In this study, we show that cultured human tenocytes undergo apoptosis following Anti-Fas treatment. Our results further show that SP, via NK-1 R, reduces Anti-Fas-induced apoptosis in a dose- and time-dependent manner in these cells, and that the anti-apoptotic effect of SP is mediated through an Akt-dependent pathway. Moreover, the results of the study seem to support the hypothesis that endogenous SP produced by tenocytes , exerts autocrine anti-apoptotic effects via NK-1 R.
The methods of culturing primary human tendon cells used in this study have successfully been used in previous studies [1, 18]. We have confirmed that the isolated cultured human tendon cells used in our study grow easily in culture and form networks with cell-cell connections (connexin 43; see ), and that they are of tenocyte phenotype in the passages used for experiments, based on the expression of tenomodulin and scleraxis, both specific markers for tenocytes [19, 20]. The cells were also shown to express a high level of collagen type I , which further confirms tenocyte characteristics . Of further interest for the results of this study, we have previously shown that these cells produce SP and express NK-1 R in passages used for these experiments .
In the present study, apoptosis was a vital end-point measurement for the hypothesis and aims. The release of LDH, which occurs when cells lose their membrane integrity, cannot distinguish between apoptosis and necrosis, nor can crystal violet assay, which is an indirect measurement of the number of cells. To confirm apoptotic events, we have therefore used TUNEL assay, which is a method designed to specifically detect cells undergoing apoptosis, as well as analysis of apoptotic events such as cleavage (i.e. activation) of caspase-3 and cleavage of PARP.
The induction of cell death by Anti-Fas, as measured with LDH, showed a clear dose- and time-dependent response. However, the dose–response shown for the effects of SP and the NK-1 R inhibitor was not as clear-cut, but a trend was seen with the most effective concentration for both SP and the NK-1 R inhibitor in accordance with the concentrations used by others [3, 17].
Anti-Fas as a model for apoptosis in cultured tenocytes
We observed that human tenocytes abundantly express the Fas Receptor (FasR), making them susceptible to stimulation with FasL (Anti-Fas treatment). Indeed, we found that exogenously added Anti-Fas induced a dose- and time-dependent cell death (increase in LDH release). Anti-Fas concurrently reduced tenocyte cell viability, and furthermore induced apoptosis-specific events such as fragmentation of DNA and cleavage of caspase-3 and PARP, confirming an apoptotic effect of Anti-Fas.
It is known that inflammatory cells produce FasL , and that in cases of tendinosis, the degree of inflammatory cell infiltrates in the paratenon, the tendon sheath, is significantly increased [7, 23, 24]. This makes it probable, that apoptosis of tendinosis in vivo  is partly explained by a FasL production by the inflammatory cells in the paratenon, which in turn stimulate the tenocytes expressing FasR to undergo apoptosis.
In our analysis of apoptotic events, cleaved caspase-3 and PARP are used. We show that Anti-Fas treatment cleaves/activates caspase-3 as well as cleaves PARP. It is well known that the main effect of c-caspase-3 is cleavage of PARP. However, the existence of c-PARP does not necessarily indicate apoptosis, as it is known to be able to occur as an independent event that can be disassociated from apoptosis . Furthermore, our study does not determine if caspase-8 directly activates caspase-3, and thereby indirectly cleaves PARP, nor if the cleavage of caspase-3 is a result of mitochondrial involvement and release of cytochrome C. Nevertheless, we show, by using the pan-caspase inhibitor (ZVAD-FMK), that the cleavage of PARP is caspase-dependent, as the pan-caspase inhibitor prevents the Anti-Fas-induced increase in cleaved PARP, and therefore we also demonstrate that the cleavage of PARP is apoptosis-dependent.
To summarize, in human tenocytes, Anti-Fas exposure is here confirmed to be a good model to induce apoptosis, at least when studying late apoptotic markers such as caspase-3 and PARP. Similar responses have also been observed in other cell types, such as epithelial cells , showing that Anti-Fas is a good apoptosis inducer, and human colonocytes , showing that apoptotic end-markers (caspase-3/PARP) are good to evaluate an apoptotic response in cultured cells.
SP reduces Anti-Fas induced apoptosis in tenocytes
In this study, we show that SP dose-dependently protects from the Anti-Fas-induced decrease in cell viability and increase in LDH release (i.e. cell death), and also that it reduces the cleavage/activity of late apoptotic markers such as caspase-3 and PARP. However, when interpreting crystal violet results, one should bear in mind that SP has also been shown to have a proliferative effect in tenocytes , as well as in other cell types (such as preadipocytes  and human skin fibroblasts ), and that this will affect the viability results. However, the LDH and crystal violet assays in combination with the specific apoptosis analyses, make it likely that the outcome from the LDH and crystal violet assays are a result of apoptosis. Nevertheless, the fact that SP in this study seems to have a less marked blocking effect on Anti-Fas-induced viability decrease, as compared with Anti-Fas-induced LDH increase, is logical and could be explained by SP's proliferative effect in these cells.
In this study, SP's preferred receptor, NK-1 R, was blocked with a NK-1 R inhibitor to confirm that the effect of SP is mediated via a NK-1 R specific pathway. Indeed, the effect of SP was reduced in a dose-dependent manner when cells were pre-treated with the NK-1 R inhibitor in different concentrations, confirming that SP mediates its effect via NK-1 R in human tenocytes. However, hypothetically, SP produced by the tenocytes [1, 4] can also bind to, and exert effects via, NK-2 R and NK-3 R; it is just a question of receptor availability and the concentration of SP . These two receptor subtypes were not investigated in this study, as NK-1 R is the dominant/preferred form [26, 27] and is also clearly expressed on tenocytes in human Achilles tendons , and as the results of the experiments here were clear-cut regarding the inhibition of SP effects when blocking the NK-1 R. Nevertheless, the fact that SP also has the possibility of binding to NK-2 R and NK-3 R, might explain why the use of the NK-1 R inhibitor did not completely abolish the effect of SP at all time.
The results that SP promotes decreased apoptosis, and also that it increases proliferation , in human tenocytes in culture, are effects corresponding to SP effects shown for other cell types as well. Thus, this double-edged response of SP has also been demonstrated in human colonocytes [2, 3], skin fibroblasts  and preadipocytes , in which SP stimulates proliferation and has an anti-apoptotic effect simultaneously.
The anti-apoptotic effects of SP is mediated via activation of Akt
We show that in human tenocytes SP time-dependently activates Akt, a protein kinase that is well known to facilitate cell survival and to prevent apoptotic cell death . This is in accordance with previous results in other cell types [3, 28] in which SP also activates Akt. Furthermore, the reduction in Anti-Fas-induced apoptosis of human tenocytes seen after SP treatment in this study (as evident by decreased cleavage of PARP), is effectively blocked with an Akt inhibitor, showing that SP's effect is mediated, at least in part, via Akt.
Following exposure to SP, a plateau of phosphorylated Akt was seen after ca. 5–10 min. As phosphorylation of Akt is in the early cascade of an anti-apoptotic effect, this prompt Akt response after SP exposure is trustworthy, and, as it should, it precedes later responses, such as decreased cell death/apoptosis and increased viability. The prompt Akt response to SP is furthermore in accordance with a study on colonocytes , which in addition also had the same time-response for the SP induced reduction in end-point apoptosis markers, such as PARP cleavage.
In this study, we also confirmed that the phosphorylation of Akt was dose-dependently reduced when cells were pre-treated with the NK-1 R inhibitor. This confirms that the effect is specifically mediated via a SP-NK-1 R specific pathway.
There are other pathways, than the Akt pathway, that might mediate an anti-apoptotic effect in response to SP stimulation, which are not examined in this study. For instance, SP is known to induce an activation of NF-kappa B  which in turn can up-regulate the caspase-8 inhibitor FLIP, resulting in increased resistance to Anti-Fas-induced cell death .
Apoptosis in tendinosis and the role of SP produced by tenocytes; concluding remarks
Tendinosis is known to be associated with a higher expression of apoptotic cells than normal tendon tissue [9, 31]. The role of apoptosis in tendinosis is, however, unknown. A cardinal feature of tendinosis is furthermore the marked hypercellularity seen in the tendon tissue . The pathophysiology of such hypercellularity, as well as its possible function, is not yet clarified. However, in an animal model, we have shown that overload of the Achilles tendon induces elevated intratendinous SP production  as well as tenocyte hypercellularity . Most interestingly, the elevation of endogenous SP production was found to precede the increase in cell number, and exogenously administered SP further accelerated the hypercellularity . The fact that tenocytes respond to mechanical stress with increased SP production has furthermore been confirmed in human tendon cells in vitro .
It is possible, that the hypercellularity seen in tendinosis is an inappropriate response to mechanical stress, leading to defective tendon metabolism, and that the increase in apoptosis is an attempt for the tendon to maintain an environment with a balanced number of tenocytes, thereby limiting a process of excessive and possibly dysfunctional collagen synthesis (see for instance ). If so, SP may, speculatively, have a dual detrimental role in the pathophysiology of tendinosis; harbouring the capacity of both stimulating proliferation of tenocytes  and inhibiting apoptosis of these cells, the latter shown by this study.
Our hypothesis is based on the assumption that there is an autocrine SP loop in human tenocytes, i.e. that the tenocytes produce SP, in response to mechanical stimuli, and this SP in turn affects the tenocytes themselves by stimulating NK-1 R on the cell surface. Several results seem to corroborate this hypothesis. Human tenocytes have been shown to produce SP in vitro (ranging in the concentration of 170 pg/2 × 106 tendon cells) , and also to express the NK-1R both in vitro  and in vivo . In this study, the phosphorylation of Akt that is seen after Anti-Fas treatment (without exogenous SP being added) is reduced when the tenocytes are pre-treated with a NK-1R inhibitor (Fig. 13). As we also show, in this study, that SP effectively phosphorylate Akt in tenocytes, it is not far-fetched to speculate that endogenously produced SP, in an autocrine manner, binds to the NK-1Rs present on the cells, thus resulting in phosphorylation of Akt, a loop that is interfered with when the NK-1R inhibitor is added. The NK-1 R inhibitor together with Anti-Fas furthermore resulted in a higher expression of cleaved caspase-3 and PARP (Fig. 13), than did Anti-Fas alone, possibly suggesting that a protective, anti-apoptotic effect of endogenous SP is constantly present and here blocked by the NK-1 R inhibitor. Nevertheless, the present study, in conjunction with previous studies here cited, only provide indications of an autocrine SP loop in tenocytes. Future studies, using for instance siRNA technique to silence the SP expression, will have to elucidate this further.
To summarize, the two here mentioned effects of the NK-1 R inhibitor (reduced P-Akt and increased c-caspase-3/c-PARP) can be explained by two alternative sequences: (1) Endogenous SP is blocked, leading to decrease in P-Akt which in turn leads to increased apoptosis (Akt being a known anti-apoptotic protein kinase ), or (2) endogenous SP is blocked, leading to increase in apoptosis (through alternative pathways than the Akt-dependent [29, 30]), which in turn leads to a caspase-dependent cleavage/inactivation of Akt .
Finally, it should here be mentioned that as for the in vivo situation, alternative sources of ligands stimulating the NK-1 Rs of the tenocytes might, in addition to the tenocytes themselves producing SP, be SP-positive nerves in or around the tendon [8, 36], or speculatively cells of the tendon producing other tachykinins. It is thus known, that the tachykinin neuorokinin A (NKA) can also bind to NK-1 R with high enough affinity to elicit a biological response . Considering the fact that NKA is shown to be present in nerve endings of the paratenon of rat Achilles tendons  it is possible that NKA is another substance that might bind to the NK-1 R expressed by human tenocytes in vivo.
In conclusion, considering both its proliferative and anti-apoptotic effects, the results of this and previous studies identify SP as a potent regulator of the marked hypercellularity seen in tendon tissue as part of the pathology of tendinosis.