Non-Metastatic Cutaneous Melanoma Induces Chronodisruption in Central and Peripheral Circadian Clocks

Abstract

The biological clock has received increasing interest due to its key role in regulating body homeostasis in a time-dependent manner. Cancer development and progression has been linked to a disrupted molecular clock; however, in melanoma, the role of the biological clock is largely unknown. We investigated the effects of the tumor on its micro- (TME) and macro-environments (TMaE) in a non-metastatic melanoma model. C57BL/6J mice were inoculated with murine B16-F10 melanoma cells and 2 weeks later the animals were euthanized every 6 h during 24 h. The presence of a localized tumor significantly impaired the biological clock of tumor-adjacent skin and affected the oscillatory expression of genes involved in light- and thermo-reception, proliferation, melanogenesis, and DNA repair. The expression of tumor molecular clock was significantly reduced compared to healthy skin but still displayed an oscillatory profile. We were able to cluster the affected genes using a human database and distinguish between primary melanoma and healthy skin. The molecular clocks of lungs and liver (common sites of metastasis), and the suprachiasmatic nucleus (SCN) were significantly affected by tumor presence, leading to chronodisruption in each organ. Taken altogether, the presence of non-metastatic melanoma significantly impairs the organism's biological clocks. We suggest that the clock alterations found in TME and TMaE could impact development, progression, and metastasis of melanoma; thus, making the molecular clock an interesting pharmacological target.

Keywords: cancer; central and peripheral clocks; chronodisruption; melanoma; tumor macroenvironment; tumor microenvironment.

Figure 1

(A,B) Non-metastatic skin melanoma affects clock gene expression in skin. Expression of (A) Per1, (B) Bmal1 in control skin, tumor-adjacent skin, and tumor samples. Letters represent significant differences in gene expression within the same group, Two-Way ANOVA followed by Tukey post-test. In (A) a ≠ b, p < 0.01, and c ≠ d, p < 0.05. In (B) a ≠ b, p < 0.01, c ≠ d, p < 0.01, and e ≠ f, p < 0.05. Significant differences among groups were demonstrated by Two-Way ANOVA followed by Bonferroni post-test, and are represented by asterisks having the control skin as reference. Differences between tumor-adjacent skin and tumor samples are indicated by brackets. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001. Values are presented as the mean expression (n = 3–7) ± SEM of the gene of interest normalized by Rpl37a RNA, relative to the minimal value of skin control group. (C) Temporal tumor melanin content. Boxplots show the median, quartiles, maximum, and minimum melanin contents. Melanin content was normalized by total protein. Letters represent statistical temporal differences in melanin content as revealed by One-Way ANOVA followed by Tukey post-test. a ≠ b, p < 0.05.

Figure 2

Non-metastatic skin melanoma affects clock-controlled gene expression in skin. Expression of (A) Pparγ, (B) Opn2, (C) Opn4 and (D) Xpa in control skin, tumor-adjacent skin, and tumor samples. Letters represent significant differences in gene expression within the same group, Two-Way ANOVA followed by Tukey post-test. In (A) a ≠ b, p < 0.01. In (B) a ≠ b, p < 0.01. In (C) a ≠ b, p < 0.05, c ≠ d, p < 0.0001, e ≠ f, p < 0.05 and g ≠ h, p < 0.01. In (D) a ≠ b, p < 0.001, c ≠ d, p < 0.01 and e ≠ f, p < 0.05. Significant differences among groups were demonstrated by Two-Way ANOVA followed by Bonferroni post-test, and are represented by asterisks having the control skin as reference. Differences between tumor-adjacent skin and tumor samples are indicated by brackets. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001. Values are presented as the mean expression (n = 3–7) ± SEM of the gene of interest normalized by Rpl37a RNA, relative to the minimal value of skin control group.

Figure 3

The molecular clock machinery is repressed in human primary melanomas compared to normal skin. Unsupervised hierarchical clustering of 557 GTEx normal (sun and non-sun exposed) skin samples and 104 TCGA primary melanomas according to the expression of clock and clock-controlled genes. Clusters were defined based on the Euclidean distance using complete linkage.

Figure 4

Non-metastatic skin melanoma affects circadian expression of clock genes and Xpa in lungs. Expression of (A) Per1, (B) Bmal1, (C) Reverb-α and (D) Xpa in lungs of control and tumor-bearing mice. Letters represent significant differences in gene expression within the same group, Two-Way ANOVA followed by Bonferroni post-test. In (A) a ≠ b, p < 0.05 and c ≠ d, p < 0.001. In (B) a ≠ b, p < 0.01. In (C) a ≠ b, p < 0.0001 and c ≠ d, p < 0.05. In (D) a ≠ b, p < 0.01. Significant differences between control and tumor-bearing mice were demonstrated by Two-Way ANOVA followed by Bonferroni post-test, and are represented by asterisks having the control lung as reference. * p <0.05, *** p <0.001, **** p <0.0001. Values are presented as the mean expression (n = 4–6) ± SEM of the gene of interest normalized by 18S RNA, relative to the minimal value found in the control group.

Figure 5

Non-metastatic skin melanoma affects circadian expression of clock genes and clock-controlled genes in liver. Expression of (A) Per1, (B) Bmal1, (C) Reverb-α, (D) Pparα, (E) Glut2 and (F) Xpa in liver of control and tumor-bearing mice. Letters represent significant differences in gene expression within the same group, Two-Way ANOVA followed by Bonferroni post-test. In (A) a ≠ b, p < 0.05 and c ≠ d, p < 0.01. In (B) a ≠ b, p < 0.001, c ≠ d, p < 0.05. In (C) a ≠ b, p < 0.0001, c ≠ d, p < 0.05. In (D) a ≠ b and c ≠ d, p < 0.05. In (E) a ≠ b, p < 0.05. In (F) a ≠ b, p < 0.05. Significant differences between control and tumor-bearing mice were demonstrated by Two-Way ANOVA followed by Bonferroni post-test, and are represented by asterisks having the control liver as reference ** p < 0.01, **** p < 0.0001. Values are presented as the mean expression (n = 4–6) ± SEM of the gene of interest normalized by 18S RNA, relative to the minimal value found in the control group.

Figure 6

Non-metastatic skin melanoma affects circadian expression of clock genes and cFos in SCN. Expression of (A) Per1, (B) Bmal1 and (C) cFos in SCN of control and tumor-bearing mice. Letters represent significant differences in gene expression within the same group, Two-Way ANOVA followed by Bonferroni post-test. In (A) a ≠ b, p < 0.01 and c ≠ d, p < 0.05. In (B) a ≠ b, p < 0.05. In (C) a ≠ b and c ≠ d, p < 0.05, e ≠ f, p < 0.01 and g ≠ h, p < 0.001. Significant differences between control and tumor-bearing mice were demonstrated by Two-Way ANOVA followed by Bonferroni post-test, and are represented by asterisks having the control SCN as reference. ** p < 0.01, **** p < 0.0001. Values are presented as the mean expression (n = 3–7) ± SEM of the gene of interest normalized by Rpl37a RNA, relative to the minimal value found in the control group.

Figure 7

Non-metastatic melanoma leads to systemic chronodisruption in peripheral and central clocks in mice. (A) In a physiological situation melanopsin-expressing retinal cells translate the photic information into glutamate release at the suprachiasmatic nucleus (SCN)—the central clock. The SCN then adjusts its molecular clock, comprised by several circadian oscillatory genes, to the environmental light-dark information (24 h duration). The SCN feeds the external time information to several regulatory regions of the brain, which control many biological processes. In fact, in a harmonic setting all biological processes within the organism are in phase and aligned to the external time, thus ensuring a homeostatic condition among all organs and systems; (B) However, this systemic harmonic condition is lost when a localized and non-metastatic melanoma is present. Our data show that the effects TME and TMaE lead, respectively, to a chronodisruption scenario in adjacent skin and in distant organs (lungs, liver, and SCN). In addition, TME and TMaE not only disrupt the molecular clock but also affect key tissue-specific regulatory genes. Based on the literature, we suggest that these tumor-induced effects are due to the presence of soluble factors which leak from the encapsulated tumor into the bloodstream, thus triggering systemic chronodisruption.