ALKBH5-mediated m6A modification of IL-11 drives macrophage-to-myofibroblast transition and pathological cardiac fibrosis in mice

Cardiac macrophage contributes to the development of cardiac fibrosis, but factors that regulate cardiac macrophages transition and activation during this process remains elusive. Here we show, by single-cell transcriptomics, lineage tracing and parabiosis, that cardiac macrophages from circulating monocytes preferentially commit to macrophage-to-myofibroblast transition (MMT) under angiotensin II (Ang II)-induced hypertension, with accompanying increased expression of the RNA N6-methyladenosine demethylases, ALKBH5. Meanwhile, macrophage-specific knockout of ALKBH5 inhibits Ang II-induced MMT, and subsequently ameliorates cardiac fibrosis and dysfunction. Mechanistically, RNA immunoprecipitation sequencing identifies interlukin-11 (IL-11) mRNA as a target for ALKBH5-mediated m6A demethylation, leading to increased IL-11 mRNA stability and protein levels. By contrast, overexpression of IL11 in circulating macrophages reverses the phenotype in ALKBH5-deficient mice and macrophage. Lastly, targeted delivery of ALKBH5 or IL-11 receptor α (IL11RA1) siRNA to monocytes/macrophages attenuates MMT and cardiac fibrosis under hypertensive stress. Our results thus suggest that the ALKBH5/IL-11/IL11RA1/MMT axis alters cardiac macrophage and contributes to hypertensive cardiac fibrosis and dysfunction in mice, and thereby identify potential targets for cardiac fibrosis therapy in patients.

Smooth muscle cells (SMCs) are reported to upregulate macrophage-associated gene programs during inflammatory responses 17 .To exclude the potential hypothesis that smooth muscle cells upregulated macrophage-associated gene programs under Ang II treatement, Myh11Cre ERT2 ; Rosa26 Td mice were utilized to test whether SM cells upregulated macrophage-associated gene programs.FACS and immunostaining revealed Td+SMA+ cells, but not Td+CD11b cells were increased in hypertensive hearts (Supplementary Fig. 3E-G).Although we observed an increased CD11b+ cells in hypertensive hearts, these cells were not derived from Td+ cells.We then performed parabiosis experiments by conjoining non-fluorescent C57BL6/J mice with Myh11Cre ERT2 ; Rosa26 Td mice and observed few Td+ cells in hearts from both PBS and Ang II treated C57BL6/J mice assessed by FACS analysis (Supplementary Fig. 3H), indicating that Myh11+ SM cells in the heart do not contribute to macrophage-like cells in this pathological process.Taken together, these indicate that hypertension induces circulating monocytederived macrophages, but not cardiac resident macrophages, to transition to myofibroblasts.

ALKBH5 upregulation is involved in hypertension-induced cardiac MMT
To determine the molecular mechanism that is involved in the regulation of MMT and pathological cardiac fibrosis in the heart, we reanalyzed the scRNA sequencing profiles.Bioinformatic analysis showed that RNA N6-methyladenosine demethylase ALKBH5 was widely expressed in all clusters in the heart with more variation in macrophage cluster (Supplementary Fig. 4A).We next analyzed the ALKBH5 expression in the macrophage subclusters, and observed a higher ALKBH5 expression in myofibroblasts compared with macrophage populations shown by dot plots (Fig. 2A), heat maps (Fig. 2B).Pseudotime analysis further suggested that ALKBH5 expression level was in parallel with the expression of Acta2 (Fig. 2C).The increased ALKBH5 expression during the process of MMT was also shown by Vilon image (Fig. 2D).
N6-methyladenosine (m6A) RNA modification, which are mediated by readers (Ythdc1, Ythdc2, Ythdf1, Ythdf2, Ythdf3 and Hnrnpa2b), writers (Mettl3, Mettl14 and Wtap) and eraser proteins (Fto and ALKBH5), plays various roles in mRNA instability, translation, splicing and phase separation 18 .m6A modification was reported to control cardiac homeostasis and hypertrophy 19 .To define whether m6A modification participates in macrophage-myofibroblast transition under hypertensive stress, we performed m6A dot blot analysis and found that m6A methylation was significantly decreased in macrophages after Ang II treatment (Fig. 2E).We also analyzed the other m6A modification proteins in the scRNA sequencing data, and found that except for ALKBH5, the others had no significant correlation to the MMT process (Supplementary Fig. 4B).We also assessed the mRNA level of m6A modification proteins in sorted Td+ cells from Cx3cr1Cre;Rosa26 Td mice by RT-qPCR, and found that ALKBH5 notably increased after AngII-infusion (Supplementary Fig. 4C).We also observed an upregulated ALKBH5 protein level in vitro cultured Td+ macrophages following Ang II treatment (Fig. 2F).We therefore hypothesized that ALKBH5 might contribute to hypertension-induced MMT, and related cardiac fibrosis and dysfunction.

Specific deletion of ALKBH5 in Cx3cr1 lineage improves Ang IIor pressure overload-induced MMT and cardiac fibrosis
We next constructed Cx3cr1cre; Rosa26 Td ; ALKBH5 flox/flox (ALKBH5 macKO-Td , KO) mice to investigate the role of macrophage ALKBH5 in MMT and pathological cardiac fibrosis.The expression of ALKBH5 was significantly decreased in peritoneal macrophages of ALKBH5 macKO-Td mice compared with control mice (Cx3cr1cre; Rosa26 Td ; ALKBH5 wt/wt , WT) (Supplementary Fig. 5A, RIP-sequencing Dataset).We firstly determined the effect of ALKBH5 deletion on MMT in vivo and in vitro.ALKBH5 deletion did not change Td+ cells in hearts under steady state or following challenge (Fig. 2G).ALKBH5 macKO-Td mice displayed a significant decrease in SMA+Td+ cells in the hypertensive hearts (Fig. 2G), as well as reduction of SMA and ECM gene collagen I and III mRNA level in sorted Td+ cells (Fig. 2I).Consistent with in vivo findings, we found that Ang II increased but ALKBH5 knockout reversed the increased SMA+Td+ myofibroblast-like cell formation and expression of SMA and ECM genes collagen I and III in cultured macrophages after Ang II treatment (Fig. 2I, J).
We then determined the effect of macrophage ALKBH5 deletion on cardiac function and fibrosis.We found that ALKBH5 deletion had no significant effect on Ang II-induced blood pressure elevation (Fig. 3A).Ang II-infusion for 14 days displayed no significant ejection fraction (EF) decrease, but E/e' ratio increase, indicating that hypertension induced cardiac diastolic dysfunction.exhibited a significant decrease of heart weight/body weight (Fig. 3B), LV mass (Fig. 3C) and E/e' ratio compared with control mice (Fig. 3F).Cardiac fibrosis evaluated by Masson's Trichrome-staining (Fig. 3F), as well as the expression of SMA and ECM genes collagen I and III (Fig. 3G), were also decreased in the heart of ALKBH5 macKO-Td mice compared with control mice.ALKBH5 macKO-Td mice also exhibited smaller cardiomyocytes shown by wheat germ agglutinin staining (Fig. 3H), as well as attenuation of cardiac fetal gene atrial natriuretic peptide (ANP) and brain natriuretic peptide (BNP) reactivation (Fig. 3I).
These data suggest that loss of ALKBH5 in macrophage inhibits MMT and improves pathological cardiac fibrosis and hypertrophy and further attenuates cardiac diastolic dysfunction.Macrophage proliferation and cardiac fibroblast activation play pivotal roles in pathological cardiac remodeling 20 .Therefore, we evaluated the effects of ALKBH5 on Cx3cr1-derived cell proliferation, and found that the proliferation (Ki67+ cells) of Cx3cr1-derived Td positive cells in the heart of ALKBH5 macKO mice was decreased (Supplementary Fig. 5B).Consistently, cell proliferation of cultured   macrophages was decreased after ALKBH5 knockout with Ang IIstimulation (Supplementary Fig. 5C).We also performed macrophage and cardiac fibroblast co-culture assay (Supplementary Fig. 5D), and found that ALKBH5 deletion in macrophages significantly inhibited SMA and ECM genes collagen I and III expression in Ang II-treated cardiac fibroblast (Supplementary Fig. 5E).
To further validate the roles of ALKBH5 in pathological cardiac remodeling, we analyzed pressure overload-induced cardiac dysfunction by utilizing transverse aortic constriction (TAC) mouse model, and found a greater decrease of E/e' ratio and increase of ejection fraction in ALKBH5 macKO-Td mice after TAC 28 days (Supplementary Fig. 6A).Further, ALKBH5 macKO-Td mice displayed decreased SMA+Td+ cells and cardiac fibrosis shown by Masson's Trichromestaining and the expression of SMA and ECM genes collagen I and III (Supplementary Fig. 6B-D).These data indicate that ALKBH5 deletion in Cx3cr1 lineage improves cardiac diastolic function under hypertensive stress.
ALKBH5 promotes MMT and cardiac fibrosis via regulating m6A modification on IL-11 mRNA in Cx3cr1-derived macrophages To investigate the mechanism by which ALKBH5 regulates MMT and cardiac fibrosis under hypertension, we performed ALKBH5 RNA immunoprecipitation-sequencing (RIP-seq) in cultured macrophages and found several genes are the direct targets of ALKBH5 (Fig. 5A).The pathway enrichment analysis was performed to show the pathway regulated by ALKBH5 responsive to Ang II.Among the top 10 enriched pathways, we observed that Tgf-beita signaling pathway was associated with ALKBH5, which further indicated that macrophage ALKBH5 participated in the process of Ang II-induced macrophage-myofibroblast transition (Fig. 5B).Moreover, Ang II significantly increased the enrichment of ALKBH5 on IL-11 mRNA, the top abundant enriched target mRNA (Fig. 5A, Supplementary Data 1).Therefore, we hypothesized that IL-11 mRNA might be the m6A target directly modulated by ALKBH5.We next did m6A RNA immunoprecipitation-sequencing in ALKBH5 knockout and control macrophages and found that the GGAC motif was highly enriched within m 6 A locus in both control and ALKBH5 knockout macrophages (Fig. 5C).The m 6 A peaks were particularly abundant near start and stop codons (Fig. 5D).m 6 A peaks were enriched on IL-11 mRNA (Fig. 5E), which was confirmed by m6A RIP-qPCR using primer on peak1 and peak2 regions (Fig. 5F).We then assessed RNA decay modulated by m6A and found that IL-11 mRNA decreased faster in ALKBH5 deficient macrophages (Fig. 5G).These data suggest that ALKBH5 deficiency in macrophages increases m6A modification on IL-11 mRNA, which results in decreased IL-11 stability.
Previous scRNA-seq analysis of fibrotic hearts demonstrated that IL-11 and its receptor (IL11RA1) pathway is a crucial determinant for cardiovascular fibrosis 21 .Our scRNA sequencing data also showed that macrophage-derived myofibroblast cluster, but not macrophage, was highly enriched for IL11RA1 under hypertensive stress (Fig. 5H).Pseudotime analysis also showed that IL11RA1 expression level was in parallel with the process of MMT in cardiac macrophages (Fig. 5I).The expression of IL11 and IL11RA1 were increased in sorted Td+ cells from Cx3cr1Cre;Rosa26 Td following Ang II-infusion (Fig. 5J).These suggested that ALKBH5 promotes cardiac macrophage-myofibroblast transition and fibrosis via directly targeting IL-11 mRNA and subsequently activating IL-11/IL11RA1 pathway in cardiac macrophages.
To identify the role of IL-11 in ALKBH5-mediated cardiac dysfunction, we utilized bone-marrow transplantation (BMT) to specifically overexpress IL11 in circulating monocytes.FACS analysis indicated the success of BMT, with Td+ cells in the reconstituted hearts of both ALKBH5 macKO-Td -BM and Cx3cr1Cre Td -BM mice (Fig. 6A).ALKBH5 deficiency or IL11 overexpression did not influence cardiac Td+ cells infiltration (Fig. 6A).FACS analysis and immunostaining analysis showed that C57BL6/J mice transplanted with BM from ALKBH5 macKO -Td mice displayed decreased percentage of SMA + in Td + cells compared to Cx3cr1Cre Td mice after AngII-infusion, which were elevated after IL11 overexpression (Fig. 6A, B).Echocardiography data showed that IL11 overexpression (IL11 OE) damaged the cardiac diastolic function shown by the increase in E/e' while no change in ejection fraction in both ALKBH5 macKO-Td -BM and Cx3cr1Cre Td -BM mice (Fig. 6C, D).IL11 overexpression reversed the decreased cardiac fibrosis (Fig. 6E) and fibrosis related genes caused by ALKBH5 knockout in macrophages (Fig. 6F).
Exogenous IL-11 recombinant protein (rIL-11) was also used in vitro to confirm the role of ALKBH5-IL11 pathway in Ang II induced MMT and subsequently proliferation of myofibroblasts.We found that rIL-11 promoted Ang II-induced SMA, Col I and Col III expression and cell proliferation in both wild type (WT) and ALKBH5 knockout (KO) macrophages (Supplementary Fig. 9A-C).Consistently, Co-cultured macrophages with primary cardiac fibroblasts demonstrated that rIL-11 administration blocked ALKBH5 knockout-mediated inhibition of SMA and ECM gene expression and proliferation of cardiac fibroblasts (Supplementary Fig. 9D-F).To further investigate the role of IL11, we inhibited IL11 signaling in cultured macrophages using neutralizing antibody against IL11.The results showed that inhibition of IL11 signaling reduced SMA and Col1 expression and cell proliferation in WT macrophages, but had no effect on ALKBH5 knockout macrophages (Supplementary Fig. 9G, I).We also assessed the expression of ALKBH5, IL-11 and IL11RA1 in the cultured fibroblasts, and found that Ang II treatment had no significant effect on ALKBH5 and IL-11 expression, but increased IL11RA1 expression in cardiac fibroblasts (Supplementary Fig. 10).Overall, these in vivo and in vitro data inferred that ALKBH5/IL-11 pathway activation is mainly existed in cardiac macrophages.Then the secreted IL-11 induces MMT, as well as promotes cardiac fibroblast activation via directly binding to the receptor IL11RA1 in both macrophages and fibroblasts.

Nanoparticle monocyte/macrophage-target delivery of ALKBH5 or IL11RA1 siRNA attenuates Ang II-induced cardiac fibrosis and dysfunction
Lipid nanoparticle (LNP)-mRNA technology is widely used for targeting immune cells, especially for monocyte-derived macrophages in various animal models 22 .We therefore generated the DiR-modified LNPs C12-200 23,24 that is successfully used for targeting circulating monocytes (Fig. 7A).Size distribution and zeta potential of LNPs were measured by dynamic light scattering and cryo-transmission electron micrographs (Fig. 7B-D).We then delivered ALKBH5 or IL11RA1 siRNA encapsulated in monocyte/macrophage-target DiR-modified LNPs C12-200 to alleviate the cardiac fibrosis and dysfunction under hypertensive stress (Fig. 7E).Fluorescence imaging combined with microCT showed that nanoparticles was successfully distributed in the heart following Ang II-infusion (Fig. 7F).ALKBH5 siRNA-encapsulated LNPs improved Ang II-induced cardiac dysfunction shown by decreased E/e' ratio (Fig. 7G, H).Masson trichrome staining showed that LNPmediated ALKBH5 silencing significantly improved cardiac fibrosis caused by Ang II-infusion (Fig. 7I).ALKBH5 silencing decreased SMA positive cells (Fig. 7J), and also expression of SMA and collagen types I and III in the Ang II-infused hearts (Fig. 7K).Similarly, IL11RA1 siRNAencapsulated LNPs treatment improved cardiac function and attenuated cardiac fibrosis (Supplementary Fig. 11A-D).These suggest that targeting macrophage-ALKBH5-IL-11/IL11RA1 pathway may provide a novel therapy for pathological cardiac remodeling.

Discussion
Macrophages play crucial roles in pathological cardiac fibrosis.However, the factors that regulate the phenotype or transition and activation of cardiac macrophages during cardiac fibrosis is not clear.Herein, we provide evidence that circulating monocyte-derived cardiac macrophages transition to myofibroblasts, and contribute to cardiac fibrosis under hypertension.ALKBH5 promotes MMT and activates cardiac fibroblasts through upregulating IL-11 in Cx3cr1 precursor-derived cells.Macrophage-targeted lipid nanoparticles, containing ALKBH5 or IL11RA1 siRNA, attenuate the transition of macrophages to myofibroblasts and improve hypertensive cardiac fibrosis (Fig. 8).Taken together, the present study provides direct evidence that transition of macrophages to myofibroblasts is involved in the process of hypertension-induced cardiac fibrosis, which is largely dependent on the ALKBH5-decreased m6A-demethylation on IL-11 in cardiac macrophages.
Transition of macrophages has been reported in different pathological processes, probably due to the plasticity of the myeloid lineage, especially monocytes/macrophages.Increasing evidence suggests that the historically defined fibroblast is actually not a cell type, but a general name for heterogeneous cells.A number of cell types, including BM-derived cells and endothelial cells, are suggested to contribute to pathological fibrosis by converting to a myofibroblast phenotype via utilizing BM chimeras, parabiosis, and lineage tracing experimental approaches 25 .A majority of cardiac fibroblasts is derived from epicardial and endothelial cells that express Tcf21, Wt1 or Tbx18 during embryonic development [26][27][28] .In addition to this, bone marrow-derived myeloid cells, fibrocytes, and small infiltrating immune cells have been suggested to generate myofibroblasts in the diseased mouse heart [29][30][31] .There are two major populations of macrophages within the healthy and diseased heart, a population of CCR2 + circulating monocyte-derived macrophages and a smaller population of resident cardiac Lyve1 + macrophages, that could be replaced by infiltrating monocyte-derived CCR2 + macrophages 32,33 .In unstressed hearts of mice, the embryonic Cx3cr1 precursor-derived resident cardiac macrophages facilitate tissue homeostasis with expressing Lyve1 10,34 .Cx3cr1 precursors also labels circulating monocytes and that infiltrates to diseased heart to generate CCR2 + macrophages, that causes an acute sterile immune response during cardiac cell therapy remodeling 12,13,32,35 .Our genetic fate mapping supported that circulating monocyte-derived cardiac macrophages transition to myofibroblasts, which contributes to cardiac fibrosis under hypertension.N6-methyladenosine (m6A), the most abundant modification on mRNA, is catalyzed by methyltransferase complex and demethylases.m6A modification controls various aspects of immunity, including immune recognition, activation of innate and adaptive immune responses, and immune cell development and fate decisions 36 .ALKBH5-mediated m6A-demethylation has been reported to be involved in the regulation of biological processes, including immune response, ROS production and viral infection [37][38][39] .Ablation of ALKBH5 in CD4 + T cells increases m6A modification on IFN-γ and CXCL2 mRNA and reduces corresponding protein expression, thereby attenuating T cell-mediated inflammation and immune response 40 .In the cardiovascular system, ALKBH5 plays different roles in different cell types.ALKBH5 promotes cardiomyocyte proliferation after myocardial infarction while attenuating endothelial cell-mediated post-ischemic   gene for the phenotype in this mode.However, we could not exclude the other ALKBH5 downstream candidate genes that may be involved in the regulation of this phenotype in the macrophages.This needs further investigation in the future.Lipid nanoparticle-mRNA technology underlies recent successes in COVID-19 vaccine development and has stimulated broad efforts to extend this therapeutic platform to address numerous pathologies.A recent study reported an immunotherapy strategy to generate transient CAR T cells that can recognize the fibrotic cells in the heart by injecting CD5-targeted LNPs 42 .In fact, cardiac monocytes and macrophage abundance are much more than the other immune cells in the cardiac pathophysiological status.Therefore, we utilized optimized LNPs to target monocytes/macrophages, and ALKBH5 or IL11RA1silencing siRNA is efficiently delivered to cardiac macrophages.This LNP-siRNA technology successfully inhibits the transition of macrophages to myofibroblasts, and improves pathological cardiac fibrosis.Taken together, our finding of ALKBH5-IL-11-IL11RA1 pathway activation-mediated transition of macrophage to myofibroblasts represents a novel mechanism that contributes to pathological cardiac fibrosis.Notwithstanding the much more advanced complexity of human pathological cardiac fibrosis in comparison with the mouse model, our novel findings suggest an attractive possibility that targeting ALKBH5-IL-11-IL11RA1 pathway in cardiac macrophages via LNP-siRNA technology might serve as a potential therapeutic tool for the prevention of cardiac fibrosis in heart diseases.
Cx3cr1Cre mice 43 were obtained from the Jackson Laboratory (Stock#: 025524).Lyve1Cre mice 44 were from the Jackson Laboratory (Stock#: 012601).Lyz2Cre 45 (Stock#: 004781), Cx3cr1Cre ERT2 46 (Stock#: 021160) and Myh11Cre ERT2 47 (Stock#: 019079) mice were purchased from the Jackson Laboratory.CCR2GFP mice were from the Jackson Laboratory (Stock#: 027619).CCR2 knockout mice were obtained by mating CCR2GFP/+ mice with each other.All mice were in C57BL/6J background and male mice aged 8-12 weeks were included in this study, and housed at 25 °C, 12-h light/dark in a specific pathogen-free environment at the University Laboratory Animal Services Centre of Fudan University.

SBP measurements
Systolic blood pressure (SBP) was measured by the noninvasive tailcuff technique (Visitech Systems, Inc).Blood pressure were averaged from at least 10 consecutive measurements for each mouse.

Transverse aortic constriction left ventricle pressure overload model
Briefly, under sterile conditions, a blunted 27-gauge needle placed parallel to the transverse aorta was tied by 7-0 silk suture.Subsequently, the needle was gently removed.The sham-operated mice were conducted with identical surgery except for tying the transverse aorta.Echocardiography was performed to evaluate the cardiac function four weeks later.
Formulation of siRNA lipid nanoparticles (LNPs) for in vivo gene expression silencing C12-200-siRNA liposomes comprised C12-200, DSPC, cholesterol, and mPEG2000-DMG.Briefly, the molar ratio of liposomes was C12-200/ DSPC/cholesterol/mPEG2000-DMG at 50:10:38.5:1.5.The lipids were dissolved in 90% ethanol and the siRNA was solubilized in 10 mM citrate, pH3 buffer at a concentration of 0.4 mg/mL.Mixed lipids were slowly added to siRNA citrate buffer and incubated at room temperature for 30 min.Ethanol removal and buffer exchange of siRNAcontaining liposomes was achieved by ultrafiltration against PBS using a 100,000 MWCO membrane.Finally, the formulation was filtered through a 0.2 mm sterile filter.Particle size was determined using a Malvern Zetasizer NanoZS (Malvern).The morphology of liposome was observed via cryo-TEM (FEI Tecnai G2 F20 electron microscope (200 kV)).The resulting particles had a mean particle diameter of 100 nm with approximately 90% siRNA entrapment efficiency.The liposomes were stained with DiR by directly adding DiR (1,1-dioctadecyl-3,3,3,3-tetramethylindotricarbocyanine iodide, 1 mol%) to the lipid mixture in ethanol during preparation.
In vivo loss-of-function study by using siRNA-lipid nanoparticles (LNP) complexes Briefly, siRNA lipid nanoparticles were diluted in PBS at a dose of 0.2 mg/ml.Formulations were administered intravenously via tail vein injection (2 mg/kg) for 14 days at an interval of 2 days.The distribution of siRNA lipid nanoparticles was captured with IVIS system.The efficiency of siRNA was determined by western blot.

Dynamic FMT-CT imaging of siRNA biodistribution
Mice were placed onto the warmed stage inside the camera box and received continuous exposure to 2.5% isoflurane to sustain sedation during imaging.The light emitted from DiR dyes were detected, integrated, digitized, and displayed by the IVIS camera system.Regions of interest from displayed images were identified and quantified as total photon counts or photons/s using Living Image® software 4.0 (Caliper, Alameda, CA.).

Echocardiography
Mouse echocardiography was performed using the Vevo2100 High-Resolution Imaging System.Two-dimensional long-axis and short-axis LV imaging were obtained for analysis.M-mode tracings were recorded through septum and posterior LV walls to measure LV dimension and wall thickness.Ejection fraction was calculated by measurement with LV end-diastolic diameter and end-systolic diameter.Pulse-wave Doppler and tissue Doppler imaging of apical 4-chamber view were recorded to measure and analyze E/e' (the ratio of peak early transmitral flow velocity to the peak early diastolic mitral annular velocity).

Bone marrow transplantation (BMT)
Briefly, donor bone marrow (BM) was isolated from ALKBH5 macKO-Td and Cx3cr1Cre Td mice, and then infected with empty vector or IL11 overexpression lentivirus (10 8 Pfu).Recipient C57BL/6 mice were subjected to 10-Gy lethal-dose irradiation from cobalt 60 source followed by tail vein injection of 5 × 10 6 congenic donor BM with or without IL11 overexpression.The chimeras were recovered for 2 weeks to reconstitute bone marrow, and then implanted osmotic mini-pumps (Alzet) containing either saline or angiotensin II (Meilunbio, MB1677-2) under back for 2 weeks.The recipient mice received autoclaved water containing 32 × 104 U/L Gentamicin Sulfate (Biosharp, Cat#: BS143) and 250 U/L Erythromycin (Aladdin, Cat#: E105345) 1 week before and until euthanasia by carbon dioxide (CO 2 ) inhalation for analysis after the BMT.

Histology
The mice euthanasia was conducted using carbon dioxide (CO 2 ) inhalation, and the hearts were rinsed in PBS and fixed in 4% paraformaldehyde (PFA) for overnight.The hearts were then dehydrated and paraffin embedded, and sectioned at 4-μm interval.The sections were stained with Masson's Trichrome (Servicebio; Cat# G1006) for detection of cardiac fibrosis and Alexa FluorTM 488 conjugated wheat germ agglutinin (WGA) (W11261, Invitrogen) for measurement of cardiomyocyte size in vivo after deparaffinization and rehydration.Slides were imaged using the Leica DM IL inverted contrasting microscope and percentage heart fibrosis and cardiomyocyte size were quantified with Image J software.
The above-stained cells were then washed three times with 1xPBS before flow cytometry analysis.Data were acquired on NovoCyte 3000 flow cytometer (Agilent Technologies) and analyzed with NovoExpress software version 1.5.0.

Peritoneal macrophage isolation
Briefly, 1 mL of 3% brewer thioglycollate broth was injected into the peritoneal cavity.After three days of injection, the macrophages were collected by injecting and aspirating 10 mL PBS from the peritoneum.The cells were replaced with fresh RPMI 1640 medium supplemented with 10% fetal bovine serum and cultured for later experiments.

Primary cardiac fibroblast isolation and culture
The hearts were dissected and cut into small pieces and digested with 1 mg/ml Collagenase II in DMEM medium at 37 °C for 1 hour.Cells were passed through a 70 µm cell strainer (BD Falcon) and centrifuged.The cell pellet was re-suspended and cultured in DMEM containing 10% FBS.

RNA isolation and qPCR
Total RNA was isolated with an RNA Purification Kit at room temperature according to the manufacturer's instructions.For analysis of mRNA expression, 1 ug RNA was reverse-transcribed into cDNA with the cDNA synthesis kit (EZBioscience, United States).Quantitative real-time PCR analysis was then executed with SYBR green qPCR master mix on the Biorad PCR system (Thermo Fisher Scientific).GAPDH was taken as endogenous control.Each sample was run at least in triplicate.The primers used for qPCR analysis are listed in Supplementary Table 1.

RNA immunoprecipitation-sequencing
For immunoprecipitation-sequencing, total RNA was isolated from peritoneal macrophage.Briefly, macrophages with and without Ang II treatment were re-suspended and lysed in RIP lysis buffer, and the lysates were used as input.Then, lysates were incubated with ALKBH5 antibody in RIP immunoprecipitation buffer at 4 °C for 4 hours.Normal Rabbit IgG (Proteintech) was used as a negative control.The lysates were immunoprecipitated by Protein A/G beads (Repligen) at 4 °C for an additional 2 hours.The magnetic bead-bound complexes were centrifuged and washed for 5 times.The precipitated RNA samples were extracted and purified with phenol, chloroform and isoamyl alcohol, and the immunoprecipitated RNA was used to construct RNA libraries.Then, library sequencing was performed by illumine HiSeq 4000 sequencer with 150 bp paired-end reads.Data analysis was performed according to the published procedure.

Methylated RNA immunoprecipitation sequencing and MeRIP-qPCR
Peritoneal macrophages were isolated from ALKBH5 knockout or control littermate mice, and total RNA was isolated with RNA Purification Kit (EZBioscience, B0004D, China) as above described.RNA samples were chemically fragmented into,100-nucleotide-long fragments by 5 min incubation at 94 °C in fragmentation buffer (10 mM ZnCl2, 10 mM Tris-HCl, pH 7).The fragmentation reaction was stopped with 0.05 M EDTA, followed by standard ethanol precipitation.Then the fragmented RNA (400 mg mRNA or 2.5 mg total RNA) was incubated for 2 hours at 4 °C with 5 mg of anti-m6A antibody (Abcam; Cat# ab151230) in IPP buffer (150 mM NaCl, 0.1% NP-40, 10 mM TrisHCl, pH 7.4) supplemented with RNase inhibitors.The mixture was then immunoprecipitated by Protein A/G beads (Repligen) at 4 °C for an additional 2 hours.After extensive washing, bound RNA was eluted from the beads with 0.5 mg/ml N6-methyladenosine (Sigma-Aldrich) in IPP buffer, and ethanol precipitated.RNA was resuspended in H 2 O and used for library generation with mRNA sequencing kit (Illumina).m6A enrichment was determined by qPCR analysis with primers in the Supplemental Table 1.

m6A dot blot assay
Total RNA was isolated from the sorted Td+ cardiac macrophages with RNA Purification Kit (EZBioscience, B0004D, China).The serially diluted mRNA was denatured at 95 °C to disrupt secondary structures for 3 min, and then chilled on ice immediately.mRNA was dropped and crosslinked onto the Hybond-N+ membrane in Stratalinker 2400 UV Crosslinker under Autocrosslink mode (1200 microjoules [x100]; 25-50 sec).The membrane was washed in 10 ml of wash buffer in a clean washing tray for 5 min at room temperature.The membrane was blocked for 1 hour at room temperature, and then incubated with anti-m6A antibody (1:250 dilution; 2 μg/ml) overnight at 4 °C with gentle shaking.The membrane was incubated with IgG-HRP (1:10,000 dilution; 20 ng/ml) for 1 hour at room temperature with gentle shaking.The immuno-positive images were visualized by enhanced chemiluminescence reagent (Millipore, USA) according to the manufacturer's instructions.The membranes were then stained with 0.02% methylene blue in 0.3 M sodium acetate (pH 5.2) for 2 hours and washed with ribonuclease-free water for 5 hours.

RNA stability assays
Macrophage were treated with actinomycin D at a final concentration of 5 μg/mL.Total RNA was extracted at 0, 1 and 2 hours after adding actinomycin D. The mRNA stability was determined by RT-qPCR and the data were normalized to that at t = 0 time point.

Parabiosis study
Cx3cr1Cre; Rosa26 Td or Cx3cr1Cre; Alkbh5 flox/flox ; Rosa26 Td donor mice were surgically joined with CCR2 GFP/GFP recipient mice with longitudinal incisions along the side of the mouse extending from approximately 1 cm behind the ear to just past the hind limb.Mice remained surgically joined for 7 days to connect the blood circulation before Angiotensin II infusion.Hearts were harvested at 14 days following Angiotensin II infusion and analyzed by Immunohistochemistry or flow cytometry.
Non-myocardial single-cell RNA sequencing Single-cell isolation.In brief, hearts from Angiotensin II treated C57BL/6 mice were isolated and stored in the sCelLiveTM Tissue Preservation Solution (Singleron Bio Com, Nanjing, China) on ice within 30 mins.The hearts were washed with Hanks Balanced Salt Solution (HBSS) 3 times and then digested with 2 ml sCelLiveTM Tissue Dissociation Solution (Singleron) by Singleron PythoN™ Automated Tissue Dissociation System (Singleron) at 37 °C for 15 mins.The cell solution was then passed through a 40 μm cell strainer to collect the non-CM cells, centrifuged at 500 × g for 5 mins and suspended softly with GEXSCOPE® red blood cell lysis buffer (Singleron, 2 ml) for another 10 mins to remove red blood cells.Finally, the samples were stained with trypan blue (Sigma, United States) and the cellular viability was evaluated microscopically.The live single suspension was run on a BD LSRII machine and then constructed library as follows.
scRNA sequencing.Sorted single-cell suspensions (1 × 10 5 cells/ml) with PBS (HyClone) were loaded into microfluidic devices using the Singleron Matrix® Single Cell Processing System (Singleron).Subsequently, the scRNA-seq libraries were constructed according to the protocol of the GEXSCOPE® Single Cell RNA Library Kits (Singleron) 48 .Individual libraries were diluted to 4 nM and pooled for sequencing.At last, pools were sequenced on Illumina HiSeq X with 150 bp paired end reads.
Pre-processing and quality control for scRNA-seq data.Raw reads were processed to generate gene expression profiles using an internal pipeline in Singleron company (Nanjing, China).Briefly, cell barcode and UMI was extracted after filtering read one without poly T tails.Adapters and poly-A tails were trimmed (fastp V1) before aligning read two to GRCh38 with ensemble version 92 gene annotation (fastp 2.5.3a and featureCounts 1.6.2) 10 .Reads with the same cell barcode, UMI and gene were grouped together to calculate the number of UMIs of genes in each cell.The UMI count tables of each cellular and barcode were employed for further analysis.Cell type identification and clustering analysis were performed by the Seurat program (Version 4.3.0) 49,50.UMI count tables were loaded into R using read.tablefunction.Afterwards, parameter resolution to 0.6 was set for FindClusters function to clustering analyses.Differentially expressed genes (DEGs) between different samples or consecutive clusters were identified with function FindMarkers 51 .RNA velocity analysis by scVelo was performed to analyze the origin of myofibroblasts.
Statistical analysis.All data were presented as mean ± SEM.Shapiro-Wilk test was used to determine the normality of data, and all data passed normality and equal variance.Comparisons between 2 groups were analyzed by 2-tailed Student's t-test.Differences between multiple groups were performed using one-way ANOVA followed by Tukey post-hoc tests.2-way ANOVA followed by Tukey post-hoc tests was used for comparisons between multiple groups when there were 2 experimental factors.Biological experimental replicates in each group are shown in Fig. legends.P < 0.05 was considered to denote statistical significance.All statistical analyses were performed by GraphPad.Prism.9software package (SPSS Inc, Chicago, IL).

Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Fig. 1 |
Fig. 1 | Hypertension increases macrophage-myofibroblast transition.A Diagram of single cell RNA sequencing of non-myocardial cells from Angiotensin II or PBS infused C57BL/6 mice.B UMAP dimensionality reduction analysis displayed clusters of non-myocardial cells.C UMAP dimensionality reduction analysis of macrophage.D Feature plots depicting single-cell gene expression of selected macrophage and myofibroblast genes by UMAP dimensionality reduction analysis.E, RNA velocity (scVelo) analysis showed macrophage to myofibroblast transition.F Representative flow cytometry analyses of cardiac CD11b + SMA + cells gated on Td + cells from PBS or Ang II treated Cx3cr1Cre; Rosa26 Td and Lyz2Cre;Rosa26 Td mice, or GFP+ cells from PBS or Ang II treated CCR2GFP mice, with representative images at left and quantification at right.n = 5.G Representative immunofluorescent images (left) and quantification (right) of SMA + Td + cells of cardiac tissues from Cx3cr1Cre; Rosa26 Td , Lyz2Cre; Rosa26 Td and CCR2GFP mice with and without Ang II infusion.n = 5.Scale bar, 50 μm.All data are presented as mean ± standard error mean.Data in F, G were analyzed by two-tailed unpaired Student's t-test.P < 0.05 was considered as statistically significant.Ang II Angiotensin II, scRNA-seq single cell RNA sequencing, UMAP uniform manifold approximation and projection, SMC smooth muscle cell, CM cardiomyocyte, EC endothelial cell, NK natural killer.

Fig. 2 |
Fig. 2 | ALKBH5 upregulation is involved in hypertension-induced cardiac MMT.A Dot plots depicting single-cell gene expression of ALKBH5 in macrophage subclusters visualized on the UMAP dimensionality reduction plots.B The heat maps showing the marker gene and ALKBH5 expression of each cell cluster.C The ALKBH5 and Acta2 expression in pseudotemporal trajectory analysis from cluster macrophages to myofibroblasts.D ALKBH5 expression in macrophages and myofibroblast-like macrophages shown by violin.E m6A dot blot assay of mRNAs in cardiac Td+ cells from Cx3cr1Cre Td mice with or without Ang II infusion (n = 5).F Expression of ALKBH5 by western blot in cardiac Td + cells from Cx3cr1Cre Td mice with and without Ang II infusion.n = 5.G Representative immunofluorescent images and quantification of SMA + Td + cells in cardiac tissues from Cx3cr1Cre Td and ALKBH5 macKO-Td mice (n = 5 for PBS group and n = 7 for Ang II group).Scale bar, 100 μm.H mRNA levels of SMA and collagen (Col) types I and III expression in cardiac Td + cells from Cx3cr1Cre Td and ALKBH5 macKO mice.n = 5 for PBS group and n = 7 for Ang II group.I Representative immunofluorescent images (left) and quantification (right) of SMA + cells in cultured Td + macrophages cells from Cx3cr1Cre Td and ALKBH5 macKO mice with and without Ang II infusion.n = 5.Scale bar, 100 μm.J Representative images and quantification of the SMA and collagen (Col) types I and III expression by western blot in cardiac tissues of Cx3cr1Cre Td and ALKBH5 macKO- Td mice (n = 5 for PBS group and n = 7 for Ang II group).All data are presented as mean ± standard error mean.Data in E, F were analyzed by two-tailed unpaired Student's t test.Data in D were analyzed by Wilcoxon signed-ranks test.Data in G were analyzed by two-way ANOVA followed by Tukey post-hoc tests.Data in H-J were analyzed by one-way ANOVA followed by Tukey post-hoc tests.n.s.indicates nonsignificant.P < 0.05 was considered as statistically significant.UMAP uniform manifold approximation and projection, Ang II Angiotensin II, FOV field of view, WT wild-type, KO knockout, Col collagen.

Fig. 3 |
Fig. 3 | Specific deletion of ALKBH5 in Cx3cr1 lineage improves Ang II induced MMT and cardiac fibrosis.A Quantification analysis of SBP in sham and Ang II treated WT and ALKBH5 macKO-Td mice (n = 5 for PBS group and n = 7 for Ang II group).B Quantification analysis of heart weight/body weight in sham and Ang II treated WT and ALKBH5 macKO-Td mice (n = 5 for PBS group and n = 7 for Ang II group).C Quantification analysis of LV mass in sham and Ang II treated WT and ALKBH5 macKO-Td mice (n = 5 for PBS group and n = 7 for Ang II group).D Quantification analysis of fractional shortening in sham and Ang II treated WT and ALKBH5 macKO-Td mice (n = 5 for PBS group and n = 7 for Ang II group).E Representative echocardiography images and quantification of ejection fraction (left) and E/e' (right) of WT and ALKBH5 macKO-Td mice after PBS or Ang II treatment for 14 days (n = 5 for PBS group and n = 7 for Ang II group).F Representative images of Masson trichrome staining in cardiac tissue and quantification of positive fibrotic area (n = 5 for PBS group and n = 7 for Ang II group).Scale bar, 100 μm.G Representative images and quantification of the SMA and collagen (Col) types I and III expression by western blot in cardiac tissues of WT and ALKBH5 macKO-Td mice (n = 5 for PBS group and n = 7 for Ang II group).H Representative images of wheat germ agglutinin staining in cardiac tissue and quantification of positive fibrotic area (n = 5 for PBS group and n = 7 for Ang II group).Scale bar, 100 μm.I mRNA levels of ANP and BNP expression in cardiac Td + cells from WT and ALKBH5 macKO mice.n = 5 for PBS group and n = 7 for Ang II group.All data are presented as mean ± standard error mean.Data in A-F and H were analyzed by two-way ANOVA followed by Tukey post-hoc tests.Data in G, I were analyzed by one-way ANOVA followed by Tukey post-hoc tests.n.s.indicates nonsignificant.P < 0.05 was considered as statistically significant.Ang II Angiotensin II, WT wild-type, ANP atrialnatriureticpeptide, BNP brain natriuretic peptide, Col collagen.

Fig. 4 |
Fig. 4 | ALKBH5 in circulating monocytes-derived macrophage contributes to hypertension-induced cardiac fibrosis and dysfunction.A Diagram of parabiosis between CCR2 KO and Cx3cr1Cre; Rosa26 Td or Cx3cr1Cre; ALKBH5 fl/fl ; Rosa26 Td mice, respectively.B Representative images and quantification of flow cytometry analyses of Td + and CD11b + SMA + cells gated on Td + cells in hearts from CCR2 KO mice cojoined with Cx3cr1Cre; Rosa26 Td or Cx3cr1Cre; ALKBH5 fl/fl ; Rosa26 Td mice.n = 5.C Representative immunofluorescent images and quantification of SMA + cells in Td + cells of cardiac tissues from CCR2 KO cojoined with Cx3cr1Cre; Rosa26 Td or Cx3cr1Cre; ALKBH5 fl/fl ; Rosa26 Td mice (n = 5).Scale bar, 100 μm.D Representative echocardiography images of ejection fraction of the CCR2 KO cojoined with Cx3cr1Cre; Rosa26 Td or Cx3cr1Cre; ALKBH5 fl/fl ; Rosa26 Td mice after Ang II treatment for 14 days, with indices of cardiac ejection fraction and E/e' at right.n = 5.E Representative echocardiography images of E/e' (E) of the CCR2 KO cojoined with Cx3cr1Cre; Rosa26 Td or Cx3cr1Cre; ALKBH5 fl/fl ; Rosa26 Td mice after Ang II treatment for 14 days, with indices of cardiac ejection fraction and E/e' at right.n = 5.F Representative images of Masson trichrome staining.n = 5.Scale bar, 100 μm.G Quantification of positive fibrotic area.n = 5.Scale bar, 100 μm.H Representative images of SMA and ECM genes collagen I and III in cardiac tissues shown by western blot (n = 5).All data are presented as mean ± standard error mean.Data in B-E and G were analyzed by two-way ANOVA followed by Tukey post-hoc tests.Data in H were analyzed by one-way ANOVA followed by Tukey post-hoc tests.n.s.indicates nonsignificant.P < 0.05 was considered as statistically significant.Ang II Angiotensin II, WT wild-type, KO knockout, Col collagen.

Fig. 5 |
Fig. 5 | Macrophage ALKBH5 directly regulates m6A modification on IL-11 mRNA.A top abundant enriched target mRNAs of ALKBH5 induced by Angiotensin II.B Top enriched pathway of ALKBH5 induced by Angiotensin II.C m6A motif identified from m6A RNA immunoprecipitation-sequencing analysis.D Metagene distribution of m6A peaks along the whole transcriptome, including 5' untranslated regions (5' UTR), covering coding sequence (CDS) and 3' UTR.E m6A peak distribution on IL-11 mRNA shown by Integrative Genomics Viewer (IGV) tool.kb means kilo base pair.F Relative fold change of m6A enrichment on IL-11 mRNA by m6A RIP-qPCR analysis (n = 4).G Degradation of IL-11 mRNA was detected in macrophages following treating with actinomycin D for indicated time (n = 3).H IL11RA1 gene signatures visualized by Feature Plots.I Expression level of IL11RA1 along the pseudotime trajectory.J Protein levels of IL11RA1 in cardiac Td + cells of the hearts following PBS and Ang II treatment shown by western blot with representative images at left and quantification at right.n = 5.All data are presented as mean ± standard error mean.Data in F were analyzed by one-way ANOVA followed by Tukey post-hoc tests.Data in G were analyzed by two-way ANOVA followed by Tukey post-hoc tests.Data in J were analyzed by two-tailed unpaired Student's ttest.n.s.indicates nonsignificant.P < 0.05 was considered as statistically significant.UTR Untranslated Regions, WT wild-type, KO knockout, UMAP uniform manifold approximation and projection, Ang II angiotensin II.

Fig. 6 |
Fig. 6 | IL-11 overexpression reverses macrophage ALKBH5 deletion-mediated improved cardiac dysfunction via increasing cardiac fibrosis.A Representative flow cytometry analyses of Td+ and CD11b+SMA+cells in hearts from C57BL6/J mice transplanted bone marrow from ALKBH5 macKO-Td -BM and Cx3cr1Cre Td -BM mice with empty vector or IL11 overexpression, with quantification at right.n = 6 for WT and ALKBH5mac KO group.n = 7 for IL11 OE group.n = 5 for ALKBH5mac KO and IL11 OE group.B Representative immunofluorescent images and quantification of SMA + cells in Td + cells of cardiac tissues from the reconstituted mice.n = 6 for WT and ALKBH5mac KO group.n = 7 for IL11 OE group.n = 5 for ALKBH5mac KO and IL11 OE group.Scale bar, 100 μm.C Representative echocardiography images and quantification of ejection fraction of the above mice after Ang II treatment for 2 weeks.n = 6 for WT and ALKBH5mac KO group.n = 7 for IL11 OE group.n = 5 for ALKBH5mac KO and IL11 OE group.D Representative echocardiography images and quantification of E/e' of the above mice after Ang II treatment for 2 weeks.n = 6 for WT and ALKBH5mac KO group.n = 7 for IL11 OE group.n = 5 for ALKBH5mac KO and IL11 OE group.E Representative images of Masson trichrome staining in cardiac tissue and quantification of positive fibrotic area.n = 6 for WT and ALKBH5mac KO group.n = 7 for IL11 OE group.n = 5 for ALKBH5mac KO and IL11 OE group.Scale bar, 100 μm.F Representative images and quantification of the SMA, collagen (Col) types I and III and IL11 expression by western blot in cardiac tissues of the reconstituted mice.n = 6 for WT and ALKBH5mac KO group.n = 7 for IL11 OE group.n = 5 for ALKBH5mac KO and IL11 OE group.All data are presented as mean ± standard error mean.Data in A-E and F were analyzed by one-way ANOVA followed by Tukey posthoc tests.n.s.indicates nonsignificant.P < 0.05 was considered as statistically significant.WT wild-type, KO knockout, OE overexpression, Col collagen.

Fig. 7 |
Fig. 7 | Nanoparticle monocyte/macrophage-target delivery of ALKBH5 siRNA attenuates Ang II-induced cardiac fibrosis and dysfunction.A Schematic diagram of lipid nanoparticle.B Size distribution of lipid nanoparticle measured by dynamic light scattering (DLS).C Zeta potential of lipid nanoparticle measured by dynamic light scattering (DLS).D The represent cryo-transmission electron micrographs (TEMS) of lipid nanoparticle.Scale bar, 50 nm.E Experimental diagram of in vivo ALKBH5 loss-of-function using C12-200 lipid nanoparticles.F Representative fluorescence imaging combined with microCT after intravenous injection of DiR-labeled C12-200 lipid nanoparticles, with quantification of fluorescent intensity at right.n = 5.G Representative echocardiography images and indices of ejection fraction of mice with scramble or ALKBH5 siRNA/LNP.n = 5.H Representative echocardiography images and indices of E/e' of mice with scramble or ALKBH5 siRNA/LNP.n = 5.I Representative images (left) and quantification of positive fibrotic area (right) of Masson trichrome staining in cardiac tissue of mice with scramble or ALKBH5 siRNA/LNP n = 5.Scale bar, 100 μm.J Representative immunofluorescent images and quantification of SMA + cells in cardiac tissues from scramble or ALKBH5 siRNA/LNP treated mice.n = 5.Scale bar, 100 μm.K Western blot analysis of IL11RA and ECM genes collagen I and III in cardiac tissues from mice with scramble or ALKBH5 siRNA/LNP.Quantitative results are shown on the right.n = 5.All data are presented as mean ± standard error mean.Data in F-K were analyzed by two-tailed unpaired Student's t-test.n.s.indicates nonsignificant.P < 0.05 was considered as statistically significant.siRNA small interfering RNA, LNP lipid nanoparticle, Col collagen.