The reduction in blood glucose by different doses of Artesunate and other artemisinin derivatives like artemether has been reported by other researchers for a long time (21, 30–32). In our recent study, we also reported that artesunate produced a blood glucose-lowering effect in female rats at both 2.9 mg/kg and 8.7 mg/kg, but differential effects in the male rats - reduction with 2.9 mg/kg treatment for 5 days but increase at both 5 days and 15 days’ treatment of 8.7 mg/kg (24). Using the same doses as in our previous study (24), we have consistently shown in this study that artesunate’s effect on the blood glucose is gender-, dose- and duration-dependent. In male rats for instance, we again observed a reduction in blood glucose by low (2.9 mg/kg) and high (8.7 mg/kg) doses of artesunate at 5 days but increase by high dose and no change by low dose of artesunate at 15 days. Our present data on female rats similarly show that blood glucose was reduced by low dose artesunate at 15 days while other groups were not affected. However, it is noteworthy that the low dose of artesunate caused hypoglycaemia on the 5th day of administration while hypoglycaemia occurred on the 15th day of administration in female rats treated with the same dose. This suggests that the onset of artesunate-induced hypoglycaemia is much earlier in male than female rats as opposed to the early occurrence of artesunate-induced haemolysis in female rats earlier reported (24). Is the artesunate-induced reduction in blood glucose related to increase in insulin level?
The mechanism of hypoglycaemia induced by many antimalarial drugs have been suggested to be related to increase in insulin secretion. For instance, cinchona alkaloids (quinine and quinidine) increases insulin secretion by blocking ATP-sensitive potassium (K+ ATP) channels in the pancreatic beta cells (30, 32). Similarly, chloroquine and halofantrine increase plasma insulin and glucose uptake, leading to hypoglycaemia (21, 33). Loss of functional β-cells, as evident from the absence of serum insulin C-peptide, is a major problem of Type-1 diabetic patients. Regeneration of patient-specific insulin-producing cells have been attempted using different cell sources including hepatic cells, embryonic stem cells, exocrine cells, induced pluripotent stem cells and endocrine cells (34–36). α-cells, which are developmentally closely related to β-cells, have the ability to replenish insulin-producing cells after extreme β-cell loss (37). During development (38) and when genetically triggered in adulthood (39), overexpression of the transcription factor Pax4 converts α-cells to β-cells. Moreover, the β-cell factor Pax4 represses the α-cell master regulatory transcription factor Arx (40), the loss of which converts α-cells into β-cells (41).
Li et al. (23) identified artemisinins as approved drugs that can confer β-cell characteristics to α-cells. Artemether and its active metabolite, dihydroartemisinin, fully inhibit the Arx overexpression phenotype in Min6 cells while also inducing insulin expression in αTC1 cells. The antagonism of Arx (by inducing translocation from the nucleus to the cytoplasm and thus depleting it from the chromatin) also causes the reduction of glucagon protein levels by more than 50% (23). The authors also convincingly showed that artemisinins induce insulin expression in the α-cells by targeting gephryin, which increases GABA signaling that will eventually suppress glucagon secretion in the α-cell. In the present study, we observed that artesunate increased insulin concentration in male rats that received low dose artesunate for 5 days, but not in other groups. Moreover, artesunate reduced glucagon concentrations of all male rats’ groups and also in female rats that received low and high doses of artesunate for 15 days. Our present study agrees with previously cited ones, and suggests that the artesunate-induced reduction in blood glucose is mediated by increase in insulin and decrease in glucagon concentrations possibly by modulating regulating proteins like Pax4 and Arx even though our present study is limited for not quantifying them.
Artesunate accumulates in the red blood cells (42) where the endoperoxidase bridge of the former is split by iron haeme present in the latter, leading to the release of ROS (43). Studies have shown that artesunate generates free radicals, increases malondialdehyde and decreases antioxidant activities like the superoxide dismutase and catalase (24, 44). Though artesunate’s moderate pro-oxidant action beneficially serves as a mechanism for its antimalarial effect by alkylating the malaria parasite’s membrane (10), its excessive pro-oxidant action can endanger erythrocytes that host the parasite (45) and cause haemolysis (44, 46). Notwithstanding its pro-oxidant effects, artesunate has been widely reported to enhance the production and release of glucose-6-phosphate dehydrogenase (G-6-PD), a widely known second line anti-oxidant enzyme (24). The G-6-PD is an important house-keeping enzyme that catalyses the first step of the pentose phosphate pathway, which is the sole source for the production of reducing capacity in the form of NADPH in the erythrocytes as erythrocytes lack the nucleus, the ribosomes, and the mitochondria (at maturity) that generate NADPH as in other cells (47, 48). The NADPH in turn activates glutathione reductase and catalase, which are part of the first line anti-oxidant system abundantly present in the red blood cells. The G-6-PD is the main cytoplasmic source of NADPH that prevents oxidative stress and haemolytic anaemia of the red blood cells and its deficiency leads to drug-dependent (e.g. antimalarial drug) and –independent haemolytic anaemia (45). In fact, the overexpression of G-6-PD decreases endothelial cell oxidative stress (49) and the risk of diabetes (50, 51), β-cell apoptosis and insulin resistance (52) are increased with deficiency or decrease of G6PD. The level of G-6-PD has been shown to increase in artesunate-treated erythrocytes of rats (22) and humans (53).
In our previous study, we observed that artesunate-induced increase in G-6-PD prevents haemolysis in male rats but not in female rats. We hypothesised that it is due to lower artesunate-induced lipid peroxidation in male rats compared to the female rats, which made G-6-PD’s scavenging of free radicals to sufficiently prevent haemolysis in male rats compared to the female rats with higher oxidative stress (24). In the present study, we observed that G6P was increased only in the male rats that received low dose artesunate for 15 days, while there was no change in any of the female rats. The G6PD, a rate-limiting enzyme in the pentose phosphate pathway, acts on glucose-6-phosphate to produce 6-phosphoglucono-δ-lactone via a dehydrogenation process, during which NADPH is formed from NADP+. That artesunate increased glucose-6-phosphate in male rats, but not in female rats, in this study suggests that artesunate increases the substrate for G6PD that leads to accumulation of NADPH which is needed to activate other anti-oxidant enzymes (e.g. catalase and glutathione reductase) to promote free radical scavenging in the red blood cell. This further explains our previous report that artesunate prevents haemolysis in male rats but not in female rats (24).
After carbohydrate-containing meal, high postprandial insulin level stimulates glycogen synthesis by enhancing glucose entry to the liver from the blood, conversion of glucose to glucose-phosphate (G6P) with the enzyme glucokinase, and the addition of G6P molecules to the ends of the chains of glycogen. After the meal digestion and during fasting, the fall in the insulin level will stimulate glycogenolysis and gluconeogenesis that will maintain release of glucose into the blood for use by the body cells, and G6Pase is a very important enzyme in both of these metabolic processes that are paramount during fasting. In glycogenolysis, various enzyme systems remove glucose molecules from glycogen stands in the form of G6P, which remains in the cell unless it is dephosphorylated (cleaved) by the enzyme G6Pase to produce free glucose and free phosphate anion. Thus, the free glucose can be transported out of the liver to cells into the blood to maintain adequate glucose supply to many body cells. Continuous fasting-induced reduction in insulin level leads to gluconeogenesis where muscle proteins and adipose tissue triglycerides are catabolised into amino acids, free fatty acids and lactic acids. More importantly, the amino acids and lactic acid are used to form new G6P in the liver cells through the process of gluconeogenesis. Interestingly, the last step of gluconeogenesis, like that of glycogenolysis, involves the dephosphorylation of G6P by G6Pase to form free glucose and phosphate. Is artesunate-induced reduction in blood glucose related to its alternation in this key enzyme?
The glycogen storage disease type 1 (GSD 1) is an inherited disease that leads to the inability of the liver to sufficiently break down its stored glycogen and thus disrupt glucose homeostasis. It is divided into two main types (GSD 1a and GSD 1b), which differ in cause, presentation and treatment. The GSD 1a is caused by a deficiency in the enzyme G6Pase while GSD 1b is caused by a deficiency in the enzyme glucose-6-phosphate translocase (G-6-P-T) that transports G6P from the cytoplasm to the microsomes (54). Since glycogenolysis is a major metabolic pathway by which the liver supplies glucose to the body during fasting, deficiency of both enzymes can cause severe low blood glucose and a corresponding excess glycogen storage in the liver. In our present study, we observed that the G6Pase concentration was reduced in male rats that received low and high doses of artesunate for 5 days, and in those that received high dose of artesunate for 15 days. We also observed that the G6Pase concentration was reduced in female rats that received low and high doses of artesunate for 5 days, but increased and unchanged in those that received low dose and high dose artesunate respectively for 15 days. Our present data also showed that glycogen was increased by both doses of artesunate at 5 days but unchanged by both doses at 15 days in both male and female rats. Taken together with our observation of reduction in glucagon level, the artesunate-induced reduction in G6Pase and increase in glycogen indicate that glycogenolysis is inhibited by artesunate, which led to glycogen accumulation in the liver especially during the 5 days’ duration. What led to the return of glycogen to normal level at 15 days in both male and female rats is not yet understood and needs further investigation.
The G6P has been reported to modulate the activity of 11β-hydroxysteroid dehydrogenase type 1 (11βHSD1). In GSD 1a for instance, G6P excess has been documented in the endoplasmic reticulum, which has been associated with an increase in the activity of 11βHSD1 (55). The 11βHSD1, an endoplasmic reticulum-bound enzyme, is typically expressed in glucocorticoid receptor-rich tissues (e.g. liver, brain, lung, and adipose tissue) and is responsible for the conversion of inactive cortisone to an active cortisol (56). The 11βHSD1 plays a key role in the development of metabolic syndrome and Cushing’s syndrome (57, 58) and 11βHSD1 knockout mice are resistant to the development of metabolic syndrome (59). In fact, a decrease in the hypothalamic-pituitary-adrenal negative feedback response has been reported in the G6PD knockout mice (60) and GSD 1a patients have reportedly shown high cortisol level (61). Interestingly, 11βHSD1 needs NADPH (generated from the G6PD-mediated conversion of G6P to 6-phosphogluconactone) as a cofactor (62). Though our present study did not determine the activity of 11βHSD1, we strongly speculate that the artesunate-induced increase in cortisol observed in all the male rats and in female rats that received low dose artesunate for 5 days in this study might be associated with the already-established artesunate-induced increases in G6P, G6PD and NADPH, all of which play parts in the activation of 11βHSD1. This is consistent with a previous report that accumulation of G6P in the endoplasmic reticulum fuels the G6PT-G6PD-11βHSD1 system that eventually leads to increased activation of glucocorticoids (63). Even though the level of cortisol was increased by artesunate, it could not elicit glycogenolysis or gluconeogenesis because of artesunate-induced reduction in G6Pase.
We investigated whether some of the gender differences associated with artesunate effects reported in this study are associated with sex hormones (testosterone and oestrogen) or not. We observed that oestrogen concentration was not affected in male and female rats that received both doses of artesunate, except in male rats that received low dose for 5 days where it was increased and in female rats that received high dose for 15 days where it was reduced. We also observed that except in male rats that received high dose of artesunate for 15 days, artesunate did not affect testosterone concentration in all other male and female groups. Our no-effect observation is consistent with the previous report of Samuel et al. (2018) that also observed that treatment with artesunate-amodiaquine and artemether-lumenfantrine for 3 and 6 days caused no significant effect on testosterone. The authors further showed that artesunate had no effect on the luteinising hormone, follicle stimulating hormone, and even the weights of the testis, epididymis, prostate, and seminal vesicule. It is worthy of note that Samuel et al. (2018) also used 2.86 and 8.58 mg/kg of artesunate-amodiaquine for 3 days and 6 days, which are similar to our own doses, while their experimental durations are also within ours.
However, while it appears that the sex hormones played no role in the effects of artesunate generally, the sharp rise in the plasma level of estrogen in a similar manner with insulin in the male rats treated with the low dose of artesunate for 5 days suggests that estrogen might have contributed to the hyperinsulinaemia and hypoglycaemia observed in those rats. A previous study has reported a decrease in fasting levels of glucose upon the administration of oral oestrogen replacement in postmenopausal women (65). In animal studies, the main steroids of the ovary, the oestrogens and the progestins, have been shown to provide a protective influence to the susceptibility to experimental diabetes (66, 67). Furthermore, an increase in basal glycaemia and an impaired glucose tolerance have been observed in ovariectomised mice as well as rats; while steroid replacement experiments indicated that a deficiency of estrogens is mainly responsible for the deterioration of glucose tolerance (68, 69). Transdermal oestradiol replacement therapy in estrogen-deficient postmenopausal women was shown to improve beta-cell function in vivo and to augment insulin secretion in response to an acute glucose challenge (70, 71). This effect was proposed to involve a tropic action of oestradiol on pancreatic islets in combination with an increase in glucose transport in the muscle and inhibition of gluconeogenesis (71). Indeed, the islets of Langerhans have been demonstrated to express estrogen receptors (El Seifi et al., 1981) and to show a tropic response to oestradiol treatment in vivo (72).