Fig 1.
dZIP13 knockdown promotes RafGOFScrib−/− tumorigenesis and malignant development.
(A) dZIP13 RNAi increased tumor overgrowth in the cephalic complex of RafGOFScrib−/− flies at day 10 after egg laying (AEL). Scale bar, 50 μm. (B-C) Relative fluorescence intensity (B) and tumor size (C) were quantified for the specified genotypes. n = 30 (RafGOFScrib−/− control), n = 28 (dZIP13 RNAi; RafGOFScrib−/−), and n = 28 (dZIP13 RNAi1#; RafGOFScrib−/−). (D) Tumor growth in the gonad was exacerbated by dZIP13 RNAi at 10 days AEL. Scale bar, 50 μm. (E-F) Quantification of relative fluorescence intensity (E) and tumor size (F) for the indicated genotypes. n = 24 (RafGOFScrib−/− control), n = 21 (dZIP13 RNAi; RafGOFScrib−/−), n = 20 (dZIP13 RNAi1#; RafGOFScrib−/−). (G) The invasive characteristics of tumors were classified into four grades based on our classification criteria at 10 days AEL. Scale bar, 50 μm. (H) Statistical analysis of tumor invasion levels at 10 days AEL. n = 32 (RafGOFScrib−/−), n =38 (dZIP13 RNAi; RafGOFScrib−/−), n = 36 (dZIP13 RNAi1#; RafGOFScrib−/−). (I) Representative confocal images showing metastatic tumor clones (GFP-positive) in distant tissues of dZIP13 RNAi; RafGOFScrib−/− flies. Scale bar, 50 μm. (J) Tumor cells invading the fat body. (K) Tumor cell migration into the body wall muscle. (L) GFP-positive metastatic clones detected in the gut. (%Control indicated that the proportion of invasion into fatbody, muscle and gut in total.) Data are presented as mean ± SEM. Statistical significance was calculated using unpaired two-tailed Student′s t-test or chi-square test as appropriate. *p< 0.05, **p< 0.01, ***p< 0.001. Genotypes used are as follows: (A–L) ey-Flp/+; Act>y+-Gal4, UAS-GFP/+; FRT82B tub-Gal80/UAS-RafGOF FRT82B Scrib−/−, ey-Flp/+; Act>y+-Gal4, UAS-GFP/dZIP13 RNAi; FRT82B tub-Gal80/UAS-RafGOF FRT82B Scrib−/− and ey-Flp/+; Act>y+-Gal4, UAS-GFP/dZIP13 RNAi1#; FRT82B tub-Gal80/UAS-RafGOF FRT82B Scrib−/−.
Fig 2.
The tumor growth and metastasis in both RafGOFScrib−/− and RafGOFScrib−/−, dZIP13 RNAi flies (10 d AEL) could be modified by dietary iron.
(A) Ferrozine assays suggested increased iron content in the cephalic complex of RafGOFScrib−/−, and dZIP13 RNAi increased the iron content even more. n = 150 cephalic complexes per group. (B) Aconitase activity demonstrated a marked increase in iron levels within the cephalic complex dZIP13 RNAi; RafGOFScrib−/− compared to RafGOFScrib−/−. n = 150 cephalic complexes per group. (C) The survival rate of RafGOFScrib−/− and dZIP13 RNAi; RafGOFScrib−/− were inhibited by BPS and enhanced by FAC. n = 50 larvae per vial, n = 6 vials per experimental group. (D) Tumor growth (indicated by red arrows) and metastasis (indicated by white arrows) in both RafGOFScrib−/− and dZIP13 RNAi; RafGOFScrib−/− flies (10 d AEL) were effectively inhibited by BPS while significantly enhanced by FAC. Scale bar, 100 μm. "NF" refers to "normal food", which was used as the standard control diet in experiments. (E-F) Quantification of relative fluorescence intensity (E) and tumor size (F) for the specified genotypes. n = 9. *p < 0.05, **p < 0.01, ***p < 0.001 and ns no significant. Genotypes used are as follows: (A–B) ey-Flp/ + ; Act > y+-Gal4 (control), ey-Flp/ + ; Act > y+-Gal4, UAS-GFP/ + ; FRT82B tub-Gal80/UAS-RafGOF FRT82B Scrib−/− and ey-Flp/ + ; Act > y+-Gal4, UAS-GFP/dZIP13 RNAi; FRT82B tub-Gal80/UAS-RafGOF FRT82B Scrib−/−. (C-F) ey-Flp/ + ; Act > y+-Gal4, UAS-GFP/ + ; FRT82B tub-Gal80/UAS-RafGOF FRT82B Scrib−/− and ey-Flp/ + ; Act > y+-Gal4, UAS-GFP/dZIP13 RNAi; FRT82B tub-Gal80/UAS-RafGOF FRT82B Scrib−/−.
Fig 3.
dZIP13 knockdown enhances tumor growth and invasion through JAK/STAT signalling pathway activation, which results from cytosolic iron accumulation.
(A) The tumor growth and invasion induced by dZIP13 RNAi were notably inhibited by disrupting the JAK/STAT signalling pathway through the co-expression of DomeDN. Scale bar, 50 μm. (B-C) Relative fluorescence intensity (B) and tumor size (C) were quantified for the specified genotypes. n = 12 (RafGOFScrib−/− control), n = 8 (DomeDN; RafGOFScrib−/−), n = 13 (dZIP13 RNAi; RafGOFScrib−/−), n = 9 (DomeDN, dZIP13 RNAi; RafGOFScrib−/−). (D) Immunohistochemical staining demonstrates that dZIP13 RNAi leads to a decrease in E-cadherin expression, which can be rescued by DomeDN. Scale bar, 200 μm. (E) Western blot analysis demonstrates that dZIP13 RNAi leads to a decrease in E-cadherin expression, which can be rescued by DomeDN. (F) Quantification the E-cadherin level of the indicated genotypes. (G) Compared with the control, dZIP13 RNAi, Mvl OE and Tsf1 OE induced 10 × STAT.GFP expression in the eye-antennal discs. Mvl RNAi or Tsf1 RNAi suppressed the induced 10 × STAT.GFP in dZIP13 RNAi. Scale bar, 100 μm. (H) Quantification of STAT.GFP intensity for the specified genotypes. *p < 0.05, **p < 0.01, ***p < 0.001. Genotypes used are as follows: (A-F) ey-Flp/ + ; Act > y+-Gal4, UAS-GFP/ + ; FRT82B tub-Gal80/UAS-RafGOF FRT82B Scrib−/−, ey-Flp/ + ; Act > y+-Gal4, UAS-GFP/dZIP13 RNAi; FRT82B tub-Gal80/UAS-RafGOF FRT82B Scrib−/−, ey-Flp/ + ; Act > y+-Gal4, UAS-GFP/UAS-DomeDN; FRT82B tub-Gal80/UAS-RafGOF FRT82B Scrib−/− and ey-Flp/ + ; Act > y+-Gal4, UAS-GFP/UAS-DomeDN,dZIP13 RNAi; FRT82B tub-Gal80/UAS-RafGOF FRT82B Scrib−/−. (G-H) ey-gal4/ + ; 10 × STAT.GFP/+ (control), ey-gal4/dZIP13 RNAi; 10 × STAT.GFP/ + , ey-gal4/Mvl OE; 10 × STAT.GFP/ + , ey-gal4/UAS-Tsf1 OE; 10 × STAT.GFP/ + , ey-gal4/dZIP13 RNAi; 10 × STAT.GFP/Mvl RNAi and ey-gal4/dZIP13 RNAi; 10 × STAT.GFP/Tsf1-RNAi.
Fig 4.
The activation of iron on JAK/STAT signalling and tumor progression depends on EZH2.
(A) qPCR showed dZIP13 RNAi and the addition of extra FAC increased the EZH2 mRNA level, whereas BPS dramatically down-regulated it in RafGOFScrib−/− tumors. n = 50 cephalic complexes per group. (B) Western blot analysis reveals that the dietary iron chelator BPS significantly reduced the levels of EZH2 in tumors, while dZIP13 RNAi and the addition of extra FAC led to an increase in these levels. n = 50 cephalic complexes per group. (C) Quantitative assessment of EZH2 levels from (B). (D) The effect of iron on tumor growth and invasion was lost following the knockdown of EZH2. Scale bar, 50 μm. (E-F) Quantification of relative fluorescence intensity (E) and tumor size (F) for the specified genotypes. n = 9 per group. (G) The influence of iron on 10 × STAT.GFP expression was absent in EZH2 RNAi. Scale bar, 100 μm. (H) Quantification of STAT.GFP intensity for the indicated genotypes. n = 6. *p < 0.05, **p < 0.01 and ***p < 0.001. Genotypes used are as follows: (A-C) ey-Flp/ + ; Act > y+-Gal4, UAS-GFP/ + ; FRT82B tub-Gal80/UAS-RafGOF FRT82B Scrib−/−. (D-F) ey-Flp/ + ; Act > y+-Gal4, UAS-GFP/ + ; FRT82B tub-Gal80/UAS-RafGOF FRT82B Scrib−/− and ey-Flp/ + ; Act > y+-Gal4, UAS-GFP/EZH2 RNAi; FRT82B tub-Gal80/UAS-RafGOF FRT82B Scrib−/−. (G-H) ey-gal4/ + ; 10 × STAT.GFP/+ (control) and ey-gal4/EZH2 RNAi; 10 × STAT.GFP/ + .
Fig 5.
Iron regulates the transcription of EZH2 through TET.
(A) The TET enzyme activity was considerably diminished in RafGOFScrib−/− fed with BPS while increased under dZIP13 RNAi or fed with FAC in RafGOFScrib−/−. n = 50 cephalic complexes per group. (B) The mRNA levels of EZH2 in tumors were significantly decreased by BPS and markedly increased by FAC. UAS-TET RNAi suppressed the regulation of iron on EZH2 expression. n = 150 cephalic complexes per group. (C) Dietary iron supplementation elevated the expression of EZH2 in the tumors, whereas iron chelator BPS reduced its expression. EZH2 level was reduced in UAS-TET RNAi, and UAS-TET RNAi inhibited the regulation of iron on EZH2 expression. (D-E) Both the activated JAK/STAT signalling (D) and tumor progression (E) were suppressed by UAS-TET RNAi, and the regulation of iron on JAK/STAT signalling and tumor progression disappeared in UAS-TET RNAi. Scale bar, 100 μm. (F-G) Quantification of relative fluorescence intensity (F) and tumor size (G) for the specified genotypes. n = 9 per group. (H-I) The regulation of iron on tumor invasion (H) and migration (I) disappeared in UAS-TET RNAi. n = 30 animals were used, derived from three separate experiments. *p < 0.05, **p < 0.01, ***p < 0.001 and ns no significant. Genotypes were as follows: (A) ey-Flp/ + ; Act > y+-Gal4, UAS-GFP/ + ; FRT82B tub-Gal80/UAS-RafGOF FRT82B Scrib−/−. (B-C) ey-Flp/ + ; Act > y+-Gal4, UAS-GFP/ + ; FRT82B tub-Gal80/UAS-RafGOF FRT82B Scrib−/− and ey-Flp/ + ; Act > y+-Gal4, UAS-GFP/UAS-TET RNAi; FRT82B tub-Gal80/UAS-RafGOF FRT82B Scrib−/−. (D) ey-gal4/ + ; 10 × STAT.GFP/+ (control) and ey-gal4/UAS-TET RNAi; 10 × STAT.GFP/ + . (E-I) ey-Flp/ + ; Act > y+-Gal4, UAS-GFP/ + ; FRT82B tub-Gal80/UAS-RafGOF FRT82B Scrib−/− and ey-Flp/ + ; Act > y+-Gal4, UAS-GFP/UAS-TET RNAi; FRT82B tub-Gal80/UAS-RafGOF FRT82B Scrib−/−.
Fig 6.
The exacerbation of tumor malignant development by dZIP13 RNAi involves the recruitment and proliferation of hemocytes, which are mediated by upds.
(A) The upds mRNA levels were increased in dZIP13 RNAi tumors. n = 50 cephalic complexes per group. (B) The upd3-lacZ reporter expression revealed an increase in upd3 protein levels within dZIP13 RNAi-associated tumors. Scale bar, 200 μm. (C) Co-immunostaining showing an increased number of hemocytes (red, anti-NimC1) adhering to the surface of eye-antennal discs in RafGOFScrib−/− and dZIP13 RNAi;RafGOFScrib−/− tumors. Scale bar: 200 μm. (D) The ratio of PH3-positive hemocytes to the total hemocyte population was quantified in RafGOFScrib−/− and dZIP13 RNAi; RafGOFScrib−/− eye-disc tumor clones.stained with an anti-PH3 antibody in eye disc. Scale bar, 200 μm. (E) Eye-antennal discs bearing dZIP13 RNAi; RafGOFScrib−/− clones were treated with 1% DMSO (control) or rapamycin. (F-G) Tumor size and invasive behavior were assessed by GFP labeling of clones. Compared to control, rapamycin treatment markedly reduced clone overgrowth and invasion into surrounding tissues. Quantification of tumor size (F) and invasion frequency (G) following treatment. n = 6. *p < 0.05, **p < 0.01, ***p < 0.001 and ns no significant. Genotypes used are as follows: (A-C) ey-Flp/ + ; Act > y+-Gal4, UAS-GFP/ + ; FRT82B tub-Gal80/UAS-RafGOF FRT82B Scrib−/− and ey-Flp/ + ; Act > y+-Gal4, UAS-GFP/EZH2 RNAi; FRT82B tub-Gal80/UAS-RafGOF FRT82B Scrib−/−. (D-G) ey-Flp/ + ; Act > y+-Gal4, UAS-GFP/ + ; FRT82B tub-Gal80/UAS-RafGOF FRT82B Scrib−/− and ey-Flp/ + ; Act > y+-Gal4, UAS-GFP/dZIP13 RNAi; FRT82B tub-Gal80/UAS-RafGOF FRT82B Scrib−/−.
Fig 7.
A schematic representation of the mechanisms by which the iron transporter dZIP13 influences tumor development.
Iron is an essential regulator of TET activity in vivo. The knockdown of dZIP13 causes iron to accumulate in the cytosol, leading to an increase in TET activity. TET subsequently induces EZH2 transcription. EZH2 regulates JAK/STAT activation in tumors, resulting in hemocyte recruitment.
Table 1.
The primers used for qPCR.