**3. Oxidative metabolism and apoptosis of neutrophils cultured in physiological concentrations of SF, oxygen and cyclic loaded pressure**

Although the presence of activated neutrophils in SF is well documented, it is still unknown how the neutrophils are activated, how they interact with other cells, and how long they persist at the site of inflammatory joint.

It was identified that addition of SF to neutrophils results in activation of neutrophils measured by a rapid chemiluminescence (CL) response (luminol- and lucigenin-dependent). Luminol-dependent CL is capable of monitoring both intracellular and extracellular ROI generation (NADPH oxidase and MPO), as luminol freely penetrates the neutrophil's cell membrane. Lucigenin-dependent CL, on the other hand, measures only the rate of extracellular ROI secretion (NADPH oxidase dependent) because lucigenin does not penetrate neutrophils, and light emission detected is independent of the activity of MPO. Pre-incubation of normal blood neutrophils in 10% SF enhanced the luminol- and lucigenin-CL, suggesting that both MPO and NADPH oxidase activity were activated in parallel during exposure to 10% SF from RA patients (Bender et al., 1986). Synovial fluid (20% concentration) isolated from RA patients activated blood neutrophils, leading to increase of luminol dependent CL over a 50 range (Nurcombe et al., 1991b). In contrast, the same fluid activated to a much lower range (two or three fold) of maximal rates of lucigenin dependent CL. All the mentioned reports were performed with SF in concentrations that did not exceed 20%. Other studies, confirmed that SF (used in concentration of 25%) produced rapid and parallel responses (luminol- and lucigenin-CL) in neutrophils.

Our recent results (Gajewski et al., 2009), in contrast to earlier studies, indicate that higher concentrations of SF (up to 80%) have a quite different effect on lucigenin-dependent and luminol-dependent CL response. Increased concentrations of SF resulted in a reduction of luminol-dependent CL response and a very significant increase of lucigenin-dependent CL, reflecting extracellular ROI generation. This effect was observed irrespective of the stimulator used and whether neutrophils were isolated from SF or blood from either RA patients or healthy subjects. This indicates that increasing SF concentration results in higher extracellular ROI secretion and lower MPO-dependent ROI production. The promotion of extracellular release of ROI observed in this experiment is likely to be associated with the high concentrations of SF used, and raises the possibility that extracellular activity of neutrophils may be a general characteristic which prolongs the inflammatory process (Gajewski et al., 2009).

Similar studies were performed by Bell et al. (1995). In these studies they examined the hypothesis that persistent inflammatory responses in RA may result from inhibition of neutrophils apoptosis by factors in SF. The effects of aging in culture and addition of SF on apoptosis was investigated using SF in a concentration range 0-75%. A significant effect of SF on promotion of apoptosis of synovial fluid neutrophils was observed at concentrations of 50% and above (Bell et al., 1995).

It has been proposed that the process of hypoxic-reperfusion injury contributes to the persistence of synovitis in the inflamed joint. The generation of pathological, exercise

The Role of Neutrophils in Rheumatoid Arthritis – Experiments *In Vitro*: A Change of Conception? 63

One of the truly remarkable discoveries in modern biology is the finding that the nervous system and the immune system use a common language for intra- and inter-system communication (Blalock, 2005). This biochemical information circuit between neurons and immune cells allows the immune system to function as a sensory organ, and completes our ability to be cognizant not only of the world we can seen hear, taste, touch and smell, but also of bacteria, viruses, antigens or tumor cells. Recognition of such "non-cognitive stimuli" by the immune system results in transmission of information to the central nervous system to cause a physiological response that is ultimately beneficial to the host and detrimental to

The idea that immune cells were a source of neuropeptides was viewed by many as heretical; as is often the case when fundamental and unexpected discoveries are made (Blalock, 2005). Concurrent with early observation on production of neuropeptides by immune cells emerging studies reported that neuropeptide/neurotransmitter receptor were

It has been proposed that activity of neutrophils in RA may be influenced by neurotransmitters, including agonists of adrenergic and cholinergic receptors. This conclusion was based on the results of pharmacological studies that suggested the existence of several subtypes of functional adrenergic and cholinergic receptors of human neutrophils. An important modulatory mechanism in neutrophils is the interplay between stimulatory and inhibitory receptors. For example, activation of neutrophils is antagonized by agents that stimulate adenylate cyclase (AC). These agents include B-adrenergic agonists, prostaglandins of the E series, adenosine and histamine. These effects appear to be mediated through adrenergic receptors, activating AC, which leads to an increase in cyclic AMP. Based on the hypothesis of the opposing actions of cAMP and cGMP, it was suggested that the agonists of cholinergic receptors might stimulate leukocyte activation (Gajewski et al.,

The effect of the adrenergic receptors isoproterenol (ISO) and the cholinergic receptor carbachol on ROI production was measured by CL. Activation of neutrophils by OZ (stimulating phagocytosis), PMA (activation of NADPH oxidase) and fMLP (stimulation of

The adrenergic receptor antagonist ISO has inhibitory effects on the CL of both peripheral blood neutrophils and SF neutrophils, preactivated by all three stimulators. The inhibitory effect was unchanged after addition of carbachol. Carbachol itself does not influence the CL of neutrophils isolated from either blood or SF, and preactivated by OZ, but PMA modulates the response of fMLP stimulated cells. It causes a significant increase of both luminol-and lucigenin CL expressed by peripheral blood neutrophils, whereas it reduces the response of SF neutrophils. For the first time, to our knowledge, we showed that the cholinergic agonist, carbachol, has been able to increase CL of peripheral human blood neutrophils. In contrast, the same treatment resulted in a decreased CL of neutrophils isolated from SF of patients with RA. These results support the hypothesis that neutrophils, upon cholinergic stimulation may affect the parasympathetic system – both its proinflammatory, and its anti-inflammatory activities. With the milieu of the cells determining

chemotactic receptors) leads to elevation of the CL reponse (Gajewski et al., 1997).

**4. Immune cells – a new, diffusely expressed adrenergic and cholinergic organ? The neuromodulatory aspects of inflamed joint cavity: the importance** 

**of cholinergic system** 

1995).

the infectious agent (Blalock, 2002, 2005).

present on the same cells (Wybran et al., 1979).

the nature of the stimulation (Gajewski et al., 1997).

induced intra-articular pressure leading to occlusion of the microcirculation is central to this mechanism (Hitchon & El-Gabalawy, 2004). Several observations show that neutrophilsmediated lysis of surrounding cells during ischemia-reperfusion is largely mediated by ROI (Smith et al., 1989). Superoxide dismutase (SOD) can significantly reduce cellular lysis and damage. Treatment with SOD prior the reperfusion reduces the concentration of the potentially injurious ROI**.** Similarly, administration of catalase decreases hydrogen peroxide concentration and reduces tissue damage during reperfusion (Smith et al., 1989). In the investigation of Nguyen (Nguyen et al., 2005) it was concluded that neutrophils play a significant role in injuring cell membranes reloading following periods of unloading, and that this membrane damage was mediated by MPO. A substantial, synergistic effect on the level of muscle lysis when both mechanical loading and neutrophils were applied to muscle cells *in vitro* was found. Loading alone caused only a 1.7% lysis of muscle cells, while coculturing with neutrophils in the absence of loading resulted in only 3.5% lysis. However, loading in the presence of neutrophils resulted in 12.6% lysis of muscle cells under otherwise identical culture conditions in the presence of SOD (Nguyen et al., 2005). These observations provide a direct link between changes in mechanical loads applied to tissue, and an increase in damage that is induced by inflammatory cells. According to these results, pressure could be also considered as an important factor in *in vitro* experiments.

Another factor that is not often considered and influences neutrophils metabolism and survival is the concentration of molecular oxygen used to incubate the cell suspensions. Most *in vitro* experiments are performed in air or air mixtures of 21% O2. Rarely, if ever, will such high concentration of oxygenation occur *in vivo*. For example, the O2 tension within SF has been reported to be low as 1-2%. It was recently showed that local O2 concentration affects neutrophils apoptosis. Only under hypoxic conditions, such as those typically found in RA joints, anti-apoptotic pathways are trigged. Neutrophils normally have a short life in the circulation (8-12h), whereas within SF with physiological oxygen concentration, neutrophils lifetime can be extended, increasing their potential to cause damage and promote inflammation (Cross et al., 2006).

The arguments that the prolonged lifespan of activated neutrophils in patients with RA may contribute to the progression of the inflammatory process to chronicity were supported by experiments in which neutrophils were exposured to SF (50%). In these conditions, in whole blood, neutrophils were stimulated with equal volume of SF and trans-differentiation of neutrophils into dendritic like cells was observed (Iking-Konert et al., 2005). It may be suggested that neutrophils from SF undergo major alterations, including transdifferentiation to cells with dendritic-like characteristics, probably induced by T cell derived cytokines. Exposure of SF, which contained considerable amounts of cytokines, induced a similar receptor pattern on blood derived neutrophils of healthy donors. This effect was also achieved by T cell supernatant alone (Iking-Konert et al., 2005).

Antigen-presenting dendritic cells provide the reducing extracellular microenvironment required for T lymphocyte activation. The mutual activation of neutrophils and T cells might contribute to perpetuation of the local inflammatory process, and eventually to the destructive process in RA (Angellini et al., 2002).

Actual findings confirm that extreme conditions within the joint have influence on neutrophils metabolism. Neutrophils in synovial tissues have different features than blood neutrophils. The proposal to use conditions, as close as possible to physiological conditions, is strongly recommended for *in vitro* experiments on neutrophils role in RA.

induced intra-articular pressure leading to occlusion of the microcirculation is central to this mechanism (Hitchon & El-Gabalawy, 2004). Several observations show that neutrophilsmediated lysis of surrounding cells during ischemia-reperfusion is largely mediated by ROI (Smith et al., 1989). Superoxide dismutase (SOD) can significantly reduce cellular lysis and damage. Treatment with SOD prior the reperfusion reduces the concentration of the potentially injurious ROI**.** Similarly, administration of catalase decreases hydrogen peroxide concentration and reduces tissue damage during reperfusion (Smith et al., 1989). In the investigation of Nguyen (Nguyen et al., 2005) it was concluded that neutrophils play a significant role in injuring cell membranes reloading following periods of unloading, and that this membrane damage was mediated by MPO. A substantial, synergistic effect on the level of muscle lysis when both mechanical loading and neutrophils were applied to muscle cells *in vitro* was found. Loading alone caused only a 1.7% lysis of muscle cells, while coculturing with neutrophils in the absence of loading resulted in only 3.5% lysis. However, loading in the presence of neutrophils resulted in 12.6% lysis of muscle cells under otherwise identical culture conditions in the presence of SOD (Nguyen et al., 2005). These observations provide a direct link between changes in mechanical loads applied to tissue, and an increase in damage that is induced by inflammatory cells. According to these results,

pressure could be also considered as an important factor in *in vitro* experiments.

promote inflammation (Cross et al., 2006).

achieved by T cell supernatant alone (Iking-Konert et al., 2005).

destructive process in RA (Angellini et al., 2002).

Another factor that is not often considered and influences neutrophils metabolism and survival is the concentration of molecular oxygen used to incubate the cell suspensions. Most *in vitro* experiments are performed in air or air mixtures of 21% O2. Rarely, if ever, will such high concentration of oxygenation occur *in vivo*. For example, the O2 tension within SF has been reported to be low as 1-2%. It was recently showed that local O2 concentration affects neutrophils apoptosis. Only under hypoxic conditions, such as those typically found in RA joints, anti-apoptotic pathways are trigged. Neutrophils normally have a short life in the circulation (8-12h), whereas within SF with physiological oxygen concentration, neutrophils lifetime can be extended, increasing their potential to cause damage and

The arguments that the prolonged lifespan of activated neutrophils in patients with RA may contribute to the progression of the inflammatory process to chronicity were supported by experiments in which neutrophils were exposured to SF (50%). In these conditions, in whole blood, neutrophils were stimulated with equal volume of SF and trans-differentiation of neutrophils into dendritic like cells was observed (Iking-Konert et al., 2005). It may be suggested that neutrophils from SF undergo major alterations, including transdifferentiation to cells with dendritic-like characteristics, probably induced by T cell derived cytokines. Exposure of SF, which contained considerable amounts of cytokines, induced a similar receptor pattern on blood derived neutrophils of healthy donors. This effect was also

Antigen-presenting dendritic cells provide the reducing extracellular microenvironment required for T lymphocyte activation. The mutual activation of neutrophils and T cells might contribute to perpetuation of the local inflammatory process, and eventually to the

Actual findings confirm that extreme conditions within the joint have influence on neutrophils metabolism. Neutrophils in synovial tissues have different features than blood neutrophils. The proposal to use conditions, as close as possible to physiological conditions,

is strongly recommended for *in vitro* experiments on neutrophils role in RA.

### **4. Immune cells – a new, diffusely expressed adrenergic and cholinergic organ? The neuromodulatory aspects of inflamed joint cavity: the importance of cholinergic system**

One of the truly remarkable discoveries in modern biology is the finding that the nervous system and the immune system use a common language for intra- and inter-system communication (Blalock, 2005). This biochemical information circuit between neurons and immune cells allows the immune system to function as a sensory organ, and completes our ability to be cognizant not only of the world we can seen hear, taste, touch and smell, but also of bacteria, viruses, antigens or tumor cells. Recognition of such "non-cognitive stimuli" by the immune system results in transmission of information to the central nervous system to cause a physiological response that is ultimately beneficial to the host and detrimental to the infectious agent (Blalock, 2002, 2005).

The idea that immune cells were a source of neuropeptides was viewed by many as heretical; as is often the case when fundamental and unexpected discoveries are made (Blalock, 2005). Concurrent with early observation on production of neuropeptides by immune cells emerging studies reported that neuropeptide/neurotransmitter receptor were present on the same cells (Wybran et al., 1979).

It has been proposed that activity of neutrophils in RA may be influenced by neurotransmitters, including agonists of adrenergic and cholinergic receptors. This conclusion was based on the results of pharmacological studies that suggested the existence of several subtypes of functional adrenergic and cholinergic receptors of human neutrophils. An important modulatory mechanism in neutrophils is the interplay between stimulatory and inhibitory receptors. For example, activation of neutrophils is antagonized by agents that stimulate adenylate cyclase (AC). These agents include B-adrenergic agonists, prostaglandins of the E series, adenosine and histamine. These effects appear to be mediated through adrenergic receptors, activating AC, which leads to an increase in cyclic AMP. Based on the hypothesis of the opposing actions of cAMP and cGMP, it was suggested that the agonists of cholinergic receptors might stimulate leukocyte activation (Gajewski et al., 1995).

The effect of the adrenergic receptors isoproterenol (ISO) and the cholinergic receptor carbachol on ROI production was measured by CL. Activation of neutrophils by OZ (stimulating phagocytosis), PMA (activation of NADPH oxidase) and fMLP (stimulation of chemotactic receptors) leads to elevation of the CL reponse (Gajewski et al., 1997).

The adrenergic receptor antagonist ISO has inhibitory effects on the CL of both peripheral blood neutrophils and SF neutrophils, preactivated by all three stimulators. The inhibitory effect was unchanged after addition of carbachol. Carbachol itself does not influence the CL of neutrophils isolated from either blood or SF, and preactivated by OZ, but PMA modulates the response of fMLP stimulated cells. It causes a significant increase of both luminol-and lucigenin CL expressed by peripheral blood neutrophils, whereas it reduces the response of SF neutrophils. For the first time, to our knowledge, we showed that the cholinergic agonist, carbachol, has been able to increase CL of peripheral human blood neutrophils. In contrast, the same treatment resulted in a decreased CL of neutrophils isolated from SF of patients with RA. These results support the hypothesis that neutrophils, upon cholinergic stimulation may affect the parasympathetic system – both its proinflammatory, and its anti-inflammatory activities. With the milieu of the cells determining the nature of the stimulation (Gajewski et al., 1997).

The Role of Neutrophils in Rheumatoid Arthritis – Experiments *In Vitro*: A Change of Conception? 65

enhanced transcription and translation (Quayle et al., 1997). Consequently, several gene products, such as IL-8 and MMP-9, are up-regulated, allowing not only the up-regulation of cell function but also development of new cellular responses, such as antigen-presentation to T cell via activated MHCII expression (Cascao et al., 2009). Neutrophils isolated from the SF of RA patients expressing MHCII, CD80 and CD86 are able to stimulate T-cell proliferation (Wright et al., 2010). Indeed, the levels of expression of MHCII and costimulatory molecules on neutrophils from SF have been reported to be equivalent to or greater than the levels of expression on monocytes and B cells (Sandilands et al., 2005). Apart from their ability to stimulate T-cells in this way, it is also possible that neutrophils can expose cryptic epitopes, as they possess different proteases than other antigenpresenting cells. Thus, their function within SF could be different from that of other antigen-

Inflammatory reprogramming may increase neutrophil viability. As was shown, 8-17% neutrophils of the global neutrophil population have the potential to persist for more than 72 h under inflammatory conditions (Chakravarti et al., 2008). This is in contrast to the circulating neutrophils whose survival is measured in hours. The mechanism of this persistence remains unknown, but it seems that the protein kinases are largely implicated in survival of these long-lived neutrophils (Cronstein et al., 1992). The phospholipids metabolic pathway leading to leukotriene B4 (LTB4) synthesis also illustrates differences between these long-lived neutrophils and circulating neutrophils. As was shown, a significant amount of the 5-lipoxygenase (5-LO) is localized to the nuclear membrane in long-lived neutrophils, in basal conditions, a phenomenon absent in circulating neutrophils

As was mentioned, neutrophils exposure to SF (50%) induces transdifferentiation of neutrophils into dendritic like cells (Iking-Konert et al., 2005). When the neutrophils were cultured with TNF-α, IFN-γ and IL-4, the resultant cells had morphologic, cytochemical, and phenotypic features of macrophages. In contrast to the starting population, they were negative for myeloperoxidase, specific esterase and lactoferrin. It appears that, in response to the cytokines present in SF, postmitotic neutrophils can become macrophages (Araki et al., 2004). Neutrophils are known to phagocytose invading pathogens and harmful particles. However, in the study of Rydel-Tormanen et al. (Rydel-Tormanen et al., 2006) it was demonstrated that neutrophils are also able to engulf apoptotic neutrophils or cell debris resulting from secondary necrosis of neutrophils. Previously, neutrophils phagocytosing apoptotic cells and nuclei have been described in blood smears from patients with systemic lupus erythematosus (SLE), a feature called LE cells (Bohm, 2004). Moreover, in inflammatory foci, apparently viable neutrophils with phagosomes enclosing were found with what appeared to be whole apoptotic neutrophils and apoptotic nuclei. Neutrophils may thereby contribute to clearance and resolution of inflammation, thus acting as a back up system in situations when the macrophages clearance system is insufficient and/or overwhelmed. It is apparent that neutrophils have the abilities needed to mimic macrophage behavior and express most, if not all, surface receptors used by macrophages in the process of phagocytosis, suggesting the mechanisms to be similar in the two types of

**6. Long-density granulocytes (LDG) and endothelial progenitor cells (EPC)**  Two studies (Denny et al., 2010, Hacbarth & Kajdacsy-Bella, 1986) have reported the presence of an abnormal subset of neutrophils in the peripheral circulation of SLE patients.

presenting cells (Wright et al., 2010).

(Chakravarti et al., 2008).

cells (Rydell-Tormanen et al., 2006).

The observed opposite effects may suggest that different subtypes of muscarinic receptors are expressed on neutrophils presented in blood and synovial fluid, respectively. Because there are five distinct genes encoding muscarinic cholinergic receptors subtypes, we investigated the expression of mRNA encoding these receptors subtypes (i.e. m1-m5) in neutrophils isolated from healthy blood donors and patients with RA. Our results demonstrated for the first time the presence of m3, m4 and m5 muscarinic receptor subtypes in human blood neutrophils (Bany et al., 1999). The lack of mRNA for m4 muscarinic receptor subtype in neutrophils isolated from SF, may contribute to the opposite responses to cholinergic stimuli observed in neutrophils from blood and SF (Gajewski et al., 1997, Bany et al., 1999).

These findings were partially confirmed by Tracey (Tracey, 2002). The molecular dovetail between the cholinergic nervous system and the innate immune system is a nicotinic αbungarotoxin-sensitive macrophage acetylcholine receptor. Exposure of human macrophages, but not peripheral blood monocytes, to nicotine or acetylcholine inhibits the release of TNF-α, IL-1 and IL-8 in response to endotoxin. Tissue macrophages, but not circulating monocytes, produce most of the TNF-α which appears systemically during an excessive inflammatory response. Interaction between the macrophage cholinergic receptor and its ligand inhibits the synthesis of pro-inflammatory cytokines (TNF-α, IL-1 and IL-18) but not anti-inflammatory cytokines (such as IL-10). Acetylcholine inhibits the expression of TNF-α protein in macrophages, but not the induction of TNF-α mRNA levels, indicating that activation of the cholinergic receptor transduces intracellular signals that inhibit cytokine synthesis at a post-transcriptional stage. As compared with macrophages, monocytes are refractory to the cytokine-inhibiting effects of acetylcholine: only supraphysiological concentrations of cholinergic agonist inhibit cytokine synthesis in monocytes (Tracey, 2002).

Further studies on the "reprogramming" of blood neutrophils (contradictory effects of adrenergic and cholinergic systems) into other cell types, SF neutrophils, (both systems inhibiting CL response) are expected to yield new insight into event related to RA therapy.
