Treating cachexia using soluble ACVR2B improves survival, alters mTOR localization, and attenuates liver and spleen responses

Abstract

Background: Cancer cachexia increases morbidity and mortality, and blocking of activin receptor ligands has improved survival in experimental cancer. However, the underlying mechanisms have not yet been fully uncovered.

Methods: The effects of blocking activin receptor type 2 (ACVR2) ligands on both muscle and non-muscle tissues were investigated in a preclinical model of cancer cachexia using a recombinant soluble ACVR2B (sACVR2B-Fc). Treatment with sACVR2B-Fc was applied either only before the tumour formation or with continued treatment both before and after tumour formation. The potential roles of muscle and non-muscle tissues in cancer cachexia were investigated in order to understand the possible mechanisms of improved survival mediated by ACVR2 ligand blocking.

Results: Blocking of ACVR2 ligands improved survival in tumour-bearing mice only when the mice were treated both before and after the tumour formation. This occurred without effects on tumour growth, production of pro-inflammatory cytokines or the level of physical activity. ACVR2 ligand blocking was associated with increased muscle (limb and diaphragm) mass and attenuation of both hepatic protein synthesis and splenomegaly. Especially, the effects on the liver and the spleen were observed independent of the treatment protocol. The prevention of splenomegaly by sACVR2B-Fc was not explained by decreased markers of myeloid-derived suppressor cells. Decreased tibialis anterior, diaphragm, and heart protein synthesis were observed in cachectic mice. This was associated with decreased mechanistic target of rapamycin (mTOR) colocalization with late-endosomes/lysosomes, which correlated with cachexia and reduced muscle protein synthesis.

Conclusions: The prolonged survival with continued ACVR2 ligand blocking could potentially be attributed in part to the maintenance of limb and respiratory muscle mass, but many observed non-muscle effects suggest that the effect may be more complex than previously thought. Our novel finding showing decreased mTOR localization in skeletal muscle with lysosomes/late-endosomes in cancer opens up new research questions and possible treatment options for cachexia.

Keywords: Activin; Acute phase response; MDSC; Myostatin; Physical activity; Protein synthesis.

Figure 1

Schematic representation of the experimental design and the treatments. C26 cells were injected on Day 0, and sACVR2B‐Fc or PBS vehicle were administered on Days ‐11, ‐7, ‐3, 1, 5, and 9 in all experiments.

Figure 2

The effects of sACVR2B‐Fc administration on survival, tissue masses and food intake in C26 cancer cachexia. (A) A 3‐week Kaplan–Meier survival curve (log‐rank (Mantel‐Cox) test). N = 6, 12, 8, and 9 in CTRL, C26 + PBS, C26 + sACVR/b, and C26 + sACVR/c, respectively. (B) Body mass and the final tumour‐free body mass, in the short‐term experiment. There was a significant time × group interaction (P = 0.006, repeated measures ANOVA). Masses of (C) tibialis anterior (TA), (D) diaphragm (DIA), (E) the heart, and (F) epididymal white adipose tissue (eWAT) normalized to the length of the tibia in mm (TL) at 11 days after C26 cell inoculation. (G) Average food intake during Days 8–10 of the short term experiment, in which N = 3–4 cages/group, 2 mice/cage. (H) Tumour mass on Day 11 after C26 cell inoculation. *, ** and *** = P < 0.05, 0.01 and 0.001, respectively. CTRL vs. C26 + PBS difference was analysed by Student's t‐test (B–G), and differences between the C26‐groups with one‐way ANOVA with Holm–Bonferroni corrected LSD (B–H). Lines without vertical ends show a pooled effect: (D) sACVR2B‐Fc combined and (G) C26‐groups combined. N‐sizes are depicted in the bar graphs.

Figure 3

Decreased protein synthesis is associated with altered mTOR localization in the tumour‐bearing mice at 11 days after C26 cell inoculation. (A) Protein synthesis analysed by SUnSET in TA, diaphragm (DIA) and the heart (left) and the representative blots (right, C = CTRL, P = C26 + PBS, Ab = C26 + sACVR/b, Ac = C26 + sACVR/c). (B) Quantification of mTOR‐LAMP2 colocalization in TA and the representative images (scale bar = 10 μm). Membranes were excluded from the analysis, but this did not have major impact on the results (data not shown). (C) Phosphorylation of rpS6 on Ser240/244 in TA (left) and the representative blots (right). (D) Correlation between mTOR‐LAMP2 colocalization and protein synthesis in CTRL and C26 + PBS groups (Pearson correlation coefficient). * and ** = P < 0.05 and 0.01, respectively. Kruskall–Wallis with Holm–Bonferroni corrected Mann–Whitney U (A, C); Student's t‐test (B, C26‐ and sACVR2B‐Fc‐effects). Lines without vertical ends show a pooled effect of all C26‐groups combined. N‐sizes are depicted in the bar graphs.

Figure 4

Home cage physical activity and muscle oxidative properties at early phase of C26 cancer cachexia. (A) Activity indexes (AU) at baseline and at Day 10 after C26 cell injection. N = 2–3 cages/group, 2 mice/cage. This result was replicated in the second short‐term experiment (data not shown). (B) Citrate synthase activities in TA, diaphragm, and the heart on Day 11 after C26 cell injection. (C) PGC‐1α and cytochrome (Cyt) c, and (D) mitochondrial OXPHOS protein content in TA on Day 11 after C26 cell inoculation. (E) Representative blots. * and ** = P < 0.05 and 0.01, respectively. C26‐effect was analysed by Student's t‐test (A, B), and group differences by Kruskall–Wallis with Holm–Bonferroni corrected Mann–Whitney U (C, D). N = 7–9/group.

Figure 5

Liver mass, protein synthesis and markers of acute phase response on Day 11 after C26 cell injection. (A) Liver mass normalized to the length of the tibia (TL). (B) Liver protein synthesis (left) and representative blots (right). (C) Phosphorylation of rpS6 on Ser240/244 and Stat3 on Tyr705, and protein contents of fibrinogen and serpinA3N in liver (left) and the representative blots (right). N = 6–9/group. ** and *** = P < 0.01 and 0.001, respectively. Student's t‐test and one‐way ANOVA with Holm–Bonferroni corrected LSD (A), Kruskall–Wallis with Holm–Bonferroni corrected Mann–Whitney U (B, C). Lines without vertical ends in (C) show a pooled effect of both sACVR2B‐Fc groups combined. N‐sizes are depicted in the bar graphs.

Figure 6

Administration of sACVR2B‐Fc attenuates C26 cancer‐induced splenomegaly independent of splenic MDSCs. (A) Spleen mass normalized to the length of the tibia (TL) on Day 11 after C26 cell injection. (B) Haematoxylin and eosin staining of the spleen on Day 13 after C26 cell injection. (C) CD11b and GR‐1 (LY‐6C/G) count in spleen on Day 13 after C26 cell injection and representative immunofluorescence images. Scale bar = 100 μm. The mRNA expression of MDSC markers (D) interleukin‐10 (Il‐10, (E) S100 calcium binding protein A8 (S100a8), and (F) the splice variant of X‐box Binding Protein 1 (Xbp1s) as well as (G) Interferon Regulatory Factor 8 (Irf8), a negative regulator of MDSCs,41 on Day 13 after C26 cell injection. *, **, and *** = P < 0.05, 0.01, and 0.001, respectively. Student's t‐test and one‐way ANOVA with Holm–Bonferroni corrected LSD (A, C, D), Mann–Whitney U (E–G). Lines without vertical ends show a pooled effect of all C26‐groups combined. N‐sizes are depicted in the bar graphs. N = 7–9/group in (E–G).

Figure 7

Haematological parameters in C26 tumour‐bearing mice and the effects of sACVR2B‐Fc. (A) Red blood cell count (RBC), (B) haemoglobin, (C) haematocrit, (D) platelet count, and (E) white blood cell count (WBC) on Day 11 after C26 cell injection. *, **, and *** = P < 0.05, 0.01 and 0.001, respectively. Kruskall–Wallis with Holm–Bonferroni corrected Mann–Whitney U (A–C) or Student's t‐test and one‐way ANOVA with Holm–Bonferroni corrected LSD (D, E). N‐sizes are depicted in the bar graphs. These results were replicated in the second short‐term experiment (data not shown).