Genetic determinants of oxidative stress-mediated sensitization of drug-resistant cancer cells

Abstract

Drug resistance in cancer is an overwhelming problem, because drug-resistant cancer cells are harder to kill with the same drug. The mechanism of drug resistance differs for various cancers based on the type of drug being used for its treatment. Most current drugs are shown to increase reactive oxygen species (ROS) in respective cancer cells that induces apoptosis, but continuous treatment with the same drug may reduce cellular ROS levels and may convert drug sensitive cancer cells into drug resistant cells. In addition, exogenous elevation of ROS in conjunction with drug resensitizes drug-resistant cancer cells. Thus, constant maintenance of higher ROS level in cancer cells may be a prerequisite for drug efficacy in certain type of cancer cells. Thus, modulation of ROS-mediated genetic pathway genes could be an efficient alternative to maintain higher ROS level in cancer cells for “combinational chemotherapy” with the drug. In this review, I discuss whether ROS reduction in drug-resistant cancer cells could be a general mechanism of drug resistance for most cancers with its specific drug, and whether elevation of ROS levels with the drug could be a valuable strategy for increasing drug efficacy in most cancers.

Cancer, a complex disease, arises as a result of a progressive accumulation of genetic aberrations and epigenetic changes that lead to the breakdown of normal cellular division checkpoints. According to the World Health Organization (WHO), an estimated 7.5 million patients worldwide die of this disease every year. What is remarkable is that most deaths occur after subsequent medical intervention with both conventional and novel, targeted anti-cancer therapies. Despite initial high response rates, a large proportion of patients relapse, resulting in a therapeutic challenge. Many of these patients are not curable with current chemotherapeutic strategies, and the goal of therapy is only to improve the quality and length of life.1 Consequently, more focus is needed to understand resistance in chemotherapy.

Biological barriers make it difficult for treatments to access specific target cancer cells. Efflux pumps expel the drugs from these cells into the blood leading to adequate levels of the drug in the bloodstream but only small concentrations actually reaching the targeted areas and cells. Moreover, isolated cellular areas, known as reservoirs, are responsible for drug accumulations.2 Current theories are related to cancer-initiating cells as a modified tissue stem cells, acquiring the property of self-protection by increasing successive drug resistance transporters.2 Therefore, improvements in the successful therapeutic management of cancer require an understanding of multidrug resistance (MDR) mechanisms, and the identification of the underlying drug resistant pathways, for successful manipulation of genes for all classes of tumors. This process is believed to be dependent on the cell and microenvironmental context that includes oncogene expression, apoptosis mechanism, cell cycle control and regulation, DNA repair and mutation, redox regulation and vascularization and many others.

Reactive Oxygen Species Content Is Higher in Almost All Cancer Cells

Reactive oxygen species (ROS) is inevitably generated through cellular metabolism. Redox regulation in cancer cells is complex and almost all cancer cells exhibit elevated levels of endogenous ROS.3, 4 Elevated ROS level is also observed in some tumors those have been implicated in HIF-1 signaling.5 Oxidative radical generation and the endogenous concentration are important for the cellular function in cancer.4

Redox Management in the Extracellular and Intracellular Tumor Microenvironment

Under physiologic conditions, the extracellular space is known to have a relatively more-oxidized state than the interior of the cell.6 In cancer, the extracellular redox state may be altered, resulting in specific proteases, soluble factors, or the extracellular matrix having altered functions or activities. Recent studies also strongly support the important relations between the extracellular redox state and cancer cell aggressiveness.7 ROS compartmentalization and its distribution within cancer cells also play a critical role in tumor progression.8

How and Why ROS Is Elevated in Tumors

The mitochondria communicate with the nucleus through the production of ROS, a byproduct of aerobic respiration within the mitochondria. Mitochondrial signaling is altered in cancer cells, where the mitochondrial genome is frequently found to harbor mutations and display altered functional characteristics leading to increased glycolysis. Mitochondrial oxidative damage triggers a glycolytic switch (Warburg effect)9 and activates redox-sensitive transcription factors, resulting in an increase in cell resistance to oxidants. Cells may regulate ROS levels by eliminating ROS-scavenging systems such as superoxide dismutases (SOD1, SOD2 and SOD3), glutathione peroxidases, peroxiredoxins, glutaredoxins, thioredoxins and catalases.10 In an oncogenic Kras-driven mouse model of lung cancer, anchorage-independent growth requires ROS generation at the mitochondrial complex III, and disruption of this mitochondrial function through knockdown of the mitochondrial transcription factor A (TFAM) reduces tumorigenesis. These results demonstrate that mitochondrial ROS generation are essential for Kras-induced cell proliferation and tumorigenesis.9 Transduction of mitochondrial signal in the cytoplasm is also mediated by several GTPase and RAS family of proteins, precisely through isoprenelation.11 As signaling molecules, ROS oxidize and inhibit p38 mitogen-activated protein kinase (MAPK) phosphatases resulting in enhanced proliferation and survival and are particularly advantageous to cancer cells. ROS also affect transcription through phosphorylation, activation and oxidation of transcription factors such as APEX1, NF-kappaB, p53 and HIF-1alpha leading to changes in target gene expression.12–15 Increased ROS production by defective cancer cell mitochondria is observed with upregulation of ETS-1 that has been increasingly associated with aggressive cancers.16

Recently, the role of antioxidants has also been demonstrated in cancer stem cell self-renewal, complex cell programming involving enhanced cell motility and in therapeutic resistance.2, 17 ROS concentrations act as a double-edged sword for tumor progression as high concentrations of ROS are needed for cancer progression but are toxic to normal cells. ROS act as a second messenger and its signaling is necessary for cancer metastasis, cellular adhesion and spreading.18, 19 According to a “canonical” view, ROS positively contribute to carcinogenesis and malignant progression of tumor cells by driving genomic damage and genetic instability, thereby transducing signaling intermediates. An attractive hypothesis about metastasis suggests that cell spreading is an integrated strategy for cancer cells to avoid oxidative damage from excess ROS at the primary tumor site, thus also explaining why redox signaling pathways are often upregulated in malignancy and metastasis.20 Therefore, intracellular ROS level in cancer cells could be exploited in two ways, by decreasing it to inhibit early tumorigenesis or by increasing it with the drug to kill cancer cells (Fig. 1).

Figure 1. Flowchart depicting effect of ROS level on drug sensitive and resistant cancer cells. ROS content in cancer cell is not only a modulator of tumor progression and metastasis but also plays a major role in the mechanism of drug sensitivity and drug resistance. When normal cells become cancerous, ROS level increases. Tumorigenesis could be inhibited by reducing ROS level through inhibiting ROS generating systems. Again, anticancer drugs initially induce ROS and kill cancer cells through apoptosis, but in prolonged drug treatment, ROS level is reduced and cells become drug-resistant. Exogenous ROS in combination with drug again resensitize drug resistant cells and kill through apoptosis.

Early Tumorigenesis Could Be Inhibited by Decreasing ROS Levels

Therapeutic antioxidants prevent early events in tumor development where ROS are important for signaling carcinogenesis.19 Thus, it seems reasonable to counteract ROS by inhibiting ROS production. Such therapeutic agents include polymeric SOD, xanthine oxidase inhibitor, heme oxygenase-1 inducers, radical scavengers (e.g., canolol)21 or synthetic antioxidants (propyl gallate).22 However, our increased understanding of genetic and genomic advances in cancer growth suggests that effective targeting of cancer cells requires identification of specific ROS-sensing signaling pathways that modulate diverse stress-regulated cellular functions.19 When A2780 ovarian cancer cells are treated only with exogenous ROS, profound changes in gene expression patterns are observed, several ROS inducing genes (Table 1) are identified and these genes are known to be involved in tumor progression.23 When the dose of exogenous ROS is well below the lethal level, expression changes in apoptotic genes are hardly noticeable. In contrast, a ROS inducing gene, CYR61, an angiogenic inducer and important regulator of cancer progression is upregulated. Its expression is not only highly induced by ROS, it acts on several cell proliferative genes that are known to promote tumor progression.23, 24 Downregulation of CYR61 by zoledronic acid induces apoptosis in prostate cancer cells with a significant reduction of intracellular ROS levels.25, 26 Zoledronic acid also shown to have clinical benefit in breast, bone, lung and renal cancer.27 Thus, these results are in concordance with the hypothesis that ROS induce progression of early stages of tumor formation and downregulation of CYR61 gene or protein or regulating its interacting small molecules could be effective for efficient drug design (Fig. 2) for the inhibition of tumorigenesis.

Figure 2. CYR61 and its modulating genes could be efficient drug target. IPA analysis identified many modulator of CYR61 gene. Dark shaded genes changes their expression in drug resistant ovarian cancer cells but could also play similar role in other cancers where CYR61 level is elevated such as breast, bone and lung cancer. Light shaded are connector genes and empty boxes are small molecules. CYR61 acts in three ways—it is regulated by several genes or agents such as CA2NA2D2, Mir205, epoprostenol and so on and it activates VEGFC and VEGFA. However, VEGFA and VEGFC expression could be further modulated by several genes and agents. Thus, effective targeting or modulation of these genes to downregulate CYR61 could be useful for treating cancer efficiently. CYR61 also induces ARHGEF6 and may play important role in maintaining ROS.

Table 1. Genes those do not or very little express in drug sensitive or resistant cells but express highly after very low doses (5 μM) of ROS treatment in drug resistant cells

Several Chemotherapeutic/Anticancer Agents Induce ROS Production Leading to Apoptosis in Various Tumor Cells

Induction of ROS through viral protein

Rat parvovirus H-1 (H-1PV) acts as an anticancer agent with oncosuppressive properties and it is nonpathogenic in humans. The viral NS1 protein induces apoptosis in cancer cells through inducing ROS that is associated with DNA damage and caspase activation.28 However, although nonpathogenic, treating cancer with whole virus would be a risk as viral RNA/DNA integration into human genes could make any gene nonfunctional and increase complications. Instead, NS1 protein itself could potentially be developed as an effective anti-cancer drug.

Induction of ROS through chemical agents

A naturally occurring ROS-inducing compound, beta-phenyl ethyl isothiocyanate selectively killed cancer cells but not normal cells.29 Another promising compound, NSC-741909 induces robust ROS generation in sensitive but not in resistant cancer cells.30 Celastrol, a triterpene, facilitates TRAIL-induced apoptosis by downregulating cell survival proteins and upregulating death receptors.31 These molecules have the potential to be effective for all cancer cells, although ROS induction through anticancer drugs is discretely studied and limited to only specific type of cancer cells. Initial ROS induction seems to be common for many anticancer agents for most of the cancers (Table 2).

Table 2. Anticancer agents those induce cell death through induction of ROS generation in various cancer cells

Breast cancer cells

Various anticancer agents, including those in the process of drug development for breast cancer, are shown to induce ROS generation through mitochondria. Rotenone and thermoquinone treatment activates ROS, resulting in apoptosis through c-jun N-terminal kinase, MAPKs and the inactivation of extracellular signal-regulated protein kinase 1/2 (ERK1/2).34 Sulforaphane, a dietary isothiocyanate, possesses anti-cancer and chemopreventive activities through ROS generation by suppressing telomerase activity via transcriptional and post-translational regulation of hTERT (human TElomerase Reverse Transriptase), which has been implicated in redox-mediated events.35, 52

Leukemia

The combination of a reducing agent (ascorbate) and a redox active quinone (menadione) generates ROS and induces apoptotic cell death in K562 cells. The mechanism underlying this cytotoxicity involves the oxidative cleavage of Hsp90 causing subsequent loss of its chaperone function, thus leading to degradation of wild-type and mutated Bcr-Abl proteins. These cancer cells are more sensitive to ascorbate/menadione than healthy cells due to their lack of antioxidant enzymes, mainly catalase.39 In contrast, rosmarinic acid sensitizes cell death in human leukemia U937 cells through suppression of ROS generation and TNF-alpha(Tumor Necrosis Factor- alpha)-induced NF-kappaB activation.40

Lung cancer cells

Tanshinone IIA and MG132 induce apoptosis in many cancer cells, including A549 cells, through increasing ROS and possibly by decreasing the mitochondrial membrane potential (MMP) and glutathione (GSH) depletion.42 Another compound, Phx3, also induces cellular apoptosis by generating ROS in these cells. Microscopic examination confirmed that Phx-3 was mainly localized in the cytoplasm and mitochondria, but not in the nucleus emphasizing its role in cellular ROS production and apoptosis.43

Ovarian cancer cells

Chlorambucil or cisplatin initially induce ROS levels in A2780 cancer cells similar to those observed in other types of cancer cells, but extensive and systematic studies revealed that continuous drug treatment reduces cellular ROS levels and that reduced ROS-containing cells became drug-resistant cells.23

Colorectal cancer cells

Colorectal cancer cell lines were most sensitive to anthocyanidins, leading to cellular ROS accumulation, inhibition of glutathione reductase and depletion of glutathione and undergoes apoptosis.45

Prostate cancer cells

As both prostate cancer and aberrant changes in ROS become more common with aging, ROS signaling may play an important role in the development and progression of this malignant cancer. ROS has been shown to mediate apoptosis-inducing activity of CDDO-Me (C-28 methyl ester derivative methyl-2-cyano-3,12-dioxooleana-1,9(11)-dien-28-oate) in PC-3 prostate cancer cell lines through loss of MMP and release of cytochrome C.46

Urinary bladder cancer cells

In oncogenic H-Ras-expressing J82 urinary bladder cells, oncogenic H-Ras expression and FK228 treatment synergistically induce the ERK pathway, resulting in differential increase of Nox-1 elevation, ROS production and GSH depletion. This leads to differential caspase activation and apoptotic cell death.50

Liver cancer cells

Chrysophenol induces necrotic damage through the production of ROS, alteration of ATP levels and kills J5 human liver cancer cells by necrosis but not by apoptosis.48

Cervical cancer cells

Pigment epithelium-derived factor, an angiogenesis inhibitor, showed an antiproliferative effect on cervical cancer cell lines by inducing apoptosis through generation of ROS and accumulation of cells in the sub-G1 phase.53

Pancreatic cancer

Pancreatic cancer cells when treated with MDA-7/IL24 and a ROS inducer (arsenic trioxide or dithiophene) showed increased apoptosis and this effect could be reversed by ROS inhibitors such as N-acetyl-L-cysteine implying the role of ROS in inducing cell death in these cells.

Drug resistance Could Occur due to Depletion of ROS in Drug-Treated Cancer Cells

Although various drugs induce ROS initially in most cancer cells, the role of ROS were not extensively investigated after the drug sensitive cells became drug resistant cells. It has been observed that cisplatin or chlorambucil also initially induces ROS production in ovarian carcinoma A2780 cells as observed similarly for other drugs in various types of cancer cells, but prolonged drug treatment with either drug actually reduces ROS level making those cells resistant to the drug. Additionally, the generation of exogenous ROS in combination with the drug increases sensitivity to these drugs in resistant cells.23, 54 Thus, a decrease of ROS level in prolonged drug-treated cells is not a secondary cellular outcome, instead a primary mechanism for specified drug resistance. Although yet to be investigated, similar underlying drug resistant mechanisms due to reduced ROS level may be presumably common for other cancers for a specific drug that initially induces ROS (Table 2). It appears that cellular ROS level in cancer cells determines whether specified drug would induce apoptosis or escape sensitivity to induce tumorigenesis and to become resistant (Fig. 3).

Figure 3. Effect of anticancer drugs in modulating ROS to activate tumorigenesis or apoptosis in drug resistant cancer cells. Through knowledge based IPA analysis, the critical roles of ROS have been demonstrated for both tumorigenesis and drug dependent apoptosis for several anticancer agents. These anticancer agents were shown to induce ROS level initially in various cancers cells and could induce apoptosis through TP53, DUSP2, HSPA1A, JUN, APEX1 and CASPs. However, prolonged treatment of these anticancer drugs could also reduce ROS level to activate tumorigenesis through tumor inducing genes, CYR61, ARHGEF6, TGFB1, CDK6, VEGFC and VEGFA.

However, how these chemotherapeutic agents after initial ROS induction further reduce ROS levels to become resistant cells, or utilize ROS to induce apoptosis or deplete intracellular ROS, is largely unknown. One of the mechanisms could be anticipated that increased ROS generation by anticancer agents increases antioxidant enzymes level in cancer cells that subsequently reduces ROS level. When an antioxidant enzyme, catalase expression in the membrane is inhibited, induction of apoptosis through intercellular ROS signaling has been restored in resistant cancer cells.55 However, such a mechanism assumes transcription/production of antioxidant enzymes must be directly or indirectly induced by ROS. In such cases, defining the effect of cellular ROS inducing (for ovarian carcinoma cells; Table 1), maintenance (such as ARHGEF6, CDK2 and p53) and redox balance pathway genes (such as APEX1 and PRKDC) or proteins on transcription of antioxidant genes is an important criterion for understanding ROS levels in resistant cancer cells with or without drug treatment.

Identification of Genetic Pathways Leading to ROS-Mediated Apoptosis in Drug-Resistant Cancer Cells

ROS have been proposed as common mediators for apoptosis and the mode of cell death depends on the severity of the oxidative damage.18 p53, the most studied tumor suppressor protein, also directly controls the metabolic traits of cells and generated a certain level of p53-inducing stress, including mitochondrial ROS generation that is associated with increased DNA damage and apoptosis.12 A study of human larynx cancer cells treated with phorbol myristate ester and ionizing radiation suggested that increase of MDR depended on increased DNA-PKC (PRKDC) activity and ROS content in the cells.56 Integrins and matrix metalloproteinases mediate ROS production through ITGB4 and MMP expressions to induce apoptosis in cancer cells.57 Thus, the necessity for activation of both signalization and stress reaction genes is important for overcoming drug resistance in tumor cells.

Both ROS generation and GSH depletion could also occur in drug-treated apoptotic cancer cells.44 This suggests that both processes are genetically interlinked, and several genes in these pathways are activated in drug resistant cancer cells.10 Rho family GTPases are critical regulators of the cytoskeleton, cell migration, cell–cell adhesion and their functions or activity depends on guanine nucleotide-bound state through regulatory proteins such as guanine nucleotide exchange factors (GEFs) and GTPase activating proteins. ROS activate GTPase RhoA proteins through two critical cysteine residues located in a unique redox-sensitive motif within the phosphoryl binding loop.58, 59 Microarray gene expression associated with IPA analysis identified several genes (Fig. 3) those are active component in the ROS maintenance pathway and are known to induce apoptosis. A GEF gene, ARHGEF6, is identified as its expression is elevated in drug-resistant ovarian carcinoma cells and decreased in conjunction with the drug and exogenous ROS, thereby reinstating sensitivity implying this gene could be a potential drug target for overcoming drug resistance.23 Gene network analysis shows that ARHGEF6 could play a dual role through inducing tumorigenesis in less ROS content cells (drug resistance) and inducing apoptosis (drug sensitivity) in higher ROS content cells (Fig. 3) and its expression could be modulated with several genes or small molecules to maintain cellular ROS level (Fig. 4).

Figure 4. Identification of small molecules and miRNA that regulate ROS maintaining genes. ARHGEF6 and CDK6 are ROS maintaining genes and they act on TP53, HSPA1A and CFLAR to induce apoptosis. These two ROS maintenance genes have critical role in ROS-reduced drug resistance mechanism and their regulation could be manipulated through various genes, small molecules or through many miRNAs for increasing drug sensitivity.

Redox signaling and impairment of DNA repair are also prerequisite for signaling apoptosis.60 A cellular redox maintaining gene, APEX1 modulate DNA repair activity and MDR in lung cancer cells implying these two processes could be related with redox dependent mechanism.15 Although, many drugs generally bind DNA directly and induce DNA damage, increased ROS signaling may be necessary for excessive DNA damage to induce apoptotic signaling.

Recent evidence demonstrated an association between a number of single nucleotide polymorphisms (SNPs) in oxidative DNA repair genes and antioxidant genes with human cancer susceptibility.61 Four SNPs in EGF, three SNPs in PRDX4 and XPC, and two SNPs in GSTA4, TGFBR2, TNFAIP2, BCL2, DPYD and EGFR are associated with docetaxol clearance in cancer cells. Although the patient number was small (24), this could pave the way for epidemiological studies that identify drug resistance genes directly in drug treated cancer patients.

Growing evidence also supports a role for oxidative stress in modulating epigenetic processes and altering gene expression.62 It is therefore of great importance to begin identification of aberrant DNA methylation patterns that could lead to identify potential biomarkers for drug resistance in cancer.

The role of oxidative stress evaluated ex vivo in cancer patients is also emerging as a surrogate marker of response to target-based agents.63 When hepatocellular carcinoma patients were treated with sorafenib, increase of SOD and decrease of ERK1/2 is associated with reduction of ROS level. Thus, by measuring ROS level, it would be possible to follow up the response of drug treatment in cancer patients.

Developing ROS-Induced Therapy for Drug-Resistant Cancer Cells

Increased generation of ROS and an altered redox status might not only be exploited for the treatment of primary tumors but also has the potential to be used for sensitizing drug-resistant cancer. Numerous antitumor agents (Table 2), as earlier described, induce ROS production and activate ROS-dependent apoptotic pathways. As observed in ovarian carcinoma cells, chlorambucil or cisplatin initially increase ROS levels, but prolonged treatment reduces ROS levels in cancer cells resulting in drug resistance. Thus, constant maintenance of higher ROS level in cancer cells is necessary for better drug efficacy. Using a “combination therapy” of drug and exogenous ROS could be an efficient alternative to increase the efficacy of drug treatments by constantly maintaining higher ROS levels in cancer cells, thereby precluding drug resistance. The drug sensitivity of drug resistant cancer cells could be achieved in combination with the drug and generating additional ROS, mainly by (i) targeting mitochondrial ROS generating systems, (ii) modulating the ROS maintaining genes or (iii) introducing ROS directly into solid tumors.

Targeting mitochondrial ROS generating systems

Manipulation of drug-resistant mechanisms by redox modulation could have significant therapeutic implications by inhibiting the antioxidative enzyme systems of tumor cells.10, 55, 64 Using optimal concentrations of the catalase inhibitor [3-aminotriazole (3-AT)], or site-specific generation of hydroxyl radicals at the cell membrane of the tumor, could prove useful for maintaining excess ROS levels in cancer cells.55

Drugs such as mitocans are being developed that selectively target the mitochondria of malignant cells without adversely affecting those of normal cells. Mitocans selectively interfere with the bioenergetic functions of cancer cell mitochondria, causing major disruptions that lead to increased ROS production and induction of intrinsic apoptotic pathways.65

Casiopeinas, a series of mixed chelate copper complexes, induce mitochondrial DNA damage and imbalance the expression of the apoproteins of the mitochondrial respiratory chain, and increase ROS production,66 thereby attracting attention as suitable mitochondria targeting anti-cancer drugs.

Modulating the expression of ROS maintaining genes in drug-resistant cancer cells

In another approach, deciphering the pathways involved in ROS maintenance genes in drug-resistant cancer cells and modulating their expression to increase cellular ROS level. Studies in drug-resistant ovarian carcinoma cells enabled identification of at least two ROS maintenance genes, ARHGEF6 and CDK6, with a role in tumor progression and spreading.23, 67 Successful manipulation of these genes, proteins or regulatory molecules of these genes/proteins such as miRNA (Fig. 4) could be efficiently tested in cellular and animal models to determine their effectiveness in inducing apoptosis in drug-resistant tumor cells.

Introducing ROS directly into solid tumors

Liposome-polycation-DNA (LPDI and LPDII) nanoparticles conjugated with a ROS generating agent are the promising discovery for treating cancer cells. LPD nanoparticles with a guanidinium containing cationic lipid such as DSAA (N,N-distearyl-N-methyl-N-2-(N′-arginyl) aminoethyl ammonium chloride) can induce ROS, downregulate MDR transporter expression (Pgp-glycoprotein-mediated drug efflux), increase doxorubixin uptake and show a significant improvement in tumor growth inhibition.68 In another approach, multicellular tumor spheroids (MCTS) also offer an excellent in vitro system that mimics the endogenous oxidative stress often observed in tumors and could be potentially exploited for ROS generation in vivo.69 Thus, designing suitable vehicles to efficiently deliver or generate ROS in conjunction with specific drugs could help to overcome drug resistance in cancer cells in vivo.

In summary, ROS balancing and distribution in cancer cells involve complex mechanisms and play a significant role in drug resistance. Most of the currently used drugs initially induce ROS generation, although ROS levels in drug-resistant cancer cells are not studied extensively. Prolonged drug treatment reduces ROS level in ovarian carcinoma cells, causing sensitive cells to gain resistance. However, it remains to be determined whether this mechanism for drug resistance is common for most cancers for any specific drug. If so, manipulation of ROS levels through the modulation of ROS generating or maintaining genes or direct administration of ROS in combination with the drug could be an efficient alternative for future cancer chemotherapy.

Acknowledgements

The author thanks Ms. Julie Maier and Dr. T. Johnson for critically reading this manuscript. He sincerely thanks anonymous reviewers for improving the quality of this manuscript.