WO2018232352A1 - In utero prevention of congenital heart disease by metabolic intervention - Google Patents

Abstract

A method of hindering the development of congenital heart disease (CHD) in a subject involves administering an inhibitor of nucleotide biosynthesis enzymes to the subject, such as by administration of the inhibitor to a pregnant female carrying the subject in utero. The inhibitor can be hydroxyurea, or a structurally similar hydroxamic acid derivative that targets histone deacetylases, matrix metalloproteinases or ribonucleotide reductase (RNR). One can also screen a pregnant female carrying the subject in utero for hyperglycemia or diabetes, and/or screen the subject for family history or a genetic indicator of CHD. A subject exhibiting structural evidence of CHD based on imaging can be treated with HU to mitigate severity of CHD. The invention additionally provides a method for hindering the development of other anomalies known to be associated with diabetic pregnancy, as well as methods for inducing maturation of cardiomyocytes.

Description

IN UTERO PREVENTION OF CONGENITAL HEART DISEASE BY METABOLIC

INTERVENTION

[0001] This application claims benefit of United States provisional patent application number 62/521,162, filed June 16, 2017, the entire contents of which are incorporated by reference into this application.

REFERENCE TO A SEQUENCE LISTING SUBMITTED VIA EFS-WEB

[0002] The content of the ASCII text file of the sequence listing named ^i,UCLA255WOU1_SL", which is 3 kb in size, was created on June 13, 2018, and electronically submitted via EFS-Web herewith the application. The sequence listing is incorporated herein by reference in its entirety. TECHNICAL FIELD OF THE INVENTION

[0003] The present invention relates to methods for preventing and treating congenital heart disease and for the preparation of cardiomyocytes, as well as compositions for performing these methods.

BACKGROUND OF THE INVENTION

[0004] With the advances in fetal diagnosis and surgical procedures, the number of patients with congenital heart disease (CHD) who survive childhood (adult CHD) is growing rapidly by nearly 5% per year. Maternal hyperglycemia is a common medical condition associated with a 2-5 fold increase in CHD. Currently, 60 million women of reproductive age (18-44 years old) worldwide, and approximately 3 million in the U.S.. have diabetes mellitus. This number is estimated to double by 2030, posing a huge medical and economic burden. Thus, prevention and treatment of CHD associated with diabetic pregnancy will be an urgent medical issue in the coming decades.

[0005] Others have attempted to decrease the risk of CHD in diabetic pregnancy by tightly controlling the blood glucose level of pregnant mothers. However, this approach has proven to be ineffective. There remains a need for methods for preventing and treating CHD.

SUMMARY OF THE INVENTION

[0006] In one embodiment, the invention provides a method of hindering the development of congenital heart disease (CHD) in a subject. The method comprises administering an inhibitor of nucleotide biosynthesis enzymes to the subject, such as by administration of the inhibitor to a pregnant female carrying the subject in utero. In one embodiment, the inhibitor is hydroxyurea (HU; also referred to as hydroxycarbamide), or a structurally similar hydroxamic acid derivative that targets histone deacetylases, matrix metalloproteinases or ribonucleotide reductase (RNR). Examples of other inhibitors of nucleotide biosynthesis enzymes include, but are not limited to, didox, tridox, motexafin gadolinium, fludarabine, cladribine, gemcitabine, tezacitabine, triapine, gallium maltolate, gallium nitrate. In another embodiment, the inhibitor is selected from the group consisting of: P-Rib-PP synthetase inhibitor (MRPP, etc.); APRTase inhibitor

(methotrexate, MMPR. etc.), GAR transformylase inhibitor (multitargeted antifolate (LY231514) , etc.); AICAR transformylase inhibitor (dmAMT, AG2009, etc.); IMP cyclohydorolase (purine nucleoside 5'-monophosphate derivatives, etc.); IMP dehydrogenase inhibitor (TAD, SAD, BAD, VX-497, etc.); Aspartate transcarbamylase inhibitor (PALA, etc.); DHO DHase inhibitor

(lapachol, etc.); ODCase inhibitor (pyrazofurin, etc.); and CTP synthetase inhibitor (3- Deazauridine, etc.).

[0007] In one embodiment, the method further comprises screening a pregnant female carrying the subject in utero for hyperglycemia or diabetes, and/or screening the subject for family history or a genetic indicator of CHD prior to administering HU or other inhibitor. In one embodiment, the screening for a genetic indicator of CHD comprises genetic testing of fetal cells in maternal blood (FCMB), DNA of fetal origin circulating in the maternal blood, preimplantation genetic diagnosis (PDG) during in vitro fertilization (IVF). transcervical retrieval of trophoblast cells, chorionic villus sampling, amniocentesis, or percutaneous umbilical cord blood sampling.

Representative examples of a genetic indicator of CHD include, but are not limited to, CHD7, ELN, GATA4, GATA6, GDF1, JAG1 , NKX2-5, NKX2-6, NOTCH 1 , NOTCH2, NR2F2, TBX1. TBX5, TBX20, SEMA3A, FOG2, Ras/Raf, and ZIC3. More extensive lists of genes associated with CHD are available, and can be useful in some situations. For example, Invitae Corporation (San Francisco, California) offers a testing panel of 82 genes, plus an additional seven preliminary evidence genes. Genetic indicators can be detected using sequencing and or detection methods such as. for example, next generation sequencing, and/or high-density targeted array to detect duplication or deletion.

[0008] In some embodiments, the method further comprises imaging the heart of the subject via echocardiogram, computed tomography (CT), and/or magnetic resonance imaging (MRI). In other embodiments, echocardiography, computed tomography (CT), and/or magnetic resonance imaging (MRI) are used to detect or confirm CHD. The subject already exhibiting structural evidence of CHD can be treated with HU in accordance with the invention to mitigate severity of CHD.

[0009] The invention additionally provides a method for hindering the development of other anomalies known to be associated with diabetic pregnancy, including caudal regression syndrome, situs inversus, ureter duplex, renal agenesis, and the like. The method comprises administering an inhibitor of nucleotide biosynthesis enzymes to the subject, such as by administration of the inhibitor to a pregnant female carrying the subject in utero.

[0010] In a typical embodiment, the administering step of the methods described herein is intravenous or oral. The HU or other inhibitor of nucleotide biosynthesis enzymes can be administered directly to the fetus, or in a typical embodiment, indirectly via administration to the pregnant mother. The HU is typically administered at a dose of 1-100 mg/kg maternal body weight. In some embodiments, the dose is about 5-50 mg/kg. In one embodiment, the HU is administered at a dose of about 10 mg/kg body weight.

[0011] Also provided is a method of screening candidate agents for treatment of or toxicity to developing cardiomyocytes. The method comprises culturing human embryonic stem cells (hESCs) and differentiating the hESCs into hESC-CMs (cardiac myocytes) in a chemically- defined medium. The method further comprises exposing the cultured hESC-CMS to a candidate agent (e.g., by contacting the cultured cells with the agent), and monitoring the expression of cardiac markers (e.g., TNNT2, NKX2-5) and other genes associated with cardiac muscle and function in response to the agent. Agents that increase the amount of cardiac markers, genes associated with cardiac muscle and function, mitochondrial maturity, and electrophysiological maturity are thereby identified as potential treatments to facilitate or protect cardiac myocyte development while agents that reduce expression of such markers are identified as potentially detrimental to developing cardiac myocytes.

[0012] The invention additionally provides a method of inducing maturation of cardiomyocytes in vitro. In one embodiment, the method comprises culturing stem cells in a chemically-defined medium containing less than 5 mM glucose for at least 14 days. In some embodiments, the stem cells are human embryonic stem cells. In some embodiments, the stem cells are human induced pi uri potent stem cells. In some embodiments, the stem cells are cultured in the medium for at least 5 days, 7 days, 10 days, 14 days, 20 days, or longer. In some embodiments, the glucose is 1 mM; in other embodiments, the glucose is 0 mM.

DESCRIPTION OF THE FIGURES

[0013] Figure 1. Illustrates that high yield hESC-CMs derived under chemically defined condition recapitulate developmental time course. (A) Representative flow cytometry analysis for a cardiac marker. MF20 (myosin), of hESC-CMs (H9) at day 14. (B) Bar chart of the percentage of MF20⁺ cardiomyocyte population by flow cytometry analysis in hESC-CMs generated from H9 (n=8) and UCLA4 (n=3, mean ± SD). (C) Images of phase-contrast and Tnnt2 (cTnT: cardiac Troponin T) immunofluorescent staining of hESC-CMs at day 28 (scale bar=50 pm). (D) Time course of gene expression profile by RNA-seq using MSigDB is shown in the heatmap. hESC; human embryonic stem cell, hMP; human mesodermal precursor, hCP; human cardiac progenitor, hCM14 or 28; human ESC-derived cardiomyocyte differentiation day 14 or 28 (n=3). (E) The heatmap of representative cardiac genes from (D) shows progressive upregulation over the time.

[0014] Figure 2. Illustrates how glucose reduction promotes maturation of hESC-CMs. (A)

Schematic representation of experimental regimen. hESC-CMs are differentiated in the medium containing 25mM glucose until day 14, when -90% of the cells already are MF20^*. Cells are analyzed at day 28 unless otherwise specified. (B) Relative mRNA expression of TNNT2, NKX2-5, and PPARGC1A by qPCR. All these markers are upregulated in hESC-CMs in glucose-deprived conditions (n=3, mean ± SD, p-value by one-way ANOVA test). (C) Pathway analysis of differentially expressed genes in 0 mM glucose (top left panel), and of ones enriched in hESC-CMs in 25 mM glucose (bottom left) based on RNA-seq data is shown. The heatmap (right panel) shows the relative expression of representative cardiac genes compared between hESC-CM cultured with 25 mM glucose or without glucose. (D) Assessment of mitotic activity by pH3 immunostaining. Representative photomicrographic images of three independent experiments. (E) Assessment of mitotic activity. Shown are representative EdU flow cytometry (left) and quantitation of %EdU^* cardiomyocytes (right. n=3, mean ± SD, p < 0.01 by t-test). (F) Assessment of the maturity of cardiomyocytes by stainings for MitoTracker (mitochondrial content) and a-Actinin. Representative photomicrographic images of three independent experiments. (G) Assessment of the maturity of cardiomyocytes by flow cytometry for MF20 and MitoTracker. Representative images of at least three independent experiments. (H) Assessment of the mitochondrial contents by quantitative PCR for mitochondrial and nuclear DNA (n=4, mean ± SD, p < 0.05 by t-test). (I) Assessment of the cell size by forward scatter (FSC) from flow cytometry data. At least 10,000 cells were measured per sample. Representative histogram from 3 flow cytometry data for each group (left) and the geometrical means of FSC (right n=3, mean ± SD, p < 0.05 by t-test). (J) Expression of cardiac markers extracted from RNA-seq data. (n=3. mean ± SD, by t-test). (K) Sarcomere length analysis. Representative trace of a-actinin and cellular architecture (left) and analysis of biological triplicates of 20-30 cells measurements. (n=3, mean ± SD, p=n.s. by t-test). (L) Impact of 2-Deoxy-D-glucose (2-DG), a competitive inhibitor of glucose, on the MitoTracker and MF20 levels measured by flow cytometry.

Representative contour plots (left) and quantitation from three independent experiments (right) are shown. (n=3, mean ± SD, p<0.01 by t-test).

[0015] Figure 3. Illustrates that glucose deprivation promotes the functional maturation of hESC-CM. (A) Representative photomicrographic images of mitochondrial membrane potential assay using JC-1 dye in hESC-CMs cultured in the presence (left) and absence (right) of glucose. Note the elongated mitochondria in the hESC-CMs cultured in 0 mM glucose. (B) Flow cytometry analyses of JC-1 of hESC-CMs cultured in different concentrations of glucose and pyruvate. Representative flow cytometry profile (left) and the quantitation of the intensity of JC-1 aggregates (right; mean intensity ± rSD, p<0.01 by one-way ANOVA test). (C) Changes in intracellular lactate level measured as YFP/CFP ratio with Laconic²⁵, a fluorescence resonance energy transfer (FRET)-based biosensor of lactate, in hESC-CMs cultured in the presence (upper panel) or absence (lower panel) of glucose. Increase in intracellular lactate level is shown downwards. Note that, in hESC-CMs cultured in the absence of glucose, addition of pyruvate (Pyr) does not lead to the increase in intracellular lactate. A representative of 2 independent experiments. (D)-(H). Oxygen consumption rate (OCR) measured with a Seahorse analyzer (D). hESC-CMs cultured in the absence of glucose show significantly greater mitochondrial respiration (E), ATP-linked respiration (F), and maximum mitochondrial respiration capacity (H) without any difference in proton leak (G) suggesting that glucose-deprivation potentiates OXPHOS (n = 19, each, mean ± SD, p < 0.01 by t-test). (I) Calcium transient assay of hESC-CMs cultured in the presence/absence of glucose. Representative waves, Vmax, AF/FO, and time to 50% decay are shown (n = 10 (G25) and 11 (GO), p values by t-test). (J) Motion speed analyses by digital image correlation. The first (·) and the second peak (▲) of the duplex represent the contraction and the relaxation speed, respectively (left). Contraction velocity, relaxation velocity, and beat rate are shown (n = 12 (G25) and 9 (GO), mean ± SD, p values by t-test). (K) Electrophysiological analyses of the maturity of hESC-CMs by multi- electrode array (MEA). Maximum dv/dt of field potential obtained from three independent measurements of properly recorded channels among 120 electrodes with representative trace of recoreded field potential; recorded before starting glucose reduction in the media, and after 10 days of glucose reduction, (n = 3, each group, mean ± SD, p-value by t-test). (L)

Representative bright field photomicrographic image of the monolayer cells plated onto the MEA chip. (M) Snapshots of recorded field potential of hESC-CM culture with or without glucose.

[0016] Figure 4. Illustrates how pentose phosphate pathway inhibits cardiac maturation. (A) Heatmap presentation of the metabolomics analysis of hESC-CMs cultured in the

presence/absence of glucose. Note the decrease in the metabolites in purine metabolism, pyrimidine metabolism, PPP in glucose-deprived condition (n = 3, each). See also (J). (B) ATP levels of hESC-CMs under G25 and GO conditions (n = 3, each, mean ± SD, p = n.s. by t-test). (C) Schematic of experimental regimen for chemical inhibition of glucose metabolic pathways. hESC-CMs are cultured in the medium containing 4 different glucose levels and chemical. See also (K). (D) Relative mRNA expression of TNNT2 and NKX2-5 in different concentration of glucose and 2-DG, a competitive inhibitor of glucose. 2-DG restored cardiac maturation under glucose presence. (E)-(H). Relative mRNA expression of TNNT2 in 0-25 mM glucose with chemical inhibitors for glucose metabolic pathways; (E) 0-10 μΜ 3PO (3-(3-pyridinyl)-1-(4- pyridinyl)-2-propen-1-one. a PFKFB3 inhibitor), (F) 0-5 mM sodium oxamate (NaOX; an LDH inhibitor), (G) 0-5 μΜ 6AN (6-Aminonicotinamide, a G6PD inhibitor), and (H) 0-50 μΜ DHEA (Dehydroepiandrosterone, a G6PD inhibitor). n=3, each, mean ± SD. p-value by one-way ANOVA. See also (L)-(Q). (I) Metabolomics analyses by mass spectrometry. Top diagram shows the comparison of metabolomics analysis by mass spectrometry of hESC-CMs cultured in the presence or absence of glucose. Each hexagon indicates a metabolite decreased (blue in online version; hatched herein) and increased (red in online version; shaded herein) in the absence of glucose. Green frame (see online version) indicates statistically significant change. Note the decrease in the metabolites in purine metabolism, pyrimidine metabolism, PPP, hexosamine pathway, and glycolysis, (n = 3, each). The heatmap of metabolite levels measured by mass spectrometry of hESC, hCM14, G+ (hCM28 cultured in 25 mM glucose), and G- (hCM28 cultured without glucose). (J) RNAi knockdown of glucose metabolic enzymes. Relative TNNT2 expression after RNAi knockdown is shown in the upper bar graph. RNAi targeting scramble, HK1 , RRM2, RRM2B, G6PD, and PFK were transfected by lipofection for 48 hours followed by 7 days' incubation with 25 mM glucose (n = 3, each group, mean ± SD, p<0.05 between RRM2B and scramble by t-test). Bottom panel indicates the RNAi knockdown efficiency for the glucose metabolic enzymes. (K) Schematic illustration showing that the pentose phosphate pathway inhibits cardiac maturation. Summary of the impact of the glucose metabolism inhibitors on the cardiac maturity is shown. Inhibitors tested are shown in box (see red box in online version), and the effective inhibitors are highlighted (shaded boxes; highlighted yellow in online version). There is a clear trend that inhibition of PPP increases the maturity of hESC-CMs. (L)-(Q) Summary of the impact of the glucose metabolism inhibitors on the cardiac maturity. (L) Relative mRNA expression of TNNT2 and NKX2-5 in conditions of 0-25 mM of glucose and 0-25 mM 2-DG. No data were available for the samples with 2-DG concentration higher than its glucose level due to the cytotoxicity of 2-DG. 2-DG dose-dependently restored cardiac maturation under glucose presence. A representative of three independent experiments. (M) Relative mRNA expression of TNNT2 in 0-25 mM glucose and 0-10 μΜ 3PO (a PFK inhibitor). 3PO failed to restore the effect of glucose deprivation. A representative of three independent experiments. (N) Relative mRNA expression of TNNT2 in 0-25 mM of glucose and 0-5 mM sodium oxamate (NaOX; an LDH inhibitor). Sodium oxamate failed to restore the effect of glucose deprivation. A representative of three independent experiments. (O) Relative mRNA expression of TNNT2 in 0-25 mM of glucose and 0-5 μΜ 6AN (a G6PD inhibitor). 6AN dose- dependently restore the effect of glucose deprivation, suggesting that pentose phosphate pathway plays a critical role in the glucose-dependent inhibition of cardiac maturation. A representative of three independent experiments. (P) Relative mRNA expression of TNNT2 in 0- 25 mM of glucose and 0-50 μΜ DHEA (a G6PD inhibitor). DHEA dose-dependently restore the effect of glucose deprivation, suggesting that pentose phosphate pathway plays a critical role in the glucose-dependent inhibition of cardiac maturation. A representative of three independent experiments. (Q) ROS level is measured by the signal intensity of DCFDA. Glucose reduction does not cause an increase in ROS level despite higher mitochondrial function of hESC-CMs cultured in low glucose medium, p-value by one-way ANOVA (n=4).

[0017] Figure 5. Nucleotide metabolism regulates cardiomyocyte maturation. (A)

Photomicrographs of glucose-deprived hESC-CMs cultured in the absence (a, b) or presence (c, d) of 25 mM uridine, and stained for pH3 (mitosis marker). The addition of uridine restored the proliferative activity even in the absence of glucose. A representative of three independent experiments. (B) Proliferation rate as pH3⁺ cells /a-Actinin⁺ of the stained images of (A) are shown in percentage (n=3, mean ± SD, p < 0.01 by t-test). (C) Relative mRNA expression of TNNT2 in hESC-CMs in 25 mM or 0 mM glucose with 0 or 25 mM uridine. Uridine dose- dependently inhibited the TNNT2 expression level in glucose-deprived condition (n=3, mean ± SD, p<0.0005 by one-way ANOVA test). See also (F). (D) Relative mRNA expression of TNNT2 and NKX2-5 in hESC-CMs cultured in 0-25 mM of glucose in the presence or absence of thymidine. Thymidine block induces the levels of TNNT2 and NKX2-5 (n=3, mean ± SD, p<0.01 by t-test between with or without 25 mM Thymidine with Glucose 25 mM. See also (G). (E) Relative mRNA expression of TNNT2 and NKX2-5 in hESC-CMs cultured in 0-25 mM of glucose and 0-2 mM hydroxyurea (HU, a ribonucleotide reductase inhibitor). HU dose- dependently induced the expression of TNNT2 and NKX2-5 at 1, 5. and 25 mM glucose. (n=3. mean ± SD, p-value by one-way ANOVA test. See also (H). (F)-(H) Nucleotide inhibits hESC- CM maturation. (F) The relative expression of TNNT2 mRNA measured by qPCR in hESC-CMs in 16 different conditions of 0-25 mM of glucose and 0-25 mM uridine. Uridine recapitulates the effect of glucose in dose-dependently inhibiting the TNNT2 expression level. A representative of three independent experiments. (G) The relative mRNA expression of TNNT2 and NKX2-5 measured by qPCR in hESC-CMs cultured in 0-25 mM of glucose in the presence or absence of thymidine. Thymidine block induces the levels of TNNT2 and NKX2-5. This is a representative graph of three independent experiments. (H) The relative mRNA expression of TNNT2 and NKX2-5 measured by qPCR in hESC-CMs cultured in 0-25 mM of glucose and 0-2 mM hydroxyurea (HU, a ribonucleotide reductase inhibitor). HU dose-dependently induced the expression of TNNT2 and NKX2-5 at 1, 5, and 25 mM glucose. A representative of three independent experiments. (I)-(J) Nucleotide deprivation, not cell cycle block, is the primary inducer of cardiomyocyte maturation. (I) The impact of the CDK4/6 inhibitor on the mRNA relative expression of TNNT2 and NKX2-5 measured byqPCR. Cell cycle block by CDK4/6 inhibitor did not significantly induce the expression of cardiac markers. A representative of three independent experiments. (J) The impact of Paclitaxel on the mRNA relative expression of TNNT2 and NKX2-5 measured by qPCR. Cell cycle block by Paclitaxel did not significantly induce the expression of cardiac markers. A representative of three independent experiments.

[0018] Figure 6. Developmental time course of cardiac glucose uptake measured by ^1SF-FDG accumulation. The radioactivity of the entire heart was measured by γ-counter after tail vein i.v. (fetus) or i.p. (neonates) injections. Values were normalized to heart weight and total body signal (heart values/(heart weight * total body values). Note a drastic decrease in glucose uptake at late gestational and neonatal stages (n = 6, each. p<0.0001 by one-way ANOVA test).

[0019] Figure 7. Hyperglycemia in utero promotes proliferation and inhibits maturation of cardiomyocytes in vivo. (A) Diagram illustrating in vivo analysis of the impact of maternal hyperglycemia on fetal heart development using diabetic mouse model (Akita). (B),(C) Cell cycle analyses of fetal and neonatal cardiomyocytes from normal and diabetic pregnancy. Tnnt2- positive cardiomyocytes from diabetic Akita mothers show a higher percentage of cells in S- phase at both E16.5 and P0. (n = 7, each, mean ± SD, P value by t-test). (D),(E) Double immunostaining for phospho-histone H3 (pH3. green in online version) and aActinin (red in online version) of the heart from normal and diabetic pregnancy at E16.5, P1 and P7

(D). %pH3⁺ cells within a-Actinin* cardiomyocytes (CM) from normal (WT) and diabetic (Akita) pregnancy at E16.5 and P1 (E). At least 10,000 cardiomyocytes were counted for each of 5 hearts, (n = 7, each, mean ± SD, p-value by t-test). (F) qPCR analysis for 7/?/?/2expression in the hearts from normal and diabetic pregnancy. The expression level is normalized to control at each stage (n=3, each. P value by t-test). (G) Histological analysis of P1 hearts from normal (WT) and diabetic (Akita) pregnancy. Ventricular wall thickness (RV, right ventricle, LV left ventricle) and interventricular septum (IVS) thickness were histologically analyzed, (scale bar = 200 Mm). (n=5 and 4 for WT and Akita, respectively, mean ± SD. p-value by t-test). (H) Cell size analysis of the cells isolated from P1 hearts from normal (WT) and diabetic (Akita) pregnancy using FSC (forward scatter) by flow cytometry. At least 25,000 cells were measured per sample. Representative histogram from 3 flow cytometry data for each group (left) and the geometrical means of FSC (right, n=3, p<0.05 by t-test).

[0020] Figure 8. Therapeutic effect of HU (hydroxyurea) on the phenotype of congenital heart disease in the neonates from diabetic Akita mother mouse. Wild type male was crossed to wild type control mothers and diabetic mothers (Akita) injected with HU or vehicle solution. HU was injected at 10mg/kg i.p. daily from E12.5 until birth. Left, representative H&E staining of the neonatal hearts from wild type (wt) mother, diabetic mother, and diabetic mother treated with HU. Right, quantitation of heart wall thickness, demonstrating that HU treatment reversed the hypertrophic phenotype. Shown are the measurements of right ventricular wall thickness. N=8- 10, each.

DETAILED DESCRIPTION OF THE INVENTION

[0021] As described herein, glucose inhibits the maturation of the fetal heart through enhancing the synthesis of nucleotides. The inventors have successfully prevented CHD in a mouse model of CHD. The invention is based on the unexpected discovery that blocking nucleotide synthesis can be used as a strategy for preventing CHD in diabetic mothers. In the murine model of CHD associated with diabetic pregnancy described in the Examples herein, injection of a non-toxic dose of hydroxyurea (HU) to diabetic mothers reduces the incidence of CHD in newborns. The methods described herein are particularly beneficial to subjects at high risk for fetal heart disease, including but not limited to, subjects having positive prenatal genetic testing and family history of CHD, as well as gestational diabetes.

Definitions

[0022] All scientific and technical terms used in this application have meanings commonly used in the art unless otherwise specified. As used in this application, the following words or phrases have the meanings specified. [0023] As used herein, "low glucose" and "glucose restriction'' refers, in the context of culturing stem cells, for example, to a glucose concentration below 5 mM. In some embodiments, the glucose concentration is 0 mM.

[0024] As used herein, "abnormality" in the context of genes and genetic indicators of disease or disorder refers to a detectable alteration in the copy number or quality of a gene as compared to its normal counterpart, as it would appear in a healthy subject. Representative examples of genetic abnormalities include deletions, mutations, and duplications.

[0025] As used herein, to "prevent" or "protect against" a condition or disease means to hinder, reduce or delay the onset or progression of the condition or disease.

[0026] As used herein, "hindering" or "mitigating" refers to impeding, preventing, reducing, or otherwise minimizing an adverse outcome.

[0027] As used herein, "a" or "an" means at least one, unless clearly indicated otherwise. Methods

[0028] In one embodiment, the invention provides a method of hindering the development of congenital heart disease (CHD) in a subject. The method comprises administering an inhibitor of nucleotide biosynthesis enzymes to the subject, such as by administration of the inhibitor to a pregnant female carrying the subject in utero. In one embodiment, the inhibitor is hydroxyurea (HU; also referred to as hydroxycarbamide), or a structurally similar hydroxamic acid derivative that targets histone deacetylases, matrix metalloproteinases or ribonucleotide reductase (RNR). Examples of other inhibitors of nucleotide biosynthesis enzymes include, but are not limited to, didox, tridox, motexafin gadolinium, fludarabine, cladribine, gemcitabine, tezacitabine, triapine, gallium maltolate, gallium nitrate. In another embodiment, the inhibitor is selected from the group consisting of: P-Rib-PP synthetase inhibitor (MRPP, etc.); APRTase inhibitor

(methotrexate. MMPR, etc.). GAR transformylase inhibitor (multitargeted antifolate (LY231514) , etc.); AICAR transformylase inhibitor (dmAMT, AG2009, etc.); IMP cyclohydorolase (purine nucleoside 5'-monophosphate derivatives, etc.); IMP dehydrogenase inhibitor (TAD, SAD, BAD, VX-497, etc.); Aspartate transcarbamylase inhibitor (PALA, etc.); DHO DHase inhibitor

(lapachol, etc.); ODCase inhibitor (pyrazofurin, etc.); and CTP synthetase inhibitor (3- Deazauridine, etc.).

[0029] In one embodiment the method further comprises screening a pregnant female carrying the subject in utero for hyperglycemia or diabetes, and/or screening the subject for family history or a genetic indicator of CHD prior to administering HU or other inhibitor. In one embodiment, the screening for a genetic indicator of CHD comprises genetic testing of fetal cells in maternal blood (FCMB), DNA of fetal origin circulating in the maternal blood, preimplantation genetic diagnosis (PDG) during in vitro fertilization (IVF), transcervical retrieval of trophoblast cells, chorionic villus sampling, amniocentesis, or percutaneous umbilical cord blood sampling. Representative examples of a genetic indicator of CHD include, but are not limited to, CHD7, ELN, GATA4, GATA6, GDF1, JAG1 , NKX2-5, NKX2-6, NOTCH 1 , NOTCH2, NR2F2, TBX1. TBX5, TBX20, SEMA3A, FOG2, Ras/Raf, and ZIC3. More extensive lists of genes associated with CHD are available, and can be useful in some situations. For example, Invitae Corporation (San Francisco, California) offers a testing panel of 82 genes, plus an additional seven preliminary evidence genes. Genetic indicators can be detected using sequencing and or detection methods such as, for example, next generation sequencing, and/or high-density targeted array to detect duplication or deletion.

[0030] In some embodiments, the method further comprises imaging the heart of the subject via echocardiogram, computed tomography (CT), and/or magnetic resonance imaging (MRI). In other embodiments, echocardiography, computed tomography (CT), and/or magnetic resonance imaging (MRI) are used to detect or confirm CHD. The subject already exhibiting structural evidence of CHD can be treated with HU in accordance with the invention to mitigate severity of CHD.

[0031] The invention additionally provides a method for hindering the development of other anomalies known to be associated with diabetic pregnancy, including caudal regression syndrome, situs inversus, ureter duplex, renal agenesis, and the like. The method comprises administering an inhibitor of nucleotide biosynthesis enzymes to the subject, such as by administration of the inhibitor to a pregnant female carrying the subject in utero.

[0032] In a typical embodiment, the administering step of the methods described herein is intravenous or oral. The HU or other inhibitor of nucleotide biosynthesis enzymes can be administered directly to the fetus, or in a typical embodiment, indirectly via administration to the pregnant mother. The HU is typically administered at a dose of 1-100 mg/kg maternal body weight. In some embodiments, the dose is about 5-50 mg/kg. In one embodiment, the HU is administered at a dose of about 10 mg/kg body weight.

[0033] Also provided is a method of screening candidate agents for treatment of or toxicity to developing cardiomyocytes. The method comprises culturing human embryonic stem cells (hESCs) and differentiating the hESCs into hESC-CMs (cardiac myocytes) in a chemically- defined medium. The method further comprises exposing the cultured hESC-CMS to a candidate agent (e.g., by contacting the cultured cells with the agent), and monitoring the expression of cardiac markers (e.g., TNNT2, NKX2-5) and other genes associated with cardiac muscle and function in response to the agent. Agents that increase the amount of cardiac markers, genes associated with cardiac muscle and function, mitochondrial maturity, and electrophysiological maturity are thereby identified as potential treatments to facilitate or protect cardiac myocyte development while agents that reduce expression of such markers are identified as potentially detrimental to developing cardiac myocytes. [0034] The invention additionally provides a method of inducing maturation of cardiomyocytes in vitro. In one embodiment, the method comprises culturing stem cells in a chemically-defined medium containing less than 5 mM glucose for at least 14 days. In some embodiments, the stem cells are human embryonic stem cells. In some embodiments, the stem cells are human induced pluripotent stem cells. In some embodiments, the stem cells are cultured in the medium for at least 5 days. 7 days. 10 days, 14 days, 20 days, or longer. In some embodiments, the glucose is 1 mM; in other embodiments, the glucose is 0 mM.

Kits

[0035] The invention provides kits comprising materials, including pharmaceutical compositions, for use with the methods described herein, such as HU or other inhibitor of nucleotide biosynthesis enzymes, prepared in unit dosage form, and optionally, one or more suitable containers containing compositions of the invention. The kit can optionally include a buffer, excipient or other materials useful for administering the composition of the invention.

Dosage and Routes of Administration

[0036] HU has been used for cancer chemotherapy. When administered orally to a pregnant woman, HU passes the placenta. Although high-dose HU causes fetal anomaly (500mg/kg in rats and rabbits), the inventors have found that low doses are safe. In our mouse study, HU administered to pregnant females successfully prevented congenital heart disease at 10mg/kg, showing that it can be safely used with pregnant mothers, and in human subjects.

[0037] In a typical embodiment, the administering is intravenous or oral. The HU is typically administered at a dose of 1-100 mg/kg body weight, based on the maternal body weight. In some embodiments, the dose is about 5-50 mg/kg. In one embodiment, the HU is administered at a dose of about 10 mg/kg body weight. In other embodiments, the dose is about 5, 15, 20, 25, 30. 40, 45, or 50 mg/kg body weight. The dose and treatment schedule can be adjusted by the treating physician in view of other factors specific to a given patient and the developmental stage of the fetus.

[0038] In a typical embodiment, the treatment is administered daily from the point of diagnosis until delivery. Adjustments and variations to this protocol can be determined by the treating physician. Likewise, other inhibitors of nucleotide biosynthesis enzymes can be administered using a comparable dosage that is low enough to avoid toxicity, yet sufficient to promote cardiomyocyte maturation.

-EXAMPLES

[0039] The following examples are presented to illustrate the present invention and to assist one of ordinary skill in making and using the same. The examples are not intended in any way to otherwise limit the scope of the invention. [Example 1 : Glucose inhibits cardiac myocyte maturation through nucleotide biosynthesis

[0040] The heart switches its energy substrate from glucose to fatty acids at birth, and maternal hyperglycemia is associated with congenital heart disease. However, little is known about how blood glucose impacts heart formation. Using a chemically-defined human pluripotent stem cell- derived cardiomyocyte differentiation system, this Example demonstrates that high glucose inhibits the maturation of cardiomyocytes at genetic, structural, metabolic and

electrophysiological levels via nucleotide biosynthesis through the pentose phosphate pathway. Even though blood glucose level in embryos is stable in utero during normal pregnancy, glucose uptake by fetal cardiac tissue was drastically reduced at late gestational stages. Perturbation of glucose dynamics during gestation in a murine model of diabetic pregnancy promoted mitosis and inhibited maturation of fetal cardiomyocytes. Thus, the metabolic switch is not only to meet the energy demand but also to induce a genetic program to facilitate cardiac maturation, providing a possible mechanistic basis for the congenital heart disease in diabetic pregnancy.

[0041] Congenital heart disease (CHD) is the most common type of birth defect affecting 0.8% of human live births (Fahed et al., 2013). Although genetic factors play a significant role in the development of CHD, current genomic technologies including exome sequencing and SNP arrays have provided a genetic diagnosis for only 11% of the probands (Pediatric Cardiac Genomics et al., 2013), highlighting the crucial role of non-genetic contributors.

[0042] Among non-genetic factors influencing the fetal heart, maternal hyperglycemia is the most common medical condition associated with 2-5 fold increase in CHD independent of genetic contributors (Centers for Disease, 1990; Simeone et al., 2015; Yogev and Visser, 2009).

Diabetic pregnancy is often accompanied by maternal complications including vasculopathy, neuropathy, and insulin resistance, which potentially affect the fetal cardiac formation indirectly.

These systemic complications are often subclinical, hindering the dissection of the

pathomechanism of CHD in diabetic pregnancy. Thus, despite the established association between maternal hyperglycemia and malformation of the fetal heart, little is known about how glucose levels impact cardiomyocyte development and how hyperglycemia affects the heart formation in diabetic pregnancy (Gaspar et al., 2014).

[0043] The metabolic environment is one potential non-genetic determinant of cell proliferation and differentiation. Cells display distinct metabolic characteristics depending on the

differentiation stage (Carey et al., 2015; Tohyama et al., 2016; Wang et al., 2009), and the fuel type of the cells serves not merely as a source of energy but also as a critical regulator of self- renewal and differentiation of stem/progenitor cells (Harris et al., 2013; Oburoglu et al., 2014; Shiraki et al., 2014; Shyh-Chang et al., 2013). However, little is known about the mechanism. Cardiomyocytes shift their energy substrate during late embryonic and neonatal stages

(Makinde et al., 1998). Glucose is the major energy source during the early developmental stages. Oxidative phosphorylation is low until E10.5 of developing rodent hearts, and rapidly increases between E10.5 and 14.5 (Cox and Gunberg, 1972). This coincides with the rapid maturation of the mitochondrial structure in the embryonic cardiomyocytes (Mackler et al., 1971). Shortly after birth, fatty acid oxidation becomes the predominant source of ATP production in order to meet the high energy demand of the maturing heart (Warshaw and Terry, 1970). These metabolic changes occur as a consequence of the changes in the expression of metabolic enzymes and transporters. However, it remains unclear whether and how these metabolic changes, in turn, regulate the cardiac differentiation program.

[0044] This Example shows that glucose not only induces cardiomyocyte proliferation, but also inhibits cardiomyocyte maturation, using in vitro human embryonic stem cell-derived

cardiomyocytes (hESC-CMs) and an in vivo murine diabetic model. Glucose can be

metabolized in multiple catabolic and anabolic pathways including glycolysis, oxidative phosphorylation, the pentose phosphate pathway (PPP) and the hexosamine biosynthesis pathway. Chemical screening revealed that the pro-mitotic/anti-maturation effect of high glucose is regulated by glucose-derived deoxynucleotide biosynthesis through PPP. In vivo

measurement of ¹⁸F-FDG accumulation revealed that glucose uptake is drastically suppressed during late gestational and early postnatal stages. Exposure to high blood glucose in a murine model of diabetic pregnancy resulted in higher mitosis and delayed maturation of fetal cardiomyocytes in vivo. Together, the data presented herein uncover how the dynamics of glucose metabolism impacts the late embryonic cardiogenesis.

Methods

[0045] Mouse and cell lines Wild-type and Akita mice were maintained on C57BLJ6 background according to the Guide for the Care and Use of Laboratory Animals published by the US National Institute of Health (NIH Publication No. 85-23, revised 1996). Housing and experiments were performed according to the Institutional Approval for Appropriate Care and Use of Laboratory Animals by the UCLA Institutional Animal Care and Use Committee (Protocol #2008-127-07). H9 (WA09) and UCLA4 (UCLA stem cell core) hESC lines were maintained as described before (Arshi A. et al 2013). Authentication of hESCs was achieved by confirming the expression of pluripotency genes and protein markers. hESCs were routinely verified as mycoplasma-free using a PCR-based assay. hESCs were grown and differentiated in a chemically-defined condition (Minami et al., 2012; Young et al., 2016; Zhu et al., 2017). Usage of all the human ES cell lines is approved by the UCLA Embryonic Stem Cell Research

Oversight (ESCRO) Committee and the Institutional Review Boards (IRB) (approval #2009-006- 04).

Table 1: hESC-CM Differentiation Medium

[0046] RNA-seq and data analyses For RNA-seq in Figures 1 D, 1 E and 2C, RNA was extracted from hESC, hMP, hCP, hCM14, hCM28, hCM28(glucose 25 mM) and hCM28(glucose 0 mM) using TRIZOL (TheroFisher) and RNeasy kit (QIAGEN). 500 ng of DNasel-treated RNA was used as input material for library preparation using the lllumina TruSeq mRNA kit (lllumina, RS-122-2001), according to manufacturer's instructions. Final libraries were sequenced as single-end 50bp on the lllumina HiSeq2000 platform (GSE84814). Libraries for RNA-Seq in Figures 1D, 1E and 2C were prepared with KAPA Stranded RNA-Seq Kit. The workflow consists of mRNA enrichment, cDNA generation, end repair, A-tailing, adaptor ligation, strand selection and PCR amplification. Different adaptors were used for multiplexing samples in one lane. Sequencing was performed on lllumina HiSeq 3000 for a paired end 2x150 run (GSE84815). Data quality check was done on lllumina SAV. De-multiplexing was performed with lllumina Bcl2fastq2 v 2.17 program. Short read sequences generated from lllumina Sequencer were aligned to the UCSC human reference genome hg19 downloaded from support.lllumina.com (see /sequencing/sequencing_software/igenome.html) using TopHat from the Tuxedo Tools. The average overall read mapping rate reached over 82 (average=82.43) percent. The output resulted in BAM files (Binary Sequence Alignment/Map format). These outputs contain information for assigning location and quantifying the short-read alignments obtained from the RNA-seq samples. This is necessary for downstream analyses such as annotation, transcript abundance comparison and polymorphism detection. Counts/gene expression matrices were generated using HTSeq which quantified the reads per transcript. The expression matrices were log transformed and normalized using the riog function in Deseq2 package. The normalized gene expression matrices were used as input for SaVanT (Signature Visualization Tools) which allowed for the visualization of molecular signatures directly related to heart development as seen in Figures 1D, 1E and 2C. [0047] Flow cytometry hESC-CMs and mouse embryonic hearts were washed three times with PBS and incubated at 37°C in a dissociation enzyme solution with occasional pipetting to a single-cell suspension. The enzyme solution contained 1% Penicillin/Streptomycin

(ThermoFisher, 15140-122), 10%Fetal Bovine Serum (Hyclone), collagenase 2 mg/ml

(Worthington, CLS-2), dispase 0.25mg/ml (Gibco, 17105-041), DNAase I (ThermoFisher) in PBS. The cells were analyzed with the following antibodies: MF20 (mouse, 1:100, Hybridoma Bank), Tnnt2 (rabbit, 1:250, Sigma-Aldrich), MitoTracker Orange CMTMRos (ThermoFisher), JC-1 (Abeam), and EdU (100 μΙ of 10 mM EdU solution per 10 g of mouse, ThermoFisher). For MF20 and Tnnt2, FITC-conjugated anti-mouse IgG secondary antibody (BD Biosciences) was used. Stained cells were analyzed by a flow cytometer (LSRII, BD Biosciences). Data analysis was performed using FACSDiva (BD Biosciences).

[0048] Immunocytochemistry and morphological analysis Cells were fixed with 4% paraformaldehyde, blocked for an hour with 5% normal goat serum, incubated with mouse alpha actinin antibody (Sigma), followed by Alexa fluor 488-conjugated secondary antibody

(ThermoFisher). Images were taken with Zeiss LSM780 confocal microscopy. Sarcomere length were analyzed using Zeiss Zen software.

[0049] mtDNA-to-nDNA ratio analysis Total DNA including mtDNA was extracted from cells using PureLink DNA kit (ThermoFisher), and DNA purity and quantity was determined by spectrophotometer. To determine the ratio between mitochondrial and nuclear DNA, qRT-PCR was performed on Roche Lightcycler 480 using SYBR Green dye. Mitochondrial gene expression was corrected for nuclear gene expression values, and normalized to the value of the control group per experiment as described before. Forward and reverse primer sequences are as follows: UUR forward, CAC CCA AGA ACA GGG TTT GT (SEQ ID NO: 1), UUR reverse, TGG CCA TGG GTA TGT TGT TA (SEQ ID NO: 2) for mt DNA, B2-microglobulin forward, TGC TGT CTC CAT GTT TGA TGT ATC T (SEQ ID NO: 3), B2-microglobulin reverse, TCT CTG CTC CCC ACC TCT AAG T (SEQ ID NO: 4).

[0050] Ca²⁺ transient assay Ca²⁺ transient was measured as described (Shimizu et al., 2015). Briefly, hESC-CMs cultured in the presence/absence of glucose were loaded with 5 μΜ fluo-4 AM and imaged in Tyrode buffer containing 138.2 mM NaCI, 4.6 mM KCI, 1.2 mM MgCI, 15 mM glucose, 20 mM HEPES according to manufacturer's instruction. Images were recorded on a Zeiss LSM 780 confocal microscope. Data analysis was carried out using the Zeiss Zen and ImageJ.

[0051] In vitro contractility assay Contractility assessments were performed by utilizing a video-based technique with the UCSF Gladstone-developed Matlab program MotionGUI (Huebsch et al., 2015). The videos were converted from .mts to .avi format at native resolution using commercially-available software and loaded into the MotionGUI program. The conversion between pixels and real distance was performed within the MotionGUI program using a reference image with unit divisions of 100 um, taken under the same objective and video zoom settings as the cell videos, to yield a pixel size of 0.681125. This pixel size was used for all contractility assessments. Motion vectors were calculated, and the data was evaluated upon completion. All samples were subjected to the same post-processing procedures in order to ensure consistency during comparative analysis. Each video sample was post-processed using neighbor-based cleaning with the vector-based cleaning criterion within the program. The threshold for this post-processing method was set to 2 for all samples and was adequate for improving the signal-to-noise ratio enough to clearly identify peaks corresponding to beating events in most samples. A small number of videos suffering from significant noise issues were separately subjected to fast Fourier transform (FFT) frequency domain cleaning with a cutoff frequency of 1 Hz. Only one post-processing method was applied to one video at one time. All other parameters of the MotionGUI program not outlined here were set to their respective default values.

[0052] Measurement of Intracellular lactate level by Laconic The lactate biosensor Laconic is a gift from Dr. Barros (San Martin et al., 2013). Overexpression of Laconic in hESC-CM was achieved with engineered adenoviruses encoding the construct. Expression of the construct was sufficiently high after 36-48 h for FRET experiments or microscopy imaging. All cells were imaged live without fixation. Images (16-bit) were acquired using a microscope (Eclipse TE300; Nikon) fitted with a 60K (1.4 NA) oil immersion lens (Nikon) and equipped with a filter cube comprising a CFP band-pass excitation filter, 436/20b, together with a longpass dichroic mirror (455DCLP; Chroma Technology Corp). Two LEDs (Philips Lumileds), one emitting at 455 ± 20 nm (royal blue) and the other emitting at 505 ± 15 nm (cyan) were used as light sources.

Ratiometric FRET measurements were obtained from the YFP and CFP images acquired simultaneously using a Dual View image splitter (Optical Insights) equipped with a 505-nm longpass dichroic filter to separate the CFP and YFP signals, a CFP emission filter (480/30), and a YFP emission filter (535/40) (John et al., 2008. 2011). Images were captured with a Cascade 512B digital camera (Photometries). Reagents indicated in figure 3C were added and followed by washing.

[0053] XF24 extracellular flux analyzer hESC-CMs were seeded onto a matrigel-coated XF24 Cell Culture Microplate (Seahorse Bioscience) at 2-7.5 10⁴ cells/well with or without glucose (25 mM Glucose of cardiac differentiation media). Oxygen consumption rate (OCR) was measured using an XF24 Extracellular Flux Analyser (Seahorse Bioscience) in unbuffered DMEM assay medium supplemented with 1 mM pyruvate, 2 mM glutamine and with or without 25 mM glucose. OCR was measured before and after the sequential addition of 0.75 μΜ oligomycin, 0.5 μΜ FCCP and 0.75 μΜ of rotenone/myxothiazol. OCR was normalized to protein concentration using a Bradford assay (Bio-Rad). Mitochondrial respiration was calculated as the difference between total and rotenone/myxothiazol rates. Maximal respiration was the response to FCCP. ATP-linked respiration was the oligomycin-sensitive respiration while uncoupled respiration was the difference between oligomycin and rotenone/myxothiazol rates.

[0054] Multi-electrode array hESC-CMs at the stage of hCM14 were plated on microelectrode arrays (MEAs) containing 120 integrated TiN electrodes (30 μηη diameter, 200 μπι interelectrode spacing). The MEAs were placed in an incubator with a temperature of 37 °C and 5% CO2- Two days were given to allow the cardiomyocytes to well attach the MEAs before starting recording. Local field potentials at each electrode were collected over a period of 5 minutes every day in total with a sampling rate of 1 KHz using the MEA2100-HS120 system (Multichannel systems, Reutlingen, Germany). Data analysis was carried out using the MC_DataTool (Multichannel Systems), Origin (OriginLab Corporation) and Matlab (MathWorks). Data shown are based on three independent hESC-CM prep.

[0055] Mass speetrometry-based metabolic measurements The experiments were performed as described (Krall et al., 2016). Briefly, cells were seeded in 6well plates, so that the final cell count at the time of metabolite extraction was about 7* 10^s and even across all cell lines. To extract intracellular metabolites, cells were briefly rinsed with cold 150 mM ammonium acetate (pH 7.3), followed by addition of 1 ml cold 80% MeOH on dry ice. Cell scrapers were used to detach cells, and the cell suspension was transferred into Eppendorf tubes. Extracted metabolites were transferred into glass vials and dried down under vacuum. For the LC-MS- based analysis, the samples were resuspended in 70% acetonitrile 50 μΙ were injected onto a Luna NH2 (150 mm x 2 mm, Phenomenex) column. Separation was achieved using A) 5 mM NH4ACO (pH 9.9) and B) ACN. The gradient started with 15% A) going to 90% A) over 18 min, followed by an isocratic step for 9 min and reversal to the initial 15% A) for 7 min. Metabolites were quantified with TraceFinder 3.3 using accurate mass measurements (≤ 3 ppm) and retention times of pure standards. Data analysis was performed using the statistical language R.

[0056] Gene expression analysis by quantitative reverse-transcriptase PCR. RNA was extracted from the tissue or the cells cultured with a specific concentration of glucose together with or without titrated metabolic pathway inhibitors using Direct-zol RNA mini prep kit (Zymo research). RNA was reverse-transcribed into complementary DNA using the qScript cDNA synthesis kit (Quanta Biosciences). Quantitative reverse-transcriptase PCR was performed using Viia7 (Applied Biosystems/ThermoFisher). In Figures 1(B), 2(B),4(C)-(E), Figure 4(L)-(P), and Figure 5(F)-(J), each bar represents the average of biological duplicates with at least 3 independent wells, each of which is triplicated for qPCR reaction. The relative mRNA level is normalized to the expression level of 25 mM glucose without any chemicals (white bar).

Forward and reverse primer sequences are as follows: GAPDH forward.

TTGAGGTCAATGAAGGGGTC (SEQ ID NO: 5), GAPDH reverse,

GAAGGTGAAGGTCGGAGTCA (SEQ ID NO: 6), TNNT2 forward,

CAGAGCGGAAAAGTGGGAAGA (SEQ ID NO: 7), TNNT2 reverse. TCGTTGATCCTGTTTCGGAGA (SEQ ID NO: 8), NKX2-5 forward,

GTTGTCCGCCTCTGTCTTCT (SEQ ID NO: 9), NKX2-5 reverse,

TCTATCCACGTGCCTACAGC (SEQ ID NO: 10). PPARGC1A forward,

GGTGCCTTCAGTTCACTCTCA (SEQ ID NO: 11), PPARGC1A reverse,

AACCAGAGCAGCACACTCGAT (SEQ ID NO: 12).

[0057] "F-FDG measurement by counter -⁸F-FDG was obtained from the UCLA Department of Nuclear Medicine. Warmed pregnant mice or pups were injected intravenous or

intraperitoneal (respectively) with ~90 microCi (~3.33 MBq) of ¹⁸F-FDG. After 2 hours, the mice were sacrificed. Preliminary experiments suggested that 2 hours was sufficiently long for ¹⁸F- FDG to reach maximum accumulation in each organ and embryo. Fetal or neonatal hearts were separated from the other tissue (carcass) and the mass and radioactivity in both the hearts and the carcasses were measured using a standard balance and a Wizard 3" automatic gamma counter (Perkin Elmer), respectively. The radioactivity levels in the pup carcasses were higher than the detection limit of the gamma counter and instead the expected gamma counter values for the pup carcasses were calculated based on the decay-corrected injected dose of ¹⁸F-FDG and known conversion values between microCi and CPM on the gamma counter. To calculate the "Normalized FDG accumulation", radioactive accumulation in each heart was divided by heart weight and then further divided by the total radioactivity in each embryo or pup. This last normalization is to account for differences in ¹⁸F-FDG injected dose and accessibility to the embryos and pups. Averages and standard errors of the mean (sem) were calculated, and the values were normalized such that E10.5 embryo FDG accumulation was set to 100.

[0058] Immunostaining Mouse embryos were isolated in cold PE3S and fixed in 4% PFA for 1~2 hours, followed by equilibration in 30% sucrose in PBS solution overnight. The tissues were placed in 1:1 30% sucrose/OCT (Tissue-Tek, Electron Microscopy Sciences) solution for 1-2 hr, in 100% OCT compound for 1 hr at 4°C, and embedded in 100% OCT compound, carefully oriented in Cryomolds (Ted Pella). The blocks were immediately frozen on dry ice with isopropanol and stored at -80°C. The sections were cut 5μπι with a Leica CM3050 S cryostat. The following primary and secondary antibodies were used: oActinin (mouse, 1:200, Sigma- Aldrich), Phospho-Histone 3 (rabbit, 1:250, Millipore), Alexa Fluor 488 (green). Alexa Fluor 594 (red)-conjugated secondary antibodies specific to the appropriate species were used (1 :500; ThermoFisher) for fluorescent staining. Sections were mounted with antifade mounting medium with DAPI (ThermoFisher), and analyzed by using Axiolmager D1 (Carl Zeiss Microimaging, Inc).

[0059] Statistical analysis ANOVA and Student's t-test were used to determine whether any statistically significant difference exists among independent groups. Results

[0060] Glucose reduction promotes hESC-CM differentiation

[0061] hESC-CMs were differentiated in monolayer in a chemically-defined condition reproducibly yielding ~90% of MF20⁺ cardiomyocytes at day 14 with multiple cell lines including WA09 (H9) and UCLA4 hESCs (Figures 1A-C) (Arshi et al., 2013; Minami et al., 2012). hESC- CMs start to beat synchronously at around day 6-7 in our system. To characterize the differentiation stages, mRNA expression profiles from H9 hESC-CMs was serially examined by RNA-seq at 5 distinct stages (GSE84815); undifferentiated hESC (day 0). mesodermal precursor stage (hMP, day 2), cardiac progenitor stage (hCP, day 5), immature cardiomyocyte (hCM14) and hESC-CMS differentiated for 14 additional days (hCM28). The expression data were analyzed using signatures collected from MSigDB, a body atlas and primary cell atlas (Mabbott et al., 2013; Su et al., 2004: Subramanian et al., 2005). As expected, the stem cell signature decreases during these five stages, while signatures associated with heart and smooth muscle increase, further suggesting that our protocol leads to highly enriched cardiomyocytes (Figures 1 D and 1 E). This differentiation course is comparable to the previous reports (Paige et al., 2012; Wamstad et al., 2012).

[0062] To examine the impact of glucose levels on cardiac differentiation, hESC-CMs were cultured in media containing various concentration of glucose starting at the hCM14 stage, when cells are already differentiated to immature cardiomyocytes (Figure 2A). The basal differentiation medium contains 25 mM glucose, 0.9 mM pyruvate, essential and nonessential amino acids, and human albumin (G25 medium; Table 1). Interestingly, glucose dose- dependently suppressed the expression of TNNT2 (cardiac marker), NKX2-5 (cardiac marker), and PPARGC1A (mitochondrial marker) (Figure 2B). Gene expression profiling by RNA-seq revealed that genes related to cardiac muscle and function are enriched in hESC-CMs in low glucose medium and genes associated with mitosis and cell cycle are enriched in high-glucose group genome-wide (Figure 2(C), (K); GSE84814). These data suggest that low glucose after day 14 induces the differentiation and suppresses cell cycle of hESC-CMs.

[0063] To validate these results, hESC-CM proliferation was analyzed by pH3 staining and EdU flow cytometry analysis. Low glucose decreased mitotic activity at day 28 without affecting the viability of hESC-CMs (Figure 2D, E). In addition, hESC-CMs in low glucose medium showed more robust staining of a-actinin, although the sarcomere length did not significantly change (Figure 2F, K). MitoTracker staining and flow cytometry analyses revealed that hESC-CMs cultured in low glucose media have increased mitochondrial contents and inter-myofibrillar distribution of mitochondria characteristic of differentiated cardiomyocytes in a dose-dependent manner (Figures 2F and 2G). Addition of 2-DG (2-Deoxy-D-glucose), a competitive inhibitor of glucose phosphorylation, induced higher levels of MitoTracker and MF20 expression even in the presence of 5 or 25 mM glucose (Figure 2L), suggesting that the effect is specific to glucose and not to changes in osmotic pressure. Consistently, flow cytometry showed significant increase in cell size in glucose-restricted condition (Figure 21). Together, these results demonstrate that glucose dose-dependently suppresses the maturation of cardiomyocyte cellular architecture and the upregulation of cardiac genes in hESC-CMs.

[0064] Glucose reduction promotes functional maturation of hESC-CMs

[0065] We next compared the metabolic and functional maturity of hESC-CMs cultured in the presence and absence of glucose by six methods. First, hESC-CMs were stained with JC-1, a green fluorescent dye that generates red fluorescence upon formation of aggregates in active mitochondria. The level of red fluorescence is often used as an indicator of mitochondrial membrane potential and, therefore, mitochondrial activity. Immunofluorescent staining revealed that mitochondria in glucose-reduced hESC-CMs are more elongated (Figure 3A), and flow cytometry analysis revealed that the JC-1 aggregate is significantly higher in glucose-reduced hESC-CMs cultured in both regular 0.9 mM and 10 mM pyruvate conditions. (Figure 3B). These data suggest that high glucose inhibits the functional maturation of mitochondria.

[0066] Second, intracellular lactate levels of hESC-CMs were measured using Laconic, a FRET (Fluorescence Resonance Energy Transfer)-based lactate nanosensor (San Martin et al., 2013). After glucose deprivation, the Laconic construct was introduced to hESC-CMs via adenovirus as described (John et al., 2008). In hESC-CMs differentiated in standard 25 mM glucose, bath- applied lactate (4 mM) caused a fast increase in intracellular lactate (decrease FRET ratio), demonstrating the efficacy of the probe. Subsequent addition of pyruvate evoked a similar, but smaller elevation of intracellular lactate. Under these conditions inhibition of mitochondrial respiration with sodium cyanide (NaCN) had only a minor effect on the intracellular lactate level. This result suggests that hESC-CMs differentiated in standard 25 mM glucose do not actively metabolize pyruvate (Figure 3C, upper panel). In contrast, addition of 4 mM pyruvate to glucose-reduced hESC-CMs did not cause intracellular lactate accumulation, and addition of NaCN in the presence of pyruvate resulted in a substantial increase in lactate level. This result is consistent with pyruvate utilization by mitochondria (Figure 3C, lower panel). Together, these data suggest that mitochondria metabolize pyruvate in hESC-CMs that are cultured in glucose- reduced conditions, but not in the presence of glucose.

[0067] Third, we assessed cellular respiration of hESC-CM using the XF24 Extracellular Flux Analyzer (Seahorse Bioscience), in which oxygen consumption rate (OCR) was measured in real time in a basal state and in response to oligomycin (ATP synthase inhibitor), FCCP

(mitochondrial uncoupler), and rotenone/myxothiazol (complex l/lll inhibitors, respectively) (Figure 3D). Although base-line mitochondrial respiration was not changed (Figure 3E), ATP- linked respiration was elevated in the no glucose condition (Figures 3F and 3G). Glucose- reduced hESC-CMs also demonstrated substantially larger maximum respiration capacity as indicated by the response to FCCP (Figure 3H). These results corroborate the greater capacity of cellular respiration in hESC-CMs cultured under low glucose.

[0068] Fourth, the Ca^2* kinetics of hESC-CMs were assessed using a Ca²⁺ transient assay (Shimizu et al., 2015). While the peak amplitude of the transient (AF/FO) did not show a significant difference, the maximum upstroke (Vmax) was significantly faster and the time to 50% decay was significantly shorter in glucose-reduced group (Figure 3I). This pattern is consistent with the previous report demonstrating the role of thyroid hormone on the hiPSC-CM maturation (Yang et al., 2014), and suggestive of an inhibitory role of high glucose on hESC-CM

maturation.

[0069] Fifth, taking advantage of our monolayer culture system, we examined

electrophysiological properties with multi-electrode array (MEA) culture plate as reported (Zhu et al., 2017). Maximum upstroke velocity (dV/dW) of field potential is a reliable parameter for the electrophysiological maturity of cardiomyocytes derived from pluripotent stem cells (Haase et al., 2009; Ma et al., 2011; Zhang et al., 2009). Compared with hESC-CMs cultured in 25mM glucose, those cultured in the absence of glucose displayed a significant increase in dV/dW in glucose deprivation condition (Figure3K-3M).

[0070] Finally, monolayer culture method allowed for the measurement of contractility by digital image correlation using MotionGUI program (Huebsch et al., 2015). Both average max contraction speed and average max relaxation speed were higher in hESC-CMs cultured in OmM glucose (Figure 3J). Together, these data suggest that glucose reduction promotes functional maturation of hESC-CMs at metabolic, electrophysiological, and biomechanical levels.

[0071] Glucose blocks cardiac maturation via pentose phosphate pathway

[0072] Having determined that glucose reduction induces cardiac maturation at morphological, genetic metabolic, and functional levels, we next sought to analyze the mechanism by which glucose blocks cardiac maturation. Glucose is metabolized by multiple pathways facilitating both catabolic reactions (anaerobic glycolysis and aerobic TCA cycle) and anabolic reactions (PPP, hexosamine pathway, etc.). We first examined the impact of glucose reduction on the global metabolomics signature. Mass spectrometry revealed that glucose deprivation resulted in a significant decrease in the levels of the metabolites in purine metabolism, pyrimidine

metabolism, PPP, hexosamine pathway, and glycolysis, while lipid precursors, amino acids, glutamine, and glutamate, as well as urea cycle metabolites, were not significantly affected (Figures 4A and Figure 4I). ATP level was not significantly different in glucose-restricted condition (Figure 4B) nor was there any specific stress pathways significantly increased by RNA-seq, suggesting that the cells were not energy-starved in the absence of glucose in the cell culture media. [0073] To identify the metabolic pathway responsible for the improved cardiac maturation by glucose reduction, we conducted a systematic screening using chemical inhibitors for the various glucose metabolic pathways in monolayer 384-well format (Figures 4C). Consistent with flow cytometry for MitoTracker and MF20 (Figure 2L), 2-DG dose-dependently abolished the glucose-dependent inhibition of TNNT2 and NKX2-5 mRNA levels in hESC-CM (Figures 4D and 4L). 3PO (S-iS-PyridinyO-l-t^pyridinyO^-propen-l-one), an inhibitor of PFKFB3, a regulator of PFK1, did not affect the level of hESC-CM maturity at any concentration of glucose (Figures 4E and 4M). The failure of 3PO to recapitulate the effect of glucose deprivation suggests that glucose metabolites downstream of PFK1 are not essential for the glucose-dependent inhibition of cardiac maturation. Consistently, sodium oxamate (LDH inhibitor) did not block the inhibitory effect of glucose (Figures 4F and 4N). Interestingly, however, 6AN (6-(cyclohexa-2,5-dien-1- ylideneamino) naphthalene-2-sulfonate) and DHEA (didehydroepiandrosterone), both inhibitors of G6PD in the oxidative arm of the PPP, recapitulated glucose reduction (Figures 4G, 4H, 40, 4P). As summarized in Figure 4(K), our chemical inhibitor screening suggests that the PPP plays a critical role in the inhibition of cardiac maturation and that blocking this pathway by either chemical inhibitors or glucose deprivation induces cardiac maturation. Although mitochondria are a major source of reactive oxygen species (ROS) and physiological levels of ROS promote cellular differentiation (Crespo et al., 2010), the level of ROS measured by DCFDA (dichlorodihydrofluorescein diacetate) did not increase in the absence of glucose nor did ROS inhibition had a significant impact on TNNT2 expression level (Figure 4Q). suggesting that the increase in ROS is not responsible for the induction of cardiac maturation.

[0074] Nucleotide metabolism regulates cardiomyocyte maturation

[0075] The oxidative arm of the PPP generates two major products; reducing power in the form of NADPH and 5-carbon sugars which supply the backbone for nucleotide biosynthesis. To test whether glucose level impacts cardiac maturation via nucleotide biosynthesis, we rescued nucleotide synthesis by adding uridine to hESC-CMs cultured in low glucose media. Under glucose starvation, supplementation of nucleotide by the addition of uridine is known to rescue the growth of bacteria, yeast and malignant cells (Linker et al., 1985). In our hESC-CM culture system, uridine restored the cell proliferation even in the low glucose (Figures 5A and 5B). Interestingly, uridine dose-dependently reduced the level of TNNT2 in glucose-deprived condition (Figures 5C and 5F). suggesting that glucose-mediated inhibition of cardiac maturation is dependent on the supply of nucleotides and not NADPH.

[0076] To test whether nucleotides are necessary for the glucose-dependent inhibition of cardiac maturation, nucleotide biosynthesis was blocked by multiple methods. Contrary to uridine, addition of an excess amount of thymidine blocks the synthesis of DNA by inhibiting the formation of deoxycytidine (i.e., the thymidine block method), which is commonly used to synchronize the cell cycle (Reichard et al., 1960; Xeros, 1962). When excess thymidine was added to hESC-CMs, the expression of TNNT2 and NKX2-5 were increased (Figure 5D and Figure 5G). To further confirm this effect, we blocked deoxynucleotide synthesis by hydroxyurea (HU), an inhibitor of ribonucleotide reductase (RNR) that catalyzes the formation of deoxyribo- nucleotides. Consistent with the results from thymidine block, HU dose-dependently induced the expression of TNNT2 and NKX2-5 (Figures 5E and 5H). RNAi-based knockdown of RRM2B, a key subunit of RNR, also showed a significant increase in TNNT2 expression level even in the presence of 25 mM glucose (Figure 4J). Together, these gain- and loss-of-function data suggest that nucleotide biosynthesis is a key regulatory pathway of the pro-mitotic/anti-maturation effect of glucose.

[0077] Nucleotide deprivation, not cell cycle block, induces cardiomyocyte maturation

[0078] Nucleotide synthesis is a key step in DNA replication and thus cell cycle activity.

Therefore, it is not clear whether the maturation of hESC-CMs by deprivation of glucose is due to the cell cycle block in general or the effects of nucleotides themselves. To examine whether cell cycle arrest in general is an essential trigger of cardiac maturation, we blocked the mitotic activity of hESC-CMs by a CDK4/6 inhibitor and paclitaxel (Taxol®; inhibitor of microtubule breakdown), which both block cell cycle without directly inhibiting the nucleotide kinetics.

Interestingly, neither CDK4/6 inhibitor nor paclitaxel induced cardiac maturation (Figures 5I and 5J). Together, these data suggest that cell cycle arrest by itself is not critical for the promotion of cardiac maturation. Rather, nucleotide deprivation is a key mechanism for cardiac maturation.

[0079] Glucose uptake is progressively suppressed during physiological cardiogenesis in utero

[0080] These in vitro data suggest that glucose reduction promotes cardiac maturation while inhibiting the proliferation. An intriguing possibility is that the same mechanism underlies the cardiac maturation in the in vivo natural counterpart However, during normal embryogenesis, the blood glucose level is primarily regulated by maternal metabolism and stays relatively stable in utero, leading us to hypothesize that cellular glucose uptake becomes restricted at late fetal stages. To test this possibility, we measured the glucose uptake in the fetal hearts using ¹⁸F- labeled 2-deoxy-2-fluoroglucose (FDG). ¹⁸F-labeled FDG was injected intravenously via the maternal tail vein at E10.5, 13.5, and 15.5 or intraperitoneal to P1 and P7 pups. After 2 hours, the mice were imaged by PET/CT, the hearts were dissected, and cardiac ¹⁸F-FDG

accumulation was measured quantitatively. Interestingly, the normalized cardiac accumulation progressively and rapidly decreases from E10.5 to P7 with 0.11% and 0.05% uptake at P1 and P7 heart, respectively, compared to E10.5 heart (Figure 6). These data suggest that cardiac glucose uptake becomes significantly restricted at late gestational and early postnatal stages, creating an intracellular glucose deprivation condition during natural in vivo development.

[0081] Hyperglycemia promotes proliferation and inhibits maturation of cardiomyocytes in utero [0082] We next tested whether hyperglycemia promotes proliferation and inhibits maturation of cardiomyocytes in vivo using fetuses and neonates from diabetic pregnancy. Akita

heterozygous mice carry a single amino acid substitution in the Ins2 gene and exhibit multiple disorders associated with maturity-onset diabetes of the young (MODY) (Barber et al., 2005; Fujita et al., 2001 ; Wang et al., 1999; Yaguchi et al., 2003; Yoshioka et al., 1997). By crossing an Akita female with a wild-type male, we created a diabetic pregnancy condition in which wild- type fetuses (half of the litters) are exposed to hyperglycemia (Figure 7A). In the C57BLJ6 background, the average blood glucose levels of Akita mothers we used were significantly higher than sex-matched control littermates (215±84 vs 71±12 mg/dl. respectively; p < .0005). Fetal and neonatal Tnnt2* cardiomyocytes from wild-type hearts from wild-type mothers and wild-type hearts from Akita mothers were examined for the mitotic activity by in vivo EdU incorporation assay at E16.5 and P0 stages when cardiomyocytes are not yet

multinucleated/multiploidic. As shown in Figure 7B and 7C, the number of cardiomyocytes in S phase were significantly higher at both E16.5 and P0 in diabetes group. Histological analyses showed that the number of phosphorylated histone H3-positive cardiomyocytes (pH3VTnnt2*) are higher at E16.5 in the embryos from diabetic pregnancy (Figures 7D and 7E). These data suggest that fetal cardiomyocytes are more mitotic when exposed to maternal hyperglycemia.

[0083] To examine whether hyperglycemia inhibits the maturation of fetal cardiomyocytes in vivo, we analyzed the fetal/neonatal heart from diabetic pregnancy. The level of Tnnt2 expression was significantly lower in the heart from the diabetic pregnancy (Figure 7F). A hallmark of the congenital heart disease associated with diabetic pregnancy is asymmetric cardiac hypertrophy. Although heart weight/body weight ratio did not show a difference in our mouse model, the thickness of left and right ventricular free walls was significantly increased in the hearts from diabetic pregnancy at P1 (Figure 7G). Consistently, the cardiomyocyte size measured by flow cytometry was significantly smaller in diabetic pregnancy group (Figure 7H). These data suggest that overproliferation and/or delayed maturation underlie the pathological mechanism of cardiomyopathy associated with diabetic pregnancy.

Discussion

[0084] This Example shows (1) that glucose dose-dependently inhibits cardiac maturation in hESC-CMs, (2) that the maturation-inhibitory effect is dependent on nucleotide biosynthesis via the PPP, (3) that the developing heart accomplishes intracellular glucose starvation by limiting the glucose uptake at late gestational stages during normal embryogenesis, and (4) that perturbation of the environmental glucose level in diabetic pregnancy affects natural cardiomyocyte maturation in vivo.

[0085] Cardiomyocytes switch their main energy substrate from glucose (or other

carbohydrates) to fatty acids shortly after birth. This metabolic switch has long been believed to be an adaptation of cardiomyocytes to facilitate more efficient production of ATP. However, our study has discovered that a drastic suppression of glucose uptake occurs during gestational stages, long before the metabolic switch after birth (Figure 5). Our in vitro hESC-CM glucose- deprivation experiments mimic this in vivo glucose starvation. Our results suggest that glucose deprivation induces cardiac maturation at genetic, morphological, metabolic and

electrophysiological levels (Figures 2 and 3). Highly proliferative cells including cancer cells and pluripotent stem cells often exhibit high rates of glycolysis even in the presence of ample oxygen. This phenomenon, known as the Warburg effect, is considered to be an adaptation of cells to potentiate anabolism of nutrients into biomaterials. In an environment with plentiful glucose and oxygen, the limiting factor for cell proliferation is not ATP but the building blocks for biomaterials including nucleotides, amino acids, and lipids (Vander Heiden et al., 2009). Recent studies suggest that the specific metabolites play a regulatory role in the differentiation of stem cells (Shiraki et al., 2014; Shyh-Chang et al., 2013). Thus, the metabolic environment is an important determinant of cell behavior. Our study revealed that glucose is a negative regulator of the maturation as well as a positive regulator of the mitotic activity of fetal cardiomyocytes in vitro and in vivo. Perturbation of the natural glucose starvation results in higher mitotic activity and lower maturity of cardiomyocytes in vivo (Figure 7). Together, our results suggest that the metabolic switch during perinatal stages is not only to meet the energy demand but also to induce the genetic program that facilitates the maturation of the cardiomyocytes in vivo. An important question yet to be answered is how the drastic suppression of intracellular glucose uptake is achieved in the fetal heart. As the fetal glucose environment is primarily determined by maternal metabolism and kept relatively constant in utero, one possibility is that the glucose uptake is limited at the glucose transporter level in fetal cardiomyocytes. In fact, fetal heart switches glucose transporter isoform at around this stage. Understanding how glucose regulates genetic program and how the glucose uptake is regulated at the genetic level will be a key to further dissect the cross-talk between genetic and non-genetic factors governing heart formation.

[0086] Nucleotide biosynthesis via the PPP as a key balancing mechanism of

proliferation and maturation Glucose is the most fundamental and commonly available nutrient to the cells. Hence, the activity of the glucose metabolic pathways is tightly regulated in cells. Glucose is broken down to extract energy through glycolysis pathway and also shunts to supply 5-carbon sugars and NADPH through the PPP. In our study, chemical inhibition of glucose metabolic pathways in hESC-CMs revealed that it is not the catabolic breakdown of glucose to extract energy but rather the anabolic use of glucose to build nucleotides that are responsible for the glucose-dependent inhibition of cardiac maturation. Most of the proliferating cells synthesize nucleotides de novo from glucose, glutamine, and CO2. In our hESC-CM experiments, blocking the PPP and nucleotide biosynthesis inhibited the glucose-mediated induction of mitosis and suppression of maturation, and supplementation of nucleotides was sufficient to recapitulate the effect of glucose (Figures 4 and 5). These data suggest that nucleotide biosynthesis via the PPP is the key regulator of the pro-mitotic/anti-maturation effect of glucose. Interestingly, cardiomyocyte maturation was not induced by blocking cell cycle by a CDK4/6 inhibitor or Paditaxel neither of which directly impact the nucleotide kinetics. Therefore, it is not the cell cycle in general but the nucleotide itself that blocks the maturation (Figure 51. 5J). It is well-documented that there is generally an inverse correlation between cell proliferation and differentiation during developmental stages (Ruijtenberg and van den Heuvel, 2016). Our data raise an intriguing possibility that nucleotide biosynthesis serves as a nodal point balancing cell proliferation and differentiation during development.

[0087] Hyperglycemia as a potential teratogen to the fetal heart Clinically, maternal diabetes can accompany multiple complications including neuropathy, microvasculopathy, nephropathy, and insulin resistance. Although meta-analysis predicts that hyperglycemia itself is a major teratogen during diabetic pregnancy (Reece et al., 1996), it is often difficult to dissect the impacts of maternal complications on CHD as they are often subclinical. To our knowledge, our in vitro study is the first to demonstrate that environmental glucose itself, if excessive, directly impacts cardiac differentiation. The heart formation is regulated by both genetic and non-genetic factors with the latter playing important roles particularly during late-stage cardiogenesis. An interesting aspect of the interaction between genetic and non-genetic mechanisms is that they seem to mutually reinforce each other. Our data suggest that the glucose metabolic environment is, on one hand, a consequence of changes in cardiac genetic program and on the other hand a cause of the changes in cardiac gene expression.

[0088] Potential application to the therapeutics Understanding the metabolic signature of hESC-CMs will potentially open new methods for purifying the hESC-CMs (Tohyama et al., 2016; Tohyama et al.. 2013) or inducing their maturation (Drawnel et al., 2014). Considering the inhibitors of the PPP and nucleotide biosynthesis have entered clinical trials for cancer treatment (Tennant et al., 2010; Vander Heiden, 2011), our data raise the possibility that manipulating this pathway may allow us to control the proliferation and maturation of cardiomyocytes for regenerative medicine.

[0089] With the advances in fetal diagnosis and surgical techniques, the number of CHD patients who survive childhood (adult CHD) is growing rapidly by nearly 5% per year (Brickner et al., 2000). Maternal hyperglycemia a common medical condition associated with 2-5 fold increase in CHD (Centers for Disease, 1990; Simeone et al., 2015; Yogev and Visser, 2009). Currently, 60 million women of reproductive age (18-44 years old) worldwide and approximately 3 million in the U.S. have diabetes mellitus. This number is estimated to double by 2030, posing a huge medical and economic burden (Gabbay-Benziv et al., 2015). Therefore,

prevention/treatment of CHD associated with gestational diabetes will be an urgent medical issue in next few decades. The findings presented herein will lay a foundation for understanding how glucose environment regulates cardiogenesis and how disturbance of non-genetic factors affects the genetic program during the pathological development of the heart.

References

[0090] Arshi, A., et al. (2013). Sci Technol Adv Mater 14. PubMed PMID: 24311969; PMCID: 3845966.

[0091] Barber, A.J., et al. (2005). Invest Ophthalmol Vis Sci 46, 2210-2218. PubMed PMID: 15914643.

[0092] Brickner, M.E., et al. (2000). N Engl J Med 342, 256-263. PubMed PMID: 10648769.

[0093] Carey, B.W., et al. (2015). Nature 518, 413-416. PubMed PMID: 25487152; PMCID: PMC4336218.

[0094] Centers for Disease, C. (1990). Perinatal mortality and congenital malformations in infants born to women with insulin-dependent diabetes mellitus-United States, Canada, and Europe, 1940-1988. MMWR Morb Mortal Wkly Rep 39, 363-365. PubMed PMID: 2111001.

[0095] Cox. S.J., and Gunberg, D.L. (1972). J Embryol Exp Morphol 28. 235-245. PubMed PMID: 4642990.

[0096] Crespo, F.L, et al. (2010). Stem Cells 28, 1132-1142. PubMed PMID: 20506541.

[0097] Drawnel, F.M., et al. (2014). Cell Rep 9, 810-821. PubMed PMID: 25437537.

[0098] Fahed, A.C., et al. (2013). Circ Res 112, 707-720. PubMed PMID: 23410880.

[0099] Fujita. H., et al. (2001). Exp Nephrol 9, 380-386. PubMed PMID: 11701997.

[0100] Gabbay-Benziv, R., et al. (2015). World J Diabetes 6, 481-488. PubMed PMID:

25897357; PMCID: PMC4398903.

[0101] Gaspar, J.A., et al. (2014). Circ Res 114, 1346-1360. PubMed PMID: 24723659.

[0102] Haase, A., et al. (2009). Cell Stem Cell 5, 434-441. PubMed PMID: 19796623.

[0103] Harris, J.M., et al. (2013). Blood 121, 2483-2493. PubMed PMID: 23341543; PMCID: PMC3612858.

[0104] Huebsch, N., et al. (2015). Tissue Eng Part C Methods 21, 467-479. PubMed PMID: 25333967; PMCID: PMC4410286.

[0105] John, S.A., et al. (2008). Pflugers Arch 456, 307-322. PubMed PMID: 18071748; PMCID: PMC4339035.

[0106] Krall, A.S., et al. (2016). Nat Commun 7, 11457. PubMed PMID: 27126896; PMCID: PMC4855534.

[0107] Linker. W., et al. (1985). Eur J Cell Biol 36, 176-181. PubMed PMID: 3996430. [0108] Ma, J., et al. (2011). Am J Physiol Heart Circ Physiol 301, H2006-2017. PubMed PMID: 21890694; PMCID: PMC4116414.

[0109] Mabbott, N.A., et al. (2013). BMC Genomics 14, 632. PubMed PMID: 24053356; PMCID: PMC3849585.

[0110] Mackler, B., et al. (1971). Arch Biochem Biophys 144, 603-610. PubMed PMID:

4328160.

[0111] Makinde. A.O., et al. (1998). Mol Cell Biochem 188, 49-56. PubMed PMID: 9823010.

[0112] Minami, I., et al. (2012). Cell Rep 2, 1448-1460. PubMed PMID: 23103164.

[0113] Oburoglu. L, et al. (2014). Cell Stem Cell 15, 169-184. PubMed PMID: 24953180.

[0114] Paige, S.L, et al. (2012). Cell 151 , 221-232. PubMed PMID: 22981225; PMCID:

PMC3462257.

[0115] Pediatric Cardiac Genomics, C, et al. (2013). Circ Res 112, 698-706. PubMed PMID: 23410879; PMCID: PMC3679175.

[0116] Reece, E.A., et al. (1996). Teratology 54, 171-182. PubMed PMID: 9122886.

[0117] Reichard, P., et al. (1960). Biochim Biophys Acta 41, 558-559. PubMed PMID:

14437039.

[0118] Ruijtenberg, S., and van den Heuvel, S. (2016). Cell Cycle 15, 196-212. PubMed PMID: 26825227; PMCID: PMC4825819.

[0119] San Martin, A., et al. (2013). PLoS One 8, e57712. PubMed PMID: 23469056; PMCID: PMC3582500.

[0120] Shimizu, H., et al. (2015). Elife 4. PubMed PMID: 25588501; PMCID: PMC4293673.

[0121] Shiraki, N., et al. (2014). Cell Metab 19, 780-794. PubMed PMID: 24746804.

[0122] Shyh-Chang, N., et al. (2013). Science 339, 222-226. PubMed PMID: 23118012;

PMCID: PMC3652341.

[0123] Simeone, R.M., et al. (2015). Am J Prev Med 48, 195-204. PubMed PMID: 25326416; PMCID: PMC4455032.

[0124] Su, A.I., et al. (2004). Proc Natl Acad Sci U S A 101, 6062-6067. PubMed PMID:

15075390; PMCID: PMC395923.

[0125] Subramanian, A., et al. (2005). Proc Natl Acad Sci U S A 102, 15545-15550. PubMed PMID: 16199517; PMCID: PMC1239896.

[0126] Tennant, D.A., et al. (2010). Nat Rev Cancer 10, 267-277. PubMed PMID: 20300106.

[0127] Tohyama, S., et al. (2016). Cell Metab 23, 663-674. PubMed PMID: 27050306. [0128] Tohyama, S., et al. (2013). Cell Stem Cell 12, 127-137. PubMed PMID: 23168164.

[0129] Vander Heiden, M.G. (2011). Nat Rev Drug Discov 10, 671-684. PubMed PMID:

21878982.

[0130] Vander Heiden, M.G., et al. (2009). Science 324, 1029-1033. PubMed PMID: 19460998; PMCID: PMC2849637.

[0131] Wamstad, J.A., et al. (2012). Cell 151, 206-220. PubMed PMID: 22981692; PMCID: PMC3462286.

[0132] Wang, J., et al. (2009). Science 325, 435-439. PubMed PMID: 19589965; PMCID:

PMC4373593.

[0133] Wang, J., et al. (1999). J Clin Invest 103, 27-37. PubMed PMID: 9884331 ; PMCID: PMC407861.

[0134] Warshaw, J.B., and Terry, M.L (1970). J Cell Biol 44, 354-360. PubMed PMID:

5415033; PMCID: PMC2107947.

[0135] Xeros, N. (1962). Nature 194, 682-683. PubMed PMID: 14008663.

[0136] Yaguchi, M., et al. (2003). Neuropathology 23, 44-50. PubMed PMID: 12722925.

[0137] Yang, X., et al. (2014). J Mol Cell Cardiol 72, 296-304. PubMed PMID: 24735830;

PMCID: PMC4041732.

[0138] Yogev, Y., and Visser, G.H. (2009). Semin Fetal Neonatal Med 14, 77-84. PubMed PMID: 18926784.

[0139] Yoshioka, M., et al. (1997). Diabetes 46, 887-894. PubMed PMID: 9133560.

[0140] Young, C.S., et al. (2016). Cell Stem Cell 18, 533-540. PubMed PMID: 26877224;

PMCID: PMC4826286.

[0141] Zhang, J., et al. (2009). Circ Res 104, e30-41. PubMed PMID: 19213953; PMCID:

PMC2741334.

[0142] Zhu, H., et al. (2017). Sci Rep 7, 43210. PubMed PMID: 28266620; PMCID:

PMC5339708.

[Example 2: Administration of hydroxyurea (HU) facilitates cardiomvocvte maturation

[0143] As shown in Figure 5 (see also Example 1), nucleotide metabolism regulates

cardiomyocyte maturation. Glucose-deprived hESC-CMs were cultured in the absence or presence of 25 mM uridine, and stained for the mitosis marker pH3. The addition of uridine restored the proliferative activity even in the absence of glucose. This was demonstrated in three independent experiments. Figure 5B shows the proliferation rate as pH3+ cells /a-Actinin+ of the stained images of Fig. 5A, shown in percentage (n=3. mean ± SD, p < 0.01 by t-test).

[0144] Relative mRNA expression of TNNT2 in hESC-CMs in 25 mM or 0 mM glucose with 0 or 25 mM uridine is shown in Figure 5C. Uridine dose-dependently inhibited the TNNT2 expression level in glucose-deprived condition (n=3, mean ± SD, p<0.0005 by one-way ANOVA test). See also Figure 5F. Figure 5D shows the relative mRNA expression of TNNT2 and NKX2-5 in hESC-CMs cultured in 0-25 mM of glucose in the presence or absence of thymidine. Thymidine block induced the levels of TNNT2 and NKX2-5 (n=3, mean ± SD, p<0.01 by t-test between with or without 25 mM Thymidine with Glucose 25 mM. See also Figure 5G.

[0145] Relative mRNA expression of TNNT2 and NKX2-5 in hESC-CMs cultured in 0-25 mM of glucose and 0-2 mM hydroxyurea (HU. a ribonucleotide reductase inhibitor) is shown in Fig. 5E. HU dose-dependently induced the expression of TNNT2 and NKX2-5 at 1, 5, and 25 mM glucose. (n=3, mean ± SD, p-value by one-way ANOVA test. See also Figure 5H. The gain- and loss-of-function data suggest that nucleotide biosynthesis is a key regulatory pathway of the pro-mitotic/anti-maturation effect of glucose.

Example 3: Preparation of cardiomvocvtes from stem cells

[0146] This Example demonstrates how differentiation of cardiomyocytes from embryonic or induced pluripotent stem cells can be improved by modulating nucleotide biosynthesis. As discussed in Examples 1 and 2 above, the metabolic switch not only affects the energy demand, but also induces a genetic program to facilitate cardiac maturation. The glucose restriction method can thus be used to induce the maturation of cardiomyocytes from human embryonic stem cells (hESCs) and human induced pluripotent stem cells (hiPSCs). Immature phenotype of hESC-CMs and hiPSC-CMs has been a hurdle to the application of human stem cells to cell transplantation, drug screening, and disease modeling. Improvement of the maturation of hESC/iPSC-CMs by this method will contribute to a better outcome of these therapeutics and research.

Example 4: Therapeutic effect of hydroxyurea on neonates with congenital heart disease

[0147] This Example demonstrates the terapeutic effect of HU (hydroxyurea) on the phenotype of congenital heart disease in the neonates from diabetic Akita mother mouse. Wild type male was crossed to wild type control mothers and diabetic mothers (Akita) injected with HU or vehicle solution. HU was injected at 10mg/kg i.p. daily from E12.5 until birth. As shown in Figure 8, HU treatment reversed the hypertrophic phenotype. Shown are the measurements of right ventricular wall thickness. n=8-10, each. p<0.05. [0148] Supporting materials confirming the data presented in the examples above (figures, data, videos) can be accessed in full color at eLife 2017;6:e29330 DOI: 10.7554/el_ife.29330. Two videos are included in these online materials: Video 1 shows beating of hESC-CMs differentiated in the medium containing 25 mM glucose from day 14 for 7 days. Video 2 shows beating of hESC-CMs differentiated in the medium containing 0 mM glucose from day 14 for 7 days. Also included with the online materials are Datasets and Additional Materials. The following data sets were generated: 'The global transcriptome analysis in the time course of hESC-derived cardiac differentiation", Atsushi N, Matteo P (2017), Publicly available at the NCBI Gene Expression Omnibus (accession no: GSE84815); and "Glucose inhibits cardiac maturation through nucleotide", Atsushi N, Matteo P (2017) Publicly available at the NCBI Gene Expression Omnibus (accession no: GSE84814). Additional files: "Transparent reporting form" doi.org/10.7554/eLife.29330.021.

[0149] Throughout this application various publications are referenced. The disclosures of these publications in their entireties are hereby incorporated by reference into this application in order to describe more fully the state of the art to which this invention pertains.

[0150] Those skilled in the art will appreciate that the conceptions and specific embodiments disclosed in the foregoing description may be readily utilized as a basis for modifying or designing other embodiments for carrying out the same purposes of the present invention. Those skilled in the art will also appreciate that such equivalent embodiments do not depart from the spirit and scope of the invention as set forth in the appended claims.

Claims

What is claimed is: 1. A method of hindering the development of congenital heart disease (CHD) in a subject, the method comprising administering an inhibitor of nucleotide biosynthesis enzymes to the subject in utero.

2. The method of claim 1, wherein the inhibitor is administered to a pregnant female carrying the subject in utero.

3. The method of claim 1 , further comprising screening a pregnant female carrying the subject in utero for hyperglycemia or diabetes, and/or screening the subject for family history or a genetic indicator of CHD prior to administering the inhibitor.

4. The method of claim 3, wherein the screening for hyperglycemia or diabetes comprises a glucose tolerance test.

5. The method of claim 3, wherein the screening for a genetic indicator of CHD comprises genetic testing of fetal cells in maternal blood (FCMB), DNA of fetal origin circulating in the maternal blood, preimplantation genetic diagnosis (PDG) during in vitro fertilization (IVF), transcervical retrieval of trophoblast cells, chorionic villus sampling, amniocentesis, or percutaneous umbilical cord blood sampling.

6. The method of claim 5, wherein the genetic indicator of CHD is an abnormality of one or more of: CHD7, ELN, GATA4, GATA6, GDF1, JAG1, NKX2-5, NKX2-6, NOTCH 1, NOTCH2, NR2F2, TBX1, TBX5, TBX20, SEMA3A, FOG2, Ras/Raf, and/or ZIC3.

7. The method of claim 2, wherein the administering is intravenous or oral.

8. The method of any one of claims 1-7, further comprising imaging the heart of the subject via echocardiogram, computed tomography (CT), and magnetic resonance imaging (MRI).

9. The method of any one of claims 1-8, wherein the inhibitor is hydroxyurea (HU), or a structurally similar hydroxamic acid derivative that targets histone deacetylases, matrix metalloproteinases or ribonucleotide reductase (RNR).

10. The method of any one of claims 1-9, wherein the inhibitor is hydroxyurea (HU).

11. The method of any one of claims 1-9, wherein the inhibitor is didox or tridox.

12. The method of any one of claims 1-9, wherein the inhibitor is selected from the group consisting of: P-Rib-PP synthetase inhibitor (MRPP, etc.); APRTase inhibitor (methotrexate, MMPR, etc.), GAR transformylase inhibitor (multitargeted antifolate (LY231514), etc.); AICAR transformylase inhibitor (dmAMT, AG2009, etc.); IMP cyclohydorolase (purine nucleoside 5'- monophosphate derivatives, etc.); IMP dehydrogenase inhibitor (TAD, SAD, BAD, VX-497, etc.); Aspartate transcarbamylase inhibitor (PALA, etc.); DHO DHase inhibitor (lapachol, etc.);

ODCase inhibitor (pyrazofurin, etc.); and CTP synthetase inhibitor (3-Deazauridine, etc.).

13. The method of claim 10, wherein the HU is administered at a dose of about 10 mg/kg.

14. The method of any one of claims 1-13, wherein the inhibitor is administered at a dose of 1 -100 mg/kg body weight of the pregnant female.

15. A method of inducing maturation of cardiomyocytes in vitro, the method comprising culturing stem cells in a chemically-defined medium containing less than 5 mM glucose.

16. The method of claim 15, wherein the stem cells are human embryonic stem cells or human induced pluripotent stem cells.

17. The method of claim 15, wherein the stem cells are cultured in the medium for at least 14 days.