Volume 236, Issue 11 pp. 7853-7873
REVIEW ARTICLE
Full Access

Targeting HDL in tumor microenvironment: New hope for cancer therapy

Tan-Jun Zhao

Tan-Jun Zhao

Division of Stem Cell Regulation and Application, School of Pharmacy, Hunan University of Chinese Medicine, Changsha, China

Search for more papers by this author
Neng Zhu

Neng Zhu

Department of Urology, The First Hospital of Hunan University of Chinese Medicine, Changsha, China

Search for more papers by this author
Ya-Ning Shi

Ya-Ning Shi

Division of Stem Cell Regulation and Application, School of Pharmacy, Hunan University of Chinese Medicine, Changsha, China

Search for more papers by this author
Yu-Xiang Wang

Yu-Xiang Wang

Division of Stem Cell Regulation and Application, School of Pharmacy, Hunan University of Chinese Medicine, Changsha, China

Search for more papers by this author
Chan-Juan Zhang

Chan-Juan Zhang

Division of Stem Cell Regulation and Application, School of Pharmacy, Hunan University of Chinese Medicine, Changsha, China

Search for more papers by this author
Chang-Feng Deng

Chang-Feng Deng

Division of Stem Cell Regulation and Application, School of Pharmacy, Hunan University of Chinese Medicine, Changsha, China

Search for more papers by this author
Duan-Fang Liao

Duan-Fang Liao

Division of Stem Cell Regulation and Application, School of Pharmacy, Hunan University of Chinese Medicine, Changsha, China

Search for more papers by this author
Li Qin

Corresponding Author

Li Qin

Division of Stem Cell Regulation and Application, School of Pharmacy, Hunan University of Chinese Medicine, Changsha, China

Correspondence Li Qin, School of Pharmacy,​ Hunan University of Chinese Medicine, 300 Xueshi Rd, Hanpu Science and Education District, Changsha, 410208 Hunan, China.

Email: [email protected]

Search for more papers by this author
First published: 21 May 2021
Citations: 9

Tan-Jun Zhao contributed equally to this work.

Abstract

Epidemiological studies have shown that plasma HDL-C levels are closely related to the risk of prostate cancer, breast cancer, and other malignancies. As one of the key carriers of cholesterol regulation, high-density lipoprotein (HDL) plays an important role in tumorigenesis and cancer development through anti-inflammation, antioxidation, immune-modulation, and mediating cholesterol transportation in cancer cells and noncancer cells. In addition, the occurrence and progression of cancer are closely related to the alteration of the tumor microenvironment (TME). Cancer cells synthesize and secrete a variety of cytokines and other factors to promote the reprogramming of surrounding cells and shape the microenvironment suitable for cancer survival. By analyzing the effect of HDL on the infiltrating immune cells in the TME, as well as the relationship between HDL and tumor-associated angiogenesis, it is suggested that a moderate increase in the level of HDL in vivo with consequent improvement of the function of HDL in the TME and induction of intracellular cholesterol efflux may be a promising strategy for cancer therapy.

1 INTRODUCTION

The tumor microenvironment (TME) is commonly a complex, acidic, and hypoxic environment, and is mainly composed of tumor cells, nontumor cells, and extracellular matrix (ECM). Through the in-depth study on tumors, it is found that tumor cells can promote their development and progression by recruiting and inducing resident cells to build an adaptive growth environment (Mattiuzzi & Lippi, 2019). TME accelerates tumor growth, invasion, and metastasis by suppressing immunity, promoting angiogenesis, and inhibiting inflammatory response (Hinshaw & Shevde, 2019). Currently, most of the therapeutics target the tumor itself with unsatisfying outcomes in many circumstances, including poor prognosis, adverse reactions, and drug resistance. Therefore, in-depth exploration of the roles of different compositions in the TME for targeting nontumor cells in the TME has become an attractive approach for developing novel therapeutic strategies for cancer treatment.

As a derivative of cyclopentane phenanthrene, cholesterol is closely related to the formation of cholic acid and the synthesis of hormones crucial for the sustainability of life. During tumor development, cholesterol is involved in the formation of the cell membrane, regulation of cell functions, and modulating the activation of transmembrane receptors (Pike, 2003). Accumulation of cholesterol inside tumor cells can be achieved by abnormal in situ synthesis of cholesterol, increase in uptake, and decrease in efflux (Tosi & Tugnoli, 2005), which, in turn, may promote tumor cell proliferation, invasion, and migration (Zhuang et al., 2005). In addition, cholesterol homeostasis in immune cells can also affect tumor growth in vivo (Sag et al., 2015). Therefore, when evaluating the effect of cholesterol on the cellular components of TME, not only tumor cells but also nontumor cells should be considered. As one of the key lipoproteins in cholesterol transportation, HDL and its related components play an important role in maintaining and regulating the cholesterol homeostasis between tumor cells and nontumor cells. Meanwhile, HDL can promote the clearance of cholesterol and its metabolites from tumor cells, thereby inhibiting the occurrence and progression of tumors (Ganjali et al., 2019). Furthermore, HDL can affect the function and activity of infiltrating immune cells and angiogenic vascular cells in the TME constructed by tumor cells. Thus, understanding the role of HDL in the TME may be of great significance to the prevention and treatment of cancer development and progression.

2 CORRELATION BETWEEN HDL AND TUMORS

High-density lipoprotein (HDL) protein composition includes lipid transport-related proteins, lipolytic enzymes, acute-phase reaction proteins, and multifunctional receptors that mediate cholesterol transport. Among them, apolipoprotein A-I (apoA-I) and the scavenger receptor type 1 (SR-B1), a multifunctional receptor that mediates cholesterol transport, play an important role in the regulation of cholesterol in tumor and nontumor cells. The main function of HDL/apoA-I is to play a leading role in reverse cholesterol transport (RCT). Cholesterol is a key component of the cellular membrane for both tumor and nontumor cells. During tumor development, a large amount of cholesterol accumulates in tumor cells. HDL/apoA-I interacts with ATP-binding cassette (ABC) transporters ABCA1/ABCG1 and SR-B1 to mediate the efflux of cholesterol and phospholipids (PLs) in tumor cells, thereby inhibits the development and progression of tumors (Ganjali et al., 2019). Furthermore, HDL also possesses anti-oxidation, anti-inflammation, antiapoptosis, immune-regulation, endothelial protection, and other activities that may be related to cancer biology. And oxidative, inflammation, and dyslipidemia in cancers can alter the anticancer properties of HDL by adding or removing neutral lipids, remodeling, and mutual transformation of PLs and apolipoprotein components.

Recently, epidemiological studies have reported a decreased level of plasma HDL-C in cancer patients (Chawda et al., 2011; Poorey & Thakur, 2016). However, most observational studies, case-control studies, prospective studies, and meta-analysis of randomized controlled trials have been affected by different risk factors, including tobacco exposure, obesity, metabolic syndrome, hyperinsulinemia, and environmental factors (Dilman et al., 1981), which leads to an unclear link between HDL-C levels and carcinogenesis/cancer development. Interestingly, a meta-analysis of randomized controlled trials between tumor risk and plasma HDL-C levels showed that for every 10 mg/dl increase in plasma HDL-C levels, the risk of cancer incidence was significantly reduced by 36% (Jafri et al., 2010). Possible roles of HDL in the development of different types of cancer are illustrated in detail below (Table 1). In short, the relationship between plasma HDL-C levels and cancer remains to be addressed. Variable results from different clinical trials may be influenced by factors, such as age, gender, race, number of cases, follow-up time, as well as cancer type and stage of the participating cases. Meanwhile, the difference between HDL-C and HDL leads to the possibility that the relationship between HDL levels or function and tumors may not be related to HDL-C alone (Hutchins et al., 2014). In addition, HDL may exhibit different functions in different tumors. Moreover, the impacts of HDL on tumor biology may be caused not only by affecting tumor tissue, but it also its close relationship with other components in the TME, leading to the diversity in tumor progression.

Table 1. Clinical trials related to the relationship between HDL-C/HDL components
The correlation between HDL and tumors Cancer Study design Number of patients OR/RR/HR Main outcomes Ref
Negative correlation Colorectal cancer Prospective 6,296,903 HR: 1.80 Metabolic syndrome positively correlated with the occurrence of colorectal cancer: a coexistence of abdominal obesity, impaired glucose tolerance, and reduced HDL-C levels Choi et al. (2018)
Prospective 10,876 RR: 0.84 Reduced TG and TC levels are negatively related to the risk of colorectal cancer Yao and Tian (2015)
Meta-analysis 17,387 NA Negative association between HDL-C levels and the risk of colon adenoma Passarelli and Newcomb (2016)
Prospective 1238 RR: 0.78 Negative association between apoA levels and colon cancer risk van Duijnhoven et al. (2011)
Retrospective 667 HR: 1.73 Worse prognosis and overall survival in patients with low preoperative HDL-C levels Y. Wang, Sun, et al. (2019)
Case-control 86 OR: 0.263 Plasma resistin is negatively correlated with HDL-C concentration Mihajlovic et al. (2019)
Lung cancer Prospective 14,547 HR: 1.45 Weak negative association between HDL-C levels and lung cancer risk Kucharska-Newton, Rosamond, Schroeder et al. (2008)
Meta-analysis 3067 RR: 0.76 Negative association between TC levels and the risk of lung cancer Lin et al. (2017)
Case-control 459 HR: 0.54 Direct bilirubin and total cholesterol are important predictors of overall survival in advanced non-small-cell lung cancer patients with EGFR mutations Zhang et al. (2017)
Case-control 424 OR: 0.288 Positive association between TG levels and the risk of lung cancer Hao et al. (2018)
Case-control 140 NA Platelet, PCT, NLR, PLR, albumin, HDL, and CRP levels may serve as the diagnosis of lung cancer and detection of advanced disease Şahin and Aslan (2018)
Breast cancer Meta-analysis 1,189,635 HR: 0.73 TG levels are negatively correlated with breast cancer risk and high HDL-C levels can prevent breast cancer in postmenopausal women Ni et al. (2015)
Meta-analysis 240,137 HR: 0.86 TC and HDL-C levels are significantly negatively correlated with breast cancer risk Touvier et al. (2015)
Case-Control 30,546 RR: 0.73 Low HDL-C is an independent predictor of increased risk of breast cancer in overweight and obese postmenopausal women Furberg et al. (2004)
Retrospective 1044 HR: 0.528 Preoperative serum TG and HDL-C levels may be independent factors in predicting the prognosis of breast cancer patients Li et al. (2017)
Case-Control 690 OR: 0.73 HDL-C levels reduce the risk of breast cancer in premenopausal women Kim et al. (2009)
Case-Control 359 HR: 0.18 Low HDL-C in premenopausal women may be a marker of increased breast cancer risk Kucharska-Newton, Rosamond, Mink et al. (2008)
Case-Control 181 RR: 1.60 Serum HDL-C and serum TG are most strongly associated with breast cancer risk Agnoli et al. (2010)
Endometrial cancer Prospective 225,432 NA TG levels are positively correlated with the risk of endometrial cancer Seth et al. (2012)
Prospective 8303 RR: 0.61 Metabolic abnormalities and obesity may synergize to increase the risk of endometrial cancer Cust et al. (2007)
Case- control 196 NA All patients with endometrial cancer have lower HDL-C compared with controls Qadir and Malik (2008)
Biliary tract cancer Case-control 1086 NA TC, LDL, apoA, and apoB levels are associated with a high risk of biliary tract cancer He et al. (2014)
Case-control 460 NA Compared with the control group, patients with low HDL-C levels (<30 mg/dl) have a 4.2-fold increased risk of gallstones, 11.6-fold increased risk of gallbladder cancer, and 16.8-fold increased risk of cholangiocarcinoma Andreotti et al. (2008)
Prostate cancer Retrospective 302 HR: 1.86 TC and LDL levels are not associated with anticastration prostate cancer Dambal et al. (2020)
Retrospective 100 OR: 1.13 The LDL/HDL ratio predicts risk in prostate cancer patients receiving androgen deprivation therapy Oka et al. (2018)
Renal cancer Retrospective 308 HR: 0.32 Elevated serum HDL-C levels may predict better overall survival and tumor-specific survival in patients with ccRCC Hao et al., 2019)
Case-Control 248 HR: 0.175 Elevated serum cholesterol, HDL-C and LDL-C levels may reduce cancer risk Zhang et al. (2013)
Positive correlation Colorectal cancer Retrospective 703 NA Elevated HDL levels have a certain predictive value for ocular metastasis in colorectal cancer patients Zhu et al. (2019)
Breast cancer Mendelian randomization 101,424 OR: 1.12 Circulating HDL-C associated with an increased risk of breast cancer, LDL-C not associated with breast cancer risk Beeghly-Fadiel et al. (2020)
Case-Control 279 NA Higher serum HDL-C and apoA-1 levels and lower non-HDL-C and apoB levels are associated with a 20%–30% increased risk of developing estrogen receptor-positive breast cancer Martin et al. (2015)
No correlation Lung cancer Prospective 109,798 HR: 1.01 Levels of TC, TG, and LDL-C are linearly correlated with an increased risk of lung cancer Lyu et al. (2019)
Colorectal cancer Mendelian randomization 27,910 OR: 0.94 A causal relationship between higher TC levels and colorectal cancer risk Rodriguez-Broadbent et al. (2017)
Prostate cancer Meta-analysis 23,142 RR: 0.93 Neither TC nor LDL levels are associated with overall or high-grade prostate cancer risk YuPeng et al. (2015)
Mendelian randomization 22,249 OR: 0.99 Increased levels of TG and LDL are proportional to the risk of prostate cancer invasion Bull et al. (2016)
  • Abbreviations: CRP, C-reactive protein; HDL, high-density lipoprotein; LDL, low-density lipoprotein; NA, not available; NLR, neutrophil/lymphocyte ratio; PCT, plateletcrit; PLR, platelet/lymphocyte ratio; TC, total cholesterol; TG, total glyceride.

3 COMPOSITION AND CHARACTERISTICS OF TME

The TME is composed of cellular and noncellular components, including angiogenic vascular cells (i.e., pericytes and endothelial cells), infiltrating immune cells (i.e., T and B lymphocytes, neutrophils, myeloid-derived suppressor cells (MDSCs), natural killer (NK) cells, mast cells, antigen-presenting macrophages and dendritic cells (DCs), tumor-associated fibroblasts (CAFs), adipocytes, as well as ECM. TME serves as a “fertile soil” to support tumor cells for their proliferation, invasion, and migration (Pearce et al., 2018). The characteristics of TME, including hypoxia, acidity, increased lactate, and decreased glucose concentrations, as well as stromal and immune cell recruitment, may be involved in tumor progression (Lyssiotis & Kimmelman, 2017). Nontumor cells in the TME play a variety of roles in the biological functions of tumors development by secreting cytokines and other factors (i.e., collagen, elastin, fibronectin, laminin, and proteoglycan) for maintaining proliferation signals, avoiding growth inhibition, evading epidemic killing, resisting cell death, inducing angiogenesis, and activating invasion and metastasis (Figure 1) (Hanahan & Weinberg, 2011).

Details are in the caption following the image
Effect of TME-related cells on tumorigenesis and cancer development. Eight of the ten hallmark capabilities of tumors are closely related to stromal cells in the TME, while the other two are related to the replication immortality and genome instability/mutation of tumor cells. Three types of stromal cells, including infiltrating immune cells, tumor-associated fibroblasts, and angiogenic vascular cells can promote tumor growth, invasion, and metastasis by secreting different cytokines. The role of stromal cells in tumor development depends on tumor types and organs where the tumor is located. CCL2, chemokine (C-C motif) ligand 2; CCP, cysteine cathepsin proteases; CXCL12, C-X-C chemokine ligand 12; EGF, epidermal growth factor; FGF, fibroblast growth factor; GF, growth factorHGF, hepatocyte growth factor; HIF, hypoxia-induced factor; IGF-1, insulin-like growth factor-1; IL, interleukin; iNOS, inducible nitric oxide synthase; LFAI, Ligands for antiapoptosis integrins; MEGF, mitogenic epithelial growth factors; MMP, matrix metalloproteinase; PDGF, platelet-derived growth factor; PDGF, platelet-derived growth factor; PIGF, placental growth factor; ROS, reactive oxygen species; SDF-1, stromal cell-derived factor-1; SF, survival factor; SP, serine proteases; TGF-β, transforming growth factor-β; TNF-α, tumor necrosis factor-α; VCAM-1, vascular cell adhesion molecule-1; VEGF, vascular endothelial growth factor

4 THE EFFECT OF HDL ON TUMOR IMMUNE MICROENVIRONMENT

Innate and adaptive immune cells are two double-edged swords to cancer. Immune cells can trigger proinflammatory responses to inhibit tumor development, and they can also be reprogrammed through the inhibitory signals in the TME to promote the progression of tumors (Hinshaw & Shevde, 2019), including assisting in immune evasion, invasion, migration, and angiogenesis of tumor cells (Ben-Neriah & Karin, 2011). The metabolic status of immune cells in the TME and how HDL affects its function are summarized in Table 2 and Figure 2.

Details are in the caption following the image
HDL and Tumor immune microenvironment. HDL has major effects on distinct immune cell types (boxes) in the TME and can stimulate (black arrows) or inhibit (red lines) the activities of diverse immune cell types
Table 2. HDL and immune cells in the TME
Cell type Normal functions Stimulatory cytokines in the TME Cytokine chemokine secretion Effect of tumor Correlation with HDL Ref
Macrophages
M1 Activation of Th1 response, Phagocytosis IFNγ, LPS IL-6/12/23/1β/, CCL2/3/4/5/8/9/10/11 Antitumor Protransformation Liu and Yang (2013); Sag et al. (2015); Smythies et al. (2010); Zamanian-Daryoush et al. (2013)
M2 Activation of Th2 response, Immune regulation, Angiogenesis, Matrix remodeling IL3/4/6/10/13, CCL2/3/14, TNF-α, CSF1 IL-1/4/6/10/12, CCR2, TGF-β, PGE2, VEGF-A, TNF-α, CXCL10/16, CCL1/5/17/22/24, l-arginine Protumor Antitransformation
Neutrophils
N1 Phagocytosis, Release of inflammatory cytokines and ROS, Promotion of tumor cell apoptosis NA TNF-α, IL1, IFNs, MMP-8, Defensins, ROS Antitumor Inhibit the activation of neutrophils Cettour-Rose et al. (2005); Lynn et al. (1991); Murphy et al. (2011); Yvan-Charvet et al. (2010)
N2 Supporting angiogenesis, Tumor cell migration and invasion, Immune surveillance and metastasis, and secretion of chemokines and cytokines TGF-β, Angiotensin II Oncostatin-M, MMP-9, CXCL1/6/8, CCL-3, Neutrophil, elastase (NE), TGF-β, PGE2 Protumor Inhibit the activation of neutrophils
T cell
CD8+ T cell Promote cell lysis IL-2 IFNG, GZMB, CCL3, perforin, IFNγ Antitumor Neutral Kadiyala et al. (2019)
Treg Blocking CD8+ T cell activation NA CTLA4, IL-2 receptors Protumor Antiactivation Liu et al. (2010); Newton and Benedict (2014)
Dendritic cell
Immature DC Recognition of antigens, Phagocytosis, Weakening APC, Inducing T cell anergy and promoting Th2 and Treg responses NA NA Tumor type Inhibit the activation, maturation, and differentiation of dendritic cells Kim et al. (2005); Perrin-Cocon et al. (2012); Wang et al. (2012)
cDC1 APC to CD8 T cells, cross presentation, secretion of IL12 IRF8, BATF3 IL12, TNF-α, IFNγ Tumor type
cDC2 APC to CD4 T cells IRF4, ZEB, Notch2/KLF4 TGF-β, TNF-α, IL1/6/8/10/12/23 Tumor type
pDC Secretion of IFN1 NA IFN-1, TNF, IL6 Tumor type
MoDC Proinflammatory cytokine production, TNF, IL-6/12 NA TNF-α, IL1/12/23 Antitumor
Tolerogenic DC APC reduction, stimulation of Th2 response, and induction of tolerance by Tregs PGE2, TGF-β, VEGF, IL10, TNF-α TGF-β Protumor
Natural killer cell
CD56hiCD16± Producing inflammatory cytokines TGF-β, PGE2, IDO, IL10 IFNγ, TNF-α Tumor type Antiactivation De Sanctis et al. (1995); Qin et al. (2020)
CD56loCD16hi Promote antibody-dependent cytotoxicity, high perforin production, and enhanced killing TGF-β, PGE2, IDO, IL10 IL10/22 Tumor type Antiactivation
Myeloid-derived suppressor cell
M-MDSC Inhibition of innate and adaptive immune responses CSF-1, CCL2/5/7, HIF-1α, CXCL1/12 NO, CCL3/4/5, Arg1, PGE2, IL4 Protumor Antiactivation Kujawski et al. (2008); Nagaraj and Gabrilovich (2012); Plebanek et al. (2018); Shojaei et al. (2007); Zamanian-Daryoush et al. (2013)
PMN-MDSC Inhibition of innate and adaptive immune responses CXCL5, CCL15 ROS, Arg1, PGE2, IL4 Protumor Antiactivation
eMDSC Inhibition of innate and adaptive immune responses NA NA Protumor Antiactivation
  • Abbreviations: 7-KC, 7-ketocholesterol; APC, antigen-presenting cell; apoA-I, apolipoprotein A-I; Arg1, arginase 1; BATF3, basic leucine zipper transcriptional factor ATF-like 3; bFGF, basic fibroblast growth factor; CAF, tumor-associated fibroblasts; CCR2, C-C chemokine receptor type 2; CSF1, colony-stimulating factor 1; CTLA-4, cytotoxic T lymphocyte protein 4; CTLA4, cytotoxic lymphocyte antigen 4; DC, dendritic cell; ECM, extracellular matrix; G-CSF, granulocyte colony-stimulating factor; GZMB, granzyme B; IDO, indoximod; IFNG, interferon gamma gene; IFNγ, interferon-γ; IRF4, interferon regulatory factor 4; KLF4, Krüppel-like factor4; LPS, lipopolysaccharide; MDSC, myeloid-derived suppressor cell; MIP-2, macrophage inflammatory protein-2; NA, Not available; PD-1, programmed death 1; PD-L1, programmed death ligand 1; PGE2, prostaglandin E2; PL, phospholipid; TAM, tumor-associated macrophage; TAN, tumor-associated neutrophil; TCR, T cell receptor; ZEB, zinc finger E-box binding homeobox.

4.1 Macrophages

Tumor-associated macrophages (TAMs) are mainly originated from monocyte differentiation and bone marrow–derived macrophages (Chen et al., 2017). Monocytes can be triggered by Th1 cytokine interferon γ (IFN-γ) and bacterial lipopolysaccharide (LPS) to M1-phenotypic TAM (M1-TAM) for inhibiting angiogenesis, tumor survival, and metastasis, or by Th2 cytokines (IL-4/6/10/13) to M2-phenotypic TAM (M2-TAM) for promoting cancer progression (Liu & Yang, 2013). Clinical data have shown a positive association between high TAM levels in the TME and poor prognosis of patients with breast cancer or prostate cancer (Lewis & Pollard, 2006). TAMs promote tumor metastasis via a variety of mechanisms, including promoting angiogenesis, accelerating tumor growth, and enhancing tumor cell invasion and migration (Yang et al., 2018). In addition, TAM also promotes tumor development through antitumor immune mechanisms, including secretion of programmed death ligand 1 (PD-L1) and PD-L2, to inhibit the antitumor cytotoxic CD8+ T cell response (Kuang et al., 2009), and secretion of cytokines and proteases, for example, arginase 1 (Arg1), IL-10, TGF-β, and prostaglandin, to prevent T cell activation (Shimizu et al., 2018). Recent studies have demonstrated that HDL plays a key regulatory role in the phenotypic transformation of macrophages. In a transgenic melanoma mouse model expressing apoA-I, HDL/apoA-I displayed a strong immunomodulatory effect in the TME and converted M2-TAM into M1-TAM (Zamanian-Daryoush et al., 2013). Furthermore, HDL/apoA-I also induces macrophages to differentiate into M2 phenotypic (Smythies et al., 2010), affects antigen presentation and T cell receptor (TCR) signaling (Gruaz et al., 2010), thereby promoting tumor development. At the cellular level, HDL/apoA-I was observed to promote cholesterol efflux from macrophages, leading to cholesterol depletion, lipid raft rupture, and upregulation of IL-10 expression, thereby inhibiting monocyte chemotaxis, reducing TAM content, and weakening the promotion of TAM on tumor progression (Smythies et al., 2010). In vitro, the incubation of monocytes with HDL/apoA-I suppresses the expression of CD11b as well as cellular adhesion function, which might indicate that HDL/apoA-I may prevent the activation and recruitment of monocytes, thereby inhibiting their differentiation into TAM (Murphy et al., 2008). Another study shows that the accumulation of cholesterol in ABCG1-/- TAMs can suppress the growth of melanoma and bladder cancer cells, associated with the shift of macrophages from M2 subtype to M1 subtype (Sag et al., 2015). A possible explanation is that ABCG1 deficiency causes the inefficiency of oxysterol (i.e., 7-ketocholesterol (7-KC)) clearance, and consequently, the accumulation of oxysterol inside TAM induces apoptosis (Tarling et al., 2010). Moreover, ABCG1-/- TAM results in accumulation of cholesterol, which, in turn, activates IκB kinase/NF-κB signaling and enhances polarization of TAMs to M1-TAM phenotype, as well as produces TNF-α and NO to mediate the lysis of tumor cells (Escórcio-Correia & Hagemann, 2011). Transcriptional regulator transcription factor 3 (ATF3), as an HDL-inducible target gene in macrophages, was found to downregulate the expression of proinflammatory cytokines induced by Toll-like receptor (TLR) and inhibit the inflammatory response and tumor progression (De Nardo et al., 2014). Therefore, inhibition of tumor progression by increasing the content of HDL or improving the function of HDL may be an effective antitumor strategy to promote the consumption of TAM in the TME or the conversion of M2-TAM to M1-TAM.

4.2 Neutrophils

Similar to TAM, tumor-associated neutrophils (TANs) have plasticity related to the type and progression of tumors (Aras & Zaidi, 2017). Generally, TANs can be divided into antitumor N1 inflammatory phenotypes and tumor-promoting N2 immunosuppressive phenotypes (Fridlender & Albelda, 2012). N1-TAN can produce ROS, exert antitumor role in the early stage by activating the apoptosis signal (Liou & Storz, 2010). However, during tumor development, a low level of ROS secreted by N1-TAN is incapable of killing tumor cells, instead, it can indirectly promote tumor growth through genome instability (Uribe-Querol & Rosales, 2015). In addition, inflammatory TME and TGF-β promote tumor growth by increasing the polarization of TANs to the N2 phenotype (Galdiero et al., 2013). One of the main mechanisms of N2-TAN to promote tumor growth is to inhibit adaptive immune cells, especially cytotoxic T lymphocytes (Y. Zhang et al., 2020). In the TME, neutrophil elastase (NE) and metalloproteinases (MMP-8/9) secreted by TANs reconfigure the ECM, while ROS/RNS, NE, and MMP-9 promote cancer cell invasion, VEGF promotes angiogenesis, and PGE2 promotes tumor progression (Demers et al., 2016). MMPs, NE, and cathepsin G can degrade proinflammatory cytokines and reconfigure TME to promote tumor progression and metastasis (Schauer et al., 2014). Increased TAN is associated with poor prognosis of the patients with melanoma (Schmidt et al., 2005) and renal cancer (Atzpodien & Reitz, 2008). Moreover, HDL/apoA-I can reduce the abundance of lipid rafts in the neutrophil membranes through ABCA1 and SR-B1, suppress the expression CD11b levels, and inhibit the activation, adhesion, and migration of neutrophils (Murphy et al., 2011), indicating that HDL/apoA-I is involved in inhibiting TAN activation in TME. Furthermore, HDL/apoA-I can attenuate the expression of neutrophil activators, for example, PMA and fMLP, and the priming effect of LPS on neutrophils (Cettour-Rose et al., 2005), thereby suppressing the activation of neutrophils. Meanwhile, neutrophils can be produced in mice with ABCA1/ABCG1 knockout (Yvan-Charvet et al., 2010). Cytokines are synthesized by neutrophils, such as granulocyte colony-stimulating factor (G-CSF) and macrophage inflammatory protein-2 (MIP-2). G-CSF are the main regulator to stimulate the development and differentiation of neutrophils (Eyles et al., 2006), which can weaken the clearance efficiency of neutrophils in inflammatory tissues, aggravate the inflammatory response (Roberts, 2005), and promote the adhesion of neutrophils, and so forth (Kaushansky, 2006). In the mouse model of apoA-I-/-, the levels of neutrophils and G-CSF are significantly increased, with the G-CSF antibody being able to suppress the increment of neutrophils in this model (Dai et al., 2012). Meanwhile, injection of apoA-I mimetic peptide can also decrease neutrophils and G-CSF levels (Dai et al., 2012). Therefore, apoA-I inhibits the neutrophil-mediated inflammatory response by downregulating the expression of G-CSF. Another cytokine, MIP-2, binds to its specific receptors and regulates the activation of neutrophils through the p38 mitogen-activated protein kinase pathway (Qin et al., 2017). It has been shown that HDL can reduce neutrophils infiltration in the animal model of kidney hemorrhage/resuscitation by inhibiting MIP-2 expression (Cockerill et al., 2001). Therefore, inhibiting the recruitment and activation of neutrophils and inducing the polarization of TANs toward the N1 phenotype is a potential therapeutic strategy. More studies on the effect of HDL on TAN differentiation are warranted.

4.3 T cells

T cells belong to one of the important components of adaptive immunity, and their infiltration and function are essential to suppress tumor occurrence and progression. T cells can differentiate into two different types, known as CD8+ T cells and CD4+ T cells. CD4+ T cells can further be divided into antitumor and proinflammatory T helper cell 1 (Th1), immunosuppressive T-helper cell 2 (Th2), unclear T helper cell 17 (Th17), or immunoregulatory T cells (Tregs). HDL can inhibit T cell proliferation and reduce its survival rate (Newton & Benedict, 2014).

CD8+ T cells can induce tumor cell lysis by recognizing tumor-specific antigens, secreting perforin, granzyme B (GZMB), and other effector molecules. Injection of reconstituted high-density lipoprotein (rHDL) particles can promote the activation of CD8+ T cells in hepatocellular carcinoma, colloid blastoma, and other tumor models (Kadiyala et al., 2019). T cells induce signal transduction due to prolonged exposure to homologous antigens produced by tumor cells in the TME, thereby increase the expression of inhibitory molecules, including cytotoxic T lymphocyte protein 4 (CTLA-4) and programmed death 1 (PD-1) (Meng et al., 2018). Accumulating evidence indicates that the interaction between PD-1 and PD-L1/2 can inhibit effector function to promote T cell apoptosis (Ahmadzadeh et al., 2009). Clinical samples of liver cancer and cervical cancer also show that the elevated expression of PD-1 and PD-L1 on CD8+ T cells correlates to T cell apoptosis and cancer development (Kim et al., 2018; Meng et al., 2018). In addition, chemotherapy can also promote the expression of PD-1 on CD8+ T cells (Maude et al., 2018), suggesting that PD-1-mediated inhibition of CD8+ T cells may be the reason for the failure of chemotherapy. In the acidic environment of TME, the accumulation of lactate can lead to tumor acidosis, thereby impairing the expression of TCR coreceptor and making T cells ineffective (Calcinotto et al., 2012). In clinical samples, high serum lactate dehydrogenase or lactate levels usually predict a poorer prognosis for cancer patients (Thompson et al., 2010). Hypoxia does not inhibit CD8+ T cell activity, but promotes T cell survival and enhances T cell-mediated tumor regulation (Gropper et al., 2017). Hence, promoting CD8+ T cells activation and inhibiting PD-1 expression are potential antitumor therapeutic manner. In addition, IL-2, a cytokine essential for the proliferation and differentiation of CD8+ T cells, may invalidate the inhibition mediated by PD-1. Conversely, PD-1 may indirectly inhibit T cell function and survival through IL-2 (Francisco et al., 2010), while HDL can reduce IL-2 levels to inhibit PD-1 and CD8+ T cell activation (Benghalem et al., 2017), indicating that reduction of IL-2 by HDL might be a double-edged sword.

Tregs, as immunosuppressive cells, inhibit the activation of T cells through multiple pathways. Tregs play a role in tumor cell immune evasion by inhibiting cytotoxic activity and support other protumor cell types, for example, M2-TAM (Miska et al., 2019). Tregs can block CD8+ T cell activation by expressing IL-2 receptors (CD25 and CD132), as well as binding and depleting IL-2 (McNally et al., 2011). Tregs can also inhibit DC expansion and immunogenicity to hinder CD8+ T cell activation (Jang et al., 2017). In addition, cytotoxic lymphocyte antigen 4 (CTLA4) and transforming growth factor β (TGF-β) on the surface of Tregs can inhibit CD8+ T cells function and activity (Shabaneh et al., 2018). It is reported that high Treg levels in the TME are positively associated with poorer prognosis of the patients with ovarian cancer (Preston et al., 2013), breast cancer (Shou et al., 2016), and bladder cancer (Togashi et al., 2019). HDL inhibits the differentiation of T cells into CD4+ T cells and reduces CD4+ T cell viability and proliferation ability (Newton & Benedict, 2014). Another study has found that HDL/apoA-I reduces cholesterol levels in T cell membrane lipid rafts, thereby blocking MHC- II expression and inactivating CD4+ T cells (Parra et al., 2015). Lipid raft membranes are rich in sphingomyelin phosphate and able to be phosphorylated by sphingosine kinase (SPHK) to form HDL-related bioactive lipid sphingosine-1-phosphate (S1P), which, as a switch factor, can inhibit the differentiation of Tregs by activating its receptor (Liu et al., 2010). Based on these findings, it is speculated that HDL might reduce the content of CD4+ T cells in the TME to decrease the density of Tregs. Contrarily, another study has found that HDL promotes Treg survival in a concentration-dependent manner, associated with increased SR-B1 levels (Rueda et al., 2017). This discrepancy may be due to the different functions of HDL.

4.4 Dendritic cells

DCs, as important antigen-presenting cells (APCs), are central regulators linking innate and adaptive immunities and are differentiated from monocytes. According to their development and phenotype, mature DCs are divided into conventional type 1 DC (cDC1), conventional type 2 DC (cDC2), plasmacytoid DC (pDC), and monocyte-derived DC (MoDC). The functions of different subpopulations stem from the differences in their receptor presentations. pDCs can initiate CD8+ T cell differentiation (Tel et al., 2013). cCD1s can process and cross-present exogenous antigens on MHC I to activate CD8+ T cells and Th1 cells (Mildner & Jung, 2014). cCD2s can induce CD4+ T cell polarization of different subsets and activate CD8+ T cells (Binnewies et al., 2019). MoDCs can promote the phenotypic differentiation of CD4+ T cells into Th1 cells, Th2 cells, or Th17 cells (Segura et al., 2013). In the TME, tumor-infiltrating dendritic cells (TIDCs) can manifest as immunogenic or tolerant. Immunosuppressive cytokine, VEGF, IL-10, TGF-β, and PGE2 can inhibit DC maturation into immunogenic cells and promote its development into tolerant phenotypes, thereby inhibiting its Th1 priming and endowing its promoting Th2 and Treg (Motta & Rumjanek, 2016). Moreover, the appearance of tumor-associated antigen (TAA) in the absence of costimulatory signals leads to T cell anergy (Steinman, 2012). Of course, the plasticity of DCs varies in different tumors. It has been reported that DCs are tumor suppressive in the early stage and become tumor-promoting as the tumor progresses (Chaput et al., 2008). Another study finds that DC phenotypes differ between subtypes of the same tumor type (Michea et al., 2018).

HDL/apoA-I induces the production of prostaglandin E2 (PGE2) and IL-10 while decreases IL-12 and IFN-γ secretion in monocytes, and results in the suppression of DCs activation, maturation, and differentiation (Kim et al., 2005). HDL reduces IL-12 secretion in stimulated mature DCs and attenuates the ability of T cells to differentiate into Th1 subtypes (Kim et al., 2005). Under the stimulation of TLR4, HDL inhibits the ability of DCs to induce T cell Th1 response, characterized by high IFN levels secretion, and HDL can also inhibit the Th1 function of DCs stimulated by TLR1/2 and TLR2/6 ligands (Perrin-Cocon et al., 2012). After incubation of DCs with HDL/apoA-I, DCs show a reduced potential to activate T cells due to cholesterol efflux and disruption of membrane lipid rafts in DCs (Wang et al., 2012). HDL/apoA-I inhibits T cell activation by destroying DC membrane lipid rafts, reducing cholesterol content in lipid rafts as well as the density of MHCII with antigen presentation and signal transduction functions, while cholesterol supplementation may restore T cells activation (Wang et al., 2012). HDL/apoA-I not only inhibits DCs to activate T cells but also mediates IL-2 level reduction to inhibit CD8+ T cell activation by reducing lipid raft content and MHC-II density (Wang et al., 2012). Therefore, the composition and function of DCs are greatly influenced by tumor type or subtype and its unique TME, leading to the complicated regulation of DCs function in certain tumors.

4.5 Natural killer cells

NK cells are cytotoxic innate immune cells. Proinflammatory cytokines promote NK cells to be recruited into TME and activate them for recruiting other immune cells (Paul & Lal, 2017). NK cells have two phenotypes, CD56hiCD16± and CD56loCD16hi (Stabile et al., 2017), collectively known as tumor-infiltrating natural killer cells (TINKs) in the TME. CD56hiCD16± TINK secretes inflammatory cytokines, while CD56loCD16hi TINK is cytotoxic and possesses a cell-mediated killing function (Glasner et al., 2018). NK cells can inhibit tumor cell growth and metastasis by utilizing death receptor-mediated apoptosis and perforin/granzyme-mediated cytotoxicity (Langers et al., 2012). Tumor cells secrete immunosuppressive soluble factors (miR29b, CXCL10, etc.) that inhibit the production and cytotoxicity of TINK inflammatory cytokines to protect themselves from NK-mediated killing in the TME (Al Absi et al., 2018; Scoville et al., 2018). Furthermore, lactic acid, a by-product of glycolysis by cancer cells in the TME, also weakens NK cell activity and reduces its killing effect on tumor cells (Brand et al., 2016).

In an animal model of hepatocellular carcinoma, NK cells were found to enhance antitumor properties by accumulating cholesterol, promoting the formation of lipid rafts, and activating immune signaling pathways (Qin et al., 2020). It is possible that inhibition of HDL/apoA-I and its related transporters protein might be beneficial to reduce cholesterol efflux and promote cholesterol accumulation in NK cells. An observation that HDL inhibits NK cell killing in a dose-dependent manner is of clinical relevance although no significant correlation between NK cell activity and HDL-C levels has been found (Kim et al., 2019). This may be due to the differences in tumor types and their TME. Therefore, increasing cytotoxic NK cells and/or targeting TINKs may be effective for inhibiting tumor progression. However, the relationship between HDL and TINK has not been deeply investigated.

4.6 Myeloid-derived suppressor cells

MDSCs support tumor progression by promoting tumor cell survival, angiogenesis, invasion, and metastasis mainly through the three phenotypes: monocyte-bone marrow mesenchymal stem cells (M-MDSC), polymorphonuclear MDSC (PMN-MDSC), and early MDSC (eMDSC) (Gabrilovich et al., 2012). Tumor cells promote M-MDSC recruitment to tumors by secreting various chemokines, including CCL2, CCL5, and CXCL12 (Qian et al., 2011), and promote PMN-DMSC recruitment to tumors by secreting CXCL5 and CCL15. (Inamoto et al., 2016). In addition, increased expression of immunosuppressive molecules in the TME enhanced the immunosuppressive properties of PMN-DMSC and M-MDSC (Kumar et al., 2016). Recent studies have shown that the above characteristics are more strongly associated with M-MDSC (Mao et al., 2014), and HIF-1α induces the differentiation of M-MDSC into immunosuppressive TAM under hypoxic TME (Liu et al., 2014). MDSC can also produce immunosuppressive cytokines, such as IL-10 and TGF-β (Ostrand-Rosenberg & Sinha, 2009), induce Tregs (Pan et al., 2010), and affect NK cell function (Mao et al., 2014). In apoA-ITg mice treated with apoA-I reduced tumor-associated angiogenesis is associated with reduced numbers of MDSCs with angiogenic activity (Shojaei et al., 2007). The proliferation of myeloid cells (MCs) is increased in ABCA1/G1-/- mice. The proliferation phenotype of MCs reduced to baseline level after administration of exogenous HDL, indicating that HDL inhibits the proliferation of MCs independent of ABCA1/G1 (Yvan-Charvet et al., 2010). Through a mouse model of apoA-I-/-, researchers have found that injection of apoA-I can reduce metastasis and tumor development, but the number of MDSCs in the TME is decreased upon apoA-I treatment (Zamanian-Daryoush et al., 2013). Based on the abovementioned discussion on the HDL-mediated inhibition of MC proliferation, it is suggested that the anticancer effect of apoA-I might be due to apoA-I is maturated into cholesterol-rich HDL in vivo, then binds to SR-B1 to inhibit MDSCs. Plebanek et al. demonstrated the above speculation that in melanoma animal and lung cancer animal models, high-density lipoprotein nanoparticles (HDL-NPs), whose surface is endowed with apoA-I and phospholipid, combined with SR-B1 decreased MDSCs to express Arg1, NOS2, and CCL5, weakened MDSCs to inhibit T cell-mediated cancer immune response, and enhanced CD8+ and CD4+ T cells proliferation, mattering more to CD8+ T cells (Plebanek et al., 2018). Therefore, blocking the recruitment and proliferation of MDSCs might be an effective approach to inhibit tumor development.

5 EFFECTS OF HDL ON ANGIOGENESIS IN TME

Epidemiological studies show that in many circumstances, tumor micro-vessel density is negatively correlated with patient survival (Uzzan et al., 2004), while HDL levels are positively associated with the prognosis of patients with angiogenesis-related diseases (Tan et al., 2015). These findings indicate that HDL may be involved in inhibiting tumor angiogenesis. The occurrence and development of tumors depend on tumor angiogenesis. New vessels are important to the tumor as they provide oxygen and nutrients to sustain tumor proliferation, invasion, and migration. Some well-known vascular biological effects of HDL are mainly attributed to its anti-inflammatory, antioxidant, and unique role in RCT (Figure 3).

Details are in the caption following the image
HDL and angiogenesis in the TME. In hypoxic TME, HDL promotes the binding of HIF-1α to HRE in the nucleus to drive the transcription of angiogenic factors and induce angiogenesis by activating the PI3K/Akt pathway in endothelial cells. In addition, HDL augments VEGFR2 phosphorylation and triggers downstream activation of ERK1/2 and p38 MAPK to induce angiogenesis. In inflammatory TME, HDL binds to SR-B1 to inhibit the NF-κB pathway, binds to HRE to inhibit macrophage recruitment, and reduces CCL2 expression through SIP to inhibit angiogenesis. In addition, HDL also promotes ATF3 expression in macrophages and reduces the inflammatory response induced by TLR, thereby inhibiting angiogenesis. ABCA1, ATP-binding cassette A1; ATF3, regulator transcription factor 3; NF-κB, nuclear factor κ-B; SR-B1, scavenger receptor type 1; TLR, toll-like receptor

In inflammatory TME, HDL may directly affect endothelial cells and/or indirectly affect macrophages to inhibit inflammation-driven pathological angiogenesis. The IκB kinase β subunit can phosphorylate, ubiquitinate, and degrade IκB in endothelial cells (ECs) to release NF-κB into the nucleus for binding its response elements (Mercurio et al., 1997). Subsequently, the transcription of matrix metalloproteinase 9 (MMP9), vascular cell adhesion molecule-1 (VCAM-1), and intercellular adhesion molecule-1 (ICAM-1) (Chen, 2005) are initiated to promote macrophage recruitment and angiogenesis. It has been found that rHDL can inhibit the key steps of NF-κB pathway activation, including IKK activity, IκB phosphorylation, and NF-κB nuclear translocation (Bursill et al., 2010; Prosser et al., 2014), downregulate the expression of MMP9, VCAM-1, and ICAM-1, as well as reduce macrophage number and suppress macrophage activity via binding to its receptor scavenger receptor class B type I (SR-B1) on the surface of the endothelial cell membrane (Patel et al., 2010). In addition, the proangiogenic growth factor VEGF and its receptor VEGFR2 have NF-κB response elements and HIF-1α response elements in its promoter region, which could be activated under inflammatory conditions. Recent studies have shown that rHDL can reduce VEGF and VEGFR2 levels under inflammatory conditions, and downregulate MCP-1/CCL2 and TNF-α expression, significantly inhibit tubulogenesis, EC migration, and proliferation, thereby suppressing angiogenesis (Bursill et al., 2010). HDL and ABCA1/ABCG1-mediated cholesterol efflux inhibits angiogenesis by affecting lipid rafts, disrupting VEGFR2 dimer, and reducing inflammatory response (Fang et al., 2013). HDL-associated sphingosine-1-phosphate (S1P) can reduce angiogenesis by inhibiting the expression of E-selectin and CCL2/MCP-1 in ECs induced by TNF-α (Tölle et al., 2008). As an important coordinator of inflammation-driven angiogenesis, macrophages can secrete inflammatory cytokines, such as TNF-α, differentiation inhibitor-1 (Id1), VEGF, and thrombospondin 1 (THBS1) by activating Toll-like receptors (TLRs) to promote angiogenesis (Medzhitov & Horng, 2009; Takeuchi & Akira, 2010). TLR activation can induce transcriptional regulator ATF3 to inhibit the overexpression of proinflammatory factors through negative feedback regulation, and ATF3 is related to inflammatory angiogenesis (Gilchrist et al., 2006; Hai et al., 2010). As one of the downstream inhibitory targets of ATF3, Id1 can regulate the expressions of THBS1 and VEGF to regulate angiogenesis (Kang et al., 2003; Okamoto et al., 2006). Moreover, it is reported that HDL can inhibit TLR ligand-induced inflammatory response by activating ATF3 to reduce angiogenesis (De Nardo et al., 2014). In addition, HDL can also inhibit the expression of chemokine receptors (CCR2 and CX3CR1) in macrophages, thereby reducing macrophage recruitment and inhibiting angiogenesis (Bursill et al., 2010).

In hypoxic TME, HDL promotes angiogenesis by inducing migration and proliferation of endothelial cells. Under hypoxia, HDL can enhance stability to regulate the posttranslational regulation of HIF-1α to facilitate angiogenesis. The interaction between HDL and SR-B1 mediates the activation of the PI3K/Akt signaling pathway and induces the expressions of ubiquitin ligases Siah1 and Siah2. The Siah1/2 complex enhances the stability of HIF-1α by inhibiting the expression of proline hydroxylase PHD2 and PHD3 (Nakayama et al., 2004; Tan et al., 2014). HIF-1α then translocates into the nucleus, where it complexes with the HIF-β subunit and binds to the hypoxia response element (HRE) to promote the expressions of angiogenesis-related proteins, such as VEGF and VEGFR2, thereby inducing endothelial cell proliferation and migration to ischemic sites and promote angiogenesis (Prosser et al., 2014). HDL can enhance the phosphorylation of ERK1/2 and may therefore contribute to the proliferation of HDL under hypoxia (Prosser et al., 2014). VEGFR2 attenuates HDL-induced ERK1/2 phosphorylation during hypoxia, suggesting that HDL-induced ERK1/2 activation depends on VEGFR2. VEGFR2 can also activate the p38 MAPK signaling pathway, which mediates endothelial cell migration, a key cellular process required for angiogenesis (Pagès et al., 2000). It indicates that HDL can also promote angiogenesis by enhancing the phosphorylation of VEGFR2 and triggering the downstream activation of ERK1/2 and p38 MAPK.

In summary, HDL can promote angiogenesis in a hypoxia state, and inhibit angiogenesis in an inflammatory state (Tan et al., 2015). Regarding the hypoxic and inflammatory characteristics of TME, the effect of HDL on tumor neovascularization and growth may be related to the state in the TME or to the type of cancers, as the densities of blood vessels in different cancers and the pathological stimuli that drive their growth are different. Whether the inhibition of angiogenesis is one of the mechanisms by which HDL exerts antitumor effects is not known yet.

6 HDL AS A PHARMACOLOGICAL TARGET FOR TUMORS

In addition to epidemiological data, there seems to be pharmacological evidence supporting the relationship between HDL and tumors based on the clinical effects of drugs that can increase HDL levels. These drugs mainly include statins, fibrates, niacin, and CETP inhibitors.

Statins have been demonstrated to reduce the incidence of colorectal cancer, breast cancer, prostatic cancer, and so on (Bonovas, 2014; Chae et al., 2015; Pisanti et al., 2014). Another study has found that cancer patients taking fibrates have significantly lower mortality compared with untreated patients with dyslipidemia (Gardette et al., 2009). Niacin has been reported to increase HDL levels by 20%–25% (Kamanna & Kashyap, 2008). However, a 15-year follow-up study found that niacin had no effect on cancers (Canner et al., 1986). Traditional Chinese monomers have shown certain effects in improving HDL levels, including curcumin (Ganjali et al., 2017), berberine (Pirro et al., 2017), tanshinone (Jia et al., 2016), silybin (Rui, 1991), and other phytochemicals. Although the anticancer effects of curcumin (Momtazi et al., 2016), berberine (Pan et al., 2017), tanshinone (Luo et al., 2019), and silybin (Bosch-Barrera & Menendez, 2015) have also been described, whether these effects may be related to increased HDL levels has not been explored. Barter et al. Clinical studies have found that after treatment with the CETP inhibitor Torcetrapib, HDL levels have increased by 72%, but cancer patients have a higher mortality rate (Barter et al., 2007). This positive correlation between Torcetrapib and cancer risk remains unclear, and the relationship between Evacetrapib or Dalcetrapib and cancers has not been demonstrated. The only exception is that the increase in HDL-C caused by Anacetrapib is not associated with cancer incidence (Bowman et al., 2017).

Srivastava et al. indicated that very high HDL levels are directly proportional to mortality (Srivastava, 2018). Therefore, whether CETP inhibitors or other drugs increase HDL levels, evidence of benefit from cardiovascular events and/or mortality is required. In addition, HDL isolated from diabetic patients can induce the proliferation of breast cancer cell line MCF-7, while HDL from healthy controls can not induce MCF-7 proliferation. Furthermore, HDL from diabetic patients can increase the adhesion of cancer cells to endothelial cells, thereby promoting the metastasis of breast cancer (Huang et al., 2016). The above evidence confirms that improving HDL functions may be more effective than simply increasing HDL levels to treat tumors (Boden et al., 2011), indicating that HDL functions may be more important than HDL levels (Rader & Tall, 2012).

rHDL is a disc-shaped particle composed of phospholipids and apoA-I (or synthetic apoA-I mimetic peptide), which can transport anticancer molecules, and so forth (Krause & Remaley, 2013). Through a variety of tumor models in vivo and in vitro, Morin et al. (2018) found that rHDL plays an antitumor effect mainly by enhancing anti-inflammatory activity, inhibiting angiogenesis and eliminating proliferation, migration, and invasion induced by growth factors. Moreover, rHDL can effectively respond to TME by reducing MDSC, recruiting M1-TAM and cytotoxic CD8 + T cells (Zamanian-Daryoush & DiDonato, 2015). Another study showed that synthetic high-density lipoprotein (sHDL) can deliver antitumor drugs to the matrix sol of tumor cells in patients with hepatocellular carcinoma, promote DC maturation and induce CD8+ T cell responses, thereby jointly making tumors respond to PD-1 blockers sensitivity (J. Wang, Meng, et al., 2019).

7 CONCLUDING REMARKS AND FUTURE PERSPECTIVES

The relationship between plasma HDL-C levels and cancer incidence and mortality remains controversial. The majority of clinical studies support a lower incidence of cancer in patients with increased HDL-C levels. Although studies have shown that HDL is closely related to tumor progression, several key factors should be considered. Different HDL levels have a different correlation with tumor development and progression. HDL levels can inhibit the development of tumors within a certain range in vivo. The relationship between the therapeutic increase of HDL levels and the enhanced cancer treatment is an important research topic, as extremely high levels of HDL can also lead to an increase in the mortality rate of cancer patients, and low HDL is positively correlated with the incidence of cancer, indicating a possible U-shaped association between plasma HDL levels and cancer mortality. Therefore, the development of drugs targeting HDL needs to consider the extent of HDL increase. Cancer patients usually suffer from obesity, hypertension, diabetes, and other complications, or cancer-related hobbies, such as smoking and alcohol abuse. These confounding factors may lead to lower HDL levels in cancer patients, rather than cancer itself. In addition, studies in vitro have found that HDL indirectly inhibits the growth of tumor cells or promotes their apoptosis through inhibitory effects on other components of TME rather than on tumor cells themselves. The clinical failure of raising HDL levels by CETP inhibitors suggests that it may be more effective to improve the functions of HDL than only to raise the levels of HDL to affect nontumor cells in the TME.

The individual heterogeneity of cancer patients should also be considered. The technological development of informative tumor biopsy has made the precise composition of the stromal cell types in (primary and/or metastatic) the TME of the patients be revealed, allowing fine-tuning of therapeutic strategies with greater potential benefiting the patients. Revealing the precise composition of TME makes it possible to influence TME treatment against tumors through combination strategies. In addition to the internal pathways of tumor cells, the pathways and mediators of nontumor cells in the TME, including the inhibition of angiogenesis, inflammation, and immunosuppression can be the targets of HDL. The mechanisms underlying HDL-induced angiogenesis, inflammation, and immune response are not fully understood, and further addressing the crosstalk of different cells and pathways related to HDL in the TME is particularly important. Being an innovative area of research, HDL-like particles are used to influence various components in the TME. Clinical observations and experimental results are making the HDL-cancer story more interesting, attractive, and fascinating.

ACKNOWLEDGEMENTS

This study was funded by grants from the National Natural Sciences Foundation of China (Nos.: 81973668, 81774130, and 81270359), the National Science Fund of Hunan Province for Distinguished Young Scholars (No.: 2018JJ1018), and the First-Class Discipline of Pharmaceutical Science of Hunan.

    CONFLICT OF INTERESTS

    The authors declare that there are no conflict of interests.

    AUTHOR CONTRIBUTIONS

    All authors contributed to the development of this review article. Critical analysis and review of the literature were performed by Tan-Jun Zhao. The manuscript was written by Tan-Jun Zhao with revisions provided by Neng Zhu, Ya-Ning Shi, Yu-Xiang Wang, Chan-Juan Zhang, Chang-Feng Deng, Duan-Fang Liao, and Li Qin. @@All authors have read and approved the final version of the manuscript and gave consent for publication.

      The full text of this article hosted at iucr.org is unavailable due to technical difficulties.