• Cancer
• Colorectal Cancer

Colorectal cancer (CRC) is the second leading cause of cancer-related deaths in North America and Europe. Happily, the incidence and mortality of CRC have declined by 3.4% and 3.0%, respectively, from 2003 to 2007 in the USA.1 The decline in incidence is attributed to the public health emphasis on the importance of CRC screening and the reduction of behavioural risk factors. For the majority of people who do not use tobacco and alcohol, the most important determinants of CRC risk factors are weight control, dietary choice and level of physical activity. A number of investigations have demonstrated the benefit of exercise on CRC prevention, and guidelines released by governmental and non-governmental organisations also emphasise the importance of regular exercise for decreasing CRC incidence.

Although the correlation between exercise and CRC prevention is conclusive, the molecular mechanism for the protective effect of exercise remains largely unresolved. Epidemiological and experimental studies have proposed several possible biological mechanisms to explain the exercise–malignancy relationship, including regulation of metabolic activity, production of sex hormones, alterations of antioxidant enzymes and growth factors, body weight control, and alteration of immune system.2 Recent studies have shown interleukin-6 (IL-6), IL-8, IL-15, brain-derived neurotrophic factor and leukaemia inhibitory factor to be exercise-induced myokines that mediate several biological processes.3 However, none of these myokines is directly linked to the prevention of CRC tumorigenesis.

In the current study, Aoi et al assessed the transcriptome of the gastrocnemius muscle in mice that received regular exercise for 4 weeks as compared with muscle from sedentary mice.4 In addition, transcriptome analysis was performed on young and old sedentary mice to evaluate the effect of age. By comparing the microarray data from these two experiments, the authors identified a novel exercise-induced myokine, secreted protein acidic and rich in cysteine (SPARC). The secretion of SPARC from the skeletal muscle of exercised mice was confirmed in ex vivo experiments. Surprisingly, the plasma levels of SPARC in exercised and sedentary mice were not significantly different. It is noteworthy that the blood samples were collected 24 h after the last round of exercise. However, SPARC protein was immediately elevated in muscle tissues and in the plasma of the mice that received a single session of exercise, and the plasma SPARC protein gradually returned to the pre-exercise level within 6 h after exercise. Similar results were also observed in humans, whose plasma SPARC was transiently increased after exercise but returned to a steady-state level at rest.

SPARC, also known as osteonectin, is a matricellular protein that binds to the components of extracellular matrix (ECM) and is involved in cell-cell interaction, growth factor function, matrix metalloproteinase expression, cell shape alteration and cell differentiation.5 The role of SPARC in cancer remains controversial and is highly cancer type-dependent and context-dependent.6 However, SPARC has been consistently considered to be a tumour suppressor in CRC according to the evidence from experimental cell models, SPARC knockout mice and clinical cohort studies,7 although an inverse correlation of stromal SPARC with poor prognosis in CRC was reported recently.8 Aoi et al found that regular low-intensity exercise significantly reduced the number of chemically induced aberrant crypt foci (ACF) and aberrant crypts (AC) in the colons of wild-type mice, but not in SPARC-null mice. Furthermore, the injection of low-dose or high-dose recombinant SPARC also prevented the formation of chemically induced ACF and AC in the colons of wild-type mice. However, a possible paracrine effect of exercise-induced SPARC in the muscle tissue for colon cancer prevention cannot be ruled out. The authors’ in vitro and in vivo studies indicated that the protective function of exercise-induced SPARC might result from the increased apoptosis and decreased proliferation observed in colon cancer cell lines. However, more experiments are required to examine the molecular mechanism of SPARC-regulated cell growth.

The data from Aoi et al reasonably indicate that exercise-induced SPARC released from muscle tissue into the circulation prevents the tumorigenesis of CRC (figure 1). However, two questions are raised from this finding. The first question pertains to how SPARC protein was rapidly released from skeletal muscle cells. In this study, the application of cycloheximide, a translation inhibitor, but not actinomycin D, a transcription inhibitor, blocked cyclic stretching-induced SPARC secretion from C2C12 muscle cells, suggesting that the rapid increase of SPARC secretion from muscle cells may be due to an acceleration of SPARC translation. However, in the animal experiment, the SPARC mRNA levels were significantly increased in the skeletal muscle of the mice that received regular exercise for 4 weeks. From these results, we may assume that both transcriptional and translational mechanisms potentiate the production and secretion of SPARC protein when people adopt regular exercise. The other question that remains is how a transient increase of SPARC in the circulation is able to influence tumour formation in the colon. In this study, single-session exercise-induced plasma SPARC was decreased at the first hour and returned to the basal level at 6 h post-exercise, suggesting that plasma SPARC is not sustained during rest. Despite the fact that SPARC can directly induce the apoptosis of colon cancer cells and inhibit their proliferation, the possibility cannot be excluded that SPARC may indirectly prevent tumorigenesis by regulating the microenvironment in the colonic tissue.9 Through binding with ECM proteins, SPARC potentially contributes to the organisation of matrix in the connective tissue and basement membrane, which may change the biophysical and biochemical matrix properties that guide cell transformation and differentiation.10 These possibilities might apply in different situations that allow cells of various types to respond to SPARC immediately after exercise.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Exercise prevents tumorigenesis of colon cancer through muscle-derived SPARC. Animal and human experiments reveal exercise induces SPARC production and release from skeletal muscle tissue that transiently increases SPARC protein levels in plasma. Elevation of SPARC in plasma decreases the formation of chemical-induced aberrant crypt foci in mouse colon, which may be through the increase of apoptosis of colon cells.

Understanding the molecular mechanisms by which exercise results in CRC prevention will be of great help for designing comprehensive lifestyle guidelines and therapeutic interventions against CRC. Aoi et al have identified SPARC as a novel myokine that provides an underlying molecular mechanism for the correlation between exercise and CRC prevention. Based on their findings, new strategies could be developed to prevent CRC by adjusting exercise levels to elevate the plasma SPARC level.