An official website of the United States government
NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.
Vink R, Nechifor M, editors. Magnesium in the Central Nervous System [Internet]. Adelaide (AU): University of Adelaide Press; 2011.
The relationship between Mg and cancer is still a puzzle to disentangle. The knowledge derived from preclinical studies reveals a complex scenario in which low magnesium has both anti- and pro-tumour effects, such as inhibition of tumour growth at its primary site and facilitation of tumour implantation at its metastatic sites. In different cell types, neoplastic transformation dramatically disrupts the controlled and coordinated fluctuations of intracellular magnesium, an event that offers selective advantages to the cells. It is difficult to translate the lesson learnt from experimental models to humans. Based on epidemiological studies, Mg deficiency seems to be linked to increased risk of some types of cancers. The demonstration of an impairment of magnesium homeostasis in oncologic patients further complicates the field. We need more translational and clinical data to draw firm conclusions about the contribution of magnesium to tumours.
Although cancer mortality has declined in the last two decades, particularly in western countries (Jemal et al., 2010), cancer exacts a very high toll as a leading cause of death all over the world. Great strides have been made to disclose the molecular bases of neoplasia, but how to prevent or cure cancer still remains a largely unanswered question.
There is general agreement about the environ- mental origin of cancer. It has been estimated that more than two-thirds of human cancers can be prevented by changes in lifestyle. Statistical and epidemiological data point to diet as responsible for about 35% of human cancer mortality (Doll and Peto, 1981). While many dietary constituents transform normal cells into malignant ones in vitro and stimulate the development, growth and spread of tumours in vivo, evidence from population and laboratory studies indicate that many microconstituents inhibit, retard or reverse tumorigenesis (Manson, 2003). There is general agreement about the inverse correlation between regular consumption of fruit, cereals and vegetables and the risk of cancer. Edible plant matter contains many beneficial microconstituents, vitamins and minerals. In particular, magnesium (Mg), which is predominantly obtained by eating unprocessed grains and green leafy vegetables, is an essential micronutrient implicated in a large array of regulatory, metabolic and structural activities (Wolf and Trapani, 2008). Evidence suggests that the occidental diet is relatively deficient in Mg because of a preference for calorie-rich, micro- nutrient-poor foods, low Mg content in water and soil, and processing of many food items (Ford and Mokdad, 2003). Inadequate dietary intake of Mg is not the only cause of Mg deficiency. Mg homeostasis is very tightly tuned in the intestine and the renal tubules through a complex network of transporters, some of which have recently been defined at the molecular level (Quamme, 2010). Consequently, Mg deficiency accompanies chronic gastrointestinal and renal diseases, therapies with some classes of diuretics or anticancer drugs, and also complicates diabetes mellitus. In addition, it is common in alcoholics and in the elderly (Ford and Mokdad, 2003).
Considerable debate remains about the role of Mg in human tumours, partly due to the lack of reliable and selective analytical procedures to measure Mg. However, there has been a recent resurgence of interest in the relationship between Mg and tumours both in experimental and in clinical oncology, leading to the question whether Mg is Dr Jekyll or Mr Hyde in tumour biology and clinics.
Although not conclusive, a clinical study indicated a potential benefit of Mg deficiency in patients with malignant tumours (Parsons et al., 1974). These results have been sustained by the fact that Mg is required in all the steps involved in cell growth, from receptor-mediated intracellular signalling and transphosphorylation reactions to gene transcription, protein synthesis, DNA duplication and cell division (Wolf and Trapani, 2008). Another study showed that the serum level of Mg is frequently decreased in patients with solid tumours, independent of therapies, and that this decrease correlated to the stage of malignancy (Sartori et al., 1992). This finding seems to be due to the avidity of the tumours for Mg, which behave as Mg traps. Also, therapies influence Mg homeostasis. It has been shown that serum Mg decreases at the end of the first week of radiotherapy (Cohen and Kitzes, 1985) as well as after treatment with different chemo- therapeutics that induce Mg waste. Indeed, cisplatin is nephrotoxic (Yao et al., 2007), and cetuximab, a monoclonal antibody against the Epidermal Growth Factor (EGF) receptor, reversibly and specifically inhibits Mg reabsorption in the renal distal convoluted tubule (Muallem and Moe, 2007). An obvious question is whether radiation- or drug-induced hypomagnesemia amplifies the effect of DNA damaging cancer treatments by acting as a chemo- and radio- sensitizer. Recently, decreased serum Mg has been suggested to contribute to the therapeutic effects of cetuximab in patients with colon carcinoma (Vincenzi et al., 2008). However, it is still controversial whether to supplement severely hypomagnesemic oncologic patients with Mg or not (Wolf et al., 2009).
Several epidemiological studies have provided evidence that a correlation exists between dietary Mg and various types of cancer. A high content of Mg in drinking water seems to protect from liver and oesophageal cancer (Tukiendorf and Rybak, 2004; Yang et al., 2002). In addition, Mg content in drinking water was inversely correlated with death by breast, prostate, and ovarian cancers, whereas no correlation existed for other tumours (Yang et al., 2000a; 2000b; 2002; Chiu et al., 2004). In particular, a high serum calcium Ca:Mg ratio has been suggested as a novel risk factor which increases the development of postmenopausal breast cancer (Sahmoun and Singh, 2010), the most commonly diagnosed cancer among women in North America and Europe, with a mortality that ranks second only to lung cancer (Jemal et al., 2010). Indeed, in western populations, and even more in postmenopausal women who are taking Ca supplements to prevent osteoporosis, a high Ca:Mg intake is rather common and determines a negative Mg balance, since the two minerals compete for the same transporters in almost all tissues. Ca and Mg have equally important roles in regulating cell growth (McKeehan and Ham, 1978) and Mg is considered the natural Ca antagonist. It is therefore feasible that alterations in Ca:Mg ratio might lead to dysregulated cell functions. The ratio of Ca:Mg is also to be considered in relation to colon cancer, one of the three big killers, together with breast and lung cancer, in western society. Ca has been indicated as a chemopreventive agent for colon cancer (Lamprecht and Lipkin, 2003).
It should also be pointed that Mg might have a role in colon cancer prevention. Indeed, large epidemiological studies in Sweden, in the Netherlands, in the USA and in Taiwan have demonstrated an association between low intake of Mg and the risk of colon cancer (Larsson et al., 2005; Folsom et al., 2006; van den Brandt et al., 2007; Chiu et al., 2010) and a large population- based prospective study in Japan showed a significant inverse correlation between dietary intake of Mg and colon cancer in men, but not in women (Ma et al., 2010). Intriguingly, the association between low intake of Mg and colon cancer seems to be due, at least in part, to the increased formation of N-nitroso compounds, most of which are potent carcinogens (Chiu et al., 2010). That Mg has a role in colon neoplasia is supported by the association of adenomatous and hyperplastic polyps with a genetic polymorphism of Transient Receptor Potential Melastatin (TRPM)-7 (Dai et al., 2007), a ubiquitous ion channel with a central role in Mg uptake and homeostasis (Schmitz et al., 2003). Indeed, the subjects carrying the Thr1482-Ile variant allele are at high risk of Mg deficiency and, in turn, of colorectal neoplasia (Dai et al., 2007). While total Mg consumption has been linked to a significant lower risk of colorectal adenoma and hyperplastic polyps (Dai et al., 2007), unequivocal evidence that Mg reduces the risk of cancer will only be obtained by demonstrating that Mg supplementation prevents colorectal cancer.
The role of Mg in lung cancer is rather controversial. The first, case-control study correlated low dietary Mg with poor DNA repair capacity and increased lung cancer risk both in men and women (Mahabir et al., 2008). This link was more evident in the elderly, current smokers, drinkers and in those with a late-stage disease. Apart from its role in maintaining genomic stability, the authors proposed several additional mechanisms by which Mg might have protected against lung cancer, i.e. by regulating cell multiplication and protecting against oxidative stress, invariably associated with Mg deficiency (Guerrero-Romero and Rodriguez-Moràn, 2006). Unexpectedly, these results were not supported by a recent prospective analysis which showed that dietary Mg intake slightly increased lung cancer risk (Mahabir et al., 2010). These contrasting data could result from recall bias. In addition, residual confounders may account for weak associations because smoking is a very strong risk factor for lung cancer, and diet might not be perfectly measured. More studies of dietary Mg and lung cancer risk are required to carefully consider measurement error issues and confounders such as smoking.
A last topic to consider is the implication of inflammation in the initiation and development of cancer in Mg deficient individuals. The inflammation-cancer connection is a well- established paradigm (Colotta et al., 2009) and low Mg status has been clearly linked to increased inflammatory stress in humans (Nielsen, 2010). While in some context, such as lymphomas, inflammation exerts an anti-neoplastic role, cancer-related inflammation is involved in the early and late steps of the most common solid tumours. Indeed, inflammatory mediators induce genetic instability and contribute to the proliferation of tumour cells, promote invasion, metastasis and angiogenesis as well as an impaired response to therapies (Colotta et al., 2009).
In summary, Mg deficiency seems to be linked to an increased risk of some types of cancers in humans, but population studies are rather complicated to interpret. Too many issues need to be considered – genotype, risk factors, interactions among different nutrients and, in particular, the Mg:Ca ratio, the degree of the inflammatory response, etc – to unravel the role of low Mg in the complex, multistep process of human cancerogenesis.
Two main questions have been addressed in animal models: 1) whether a relationship exists between Mg and the sensitivity to various carcinogens; and 2) whether primary tumour growth and metastasis are affected by Mg deficiency. Several studies indicate that Mg exerts a protective effect in the early phases of chemical cancerogenesis. Mg prevents lead- and nickel-induced lung tumours in mice (Poirier et al., 1984), inhibits nickel-induced carcinogenesis in the rat kidney (Kasprzak et al., 1994) and protects against 3-methyl-cholantrene induced fibrosarcomas in rats (Patiroğlu et al., 1997). Analogous with data in humans, Mg acts as a protective agent in colorectal cancer by inhibiting c-myc expression and ornithine decarboxylase activity in the mucosal epithelium of the intestine (Mori et al., 1997). Whether Mg acts directly or by controlling inflammation is an issue that has not received adequate consideration until now. However, it should be pointed out that the inflammatory response in Mg deficient rodents is very accentuated (Mazur et al., 2007) and, as mentioned above, inflammation markedly contributes to cancer.
Back to the second question, i.e. whether Mg deficiency impacts on tumour growth in vivo, a study showed that dietary Mg deprivation inhibited tumour growth in rats by limiting glutathione (GSH) synthesis (Mills et al., 1986), for which Mg is an obligatory cofactor. More recently, in mice subcutaneously injected with Lewis lung carcinoma, mammary adeno- carcinoma and colon carcinoma cells, a low Mg- containing diet was shown to inhibit primary tumour growth, an effect that was promptly reversed by re-introducing Mg in the diet (Nasulewicz et al., 2004). Surprisingly, Mg deficient mice developed far more lung metastases than controls. Some cellular and molecular aspects were investigated to explain how a low Mg diet inhibited the development of primary tumours while enhancing metastatization. Regarding primary tumours, it is noteworthy that Mg deficiency directly inhibited tumour cell growth by downregulating cyclin B and D3, crucial for the progression through the cell cycle, and by upregulating p21, p27 and Jumonji, all involved in braking cell proliferation (Nasulewicz et al., 2004; Maier et al., 2007). Mg deficiency also affects tumour development indirectly, because it promotes DNA oxidative damage (Maier et al., 2007) and impairs angiogenesis, one of the hallmarks of cancer (Hanahan and Weinberg, 2000). Indeed, Mg- deficient mice develop tumours that are significantly less vascularized than controls (Maier et al., 2007). The angiostatic effect of low Mg can be ascribed to two different mechanisms. First, low Mg inhibits endothelial growth and migration, pivotal steps in the formation of new vessels (Bernardini et al., 2005). Second, it has recently been shown that Mg deficiency suppresses hypoxia-inducing factor (HIF)-1α activity in paraganglion cells (Torii et al., 2009). HIF1α is fundamental to initiate the cell response to hypoxia and, therefore, the release of angiogenic factors. If this finding will also be confirmed in in vivo models, it might be involved in impairing angiogenesis upon Mg deficiency. In vitro, the suppression of HIF-1α activity is due to the upregulation of IPAS, which acts as a dominant negative inhibitor by preventing HIF-1α DNA binding. Intriguingly, IPAS upregulation is caused by the activation of NFκB, the transcription factor orchestrating the inflammatory response and also implicated in tumorigenesis (Karin et al., 2006). Inflammation, which is very intense in Mg deficient mice (Mazur et al., 2007), might impact both on the growth of primary tumours and on metastases. Indeed, inflammatory mediators induce genetic instability (Colotta et al., 2009) and might synergize with low Mg in causing mutations thus allowing the generation of highly aggressive cells. Tumour Necrosis Factor (TNF)α, the target of NFκB and prototypical pro-inflammatory cytokine, is increased in the serum of Mg deficient rodents (Mazur et al., 2007). In spite of its name, TNFα has several marked pro-tumoural functions, because it facilitates the epithelial to mesenchymal transition, it enhances tumour invasion and recruits leukocytes within the tumour, thus amplifying the inflammatory process (Drutskaya et al., 2010). Moreover, TNFα contributes to immune suppression (Drutskaya et al., 2010). Together with TNFα, interleukins (IL) 1 and 6, both induced under Mg deprivation, augment the capacity of cancer cells to metastatize (Royuela et al., 2008). TNFα and IL1s also upregulate endothelial adhesion molecules in lung capillaries, thus facilitating the tethering of metastatic cells to the vessel wall and their subsequent transmigration and colonization in the adjacent tissues. In addition, Mg is an absolute requirement for the function of the metastasis-suppressor gene product NM23-H1 (Ma et al., 2008). NM23-H1 knockout mice showed accelerated metastasis (Boissan et al., 2005). It is therefore possible that low Mg availability impairs NM23-H1 anti-metastatic activity in mice.
Briefly, in rodents, Mg deficiency seems to participate both in early (initiation) and in late (progression) phases of tumorigenesis. The following scenario is feasible. Low Mg promotes oxidative stress and inflammation which generate genetic instability and, therefore, increases the risk of mutations. Since Mg is an essential cofactor in almost all enzymatic systems involved in DNA synthesis and repair, under low Mg availability mutations may become permanent thus generating the so-called “initiated” cell. The persistence of oxidative stress and inflammation together might generate further mutations, which render the cell immortal and self sufficient in terms of proliferation. Some cells might also acquire an invasive, metastatic phenotype and colonize distant organs (Figure 1).
Even though studies on cultured cells can only focus on some isolated aspects of the complex events implicated in the development of tumours, they have produced an extraordinary body of knowledge. The field of Mg research in cancer makes no exception.
Based on the results obtained in humans and in animal models, the effects of low extracellular Mg on tumour cells has been investigated. While normal diploid cells are very sensitive to low extracellular Mg and are rapidly growth retarded, neoplastic cells are rather refractory to Mg deprivation (Wolf et al., 2009). Indeed, tumour cells require longer times and much lower concentrations of extracellular Mg than normal cells to slow their proliferation rate (Wolf, 2004; Maier et al., 2007). Another peculiarity of tumour cells is their avidity for Mg, which accumulates within the cells even when extracellular Mg concentrations are low. The molecular mechanisms involved have not been completely disclosed, but it is known that Na-dependent Mg extrusion is impaired in tumour cells and contributes to elevate intracellular Mg (Wolf et al., 2007). In addition, TRPM7 is overexpressed in breast tumours and in neoplastic cultured cells (Guilbert et al., 2009), and this feature may well account for the avidity of cancer cells for extracellular Mg. Also nuclear Mg content is higher in neoplastic than in normal cells (Lukacs et al., 1996). Considering that Mg acts a second messenger and activates a vast array of enzymes, it is noteworthy that not only does intracellular Mg increase, but it is also differently distributed in transformed vs normal cells.
A very simple question is then raised: does intracellular Mg impact on different aspects of the neoplastic phenotype? Before addressing this issue, it is useful to rapidly review the latest hypotheses proposed to explain the contribution of Mg to normal cell proliferation. Since Mg participates in all the major metabolic processes as well as to redox reactions, it is no surprise that it has a direct role in controlling cell survival and growth. Cell survival is strictly related to attachment to the extracellular matrix. Adhesion and consequent spreading are mediated by diverse molecules, most notably by integrins which require divalent cations such as Ca and Mg to function (Rouslathi and Pierschbacher, 1987). Once the cells adhere and spread, they are alive, metabolically active and ready to respond to growth factors. According to a recently proposed model (Rubin, 2007), intracellular Mg levels vary in quiescent vs growth factor-activated, proliferating cells. In quiescent cells, cytosolic Mg stabilizes the plasma membrane and subcellular organelles by complexing with negatively charged phospholipids (Suh and Halle, 2007), while nuclear Mg stabilizes the double helix and retains the chromatin structure (Figure 2). Binding of growth factors to their cognate receptors perturbs the membranes and, consequently, decreases Mg binding affinity for the membrane, thus resulting in increased focal concentrations of intracellular Mg (Rubin, 2007). Also Mg influx from the extracellular milieu contributes to increase total intracellular Mg in response to growth factors. In particular, EGF regulates Mg transepithelial transport by increasing TRPM6 activity and surface expression (Thebault et al., 2009). Intracellular Mg then behaves as a second messenger and regulates a wide variety of reactions ultimately leading to cell division.
Neoplastic transformation dramatically disrupts these controlled and coordinated fluctuations of intracellular Mg (Figure 2). Membrane perturb- ations that are typically associated with the neoplastic phenotype determine the release of intracellularly bound Mg. Accordingly, a seminal study showed a significant increase of intra- cellular Mg in human brain tumours by 31P-NMR (Taylor et al., 1991).
Having high intracellular Mg might represent a selective advantage for the transformed cells. Mg forms complexes with ATP, ADP and GTP, being necessary for the activity of enzymes implicated in the transfer of phosphate groups such as glucokinase, phosphofructokinase, phospho- glycerate-kinase and pyruvate kinase (Gunther, 2008), enzymes of glycolysis, which is known to be the preferential pathway utilized by neoplastic cells to produce energy despite the presence of oxygen. This issue has recently been reappraised. The expression of TRPM7 and, therefore, the capability of the cells to take up Mg has been associated with the transition from a quiescent to a proliferative metabolic state in which it is aerobic glycolysis that generates the energy necessary for cell growth (Sahni et al., 2010).
Mg also complexes with DNA polymerase, ribonucleases, adenylcyclase, phosphodiesterases, guanylate-cyclase, ATPases and GTPases, being therefore implicated in the metabolism of nucleic acids and proteins and in signal transduction. In addition, Mg activates telomerase (Lue, 1999), a ribonucleoprotein that is responsible for maintaining the terminal repeats of telomeres and conferring limitless replicative potential. It is because of the reactivation of telomerase activity that transformed cells become immortal (Hanahan and Weinberg, 2000).
Intriguingly, the nuclear Ser/Thr phosphatase PPM1D (also known as WIP1), which is over- expressed in human primary breast, gastric, pancreatic, ovarian and neuroblastoma tumours, requires Mg for its activity. PPM1D is involved in the regulation of several essential signalling pathways that are implicated in cancer pathogenesis (Le Guezennec and Bulavin, 2010). The oncogenic properties of PPM1D were originally thought to stem from its ability to dephosphorylate and, consequently, inactivate the p53 tumour suppressor gene. However, recent studies have shown that PPM1D also targets other key stress response kinases that function in DNA damage response and repair. In addition, PPM1D complements several oncogenes, such as Ras, Myc, and HER-2/neu, for cellular trans-formation both in vitro and in vivo. Therefore, a rise in intracellular Mg might result in a permanent activation of PPM1D and potently contribute to several steps of tumour development.
A last issue needs consideration. Tumours should be regarded as complex tissues in which normal cell types within the stroma serve as active collaborators in cancer progression. Many of the growth signals driving the proliferation and invasion of carcinoma cells originate from the stromal cell components of the tumour mass. Indeed, low Mg modulates the functions of normal cells present in the tumour micro- environment. In particular, endothelial cells cultured in low Mg release higher amounts of metalloproteases and growth factors (Ferrè et al., 2010). Similar results were obtained in cultured human fibroblasts (Maier et al., unpublished results). In addition, low Mg promotes endothelial and fibroblast senescence (Killilea and Maier, 2008) and senescent cells can modify the tissue environment in a way that synergizes with oncogenic mutations to promote the progression of cancers (Campisi, 2003). On the other hand, low Mg impairs the acquisition of an angiogenic phenotype in vitro. Yet, the question about the role of Mg deficiency in cancer remains: good or bad?
How do we reconcile this wealth of information? Although some results are controversial, it is possible to conclude that most of the data available point to low Mg as a contributing factor to tumorigenesis. Mg deficiency increases oxidative stress, which causes DNA damage, and impairs DNA repair mechanisms. It therefore may lead to mutations that produce oncogenes with dominant gain of function, and tumour suppressor genes with recessive loss of function (Hanahan and Weinberg, 2000). Upon transformation, the cells lose the capability to regulate Mg homeostasis. Mg also accumulates in the cell when Mg availability is low, its distribution is altered and is not subject to coordinated fluctuations in response to various stimuli. Low extracellular Mg also affects normal stromal cells that can release growth factors and proteases, thus potentiating tumour growth and development. In addition, in vivo low Mg activates an inflammatory response, which has many tumour-promoting effects.
The connection between Mg and cancer is rather strong, but several questions remain. First, it might be asked whether Mg deficiency is sufficient for the development of cancer. Indeed, low Mg increases the levels of reactive oxygen species, which mutate DNA and, therefore, generate genetic instability. Another candidate as a potential carcinogen in Mg deficiency is inflammation. However, it is more likely that a low Mg state only contributes to tumorigenesis by synergizing with many different factors.
A second aspect that is often underevaluated is the effect on the tumour of an aberrant Ca:Mg ratio that is inevitably created in Mg deficiency. Even though Ca and Mg have become the gold standard when discussing nutritional supplements, the balance between the two cations needs to be carefully evaluated both in vitro and in vivo.
A third question arises about whether outcomes in the cancer mouse model can predict a real human disease mechanism. Mg deficiency retards primary tumour growth but enhances metastases in mice. It would be relevant to consider the balance between cancer-promoting and inhibiting responses by low Mg in humans also.
The last big question is whether knowledge about the connection between low Mg and cancer can be translated into useful approaches to prevent and treat cancer. Most of the data available suggest that Mg could be a chemo-preventive agent. Therefore, correcting Mg intake might represent an effective and low cost preventive measure to reduce cancer risk. The picture is different in oncologic patients, because hypomagnesemia might be beneficial since it sensitizes neoplastic cells to radiation or chemotherapeutics. Therefore the “hamletic” doubt remains: to treat or not to treat hypo- magnesemic oncologic patients?
Future research should focus on elucidating how Mg is involved in preventing cancer by modulating crucial events in signal transduction pathways and the cell cycle. Hints might derive from nutrigenomics, i.e. the use of high- throughput genomic methods in nutritional studies, which can provide information about the cross-talk between Mg and genes. The next step should lead to translation of this basic research into nutritional preventive protocols for human tumours, an approach that will involve many scientific disciplines. The challenge is great, but the reward might be enormous.
This work is supported by MIUR-PRIN 2007 grant number 2007ZT39FN.
This book is copyright. Apart from any fair dealing for the purposes of private study, research, criticism or review as permitted under the Copyright Act, no part may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise without the prior written permission. Address all inquiries to the Director at the above address.
Your browsing activity is empty.
Activity recording is turned off.
See more...
