Apelin Resistance Contributes to Muscle Loss during Cancer Cachexia in Mice

Figures

• 

  Figure 1. Different tumors induce body weight loss and muscle wasting in mice with different timing and tumor growth rate. In orange, 9–10 week-old male C26-bearing BALB/c mice start losing body weight already after 10 days from tumor injections and have lost 20% of body weight 12 days later, n = 9–10 (A). In blue, 9–10 week-old male MCG101-carrying C57BL6/J mice lose about 10% of their initial body weight after 20 days from tumor inoculation, n = 10–11 (B). In green, 9–10 week-old male LLC-bearing C57BL6/J mice present about 8% body wasting after 20 days, but this was not significant, n = 10 (C). Two-way ANOVA, * p ≤ 0.05, *** p ≤ 0.001, **** p ≤ 0.0001. In black, PBS-injected mice are indicated as controls (A–C). Tumor weights measured manually with a caliper over time are plotted (D). Ordinary one-way ANOVA, Tukey’s multiple comparisons test, not significant, n = 10–11. The weights of gastrocnemius (E) and tibialis anterior muscles (F) are reduced in all the tumor-bearing mice, but to different extents. In orange, C26 n = 18–20, in blue, MCG101 n = 20–22 and in green, LLC n = 20. Unpaired t-test, * p ≤ 0.05, ** p ≤ 0.01, **** p ≤ 0.0001.

• 

  Figure 2. In the atrophied muscle of all the three animal models of cancer cachexia and cancer patients, apelin expression falls when MuRF1 increases. Q-PCR shows that mRNA levels of the ubiquitin ligase MuRF1 are highly induced in the TA of mice carrying the tumor C26 (A) or MCG101 (B) or LLC (C). Unpaired t-test, **** p ≤ 0.0001 (A); Mann–Whitney test, *** p ≤ 0.001 (B). Unpaired t-test, * p ≤ 0.05 (C), n = 7–10 (A–C). Apelin (Apln) expression is drastically decreased in TA from mice bearing C26 (D) or MCG101 (E) or LLC (F). Unpaired t-test, * p ≤ 0.05 (D), **** p ≤ 0.0001 (E), *** p ≤ 0.001 (F), n = 7–10. The mRNA levels of MuRF1 are not induced in the TA of mice carrying the non-cachectic tumor 4T1 (G) and apelin expression is not altered in this tumor model (H). 3HK, three housekeeping genes (TBP, IPO8, and Gusb), IPO8 or Importin 8 and 18S or 18S ribosomal RNA, were used to normalize the data. Unpaired t-test, not significant, n = 5. In rectus abdominis muscles from cancer patients, as soon as MuRF1 increases, apelin decreases (I). The scatter plot shows the inverse correlation between the expression of apelin and MuRF1. These expression data are log2-transformed. Spearman’s test,…

• 

  Figure 3. Apelin protects myotubes from dexamethasone-induced atrophy. Four day-differentiated myotubes were exposed for 24 h to vehicle (Ctrl) or 10 μM dexamethasone (Dexa) and representative brightfield images are shown. Scale bar, 15 μm (A). In radioactive-based assays, we measured long-lived protein degradation of myotubes treated with vehicle or 10 μM Dexa for 36 h; it was induced by 25% by Dexa (B). Multiple t-tests, **** p ≤ 0.0001, n = 6. Dexa-treated myotubes have lower total protein content than controls (Ctrl) (C). Unpaired t-test, *** p ≤ 0.001, n = 6. Like in vivo, Q-PCR shows that MuRF1 expression is induced (D), while apelin expression is reduced (E) in myotubes treated for 24 h with 10 μM Dexa. TBP was used as housekeeping gene. Unpaired t-test, **** p ≤ 0.0001, n = 6. Four day-differentiated myotubes transfected with the empty vector pLKO, as control, or preproapelin-encoding plasmid for 24 h were treated with 10 μM Dexa for 24 h. Exogenous preproapelin partially restrained the Dexa-induced degradation of long-lived proteins (F). Ordinary one-way ANOVA, * p ≤ 0.05, n = 6, ** p ≤ 0.01. Myoblasts were transfected with a plasmid in which Firefly luciferase (FLuc) is under the…

• 

  Figure 4. While L-apelin 13 is degraded in mouse serum in three hours, D-apelin 13 is stable for days. The peptide [Pyr1]-apelin 13 (MW 1533.81 g/mol) in L version (A) or D version of apelin 13 (B) were incubated at the same concentration in mouse serum at 37 °C. After different times from three hours to five days, a sample was analyzed using a MALDI-TOF mass spectrometer in linear mode. The molecular weights were compared with the original peptides at time 0. The highlighted peak that detects L-apelin 13 disappeared after 3 h (A), when additional peaks corresponding to lower molecular weights appear, indicating its fragmentation. Instead, D-apelin 13 was more stable in the mouse serum, because its peak (MW 1550 g/mol) persisted for five days (B). Myoblasts were transfected with a plasmid in which Firefly luciferase (FLuc) is under the control of cAMP responsive elements (CRE) and another for Renilla luciferase (RLuc) under the control of thymidine kinase promoter, to normalize the data. After 24 h, myoblasts were treated for 5 h with vehicle or 1 μM L-apelin 13 or D-apelin 13, in combination with vehicle or 50 μM forskolin and the ratio of FLuc/RLuc is plotted. L-apelin 13 and D-apelin 13…

• 

  Figure 5. Intramuscular injection of preproapelin-expressing adenoassociated viruses or systemic treatment with D-apelin 13 do not protect muscles from cancer-induced wasting. Intramuscular injection of the preproapelin-expressing adenoassociated viruses type 9 (AAV9) does not preserve the TA from atrophy (A), even though the apelin gene is overexpressed in the muscle as shown by Q-PCR (B). C26-induced MuRF1 is not restrained by apelin overexpression (C). IPO8 is used to normalize the data. Ordinary one-way ANOVA, Kruskal–Wallis test, * p ≤ 0.05, *** p = 0.0001, **** p ≤ 0.0001, n = 8–18. Vehicle (H2O) or 0.5 or 1.5 μmol/Kg D-apelin 13 were injected daily intraperitoneally in BALB/c mice and body weighs monitored over time. After 13 days, there were no differences in BWL among groups, excluding overt toxicity. Mouse body weights are plotted over time (D). One-way ANOVA for repeated measures, not significant, n = 3–4. C26 tumor cells subcutaneously injected in BALB/c mice cause BWL after 10 days from the injection, and D-apelin 13 does not preserve mouse body weight (E). One-way ANOVA for repeated measures, * p ≤ 0.05, n = 15. D-apelin 13 does not preserve muscles from atrophy induced by…

• 

  Figure 6. Muscles from mice with cancer cachexia display apelin resistance. Proteins from TA muscles of C26-carrying mice injected with preproapelin-expressing AAV9 were analyzed in Western blot for p-AKT, AKT, p-4EBP1, 4EBP1, MuRF1, and vinculin as loading control (A). Quantitations for p-AKT/AKT/vinculin (B), p-4EBP1/4EBP1/vinculin (C), and MuRF1/vinculin (D) are shown. Ordinary one-way ANOVA, Tukey’s multiple comparison test, * p ≤ 0.05, ** p ≤ 0.01, **** p ≤ 0.0001, n = 3–4. The APJ-GFP-encoding plasmid was electroporated for 14 days in mice that were injected with C26 tumor cells the day after. A representative image is shown (E). Scale bar, 50 μm. At sacrifice, TA were dissected, frozen, and cut and the mean fiber CSA of APJ-GFP-expressing fibers was found bigger than that of adjacent non-expressing ones (F). Unpaired t-test, Mann–Whitney test, * p ≤ 0.05, seven mice and 89 fibers.

• 

  Figure 7. Plasma apelin is increased in cachectic cancer-bearing mice and it seems tumor-derived. ELISA was done on the plasma of mice with the cachectic tumors C26 (A), MCG101 (B) or LLC (C). Apelin was high in plasma of all three cancer types-bearing animals. Mann–Whitney test, * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001. PBS, C26, n = 4–5; PBS, MCG101, n = 10–14; PBS, LLC, n = 6–10. Proteins from 4T1 or C26 tumors excised from mice were analyzed in Western blot for apelin and GAPDH as loading control (D). Quantitation for apelin/GAPDH is shown (E). Unpaired t-test, ** p ≤ 0.01, n = 4–5. TA weight of non-cachectic and cachectic pancreatic cancer bearing-mice is shown (F), Unpaired t-Test, * p ≤ 0.05, n = 5–6. Q-PCR showed that the tumoral mRNA levels of apelin were higher in cachectic than in non-cachectic mice bearing pancreatic tumors and sacrificed, respectively, 31 vs. 26 days after tumor transplantation (G). Unpaired t-test, * p ≤ 0.05, n = 5–6.