Figure 1.
Circadian oscillations and their LAN-induced circadian disruption in cancer host levels of hormonal, metabolic and dietary factors.
(A–F) Plasma levels of melatonin (A), TFAs (B), LA (C), glucose (D), lactate (E) and insulin (F) were measured in nude female host rats bearing tissue-isolated SR- MCF-7 human breast cancer xenografts under LD,12∶12 (solid black circles) or LD,12∶12+ LAN (0.2 lux) (solid red circles). Arterial blood samples were obtained via cardiac puncture at six different circadian time points during a single 24-hour period. The same single 24-hour pattern for each plasma analyte is displayed twice to illustrate rhythm continuity. Zeitgeber time (ZT) represents hours after lights on (ZT0 or 0600 hours); lights off at ZT12 (1800 hours). Black bars at bottom of figures indicate the dark phase and red bars indicate LAN. Solid black or red circles are mean ±SD; n = 6 at each time point. Significant rhythmic patterns under LD,12∶12 conditions in A – F, p<0.001; significant but disrupted rhythmic patterns under dim LAN conditions A – C and F only, p<0.001 (one-way ANOVA). *p<0.01; **p<0.05 LAN vs LD,12∶12 (Student’s t test).
Figure 2.
Circadian oscillations and their LAN-induced disruption in tumor cAMP signaling, fatty acid uptake and metabolism, the Warburg effect and proliferative activity and their impact on tumor growth.
(A–H) Tumor levels of cAMP (A), TFA uptake (B), LA uptake (C), 13-HODE formation (D), glucose uptake (E), lactate formation (F), [3H]thymidine incorporation into DNA (G), and DNA content were measured under LD,12∶12 or LD,12∶12+ LAN; see legend for Figure 1 for further experimental conditions. (I) Tumor growth in both groups was measured over the course of the experiment. Solid black or red circles are mean ±SD estimated tumor weight; n = 6 for estimated tumor weight. Cosinor analysis revealed robust and highly significant rhythmic patterns under LD,12∶12 conditions in tumor analytes; no significant daily rhythmic patterns were detected under dim LAN (see summary Table 3). *p<0.01; **p<0.05 LEN vs LD,12∶12 (Student’s t test). Tumor growth curves in I, p<0.01 slope of tumor growth in LAN group vs LD, 12∶12 group (linear regression).
Figure 3.
Circadian oscillations and their LAN-induced circadian disruption in tumor signaling and transcriptional regulatory molecules involved in the Warburg effect (e.g., AKT, c-MYC and HIF-1α).
(A–F) Tumor levels of phospho-AKTser473 (A & B), cMyc (C & D), and HIF-1α (E & F) were measured under LD,12∶12 or LD,12∶12+ LAN; see legend for Figure 1 for further experimental conditions. Solid black or red circles represent the mean ±SD relative expression (derived from the densitometric quantitation of the immunoblots) of either pAKTser473, c-MYC and HIF-1α at each circadian time point; n = 3 representative tumor samples at each time point. Relative expression of pAKTser473 represents the ratio of pAKTser473 protein to total (t)AKT protein; relative expression of c-MYC and HIF-1α represents the ratio of these proteins to tubulin and GAPDH, respectively. Cosinor analysis revealed robust and highly significant rhythmic patterns under LD,12∶12 conditions in tumor analytes; except for c-MYC, no significant daily rhythmic patterns were detected under dim LAN (see summary Table 4). Statistical comparisons were unable to be made between corresponding time points between the LD,12∶12 and LAN groups since immunoblots were performed for the two groups in separate runs. Only a single 24-hour cycle of immunoblots were run and displayed, whereas the densitometric representation of those same immunoblots was displayed twice to illustrate rhythm continuity as indicated in the legend of Fig. 1.
Figure 4.
Response of signaling and transcriptional regulatory factors involved in the Warburg effect (e.g., AKT, c-MYC and HIF-1α) in tissue-isolated SR- MCF-7 human breast cancer xenografts to short-term perfusion with nocturnal physiological concentrations of melatonin either in the absence or presence of melatonin receptor antagonist S20928 or 13-HODE.
Xenografts were perfused in situ for 60 min with rat donor whole blood containing either vehicle, melatonin (500 pM or 1 nM), S20928 (1 nM), 13-HODE (12 µg/ml), melatonin+S20928 or melatonin +13-HODE. (A – H) Tumor perfusions were performed between 2–4 hours after lights on (0600 hours) under LD,12∶12 conditions. In the perfusions A & B, a melatonin concentration of 1 nM was used whereas in all other perfusions (C – H) melatonin was used at 500 pM. Vertical bars represent the mean (±SD) relative expression (derived from the densitometric quantitation of the immunoblots) in tumors (n = 3–4) of either pAKTs473 (A–D), c-MYC (E & F), or HIF-1α (G & H). Relative expression of pAKTser473 represents the ratio of pAKTser473 protein to total (t)AKT protein; relative expression of c-MYC and HIF-1α represents the ratio of these proteins to GAPDH. Statistical analysis was performed using one-way ANOVA followed by Bonferroni’s Multiple Comparison Test to make multiple comparisons among all groups indicated by the brackets within each bar graph: * over each bracket indicates that differences are statistically significant at p<0.05.
Table 1.
Tumor glucose uptake and lactate release (Warburg effect), arterial and venous oxygen (O2) and carbon dioxide (CO2), tension O2 uptake and CO2 production measured across MCF-7(SR−) human breast cancer xenografts perfused for 60 minutes with rodent donor whole blood in situ.
Table 2.
Tumor cAMP levels, TFA uptake, LA uptake, 13-HODE formation, [3H]thymidine incorporation into DNA and DNA content in tissue-isolated MCF-7 (S−) human breast cancer xenografts perfused for 60 minutes with rodent whole blood in situ.
Table 3.
Circadian oscillations and their LAN-induced disruption in tumor cAMP signaling, fatty acid uptake and metabolism, the Warburg effect and proliferative activity and their impact on tumor growth.
Table 4.
Circadian oscillations and their LAN-induced circadian disruption in tumor signaling and transcriptional regulatory molecules involved in the Warburg effect (e.g., AKT, c-MYC and HIF-1α).
Figure 5.
A provisional mechanistic scheme by which the host circadian system regulates cancer metabolism by linking lipid signaling with the Warburg effect in human breast cancer xenografts via the nocturnal melatonin signal.
During the light phase, host blood levels of linoleic acid are low and melatonin concentrations are nil; however, cAMP-stimulated tumor uptake of linoleic acid and its metabolism to the mitogenic signaling molecule 13-HODE is maximal. Increased production of 13-HODE stimulates increased levels and phosphorylation/activation of AKT leading to enhanced aerobic glycolysis, cell proliferation and tumor growth. The introduction of dim LAN induces the continued operation of the tumor growth stimulatory mechanisms observed during the light phase. During the dark phase, host blood levels of linoleic acid are high and melatonin concentrations are maximal. Melatonin, by acting through tumor MT1 receptors [12], [27], down-regulates cAMP formation and blocks linoleic acid uptake and its metabolism to 13-HODE. Decreased production of 13-HODE results in an attenuation of AKT signaling leading to diminished aerobic glycolysis, cell proliferation and tumor growth. Although the exact role of melatonin-induced HIF-1α expression during the dark phase is unclear, it may involve newly discovered transcription-independent mechanisms that paradoxically contribute to the inhibition of cell proliferation and cell survival [32]–[34].