C57BL/6J (strain #000664) and C57BL/6J-Apc^Min/J (Apc^Min/+, strain #002020) were originally purchased from The Jackson Laboratory (Bar Harbor, ME, USA). C57BL/6J female mice were bred with C57BL/6J-Apc^Min/J male mice to obtain age- (17-44 weeks of age) and sex-matched littermate controls. C57BL/6J (male and female) and Apc^Min/+ (male and female) mice were weaned and housed together in ventilated cages to minimize environmental and microbiome-related differences. Lewis Lung Carcinoma or LLC (ATCC, Manassas, VA, USA; CRL-1642) cells were implanted (1X 10⁶, 3^rd passage) in the right flank area of C57BL/6J male mice (16-20 weeks of age) age purchased from Jackson Laboratory. Colon 26 adenocarcinoma or C26 cells (donated from Dr. Andrea Bonetto) were implanted (1X 10⁶, 3^rd passage) at 15 weeks of age in the subscapular region of CD2F1 male mice (Charles River Laboratory, Raleigh NC, USA). CD2F1 mice were euthanized two weeks after C26 cell implantation. The breeding protocol and experiments conducted in this study were approved by the Institutional Animal Care and Use Committee (IACUC) of the University of South Carolina. Mice were fed purified AIN-76A diet ad libitum (Bio-Serv, Frenchtown, NJ, USA) and room lighting followed a 12-hour light and 12-hour dark cycle with a temperature average of 22°C and humidity of 50%. Weekly body weight measurements were recorded to assess body weight loss from their peak body weight. Apc^Min/+ mice with less than 5% body weight loss was assigned to the pre-cachectic group, while those with body weight loss exceeding 10% were allocated to the severely cachectic group. Both groups, along with C57BL6/J littermate control mice, were euthanized with an overdose of isoflurane once the desired body weight loss was reached in Apc^Min/+ mice. Blood and tissue samples were collected for further analysis. LLC-tumor bearing mice were euthanized twenty-five days post-implantation and stratified based on body weight loss after tumor excision. Meanwhile, C26-tumor bearing mice were euthanized two weeks after tumor implantation and exhibited >10% body weight loss after tumor removal.

Behavioral and metabolic phenotyping of C57BL/6J (n=9) and Apc^Min/+ pre-cachectic (n=7) mice were performed between twelve and fifteen weeks of age for one week using the Promethion System (Sable System International, Las Vegas, NV, USA) as previously described (40). Severely cachectic Apc^Min/+ mice were not included in this assessment as survival was not an end point in this experiment. Food and water consumption, body mass, locomotor activity, energy expenditure, and sleeping patterns during the light and dark cycle were collected and analyzed. Metabolic parameters were analyzed using the lean body mass (Dual-Energy X-ray Absorptiometry) as a covariate.

Forelimb strength and neuromuscular function of C57BL/6J (n=9) and Apc^Min/+ pre-cachectic (n=7) mice were determined using grip strength (Bioseb, Pinellas Park, FL, USA) and rotarod (Panlab, Harvard Apparatus, Holliston, MA) assessments. For grip strength, mice were placed in a grid and their tail was gently pulled backward until they lost their grip. The average peak force from five trials was recorded and graphed. In the neuromuscular function test, mice were habituated to the rotarod, which is an elevated rotating rod. On the test day, mice were placed on the rod and subjected to a ramping protocol (0-25 rpm, 0.02 x g) over a period of 120 sec or until they fell from the rod. This protocol was repeated three times with a 2-minute waiting interval. The best performance time was recorded and graphed.

Blood collected during euthanasia was analyzed using the VetScan Hm5 (Abaxis, Union City, CA, USA) for hematology assessment. The complete blood count and 3-part differential were utilized to measure various parameters, including white blood cells (WBC), lymphocytes (LYM), monocytes (MON), neutrophils (NEU), red blood cells (RBC), Hemoglobin (HGB), hematocrit (HCT), mean corpuscular volume (MCV), mean corpuscular hemoglobin (MCH), red cell distribution width (RDW), platelets (PLT), mean platelet volume (MPV), and platelet distribution width (PDWs).

Small and large intestines were dissected and rinsed in PBS. Polyp numbers were counted under a dissection microscope after euthanasia.

The gastrocnemius muscle of Apc^Min/+mice and C57BL6/J littermate controls was homogenized in Trizol for RNA extraction. The isolated RNA was quantified using a nanodrop, and the iScript cDNA synthesis kit was used to transcribe and analyze the expression of specific genes, including: Ly6g, F4/80, CD206/mrc1, IL-10, IL-1β, IL-4, IL-6, TNFα, TGFβ1, TGFβ2, TGFβ3, SMAD2, SMAD3, and fibrinogen. Reference genes (Nono, Gusb, Hmbs, RplpO, 18S, H2afv) were also run, and stable genes (Nono, Hmbs, RplpO, H2afv) were selected as the normalization factors for each analysis using Qbase + software (Biogazelle, Ghent, Belgium).

The data were analyzed using Prism 8 (GraphPad Software, La Jolla, CA). One-way ANOVA or mixed-effects model tests were performed, followed by Tukey’s multiple comparison test or Sidak’s multiple comparisons test, respectively. Unpaired two-tailed t-tests were used to compare the number of polyps between pre-cachectic and severely cachectic Apc^Min/+ mice, as well as functional tests (grip strength and rotarod test) between C57BL/6J and Apc^Min/+ pre-cachectic mice, and platelet count between non-cancer and tumor-bearing mice. If Bartlett’s test for equal variance failed, the data were log-transformed and reanalyzed. Metabolic parameters were analyzed using an ANCOVA (MMPC National Mouse Metabolic Phenotyping Center) (41) with lean body mass as a covariate. The data are presented as means ± standard error, and a significance level of p<0.05 was used.

To confirm the progression of cancer cachexia, we measured body weight, polyp number, as well as mass of hindlimb muscles, spleen, liver, and fat. As expected, severely cachectic Apc^Min/+ mice (with an average body weight loss of twenty-four percent) weighed significantly less than Apc^Min/+ pre-cachectic (with an average body weight loss of four percent) and C57BL/6J mice at euthanasia ( Figure 1A , p=0.0425 and p<0.0001). The mass of the soleus, plantaris, gastrocnemius, extensor digitorum longus (EDL), tibialis anterior, and quadriceps muscles was also reduced in severely cachectic Apc^Min/+ mice compared to C57BL/6J mice ( Figures 1B–G , p=0.0022, p<0.0001, p<0.0001, p=0.0004, p=0.0028, and p<0.0001 respectively). Moreover, significant differences in the mass of the plantaris, gastrocnemius, EDL, and quadriceps were observed between pre-cachectic Apc^Min/+ and severely cachectic Apc^Min/+ mice (p=0.0228, p<0.0001, p=0.0079, and p=0.0002 respectively). Splenomegaly was observed in both pre-cachectic Apc^Min/+ and severely cachectic Apc^Min/+ mice when compared to C57BL/6J mice ( Table 1 , p=0.0029 and p<0.0001). No significant difference in liver weight was observed between the groups. There was no difference in fat pad mass between C57BL/6J mice and pre-cachectic Apc^Min/+ mice. However, a significant reduction in gonadal and retroperitoneal fat masses were observed between severely cachectic Apc^Min/+ mice and C57BL/6J mice (p=0.0010 and p=0.0032). No significant differences were observed in fat pad mass and polyp number between pre-cachectic Apc^Min/+ and severely cachectic Apc^Min/+ mice. Similarly, previous studies have shown no changes in polyp size relative to cachectic stage (42–46).

To characterize the hematological profile during cancer cachexia progression, we measured the number of white blood cells, red blood cells, and platelets in the blood of pre-cachectic Apc^Min/+, severely cachectic Apc^Min/+, and C57BL/6J mice. The total number of white blood cells did not differ between groups ( Figure 2A ). However, severely cachectic Apc^Min/+ mice showed a significant reduction in lymphocyte number compared to C57BL/6J mice (p=0.0071), but not compared to pre-cachectic Apc^Min/+ mice ( Figure 2B ). There was no difference in monocytes between groups ( Figure 2C ). A significant increase was observed in the number of neutrophils in severely cachectic Apc^Min/+ mice compared to C57BL/6J mice (p=0.0082), but not compared to pre-cachectic Apc^Min/+ mice ( Figure 2D ). When the neutrophil:lymphocyte was calculated, a significant increase was observed in severely cachectic Apc^Min/+ mice compared to C57BL/6J mice ( Figure 2E , p=0.0003). There was a trend towards an increase neutrophil:lymphocyte in pre-cachectic Apc^Min/+ compared to C57BL/6J mice (p=0.0550).

Anemia is one of the hallmarks of cancer cachexia (47). Therefore, we proceeded to identify if anemia precedes muscle mass loss. A reduction in the number of red blood cells, hemoglobin, and hematocrit was observed in both pre-cachectic Apc^Min/+ and severely cachectic Apc^Min/+ mice compared to C57BL/6J mice ( Figures 3A–C , p<0.0001, p<0.0001, and p<0.0001). Additionally, an increase in red blood cell indices was observed between Apc^Min/+ and C57BL/6J mice ( Figures 3D–G ). Mean corpuscular volume, or the size of the red blood cell, was significantly increased in pre-cachectic Apc^Min/+ and severely cachectic Apc^Min/+ mice compared to C57BL/6J mice ( Figure 3D , p=0.0473 and p<0.0001). Mean corpuscular hemoglobin, or the amount of hemoglobin per red blood cell, was significantly elevated in pre-cachectic Apc^Min/+ and severely cachectic Apc^Min/+ mice compared to C57BL/6J mice ( Figure 3E , p=0.0085 and P<0.0001). Additionally, a difference in mean corpuscular hemoglobin was observed between pre-cachectic and severely cachectic Apc^Min/+ mice (p=0.0127). An increase in the average concentration of hemoglobin per red blood cell, or mean corpuscular hemoglobin concentration, was observed in severely cachectic Apc^Min/+ mice when compared to C57BL/6J mice (p=0.0013) ( Figure 3F ). Additionally, increased variation in the size of red blood cells, or red cell distribution width, was observed in pre-cachectic Apc^Min/+ and severely cachectic Apc^Min/+ mice when compared to C57BL/6J mice (p<0.0001 and p<0.0001) ( Figure 3G ).

Since a very high platelet count has been associated with increased risk of developing cachexia-related cancers (32) and correlated with one-year overall survival in patients with cancer cachexia (17, 18), we sought to measure platelets at different stages of cancer cachexia. We observed that platelet count in pre-cachectic Apc^Min/+ mice was significantly elevated compared to C57BL/6J mice ( Figure 4A , p=0.0234). There was a trend towards an increased platelet count in severely cachectic Apc^Min/+ mice (p=0.0551). In terms of the size of platelets, severely cachectic Apc^Min/+ mice showed an increase in mean platelet volume compared to C57BL/6J mice ( Figure 4B , p=0.0142). Elevated mean platelet volume is indicative of platelet activation (48). There was no variation in platelet distribution observed between groups ( Figure 4C ).

To better understand the behavioral and physiological profile of pre-cachectic Apc^Min/+ mice we assessed ambulatory activity, sleep patterns, forelimb strength, and neuromuscular function. Pre-cachectic Apc^Min/+ mice showed a significant decrease in directed locomotor activity during the night cycle (active cycle) compared to C57BL/6J mice ( Figure 5A ). However, no change in stereotypical behavior (grooming, rearing, etc.) was observed between pre-cachectic Apc^Min/+ and C57BL/6J mice ( Figure 5B ). There was a significant main effect of time for increased sleep behavior during the night cycle in pre-cachectic Apc^Min/+ mice ( Figure 5C ). No changes were observed in grip strength and neuromuscular function between pre-cachectic Apc^Min/+ and C57BL/6J mice ( Figures 5D, E , p=0.1764 and p=0.8019).

Since previous studies in mice have indicated that physical inactivity and systemic metabolic dysfunction are associated with accelerated cachexia (49), we sought to investigate the impact of pre-cachexia on indirect calorimetry outcomes. We observed that pre-cachectic Apc^Min/+ mice significantly decreased total and resting energy expenditure, oxygen consumption, and carbon dioxide production compared to C57BL/6J mice ( Table 2 ). No changes were observed in mean respiratory exchange ratio, energy intake, and mean sleep. However, a significant decrease was observed in mean distance traveled in pre-cachectic Apc^Min/+ mice compared to C57BL/6J mice.

To determine the possible involvement of immune cells in the progression of cancer cachexia, we measured the gene expression of nonexclusive conventional immune cell markers in the skeletal muscle of non-cancer C57BL/6J mice, as well as pre-cachectic and severely cachectic Apc^Min/+ mice. We observed a significant increase in the expression of Ly6g (neutrophil marker) in severely cachectic Apc^Min/+ mice compared to pre-cachectic Apc^Min/+ mice ( Figure 6A , p=0.0178), but not C57BL/6J mice. Interestingly, severely cachectic Apc^Min/+ mice showed an increase in CD206 ( Figure 6C ) (mannose receptor c-type 1, M2 macrophage marker, p=0.0464) and IL-10 (Figure 6D) (anti-inflammatory cytokine related to M2 macrophages, p=0.0216), but no change in F4/80 (a general marker of macrophages) gene expression compared to C57BL/6J mice ( Figure 6B , p=0.1450). No difference in the gene expression of IL-4 was observed between the groups ( Figure 6E , p=0.8874). We observed an increase in the gene expression of the pro-inflammatory cytokine IL-1β in pre-cachectic Apc^Min/+ mice compared to C57BL/6J mice (p=0.0021) and severely cachectic Apc^Min/+ mice ( Figure 6F , p=0.0387). However, no changes in IL-6 and TNFα were observed between the groups ( Figures 6G, H , p=0.6966, and p=0.6415).

Since platelets can promote the release of various growth factors linked to fibrosis (26), and collagen deposition has been observed in cancer cachexia (27, 50), we evaluated whether the TGFβ signalling pathway was modulated in the pre-cachectic and severely cachectic stages. No difference in the gene expression of TGFβ-1 was observed between the groups ( Figure 7A , p=0.1816). However, a significant increase in the gene expression of TGFβ-2 and TGFβ-3 was observed in the skeletal muscle of pre-cachectic Apc^Min/+ mice compared to C57BL/6J mice ( Figures 7B, C , p=0.0454 and p=0.0021). Additionally, severely cachectic Apc^Min/+ mice exhibited a significant decrease in TGFβ-3 compared to pre-cachectic Apc^Min/+ mice (p=0.0056). We then sought to examine the downstream canonical pathways of TGFβ. No change was observed in SMAD2 between the groups ( Figure 7D , p=0.1541). However, an increase in SMAD3 expression was observed in pre- cachectic Apc^Min/+ mice compared to severely cachectic Apc^Min/+ mice ( Figure 7E , p=0.0475). Finally, we investigated the expression of fibrinogen gamma chain, a molecule essential for platelet aggregation (51), whose synthesis has been elevated in patients with cancer and cachexia (52, 53). A significant increase in the expression of fibrinogen gamma chain was observed in pre-cachectic Apc^Min/+ mice (p=0.0312) compared to severely cachectic Apc^Min/+ mice ( Figure 7F ).

To confirm whether the increase in platelet number or platelet activation occurs in other models of muscle wasting, we examined the platelet profile in the Lewis lung carcinoma (LLC) and colon 26 adenocarcinoma (C26) models of cancer cachexia. Tumor weight in LLC-tumor bearing mice with less than five percent body weight loss (2.353 ± 0.6105 grams) and more than ten percent body weight loss (1.771 ± 0.5006 grams) were not significantly different (p=0.5129, data not shown). No difference was observed in the platelet count of LLC-tumor bearing mice ( Figure 8A , p=0.0756). However, a significant increase in mean platelet volume was observed in severely cachectic LLC-tumor bearing mice compared to non-cancer C67BL/6J mice and LLC-tumor bearing mice with less than five percent body weight loss ( Figure 8B , p=0.0079, p=0.0270). No significant difference was observed in platelet distribution width between the groups ( Figure 8C , p=0.0983). In the case of C26-tumor bearing mice, an increase in platelet count was observed compared to non-cancer CD2F1 mice ( Figure 9A , p<0.0001). There were no changes observed in mean platelet volume and platelet distribution width between C26-tumor bearing mice and non-cancer CD2F1 mice ( Figures 9B, C , p=0.4020 and p=0.3373).