Berberine, an isoquinoline alkaloid isolated from a number of plants, has a rich history of use in Chinese medicine. It seems clear, considering the wide spectrum of therapeutic applications such as treatments for parasitic infection, bacterial diarrhea, inflammation, cardiovascular disease, hypercholesteremia, and diabetes, that berberine must trigger a number of physiological responses in biological systems.
Recent studies have documented berberine's effectiveness in the treatment of both human diabetes patients [3
]and animal models of diabetes [6
]. This has lead to numerous studies designed to understand berberine's mechanism of action regarding its hypoglycemic effect. The results of these studies are somewhat confusing and sometimes conflicting. Some studies indicate that berberine works through the insulin pathway by enhancing either insulin secretion [11
] or the signaling cascade [8
]. Others suggest berberine stimulates AMPK in an insulin pathway-independent fashion [6
] while others suggest that berberine has effects on both pathways [7
]. In most studies the berberine-enhanced glucose uptake has been attributed to increased GLUT4 activity. Some studies indicated that berberine increases GLUT4 production [37
] or translocation [6
], however other studies were unable to detect changes in GLUT4 [15
The acute effects of berberine on the transport activity of GLUT1 remain largely unexplored. A study in 3T3-L1 adipocytes showed that 6 hour exposure to berberine increases GLUT1 synthesis with no change in GLUT4 [15
]. However, a second study reported that 24 hour exposure to berberine did not change either GLUT1 or GLUT4 content and suggested that berberine increases GLUT1 activity [16
]. To explore the effects of berberine on GLUT1 activity, we measured glucose uptake in the presence of berberine in L929 fibroblast cells, a cell line that only expresses GLUT1 [30
]. We report that berberine significantly activates transport at 10 μM and reaches a maximum stimulation of about 5-fold at concentrations above 40 μM (see ). Significant activation occurs within 5 minutes and plateaus at about 30 minutes. This berberine activation of glucose uptake is faster than previously reported in L6 or 3T3-L1 cells, which did not see an effect until after 2-6 hours of exposure [7
]. Activation was not sensitive to cyclohexamide, indicating that the berberine effect does not depend on new protein synthesis. Recovery from berberine activation is slower than the activation with about 90% of the activation remaining after 30 minutes and 50% after 60 minutes.
Previous studies have suggested two distinct mechanisms for the activation of glucose uptake in L929 cells, one mechanism illustrated by the effects of methylene blue or sodium azide and the other by glucose deprivation [24
]. The kinetic analysis of the berberine activation indicates that the Km
of glucose transport is decreased, but the Vmax
is unaffected, suggesting that sensitivity of the system is enhanced, but not the responsiveness. This kinetic behavior is similar to that observed upon glucose deprivation in this cell line [26
]. To further explore the mechanism of berberine activated glucose uptake we explored the additivity of berberine's effects with maximally effective concentrations of methylene blue and azide or glucose deprivation. The results () indicate that among these stimulants, berberine induces the most robust activation of glucose uptake and that its effects are not enhanced by the other stimulants. This suggests that the mechanism of berberine activation shares common steps with both pathways.
Key studies in other cell lines have strongly suggested that berberine activation of glucose uptake is mediated through activation of AMPK. To explore this possibility we measured the dose dependent effects of Compound C, an inhibitor of AMPK, on both basal and berberine-stimulated glucose uptake. Compound C had no effect on glucose uptake, suggesting AMPK is not involve in the stimulation of glucose uptake in L929 cells. This is consistent with previous work in L929 cells that reported that AICAR, an activator of AMPK, inhibited rather than activated glucose uptake [24
]. This result was somewhat surprising given that previous work in L6 myotube cells had shown that the activation of glucose uptake by berberine could be completely blocked by compound C [14
]. We were able to repeat those results with L6 myotube cells in our laboratory as well (data not shown) suggesting that AMPK is not involved in the activation of glucose uptake in L929 cells. Also, since L6 myotube cells express both GLUT1 and GLUT4, this may suggest that AMPK activation is required for the activation of GLUT4, but not for GLUT1. The lack of involvement of AMPK in regulating glucose uptake in L929 fibroblast cells indicate that additional studies need to be done investigate the expression levels and role of AMPK in this cell line.
Studies that support AMPK activation as a key step in berberine's effects are not all in agreement regarding which downstream kinases are involved. Some studies implicate activation of ERK kinase [15
], while others provide evidence of berberine's activation of p38 MAPK [6
]. To investigate the involvement of these two kinases we measured the effects of SB203580 (inhibitor of p38 MAPK), and PD98059 (inhibitor of ERK kinase) on berberine stimulated glucose uptake. SB203580 had no effect on submaximally effective concentrations of berberine, but reduced maximum stimulation by 26%, while PD98059 completely blocked submaximal berberine stimulation and reduced maximum stimulation by 55%. The combination of the two inhibitors did not further inhibit berberine stimulation of glucose uptake (data not shown). These data suggest that ERK kinase pathway may be more involved than p38 MAPK, but neither is able to account for the full stimulatory effect of berberine. It is important to note that these inhibitory studies should be considered to be just the initial steps in discerning the mechanism of berberine activation of GLUT1. One concern is a potential cross reactivity of the inhibitors. For example, recent work in striated muscle has demonstrated that SB203580 also significantly inhibits the activation of Akt, a serine/threonine protein kinase, activated in the insulin signaling system [39
]. While this may not be important in L929 fibroblast cells, since this cell line is not insulin sensitive, it does point to potential complications in interpreting the results. Thus, future studies should look for a berberine stimulation of the phosphorylation of these kinases as well as identify other factors to account for the full activation of berberine.