ACVR2B antagonism as a countermeasure to multi-organ perturbations in metastatic colorectal cancer cachexia

Abstract

Background: Advanced colorectal cancer (CRC) is often accompanied by the development of liver metastases, as well as cachexia, a multi-organ co-morbidity primarily affecting skeletal (SKM) and cardiac muscles. Activin receptor type 2B (ACVR2B) signalling is known to cause SKM wasting, and its inhibition restores SKM mass and prolongs survival in cancer. Using a recently generated mouse model, here we tested whether ACVR2B blockade could preserve multiple organs, including skeletal and cardiac muscle, in the presence of metastatic CRC.

Methods: NSG male mice (8 weeks old) were injected intrasplenically with HCT116 human CRC cells (mHCT116), while sham-operated animals received saline (n = 5-10 per group). Sham and tumour-bearing mice received weekly injections of ACVR2B/Fc, a synthetic peptide inhibitor of ACVR2B.

Results: mHCT116 hosts displayed losses in fat mass ( - 79%, P < 0.0001), bone mass ( - 39%, P < 0.05), and SKM mass (quadriceps: - 22%, P < 0.001), in line with reduced muscle cross-sectional area ( - 24%, P < 0.01) and plantarflexion force ( - 28%, P < 0.05). Further, despite only moderately affected heart size, cardiac function was significantly impaired (ejection fraction %: - 16%, P < 0.0001; fractional shortening %: - 25%, P < 0.0001) in the mHCT116 hosts. Conversely, ACVR2B/Fc preserved fat mass ( + 238%, P < 0.001), bone mass ( + 124%, P < 0.0001), SKM mass (quadriceps: + 31%, P < 0.0001), size (cross-sectional area: + 43%, P < 0.0001) and plantarflexion force ( + 28%, P < 0.05) in tumour hosts. Cardiac function was also completely preserved in tumour hosts receiving ACVR2B/Fc (ejection fraction %: + 19%, P < 0.0001), despite no effect on heart size. RNA sequencing analysis of heart muscle revealed rescue of genes related to cardiac development and contraction in tumour hosts treated with ACVR2B/Fc.

Conclusions: Our metastatic CRC model recapitulates the multi-systemic derangements of cachexia by displaying loss of fat, bone, and SKM along with decreased muscle strength in mHCT116 hosts. Additionally, with evidence of severe cardiac dysfunction, our data support the development of cardiac cachexia in the occurrence of metastatic CRC. Notably, ACVR2B antagonism preserved adipose tissue, bone, and SKM, whereas muscle and cardiac functions were completely maintained upon treatment. Altogether, our observations implicate ACVR2B signalling in the development of multi-organ perturbations in metastatic CRC and further dictate that ACVR2B represents a promising therapeutic target to preserve body composition and functionality in cancer cachexia.

Keywords: Activin signalling; Bone; Cachexia; Colorectal cancer; Heart; Liver metastases; Skeletal muscle.

Figure 1

ACVR2B/Fc preserves body weight (BW) in mHCT116 hosts. (A) BW curves, (B) BW change at time of sacrifice (vs. Day 1), (C) carcass weights, (D) gonadal fat normalized to initial body weight (IBW), (E) fat content, and (F) lean content determined by EchoMRI (E and F) of NSG male mice (8 weeks old) intrasplenically injected with HCT116 tumour cells (1.25 × 10⁵ cells per mouse in sterile PBS: T) or an equal volume of vehicle (sham: S) and administered ACVR2B/Fc (A) (n = 5–10). (G) Liver weights normalized to IBW, (H) representative haematoxylin and eosin staining, and (I) tumour area quantification of liver tissue from S, A, T, and A + T mice. Black arrows indicate tumours, and images were taken at ×5 magnification. Scale bars: 200 μm. Data are expressed as mean ± SD. Significance of the differences: *P < 0.05, ***P < 0.001, ****P < 0.0001 vs. S; ^### P < 0.001, ^#### P < 0.0001 vs. T.

Figure 2

ACVR2B/Fc preserves muscle mass and strength in mHCT116 hosts. (A) Gastrocnemius, (B) tibialis anterior, and (C) quadriceps muscles normalized to initial body weight (IBW) in NSG male mice (8 weeks old) intrasplenically injected with HCT116 tumour cells (1.25 × 10⁵ cells per mouse in sterile PBS: T) or an equal volume of vehicle (sham: S) and administered ACVR2B/Fc (A) (n = 5–10). (D) Cross‐sectional area (CSA) frequency distribution, (E) mean CSA, and (F) representative images of dystrophin‐stained tibialis anterior muscles for CSA assessment (n = 4–6). Images were taken at ×20 magnification. Scale bars: 100 μm. (G) Grip strength assessment (n = 5–10). (H) In vivo force–frequency plantarflexion curve (n = 4). Data are expressed as mean ± SD. Significance of the differences: *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 vs. S; ^#### P < 0.0001 vs. T. For (H): *P < 0.05 vs. all other groups 80–150 Hz.

Figure 3

ACVR2B/Fc improves circulating IGF‐1 in mHCT116 hosts. (A) IL‐6 and (B) IGF‐1 plasma levels assessed by magnetic multiplex assay in NSG male mice (8 weeks old) intrasplenically injected with HCT116 tumour cells (1.25 × 10⁵ cells per mouse in sterile PBS: T) or an equal volume of vehicle (sham: S) and administered ACVR2B/Fc (A) (n = 5–10). Data are expressed as mean ± SD. Significance of the differences: *P < 0.05, ****P < 0.0001 vs. S; ^## P < 0.01 vs. T.

Figure 4

ACVR2B/Fc prevents the changes in markers of anabolism and catabolism in mHCT116 hosts. Representative western blotting and quantification (expressed as fold change vs. S) for (A) phospho‐Stat3, Stat3, (B) phospho‐AKT, AKT, (C) phospho‐ERK1/2, ERK1/2, (D) phospho‐p38, p38, (E) ubiquitin and tubulin from quadriceps muscle in NSG male mice (8 weeks old) intrasplenically injected with HCT116 tumour cells (1.25 × 10⁵ cells per mouse in sterile PBS: T) or an equal volume of vehicle (sham: S) and administered ACVR2B/Fc (A) (n = 4–10). Gene expression levels for (F) Murf1 and (G) Atrogin‐1 (normalized to TBP) (n = 5–10). Data are expressed as mean ± SD. Significance of the differences: *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 vs. S; ^# P < 0.05, ^#### P < 0.0001 vs. T.

Figure 5

ACVR2B/Fc preserves oxidative metabolism while unaltering mitochondrial proteins in mHCT116 tumour hosts. (A) Representative western blotting and quantification (expressed as fold change vs. S) for OPA1, PGC1α, PGC1β, Mitofusin‐2, DRP1, VDAC, cytochrome‐C, CoxIV, and tubulin from quadriceps muscle in NSG male mice (8 weeks old) intrasplenically injected with HCT116 tumour cells (1.25 × 10⁵ cells per mouse in sterile PBS: T) or an equal volume of vehicle (sham: S) and administered ACVR2B/Fc (A) (n = 4–10). (B) Enzymatic activity for pyruvate dehydrogenase (PDH) and (C) succinate dehydrogenase (SDH) from quadriceps muscles and (D and E) SDH staining and quantification on tibialis anterior muscles (n = 4–6). Images were captured at a magnification of ×20. Scale bars: 100 μm. Data are expressed as means ± SD. Significance of the differences: *P < 0.05, **P < 0.01, ***P < 0.001 vs. S; ^# P < 0.05 vs. T.

Figure 6

ACVR2B/Fc preserves cancellous bone in mHCT116 hosts. Representative three‐dimensional rendering of μCT scanned images and quantification of bone volume fraction (A) (BV/TV), (B) trabecular thickness (Tb.Th), (C) trabecular separation (Tb.Sp), (D) trabecular number (Tb.N), (E) trabecular pattern factor (Tb.Pf), and (F) trabecular connectivity density (Conn.Dn) of femur bones from 8‐week‐old NSG male mice (8 weeks old) intrasplenically injected with HCT116 tumour cells (1.25 × 10⁵ cells per mouse in sterile PBS: T) or an equal volume of vehicle (sham: S) and administered ACVR2B/Fc (A) (n = 4–9). Scale bars: 1 mm. Data are expressed as mean ± SD. Significance of the differences: *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 vs. S; ^# P < 0.05, ^### P < 0.001, ^#### P < 0.0001 vs. T.

Figure 7

ACVR2B/Fc preserves cardiac function in mHCT116 hosts. (A) Heart weights normalized to initial body weight (IBW) in 8‐week‐old NSG male mice (8 weeks old) intrasplenically injected with HCT116 tumour cells (1.25 × 10⁵ cells per mouse in sterile PBS: T) or an equal volume of vehicle (sham: S) and administered ACVR2B/Fc (A) (n = 5–10). (B) Left ventricular (LV) mass, (C) anterior wall thickness (AWT) at systole, (D) AWT at diastole, (E) posterior wall thickness (PWT) at systole, (F) PWT at diastole, (G) ejection fraction percentage (EF%), and (H) fractional shortening percentage (FS%) determined via conscious echocardiography (n = 3–5). Representative images are taken from M‐mode. Data are expressed as mean ± SD. Significance of the differences: *P < 0.05, **P < 0.01, ****P < 0.0001 vs. S; ^## P < 0.01, ^### P < 0.001, ^#### P < 0.0001 vs. T.

Figure 8

ACVR2B/Fc counteracts differentially expressed genes in hearts of mHCT116 hosts. (A) Heatmap of differentially expressed genes between NSG male mice (8 weeks old) intrasplenically injected with HCT116 tumour cells (1.25 × 10⁵ cells per mouse in sterile PBS: T) or an equal volume of vehicle (sham: S) and administered ACVR2B/Fc (A) (n = 3). (B) Top 25 up‐regulated and 25 down‐regulated differentially expressed genes in T vs. S and A + T vs. T. (C) Functional enrichment analysis based on the differentially expressed genes between S and T. (D) Differential expression of genes involved in heart development, heart contraction, and oxidative stress response in T vs. S and A + T vs. T.