Histidine decarboxylase inhibition attenuates cancer-associated muscle wasting

Cachexia is a devastating and multifactorial syndrome that increases mortality and decreases quality of life of cancer patients (Fearon et al., 2012). This syndrome is characterized by significant loss in weight or body mass index, which includes but is not limited to lean mass/muscle loss. At 80%, pancreatic cancer patients have the highest incidence of cachexia relative to patients with other cancer types. One third of pancreatic cancer patients die due to cachexia-associated complications, such as muscle and adipose tissue wasting, respiratory failure, or cardiac arrest (Tan et al., 2014). Moreover, cachexia worsens prognosis and limits the efficacy of many cancer-directed chemotherapeutic regimens (Bachmann et al., 2008; Dewys et al., 1980). Although there are many ongoing efforts to develop and test wasting-directed interventions, there are presently no FDA-approved therapies to treat cachexia in the clinic (Crawford, 2019).

A major obstacle hampering effective clinical interventions is a disconnect between promising data in preclinical murine models and disappointing efficacy in clinical trials (Baracos, 2018). One contributing factor to this disconnect may be the degree to which preclinical models recapitulate key aspects of the human condition. Indeed, more precise modeling may reveal targets that more readily translate into effective clinical interventions. In our previously published work, we developed and utilized an age-appropriate murine model of pancreatic cancer to investigate transcriptomic signatures of young and aged tumor-bearing skeletal muscle (Dasgupta et al., 2022). Our comparisons revealed novel targets that are currently being developed to target both muscle wasting and tumor progression in pancreatic cancer models (Dasgupta et al., 2022).

Beyond gene expression alterations, positron emission tomography imaging shows that pancreatic tumors exhibit elevated glucose uptake—a phenomenon that profoundly impacts host metabolism (Deberardinis et al., 2008; Serrano et al., 2010). Examples of tumor-associated host metabolic rewiring include depletion of glycogen reserves by gluconeogenesis in the liver, skeletal muscle breakdown to provide amino acids, and enhanced lipolysis of adipose tissue to liberate free fatty acids (Agustsson et al., 2007; Porporato, 2016). These examples highlight the degree to which cancer induces a systemic metabolic imbalance. Given notable changes in peripheral tissue metabolism during normal aging, we queried metabolic differences in young and aged cachectic muscles as a starting point to investigate metabolic drivers of cancer-associated muscle wasting.

Several wasting-associated metabolic alterations have been reported in preclinical murine models and include dysregulated amino acid metabolism, excessive fatty acid oxidation, lactate accumulation, and altered glucose homeostasis (Chen et al., 2024; Eley et al., 2007; Fukawa et al., 2016; Liu et al., 2024a; Raun et al., 2022). However, trials seeking to systemically modulate these metabolic pathways in cachectic cancer patients have not achieved widespread clinical success thus far (Dobs et al., 2013; Lundholm et al., 2007). We therefore sought to investigate novel metabolic targets that may have been overlooked in prior cachexia studies. Our preliminary data revealed histamine elevation in cachectic murine and human muscle. Whereas the source of histamine has been historically attributed to mast cells, our study reveals the contribution of mast cell–independent generation of histamine by induction of histidine carboxylase in skeletal muscle. Moreover, our data implicate histamine as a targetable mediator of cancer-associated muscle atrophy. Notably, antihistamines have demonstrated efficacy in improving survival outcomes in cancer patients via a mechanism of action that is not entirely clear. Our study provides novel mechanistic insights into histamine metabolism in wasting muscle as well as proof-of-concept data in support of histamine-targeting strategies to preserve muscle mass in cachectic patients. This study further highlights the benefits of refining cancer cachexia models to reduce confounding physiological artifacts (e.g., skeletal muscle age), thus enabling more precise identification of bona fide cachexia/muscle-wasting factors.

Our previous study characterized the cachectic phenotype and survival of young (8 wk) and aged (78 wk) control and T4 LSL-Kras^G12D/+; LSL-Trp53^R172H/+;Pdx-1-Cre (KPC) tumor-bearing mice (Dasgupta et al., 2022). Using samples from that study, we performed untargeted metabolomics analyses on frozen skeletal muscle collected from young saline (control), aged saline (control), young tumor/KPC-bearing, and aged tumor/KPC-bearing mice. Principal component analyses (PCA) revealed distinct clustering of young control versus aged control samples (Fig. S1 A), corroborating previous reports of age-associated alterations in skeletal muscle metabolism (Uchitomi et al., 2019; Zhou et al., 2021). Enrichment analyses of differentially abundant metabolites (DAMs) between these cohorts revealed significant differences in multiple pathways corresponding to amino acid catabolism/metabolism (Fig. S1, B and C). We also observed differences in a cluster of lipid metabolites as well as metabolites associated with glycolysis, glycogen metabolism, purine/pyrimidine, and primary bile acid metabolism (Fig. S1, D–H and Fig. S2, A–G). We next compared the young/aged saline controls to their tumor-bearing counterparts. A PCA plot of all groups revealed four distinct clusters (Fig. 1 A). We analyzed significant (P < 0.05, ANOVA contrast) DAMs across these groups and found that while there was significant overlap between aged KPC versus saline and young KPC versus saline DAMs, there appeared to be a DAM subset exclusively upregulated in an aged context (Fig. 1 B). Enrichment analysis of DAMs in young/aged KPC muscles versus their saline counterparts implicated multiple amino acid metabolism pathways as among the most highly dysregulated (Fig. 1 C). We then narrowed down pathways/metabolites of interest by focusing on DAMs altered as a function of tumor status as well as age. 11 metabolites (Fig. S3, A–J and Fig. 1 D) met these criteria, and we chose to focus on histamine for the following reasons. First, a recent report found histidine (a precursor metabolite that is converted to histamine by the enzyme histidine decarboxylase [HDC]) to be significantly negatively correlated with levels of IL-6, which is a known cachexia driver (Sirniö et al., 2019). Second, analysis of a previously published targeted metabolomics dataset (Metabolomics Workbench Project ID PR000680) revealed lower serum concentrations of histidine in weight-losing male cancer patients compared to weight-stable patients (Fig. 1 E). Third, independent validation of histamine abundance in frozen muscle samples confirmed increased histamine accumulation in aged KPC tumor-bearing mice (Fig. 1 F). Finally, exogenous addition of histamine to C2C12 myotubes resulted in dose-dependent increases in expression of the atrophy-related genes Trim63 and Fbxo32 (Fig. 1, G and H). These results indicated that histamine metabolism may be playing an important role in the etiology of cancer-associated muscle wasting.

The most common source of histamine is activated mast cells. We therefore stained muscles from control and tumor-bearing cohorts with toluidine blue to detect heparin-containing granules known to be enriched in mast cells (Fig. 1 I). While we observed an increase in toluidine blue staining in both cachectic cohorts compared with their respective age-matched controls, we did not observe a statistical difference between the young and aged tumor-bearing cohorts (Fig. 1 J). These data suggest that mast cell abundance alone is not sufficient to explain the specific increase in histamine abundance in aged muscle from tumor-bearing mice. We then measured the activity of the enzyme HDC, which is responsible for the conversion of histidine into histamine. We observed elevated HDC activity exclusively in muscles from aged tumor-bearing mice (Fig. 1 K). Separately, muscle samples obtained from pancreatic cancer patients were queried to assess histamine abundance. We observed a significant increase in histamine levels in cachectic patients compared with non-cachectic controls (Fig. 1 L). Together, these data suggest that increased HDC activity contributes to the elevated histamine content in aged wasting muscle.

We next utilized an inhibitor of HDC activity, (S)-α-fluoromethylhistidine (FMH) HCl, to explore the role of HDC in mediating muscle atrophy programs (Fig. 2 A). Leveraging a well-established in vitro model of cancer-induced myotube atrophy (Fig. 2 B), we added conditioned media (CM) from either control pancreas cell line (HPNE) or pancreatic cancer cell lines (T4 and T3 KPC) to differentiated C2C12 myotubes. We observed that CM-induced myotube atrophy (quantified via measuring myotube width) was decreased upon treatment with FMH (Fig. 2, C and D). Further, FMH reduced CM-associated increases in the mRNA expression of Trim63 and Fbxo32 (Fig. 2, E and F). We also observed that myotubes generated histamine upon stimulation with T4 KPC CM and that FMH treatment blocked this effect (Fig. 2 G). Moreover, FMH was able to reduce CM-associated upregulation of HDC activity in myotubes (Fig. 2 H). Finally, histamine quantification revealed increased histamine abundance in cancer CM (T4 and T3 KPC) as compared with control CM (HPNE) (Fig. S3 K). Together, these data suggest that both muscle and tumors contribute to net histamine accumulation in wasting muscle and that these contributions are separate from any mast cell involvement.

We next queried the potential of HDC inhibition as a strategy to limit cancer-associated muscle wasting in vivo. T4 KPC cells were orthotopically implanted into the pancreas of aged mice and were treated with either vehicle or FMH every 3 days, starting from day 7 post-cell engraftment (Fig. 3 A). We observed preservation of muscle but not of fat mass in tumor-bearing mice treated with FMH as compared with vehicle treatment (Fig. 3, B and C). Notably, there was no significant difference in tumor weights upon FMH treatment (Fig. 3 D). Tibialis anterior (TA), gastrocnemius (GR), and heart weights were all preserved in the tumor + FMH cohort (Fig. 3, E–G). Corresponding to muscle mass preservation, tumor-induced expression of the atrogenes Trim63 and Fbxo32 was reduced upon FMH treatment (Fig. 3, H and I). Accordingly, TA cross-sectional area (CSA) and minimum feret diameter measurements were larger in FMH-treated tumor-bearing mice compared with vehicle-treated mice (Fig. 3, J–L), and no discernable myofiber-type switching was observed (Fig. 3 M). Finally, FMH treatment decreased levels of histamine and HDC activity in muscles isolated from tumor-bearing mice (Fig. 3, N and O).

We next performed reverse phase protein array analyses on myotubes treated with histamine, HPNE CM, T4 KPC CM, or T4 KPC CM + FMH to gain insights into histamine-dependent atrophy induction. We compared significant differentially abundant proteins between (1) control versus histamine, (2) HPNE CM versus T4 KPC CM, and (3) T4 KPC CM versus T4 KPC CM + FMH groups. There were 13 commonly altered proteins across these three comparisons (Fig. 4 A). Out of these 13, we focused on proteins altered by histamine and T4 CM treatment in an FMH-dependent manner. Using these criteria, we found the pro-survival factor BCL2A1 (Vogler, 2012) down in histamine and T4 CM groups while up/remaining elevated in response to FMH (Fig. 4, B and C). We observed a similar, though inverse, trend for the ER stress marker calnexin (Ryan et al., 2016). Given (1) the role of calnexin as a calcium-binding protein (Wada et al., 1991) and (2) transcriptomics-based evidence indicating deregulated calcium-related signaling between aged saline-injected and tumor-bearing mice (Fig. 4 D), we assessed calcium levels in our treated myotubes. We observed increased basal calcium levels in myotubes treated with histamine and T4 KPC CM, which decreased upon FMH treatment (Fig. 4, E and F). Histamine treatment reduced the myotube response to KCl (a depolarizing agent) as determined using Cal-590 (Fig. 4, G and H) as well as calcium 6 (Fig. 4 I)-based measurements. We posited that these effects may result from histamine-associated reductions in voltage-gated channel activity and further hypothesized that increased calcium levels were inducing myotube atrophy via calcineurin (Shimizu et al., 2017). We therefore treated myotubes with histamine +/− FK-506 (a calcineurin inhibitor). We observed increased expression of atrogenes Trim63, Fbxo32, Foxo1, and Foxo3 upon histamine treatment—increases that were reduced with concomitant FK-506 treatment (Fig. 4 J). Moreover, while T4 KPC CM increased both PKA and PKC activity in myotubes, histamine only induced PKA activity (Fig. 4, K and L). This suggests that histamine, at least in skeletal muscle/myotubes, likely signals via H2R (Flamand et al., 2004), whereas other wasting factors present in T4 KPC CM can signal via both H1R and H2R. Finally, our previous data found increased levels of IL-3 in the serum of aged tumor-bearing mice compared with saline-injected controls (Dasgupta et al., 2022). Intriguingly, IL-3 is known to induce HDC expression in bone marrow cells (Dy et al., 1993). Notably, exogenous addition of IL-3 to C2C12 myotubes was sufficient to (1) increase expression of HDC, (2) decrease myotube width, and (3) increase expression of atrogenes (Fig. 4 M and Fig. S3, L–N).

We present evidence implicating an HDC/histamine/calcium signaling axis as an important contributor to cancer-associated muscle atrophy (Fig. 4 N). This aligns with a previous report describing increased calcium-associated proteolysis in cachectic muscle (Costelli et al., 2001). Dysregulation of calcium homeostasis is also linked to chemotherapy-induced cachexia (Conte et al., 2017). That said, given the ubiquitous nature of calcium-associated biological pathways/processes, directly targeting calcium might lead to significant off-target cytotoxicity. Therefore, a potentially more pragmatic means by which to target this signaling axis could be to target the upstream factors responsible for atrophy-associated histamine/calcium signaling induction. For example, several cytokines (IL-3, IL-18, and IL-12) are implicated in HDC upregulation in bone marrow, spleen, and liver (Dy et al., 1993; Yamaguchi et al., 2000). In our cohort of young/aged mice, we found IL-3 to be uniquely upregulated in aged tumor-bearing mice, and we demonstrated its ability to induce HDC expression, elevate histamine, and stimulate myotube atrophy. Further investigation into upstream regulators of HDC/histamine (such as IL-3) could permit more precise targeting of this signaling axis in wasting muscle.

Among its many roles as a pleotropic metabolite, histamine is a signaling molecule linked to neuronal signaling in cachexia (Zwickl et al., 2019) as well as to exercise-induced fatigue/weakness and pain (Ely et al., 2017; Obara et al., 2020). Other examples highlighting the role of histamine/HDC signaling in skeletal muscle include observations that HDC is induced after prolonged walking (Endo et al., 1998) as well as a study linking histamine receptor biology to integrated training responses in humans (Van der Stede et al., 2021). Our study, to the best of our knowledge, is the first implicating HDC/histamine in the etiology of cancer-associated muscle wasting. In accordance with studies demonstrating the ability of histamine to alter calcium flux in multiple cell types (Berra-Romani et al., 2022; Kotlikoff et al., 1987), we found a link between HDC/histamine and calcium signaling—the latter having established relationships to cancer and atrophy pathways. For example, in a model of bone metastasis, investigators observed increased oxidation of the calcium ion channel RyR1, which led to altered calcium homeostasis and subsequent functional impairments (Waning et al., 2015). Further, calcium levels are known to influence multiple protein degradation pathways such as calpain, ubiquitin–proteasome, autophagy, and mitophagy pathways (Costelli et al., 2002; Huang and Forsberg, 1998; Hyatt and Powers, 2020; Shimizu et al., 2017; Smith et al., 2008). These studies along with our data depicting increased calcium flux in histamine-treated myotubes led us to explore histamine-calcium signaling in more detail. In our model, we found a role for calcineurin-FoxO signaling downstream of histamine/HDC. The extent to which this signaling axis is activated/involved in other models of pancreatic cancer cachexia (or more broadly in cachexia associated with other cancer types) remains to be seen and is an important next step for this work.

Interestingly, histamine (1) is positively correlated with tumor progression (Masini et al., 2005), (2) is known to confer resistance to immunotherapies, and (3) antihistamine treatment increased survival outcomes in lung cancer and melanoma patients undergoing immunotherapy (Li et al., 2022). These and other studies underscore the potential value of targeting histamine to improve multiple cancer outcomes. Notably, the antihistamine loratidine was recently shown to increase survival in lung cancer patients (Liu et al., 2024b). In vitro and in vivo studies demonstrated increased tumor cell apoptosis and pyroptosis upon loratidine treatment (Liu et al., 2024b). Considered alongside our observations, these data suggest that future work should investigate the efficacy of antihistamines in cachectic cancer patients to improve both cancer and wasting-associated outcomes.

Recent studies in preclinical models have reported that blocking muscle wasting not only increases survival but also slows down tumor progression (Neyroud et al., 2023). This implies that muscle loss is not only a symptom of the disease but also a potential disease driver. This observation, alongside decades of work implicating cachexia as a negative predictor of multiple cancer outcomes, underscores the urgency to discover effective therapeutic targets. Utilizing models that faithfully recapitulate the human condition is one impactful way of approaching this issue. The field is constantly striving to utilize and generate models in which targets of cachexia can be determined with clarity and with minimal background artifacts (Arneson-Wissink et al., 2020; Dasgupta et al., 2022; Gicquel et al., 2024; Snoke et al., 2024, Preprint; Suzuki et al., 2020; Talbert et al., 2019). For example, existing preclinical models of cachexia consider the source of the tumor (human and mouse), implantation location (orthotopic, subcutaneous, and tail vein), and tumor type/stage (primary and metastasis) and have revealed several context-specific nuances with respect to cachexia etiology. That said, there remain opportunities for introducing patient-relevant factors such as treatment regimens, psychological/physical stress, and biological age into cachexia models to maximize translational relevance. By considering advanced age in a gold standard orthotopic pancreatic cancer model, we identified histamine as a novel wasting target with significant clinical potential. We acknowledge that a key study limitation is that we did not explore potential sex differences in our aged models. This is notable given reports highlighting sexual dimorphism in cachexia, and we agree that potential sex effects need to be queried in future work. Ultimately, we contend that continuous reflection on and refinement of wasting models is needed to help bridge the gap between preclinical studies and clinical trial outcomes, thus accelerating impactful research with the potential to directly benefit patients with cancer cachexia.

The human pancreas cell line, HPNE, was a gift from Dr. Martin Fernandez-Zapico (Mayo Clinic, Rochester, MN, USA). Pancreatic cancer cell lines (T4 KPC and T3 KPC) were derived from KPC mice as described previously (Boj et al., 2015) and were a gift from Dr. Gina Razidlo (Mayo Clinic, Rochester, MN, USA). C2C12 myoblasts were purchased from ATCC. All cell lines were cultured in DMEM (Gibco) with 10% FBS, 100 IU/ml penicillin, and 100 µg/ml streptomycin and incubated at 37° in a humidified incubator with 5% CO₂. C2C12 myoblasts were cultured in this media until confluent and then differentiated in DMEM with 2% horse serum (HS) and 1 µg/ml insulin for 72 h, as previously described (Dasgupta et al., 2022). (S)-FMH HCl was purchased from Medkoo biosciences (cat. #525854). FK-506 was purchased from Cayman Chemical (cat. #10007965). IL-3 was purchased from Peprotech (cat. # 213-13-10UG).

All animal experiments performed in this study were approved by the Mayo Clinic Institutional Animal Care and Use Committee and performed at Mayo Clinic. C57/BL6J mice were obtained from Jackson Laboratories.

8- and 78-wk male mice were orthotopically implanted with 0.5 × 10⁴ T4 KPC cells as previously described (Dasgupta et al., 2022). Sham surgeries were performed on a separate set of 8- and 78-wk aged male mice. After necropsy, the GR muscle from each mouse was flash frozen and processed for metabolomics analyses.

78-wk male mice were orthotopically implanted with 0.5 × 10⁴ T4 KPC cells. Mice were randomized, and 7 days after implantation, one cohort (n = 10) was treated with FMH (25 mg/kg) and the other with vehicle control (n = 10) every 3 days until endpoint. Study endpoints were low body score and/or tumor volume above 10% body weight. Study animals were weighed, had their tumors palpated and measured with calipers, and subjected to EchoMRI-based body composition analysis weekly. At study endpoints, tumor tissue, multiple skeletal muscles, and heart tissues were flash frozen in liquid nitrogen, formalin fixed, or incubated in 30% sucrose for further analysis.

A minimum of 40 mg of GR muscle was sent to Metabolon Inc. for untargeted metabolomic analysis. According to their standard pipeline, Metabolon used an ultrahigh performance liquid chromatography-tandem mass spectroscopy unit. Statistical analyses from GR metabolomics were performed by Metabolon using ANOVA contrast. Metabolites were uploaded to MetaboAnalyst 5.0 (Pang et al., 2021) to log transform and Pareto scale data to produce a 3D PCA plot.

HPNE and KPC cell lines were cultured in DMEM with 10% FBS as previously described (Dasgupta et al., 2022). For CM preparation, cells were seeded and allowed to reach 70% confluency. Then cells were washed with 1X PBS, and serum-free media was added for 24 h. Media was collected and centrifuged at 3,000 rpm for 10 min, and supernatant was collected in a fresh tube. CM was either used immediately or stored at −80° for future use. Equal number of cells were utilized to prepare batches of CM and then reconstituted with 2% HS.

Total RNA was extracted from C2C12 myotubes and tissues as previously described by using TRIzol reagent (Invitrogen) (Dasgupta et al., 2023). RNA was then purified utilizing RNAeasy columns as per the manufacturers protocols. Next, the purified RNA was utilized for qRT-PCR or RNA sequencing (RNA-seq). cDNA synthesis kit (Applied Biosystems) was utilized to obtain cDNA according to the manufacturer’s protocol. qRT-PCR was performed using SYBR Green Master Mix (BioRad). Tubulin was used as internal controls. Relative gene expression analysis was performed using delta–delta Ct method, as previously described (Dasgupta et al., 2022).

C2C12 myotubes were washed with PBS and fixed in 6-well plates using 4% paraformaldehyde solution. Myotubes were then immunolabeled using anti-myosin heavy chain 1 (clone MF20 from Iowa Hybridoma) as previously described (Schmitt et al., 2022). Alexa Fluor 488 secondary antibody (Invitrogen) was utilized to detect MF20 staining. Nuclei were stained with Hoechst. Images were captured with Hamamatsu ORCA-Flash 4.0 LT CMOS camera using Nikon NIS-elements. Myotube widths were measured/quantified using Fiji/ImageJ (Schneider et al., 2012) software.

Murine muscle tissue cross sections were immunostained to detect laminin as previously described (Dasgupta et al., 2023). Briefly, tissues were placed in a sucrose sink (30%) overnight. Tissues were then frozen in OCT, and sections (8 µm) were cut. These sections were fixed in 4% paraformaldehyde and then incubated with anti-laminin (4HB-2; Sigma-Aldrich). Alexa fluorescent conjugate 488 (Invitrogen) was used as a secondary antibody. Myovision software (Wen et al., 2018) was then used to analyze the myofiber images for minimum feret diameter and CSA. The software mapped and calculated the CSA of fibers between the range of 150 and 5,000 µm³.

TA cross sections were stained to measure the presence of different MHC isoforms, thus myofiber types, as previously described (Dasgupta et al., 2023). Slides were incubated with BA-D5 (type I myofibers, supernatant, 1:100; DSHB), SC-71 (type IIa myofibers, supernatant, 1:100; DSHB), BF-F3 (type IIb myofibers, concentrate, 1:100; DSHB), and laminin (1:250; Sigma-Aldrich) diluted in 1% BSA. Alexa Fluor secondary antibodies (Invitrogen) were diluted 1:250 in 1% BSA: anti-mouse IgG2b-405, anti-mouse IgG1-488, anti-mouse IgM-594, and anti-rabbit IgG-647. ImageJ (Schneider et al., 2012) cell counter was used to manually assess myofiber types for quantification. All TA sections were negative for type I myofibers and were therefore not included in the myofiber-typing quantification graphs.

Cytosolic Ca²⁺ dynamics in C2C12 myotubes were measured using the FLIPR Calcium 6 Assay Kit and a FlexStation 3 plate reader (Molecular Devices) as previously described (Kono et al., 2018). To measure the cytosolic Ca²⁺ response to KCl, C2C12 myotubes were loaded with calcium 6 in DMEM + 2% HS for 2 h. Immediately prior to Ca²⁺ imaging, cells were incubated in HBSS with the following composition: 138 mM NaCl, 5.3 mM KCl, 0.34 mM Na₂HPO₄, 0.44 mM KH₂PO₄, 4.17 mM NaHCO₃, 2 mM CaCl₂, 1 mM MgCl₂, and 5.5 mM glucose. Baseline (F0) fluorescence was measured for a minimum of 10 s. The elevation of calcium 6 intensity in response to KCL addition (30 mM) was monitored, which was normalized to the basal F0, according to the formula ΔF/F0.

We performed in vitro confocal imaging of C2C12 myotubes, loaded with the calcium-sensitive fluorescent dye Calbryte 590 AM in HBSS buffer described above. Myotubes were stimulated with 30 mM KCl. Calbryte 590 data are analyzed from n > 10 myotubes. Confocal images were acquired with a Zeiss LSM 800 confocal imaging system (Carl Zeiss). Imaging was performed using a 561-nm single excitation laser line, and fluorescent emission was collected with 550–620-nm emission slit widths. Images were analyzed using Fiji/ImageJ software.

Histamine quantification was performed using Histamine Assay Kit (ab235630). Levels of histamine were normalized by total protein content of each sample. HDC activity was measured in the cell and tissue lysates by utilizing the Histidine Decarboxylase Activity Kit (#K2082-discontinued; BioVision).

PKA and PKC activity was quantified using PKA and PKC activity assay kits (cat # ab139435 and cat # ab139437; Abcam). Briefly, protein from C2C12 myotubes was extracted according to the manufacturer’s recommendations, enzyme activity was measured, and data normalized to total protein content.

Human muscle specimens from donors who had previously been diagnosed with pancreatic ductal adenocarcinoma were obtained from the UNMC Tissue Bank through the RAP in compliance with Institutional Review Board 091-01-EP. Samples were categorized as cachectic or non-cachectic according to UNMC pathologists. To ensure specimen quality, organs were harvested within 3 h after mortem, and the specimens were flash frozen in liquid nitrogen or placed in formalin for immediate fixation.

Prism GraphPad 10 was utilized to perform statistical tests. Individual tests have been mentioned in figure legends. Statistical comparisons of metabolite abundance were performed by Metabolon using ANOVA contrasts. RNA-seq accession numbers are as previously reported (Dasgupta et al., 2022).

Additional figures and text are available in “Supplemental material.” Figs. S1 and S2 highlight the metabolic differences in the muscles of healthy young and aged mice. Fig. S3 shows (1) the age- and pathology-associated muscle metabolites, (2) histamine abundance in CM, and (3) the effect of IL3 on C2C12 myoblasts.

Metabolomics data are available at the NIH Common Fund’s National Metabolomics Data Repository website, the Metabolomics Workbench (Sud et al., 2016), https://www.metabolomicsworkbench.org, where it has been assigned Study ID ST003927. The data can be accessed directly via the Project DOI: https://dx.doi.org/10.21228/M8NR8D. All other data are presented within the manuscript, and corresponding authors can be contacted as needed for clarification.

We want to thank all the members of the Doles lab.

This work was supported by Mayo Clinic, Indiana University, and an American Association for Cancer Research/Pancreatic Cancer Action Network Career Development Award to J.D. Doles. A. Dasgupta is supported by National Institutes of Health (NIH), K99CA283277. We also would like to thank the UNMC Tissue Bank RAP supported by Pancreatic Cancer Detection Consortium, U01CA210240, National Cancer Institute (NCI) Cancer Center Support Grant, P30CA36727, and NCI Research Specialist, R50CA211462. We also acknowledge the Islet and Physiology Core of the NIH–funded Indiana Diabetes Research Center (P30DK097512). Metabolomics workbench is supported by NIH grant U2C-DK119886 and OT2-OD030544.

Author contributions: A. Dasgupta: conceptualization, data curation, formal analysis, funding acquisition, investigation, methodology, project administration, supervision, validation, visualization, and writing—original draft, review, and editing. R.E. Schmitt: investigation and visualization. T. Kono: data curation, formal analysis, investigation, methodology, resources, software, validation, visualization, and writing—original draft, review, and editing. C.-C. Lee: methodology. M.I. Zoberi: investigation. S.A. Epstein: investigation and validation. J.Z. Schneider: data curation and investigation. A. Hernandez: investigation. P.M. Grandgenett: methodology, resources, and writing—review and editing. T.C. Caffrey: resources. D.J. DiMaio: resources. M.A. Hollingsworth: resources, supervision, and writing—review and editing. J.D. Doles: conceptualization, data curation, formal analysis, funding acquisition, methodology, project administration, resources, supervision, validation, visualization, and writing—original draft, review, and editing.