GSK 2837808A

The expression and role of glycolysis-associated molecules in infantile hemangioma

Jian Chen, Dan Wu, Zuoqing Dong, Anwei Chen, Shaohua Liu PII: S0024-3205(20)30967-X
DOI: https://doi.org/10.1016/j.lfs.2020.118215
Reference: LFS 118215

To appear in: Life Sciences

Received date: 16 June 2020
Revised date: 30 July 2020
Accepted date: 3 August 2020

Please cite this article as: J. Chen, D. Wu, Z. Dong, et al., The expression and role of glycolysis-associated molecules in infantile hemangioma, Life Sciences (2020), https://doi.org/10.1016/j.lfs.2020.118215

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

© 2020 Published by Elsevier.

The expression and role of glycolysis-associated molecules in infantile hemangioma
Jian Chen a,b, Dan Wu a,b, Zuoqing Dongb, Anwei Chenb, Shaohua Liua,b

a: Department of Oral and Maxillofacial Surgery, School and Hospital of Stomatology, Cheeloo College of Medicine, Shandong University, Jinan, Shandong,250012, China. b: Department of Oral and Maxillofacial Surgery, Qilu Hospital, Cheeloo College of Medicine, and Institute of Stomatology, Shandong University, Jinan, Shandong,250012, China.

Corresponding Author:
Name: Shaohua Liu
Mailing address: Qilu Hospital, Cheeloo College of Medicine, Shandong University, 250012, No.107, Wenhuaxi-road, Jinan, Shandong, China
Telephone: +86 0531-82166775,82166771
Fax:+86 0531-82169075
E-mail address: [email protected]

Abstract
Aims: Infantile hemangioma (IH) is one of the most common tumors in infancy, which etiology and pathogenesis has not been fully elucidated, hypoxia and abnormal glucose metabolism is regarded as critical pathogenic factors. This study investigated the expression and function of glycolysis-associated molecules (GLUT1, HK2, PFKFB3, PKM2, and LDHA) under normoxic and hypoxic conditions to further understand the pathogenesis of IH.

Main methods: Hemangioma-derived endothelial cells (HemECs) were isolated from proliferating phase infantile hemangiomas and identified by immunofluorescence. HemECs and human umbilical vein endothelial cells (HUVECs) were cultured under normoxic and hypoxic conditions. RNA and protein expression of glycolysis-associated molecules were analyzed by quantitative real-time RT-PCR, western blotting, and immunohistochemistry. Glucose consumption, ATP production and lactate production were measured. Glycolysis-associated molecules were inhibited by WZB117, 3BP, 3PO, SKN, and GSK 2837808A and the resulting effects on HemECs proliferation, migration, and tube formation were quantified.
Key findings: Glycolysis-associated molecules were highly expressed at both mRNA and protein levels in HemECs compared with HUVECs (P<0.05). Glucose consumption and ATP production were higher in HemECs than in HUVECs, while lactate production in HemECs was lower than in HUVECs (P<0.05). Inhibition of some glycolysis-associated molecules reduced the proliferation, migration, and tube formation capacity of HemECs (P<0.05). Significance: Our study revealed that glycolysis-associated molecules were highly expressed in IH. Glucose metabolismin HemECs differed from normal endothelial cells. Altering the expression of glycolysis-associated molecules may influence the phenotype of HemECs and provide new therapeutic approaches to the successful treatment of IH. Key words hemangioma, glycolysis, endothelial cells, glycolysis-associated molecules Word count Abstract: 245 words Introduction: 463 words Discussion: 1203 words Conclusion: 58 words Total：5433 words Figure count : 5 figures 1. Introduction Infantile hemangioma (IH) is one of the most common benign vascular tumors in infancy with a prevalence rate ranging from 3% to 10%[1, 2]. The neoplasms are characterized by rapidly proliferating regions of disorganized angiogenesis, followed by unpredictable spontaneous involution[3, 4]. Common treatment options for IH include oral and injectable medications, laser therapy, and surgical treatment[5]. At present, oral propranolol has emerged as the first-line treatment option[6]; however, the specific etiology and pathogenesis of IH has not been fully elucidated and further research is needed to guide accurate treatment. Energy metabolism is an important component of cell metabolism and serves as the foundation for cellular functions. Altered energy metabolism is closely related to various abnormal cell behaviors. Normal cells produce energy through mitochondrial oxidative phosphorylation under aerobic conditions; however, cancer cells transform glucose into lactate to produce ATP under aerobic conditions via a process known as aerobic glycolysis or Warburg effect[7]. The induction of aerobic glycolysis has long been described as an important hallmark of cancer[8]. Interestingly, endothelial cells also produce most (85%) of their ATP through aerobic glycolysis despite the relative abundance of oxygen in the vascular compartment[9]. Aggregation of immature endothelial cells is a main feature of proliferating hemangiomas, and hemangioma-derived endothelial cells (HemECs), which differentiate from progenitor cells, constitute the major component of IH[10, 11]. Previous studies have shown that the biological functions of HemECs are inconsistent with those of normal endothelial cells. Properties such as adhesion, migration, and proliferation capacity induced by endostatins were different in HemECs compared to human dermal microvascular endothelial cells[10]. HemECs differ from human dermal microvascular endothelial cells in their rates of proliferation and migration in vitro[12]. Angiogenesis and vasculogenesis mediated by endothelial cells are considered to be pivotal pathogenic mechanisms driving new vessel formation in IH[6, 13]. Understanding differences in glycolysis metabolism between HemECs and normal endothelial cells may increase our understanding of the pathogenesis of IH and lead to the development of novel therapies that exploit these differences. The glycolytic pathway comprises a series of reactions. Transcriptional regulation of glycolytic enzymes is known to promote aerobic glycolysis (the Warburg Effect) [14]. Aberrant glucose metabolism by several glycolytic enzymes and glucose metabolism related molecules is believed to be an important pathogenic mechanism. In this study, we examined the glycolysis-associated molecules glucose transporter 1 (GLUT1), hexokinase 2 (HK2), 6-phosphofructo-2-kinase/fructose-2,6-bisphosphatase 3 (PFKFB3), pyruvate kinase M2 isoform (PKM2) and lactate dehydrogenase A (LDHA) because they have been shown to be vital molecules regulating glycolytic flux and play important roles in carcinogenesis[15-17]. We focused on the expression of glycolysis-associated molecules of IH with emphasis on HemECs and detected glycolytic metabolism changes and their influence on proliferation, angiogenesis and migration underlying the inhibition of glycolysis. We hypothesized that targeting glycolysis in HemECs may provide alternative therapeutic opportunities for IH. 2. Materials and methods 2.1 Isolation and cultivation of HemECs Hemangioma specimens were obtained with signed informed consent and approval of the Ethics Committee of Qilu Hospital, Shandong University. Clinical diagnoses were confirmed in the Department of Pathology at Qilu Hospital. HemECs were isolated from proliferating phase specimens using previously described methods[18]. Cells were cultured in endothelial cell medium (ECM; ScienCell™ Research Laboratories, Carlsbad, CA, USA) containing 10% fetal bovine serum (FBS; ScienCell), 100 U/ml penicillin (ScienCell), and 100 μg/ml streptomycin (ScienCell) at 37°C, 5% carbon dioxide, and 95% air. Cells collected from passages three to five were harvested for experiments. Human umbilical vein endothelial cells (HUVECs) were purchased from the Chinese Academy of Sciences (CAS) Cell Bank (Shanghai, China). Culture conditions were identical to those used for HemECs. For controls, the HemECs and HUVECs were cultured at 1% oxygen, 5% carbon dioxide and 94% nitrogen to simulate a hypoxic environment. 2.2 Immunofluorescence characterization of HemECs HemECs and HUVECs were seeded in 24-well plates (5x103 cells/well). Cells were fixed with 4% paraformaldehyde for 10 min, permeated with 0.1% Triton X-100 for 5 min, then blocked with 10% goat serum for 30 min. The primary antibodies anti-von Willebrand Factor Rabbit Monoclonal Antibody (vWF; BOSTER, Wuhan, China, 1:100), anti-cluster of differentiation 31 (CD31; BOSTER, 1:100), Rabbit Monoclonal Antibody (BOSTER, 1:100), and anti-α-Smooth Muscle Actin antibody (α-SMA; BOSTER, 1:100) were incubated at 4°C overnight, then incubated with Goat Anti-Human IgG Secondary Antibody (BOSTER, 1:50) conjugated with fluorescein isothiocyanate (FITC) fluorescence for 60 min. Nuclei were stained with 4’,6-diamidino-2-phenylindole (DAPI; BOSTER) for 5 min. Cells were then photographed with a fluorescence microscope (Leica, Wetzlar, Germany). 2.3 Immunohistochemical detection of GLUT1, HK2, PFKFB3, PKM2 and LDHA in IH tissue Immunohistochemical (IHC) staining was performed to investigate protein expression in IH tissue. Normal skin and subcutaneous tissue around the tumor were used as controls. Tissues embedded in paraffin were dried at 65˚C for 1 h, deparaffinized in xylene and rehydrated in graded alcohol. Antigen retrieval was carried out with citrate buffer for 20 min at high temperature. Endogenous peroxidase activity was inhibited with 3% hydrogen peroxide for 10 min. Non-specific binding was blocked with 10% goat serum for 10 min at room temperature. The primary antibodies were incubated at 4°C overnight, then incubated with the secondary antibody and stained using the ready-to-use SABC-AP kit (BOSTER). Staining was observed under a phase-contrast microscope (Leica). The primary antibodies were: Anti-GLUT1 Rabbit Monoclonal Antibody (GLUT1; BOSTER, 1:200); Anti-Hexokinase II/HK2 Antibody (HK2; BOSTER, 1:200); Rabbit Anti-PFKFB3 (PFKFB3; Abcam, Cambridge, UK, 1:200); Anti-PKM2 Rabbit Monoclonal Antibody (PKM2; BOSTER, 1:200); Anti-Lactate Dehydrogenase Rabbit Monoclonal Antibody (LDHA; BOSTER, 1:200). 2.4 Total RNA extraction and reverse transcription-quantitative PCR (RT-qPCR) RNA was quantitatively analyzed to compare the expression of glycolysis-associ ated molecules in the cultured HemECs and HUVECs. Total RNA was extracte d from HemECs and HUVECs using the Trizol reagent (Invitrogen, Waltham, MA, USA) and then reverse transcribed into complementary DNA (cDNA) by the ReverTra Ace™ qPCR RT kit (TOYOBO, Shanghai, China). The mRNA e xpression levels of GLUT1, HK2, PFKFB3, PKM2 and LDHA in the cultured HemECs and HUVECs were detected by SYBR Green Realtime PCR Master Mix (TOYOBO) on a Mastercycler® ep realplex assay system (Eppendorf, Ham burg, Germany). The data were normalized to β-Actin as the housekeeping gen e. Relative gene expression was calculated by the 2(−ΔΔCt) method. Primer seque nces are listed below. β-Actin, forward, 5′- GTCATTCCAAATATGAGATGCG T -3′, reverse, 5′- GCTATCACCTCCCCTGTGTG-3′. GLUT1, forward, 5′- ATG GGCTTCTCGAAACTGGG-3′, reverse, 5′- CAGGTCCTTGTTGCCCATGA-3′. H K2, forward,5′AACAGCCTGGACGAGAGCATC-3′, reverse, 5′-AGGTCAAACT CCTCTCGCCG-3′. PFKFB3, forward, 5′-AGCCCGGATTACAAAGACTGC-3′, r everse, 5′-GGTAGCTGGCTTCATAGCAAC-3′. PKM2, forward: 5′-GCCATAAT CGTCCT CACCAAGT-3′, reverse: 5′-GCACGTGGGCGGTATCTG-3′. LDHA, f orward, 5′-AGCTGTTCCACTTAAGGCCC-3’, reverse, 5′-TGGAACCAAAAGG AATCGGGA-3′. 2.5 Total protein extraction and Western blotting Protein levels were analyzed to compare the expression of glycolysis-associated molecules in the cultured HemECs and HUVECs. HemECs and HUVECs were lysed using RIPA/PMSF Buffer (Beyotime Biotechnology, Shanghai, China) and protein concentrations were measured by a BCA Protein Assay kit (Thermo Fisher Scientific, Waltham, MA, USA). Subsequently, protein samples (20 μg/lane) were separated by 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE; Beyotime Biotechnology) and then transferred to nitrocellulose membranes (Millipore, Billerica, MA, USA). The membranes were blocked with 5% skimmed milk for 1 h at room temperature, then incubated with the primary antibody overnight at 4°C. Next, the membranes were incubated for 1 h at room temperature with horseradish peroxidase (HRP)-conjugated goat anti-rabbit IgG secondary antibody (BOSTER, 1:10000) or goat anti-mouse IgG secondary antibody (BOSTER, 1:10000). Finally, protein signals were visualized by an enhanced chemiluminescence (ECL) detection system (Thermo Fisher Scientific). Images were captured by a Tanon detection system (GE Healthcare, Waukesha, USA). Protein band density was estimated by ImageJ software (National Institutes of Health, Bethesda, MD, USA). β-Actin was used as a loading control. Variation in band density was shown as fold changes compared to the control in the blot after normalization to β-Actin. The primary antibodies were: Anti-beta-Actin Rabbit Monoclonal Antibody (β-Actin; BOSTER, Wuhan, China, 1:1000); Anti-GLUT1 Rabbit Monoclonal Antibody (1:500); Anti-Hexokinase II/HK2 Antibody (1:1000); Rabbit Anti-PFKFB3 (1:2000) antibody Anti-PKM2 Rabbit Monoclonal Antibody (1:1000); Anti-Lactate Dehydrogenase Rabbit Monoclonal Antibody (1:400) 2.6 Measurement of glucose consumption, ATP production and lactate production HemECs and HUVECs were incubated in 6-well plates with 1 × 105 cells/well for 48 h under normoxic and hypoxic conditions. Culture medium from the HemECs and HUVECs was then collected to determine glucose consumption and lactate production. Cell lysate was collected to measure ATP production. Glucose, ATP and lactate levels were measured by the Glucose Uptake Colorimetric Assay kit (BioVision, Milpitas, CA, USA), ATP Colorimetric/Fluorometric Assay kit (BioVision) and Lactate Colorimetric/Fluorometric Assay kit (BioVision) according to the manufacturer’s instructions. 2.7 Drugs We used 2-fluoro-6-(m-hydroxybenzoyloxy) phenyl m-hydroxybenzoate (WZB117)[19] as a specific inhibitor of GLUT1, 3-bromopyruvate (3BP)[20] as an inhibitor of HK2, 3-(3-pyridinyl)-1-(4-pyridinyl)-2-propen-1-one] (3PO)[21] as a small molecule inhibitor of PFKFB3, shikonin (SKN)[22] as a specific inhibitor of PKM2, and GSK 2837808A[23] as an inhibitor of LDHA. The inhibitors were purchased through Target Molecule (Target Molecule, Wellesley, MA, USA). All inhibitors were used at a concentration of 10 umol/L.

2.8 Cell Counting Kit-8 (CCK-8) assay
Cell proliferation was evaluated using the Cell Counting Kit-8 (CCK-8; Corning Corporation, Corning, NY, USA) following the manufacturer’s protocols. HemECs were planted in 96-well plates with 5 × 103 cells/well for 24 h. Following medium exchange, glycolysis-associated molecule inhibitors were added to the new culture medium and the cells were incubated for 24 h. CCK-8 solution (10 μl) was then added to each well and incubated for another 2 h at 37°C. The cell optical density value at

450 nm (OD450) was determined by a microplate reader (Molecular Devices, Sunnyvale, CA, USA).

2.9 Wound healing assay
Wound healing assays were performed to evaluate the effect of glycolysis-associated molecules on HemECs migratory capacity. HemECs were planted in 6-well plates with 2 × 105 cells/well and grown to near 90% confluency. A straight scratch was made across the middle of each well with a 100 μl pipette. The culture medium was removed and new medium containing glycolysis-associated molecule inhibitors was introduced to each well. After a 48 h incubation, migration of HemECs was observed under a phase-contrast microscope (Leica). Results were calculated as the percentage of the original wound enclosed by cells, and a picture of each wound was quantified using ImageJ software.

2.10 Tube formation assay
Tube formation assays were used to evaluate the effect of glycolysis-associated molecules on the angiogenesis capacity of HemECs. A 50 μl aliquot of Matrigel™ (BD Biosciences, Franklin Lakes, NJ, USA) was added to each well of cold 96-well plates and incubated at 37°C for 30 min to allow gel polymerization. The HemECs suspensions (2 × 104 cells/well) containing glycolysis-associated molecule inhibitors were plated on the matrigel cover and incubated at 37°C for 12 h. The canalization assay of HemECs was observed under a phase-contrast microscope (Leica). The total junctions, total branching length, total meshes, and mean mesh size were quantified by ImageJ software.

2.11 Statistical analysis
Each experiment was performed in triplicate and data are presented as mean ± standard deviation (SD). Statistical computations were performed by GraphPad Prism
8.0 (GraphPad Software, San Diego, CA, USA). Statistical analyses consisted of

t-tests for comparisons between two groups or one-way analysis of variance (ANOVA) for multiple groups. The Student–Newman–Keuls (SNK) test for ANOVA was applied to determine levels of significance between groups. A P-value <0.05 was considered statistically significant. 3. Results 3.1 Immunohistochemical detection of GLUT1, HK2, PFKFB3, PKM2 and LDHA in IH tissue Immunohistochemistry was used to determine the expression of glycolysis-associated molecules in IH tissues. Normal skin and subcutaneous tissues were used as controls. Elevated immunohistochemical staining was detected in IH tissues, while staining was minimal in normal skin and subcutaneous tissues, except for some basal layer cells and vessel wall cells (Fig 1). The expression of GLUT1 was limited to luminal cells on the inner vessel walls. Staining was mainly cytoplasmic, with no obvious staining of nuclei. PKM2 was mainly expressed in the cytoplasm of cells which around the irregular vascular lumens. HK2, PFKFB3 and LDHA were expressed in multiple lumens and surrounding tissues. Most of the HK2 and LDHA positive staining was observed in the cytoplasm, and PFKFB3 stained in both the cytoplasm and nuclei. 3.2 Identification of HemECs CD31 and vWF were used as endothelial cell markers. α-SMA was used as a marker of mesenchymal cells because they are the most common contaminants in endothelial cell cultures from IH tissues. HUVECs served as controls for normal endothelial cells. The expression of α-SMA, CD31 and vWF was examined by immunofluorescence staining (Fig 2). HemECs formed polygonal structures similar to "paving stones". The α-SMA assays showed negative expression in both HUVECs and HemECs. Expression of CD31 and vWF was consistent between HemECs and HUVECs; the cytoplasm was stained by FITC (green) and cell nuclei were stained with DAPI (blue). 3.3 The glycolytic metabolism of HemECs and HUVECs We analyzed glycolytic metabolism in HemECs and HUVECs under normoxic and hypoxic conditions to measure glycolytic activity. Glucose consumption (P<0.05) and ATP production (P<0.01) were higher in HemECs than in HUVECs, while lactate production in HemECs was lower than that observed in HUVECs (P<0.01). Glycolytic metabolism in HemECs was not significantly different between normoxic and hypoxic conditions (Fig 3A, B, C). 3.4 Expression of glycolysis-associated molecules in cultured HemECs and HUVECs To evaluate variation in glycolytic metabolism between HemECs and normal endothelial cells, we examined expression of GLUT1, HK2, PFKFB3, PKM2 and LDHA under normoxic and hypoxic conditions in HemECs and HUVECs by RT-qPCR and WB methods. Total RNA and protein were extracted from cells cultured for 48 h in normoxic and hypoxic conditions. Relative mRNA and protein expression levels of GLUT1, HK2, PFKFB3, PKM2, LDHA and HIF-1a were normalized to that of β-Actin. The results showed that HemECs had higher glycolysis-associated molecules mRNA expression compared with HUVECs (GLUT1, P<0.01; HK2, P<0.05; PFKFB3, P<0.001; LDHA, P<0.05; HIF-1a, P<0.05) (Fig 4A). Under normoxic conditions, levels of GLUT1 (P<0.01), HK2 (P<0.01), PFKFB3 (P<0.001), PKM2 (P<0.001), LDHA (P<0.05) and HIF-1a (P<0.001) protein were greater in HemECs than HUVECs. Under hypoxic conditions, protein levels of HK2 (P<0.001), PFKFB3 (P<0.05), PKM2 (P<0.01) and HIF-1a (P<0.001) were elevated in HemECs; no significant differences were observed for GLUT1 and LDHA under normoxic versus hypoxic conditions in HemECs (Fig 4B). Representative bands of glycolysis-associated molecules expression in HemECs and HUVECs under different culture conditions are shown in Fig 4C. 3.5 Glycolysis-associated molecules regulate the proliferation, angiogenesis, and migration of HemECs To further investigate the function of glycolysis-associated molecules in regulating the biological function of IH, we assessed proliferation, migration, and angiogenesis in HemECs. A CCK8 assay used to detect cell proliferation of HemECs treated with glycolysis-associated molecule inhibitors for 24 h revealed that proliferation of HemECs may be attenuated by inhibiting HK2 (P<0.001) and PFKFB3 (P<0.001). Inhibition of GLUT1, PKM2 and LDHA produced no obvious effects (Fig 5A). The wound healing assay, used to evaluate the migratory capacity of HemECs treated with glycolysis-associated molecule inhibitors for 48 h, showed that migration of HemECs was attenuated by inhibiting GLUT1 (P<0.001), HK2 (P<0.01), PFKFB3 (P<0.01), PKM2 (P<0.05) and LDHA (P<0.01) (Fig 5B, C). In addition, tube formation assays on Matrigel® showed that inhibiting glycolysis-associated molecules decreased the amount of vessel formation in HemECs. Vessel junctions were significantly reduced by inhibiting GLUT1 (P<0.001), HK2 (P<0.001), PFKFB3 (P<0.001) and PKM2 (P<0.01). Inhibition of LDHA produced no significant changes. Total branching length was reduced by glycolysis-associated molecule inhibitors and significantly lowered by inhibiting HK2 (P<0.001) and PFKFB3 (P<0.001). The number of meshes was reduced by interference from the inhibitors, while mean mesh size increased. Meshes in the control group were dense and slender, while meshes in the experimental groups were larger and sparser, except when LDHA was inhibited (Fig 5D, E). Taken together, the data indicated that inhibition of glycolysis-associated molecules modulated the proliferation, migration, and angiogenesis of HemECs. 4. Discussion Although glycolysis has been extensively studied in cancer, no information is available pertaining to the expression of glycolysis-associated molecules and their biological function in IH. In this study, we (1) examined the expression of crucial glycolytic associated molecules (GLUT1, HK2, PFKFB3, PKM2 and LDHA) in IH, (2) detected differences in expression between HemECs and normal endothelial cells, (3) assayed differences in glucose metabolism between HemECs and normal endothelial cells, and (4) explored the influence of glycolysis-associated molecules on the biological function of HemECs. Our data indicated that the expression of glycolysis-associated molecules was upregulated in IH tissue and HemECs. Glycolytic metabolism of HemECs was different from normal endothelial cells. Glycolysis-associated molecules were essential regulators of proliferative activities, angiogenesis, and migration capacity in HemECs. Mechanisms that promote the Warburg effect operate through transcriptional regulation of glycolysis-associated molecules. Hypoxia can increase the expression of transcription factor HIF-1. Upregulated HIF-1a directly accelerates the expression of most molecules involved in the glycolytic pathway[24]. Similarly, our study showed that the expression of glycolysis-associated molecules in HemECs was elevated in an anaerobic incubation environment. Glucose metabolism is mainly mediated by glycolysis-associated molecules, and inhibition of these molecules may alter disease progression. GLUT1 belongs to the membrane-associated carrier family, which mediates the transport of glucose into cells. GLUT1 is the first rate-limiting step in glucose metabolism and is responsible for basal glucose uptake, which usually occurs in most tissues under normal conditions. In most cancers, GLUT1 expression has been associated with poor survival[16, 25]. GLUT1 is also a marker that distinguishes hemangiomas from other vascular malformations and tumors. Immunohistochemical staining has shown that GLUT1 is located in the endothelium of hemangiomas[26], which is consistent with our results. GLUT1 is also highly expressed in endothelial and epithelial cells from blood-tissue barriers[27]. Our research detected higher levels of GLUT1 expression in HemECs than in HUVECs. Overall, expression of GLUT1 was not increased under hypoxic conditions. Inhibition of GLUT1 appeared to weaken the angiogenesis and migration of HemECs, but there was no influence on proliferation of HemECs. This observation was inconsistent with a previous study showing that GLUT1 expression and tumor cell proliferation were significantly correlated based on Ki-67 labeling[28]. This was a research on pulmonary pleomorphic carcinoma, which is a malignant tumor that progresses rapidly with low survival rate, while infantile hemangioma is a benign tumor characterized by rapidly proliferation and unpredictable spontaneous involution. Glucose metabolism demand for tumor cells between them should be different. The correlation between GLUT1 and proliferation might also be inconsistent. HK2 is a key rate-limiting enzyme of glucose metabolism that catalyzes the phosphorylation of glucose to glucose 6-phosphate. HK2 is overexpressed in a variety of tumor cells[29]. Our results showed that HK2 expression was higher in HemECs compared to HUVECs. HemECs cultured under hypoxia showed elevated levels of HK2 expression. Inhibition of HK2 dramatically reduced proliferation, angiogenesis and migration of HemECs. HK2 induced cell stemness properties, proliferation, migration, invasion, and lactate production in cancer cells. Expression of HK2 has been significantly associated with advanced stage and high-grade cancers[30-32]. Furthermore, HK2 inhibits release of cytochrome c to block mitochondria-dependent apoptosis, a dominant condition for glycolysis[33]. PFKFB3, the most effective isoenzyme in the PFKFB family, synthesizes fructose 2,6-bisphosphate (Fru-2,6-P2). PFKFB3 has a higher kinase-to-phosphatase ratio (740:1) and promotes greater production of Fru-2,6-P2 compared to other isoenzymes[34]. PFKFB3 acts as the allosteric activator of 6-phosphofructo1-kinase (PFK-1), the rate-limiting enzyme for glycolysis[35]. PFKFB3 is mainly expressed in vascular endothelial cells and tumor cells[36, 37]. Recent studies have shown that genetic silencing of PFKFB3 reduced angiogenesis in physiological and pathological conditions[9, 38]. Blockade of endothelial PFKFB3 normalized tumor vessels and reduced cancer cell invasion and metastasis, but had no effect on tumor growth[39]. Researchers have shown that PFKFB3 inhibition in cancer cells significantly inhibits cell survival, growth, and invasion[40]. Similarly, we observed that (1) inhibition of PFKFB3 in HemECs substantially inhibited cell growth, angiogenesis and migration, (2) expression of PFKFB3 was increased in HemECs, and (3) in a hypoxic environment, expression of PFKFB3 was elevated. Pyruvate kinase is the final rate-limiting enzyme in glycolysis. The PKM2 isoenzyme is expressed in cells with high nucleic acid synthesis rates, such as proliferating cells, embryonic cells and cancer cells[41]. Depending on allosteric properties, PKM2 can catalyze dephosphorylation of phosphoenolpyruvate (PEP) to produce pyruvate and ATP[42] or function as a transcription factor to activate the transcription of certain genes[43]. Although PKM2 is elevated in cancer cells, inactivation of PKM2 is important for cancer cell proliferation[44]. Although PKM2 may not be required for tumor growth or cell proliferation in PKM2-deficient mice, knockout of PKM2 may promote more rapid breast cancer development[45]. Research has shown that attenuated PKM2 activity may promote the gathering of glycolytic intermediates that drive biosynthetic pathways in rapidly proliferating cells[46]. Nonmetabolic functions of PKM2 may play a substantial role in cancer cell proliferation and tumor development. [47]. Our research indicated that PKM2 expression was upregulated in HemECs compared to HUVECs, and that PKM2 was more highly expressed under hypoxic conditions. Inhibition of PKM2 reduced angiogenesis and migration in HemECs but did not alter cell proliferation. The effect of PKM2 inhibition on HemECs proliferation was consistent with previous results in PKM2-deficient mice[45]. LDHA catalyzes the terminal reaction of glycolysis by converting pyruvate to lactate[48]. Elevated levels of LDHA found in various cancers are involved in regulating cancer cell proliferation, angiogenesis, metastasis and immune escape[49]. Our research showed that LDHA expression in HemECs was not statistically different from HUVECs, even though lactate production in HemECs was lower than that in HUVECs. Previous studies showed that reduced lactate levels indicate the accumulation of pyruvate, and excess pyruvate facilitates the mitochondrial TCA cycle transport chain or lipogenesis[50]. HemECs had higher glucose uptake and ATP production but less lactate accumulation than HUVECs even though the expression of glycolysis-related molecules in HemECs was upregulated. On the other hand, glucose metabolism in hemangioma-derived endothelial cells may not be significantly altered. Glucose metabolism may be repressed under certain conditions, allowing a portion of IH lesions to spontaneously regress. Inhibition of LDHA in our research did not inhibit proliferation of HemECs or diminish the total number of angiogenic junctions. Although mean mesh size did not differ between the experimental and control groups, migration capability decreased. Thus, inhibiting LDHA had little impact on the biological function of HemECs. HemECs are directly involved in the proliferation and involution phases of IH. Endothelial cells are in high glycolytic flux, and the rate of endothelial glycolysis is equal to or greater than that of cancer cells[9]. In the treatment of hemangioma, propranolol affects endothelial cells, angiogenesis, vascular tone, and apoptosis[51]. Propranolol acts on endothelial cells to impede endothelial cell growth factor receptor-2 (VEGFR-2), resulting in reduced proliferation and migration and increased apoptosis[52]. VEGF blockade is a clinically attractive strategy that has been approved for anti-angiogenesis treatment[53]; however, VEGF blockade inhibits certain angiogenic signals while upregulating other proangiogenic factors. This process may result in an escape from anti-angiogenesis therapy and resumption of tumor vascularization[54]. Targeting glucose metabolism in HemECs, will limit the energy supply for angiogenesis. No matter how many angiogenic signal pathways are present, angiogenesis in IH will be reduced. Therefore, our results may provide a theoretical basis for the discovery of novel clinical treatments for IH. 5. Conclusion In conclusion, this study investigated alterations in glucose metabolism in HemECs compared to normal endothelial cells and changes in the biological behavior of HemECs after interference by glycolysis-associated molecules. The results provide direct evidence that glycolysis-associated molecules are crucial for the development of IH. These findings may increase our understanding of the molecular mechanisms involved in IH progression. Acknowledgments This study was supported by the Development Funding for Novel Clinical Technology, Qilu Hospital of Shandong University (2019-17), Education Reform Project of Shandong University (2019Y264)，and the Key Research & Development Project of Shandong Province (2019GSF108272). Conflict of interest statement All authors declare that there are no conflicts of interest. References [1] C Leaute-Labreze, P Hoeger, J Mazereeuw-Hautier, L Guibaud, E Baselga, G Posiunas, RJ Phillips, H Caceres, et al., A randomized, controlled trial of oral propranolol in infantile hemangioma, N Engl J Med, 372 (2015) 735-746. [2] A Munden, R Butschek, WL Tom, JS Marshall, DM Poeltler, SE Krohne, AB Alió, M Rit ter, et al., Prospective study of infantile haemangiomas: incidence, clinical characteristics an d association with placental anomalies, Br J Dermatol, 170 (2014) 907-913. [3] MM Tollefson, IJ Frieden, Early growth of infantile hemangiomas: what parents' photographs tell us, Pediatrics, 130 (2012) e314-320. [4] K Takahashi, JB Mulliken, HP Kozakewich, RA Rogers, J Folkman, RA Ezekowitz, Cellular markers that distinguish the phases of hemangioma during infancy and childhood, J Clin Invest, 93 (1994) 2357-2364. [5] ZY Chen, QN Wang, YH Zhu, LY Zhou, T Xu, ZY He, Y Yang, Progress in the treatment of infantile hemangioma, Ann Transl Med, 7 (2019) 692 [6] C Leaute-Labreze, JI Harper, PH Hoeger, Infantile haemangioma, Lancet, 390 (2017) 85-94. [7] O WARBURG, On the origin of cancer cells, Science, 123 (1956) 309-314. [8] D Hanahan, RA Weinberg, Hallmarks of cancer: the next generation, Cell, 144 (2011) 646-674. [9] K De Bock, M Georgiadou, S Schoors, A Kuchnio, BW Wong, AR Cantelmo, A Quaegebeur, B Ghesquiere, et al., Role of PFKFB3-driven glycolysis in vessel sprouting, Cell, 154 (2013) 651-663. [10] ZA Khan, JM Melero-Martin, X Wu, S Paruchuri, E Boscolo, JB Mulliken, J Bischoff, Endothelial progenitor cells from infantile hemangioma and umbilical cord blood display unique cellular responses to endostatin, Blood, 108 (2006) 915-921. [11] S Greenberger, J Bischoff, Pathogenesis of infantile haemangioma, Br J Dermatol, 169 (2013) 12-19. [12] E Boye, Y Yu, G Paranya,JB Mulliken, BR Olsen, J Bischoff, Clonality and altered behavior of endothelial cells from hemangiomas, J Clin Invest, 107 (2001) 745-752. [13] P Mabeta, Oncosuppressors and Oncogenes: Role in Haemangioma Genesis and Potential for Therapeutic Targeting, Int J Mol Sci , 19 (2018). [14] NY Spencer, RC Stanton, The Warburg Effect, Lactate, and Nearly a Century of Trying to Cure Cancer, Semin Nephrol, 39 (2019) 380-393. [15] XB Li, JD Gu, QH Zhou, Review of aerobic glycolysis and its key enzymes - new targets for lung cancer therapy, Thorac Cancer , 6 (2015) 17-24. [16] CC Barron, PJ Bilan, T Tsakiridis, E Tsiani, Facilitative glucose transporters: Implications for cancer detection, prognosis and treatment, Metabolism, 65 (2016) 124-139. [17] N Lypova, S Telang, J Chesney, Y Imbert-Fernandez, Increased 6-phosphofructo-2-kinase/fructose-2,6-bisphosphatase-3 activity in response to EGFR signaling contributes to non-small cell lung cancer cell survival, J Biol Chem, 294 (2019) 10530-10543. [18] ST Tan, M Velickovic, BM Ruger, PF Davis, Cellular and extracellular markers of hemangioma, Plast Reconstr Surg, 106 (2000) 529-538. [19] Y Liu, Y Cao, W Zhang, S Bergmeier, Y Qian, H Akbar, R Colvin, J Ding, et al., A small-molecule inhibitor of glucose transporter 1 downregulates glycolysis, induces cell-cycle arrest, and inhibits cancer cell growth in vitro and in vivo, Mol Cancer Ther, 11 (2012) 1672-1682. [20] S Cardaci, E Desideri, MR Ciriolo, Targeting aerobic glycolysis: 3-bromopyruvate as a promising anticancer drug, J Bioenerg Biomembr, 44 (2012) 17-29. [21] B Clem, S Telang, A Clem, A Yalcin, J Meier, A Simmons, MA Rasku, S Arumugam, et al., Small-molecule inhibition of 6-phosphofructo-2-kinase activity suppresses glycolytic flux and tumor growth, Mol Cancer Ther, 7 (2008) 110-120. [22] C Guo, J He, X Song, L Tan, M Wang, P Jiang, Y Li, Z Cao, C Peng, Pharmacological properties and derivatives of shikonin-A review in recent years, Pharmacol Res 149 (2019) 104463. [23] E Brighenti, D Carnicelli, M Brigotti, L Fiume, The inhibition of lactate dehydrogenase A hinders the transcription of histone 2B gene independently from the block of aerobic glycolysis, Biochem Biophys Res Commun, 485 (2017) 742-745. [24] GL Semenza, Hypoxia-inducible factors: mediators of cancer progression and targets for cancer therapy, Trends Pharmacol Sci, 33 (2012) 207-214. [25] ML Macheda, S Rogers, JD Best, Molecular and cellular regulation of glucose transporter (GLUT) proteins in cancer, J Cell Physiol, 202 (2005) 654-662. [26] L Huang, H Nakayama, M Klagsbrun, JB Mulliken, J Bischoff, Glucose transporter 1-positive endothelial cells in infantile hemangioma exhibit features of facultative stem cells, Stem Cells, 33 (2015) 133-145. [27] FQ Zhao, AF Keating, Functional properties and genomics of glucose transporters, Curr Genomics, 8 (2007) 113-128. [28] H Imai, K Kaira, H Endoh, K Imaizumi, Y Goto, M Kamiyoshihara, T Kosaka, T Yajima, et al., Prognostic Significance of Glucose Metabolism as GLUT1 in Patients with Pulmonary Pleomorphic Carcinoma, J Clin Med, 9 (2020). [29] E Claeyssen, J Rivoal, Isozymes of plant hexokinase: occurrence, properties and functions, Phytochemistry, 68 (2007) 709-731. [30] MKY Siu, YX Jiang, JJ Wang, THY Leung, CY Han, BK Tsang, ANY Cheung, HYS Ngan, KKL Chan, Hexokinase 2 Regulates Ovarian Cancer Cell Migration, Invasion and Stemness via FAK/ERK1/2/MMP9/NANOG/SOX9 Signaling Cascades, Cancers (Basel), 11 (2019). [31] W Dai, X Meng, S Mo, W Xiang, Y Xu, L Zhang, R Wang, Q Li, G Cai, FOXE1 represses cell proliferation and Warburg effect by inhibiting HK2 in colorectal cancer, Cell Commun Signal, 18 (2020) 7. [32] W Du, N Liu, Y Zhang, X Liu, Y Yang, W Chen, Y He, PLOD2 promotes aerobic glycolysis and cell progression in colorectal cancer by upregulating HK2, Biochem Cell Biol, (2019). [33] K Nakajima, I Kawashima, M Koshiisi, T Kumagai, M Suzuki, J Suzuki, T Mitsumori, K Kirito, Glycolytic enzyme hexokinase II is a putative therapeutic target in B-cell malignant lymphoma, Exp Hematol, 78 (2019) 46-55.e43. [34] SG Kim, NP Manes, MR El-Maghrabi, YH Lee, Crystal structure of the hypoxia-inducible form of 6-phosphofructo-2-kinase/fructose-2,6-bisphosphatase (PFKFB3): a possible new target for cancer therapy, J Biol Chem, 281 (2006) 2939-2944. [35] MH Rider, L Bertrand, D Vertommen, PA Michels, GG Rousseau, L Hue, 6-phosphofructo-2-kinase/fructose-2,6-bisphosphatase: head-to-head with a bifunctional enzyme that controls glycolysis, Biochem J., 381 (2004) 561-579. [36] Y Cao, X Zhang, L Wang, Q Yang, Q Ma, J Xu, J Wang, L Kovacs, et al., PFKFB3-mediated endothelial glycolysis promotes pulmonary hypertension, Proc Natl Acad Sci USA, 116 (2019) 13394-13403. [37] L Lu, Y Chen, Y Zhu, The molecular basis of targeting PFKFB3 as a therapeutic strategy against cancer, Oncotarget, 8 (2017) 62793-62802. [38] S Schoors, K De Bock, Anna R Cantelmo, M Georgiadou, B Ghesquière, S Cauwenberghs, A Kuchnio, Brian W Wong, et al., Partial and Transient Reduction of Glycolysis by PFKFB3 Blockade Reduces Pathological Angiogenesis, Cell Metab, 19 (2014) 37-48. [39] AR Cantelmo, LC Conradi, A Brajic, J Goveia, J Kalucka, A Pircher, P Chaturvedi, J Hol, et al., Inhibition of the Glycolytic Activator PFKFB3 in Endothelium Induces Tumor Vessel Normalization, Impairs Metastasis, and Improves Chemotherapy, Cancer Cell, 30 (2016) 968-985. [40] A Yalcin, S Telang, B Clem, J Chesney, Regulation of glucose metabolism by 6-phosphofructo-2-kinase/fructose-2,6-bisphosphatases in cancer, J Experimental molecular pathology, 86 (2009) 174-179. [41] R Ren, J Guo, J Shi, Y Tian, M Li, H Kang, PKM2 regulates angiogenesis of VR-EPCs through modulating glycolysis, mitochondrial fission, and fusion, J Cell Physiol, (2020). [42] G van Niekerk, AM Engelbrecht, Role of PKM2 in directing the metabolic fate of glucose in cancer: a potential therapeutic target, Cell Oncol (Dordr), 41 (2018) 343-351. [43] Z Zhang, X Deng, Y Liu, Y Liu, L Sun, F Chen, PKM2, function and expression and regulation, Cell Biosci, 9 (2019) 52. [44] EK Wiese, T Hitosugi, Tyrosine Kinase Signaling in Cancer Metabolism: PKM2 Paradox in the Warburg Effect, Front Cell Dev Biol, 6 (2018) 79. [45] WJ Israelsen, TL Dayton, SM Davidson, BP Fiske, AM Hosios, G Bellinger, J Li, Y Yu, et al., PKM2 isoform-specific deletion reveals a differential requirement for pyruvate kinase in tumor cells, Cell, 155 (2013) 397-409. [46] SY Lunt, MG Vander Heiden, Aerobic glycolysis: meeting the metabolic requirements of cell proliferation, Annu Rev Cell Dev Biol, 27 (2011) 441-464. [47] X Gao, H Wang, JJ Yang, X Liu, ZR Liu, Pyruvate kinase M2 regulates gene transcription by acting as a protein kinase, Mol Cell, 45 (2012) 598-609. [48] MK Schwartz, Enzymes as prognostic markers and therapeutic indicators in patients with cancer, Clin Chim Acta, 206 (1992) 77-82. [49] Y Feng, Y Xiong, T Qiao, X Li, L Jia, Y Han, Lactate dehydrogenase A: A key player in carcinogenesis and potential target in cancer therapy, Cancer Med, 7 (2018) 6124-6136. [50] S Sur, H Nakanishi, C Flaveny, JE Ippolito, J McHowat, DA Ford, RB Ray, Inhibition of the key metabolic pathways, glycolysis and lipogenesis, of oral cancer by bitter melon extract, Cell Commun Signal, 17 (2019) 131. [51] CH Storch, PH Hoeger, Propranolol for infantile haemangiomas: insights into the molecular mechanisms of action, Br J Dermatol, 163 (2010) 269-274. [52] J Stiles, C Amaya, R Pham, RK Rowntree, M Lacaze, A Mulne, J Bischoff, V Kokta, et al., Propranolol treatment of infantile hemangioma endothelial cells: A molecular analysis, Exp Ther Med, 4 (2012) 594-604. [53] M Singh, SS Couto, WF Forrest, A Lima, JH Cheng, R Molina, JE Long, P Hamilton,et al., Anti-VEGF antibody therapy does not promote metastasis in genetically engineered mouse tumour models, J Pathol, 227 (2012) 417-430. [54] X Li, X Sun, P Carmeliet, Hallmarks of Endothelial Cell Metabolism in Health and Disease, Cell Metab, 30 (2019) 414-433. Figure legends Fig.1 Immunohistochemical detection of glycolysis-associated molecules in IH. Abundant expression of GLUT1, HK2, PFKFB3, PKM2 and LDHA in infantile hemangioma compared with normal skin and subcutaneous tissues. (magnification × 200). Fig. 2 Identification of HemECs. The expression of α-SMA, CD31 and vWF was examined by immunofluorescence staining in cultured cells. No α-SMA expression was observed in ECs. CD31 and vWF were positively expressed in HemECs and HUVECs; the cytoplasm was stained by FITC (green) and cell nuclei were stained with DAPI (blue) (magnification × 200). Fig.3 Glycolytic metabolism in HemECs and HUVECs. Glucose uptake and ATP production were higher in HemECs than in HUVECs. Lactate production in HemECs was lower than that in HUVECs. Glycolytic metabolism in HemECs was not significantly different between normoxic and hypoxic conditions (ns = no significant difference, *p < 0.05, **p < 0.01, ***p < 0.001). Fig. 4 Expression of glycolysis-associated molecules in HemECs and HUVECs by RT-qPCR and WB methods. A. RNA expression of GLUT1, HK2, PFKFB3, PKM2, HIF-1a was higher in HemECs than in HUVECs. B. Relative protein expression was assayed through the density of bands as fold changes after normalization to β-actin. C. The bands of glycolysis-associated molecule expression in HemECs and HUVECs under different culture conditions (ns = no significant difference, *p < 0.05, **p < 0.01, ***p < 0.001). Fig. 5 Inhibition of glycolysis-associated molecules affected proliferation, migration, and angiogenesis in HemECs. A. Proliferation of HemECs was attenuated by inhibition of HK2 and PFKFB3; inhibition of GLUT1, PKM2 and LDHA had no obvious effect. B. Migration of HemECs was attenuated by inhibition of GLUT1, HK2, PFKFB3, PKM2 and LDHA. C. Microscopic images and graphical representation of Scratch changes. D. Microscopic images and graphical representation of tube formation. E. Total junctions, total branching length, total number of meshes, and mean mesh size were quantified by ImageJ software. (ns = no significant difference, *p < 0.05, **p < 0.01, ***p < 0.001).