Abstract
In the face of escalating maternal and fetal health threats, hypertensive pregnancy disorders (HPDs) is one of the leading cause of maternal and fetal morbidity and mortality. The range of HPDs include white-coat hypertension, chronic hypertension, gestational hypertension, mild-to-moderate and severe preeclampsia and eclampsia. Current evidence implicates an imbalance of circulating anti- and angiogenic factors in HPDs emanating from the placental vasculature, impacting on angiogenesis. Delivery of the fetus is thus far the only curative measure, albeit with increased risk. Resultant endothelial dysfunction caused by the excessive production of placental soluble fms-like tyrosine kinase-1 has been the basis of many studies to find a safer treatment strategy. Metformin, used historically in the treatment of diabetes mellitus has also found its therapeutic reach in many other disease states. These include, but are not limited to, improving blood flow in certain cancer types, treatment of polycystic ovarian disease, improving vasodilation, and reducing inflammation. Metformin is used to treat hyperglycemic endothelial dysfunction through the enhancement of the nitric oxide system, endothelin-derived hyperpolarizing factor and sirtuin 1. Similarly, endothelial dysfunction in preeclampsia and other HPDs leads to a hypoxic state and elevated blood pressures. Dubbed as the new “aspirin” of current times, the retardation of the antiangiogenic status by metformin provides an exciting and promising alternate strategy in treating these pregnancy disorders.
Introduction
Many women worldwide will regrettably not experience the joys of normal pregnancy and childbirth. Of the numerous underlying medical conditions that detrimentally affect maternal and fetal mortality and morbidity, an elevation in blood pressure has to be one of the most severe. Hypertensive disorders complicate approximately 8% of pregnancies worldwide and also increase the risk of cardiovascular disease (CVD) in mothers in the future. 1,2 There is a dire need for standardizing the management and treatment of these hypertensive emergencies both pre- and postpartum (Table 1). Up until 2010, collaborative efforts have reduced global maternal deaths down to 210 per 100 000 live births. 3 The current drive to decrease these rates by a further 75% will entail increased resources and scientific efforts. Locally, in South Africa, 18.8% of avoidable deaths are attributed to hypertensive pregnancy disorders (HPDs). 3 Current treatment for hypertension often involves angiotensin-converting enzyme inhibitors and corresponding receptor blockers; however, these are contraindicated during pregnancy due to the developmental risk to the fetus. 39 A more reliable, cost-effective, and pregnancy safe drug, metformin, is highly recommended as future treatment. 7
Characterization and Treatment of Hypertensive Pregnancy Disorders.a
aCurrent management is governed by the woman’s gestational age, blood flow to the placenta, and blood pressure.
bIn cases of asthma, heart disease, and congestive heart failure, it is recommended that parenteral labetalol should be circumvented.
A hypertensive emergency is defined as acute-onset and/or severe hypertension which perseveres for 15 or longer minutes, using standard accurate measurement techniques. 1 Hypertensive disorders in pregnancy consist of the following range, namely, white-coat hypertension; chronic hypertension; gestational hypertension; mild-to-moderate and severe preeclampsia (PE); eclampsia; and hemolysis, elevated liver enzymes, and low platelets (HELLP) syndrome. These disorders are the commonest direct cause of maternal deaths in South Africa reported in 2017. 1,4 Preeclampsia is unique to human pregnancies and because the exact etiology is unknown, prediction and treatment poses a clinical dilemma. Thus far, the only curative intervention is management as well as early delivery of the fetus and placenta. This leads to premature delivery, high caesarean delivery rates, and significantly high maternal morbidity and mortality rates. Despite the uncertainty around the etiopathogenesis of PE, extensive endothelial dysfunction and imbalance of circulating anti/angiogenic growth factors is a hallmark of the pathology, 5,6 attributed to a hypoxic placenta. This placental imbalance decreases proangiogenic factors and exacerbates the action of antiangiogenic factors to inhibit angiogenesis during placental endothelial hypoxia. 6,7 Hypoxic inducible factor 1α (HIF1α) is responsible for oxygen delivery by controlling vascular remodeling and angiogenesis. It is believed the upregulation of HIF1α during placental ischemia/hypoxia and facilitates the secretion of the antiangiogenic factors into the circulation. Interestingly, it has been reported that blocking HIF1α activity may decrease soluble fms-like tyrosine kinase-1 (sFlt-1) secretion using the well-known antidiabetic drug, metformin. 7 This review will focus on the subcategories of PE, the implication of angiogenic factors, and the impact of metformin on the possible amelioration of these HPDs.
Hypertensive Pregnancy Disorders
Chronic Hypertension in Pregnancy
The prevalence of chronic hypertension in pregnancy is subjective to race, body mass index (BMI), and age, with the majority of cases stemming from essential hypertension or to a lesser extent, underlying kidney disease. 9 Chronic hypertension is based on a documented history of high blood pressure prior to pregnancy or persistent increase in blood pressure (>140/90 mm Hg) measured twice in 24-hour intervals earlier than 20-week gestation. 2,10 It is also hypertension that persists for more than 12 weeks after delivery. It can further be broken down into mild and severe hypertension. Chronic hypertension can additionally increase the risk of developing superimposed PE by up to 20% to 25%. 10,11 Devastating outcomes include perinatal and/or maternal mortality, small for gestational age and premature infants, placental abruption, intrauterine death, and caesarian delivery. 8 These can be significantly decreased with appropriate obstetric and neonatal management and the use of antihypertensive drugs prior to pregnancy. 8
Gestational Hypertension
Previously termed as pregnancy-induced hypertension, it affects approximately 6% to 10% of pregnancies worldwide and, like PE of unknown etiopathology, is a major cause of maternal, fetal, and neonatal morbidity. It is defined as sudden transient hypertension after midgestation (20 weeks) with no proteinuria, in a previously normotensive woman. 2,11 Blood pressure monitored over a 24-hour period is the key diagnosis factor in distinguishing between gestational hypertension and PE. 12 It can be further classified into mild, moderate, and severe with the systolic reaching 160 mm Hg and/or diastolic at least 110 mm Hg. Treatment is determined by blood pressure levels, gestational age, associated risk factors, and symptoms. Women with a high risk of gestational hypertension are recommended to increase their calcium intake as a precautionary treatment. 10 It is only in hindsight, with the absence of developing PE that gestational hypertension is confirmed. Blood pressure also has to return to normal by the 12th week after delivery. Approximately, one-third of women with gestational hypertension between the 24th and 35th week are prone to develop PE. 8
Eclampsia
Eclampsia is a result of PE in pregnant or postpartum females, presenting with the onset of generalized tonic–clonic seizures/convulsions. The main distress is the quickness of the onset of seizures. 10 These seizures can occur antepartum, intrapartum, or postpartum. It may also very seldom present unpredictably in cases with slight elevation of blood pressure and no proteinuria. 8 This unique disease state (often confused with epilepsy) is, unfortunately, to date, unpreventable, or incurable despite exposure to the best medical care. About 60% of all eclamptic deaths are attributed to pulmonary edema, hemorrhagic stroke, and ischemia. 8,11 –13 Developing countries are a major contributor to this cause of maternal and/or perinatal deaths, due to the poor antenatal care and a lack of early prevention and treatment. 13 It is probable that endothelial damage linked with PE predisposes to eclampsia even at normalized blood pressures. 10 As metformin is found to reduce endothelial dysfunction, this could be a promising intervention. Currently, the only intervention is the administration of magnesium sulphate (anticonvulsant drug; Table 1) in the control and prevention of seizures and delivery of the fetus. Reservations of whether administration of magnesium sulphate in mild or severe PE would aid in reducing or ameliorating the risk of eclampsia have led to further investigations. 13
Preeclampsia
Preeclampsia is defined as a diverse maternal disorder characterized by new onset hypertension (≥140/90 mm Hg) and proteinuria (≥300 mg of protein in 24-hour urine collection or ≥1+ protein on a urine dipstick) after 20 weeks of gestation. 6 Further complications include inflammation, edema, and renal complications. It is important to note edema is no longer considered for the criteria in diagnosing PE. 10 In South Africa, however, PE affects 5% to 10% of pregnancies. 14 Remarkable similarities between intrauterine growth restriction (IUGR) and PE include common physiological and biological pathways as well as abnormal Doppler measurements during the first and second trimester of pregnancies. 15 Furthermore, an intense focus has been bestowed upon the imbalance of angiogenic, vascular endothelial growth factor (VEGF) and placental growth factor (PlGF), and antiangiogenic sFlt-1 in the pathophysiology of this disease. Further studies warrant the investigation of this imbalance in HPDs mentioned above. Together, these 2 antiangiogenic factors sFlt-1 and soluble endoglin (sENG) create an environment leading to IUGR, hypertension, HELLP, and multiorgan involvement. The HELLP syndrome is a variant of PE affecting 2% to 4% of pregnancies. 10,16
Primiparous women are at least 2 to 3 times more likely to be affected than multiparous women, 16 whereas nulliparous women are 70% to 75% more likely to develop PE and possibly will persist in a consequent pregnancy. 2 Other risk factors include a family history of PE, maternal age, BMI, and race. Medical ailments that enhance this risk are chronic renal failure, immunology, thrombophilia, pregestational diabetes/gestational diabetes mellitus (GDM), obesity, and factors of the metabolic syndrome. The threat of PE increases almost 4-fold in the presence of diabetes. 17,18,42 Clinical trials have also found an increased incidence rate of PE directly correlates to an increased pregnancy BMI and elevated gestational weight gain. 41
Pro- and Antiangiogenic Factors in Normal Pregnancy
Physiology of Normal Pregnancy: Vascularization and Angiogenesis
In the first trimester of pregnancy, the development of the placenta occurs under hypoxic conditions, which is necessary for embryonic and placental development. As gestation progresses, extravillous trophoblasts penetrate the myometrium, converting spiral arteries to wider bore vessels to provide sufficient nutrients and O2 to the developing fetus. 19 This is accomplished by specialized extravillous trophoblast subpopulations that interact with immune cells to cause changes in systemic hemodynamics, vascular remodeling, and anchorage of the placenta to the decidua. Vasculogenesis, angiogenesis, and pseudo-vasculogenesis involve lengthening as well as coiling of the spiral arteries in the placenta. 20 Initially, vasculogenesis causes differentiation of vessels, creating new blood vessels. On gestational day 21, angiogenesis forms new vessels from preexisting vessels. Angiogenesis is strongly controlled by a range of pro- and antiangiogenic factors. Pro-angiogenic factors (VEGF and PlGF) control structure, function, and growth of placental vasculature having a direct impact on fetal outcomes. The hypoxic environment in early gestation is essential for triggering the response cascade of VEGF. Part of the VEGF family, PlGF, augments the angiogenic action of VEGF by binding to their respective receptor sites. 21
Pathophysiology of HPDs and the Role of Angiogenic and Antiangiogenic Factors
The extensive research attempts to ascertain the exact etiology of all HPDs have fallen short, leading to a more well-described pathophysiology. Despite being true to its name as a “disease of theories,” plausible root causes include immunological, endothelial dysfunction 5,11 and most importantly the antiangiogenic state. Most of the theories surrounding this disease involve the placenta, more specifically its underperfusion and ischemia. 22,30 This hypoperfusion results in the excessive release of sFlt-1 and sENG, which are implicated in poor vascularization, endothelial dysfunction, hypertension, and the pathogenesis of PE since 2003. 11,23,30 Preeclampsia is thought to be 2-stage disorder, the first involves defective placental trophoblastic invasion (14-18 weeks of gestation) of the uterine spiral artery leading to decreased uteroplacental blood flow. 24 This decreased blood flow and decreased oxygenation leads to the release of apoptotic cells, syncytiotrophoblast microparticles debris and an imbalance of angiogenic and antiangiogenic factors. This imbalance triggers abnormal placental vascularization, inflammatory response, and widespread endothelial damage which affects all organ systems, progressing to stage 2. This multisystem disorder then gives rise to the symptoms (elevated blood pressure; increased proteinuria, low platelet count, and elevated liver function tests) seen clinically (Figure 1). 25
A brief overview showing the etiology and the 2-stage pathophysiology of preeclampsia (PE). Risk factors that predispose the first stage (poor placentation) of PE include biological, hereditary, and maternal influences. The poor transformation of spiral arteries decreases both blood flow and oxygen hence releasing trophoblastic debris into circulation. Consequently, stage 2 (maternal system dysfunction) gives rise to an imbalance of anti/angiogenic factors causing endothelial damage, presenting with elevated blood pressure and proteinuria.
Soluble fms-like tyrosine kinase-1, a spliced variant of VEGF receptor-1, is secreted primarily by syncytiotrophoblasts into circulation and its concentration is directly proportional to the severity of PE. 23 The rise in these levels, clinically, presents as hypertension and end-organ damage and are detectable up to 5 weeks prior to the onset of symptoms. The tandem levels of increased sFlt-1 and reduced free PlGF imply sFlt-1 completely binds to all PlGF sites and remains in excess. The lowered PlGF levels can be seen at about 9 to 11 weeks, before preeclamptic symptoms present. 5,23 Soluble endoglin, like sFlt-1, is expressed in the syncytiotrophoblast and involved in secondary symptoms such as seizures, liver dysfunction, and hypercoagulation. Antiangiogenic sENG specifically inhibits nitric oxide (NO)-mediated vasodilation and vascular permeability. 5 It contributes to PE by attaching to TGF-β1 cell surface receptors and decreasing endothelial nitric oxide synthase (eNOS) signalling. 23
Although these antiangiogenic factors implication in PE is well known, their regulation and production still needs elucidation. 26 Placental ischemia results from decreased PlGF and increased sFlt-1, hence causing partial remodeling of the spiral artery. The elevated maternal and/or placental sFlt-1:PlGF ratio has also been associated with PE. This ratio is being used clinically as a diagnostic tool for PE despite yielding unclear results. 27 Soluble fms-like tyrosine kinase-1 also causes vasoconstriction and endothelial dysfunction which results in fetal growth restriction and PE. Since these levels are more prominent in the second and third trimester, the other anti/angiogenic factors and ratios are better used as an investigative tool. 5,28 Quantification of these anti- and angiogenic factors serves as a potentially powerful tool to differentiate PE from other HPDs and possibly elucidate their indefinable pathophysiology. 11,23 Unfortunately, the origins of these antiangiogenic factors that cause placental damage remains elusive. Further investigation in human studies is warranted.
Hyperglycemia and Endothelial Dysfunction
In the vascular system, endothelial progenitor cells are responsible for angiogenesis and vascular protection. Additionally, in a hyperglycemic state, these progenitor cells are reduced and hence their function impaired, leading to vascular complications and endothelial dysfunction. 29 Okonkwo and DiPietro (2017) reviewed and stated that the prolonged endothelial cell exposure to elevated glucose levels decreases their integrity causing them to detach and flow into circulation. It was also found that the angiogenic response to diabetes is tissue and/or organ dependent. 30 Due to the glycolytic nature of endothelial cells, hyperglycemia increases reactive oxygen species levels, initiating the development of endothelial dysfunction. 31 Another important protein involved in vessel formation and cell proliferation is perlecan. This proteoglycan is situated in the basement membrane and binds to VEGF to present them to corresponding receptors at the cell surface. Perlecan levels were found to be significantly elevated in the third trimester of gestational diabetes. 32
Novel (non-hypoglycemic) Effects of Metformin in Pregnancy, Reproductive, and Other Disorders
Metformin, historically known as a hypoglycemic pharmaceutical drug in the treatment of type 2 diabetes as well as in patients diagnosed with gestational diabetes, has opened new opportunities in the treatment of other disorders. This includes treating polycystic ovarian disease; healing omental blood vessels, congestive heart failure, chronic kidney, and liver disease; and reducing sFlt-1 and sENG levels. 32 –34 As reported by Jamal et al, the sustained use of metformin throughout pregnancy (in females with polycystic ovary syndrome) reduces obstetric complications with no evidential risk of major malformations. Efforts are underway for its coadministration with insulin to treat type 1 diabetes. 24 This orally administered tablet improves insulin sensitivity in the liver by decreasing hepatic glucose production, while simultaneously enhancing insulin-stimulated uptake. Interestingly, it also protects against diabetes-induced vascular disease, 34 which has opened a plethora of advances in medicine. Other novel uses of this drug include the reduction of prostate and breast metastasis; enhance vasoprotective properties, improve inflammation and inducing vasodilation of diabetic rat vessels by altering angiogenesis. 7,19
Various studies have reported on the safety profile of metformin in pregnancy, due to its ability to cross the placental barrier, which raises questions about its safety because it can cross the placental barrier. Metformin was also found to decrease uterine artery impedance from 12 to 19 weeks of gestation in polycystic ovarian syndrome (PCOS)-complicated pregnancies; however, metformin coupled with aspirin improves placental circulation and reduces the risk of obtaining PE. 24 Shishavan et al quoted, “studies done in an early 90s study found a lowering effect of blood pressure using Metformin” conversely, later on, others have shown a lack of such effects. 36 Han et al vaguely alluded to prospective applications of metformin in the first trimester of pregnancy in relation to PE, such as improving inflammation, altering abnormal placentation, and altering angiogenesis in diabetic states. 19,24 However, very few studies have been done on the long-term effect of metformin on GDM and pregnancy outcomes. 19 Importantly, it has been reported that treatment with metformin in pregnant women with PCOS has yielded negligible fetal abnormalities and birth weight aberrations. 35 Metformin has proven to lower arterial blood pressure in diabetics clinically and in animal models, as well as a minor hypotensive effect in hypertensive human males without diabetes. 40 Also, numerous animal studies have shown the positive effects of metformin treatment on placental dysfunction and insulin demand in pregnant rats with a high fructose-fed diet. 35,36 However, no data exist on the risk of diabetes developing in mother or fetus or its effects in GDM after metformin treatment. 7 In a clinical randomized double-blinded study using metformin and a placebo, on pregnant women, secondary findings found a significant decrease in the incidence rate of PE. These nondiabetic pregnant females were subjected to metformin between weeks 12 and 18 of gestation until term. 41 The first ever systematic review of randomized controlled trials also discovered a lesser chance of getting PE through metformin, either in or out of combination, with insulin treatment. This review also, contrariwise, found no difference in the prevalence of this disease between metformin and a placebo. The above cohort of GDM or type 2 diabetic women taking metformin reduced gestational weight gain which is assumed to be the basis of this reduced risk. 42
Commentary on Proposed Mechanisms of Action of Metformin for the Treatment and Efficacy in HPDs
The link between the angiogenic pathway and insulin resistance
Preeclampsia and CVDs share several risk factors including insulin resistance, obesity, diabetes, and inflammation. 44 Insulin resistance appears to be a potential mechanism linking PE and CVD. Although the molecular pathways by which insulin resistance might link PE with CVD are vague, oxidative stress and inflammation, which are features of both diseases, may be involved. The previous report shows that the vascular superoxide producing nicotinamide adenine dinucleotide phosphate reduced oxidase, an enzyme that can be activated by hyperinsulinemia or associated excess free fatty acids, may be particularly important in oxidative stress and inflammation. 45,47 Miele et al reported that significant linear correlation between homeostasis model of insulin resistance and VEGF. Insulin induces the expression of VEGF mRNA 46 and reduced expression of VEGF mRNA in insulin-resistant states can be reversed with insulin. Hence, it is tempting to speculate that increased insulin resistance at baseline along with alterations in angiogenesis might act synergistically to predispose certain women to PE. 47
Indeed, the underlying pathology of this multifaceted disease is altered angiogenesis with insulin resistance being a strong risk factor. Receptors, on endothelial cells, specific for insulin cause vasorelaxation and increased uptake of L-arginine in the presence of insulin. Importantly, insulin improves VEGF expression enhancing angiogenesis. 42,43 The mutualistic cause of insulin resistance and impaired angiogenesis is attributed to inflammation and oxidative stress. These factors result in microvascular aberrations. In this light, insulin resistance or a lack thereof will cause vasoconstriction and less uptake of amino acids vitals for blood vessel dilation. Furthermore, the lack of endothelial growth factor in the absence of insulin paired with vasoconstriction is the backbone of PE and other HPDs.
Metformin affecting the NO pathway
Diabetes more often than not coexists with hypertension. Presently, the rationale behind the development to gestational hypertension as opposed to PE is blurred; however, both conditions are likely to have overlapping processes. 30 Under normal pregnancy conditions, eNOS is also expressed on villous endothelial cells. Nitric oxide produced by these cells is understood to be an important endogenous vasodilator within the placental vasculature. On the contrary, a decrease in the bioavailability of NO is associated with the progression of vascular disease. 31 It has been reported that the selected benefits of metformin may be mediated via the modulation of sirtuin 1 (SIRT1). Sirtuin 1 proliferates NO bioavailability, by eNOS deacetylation, and is involved in cell proliferation and angiogenesis. The SIRT1-mediated endothelial cell protection via enhanced eNOS activity may be an important target for metformin. This results in a decline in apoptosis and the promotion of angiogenesis. 34
Metformin, being from the biguanidine family, is believed to undergo oxidation of guanidine hence forming NO in a living organism. This NO action caused by the drug has been evident in relieving blood pressure in type 2 diabetic coupled hypertensive patients. 40 Once metformin is introduced into a hypoxic environment, it is assumed to increase dilation by exacerbating the release of NO and endothelin-derived hyperpolarizing factor (EDHF). Sustained metformin treatment improves EDHF-mediated relaxation. These claims were seen in an animal study, where diabetes-induced, spontaneously hypertensive rats treated with metformin showed a reduction in systolic blood pressure. 36 A study by Alzamendi et al investigating the effect of L-arginine infusion, the substrate for eNOS, and the generation of NO, in diabetics, revealed improved hemodynamic effects subsequent to treatment with metformin. Further experiments on human hepatocytes cell culture, treated with metformin, revealed very elevated levels of nitrotyrosine. Nitrotyrosine is a strong indicator of NO presence, inflammation, and cell damage. 40 As per studies above, metformin had very strong indirect correlations of releasing NO; therefore, it is vital to measure levels in humans.
Metformin affecting endothelial cells
In a recent review, it is suggested that metformin moderates the rate of gestational hypertension via its influence on endothelial function and its ability to decrease reactive oxygen species. 32 Metformin has, in many studies, been found to increase the number of endothelial progenitor cells hence reducing endothelial dysfunction. Unfortunately, its role in the progenitor cell function in the diabetic state still remains unresolved. 29,36 Surprisingly, metformin has pleiotropic effects on endothelial function, releasing both potent vasoconstrictors and vasodilators. Despite causing the release of an endothelial contracting factor (prostaglandin), it was hypothesized that positive effects on endothelial function were as a result of the compensatory over secretion endothelial vasodilator mediators. 36 The expression of vascular cell adhesion molecule 1, released by endothelial cells, exacerbating the preeclamptic syndrome is found to be decreased by metformin. 32 The preventative role of this drug extends its reach from impacting the antiangiogenic state to cell metabolism and other possible pathophysiological pathways. 7,32 It is based on these facts that we propose that metformin will be effective in treating PE. Using in vitro studies, Brownfoot et al showed the potent protective efficacy of Metformin in PE. Mechanistic through reduced endothelial dysfunction, mediated by the inhibitory effect of sFlt-1 and sENG via the reduction of the mitochondrial electron transport chain complex-1 (Figure 2). 7
Putative mechanism action of metformin in preeclampsia (PE). The interface between fetal placenta and the maternal uterus. In placental development, fetal cytotrophoblasts change from an epithelial phenotype to an endothelial phenotype as they invade maternal spiral arteries causing an increase in the vessels’ capacity which leads to healthy placental perfusion. In PE, the cytotrophoblasts fail to change phenotype. The result is a shallow invasion of the spiral arteries and, therefore, incomplete placental perfusion and represents the action metformin which reduces key features of endothelial dysfunction that are specific to preeclampsia. Metformin enhances angiogenesis through reduced soluble fms-like tyrosine kinase-1 (sFlt-1) and soluble endoglin (sENG) secretion by inhibiting complex 1 of the mitochondria.
Conclusion
As supported by the studies cited above, a comparable state of endothelial dysfunction, elevated oxidative stress, and altered angiogenesis is displayed in hyperglycemia, PE, and other HPDs. The evidence above also indicates that metformin reduces these complications alongside blood pressure. Therefore, utilizing the mechanism of metformin in the treatment of hyperglycemic-induced endothelial dysfunction, where SIRT-1 enhances NO production, this could possibly relieve hypertensive pregnancy–associated ailments. Metformin can be potentially classified as a “wonder drug” of current times. Its antiangiogenic hindrance gives thrilling and promising insight into the therapeutic effect on HPDs. Due to the affordable nature of this drug, further human studies are warranted to test its efficacy.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The research was supported by the National Research Foundation (NRF), South Africa (under award number #91544) to the principal investigator (I. Mackraj).
