Abstract

Background:

   Chronic alcohol consumption in adults can induce various cardiac
   toxicities such as arrhythmias, cardiomyopathy, and heart failure.
   Prenatal alcohol exposure can increase the risk of developing
   congenital heart diseases. Understanding biological mechanisms involved
   in the long-term alcohol exposure-induced cardiotoxicity is pivotal to
   the discovery of therapeutic strategies.

Methods:

   Cardiomyocytes derived from human induced pluripotent stem cells
   (hiPSC-CMs) were engineered into cardiac spheroids and treated with
   clinically relevant concentrations of ethanol (17 and 50 mM) for 5
   weeks. The cells were then analyzed for changes in mitochondrial
   features, transcriptomic and metabolomic profiles as well as integrated
   omics outcomes.

Results:

   Following chronic ethanol treatment of hiPSC-CMs, a decrease in
   mitochondrial membrane potential and respiration and changes in
   expression of mitochondrial function-related genes were observed.
   RNA-sequencing analysis revealed changes in various metabolic
   processes, heart development, response to hypoxia, and extracellular
   matrix-related activities. Metabolomics analysis revealed dysregulation
   of energy metabolism and increased metabolites associated with
   upregulation of inflammation. Integrated omics analysis further
   identified functional subclusters and revealed potentially affected
   pathways that are associated with cardiac toxicities.

Conclusion:

   Chronic ethanol treatment of hiPSC-CMs resulted in overall decreased
   mitochondrial function, increased glycolysis, disrupted fatty acid
   oxidation and impaired cardiac structural development.

   Keywords: Stem cells, cardiomyocytes, alcohol, mitochondria, RNA-Seq,
   metabolomics

INTRODUCTION

   Alcohol is the most consumed and highly addictive chemical substance,
   and chronic consumption of alcohol can lead to increased risk of health
   problems. According to the National Institute of Alcohol Abuse and
   Alcoholism, nearly one million people died from alcohol-related causes
   between 1999 and 2017, and in 2017 alcohol was involved with 2.6% of
   the ~2.8 million deaths in the United States ([36]White et al., 2020).
   The impact of alcohol on human health is more prevalent on a global
   scale: in 2018, the World Health Organization reported that the harmful
   use of alcohol contributed to more than 5% of deaths worldwide per
   annum ([37]World Health Organization, 2018). Alcohol can affect the
   cardiovascular system, causing various disease states such as
   arrhythmias ([38]Voskoboinik et al., 2019) and dilated cardiomyopathy
   ([39]Glymour, 2014). Alcohol-induced cardiotoxicity is not restricted
   to just adults: according to the Center for Disease Control and
   Prevention, about 10% of pregnant woman reported current drinking and
   about half of them reported binge drinking ([40]England et al., 2020),
   which increases the risk of fetal alcohol spectrum disorder (FASD).
   Likewise, a large, multi-community-based study estimated that the full
   range of FASDs in the US might number as high as 50 per 1,000 school
   children ([41]May et al., 2018). Consequently, there is a serious
   threat to the cardiovascular health of a significant portion of the
   population given that congenital heart disease is a common clinical
   manifestation of FASD with comorbidity rate of 28.6% ([42]Burd et al.,
   2007).

   Given the severity of alcohol-induced cardiotoxicity, understanding the
   mechanisms involved is critical to developing clinical guidelines and
   therapeutic interventions. The effects of chronic alcohol exposure have
   been mostly investigated in animal models treated with ethanol (EtOH).
   In these models, long-term EtOH treatment leads to oxidative stress,
   autophagy, apoptosis, mitochondrial dysfunction, activation of
   pro-inflammatory pathway, and cardiac contraction defects
   ([43]Nakashima et al., 2019, [44]Mouton et al., 2020, [45]Yang et al.,
   2021, [46]Wang et al., 2021b). Furthermore, EtOH exposure during
   embryonic development results in heart chamber defects, cardiac
   hypertrophy, fibrosis, apoptosis, oxidative stress, Ca^2+ overload, and
   contractile dysfunction ([47]Ren et al., 2002, [48]Nguyen et al., 2014,
   [49]Sarmah and Marrs, 2017, [50]Ninh et al., 2019). However, since
   human and animal cardiac physiologies differ significantly ([51]Guo et
   al., 2018), there is a need for a model more reflective of human
   cardiac physiology.

   Human induced pluripotent stem cell-derived cardiomyocytes (hiPSC-CMs)
   have shown great potential in addressing the main limitations of human
   primary cardiomyocytes ([52]Hnatiuk et al., 2021), which are difficult
   to obtain in large quantities and maintain in vitro for an extended
   period of time while preserving key functional characteristics
   ([53]Mitcheson et al., 1998). Indeed, hiPSC-CM models have been
   successfully used to model various cardiac disease conditions, as well
   as alcohol-induced cardiotoxicity ([54]Rampoldi et al., 2019, [55]Liu
   et al., 2021, [56]Liu et al., 2020b). As hiPSC-CMs resemble fetal
   cardiomyocytes, these cells could be used as an ideal cellular system
   to model prenatal alcohol exposure.

   Here, we report the effects of long-term treatment of hiPSC-CMs with
   EtOH. We found that long-term EtOH treatment decreased mitochondrial
   membrane potential and mitochondrial function and altered the
   expression of genes associated with mitochondrial function in a
   concentration-dependent manner. In addition, by transcriptomic and
   metabolomic analyses we found that long-term EtOH treatment resulted in
   global shifts in gene expression and/or metabolites associated with
   metabolic processes, heart development, response to hypoxia,
   extracellular matrix (ECM)-related activities, vascular smooth muscle
   function, stem cell pluripotency regulation, and regulation of
   inflammation.

Materials and Methods

   Vendor information and catalog numbers for major reagents are available
   in [57]Table S1.

Cell culture and EtOH treatment

   hiPSC line IMR90 (WiCell Research Institute) was fed daily on
   Matrigel-coated plates with mTeSR medium. hiPSCs were induced to
   differentiate towards cardiomyocytes using a growth factor driven
   differentiation protocol ([58]Jha et al., 2015, [59]Laflamme et al.,
   2007). At the day of induction (day 0), medium was replaced with RPMI
   1640 medium supplemented with 2% B27 minus insulin containing 100 ng/ml
   Activin A. After 24 h, medium was replaced with fresh RPMI supplemented
   with 2% B27 minus insulin containing 10 ng/ml BMP4. After 3 days,
   medium was changed to RPMI supplemented with 2% regular B27. At day 5
   of differentiation, the cells were dissociated and reseeded into
   AggreWell400 plates to obtain cardiac spheroids (CSs) ([60]Jha et al.,
   2016). After 24 h, the CSs were collected and transferred to
   low-adhesion dishes for suspension culture with medium refreshed every
   2 days. CSs typically started spontaneous beating by day 7 to 9 of
   differentiation. From day 14, different concentrations of
   EtOH-containing medium were freshly prepared by diluting pure ethyl
   alcohol in the culture medium to 2× test concentrations. Every 3 days,
   CSs were collected and resuspended in fresh culture medium, the same
   volume of 2× EtOH-working solutions were added into the dishes, and
   then mineral oil was overlaid on top of the medium to prevent EtOH
   evaporation. EtOH concentration of 17 mM is the legal limit in the
   United States to drive, corresponding to blood alcohol concentration of
   0.08% (g/100 mL blood); and 50 mM corresponds to the blood alcohol
   concentration of 0.24%, which is associated with the excitement stage
   of alcohol intoxication in clinic. The exposure duration of 5 weeks was
   selected according to the findings from our previous study ([61]Liu et
   al., 2021). We found that EtOH treatment of hiPSC-CMs for 5 weeks
   resulted in lower cell viability than EtOH treatment for 1 week or 3
   weeks.

Immunocytochemistry and cardiomyocyte purity quantification

   For immunocytochemistry, CSs were dissociated and reseeded into
   Matrigel-coated 96-well plates. Cells were fixed in 4% paraformaldehyde
   for 15 min and permeabilized using ice-cold methanol for 2 min at room
   temperature (RT). The cells were then incubated with 5% normal goat
   serum (NGS) in phosphate-buffered saline (PBS) at RT for 1 h and
   incubated with primary antibodies including NKX2–5, cardiac troponin T
   (cTnT), and α-actinin ([62]Table S2) in 3% NGS overnight at 4°C. After
   washing with PBS, the cells were incubated with the corresponding
   secondary antibodies at RT for 1 h in dark followed by counterstaining
   the nuclei with 7 μM Hoechst 33342. For cardiomyocyte purity
   quantification, images were acquired and quantitatively analyzed using
   ArrayScan XTI Live High Content Platform (Thermo Fisher Scientific).
   Briefly, total cell numbers and cell locations were acquired based on
   the Hoechst staining of the nuclei. Signal detection of NKX2–5 was
   restricted to the nucleus and that of α-actinin and cTnT was restricted
   to the cytoplasm with a mask that extended 7 units from the nucleus.
   The percentages of cells positive for NKX2–5, α-actinin and cTnT were
   calculated as cardiomyocyte purity.

Detection of mitochondrial membrane potential

   For tetramethylrhodamine (TMRM) staining, CSs were dissociated and
   reseeded into Matrigel-coated 96-well plates. The cells were incubated
   with 100 nM TMRM and 7 μM Hoechst working solutions in warm 0.25%
   bovine serum albumin (BSA) in PBS solution for 15 min at 37°C in dark.
   Images were acquired and analyzed using ArrayScan XTI Live High Content
   Platform. Briefly, cell locations were determined based on the Hoechst
   staining of the nuclei. Signal detection of TMRM was restricted to the
   cytoplasm with a mask that extended 7 units from the nucleus. Mean
   fluorescence intensities of TMRM of cells were quantified and used as
   readout.

Seahorse XF24 metabolic flux analysis

   The oxygen consumption rate (OCR) was measured using the Seahorse XF24
   Extracellular Flux Analyzer (Agilent Technologies, Santa Clara)
   ([63]Gentillon et al., 2019). CSs with or without EtOH treatment for 5
   weeks were dissociated and seeded into a Matrigel-coated Seahorse XF-24
   cell culture plate at a density of 2×10^5 cells/well. Cells were
   maintained for 2 days to allow them to attach and adapt to the Seahorse
   plate before the assay. During the 2-day incubation, the cells were not
   exposed to ethanol, since the Seahorse readout may be affected by
   mineral oil used on top of the medium to prevent EtOH evaporation. We
   note the potential limitation of this assay protocol: the 2-day
   incubation could cause an underestimation of the detrimental effect of
   EtOH on mitochondrial function if the effects from 5 weeks of ethanol
   exposure were reversible to some extent during the 2-day incubation.
   One hour before the assay, the cells were washed once with XF assay
   medium (non-buffered RPMI supplemented with 10 mM glucose, 2 mM Sodium
   Pyruvate and 2 mM Glutamine) and incubated in 525 μl of XF assay medium
   at 37°C in a non-CO[2] incubator. Mitochondrial function was analyzed
   using the XF Cell Mito Stress Kit. Mitochondrial inhibitors—oligomycin
   (2 μM), carbonyl cyanide p-(trifluoromethoxy) phenylhydrazone (FCCP, 1
   μM) and rotenone (0.5 μM) + antimycin A (0.5 μM)—were diluted in XF
   assay medium and sequentially added into each well during the
   measurements. Each measurement cycle comprised 3 min of mixing, 2 min
   of waiting, and 3 min measurements of OCR (pmol/min) and extracellular
   acidification rate (ECAR, mpH/min). Measurement cycles were performed
   after each injection of given compounds. The results were normalized to
   10^5 cells as determined by trypan blue assay. The data were analyzed
   using Wave 2.6 and Report Generator 4.0.

Quantification of mitochondrial translocase of outer membrane (TOM) receptor
complex component

   CSs were dissociated and reseeded into Matrigel-coated 96-well plates.
   Cells were fixed in 4% paraformaldehyde for 15 min and permeabilized
   using ice-cold methanol for 2 min at RT. The cells were then incubated
   with 5% NGS in PBS at RT for 1 h and incubated with anti-Tom20 primary
   antibody ([64]Table S2) in 3% NGS overnight at 4°C. After washing with
   PBS, the cells were incubated with the corresponding secondary antibody
   at RT for 1 h in dark followed by counterstaining the nuclei with 7 μM
   Hoechst 33342. Images were acquired and analyzed using ArrayScan XTI
   Live High Content Platform. Briefly, cell locations were determined
   based on the Hoechst staining of the nuclei. Signal detection of Tom20
   was restricted to the cytoplasm with a mask that extended 7 units from
   the nucleus. Mean fluorescence intensities of Tom20 of cells were
   quantified and used as readout.

Quantitative real-time polymerase chain reaction (qRT-PCR)

   Aurum total RNA mini kit was used to extract RNA and SuperScript VILO
   cDNA Synthesis Kit was used to reverse transcribe 1μg of RNA into cDNA,
   according to the manufacturer’s instructions. qRT-PCR was performed in
   an Applied Biosystems 7500 real-time PCR system (Thermo Fisher
   Scientific) using the iTaq SyBr green master mix. Human specific PCR
   primers ([65]Table S3) for the genes examined were retrieved from open
   access websites ([66]https://pga.mgh.harvard.edu/primerbank/). All
   samples were normalized to the level of the housekeeping gene GAPDH.
   Relative expression levels were calculated by the 2^-ΔΔCt method.

Quantification of mitochondrial DNA content

   Total genomic DNA (gDNA) was isolated using QIAamp DNA Mini Kit
   according to the manufacturer’s instructions. Following determination
   of gDNA concentration using a UV–vis spectrophotometer (NanoDrop,
   Thermo Fisher Scientific), samples were diluted to 20 ng/μl. RT-PCR was
   performed in Applied Biosystems 7500 real-time PCR systems (Thermo
   Fisher Scientific) using the iTaq SyBr green master mix for nuclear
   genes, succinate dehydrogenase subunit A (SDHA) and lipoprotein lipase
   (LPL), and mitochondrial genes, NADH dehydrogenase subunit I (ND1) and
   mitochondrial cytochrome oxidase II (MT-CO2). The mitochondrial DNA
   (mtDNA) was normalized to nuclear DNA (nDNA) ([67]Gentillon et al.,
   2019).

RNA-seq analysis

   RNA was extracted from both hiPSC-CMs with and without EtOH treatment
   using Aurum total RNA mini kit (Bio-Rad, CA) according to
   manufacturer’s instructions. Library preparation and sequencing were
   performed at the Parker H. Petit Institute for Bioengineering and
   Bioscience at the Georgia Institute of Technology. RNA quality was
   assessed on a Bioanalyzer instrument (Agilent, CA Technologies) and all
   samples had RNA integrity number (RIN) ≥ 7. cDNA was derived from RNA,
   hybridized and probe intensities were generated on Illumina HumanHT-12
   v3 (Illumina, CA). RNA sequence reads were aligned to the human
   reference genome (GRCh38) using HISAT2 (hierarchical indexing for
   spliced alignment of transcripts), followed by using UCSC reference
   annotation and HTSeq to estimate gene abundance. All downstream
   analyses were performed in R 4.1.2. Read count normalization and
   differential expression analysis were performed using DESeq2 R package
   1.34.0 ([68]Table S4, [69]S5) ([70]Love et al., 2014). Gene Ontology
   (GO) enrichment analysis of differentially expressed genes was done
   using clusterProfiler R package 4.2.2 ([71]Yu et al., 2012).

Metabolomic analysis

   hiPSC-CMs treated with or without EtOH (n=3/group, 1–2×10^6 per sample)
   were lysed and extracted with ice cold acetonitrile and water (2:1
   v/v). Cell extracts were analyzed as described previously ([72]Go et
   al., 2014). Briefly, thawed extracts were incubated at 4 °C for 30 min,
   centrifuged at 16100 g for 10 min to remove protein and transferred to
   a refrigerated (4 °C) autosampler for analysis. Samples were analyzed
   with three technical replicates using ultra-high-resolution mass
   spectrometry with hydrophilic interaction liquid chromatography (HILIC)
   [Accucore HILIC 100 × 2.1 mm columns]. Electrospray ionization was used
   in the positive ion mode on a Thermo Scientific Q-Exactive HF mass
   spectrometer (Thermo, Waltham, Massachusetts) operating with a
   resolution of 120000 and scan range of 85–1275 m/z (mass to charge).
   Analyte separation for HILIC was performed with a Waters XBridge BEH
   Amide XP HILIC column (2.1 × 50 mm^2, 2.6 μm particle size) and
   gradient elution with mobile phases including water, acetonitrile and
   formic acid as described previously ([73]Liu et al., 2020a). Raw data
   were extracted with apLCMS ([74]Yu et al., 2009) and xMSanalyzer
   ([75]Uppal et al., 2013). Significant features with partial least
   squares discriminant analysis [PLS-DA] score > 2 were further studied
   by pathway enrichment analysis using Mummichog ([76]Li et al., 2013).
   Metabolites in pathways were verified via xMSannotator ([77]Uppal et
   al., 2017). This approach protects against type 2 statistical error by
   including all features at P < 0.05 and protects against type 1
   statistical error by permutation testing.

Integrated omics analysis

   Integrated omics analysis of proteomics ([78]Liu et al., 2020b),
   transcriptomics, and metabolomics data was performed using xMWAS
   network algorithm ([79]Uppal et al., 2018), which uses sparse partial
   least squares regression, community detection algorithms and eigen
   vector centrality measures to estimate pair-wise correlations between
   omics features to identify protein-gene-metabolite communities. The
   number of protein-gene-metabolite communities was identified by
   optimizing cluster modularity, a frequently used community detection
   algorithm that partitions a network into clusters made up of densely
   connected nodes, so that nodes belonging to different clusters are
   sparsely connected ([80]Subelj and Bajec, 2011). For this dataset,
   correlation threshold of 0.70 and p-value cutoff of 0.05 were used.

Statistical analysis

   Data were analyzed and graphed in Excel, GraphPad Prism 8, and RStudio.
   Data are presented as mean ± SD. Comparisons were conducted via One-way
   ANOVA test followed by multiple comparison procedures (Dunnett method)
   with significant differences defined by P < 0.05 (*), P < 0.01 (**), P
   < 0.001 (***), P < 0.0001 (****). Sample sizes were given for each
   experiment.

RESULTS

Chronic EtOH exposure decreases mitochondrial membrane potential and
respiration of hiPSC-CMs

   To examine the effect of long-term EtOH exposure on the mitochondrial
   function of hiPSC-CMs, we first generated highly enriched hiPSC-CMs
   through cardiac induction and formation of cardiac spheroids on
   differentiation day 5 ([81]Figure 1A). Cardiac spheroids were generated
   by microtissue engineering ([82]Figure 1B), which helps better mimic in
   vivo cellular microenvironment than 2D cultures ([83]Liu et al.,
   2020b). As shown in [84]Figure 1C, 3D differentiation cultures
   contained cells that were positive for cardiomyocyte-associated markers
   including NKX2–5, α-actinin and cardiac troponin T (cTnT) at 91%, 84%,
   and 86%, respectively.

Figure 1. Experimental scheme of hiPSC-CM generation and characterization.

   Figure 1.
   [85]Open in a new tab

   (A) Overall experimental design. (B) Phase-contrast images of cardiac
   spheroids under two magnifications. (C) Representative images and
   quantification of immunocytochemistry acquired from ArrayScan revealing
   that the majority of the cells in culture were positive for cardiac
   transcription factor NKX2–5 and structural proteins cardiac troponin T
   and α-actinin, indicating highly enriched hiPSC-CMs at day 12 (n=5
   cultures).

   On differentiation day 14, cardiac spheroids were treated with EtOH at
   17 and 50 mM for 5 weeks. At the end of the treatment, we measured
   mitochondrial membrane potential by TMRM probe. Decreased TMRM signals
   were detected in cells treated with EtOH ([86]Figure 2A). The mean
   fluorescence intensity of TMRM was 10% lower in cells treated with 17
   mM EtOH and 13% lower in cells treated with 50 mM EtOH than in
   untreated (control) cells.

Figure 2. Chronic EtOH exposure decreases mitochondrial membrane potential
and respiration of hiPSC-CMs.

   Figure 2.
   [87]Open in a new tab

   (A) Representative images and quantification of TMRM staining in 3D
   hiPSC-CMs treated with EtOH for 5 weeks (n = 4). Relative mean
   fluorescence intensity (MFI) was calculated based on the average values
   of EtOH-treated group vs. untreated group. (B) Representative traces
   showing the oxygen consumption rate (OCR) of hiPSC-CMs following
   sequential addition of oligomycin (2 μM), FCCP (1 μM), and
   rotenone/antimycin A (0.5 μM). Bar charts present the quantification of
   basal respiration, maximal respiration, non-mitochondrial respiration,
   ATP production, spare respiratory capacity and coupling efficiency (n =
   4 cultures for no treatment group and n = 5 cultures for treatment
   groups). Comparisons were conducted between each treatment group and no
   treatment group via One-way ANOVA test. *, P-value < 0.05; **, P-value
   < 0.01; ***, P-value < 0.001; ****, P-value < 0.0001.

   We also measured the key parameters of mitochondrial coupling and
   respiratory by Seahorse XF24 metabolic flux analysis. As shown in
   [88]Figure 2B, changes of oxygen consumption rate (OCR) in control and
   EtOH-treated hiPSC-CMs were traced following the sequential supplements
   of the ATP synthase inhibitor oligomycin, a protonophore uncoupler
   FCCP, and electron inhibitors rotenone and antimycin A. Compared with
   the control, cells treated with 50 mM EtOH displayed dramatically
   decreased OCR in most parameters [in absolute OCR per 10^5 cells
   (pmol/min)] of the mitochondrial respiration, including basal
   respiration (84.73 ± 16.36 vs. 154.70 ± 26.75, P < 0.01), maximal
   respiration (209.00 ± 46.88 vs. 407.74 ± 60.24, P < 0.001),
   non-mitochondrial respiration (25.39 ± 9.95 vs. 55.60 ± 17.38, P <
   0.05), spare respiratory capacity (124.27 ± 33.38 vs. 253.04 ± 40.53, P
   < 0.001), and ATP production (73.25 ± 16.00 vs. 145.54 ± 19.18, P <
   0.0001). Cells treated with 17 mM EtOH had milder effects on
   mitochondrial respiration compared with those treated with 50 mM EtOH.
   Nevertheless, compared with no EtOH treatment, 17 mM EtOH exposure
   significantly decreased maximal respiration (271.25 ± 53.77 vs. 407.74
   ± 60.24, P < 0.01), spare respiratory capacity (139.88 ± 40.67 vs.
   253.04 ± 40.53, P < 0.01), and ATP production (116.17 ± 12.23 vs.
   145.54 ± 19.18, P < 0.05).

   Taken together, these observations indicate that chronic EtOH exposure
   affected mitochondrial function of hiPSC-CMs by decreasing
   mitochondrial membrane potential and damaging respiration process at a
   concentration-dependent manner.

Chronic EtOH exposure affects the expression of mitochondrial
function-related genes of hiPSC-CMs

   To investigate the molecular changes of mitochondria in hiPSC-CMs
   exposed to chronic EtOH treatment, we treated cardiac spheroids with
   EtOH at 17 and 50 mM for 5 weeks and quantified the expression of Tom20
   by immunocytochemical analysis. Tom20 is located at the surface of the
   mitochondrion outer membrane and functions as the transit peptide
   receptor responsible for the recognition and translocation of
   cytosolically synthesized mitochondrial preproteins. Compared with the
   control cells, Tom20 was expressed at lower levels in EtOH-treated
   hiPSC-CMs ([89]Figure 3A).

Figure 3. Chronic EtOH exposure affects the expression of critical
mitochondrial function-related genes of hiPSC-CMs.

   Figure 3.
   [90]Open in a new tab

   (A) Representative images and quantification of Tom20 expression in
   hiPSC-CMs treated with EtOH for 5 weeks (n = 4 cultures). (B) qRT-PCR
   analysis showing relative mRNA levels of mitochondrial function related
   genes in hiPSC-CMs treated with EtOH for 5 weeks (n = 3 cultures). (C)
   Quantification of the ratio of mitochondria-encoded complex I ND1 or
   MT-CO2 to nuclear-encoded complex II LPL or SHDA DNA in hiPSC-CMs
   treated with EtOH for 5 weeks (n = 3). Relative mean fluorescence
   intensity (MFI), mRNA levels and mt/nDNA ratios were calculated based
   on the average values of EtOH-treated group vs. untreated group.
   Comparisons were conducted between each treatment group and no
   treatment group via One-way ANOVA test. *, P-value < 0.05; **, P-value
   < 0.01***; P-value < 0.001.

   Next, we analyzed the expression of genes critical for mitochondrial
   function including genes implicated in fatty acid metabolism, glucose
   metabolism, and electron transport chain of mitochondria. The fatty
   acid metabolism-related genes included ACADVL (encoding very long chain
   acyl-Coenzyme A dehydrogenase, which catalyzes the first step of the
   fatty acid oxidation pathway), CPT1A and CPT1B (encoding carnitine
   palmitoyltransferase, which transports fatty acids from the cytoplasm
   into the mitochondria), PPARA and PPARGC1A (encoding peroxisome
   proliferator-activated receptors, which are transcriptional factors
   that regulate the genes involved in energy metabolism) and HADHA and
   HADHB (encoding the mitochondrial trifunctional protein
   hydroxyacyl-Coenzyme A dehydrogenase, which catalyzes the last three
   steps of oxidation of long chain fatty acids in mitochondria). The
   glucose metabolism-related genes included PDK1, PDK3, and PDK4
   (encoding pyruvate dehydrogenase kinases, which are important for
   modulating glucose metabolism in mitochondria). The electron transport
   chain of mitochondria-related genes included UCP3 (encoding uncoupling
   protein 3), COQ10A (encoding coenzyme Q10A), NDUFB5 (encoding
   NADH:ubiquinone oxidoreductase), OPA1 (encoding mitochondrial dynamin
   like GTPase), and MFN1 and MFN2 (encoding mitochondrial protein
   mitofusins which are essential for the maintenance and operation of the
   mitochondrial network). As shown in [91]Figure 3B and [92]Figure S1,
   compared with no EtOH treatment, the expression levels of COQ10A, MFN2,
   OPA1, HADHB, CPT1B, and PPARGC1A were 35%, 39%, 54%, 41%, 44%, and 38%
   lower, respectively, in cells exposed to 50 mM EtOH. The expression of
   MFN2, HADHB, CPT1B, and PPARGC1A was also significantly decreased in 17
   mM EtOH-treated cells. The reduced expression of COQ10A, MFN2, OPA1,
   HADAB, CPT1B and PPARGC1A in EtOH-treated cells was consistent with
   decreased mitochondrial function detected by metabolic flux analysis.

   In addition, we assessed mtDNA:nDNA ratio to evaluate the mitochondrial
   DNA content in hiPSC-CMs with or without long-term EtOH treatment. As
   shown in [93]Figure 3C, the mitochondrial DNA content was comparable in
   all the groups, indicating that chronic EtOH exposure did not affect
   mitochondrial DNA content in hiPSC-CMs.

   Together, these results show that chronic EtOH exposure affected the
   expression of genes that are crucial for mitochondrial function in
   hiPSC-CMs without altering mitochondrial DNA content.

Chronic EtOH treatment of hiPSC-CMs alters transcriptome profile identified
by RNA-seq analysis

   To more comprehensively examine changes induced by long-term EtOH
   exposure, we compared transcriptome profiles of cardiac spheroids
   treated with EtOH at 17 mM for 5 weeks with untreated cells by RNA-seq
   analysis. This analysis revealed 1703 differentially expressed genes
   (DEGs), of which 783 were upregulated and 920 downregulated (adjusted
   p-value < 0.05). From volcano plot visualization of all genes, some
   DEGs associated with cardiac structure, metabolism and extracellular
   matrix (ECM) organization were evident ([94]Figure 4A). For instance,
   glycolysis-related genes such as ENO1, GAPDH, and LDHA were upregulated
   while key cardiac structural genes MYL2 and MYH7 were downregulated.

Figure 4. RNA-seq identifies differentially expressed genes and GO terms.

   Figure 4.
   [95]Open in a new tab

   (A) Volcano plot showing differentially expressed genes. (B) GO
   enrichment using upregulated genes. (C) GO enrichment using
   downregulated genes. 3D hiPSC-CMs were treated with 17 mM ethanol for 5
   weeks and were compared with untreated cells (n=3 cultures).

   We next performed GO enrichment analysis on upregulated and
   downregulated DEGs ([96]Figure 4B, [97]4C). This analysis revealed
   upregulation of glycolytic process, carbohydrate metabolic process
   terms as well as response to hypoxia and T cell chemotaxis ([98]Figure
   4B, [99]Figure 5A, [100]Table S6). Upregulated DEGs associated with T
   cell chemotaxis included WNK1, WNT5A, ADAM10, CXCL11. This analysis
   also revealed that the downregulated DEGs were over-represented by GO
   terms related to extracellular matrix organization, blood vessel
   development, heart development, actin filament organization, cell
   growth, muscle cell development, female pregnancy, and response to
   hypoxia ([101]Figure 4C, [102]Figure 5B, [103]Table S7). Overall
   downregulation of development and growth was also evident, with
   downregulation of ECM function and cardiac development showing
   consistency with our previously published proteomics data ([104]Liu et
   al., 2020b). In particular, collagen trimer component genes (COL3A1,
   COL11A1, and COL1A1), muscle contraction genes (CRYAB, TNNI3, and
   ACTA1), apoptosis genes (PLK1 and SOD3), and pleiotropic genes included
   in multiple GO terms (SFRP1, COL1A1, and EMILIN1) showed decreased
   expression, which is consistent with previous findings ([105]Liu et
   al., 2020b).

Figure 5. RNA-seq identifies KEGG pathways as well as the association of
differentially expressed genes and GO terms.

   Figure 5.
   [106]Open in a new tab

   (A) Chord plot of upregulated pathways and its genes. (B) Chord plot of
   downregulated pathways and its genes. (C) KEGG enrichment analysis
   using upregulated genes. (D) KEGG enrichment analysis using
   downregulated genes. 3D hiPSC-CMs were treated with 17 mM ethanol for 5
   weeks and were compared with untreated cells (n=3 cultures).

   In addition, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway
   analysis revealed that EtOH treatment increased
   glycolysis/gluconeogenesis, HIF-1 signaling, fructose and mannose
   metabolism, adrenergic signaling in cardiomyocytes, carbon metabolism,
   and glucagon signaling ([107]Figure 5C, [108]Table S8). EtOH treatment
   also decreased focal adhesion, ECM-receptor interaction, protein
   digestion and absorption, dilated cardiomyopathy, Hippo signaling,
   TGF-beta signaling, hypertrophic cardiomyopathy, regulation of actin
   cytoskeleton, PI3K-Akt signaling, vascular smooth muscle contraction,
   platelet activation, signaling pathways regulating pluripotency of stem
   cells, Rap1 signaling, and growth hormone synthesis, secretion and
   action ([109]Figure 5D, [110]Table S9).

   Overall, our transcriptomics analysis revealed dysregulation of
   metabolic function, downregulation of ECM organization and disturbance
   in cytoskeletal organization and developmental pathways in EtOH-treated
   cells.

Chronic EtOH treatment of hiPSC-CMs alters metabolite levels identified by
metabolomics analysis

   To further understand the metabolic differences associated with chronic
   EtOH treatment, we compared metabolic features of cardiac spheroids
   treated with EtOH at 17 mM for 5 weeks with untreated cells by
   untargeted metabolomic analysis via liquid chromatography-mass
   spectroscopy (LC-MS). Of the 14,454 features identified, 548 were
   significantly altered (variable importance in projection [VIP] score >
   2) ([111]Figure 6A). By partial least squares discriminant analysis, a
   clear separation of metabolome was observed between chronic
   EtOH-treated cells and controls ([112]Figure 6B). This was also evident
   when significantly differentially expressed metabolites were displayed
   using heatmap ([113]Figure 6C).

Figure 6. Metabolomics identifies significantly perturbed metabolic features,
pathways and metabolites.

   Figure 6.
   [114]Open in a new tab

   (A) Volcano plot showing significantly perturbed metabolic features.
   (B) Partial least squares discriminant analysis (PLS-DA) score plot of
   group separation along the two principal components. (C) Heat map of
   significantly perturbed metabolic features. (D) Significantly perturbed
   metabolic pathways identified by Mummichog pathway analysis. The ratios
   of overlap_size to pathway_size are denoted on the graph. Red line
   indicates 0.05 p-value threshold. (E) Box plots of individual
   metabolites within the pathway of prostaglandin formation from
   arachidonic acid showing the mean intensity of each metabolite between
   the two groups. (F) Box plots of selected metabolites showing metabolic
   shift in long-term EtOH-treated cells. 3D hiPSC-CMs were treated with
   17 mM ethanol (EtOH) for 5 weeks and were compared with untreated cells
   (CTRL) (n=3 cultures).

   To identify significantly affected metabolic pathways, we performed
   pathway enrichment analysis on the selected significantly altered
   metabolic features. Pathway analysis performed on HILIC-pos data
   revealed enrichment of prostaglandin formation from arachidonate,
   androgen and estrogen biosynthesis and metabolism, glycerophospholipid
   metabolism, vitamin B12 metabolism, bile acid biosynthesis, and alanine
   and aspartate metabolism ([115]Figure 6D).

   We further analyzed individual metabolic features within enriched
   metabolic pathways using xMSannotator metabolic feature annotation tool
   ([116]Uppal et al., 2017) to characterize the regulation of each
   pathway. Prostaglandin formation pathway was upregulated as indicated
   by increased intensity of metabolites such as arachidonic acid,
   2-arachidonylglycerol, and leukotriene E4 in long-term EtOH-treated
   hiPSC-CMs compared with untreated cells ([117]Figure 6E). Furthermore,
   other metabolites not included in pathway enrichment analysis were also
   affected by long-term EtOH treatment in hiPSC-CMs. Specifically,
   essential fatty acid such as alpha-linolenic acid and various
   acylcarnitines such as 9-decenoylcarnitine, decanoylcarnitine and
   butenylcarnitine were decreased in long-term EtOH-treated cells
   compared with untreated cells ([118]Figure 6F).

   Since prostaglandin formation is associated with proinflammation, these
   findings suggest proinflammatory effects of EtOH treatment on
   hiPSC-CMs. In addition, since acylcarnitines are responsible for the
   transportation of fatty acids from the cytoplasm into the mitochondria
   to produce energy through fatty acid oxidation (β-oxidation), these
   results also suggest that long-term EtOH treatment of hiPSC-CMs
   resulted in decreased fatty acid oxidation, which is consistent with
   the observed mitochondrial dysfunction and decreased expression of
   CPT1B in long-term EtOH-treated hiPSC-CMs.

Integrated omics analysis revealed pathways associated with metabolic and
functional alterations in hiPSC-CMs upon chronic ethanol exposure

   Each of the omics dataset revealed unique functional alterations in
   hiPSC-CMs subjected to long-term EtOH exposure: transcriptomics data
   revealed increase in glycolysis, disturbance in cytoskeletal
   organization and HIF-1α pathway activation; metabolomics data revealed
   activation of proinflammatory pathways, hormonal responses and decrease
   in mitochondrial function and overall energy metabolism; and proteomics
   data from our previous publication suggested dysregulation of proteins
   involved with heart contraction and ECM regulation ([119]Liu et al.,
   2020b). To put all this information together and gain a comprehensive
   insight into changes that occurred after long-term EtOH exposure, we
   performed integrated omics analysis using xMWAS network algorithm
   ([120]Uppal et al., 2018).

   The integration of proteomics ([121]Liu et al., 2021), transcriptomics,
   and metabolomics data generated five unique protein-gene-metabolite
   communities or clusters by correlating 382 proteins, 1,701 genes, and
   14,454 metabolic features ([122]Figure 7A, [123]S2). Subsequently,
   functional enrichment analysis of omics feature set within each xMWAS
   community revealed their respective functional subclusters. More
   specifically, functional enrichment analysis performed on subclusters
   of proteomics features revealed alterations in pathways including focal
   adhesion and nucleotide sugar metabolism ([124]Table S10).
   Transcriptomic features showed alterations in metabolism, developmental
   pathways such as hippo signaling, TGF-beta signaling and MAPK
   signaling, as well as focal adhesion ([125]Table S11). Among these
   pathways, focal adhesion was identified by both transcriptomic and
   proteomic features ([126]Figures 7A, [127]S3).

Figure 7. Integrated omics analysis.

   Figure 7.
   [128]Open in a new tab

   (A) Visualization of the proteins, genes, and metabolites in clusters.
   Each cluster is indicated by a different color as shown. Squares are
   proteins, circles are genes, and triangles are metabolites, with
   positive correlations between them denoted in blue lines and negative
   correlations in red. GO terms next to square bullets are enriched by
   proteins and circle bullets by genes. (B) Putative mechanism of chronic
   EtOH-induced cardiotoxicity. 3D hiPSC-CMs cells were treated with 17 mM
   ethanol for 5 weeks and were compared with untreated cells (n=3
   cultures). Integrated omics analysis reveals key pathways affected by
   EtOH. The affected pathways are likely associated with dysregulated
   metabolism, increased mitochondrial damage, decreased contractility and
   abnormal Ca^2+ handling, which are highly relevant to clinical outcomes
   including atrial fibrillation and dilated cardiomyopathy.

   Hence, integrated omics analysis highlighted downstream specific
   pathways affected by long-term EtOH exposure that are associated with
   metabolic and functional alterations in hiPSC-CMs ([129]Figure 7B).

DISCUSSION

   By modeling chronic alcohol exposure-induced cardiotoxicity in
   hiPSC-CMs, we found that treatment with EtOH at physiologically
   relevant concentrations for an extended period caused severe
   detrimental effects on hiPSC-CMs. These effects were manifested in
   decreased mitochondrial membrane potential and mitochondrial content,
   decreased mitochondrial function, and altered expression of related
   genes. Our transcriptomics and metabolomics analyses identified key
   genes and metabolic features affected by chronic EtOH exposure, which
   will be useful in future studies to address relevant disease
   phenotypes. In addition, our integrated omics analysis revealed
   specific pathways affected by long-term EtOH exposure that were
   associated with metabolic and functional alterations in hiPSC-CMs.

   Our transcriptomics and GO enrichment analyses revealed that
   upregulated DEGs in EtOH-treated hiPSC-CMs were highly associated with
   glycolytic process and carbohydrate metabolic process as well as
   response to hypoxia. The genes related to glycolytic process such as
   PDK1, PDK3, GAPDH, LDHA, PGK1, GPI, and HK2 were increased in their
   expression levels in EtOH-treated hiPSC-CMs than in untreated cells;
   among them, PDK1 and PDK3 function as inhibitors of pyruvate
   dehydrogenase and activate glycolysis in TCA cycle. Similar metabolic
   remodeling on increasing glycolysis occurs during cardiac pathological
   response such as hypertrophy ([130]Allard et al., 1994), heart failure
   ([131]Masoud et al., 2014), and ischemia ([132]Hue et al., 2002), and
   upregulation of hypoxia pathway is indicative of increased reliance to
   glycolysis rather than using fatty acids for energy source
   ([133]Kierans and Taylor, 2021). In heart failure, increased glycolysis
   is likely due to a compensatory response to reduced mitochondrial
   function including oxidative metabolism and ATP production
   ([134]Lopaschuk et al., 2021). Such defects in mitochondrial function
   can be contributed by many factors including increased oxidative stress
   ([135]Lopaschuk et al., 2021). Consistently, we observed increased
   glycolysis, decreased mitochondrial function (in this study) and
   increased oxidative stress ([136]Rampoldi et al., 2019) in EtOH-treated
   hiPSC-CMs.

   In addition to changes in genes associated with glycolysis, hiPSC-CMs
   had reduced mitochondrial function upon long-term EtOH treatment. This
   conclusion was based on measurements of multiple mitochondrial
   features, including mitochondrial membrane potential detected by TMRM
   staining, the expression of mitochondrial protein Tom20 by
   immunocytochemistry and high-content imaging, and overall mitochondrial
   function as measured by oxygen consumption rate in the Seahorse
   MitoStress assay and gene expression. The decreased mitochondrial
   function in EtOH-treated hiPSC-CMs also supports the hypothesis that
   long-term EtOH treatment prevents normal cardiomyocyte development.
   Indeed, our RNA-seq analysis showed down-regulation of genes associated
   with heart development.

   Our RNA-seq analysis also indicated alteration of genes associated with
   inflammation in EtOH-treated hiPSC-CMs, which is consistent with
   inflammation induced by chronic EtOH consumption in many parts of the
   body such as the joints ([137]Barr et al., 2016), intestines
   ([138]Bishehsari et al., 2017), and brain ([139]Leclercq et al., 2017).
   For example, we found upregulation of T cell chemotaxis genes like
   WNK1, Wnt5A, ADAM10, and CXCL11 in EtOH-treated hiPSC-CMs. Upregulation
   of these genes provides a potential causal link to proinflammatory
   response upon chronic EtOH exposure. WNK1 is a negative regulator of
   NLRP3 inflammasome ([140]Mayes-Hopfinger et al., 2021). Wnt5A is
   involved in macrophage inflammatory activation in sepsis and is a
   target for anti-inflammatory mediators ([141]Pereira et al., 2008).
   ADAM10 is a mediator of vascular inflammation ([142]Yang et al., 2020).
   CXCL11 stimulates intestinal inflammation in response to microbial
   stimuli ([143]Liu et al., 2011), and also serves as immunostimulatory
   cytokine that is partially responsible for SARS-CoV-2 cytokine storm
   ([144]Callahan et al., 2021).

   Patients with excessive alcohol use have altered serum metabolomic
   signature when compared with healthy controls ([145]Liu et al., 2022).
   Excessive alcohol use can also induce pro-inflammatory response
   ([146]Crews et al., 2021). In our study, metabolomic profiling of
   EtOH-treated hiPSC-CMs revealed activation of proinflammatory pathways
   by arachidonic acid and its downstream metabolites. Chronic alcohol
   consumption has been shown to induce inflammation in many bodily
   contexts, as suggested by increased concentrations of arachidonic acid,
   prostaglandin, leukotriene, and other arachidonic acid metabolites
   ([147]Sonnweber et al., 2018). Indeed, arachidonic acid and its wide
   spectrum of derivatives such as prostaglandins and leukotrienes
   contribute to many metabolic pathways behind cardiovascular diseases,
   cancers, and inflammatory diseases ([148]Wang et al., 2021a, [149]Ma et
   al., 2022). Downregulation of alpha-linolenic acid is suggestive of
   defective cell membrane and cardiac function ([150]Fleming and
   Kris-Etherton, 2014). Downregulation of various acylcarnitine such as
   9-decenoylcarnitine suggests mitochondrial dysfunction, and this
   metabolite is also involved in incident atrial fibrillation and related
   to left atrium size and left ventricular mass ([151]Lind et al., 2021).

   Comprehensive analysis of omics data using xMWAS network algorithm
   identified distinct subclusters of differentially expressed genes and
   proteins in EtOH-treated hiPSC-CMs compared with untreated cells. The
   most frequently observed pathway associated with these genes and
   proteins was actin cytoskeleton regulation including focal adhesion.
   EtOH-treatment induced cytoskeleton dysregulation and changes in
   ECM-receptor interaction, focal adhesion and cytokine-cytokine receptor
   interaction. In addition, EtOH-treatment altered signaling pathways
   such as PI3K-Akt signaling, hippo signaling, and TGF-beta signaling.
   These changes in EtOH-treated hiPSC-CMs are consistent with
   observations from other in vitro and in vivo studies of chronic EtOH
   exposure ([152]Table S12). In a swine model, the myocardium of
   EtOH-treated animals had decreased expression of proadhesion proteins
   focal adhesion kinase, paxillin, integrin-β1, and TGF-beta compared
   with sucrose-treated animals ([153]Elmadhun et al., 2017). The impact
   of chronic EtOH exposure to actin cytoskeleton and ECM has also been
   documented in both cardiac cells ([154]Fernandez-Sola, 2020) and
   non-cardiac cells such as hepatocytes ([155]Shepard and Tuma, 2010). It
   is also known that dysregulation of actin cytoskeleton and ECM leads to
   inflammation and affects downstream signaling pathways ([156]Suresh and
   Diaz, 2021).

   In addition to the affected pathways identified by our integrated omics
   analysis, EtOH-treated hiPSC-CMs also had pathological phenotypes such
   as dysregulated metabolism and increased mitochondrial damage as
   observed in this study, and decreased contractility and Ca^2+ handling
   regulation as observed in previous studies ([157]Rampoldi et al., 2019,
   [158]Liu et al., 2021, [159]Liu et al., 2020b). These pathological
   phenotypes in EtOH-treated hiPSC-CMs are highly relevant to preclinical
   in vivo animal studies ([160]Steiner and Lang, 2017) and clinical
   outcomes including atrial fibrillation ([161]Voskoboinik et al., 2019)
   and dilated cardiomyopathy ([162]Fernandez-Sola, 2020) ([163]Figure 7B
   and [164]Table S12).

   Our findings also have implication for potential treatment for FASD.
   Increased mitochondrial damage observed in EtOH-treated hiPSC-CMs was
   also observed in a mouse model of FASD ([165]Nyquist-Battie and Freter,
   1988). In our hiPSC-CM model, upregulation of HIF-1α pathway together
   with decreased mitochondrial function likely contributed to increased
   glycolytic pathway identified by RNA-seq and decreased β-oxidation
   identified by metabolomics. Given that the inhibition of HIF-1α pathway
   can increase β-oxidation, ATP production, and mitochondrial function in
   cardiomyocytes ([166]Gentillon et al., 2019), targeting HIF-1α pathway
   may attenuate alcohol-induced cardiotoxicity.

   In conclusion, our study demonstrates that chronic EtOH treatment of
   hiPSC-CMs results in (1) decreased mitochondrial function, (2) altered
   metabolism including fatty acid oxidation, and (3) gene expression
   changes associated with increased glycolysis and impaired cardiac
   structural development. Our study also reveals changes in various
   signaling pathways (e.g., HIF-1α pathway) worthy of further
   investigation for potential therapeutic intervention in alcohol-induced
   cardiotoxicity including FASD.

Supplementary Material

   SUPINFO
   [167]NIHMS1884150-supplement-SUPINFO.pdf^ (1.3MB, pdf)

ACKNOWLEDGMENTS