This is the first study to comprehensively identify hemostasis related genes in both rodent and human skeletal muscle with microarray technology. We confirm the presence of a couple of previously reported genes (tPA and tetranectin) and extend this list to at least another 15 transcripts significantly expressed above background in skeletal muscle that have known roles in coagulation and fibrinolysis. Furthermore, we identified a potentially novel candidate gene, LPP1, that was sensitive to physical inactivity in both rat and human skeletal muscle, explaining why physical inactivity (too much prolonged sitting) may increase the risk for thrombosis. Because physical inactivity (e.g., sitting) is associated with DVT
] and changes in the contractile state of muscle, we first raised the hypothesis in 2007
] that a change in the expression of a key gene(s) in deep skeletal muscle tissue would be evident and it could be a distinct response to the physiology of inactivity (sitting too much) apart from recommended exercise. LPP1, with a proposed antagonistic role in platelet aggregation
] and inflammation
], was significantly decreased in both the human and rat studies of muscular inactivity. LPP1 was decreased with both acute and chronic reductions in normal daily contractile activity but was not impacted by exercise.
We used Affymetrix microarrays with probes for over 44,000 human transcripts. From this search we identified 23 transcripts from genes with functions related to hemostasis that were at least moderately expressed according to Affymetrix standards in human muscle (Table
). Previously, the protein for one of these (tPA) was reported to be detectable in human muscle biopsies
]. Importantly, the more definitive detection of tPA (and tetranectin) at the mRNA level in human muscle
] is consistent with our findings that one of the sources for these and other hemostatic factors could be within the muscle tissue itself, rather than indirect accumulation of the protein from the blood. We have here confirmed and extended these observations to other genes involved in fibrinolysis (uPA and uPA receptor, annexin A2), and also to genes involved in blood coagulation (i.e., Factors VII and VIII, thrombin receptor, and vWF). Furthermore, the mRNA for two enzymes, gamma-glutamyl carboxylase and vitamin K epoxide reductase, necessary for the synthesis of coagulation factors, were expressed in skeletal muscle. Despite some differences in muscle fiber type between species, all of the genes with anti-coagulant and fibrinolytic functions detected in the rat muscle were also expressed in the human muscle (i.e., annexin A5, ectonucleoside triphosphate diphosphohydrolase 1, protein S (alpha), thrombomodulin, tPA, annexin A2, uPA). This agrees with a general tenet in muscle physiology that skeletal muscle is relatively conserved between species compared to other tissues. This observation of similar anti-coagulant and fibronolytic genes being expressed in the skeletal muscle of both rats and humans would support the use of rodent models in translational studies of muscle and hemostasis.
The new knowledge that LPP1 is expressed in leg tissue differentially by physical activity patterns is significant because it provides the first study we are aware of on the physiological regulation of this gene. LPP1 mRNA encodes a 32 kDa transmembrane ecto-enzyme responsible for the degradation of the extracellular bioactive phospholipids by dephosphorylation. Lysophosphatidic acid (LPA), sphingosine-1-phosphate, ceramide-1-phosphate, and phosphatidic acid are all substrates for LPP1. LPA is arguably the most important of these lipids because it has the highest affinity for LPP1, and even more importantly, because LPA has recently emerged as a potent stimulator of platelet aggregation
], platelet monocyte aggregation
], tissue factor expression
], fibronectin matrix
], and inflammation
]. Similarly, sphingosine-1-phosphate has been shown to stimulate platelets to bind and assemble fibronectin
], stimulate tissue factor expression
], and inflammation
]. Both LPA and sphingosine-1-phosphate are released from activated platelets and are thought to promote a positive feedback on platelet aggregation. Haseruck et al.
] argued that the expression of LPP1 on the endothelium of tissues (such as the endothelium in muscle tissue) would attenuate local accumulation of LPA and thus limit platelet activation
]. In cell culture experiments where LPP1 expression has been manipulated, LPP1 has been shown to be important for the attenuation of several atherothrombogenic and malignant processes induced by LPA, including platelet aggregation
], the release of the proinflammatory cytokine IL-8
], and ovarian cancer cell proliferation
]. There are several lines of evidence that LPP1 can also attenuate the signaling of other factors including the prothrombotic and proinflammatory cytokine TNFα
] and thrombin
]. Therefore, if the protein is regulated by its gene expression, LPP1 would be a novel candidate for why physical inactivity is a risk factor for DVT or other types of thrombosis (Figure
Figure 6 Relationship between hemostatic gene function, physical inactivity, LPP1 and risk of deep venous thrombosis. Human skeletal muscle expresses distinct groups of genes involved in hemostasis (fibrinolysis, anti-coagulation, and coagulation factors, and (more ...)
Other genes involved in LPA metabolism were detected in skeletal muscle (Table
). An isoform of LPP1, LPP3, is significantly expressed in human skeletal muscle and is similar to LPP1 in that it degrades extracellular LPA and has been shown to attenuate LPA induced signaling in ovarian malignant epithelial cell cultures
]. In contrast to LPP1’s location on the apical side of endothelial cells, LPP3 is thought to be located on the basolateral membrane of endothelial cells
]. The enzyme responsible for the conversion of lysophosphatidylcholine to LPA, lysophospholipase I, was strongly detected in both human and rat skeletal muscle and, therefore, may play a local role in LPA production.
Although we present robust data showing that physical inactivity can reduce LPP1 transcript level, we did not determine if this was due to a reduction in transcription (possibly secondary to epigenetic changes to chromatin or DNA) or a reduction in mRNA stability or both. In addition, we report only on the mRNA level of this gene and not on the protein as there are not good antibodies yet available for this transmembrane protein. As we tested only lean and apparently healthy rats and humans we cannot determine that LPP1 expression would be responsive to physical inactivity in a more diseased population.