by ∼20% at 72 h, RCT to feces showed ∼31% increase at 24 h, 26% increase at 48 h, and ∼51% increase at 72 h in Miltefosine-fed versus chow-fed mice (Figure 4B). RCT to the liver was also increased by ∼21% in Miltefosine-fed mice (Figure 4C). These data indicated that Miltefosine increased RCT in vivo.

Figure 4.

Miltefosine increased RCT in mice

The foam cells prepared by loading of BMDMs isolated from WT C57BL6J mice with ³H-labeled cholesterol, were transplanted into age-matched WT male recipient mice fed with either chow or chow+ 20 mg/kg Miltefosine for 3 weeks.

(A) RCT to plasma determined at 24, 48, and 72 h (N = 7, mean ± SD for all groups, ∗∗∗, p = 0.004 to 0.005 by two-tailed t-test).

(B) RCT to feces determined at 24, 48, and 72 h (N = 7, mean ± SD for all groups, for 24 h group ∗∗, p = 0.0067 for chow versus Miltefosine by two-tailed t-test, for 48h group, ∗∗, p = 0.0012, and for 72 h group, ∗∗∗, p = 0.0001 for chow versus Miltefosine by two-tailed after t-test.

(C) RCT to liver determined at 72 h (N = 7, mean ± SD, ∗∗, p = 0.0018 for chow versus Miltefosine by two-tailed t-test.

Miltefosine reduced atherosclerosis in hyperlipidemic mice

To determine if Miltefosine can reduce atherosclerosis, 5-week-old apoE^−/− mice were weaned onto the atherogenic western type diet (WTD) or WTD containing 50 mg/kg/day Miltefosine for 9 weeks. The Miltefosine-fed mice gained significantly less body weight versus the WTD-fed group (Figures 5A andS2). No significant differences were found in food intake in WTD versus WTD + Miltefosine group (Figure 5B). The oil red O positive aortic root atherosclerotic lesion areas were significantly reduced in Miltefosine-fed group with ∼50% reduction in atherosclerotic lesions in both males and females (Figures 5C and 5D). There was a trend toward lowered total cholesterol and higher HDL-cholesterol in the plasma of the Miltefosine-fed group, but the changes did not achieve levels of significance (Figure 5E). These data indicated that Miltefosine reduces atherosclerosis and this effect may not be completely dependent on the reduction in cholesterol load.

Figure 5.

Miltefosine reduced atherosclerosis in hyperlipidemic mice

The 5-week-old apoE^−/− mice from both sexes were treated with either WTD or WTD +50 mg/kg Miltefosine diet for 9 weeks (N = 12). The control apoE^−/− mice were fed chow diet throughout the course of the study.

(A) The body weight (BW) gain during course of study. The gain in BW was calculated by subtracting BW at end-point (week 14) from BW at beginning of study (week 5) and was plotted, ∗∗∗∗ indicate p < 0.0001 for chow versus WTD, ∗∗∗∗ indicate p < 0.0001 for chow versus WTD + Miltefosine, ∗∗ indicate p < 0.001, and ∗ indicate p < 0.05 with ANOVA.

(B) Average food intake in mice fed with WTD or WTD +50 mg/kg Miltefosine. The measurements were taken weekly until the end of the study, p = n.s. (non-significant).

(C) Quantification of aortic root lesions in male mice fed with either WTD or WTD +50 mg/kg Miltefosine, ∗∗∗ indicate p < 0.0005, by two-tailed t-test).

(D) Quantification of aortic root lesions in female mice fed with either WTD or WTD +50 mg/kg Miltefosine, ∗∗∗ indicate p < 0.0005, by two-tailed t-test).

(E) Plasma cholesterol (total and HDL) in apoE^−/− mice fed with WTD or WTD + Miltefosine diet was determined by Stanbio kit following manufacturer’s instruction (N = 8 for males and females, n.s. = non-significant by two-tailed t-test).

To determine if Miltefosine reduced inflammasome activity in atherosclerotic plaques, the aortic plaque protein extracts were probed for cleaved forms of IL-1β and Gasdermin D (GsdmD). The WTD + Miltefosine fed mice showed reduced levels of cleaved IL-1β and cleaved GsdmD (Figures 6A–6C). The progression of atherosclerosis is promoted by continuous recruitment of immune cells, such as macrophages, to the plaque area. Expression of cell adhesion molecules, such as VCAM-1, facilitates inflammation-associated vascular adhesion and the trans-endothelial migration of macrophages. Miltefosine-fed mice showed reduced expression of VCAM-1 in atherosclerotic lesions versus mice fed with WTD alone (Figures 6A and 6D). Miltefosine-fed mice showed significantly reduced macrophage infiltration and necrotic area versus mice fed with WTD alone (Figures S3A–S3C). These data indicated that Miltefosine anti-atherosclerosis properties may be partially because of dampened inflammasome activity and reduced macrophage infiltration in atherosclerotic lesions.

Figure 6.

Miltefosine dampens inflammasome activity in atherosclerotic plaques

(A) The 5-week-old apoE^−/− mice from both sexes were treated with either WTD or WTD +50 mg/kg Miltefosine diet for 9 weeks. The atherosclerotic-plaque bearing aortas were excised and protein extracts were prepared using T-PER tissue protein extraction reagent (Thermo Fisher #78510), followed by western blot analysis using mouse antibody specific for cleaved N-terminal fragment of GsdmD or cleaved form of IL-1β or VCAM-1. The star (∗) indicate non-specific band that was present in all samples probed with mouse-specific antibody against cleaved GsdmD.

(B) Quantification of western blot bands with graph showing ratio of cleaved GsdmD and β-actin expression with ∗∗∗∗p < 0.0001 by two-tailed t-test.

(C) Quantification of western blot bands with graph showing ratio of cleaved IL-1β and β-actin expression with ∗∗∗∗p < 0.0001 by two-tailed t-test.

(D) Quantification of western blot bands with graph showing ratio of VCAM-1 and β-actin expression with ∗∗p < 0.005 by two-tailed t-test.

Altered gut microbiota in Miltefosine treated mice

Gut microbiota play a major role in the progression of atherosclerosis and CVD.^32,45,46 Miltefosine was originally identified as an anticancer compound but was later shown to be effective against a variety of microbes including bacteria.⁸ Thus, we determined if Miltefosine modulates the gut microbiota composition by performing a 16s ribosomal DNA (16S rDNA)-based qPCR sequencing. Fresh fecal samples were collected from apoE^−/− mice fed for 9 weeks with either chow diet, WTD, or WTD + Miltefosine. The DNA isolated from these samples was analyzed for the microbial species profile and alpha and beta diversity across different groups. Alpha diversity is a measure of microbiome diversity/complexity in each sample, whereas the beta diversity is a measure of similarity or dissimilarity between groups. The gut microbiota profile was significantly different in three groups with alterations in alpha and beta diversity in chow versus other groups and WTD-fed versus WTD + Miltefosine-fed group (Figures 7A and 7B). The WTD-fed mice showed a marked increase in Romboutsia species versus chow-fed mice (Figure 7C) and this effect was blunted in Miltefosine-treated mice. The Miltefosine-treated group showed increased levels of Bacteroides species versus the WTD-fed group (Figure 7C). These data indicate that Miltefosine alters the gut microbiota profile and this alteration could serve as one of the mechanisms for Miltefosine’s anti-atherosclerotic property.

Figure 7.

Miltefosine alters gut microbiota profile in hyperlipidemic mice

The 5-week-old apoE^−/− mice were fed with either WTD or WTD +50 mg/kg Miltefosine diet for 9 weeks. The control apoE^−/− mice were fed chow diet throughout. Fresh feces were collected and 16s rDNA sequencing was performed.

(A) Alpha diversity in gut microbial profile of apoE^−/− mice fed with different diets.

(B) Beta diversity of microbiota in gut microbial profile of apoE^−/− mice fed with different diets.

(C) Total diversity in gut microbial profile of mice fed with different diets.

Discussion

Miltefosine is an alkyl-lysophospholipid analog with in vitro activity against various Leishmania species. The activity of NLRP3 inflammasome and release of IL-1β during leishmaniasis had been reported in mouse models as well as in human patients.^3,4,5,6 The activation of the NLRP3 inflammasome, instead of serving as a tool to clear the parasitic infection, was found to be detrimental during leishmaniasis and the mice lacking NLRP3, ASC, or caspase 1 were shown to be resistant to cutaneous infection.⁴ Furthermore, the amount of IL-1β positively correlates with areas of necrosis in cutaneous leishmaniasis patients.⁷ Miltefosine reduced in vivo NLRP3 inflammasome assembly and IL-1β release in macrophages and other immune cells.^10,11 Thus, in addition to killing the pathogen, Miltefosine may also be modulating host immune responses to Leishmania infection by regulating inflammasome activity.

Miltefosine displays a range of activities such as anticancer, antimicrobial, effects on lipid metabolism homeostasis, and immune cell function,^8,12,47,48 but the exact mechanism of action of Miltefosine is not fully understood and its activities seem to be cell-type dependent. Miltefosine is known to perturb cellular lipid homeostasis by altering phosphatidylserine (PS) and phosphatidylinositol 4,5-bisphosphate (PIP2) localization in macrophages,¹⁰ and by negatively affecting phosphatidylcholine (PC) and sphingolipid biosynthesis.^48,49 Previous studies have shown that the uptake of PC biosynthetic precursors, such as choline, precedes NLRP3 inflammasome assembly and IL-1β release.^37,38 The cells exposed to LPS stimuli upregulate the expression of choline transporter to import more choline for generating PC. We found that Miltefosine reduced choline uptake in LPS-induced macrophages (Figure 1). Thus, Miltefosine may be dampening inflammasome assembly because of combinatorial inhibition of PC biosynthesis and choline uptake.

Priming of the inflammasome pathway by LPS is dependent on TLR receptors, which are enriched in membrane lipid rafts. Miltefosine-mediated disruption of lipid rafts can attenuate TLR signaling pathway,^10,50 thus we tested the effects of Miltefosine on direct NLRP3 inflammasome activity in live animals. We chose the peritoneal cavity as the site for NLRP3 inflammasome activity as it contains the liver, spleen, GI tract, and a variety of immune cells. The high percentage of naive tissue-resident macrophages in the peritoneal cavity also makes it a suitable site for testing in vivo NLRP3 inflammasome activity. We found that mice fed with Miltefosine had lower IL-1β levels in peritoneal lavage as well as in plasma (Figure 2). Low-grade systemic inflammation is known to promote atherosclerosis and sepsis survivors are at higher risk of death from CVD. Thus, we tested Miltefosine effects in a mouse model of systemic inflammation and we demonstrate that Miltefosine-fed mice had significantly less proinflammatory cytokines in plasma (Figure 3). Miltefosine, thus, may not only be involved in clearing leishmanial infection in human hosts by directly killing the pathogen but also by dampening the detrimental NLRP3-IL-1β pathway to prevent tissue injury.

The limitation of the in vivo inflammasome study is that the individual contribution of B cells, T cells, or macrophages in Miltefosine-mediated blockage of NLRP3 activity is not dissected.

Previous studies showed that Miltefosine promoted cholesterol removal from the cells,^9,10 but the effects of Miltefosine on cholesterol efflux in live animals are not clear. We found that Miltefosine-fed mice showed increased removal of cholesterol from the transplanted foam cells (Figure 4). We speculate that the in vivo cholesterol-removing activity of Miltefosine may be useful to treat metabolic diseases caused by hyperlipidemia. We tested if Miltefosine can reduce atherosclerosis, a disease promoted by hyperlipidemia and inflammation. The apoE^−/− mice were used, as these mice are prone to severe atherosclerosis on feeding with a cholesterol-rich WTD diet. The apoE^−/− mice fed with WTD + Miltefosine showed a significant reduction in weight gain and atherosclerotic plaque formation was significantly reduced in WTD + Miltefosine versus WTD fed mice (Figure 5). One of the reasons of reduced weight gain could be because of increased excretion of cholesterol from body, as the mice fed with Miltefosine showed increased reverse cholesterol transport to feces (Figure 4).

Presence and role of cleaved GsdmD, the final executor of inflammasome activity, in human and mice atherosclerotic plaques is an emerging field^27,31,51 and may become a potential target for future anti-CVD therapeutics. Miltefosine-fed mice showed reduced inflammasome activity in atherosclerotic plaques (Figure 6). Reduced levels of cleaved GsdmD and mature IL-1β ιν atherosclerotic πλαθυεσμαψσερϖε as one of the mechanisms for anti-atherosclerosis activity of Miltefosine.

The chemical structure of Miltefosine is similar to lyso-PC⁸ and Miltefosine negatively affects PC biosynthesis.⁸ Dietary PC is metabolized by gut microbiota and converted to choline and atherogenic metabolite trimethylamine N-oxide (TMAO).^32,45 Given the antimicrobial activity of Miltefosine, we tested if Miltefosine can alter the gut microbiota of hyperlipidemic mice. We found that gut microbiota from the mice fed with WTD had lower alpha diversity compared to chow-fed mice, whereas the mice fed with WTD + Miltefosine showed increased alpha diversity versus mice fed with WTD. Low alpha diversity of gut microbiota has been observed in several metabolic diseases such as obesity, hyperinsulinemia, and dyslipidemia. The WTD-fed mice showed increased levels of Enterococcus versus chow-fed mice, whereas Miltefosine feeding reduced levels of Enterococcus versus WTD-fed mice. Miltefosine also increased levels of Bacteroides species versus WTD. We also found a marked alteration in levels of Romboutsia, with WTD-fed mice showing an increase versus chow-fed mice, whereas mice fed with WTD + Miltefosine showed marked reduction in Romboutsia levels versus WTD-fed mice (Figure 7). Previous studies have shown a differential abundance of Enterobacteriaceae, Bacteroides, and Romboutsia in atherosclerosis.^52,53,54,55 One of the mechanisms by which Miltefosine can alter gut microbiota is via its anti-bacterial properties.⁸ Miltefosine may selectively promote the growth of athero-protective gut microbes while inhibiting the growth of athero-promoting bacterial species.

The limitations of our study are that we did not measure levels of Miltefosine in mouse plasma or determined the tissue-specific distribution or in vivo half-life of Miltefosine. There is no straightforward assay to measure Miltefosine in mouse plasma, and the mass spectrometry methods are not well standardized with reported high variability and only a handful of studies using this method.⁵⁶ Studies using radioactive Miltefosine have shown that it has a wide distribution in the body with high levels in the kidney, intestinal mucosa, liver, and spleen with a half-life of >6 days.⁸ We did not provide the exact mechanism through which Miltefosine elicits an anti-atherosclerotic effect or provide direct evidence that Miltefosine effects are mediated via gut microbial processes. These changes may only be associated with the other observed phenotypes of Miltefosine such as reduction in lipids and atherosclerosis. To prove a mechanistic link would require a transplantation of cecal microbes and transmission of Miltefosine-dependent anti-atherosclerotic effects in germ-free (GF) mice. The successful transmission would also only show some of the effects of the Miltefosine, in part, are transmissible with microbial transplantation to GF recipients, but it still would not provide evidence of particular microbial metabolites, such as TMAO, being the sole mediator of Miltefosine activity amongst all the other numerous microbial processes that were transplanted.

Miltefosine is a broad-spectrum antimicrobial agent that was originally developed in the 1980s as an anticancer drug. Given that Miltefosine had been used in humans for decades, it can be potentially used in lower doses as a stand-alone or as an adjuvant to LDL lowering therapies to treat inflammatory diseases such as atherosclerosis. In support of Miltefosine as a potential anti-atherosclerotic molecule, previous studies have shown that alterations in gut microbial species, either via dietary interventions with chemical compounds such as Metformin and resveratrol or via gavage inoculation, can impact the progression of atherosclerosis.^52,55,57,58 Further studies are required to determine the efficacy of Miltefosine for treating inflammatory metabolic diseases in humans.

Limitations of the study

Although our study showed reduced inflammation and atherosclerosis in Miltefosine-fed mice, it did not provide evidence that reduced inflammation is mechanism by which Miltefosine increases RCT or reduces atherosclerosis. Miltefosine exhibit pleiotropic effects on lipid metabolism and our study did not provide evidence for either choline or cholesterol as the sole mediator of Miltefosine effects on atherosclerosis. Other limitation of the study is that we did not provide the exact mechanism by which Miltefosine reduced weight gain in hyperlipidemic mice and we did not measure energy expenditure or oxygen consumption. The study did not provide evidence that changes in gut microbiota is responsible for anti-atherosclerotic effects of Miltefosine. The study did not show if plasma levels of TMAO, an atherogenic molecule released by gut microbiota, are altered in Miltefosine-fed hyperlipidemic mice. We believe that mechanism of Miltefosine anti-atherosclerotic property involves multiple pathway including choline metabolism, cholesterol efflux, inflammasome activity, and gut microbiota.

STAR★Methods

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Chemicals,peptides, andrecombinantproteins
RPMI-1640	Millipore Sigma	R0883
FBS	Gibco	10082147
2-Mercaptoethanol	Millipore Sigma	M6250
12-myrisate 13 acetate	Sigma Aldrich	P8139
Penicillin streptomycin	Cell services	721
Ethanol	Fisher Scientific	BP28184
Bovine serum albumin	Sigma Aldrich	A9647-100G
Hematoxylin and Eosin	Vector Lab	H-3401
Xyelene	Fisher Scientific	EW-88043-30
Tween-20	Sigma Aldrich	P7949
Novex 4–20% Tris Glycine Gels	Invitrogen	XP04200BOX
LPS	Sigma Aldrich	L2880
TPER	Thermo scientific	78510
Mouse IL beta ELISA Kit	R and D systems	MLB00C
Pierce BCA	Thermo fisher	23225
Adenosine Triphosphate	Sigma Aldrich	A2383
KRH buffer	Thermofisher	AAJ67795-K2
Total Cholesterol reagent	Stan Bio	1010-225
3HCholineChloride	PerkinEmler Life Sciences	NET1090001MC
Miltefosine	Avanti	850337P
Cryostat	Leica Biosystems	CM1860
HDL precipitation reagent	Stan Bio	0599020
Westerntype diet	Envigo	TD 88137
3H Cholesterol	PerkinElmer Life Sciences	NET139001
Antibodies
Cleaved Il-1beta	Cell signaling technology	63124
Cleaved Gasdermin D	Cell signaling technology	50928
beta Actin	Cell signaling technology	8457
VCAM-1	Santa Cruz	SC-13160
CTL-1(SLC44A1)	Invitrogen	PA5-102034
Horseradish peroxidase-Conjugated goat anti rabbit secondary	Bio Rad	1662408EDU
Criticalcommercialassays
DNAeasy Powersoil ProKit	Qiagen Inc Valencia, CA	47016
Microscope
Inverted microscope	Leica Microsystems, GmbH,Wetzlar, Germany	Leica DMI6000
Hamamatsu Orca Flash4 camera	Hamamatsu Photonics, Shizuoka,Japan	Leica DMI6000
Software andalgorithms
Image ProPlus	Media Cybernetics, Inc,Rockville, Maryland USA	https://www.mediacy.com/imagepro
GraphPad Prism	GraphPad	https://www.graphpad.com/

Resource availability

Lead contact

Information regarding this work and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Dr. Kailash Gulshan (k.gulshan@csuohio.edu).

Materials availability

This study did not generate unique reagents.

Method details

Cell culture

The bone marrow derived macrophages (BMDMs) were cultured in BMDM growth media (DMEM with 7.6% fetal bovine serum, 15% L-cell conditioned media, and 0.76% penicillin/streptomycin mixture). Miltefosine (850337P) was obtained from Avanti Polar. Radioactive ³H-cholesterol (NET139001) and ³H-choline chloride (NET109001) were obtained from Perkin-Elmer Life Sciences. The antibodies used were cleaved IL-1β (Cell Signaling #63124), cleaved GsdmD (Cell Signaling #50928), and β−actin (Cell Signaling #8457). VCAM-1 antibody was purchased from Santa Cruz Biotechnology and western blot reagents were from Pierce. The PBS-washed cell pellet was lysed in MPER lysis buffer containing protease inhibitors and PMSF. After discarding the nuclear pellet, the protein concentration was determined using the BCA protein assay (Pierce). 10–50 μg of cell protein samples were resolved on Novex 4–20% Tris- Glycine Gels (Invitrogen) and transferred onto polyvinylidene fluoride membranes (Invitrogen). Blots were incubated sequentially with either 1:1000 secondary antibodies or HRP-conjugated beta actin antibody (Sigma). The signal was detected with an enhanced chemiluminescent substrate (Pierce). The membranes were imaged using iBright™ CL750 Imaging System (Life Technologies, A44116) and bands were quantified via densitometry using the iBright Analysis Software.

Mice and diets

All animal experiments were performed in accordance with protocols approved by the Cleveland State University and Cleveland Clinic Institutional Animal Care and Use Committee. The C57BL6J mice were purchased from the Jackson laboratory and the apoE^−/−C57BL6J mice were bred in-house. The animals were maintained in a temperature-controlled facility with a 12-h light/dark cycle with free access to food and water. The standard chow diet (SD, 20% kcal protein, 70% kcal carbohydrate and 10% kcal fat, Harlan Teklad) was used for regular maintenance and breeding. For generating hyperlipidemia for atherosclerosis studies, mice were fed an atherogenic Westerntype diet (WTD) (Envigo, 0.2% cholesterol with 42% adjusted calories from fat, TD.88137). Miltefosine was milled in the standard chow or Westerntype diet by Envigo.

Isolation of bone marrow derived macrophages

The WT C57BL6J mice were maintained on chow diet and sacrificed at 16 weeks of age. WT C57BL6J mice were maintained on chow diet and sacrificed at 16 weeks of age. Femurs were collected to isolate and culture bone marrow macrophages using conditioned L-cell media. Mice were euthanized by CO₂ inhalation and femoral bones were removed. The marrow was flushed out of the bones into a 50 mL sterile tube using a 10 mL syringe with a 26-gauge needle filled with sterile DMEM. Cells were centrifuged for 5minat 1,800rpmat 4°C, followed by two washes with sterile PBS. The cells were resuspended in sterile-filtered BMDM growth media (DMEM with 7.6% fetal bovine serum, 15% L-cell conditioned media, and 0.76% penicillin/streptomycin mixture) and plated in 10 cm culture dishes and incubated at 37°C for 14 days. Cell media was replaced every 2–3 days for 2 weeks. The cells were routinely visualized under microscope for proliferation and differentiation into confluent BMDMs.

Radioactive choline uptake assay

Choline uptake was determined by measuring intracellular ³H-choline chloride (PerkinElmer Life Sciences) in mouse BMDMs over time. Choline uptake was determined by measuring intracellular ³H-choline chloride (Perkin-Elmer Life Sciences) over time. BMDMs were seeded into 12-well plates and treated with or without 7.5 μM Miltefosine for 16h. Treated and control cells were washed twice with sterile PBS, followed by incubation with Krebs-Ringer-HEPES (KRH) buffer (130 mM NaCl, 1.3 mM KCl, 2.2 mM CaCl₂, 1.2 mM MgSO₄, 1.2 mM KH₂PO₄, 10 mM HEPES, pH 7.4 and 10 mM glucose) for 1h. The cells were washed again, followed by the addition of KRH buffer containing 2.5 μCi/mL ³H-choline chloride and incubation at 37°C for 20 minutes. Cells were washed twice with ice-cold KRH buffer and were lysed with 0.1M NaOH and the radioactivity was determined by liquid scintillation counting. The uptake was plotted as dpm/mg of protein.

In vivo NLRP3 inflammasome activity

In vivo NLRP3 inflammasome assembly was induced by LPS and ATP injections in mice, as described earlier.²⁷ Mice were fed with either chow, chow+ Miltefosine, WTD or WTD + Miltefosine for 3 weeks. Mice were I.P. injected with either 5 μg LPS, or sterile PBS. After 4 h of LPS or PBS injection, the NLRP3 inflammasome assembly was induced by I.P. injection of ATP (0.5 mL of 30 mM, pH 7.0). The mice were euthanized after 30 min of ATP injection and peritoneal cavity was lavaged with 5 mL PBS. Approximately 3.5 mL peritoneal lavage fluid was recovered from each mouse and centrifuged at 15 K rpm for 10minat room temperature. The supernatant was subjected to IL-1β ELISA, using mouse IL-1βQuantikine ELISA kit (MLB00C, R&D systems) and following manufacturer’s instructions.

Mice RCT assay

RCT assays were performed as described earlier.²⁷ WT C57BL6J mice were euthanized by CO2 inhalation and femoral bones were removed. The marrow was flushed out of the bones into a 50 mL sterile tube using a 10 mL syringe with a 26-gauge needle filled with sterile DMEM. Cells were centrifuged for 5minat 1,800rpmat 4°C, followed by two washes with sterile PBS. The cells were cultured for 11 to 14 days in DMEM supplemented with 20% L-cell conditioned medium, 10% fetal bovine serum and 1% penicillin/streptomycin. To generate foam cells, the tritium labeled ³H-cholesterol (2 μCi/mL; Perkin Elmer, Norwalk, USA) and 100 μg/mL acetylated LDL were mixed and incubated at 37°C. This cholesterol mixture was combined with DMEM containing 20% L-cell conditioned medium and BMDMs were incubated with cholesterol-labeled media for 48 h to generate foam cells. The cholesterol loaded foam cells were washed twice with DMEM prior to harvesting for in vivo injection, and ∼2 million ³H-cholesterol dpm in a volume of 0.25 mL were transplanted subcutaneously on the upper back of recipient mice. At 24, 48, and 72 h post transplantation, plasma and feces samples were collected. Plasma radioactivity was determined, and total plasma dpm was calculated by estimating blood volume to be equal to 7% of the body weight and plasma to be 55% of the blood volume. RCT to the plasma was calculated as the % (dpm appearing in plasma/total dpm injected) of injected radioactivity. Collected feces were allowed to dry overnight at 55°C and then weighed. Feces were then hydrated in 10 mL of 50% ethanol solution followed by homogenization then an internal recovery standard of 10,000 dpm of ¹⁴C-cholesterol (Perkin-Elmer, Norwalk, USA) was added to each sample. The radioactivity was quantified as described in detail above. Upon sacrifice at 72 h, the liver was removed and weighed. A piece of liver, ∼0.2 g, was isolated, weighed, suspended in PBS, homogenized, and a known amount of ¹⁴C-cholesterol radioactivity was added as a recovery standard. The radioactivity in an aliquot of 0.3 mL of the liver homogenate was measured by liquid scintillation counting The ¹⁴C-cholesterol dpm was used to back calculate the [³H] recovery for the entire liver homogenate, which was further used to calculate the total amount of radioactivity in the liver. In addition to ¹⁴C-cholesterol internal standard, all RCT data was standardized to mouse body or fecal weight each harvest time point.

Cholesterol measurements

Total cholesterol was measured by using Stan Bio Total cholesterol reagent (#1010-225) and plasma HDL-C was determined using ultracentrifugation and precipitation with HDL precipitation reagent (StanBio # 0599020), following manufacturer’s instructions.

Atherosclerotic lesion and Mac-3 staining and quantification

Mice were sacrificed by CO₂ inhalation and weighed at 21 weeks of age. Whole blood was collected from the retroorbital plexus into a heparinized glass capillary, mixed with EDTA and spun in a microfuge to obtain plasma. The circulatory system was perfused with 10 mL PBS and the heart was excised and preserved in 10% phosphate buffered formalin. Hearts were sectioned using Leica cryostat (CM1860) and sections containing aortic sinus were embedded in OCT medium. Quantitative assessment of atherosclerotic plaque area in the aortic root was performed and lesion areas were quantified as the mean value in multiple sections at 80 μm intervals using Image Pro software (Media Cybernetics). For Mac-3 staining, antigen retrieval was performed using the Antigen-unmasking solution (Vector Labs #H-3301). Samples were washed with water, quenched for endogenous peroxidase activity using Bloxall (Vector Labs #SP-6000), then blocked with 5% BSA in PBST overnight in a humidified chamber at 4°C. Sections were then probed with CD107b/Mac3(BD biosciences #550292) antibodies at 1:200 dilution for 1h in animal free blocker and diluent (Vector Labs #SP-5035). Sections were washed with 0.05% Tween-20 in PBS and subjected to Vector lab VECTASTAIN Elite ABC-HRP Kit (#PK-6104) protocol and followed by development using the Vector NovaRED kit (Vector Labs #SK-4800), both according to manufacturer instructions. Tissue sections were then counterstained using hematoxylin (Vector Labs #H-3401) according to manufacturer instructions, then mounted and imaged at 4x under bright-field microscopy.

16S rDNA sequencing and analysis

The fresh feces samples were collected from mice using sterile forceps and spatula, and the microbial DNA was extracted using Qiagen DNeasy PowerSoil ProKit (47016), following manufacturer’s instructions. The DNA samples were subjected to 16S rRNA gene amplification and sequencing using methods explained earlier.⁵⁹ Raw 16S amplicon sequence and metadata, were demultiplexed using split_libraries_fastq.py script implemented in QIIME1.9.1.⁶⁰ Demultiplexed fastq file was split into sample specific fastq files using split_sequence_file_on_sample_ids.py script from Qiime1.9.1.⁶⁰ Individual fastq files without non-biological nucleotides were processed using Divisive Amplicon Denoising Algorithm (DADA) pipeline.⁶¹

Quantification and statistical analysis

Data are shown as mean ± SD. Comparisons of 2 groups were performed by a 2-tailed t test, and comparisons of 3 or more groups were performed by ANOVA with Bonferroni posttest. All statistics were performed using Prism software (GraphPad). For microbiome statistical analysis, the output of the dada2 pipeline (feature table of amplicon sequence variants (an ASV table)) was processed for alpha and beta diversity analysis using phyloseq,⁶² and microbiomeSeq (http://www.github.com/umerijaz/microbiomeSeq) packages in R. Alpha diversity estimates were measured within group categories using estimate_richness function of the phyloseq package.⁶² Multidimensional scaling (MDS, also known as principal coordinate analysis; PCoA) was performed using Bray-Curtis dissimilarity matrix⁶³ between groups and visualized by using ggplot2 package.⁶⁴ We assessed the statistical significance (p< 0.05) throughout and whenever necessary, we adjusted p-values for multiple comparisons according to the Benjamini and Hochberg method to control False Discovery Rate⁶⁵while performing multiple testing on taxa abundance according to sample categories. We performed an analysis of variance (ANOVA) among sample categories while measuring the of α-diversity measures using plot_anova_diversity function in microbiomeSeq package (http://www.github.com/umerijaz/microbiomeSeq). Permutational multivariate analysis of variance (PERMANOVA) with 999 permutations was performed on all principal coordinates obtained during PCoA with the ordination function of the microbiomeSeq package. Linear regression (parametric test), and Wilcoxon (Non-parametric) test were performed on ASVs abundances against meta-data variables levels using their base functions in R.⁶⁶

Published: March 20, 2023

Data and code availability

This paper does not report original code.