5FU has been the first-choice chemotherapy drug for several cancer types;however, its efficacy is diminished by patient acquired resistance and pervasive side effects contributing to reduced life quality and poor treatment outcomes(6-9).Given 5FU’s deleterious effects on circulating leukocytes, the purpose of ourstudywas to investigate the acute effects of 5FU on resident and infiltrating skeletal muscle monocytes and inflammatory mediators in addition to examining the effects of 5FU on circulating and bone marrow immune cells. Our results extend previous studies to identify that 1 cycle of a clinically translatable dose of 5FU significantly reduced CD45+ immune cells and infiltrating/activated CD11b+Ly6CHigh monocytes in skeletal muscle that was associated with a decrease in select skeletal muscle inflammatory mediators. Additionally, the reduction in skeletal muscle and circulating immune cells wasaccompanied by a reduction in bone marrow monocytes and an increase in cell cycle arrest. These resultsidentify novel off-target effects of 5FU on skeletal muscle and the skeletal muscle microenvironment.
Our understanding of chemotherapy-induced body weight and function loss, termed cachexia, has improved over the last decade(15, 43, 44).Our investigation of the acute (1 cycle) effects of 5FU demonstrated that 5FU induced clinically relevant body weight loss (>5%)(45), which was accompanied with signs of anorexia, but not skeletal muscle mass loss. Others have demonstrated that 5 weeks of 5FU combination therapy, Folfiri (leucovorin, 5FU, Irinotecan), demonstrated decreased body weight and lean mass over time, with corresponding reductions in several hindlimb weights(15).Interestingly, mice given 5 weeks of Folfox (leucovorin, 5FU, oxaliplatin) rather than Folfiri maintained body weight and lean mass, and only showed reduced quadriceps weight. Additionally, similar to the doxorubicin effects on skeletal muscle(43, 46),only Folfiri reduced skeletal muscle specific force (strength per muscle unit area) – which may occur through several mechanisms including fibrosis(15).Together this evidence suggests that significant muscle mass loss only occurs after sustained 5FU treatment(15, 47),given that 1 week of 5FU was unable to reduce hindlimb muscle weight. We then hypothesize that a loss of water weight is likely to contribute to the observed body weight loss as anorexia and dehydration with 5FU has been reported(48, 49).However, pathologies that occur early in the treatment regimen may provide insight into the etiology of muscle mass and strength losses with 5FU. To this end, 5FU-induced anemia and leukopeniaare likely to contribute to the observed functional pathologies that occur late with 5FU treatment.
The role of immune cells, particularlymacrophages, in skeletal muscle regeneration, repair, and remodeling has been well characterized(29); however, chemotherapy’s effects on these processes is not well known. Following skeletal muscle insult (e.g. damage, ischemia, exercise), there is an initial influx of neutrophils which in turn recruit naïve monocytes primarily through the release of MCP-1(29, 33). We have previously shown that 5FU induced circulating MCP-1 after 14 days of treatment, which was associated with reduced voluntary physical activity(23). The monocytes recruited by MCP-1 are primarily recruited as CD11b+Ly6CHigh monocytes which can either remain as such or differentiate and polarize to a pro-inflammatoryM1-like F4/80+CD11c+CD206-macrophage(34, 50, 51).Following an acute 5FU regime (1 week) we document a reduction in total and relative Ly6CHigh monocytes as well as totalM1-like F4/80+CD11c+CD206-macrophages in skeletal muscle despite no changes in skeletal muscle pro-inflammatory MCP-1, IL-6 and TNFα levels. However, we did observe decreased expression of pro-inflammatory genes associated with M1-like macrophages, IL-1β and IFNγ, but on the other hand did not observe corresponding changes to total M1-like macrophage cell surface marker, Itgax, more commonly known as CD11c(52).These discrepancies between the flow cytometry and gene transcription require additional work and thus, interpretations should be taken with caution; however, flow cytometry remains the gold standard for the assessment of immune cells, and it appears evident that 5FU has deleterious effects on the pro-inflammatory monocytes and macrophages.A loss of pro-inflammatory or phagocytic M1-like macrophages could negatively impact skeletal muscle remodeling and repair(50). Chemotherapeutic doxorubicin has been shown to blunt the pro-inflammatory response following exercise which mitigated the muscle’s response to exercise(53). Furthermore, while repeated muscular contractions were able to improve muscle mass in cancer patients undergoing treatment, patients did not obtain the functional and metabolic improvements that has been previously seen with exercise(54, 55).While chemotherapeutics 5FU and doxorubicin mechanisms of action differ, we can still glean potential mechanisms and clinical manifestations. To the best of our knowledge, we are the first to identify that 5FU disrupts skeletal muscle’s pro-inflammatory immune cell environment. It is also important to note that these cell surface markers and the M1/M2 dichotomous classification of macrophages does not properly reflect the true diversity and nature of resident/infiltrating macrophages and should again be interpreted cautiously(29, 34, 56, 57).
Tissue resident macrophages are classically CD206+ anti-inflammatory, pro-fibrotic surveying macrophages;(34, 37, 58, 59) however, macrophages are plastic and as skeletal muscle repair progresses theinfiltratedM1-like F4/80+CD11c+ macrophages can reduce the gene expression and release of pro-inflammatory mediators and become more phenotypically M2-like to promote extracellular matrix remodeling and angiogenesis.(60-62)Others have proposed that resident macrophages are predominantly M0 (CD11c-CD206-) which are self-maintained, proliferate, and polarize to an M1-like phenotype upon activation during the initial stages of injury repair(29, 59). Regardless, our results demonstrate that the relative phenotype of skeletal muscle macrophages is not changed by 5FU treatment; however, the total number of M1-like (CD11c+CD206-), M0-like (CD11c-CD206-), and M1-M2-like transitional macrophages were reduced with 5FU while M2-like macrophages appear spared from 5FU’s cytotoxicity – at least following 1 week of 5FU. Additionally, anti-inflammatory IL-10, pro-fibrotic TGFβ, and M2-like macrophage cell surface marker Mrc1, commonly known as CD206, gene transcription were not changed by 5FU treatment. The potential for 5FU to target M1-like macrophages rather than M2-like, points to a pro-fibrotic skeletal muscle microenvironment. 5FU combination therapy Folfiriwas shown to reduce skeletal muscle specific force (force per unit area); however, neither fibrosis nor an increase in fibrotic genes (TGF-β associated ligands) were apparent(15). Therefore, it is likely thatthese pro-fibrotic M2-like cells remain at a physiological abundance during 5FU treatment and may not be contributing to a skeletal muscle pathology directly.Interestingly,TAMs phenotypically reflect M2-like macrophages promoting immunosuppression, fibrosis, and angiogenesis, within the tumor microenvironment and have been associated with 5FU acquired resistance(63). The potential for M2-like macrophages to be protected against 5FU requires significant attention in the cancer domain.
Chemotherapy has been shown to mitigate the inflammatory response with exercise(53, 64), induce leukopenia/cytopenia(19), and disrupt cardiacmacrophage infiltration(65). To the best of our knowledge, this is the first study to demonstrate that chemotherapeutic 5FU has deleterious effects on immune cell abundance in otherwise healthy uninjured skeletal muscle. The absolute reduction in macrophage number rather than relative changes in abundance remains relevant given the physiological importance of the overall immune response in repair and remodeling(59, 66-70). The mean age of cancer patients is ~65yrs and overlapping sarcopenic and cachectic factors along with chemotherapy may contribute to disrupted skeletal muscle immune regulation(71).Disrupted skeletal muscle repair associated with changes in macrophages has been reported with aging(38), cancer(70), and chemotherapy(53).The effects of aging on skeletal musclemacrophages has demonstrated that reloading aged skeletal muscle had a blunted hypertrophy response associated with a lower number of M1-like macrophages at baseline and blunted M1-like macrophage infiltration (early) and M2-likemacrophage transition (late)(38).Surprisingly, while inflammation is a hallmark of cancer cachexia associated with muscle weakness and fatigue(72, 73), total macrophage number was reduced in damaged muscle of C26tumor-bearing mice compared to a non-cachectic tumor-bearing control(70). Additionally, macrophages were shown to regulate skeletal muscle signal transducer and activator of transcription 3 (STAT3) – downstream target of IL-6 and key regulator of skeletal muscle mitochondrial homeostasis and proteostasis(73-77)– during pancreatic cancer cachexia(78).Further work is needed to understand these potentially overlapping mechanisms with cancer and chemotherapy on skeletal muscle immune cells.
Chemotherapy’s effects on systemic inflammatory mediators(2, 4, 10, 79) and intrinsic skeletal muscle inflammatory signaling(12-14, 26-28) are continuing to be unearthed; however,our study is the first to identify that 5FU-induced leukopenia extends beyond circulation to impact the skeletal muscle microenvironment.Our results indicate that 5FU’s toxic effects on skeletal muscle leukocytes are not necessarily specific to monocytes shown by no change in both circulating monocyte count or relative abundance of skeletal muscle CD11b+ monocytes within the CD45+ population. This is not to say that monocytes aresparedfrom 5FU as the total number of skeletal muscle monocytes are reduced. Bone marrow CD11b+ monocytes and the relative abundance of infiltrating Ly6CHigh monocytes in skeletal muscle were reduced with 5FUwhich is supported by the established deleterious effects of 5FU on circulating leukocytes and the hematopoietic system(8, 10, 27). In conjunction with previous studies, our results support that 5FU’s toxicity is predominantly associated with pro-inflammatory mediators extending beyond the hematopoietic system to impact the skeletal muscle microenvironment. Another potential mechanism for the observed impact of 5FU on skeletal muscle immune cells is the potential forreduced proliferation of pro-inflammatory macrophages within the muscle microenvironment. Given that circulating and skeletal muscle monocytes/macrophages are not proportionally reduced it is possible that 5FU increased maturation of monocytes within the skeletal muscle as well as increased proliferation of M2-like macrophages.