**5.1 Pulmonary fibrosis**

Pulmonary fibrosis is the final result of a numerous and heterogeneous group of disorders known as interstitial lung diseases (ILD). Lung fibrotic remodeling is characterized by fibroblast/myofibroblast activation, and excessive extracellular matrix accumulation leading to progressive destruction of the lung architecture and usually terminal outcome (Pardo & Selman, 2002). Idiopathic pulmonary brosis (IPF), the most common form of the idiopathic interstitial pneumonias, is a chronic, progressive, irreversible, and usually lethal lung disease of unknown cause (King, Pardo & Selman 2011). IPF is characterized by the presence of clusters of fibroblasts and myofibroblasts circumscribed from surrounding cells (fibroblastic foci), which represent sites of active fibrogenesis (Selman, King & Pardo, 2001).

During a long time, proliferation of local (resident) fibroblasts and differentiation to myofibroblasts were considered the main source of extracellular matrix deposition in pulmonary fibrosis. The first report of the possible participation of mesenchymal stem cells in the pathogenesis of pulmonary fibrosis, described that collagen-producing cells with broblast characteristics were derived from BM progenitor cells, in a model of bleomycin induced pulmonary fibrosis (Hashimoto et al., 2004). The mice in this model were engrafted with BM from GFP transgenic mice that allow to easily follow the fate of these BM-derived cells. Though this group did not prove that these cells were actually fibrocytes, they recognize the possibility of this premise. Not much later, a work that showed that human CD45+Col I+CXCR4+ circulating fibrocytes were able to migrate to the lung of mice treated with bleomycin was published (Phillips et al., 2004). These authors also described that maximal intrapulmonary recruitment of CD45+Col I+CXCR4+ fibrocytes directly correlated with increased collagen deposition in the lungs. Likewise, they identified a second fibrocyte population that is CD45+Col I+CCR7+ and also traffics to the lungs of bleomycin-treated mice; interestingly the absolute number of CCR7+ fibrocytes found in the fibrotic lung was two to three fold lower than the number of CXCR4+ fibrocytes present under similar conditions, indicating that CXCR4 predominates for the recruitment of fibrocytes to injured lungs (Phillips et al., 2004).

In contrast to acute inammatory reactions, which are characterized by rapidly resolving events; brosis typically results from chronic unsolved inammation or aberrant epithelial activation (King, Pardo & Selman, 2011). Despite having distinct etiological and clinical manifestations, fibrotic remodelling is characterized by fibroblast/myofibroblast activation, and excessive extracellular matrix accumulation leading to scarring formation and

Fibrocytes have become the focus of research of a wide variety of focal and diffuse fibrosing disorders in diverse organs including lung, heart, liver, and kidney (Barth et al., 2005; Sakai et al., 2006, 2008, 2010; Andersson-Sjöland et al., 2008; Scholten et al., 2011); primarily because of their ability to home into tissues and secret extracellular matrix components. More recently however, a large and varied amount of new knowledge about fibrocytes biology has emerged, rising new hypothesis that have enriched the understanding of these

Pulmonary fibrosis is the final result of a numerous and heterogeneous group of disorders known as interstitial lung diseases (ILD). Lung fibrotic remodeling is characterized by fibroblast/myofibroblast activation, and excessive extracellular matrix accumulation leading to progressive destruction of the lung architecture and usually terminal outcome (Pardo & Selman, 2002). Idiopathic pulmonary brosis (IPF), the most common form of the idiopathic interstitial pneumonias, is a chronic, progressive, irreversible, and usually lethal lung disease of unknown cause (King, Pardo & Selman 2011). IPF is characterized by the presence of clusters of fibroblasts and myofibroblasts circumscribed from surrounding cells (fibroblastic foci), which represent sites of active fibrogenesis (Selman,

During a long time, proliferation of local (resident) fibroblasts and differentiation to myofibroblasts were considered the main source of extracellular matrix deposition in pulmonary fibrosis. The first report of the possible participation of mesenchymal stem cells in the pathogenesis of pulmonary fibrosis, described that collagen-producing cells with broblast characteristics were derived from BM progenitor cells, in a model of bleomycin induced pulmonary fibrosis (Hashimoto et al., 2004). The mice in this model were engrafted with BM from GFP transgenic mice that allow to easily follow the fate of these BM-derived cells. Though this group did not prove that these cells were actually fibrocytes, they recognize the possibility of this premise. Not much later, a work that showed that human CD45+Col I+CXCR4+ circulating fibrocytes were able to migrate to the lung of mice treated with bleomycin was published (Phillips et al., 2004). These authors also described that maximal intrapulmonary recruitment of CD45+Col I+CXCR4+ fibrocytes directly correlated with increased collagen deposition in the lungs. Likewise, they identified a second fibrocyte population that is CD45+Col I+CCR7+ and also traffics to the lungs of bleomycin-treated mice; interestingly the absolute number of CCR7+ fibrocytes found in the fibrotic lung was two to three fold lower than the number of CXCR4+ fibrocytes present under similar conditions, indicating that CXCR4 predominates for the recruitment of fibrocytes to injured

**5. Role of fibrocytes in the pathogenesis of fibrotic disorders** 

progressive dysfunction of a given organ.

cells and their participation in fibrotic diseases.

**5.1 Pulmonary fibrosis** 

King & Pardo, 2001).

lungs (Phillips et al., 2004).

Fibrocyte recruitment to damaged lungs has been proved to be mediated by several chemokine/chemokine receptor interactions. Thus, in a model of fluorescein isothiocyanate (FITC)-induced lung fibrosis, it was demonstrated that significantly higher numbers of fibrocytes are present in the airspaces of fluorescein isothiocyanate-injured CCR2+/+ mice compared to CCR2- /- mice (Moore et al., 2005; 2006). Fibrocytes isolated from the lung expressed CCR2 and migrated toward CCL2 and CCL12 ligands. Interestingly, CCL2 stimulated collagen secretion by lung fibrocytes, which differentiated towards a myofibroblast phenotype, transition that was associated with loss of CCR2 expression (Moore et al 2005).

Importantly, interruption of the chemokine axis attenuated both brocyte accumulation and pulmonary brosis (Phillips et al., 2004; Moore et al., 2006), strengthening the notion that these chemokine/chemokine receptor axis are the main responsible of fibrocytes trafficking to the lungs; however, under which biological/pathological conditions one or other chemokine/chemokine receptor system is activated, or if they represent redundant mechanisms, yet remains to be elucidated.

Recently several independent research groups have identied brocytes in different forms of brotic human lung disease. In an initial study, it was reported that circulating brocytes expressing CXCR4 and both lung and plasma levels of CXCL12 were elevated in IPF patients (Mehrad 2007). CXCL12 levels showed a positive correlation with higher number of circulating brocytes in the peripheral blood of these patients. Later, Andersson-Sjöland et al., evaluated the presence of brocytes in the lung of patients with idiopathic pulmonary brosis by immunouorescence and confocal microscopy. Fibrocytes were identied with different combinations of markers in most brotic lungs; interestingly, no brocytes were identied in normal lungs. They also found a positive correlation between the abundance of broblastic foci and the amount of lung brocytes and a negative correlation between plasma levels of CXCL12 with lung function tests (lung diffusing capacity for carbon monoxide and oxygen saturation on exercise) (Andersson-Sjöland et al., 2008). These ndings indicate that circulating brocytes may contribute to the expansion of the broblast/myobroblast population in idiopathic pulmonary brosis.

On the other hand, as mentioned earlier in this chapter, fibrocytes constitutively synthesize and release to the medium important amounts of MMP-2, MMP-7, MMP-8, and MMP-9 (García-de-Alba et al., 2010).

MMPs consist of a large family of zinc endoproteases, collectively capable of degrading all ECM components (Pardo et al., 2006). However, ECM represents only a fraction of their proteolytic targets, and moreover, a given MMP can act on various proteins and, in turn, affect a variety of processes. Gelatinases (MMP-2 and MMP-9) have been found upregulated in human pulmonary brosis and animal models of lung brosis (Swiderski et al., 1998; Selman et al., 2000; Oikonomidi et al., 2009). The overexpression of MMP-2 and MMP-9 has been mainly associated with their capacity to provoke disruption of alveolar epithelial basement membrane and enhanced broblast invasion into the alveolar spaces (Ruiz et al., 2003; Pardo et al., 2006). In the case of fibrocytes, these MMPs may facilitate the process of migration from the circulation to the interstitial and alveolar spaces in response to SDF-1/CXCL12 synthesized by alveolar epithelial cells (Andersson-Sjöland et al., 2008; Garcíade-Alba et al., 2010). TGF-β1–stimulated brocytes signicantly increase gene and protein

Hematopoietic Derived Fibrocytes: Emerging Effector Cells in Fibrotic Disorders 331

(BHR) and airow obstruction. It is among the commonest chronic conditions in Western

It has long been known that architectural and structural remodeling occur in the airways of asthmatic patients. These changes include increased collagen (type III and IV) and bronectin deposition, increased thickness of subepithelial basement membrane, angiogenesis, and brosis. All these processes collectively contribute to severe alterations of the normal bronchial architecture in response to the inammatory tissue injury, leading to progressive airway obstruction and a permanent impairment in respiratory function (Holgate et al., 2009, Hamid & Tulic 2009). Pathologic examination of these tissues demonstrates subepithelial brosis and myobroblast accumulation. Fibrocytes have been identified in the airways of patients with asthma, and it has been reported that allergen exposure induced an increment of brocyte-like cells in the bronchial mucosa of patients with allergic asthma (Shmidt et al., 2003). In a mouse model of allergic asthma, fibrocytes were recruited into the bronchial tissue following allergen exposure and differentiated into myobroblasts providing evidence for the first time that these cells might be a source of myofibroblasts in allergic asthma (Shmidt et al., 2003). Nihlberg and his group showed that fibrocytes in patients with mild asthma were primarily localized, either individually or in clusters, close to the epithelium and to blood vessels. Fibrocyte numbers correlated to the thickness of the basement membrane, supporting that these cells may participate in airway wall remodeling. The increase number of fibrocytes expressing α-SMA seen in patients with increment in the basement membrane thickness may indicate a more differentiated phenotype (Nihilberg et al., 2006). More recently, in two different works, fibrocytes percentages in peripheral blood were shown to be increased in patients with asthma with chronic airway obstruction and severe refractory asthma (Saunders et al., 2008; Chun-Hua et al., 2009). Additionally, a yearly decline in lung function has been significantly associated with the percentage of circulating fibrocytes in patients with chronic obstructive

Renal tubulo-interstitial fibrosis is a non-specific process, representing the common end-stage for kidney diseases, regardless of their etiology. The histological characteristics include the presence of tubular atrophy and dilation, interstitial leukocyte inltration, accumulation of broblasts, and increased interstitial matrix deposition (Strutz et al., 2006). Fibrocytes have also been implicated in the pathogenesis of renal fibrosis in diverse models. For example, in an experimental model of unilateral ureteral obstruction, brocytes appeared in injured parenchyma in a time dependent fashion. Thus, a remarkable number of brocytes dualpositive for CD45 or CD34 and type I collagen inltrated the interstitium, reaching a peak at day 7. Morphological interstitial brosis and collagen content were reduced by almost 50% in mice treated with anti-CCL21 antibodies 7 days after ureteral ligation. A similar reduction was observed in CCR7-null mice (Sakai et al., 2006). Interestingly, most brocytes were positive for CCR7 and CCL21, and the blockade of CCR7 reduced the number of inltrating brocytes indicating that for this organ, CCR7/CCL21 might be the main recruitment axis. The same investigators showed later that fibrocytes might contribute to brosis by an angiotensin II dependent pathway (Sakai et al., 2008). Using two models of renal brosis (unilateral ureteral obstruction and chronic angiotensin II infusion), angiotensin II type 2 receptor (AT2R) decient mice developed increased renal brosis and brocyte inltration and a concomitant

countries affecting 1 from 7 children and 1 from 12 adults (Holgate et al., 2009).

asthma (Saunders et al., 2008).

**5.3 Renal fibrosis** 

expression of both MMP-2 and MMP-9 in vitro. Another putative pathogenic role of these two enzymes is that cell surface localized MMP-2 and MMP-9 can activate latent TGF-β, and this constitutes a mechanism that may operate in normal tissue remodeling as well as in brosis, tumor growth, and invasion (Yu et al., 2000). Fibrocytes also synthesize MMP-7 and MMP-8; the presence of MMP-7 is interesting because this metalloproteinase has been associated with pulmonary brosis since is one of the most up-regulated genes in IPF and display several profibrotic activities (Zuo et al., 2002, Pardo et al., 2005). Moreover, MMP-7 and MMP-1 have been related to alveolar and bronchiolar cell migration over different matrices during IPF lung remodeling (Oikonomidi, 2009). In addition, MMP-7 cleave Ecadherin, which may inuence several aspects of cell behavior, such as epithelial-tomesenchymal transition, which is a well-recognized event that recently has gained importance as a mechanism in the pathogenesis of brosis (Lochter et al.,1997; Noe et al., 2001; Hinz et al., 2007). MMP-8 or collagenase-2 specically degrades brillar collagen types I, II and III, and is known to play an important regulatory role in both acute and chronic inammation (Prikk et al., 2001). In the context of fibrocytes it seems to have an important role in the transmigration of these cells through collagen I (García-de-Alba et al., 2010).

Recently it has been suggested that circulating fibrocytes could have a role as biomarkers for disease severity in IPF; Moeller and coworkers quantified circulating fibrocytes from patients with IPF and found that high percentages of these cells in blood were predictive of poor clinical outcomes; they compared fibrocyte levels in peripheral blood from patients with idiopathic pulmonary fibrosis (stable and during an exacerbation), patients with acute respiratory distress syndrome, and normal controls. Fibrocytes were significantly elevated in patients with stable idiopathic pulmonary fibrosis compared with normal controls, but showed a prominent increase during acute exacerbations of the disease. The number of fibrocytes in patients with acute respiratory distress syndrome was not significantly different from patients with stable idiopathic pulmonary fibrosis or normal controls (Moeller et al., 2009). These data suggest that serial measurements of brocyte percentages may predict acute exacerbations (Moore, 2009). This work was the first to bring up the notion that fibrocytes measurements may be a useful biomarker in this disease but larger studies are needed to confirm this hypothesis.

Finally, a recently published work exploring senescence-accelerated prone mice found increased levels of CXCR4 expressing brocytes in the blood of these mice when compared to wild type controls. The senescence-prone mice also displayed increased lung brosis when exposed to intratracheal bleomycin, suggesting the possibility that the increased number of brocytes contributed to disease. This is an interesting observation since IPF is considered an age related disease (Selman M et al., 2010). Actually, unpublished data from Mathai et al (Mathai et al., as cited in Herzog & Bucala 2010) indicates that the blood of healthy aged individuals contain increased concentrations of CD45+/Col-1+ brocytes and high circulating levels of MCP-1 and IL-13, suggesting that brocytes may be associated with certain aging processes.

#### **5.2 Asthma**

Asthma is an inammatory disorder of the conducting airways which undergo distinct structural and functional changes, leading to non-specic bronchial hyper-responsiveness (BHR) and airow obstruction. It is among the commonest chronic conditions in Western countries affecting 1 from 7 children and 1 from 12 adults (Holgate et al., 2009).

It has long been known that architectural and structural remodeling occur in the airways of asthmatic patients. These changes include increased collagen (type III and IV) and bronectin deposition, increased thickness of subepithelial basement membrane, angiogenesis, and brosis. All these processes collectively contribute to severe alterations of the normal bronchial architecture in response to the inammatory tissue injury, leading to progressive airway obstruction and a permanent impairment in respiratory function (Holgate et al., 2009, Hamid & Tulic 2009). Pathologic examination of these tissues demonstrates subepithelial brosis and myobroblast accumulation. Fibrocytes have been identified in the airways of patients with asthma, and it has been reported that allergen exposure induced an increment of brocyte-like cells in the bronchial mucosa of patients with allergic asthma (Shmidt et al., 2003). In a mouse model of allergic asthma, fibrocytes were recruited into the bronchial tissue following allergen exposure and differentiated into myobroblasts providing evidence for the first time that these cells might be a source of myofibroblasts in allergic asthma (Shmidt et al., 2003). Nihlberg and his group showed that fibrocytes in patients with mild asthma were primarily localized, either individually or in clusters, close to the epithelium and to blood vessels. Fibrocyte numbers correlated to the thickness of the basement membrane, supporting that these cells may participate in airway wall remodeling. The increase number of fibrocytes expressing α-SMA seen in patients with increment in the basement membrane thickness may indicate a more differentiated phenotype (Nihilberg et al., 2006). More recently, in two different works, fibrocytes percentages in peripheral blood were shown to be increased in patients with asthma with chronic airway obstruction and severe refractory asthma (Saunders et al., 2008; Chun-Hua et al., 2009). Additionally, a yearly decline in lung function has been significantly associated with the percentage of circulating fibrocytes in patients with chronic obstructive asthma (Saunders et al., 2008).

#### **5.3 Renal fibrosis**

330 Advances in Hematopoietic Stem Cell Research

expression of both MMP-2 and MMP-9 in vitro. Another putative pathogenic role of these two enzymes is that cell surface localized MMP-2 and MMP-9 can activate latent TGF-β, and this constitutes a mechanism that may operate in normal tissue remodeling as well as in brosis, tumor growth, and invasion (Yu et al., 2000). Fibrocytes also synthesize MMP-7 and MMP-8; the presence of MMP-7 is interesting because this metalloproteinase has been associated with pulmonary brosis since is one of the most up-regulated genes in IPF and display several profibrotic activities (Zuo et al., 2002, Pardo et al., 2005). Moreover, MMP-7 and MMP-1 have been related to alveolar and bronchiolar cell migration over different matrices during IPF lung remodeling (Oikonomidi, 2009). In addition, MMP-7 cleave Ecadherin, which may inuence several aspects of cell behavior, such as epithelial-tomesenchymal transition, which is a well-recognized event that recently has gained importance as a mechanism in the pathogenesis of brosis (Lochter et al.,1997; Noe et al., 2001; Hinz et al., 2007). MMP-8 or collagenase-2 specically degrades brillar collagen types I, II and III, and is known to play an important regulatory role in both acute and chronic inammation (Prikk et al., 2001). In the context of fibrocytes it seems to have an important role in the transmigration of these cells through collagen I (García-de-Alba et al., 2010).

Recently it has been suggested that circulating fibrocytes could have a role as biomarkers for disease severity in IPF; Moeller and coworkers quantified circulating fibrocytes from patients with IPF and found that high percentages of these cells in blood were predictive of poor clinical outcomes; they compared fibrocyte levels in peripheral blood from patients with idiopathic pulmonary fibrosis (stable and during an exacerbation), patients with acute respiratory distress syndrome, and normal controls. Fibrocytes were significantly elevated in patients with stable idiopathic pulmonary fibrosis compared with normal controls, but showed a prominent increase during acute exacerbations of the disease. The number of fibrocytes in patients with acute respiratory distress syndrome was not significantly different from patients with stable idiopathic pulmonary fibrosis or normal controls (Moeller et al., 2009). These data suggest that serial measurements of brocyte percentages may predict acute exacerbations (Moore, 2009). This work was the first to bring up the notion that fibrocytes measurements may be a useful biomarker in this disease but larger

Finally, a recently published work exploring senescence-accelerated prone mice found increased levels of CXCR4 expressing brocytes in the blood of these mice when compared to wild type controls. The senescence-prone mice also displayed increased lung brosis when exposed to intratracheal bleomycin, suggesting the possibility that the increased number of brocytes contributed to disease. This is an interesting observation since IPF is considered an age related disease (Selman M et al., 2010). Actually, unpublished data from Mathai et al (Mathai et al., as cited in Herzog & Bucala 2010) indicates that the blood of healthy aged individuals contain increased concentrations of CD45+/Col-1+ brocytes and high circulating levels of MCP-1 and IL-13, suggesting that brocytes may be associated

Asthma is an inammatory disorder of the conducting airways which undergo distinct structural and functional changes, leading to non-specic bronchial hyper-responsiveness

studies are needed to confirm this hypothesis.

with certain aging processes.

**5.2 Asthma** 

Renal tubulo-interstitial fibrosis is a non-specific process, representing the common end-stage for kidney diseases, regardless of their etiology. The histological characteristics include the presence of tubular atrophy and dilation, interstitial leukocyte inltration, accumulation of broblasts, and increased interstitial matrix deposition (Strutz et al., 2006). Fibrocytes have also been implicated in the pathogenesis of renal fibrosis in diverse models. For example, in an experimental model of unilateral ureteral obstruction, brocytes appeared in injured parenchyma in a time dependent fashion. Thus, a remarkable number of brocytes dualpositive for CD45 or CD34 and type I collagen inltrated the interstitium, reaching a peak at day 7. Morphological interstitial brosis and collagen content were reduced by almost 50% in mice treated with anti-CCL21 antibodies 7 days after ureteral ligation. A similar reduction was observed in CCR7-null mice (Sakai et al., 2006). Interestingly, most brocytes were positive for CCR7 and CCL21, and the blockade of CCR7 reduced the number of inltrating brocytes indicating that for this organ, CCR7/CCL21 might be the main recruitment axis. The same investigators showed later that fibrocytes might contribute to brosis by an angiotensin II dependent pathway (Sakai et al., 2008). Using two models of renal brosis (unilateral ureteral obstruction and chronic angiotensin II infusion), angiotensin II type 2 receptor (AT2R) decient mice developed increased renal brosis and brocyte inltration and a concomitant

Hematopoietic Derived Fibrocytes: Emerging Effector Cells in Fibrotic Disorders 333

Deposition and remodeling of connective tissue in the heart plays a critical role in cardiac repair and response to injury. Fibrosis also occurs on a reactive basis around coronary vessels (perivascular fibrosis) and in the interstitial space (Haudek et al., 2006). It is generally considered that both reactive and reparative fibrosis may contribute to adverse remodeling. A number of studies have supported the contribution of bone marrow progenitor cells or fibrocytes to remodeling in diverse areas of the cardiovascular system

In a mouse model of brotic ischemia/reperfusion cardiomyopathy (I/RC) it was observed a prolonged elevation of MCP-1, and concomitantly a population of small spindle-shaped broblasts with a distinct phenotype appeared in the sites of lesion. These cells were highly proliferative and expressed collagen I and α-smooth muscle actin as well as CD34, and CD45; these cells represented 3% of all non myocyte live cells. Haudek and coworkers confirmed the bone marrow origin of these cells creating a chimeric mice expressing lacZ; I/RC injury resulted in a large population of spindle-shaped broblasts containing lacZ. Interestingly, the administration of SAP in vivo markedly reduced the number of proliferative spindle-shaped broblasts and completely prevented I/RCinduced brosis and global ventricular dysfunction (Haudek et al., 2006). Similar results were reported later, in a model induced by Ang-II. Ang-II infusion resulted in the appearance of bone marrow-derived CD34+/CD45+ broblasts that expressed collagen type I and the cardiac broblast marker DDR2 while local broblasts were CD34−/CD45−. Genetic deletion of MCP-1 (MCP-1-deficient mice) prevented the Ang-IIinduced cardiac brosis and the appearance of CD34+/CD45+ broblasts. Interestingly, Ang-II-treated hearts showed induction of types I and III collagens, TGF-β1, and TNF mRNA expression. Apparently the differentiation of a CD34+/CD45+ broblast precursor population in the heart is induced by Ang-II and mediated by MCP-1 (Haudek et al.,

Neointimal hyperplasia is a common feature of various cardiovascular diseases such as atherosclerosis, postangioplasty restenosis and transplant arteriopathy. Neointima usually consists of smooth muscle cells and deposited extracellular matrix. In an in vivo ovine model of carotid artery synthetic patch graft, circulating leukocytes were shown to express collagen and α-SMA. Importantly, these cells also expressed markers unique to fibrocytes (CD34, CD45, vimentin; Varcoe et al., 2006), suggesting an association between intimal hyperplasia and fibrocyte migration. In other work performed in a rat model of transplant vasculopathy, accelerated transplant vasculopathy was associated with increased levels of host-endothelial chimerism and increased neointimal smooth muscle cell proliferation; moreover, accelerated transplant vasculopathy was associated with increased frequency of

CD34+ brocyte-like cells are detectable in normal mitral valves. In cases of myxomatous degeneration CD34+ brocytes make up the majority of mitral valve stromal cells (Barth et al., 2005). Since major factors in the development of myxomatous valve degeneration are the MMP-9 and collagen I and III, which are secreted by CD34+ brocytes, they propose that these cells might be involved in the pathogenesis of myxomatous mitral valve (Barth et al.,

**5.5 Cardiovascular disease** 

2010).

2005).

where fibrotic response seems to be a common feature.

circulating CD45+vimentin+ brocytes (Onauta et al., 2009).

upregulation of procollagen type I compared with wild-type mice. Pharmacologic inhibition of angiotensin II type 1 receptor (AT1R) with valsartan reduced the degree of renal brosis and the number of brocytes in both the kidney and the bone marrow. In isolated human brocytes, inhibition of AT2R signalling increased the angiotensin II-stimulated expression of type I collagen, whereas inhibition of AT1R decreased collagen synthesis. These results suggest that AT1R/AT2R signalling may contribute to the pathogenesis of renal brosis by at least two fibrocytes-related mechanisms: by regulating the number of brocytes in the bone marrow, and by activation of these cells in the tissues (Sakai et al. 2008).

More recently, the presence of fibrocytes was investigated by immunohistochemistry in kidney biopsy specimens from 100 patients with chronic disease; in addition 6 patients with thin basement membrane disease were studied as a disease control. In patients with chronic kidney disease, the infiltration of fibrocytes was observed mainly in the interstitium and their numbers were higher than that in patients with thin basement membrane disease. Moreover, there was an inverse correlation between the number of interstitial fibrocytes and kidney function at the time of biopsy (Sakai et al., 2010). These results suggest that fibrocytes may be involved in the pathogenesis of human chronic kidney disease though the mechanisms involved in their participation are yet to be studied.

CD34+spindle-shaped cells have also been detected in tubulointerstitial lesions in patients with renal interstitial fibrosis. Although in this work the complete phenotype corresponding to fibrocytes was not documented, it is possible that the described CD34+ cells were actually fibrocytes (Okona et al., 2003).

### **5.4 Liver fibrosis**

Hepatic brogenesis represents a wound-healing response of liver to a variety of insults. The net accumulation of extracellular matrix (ECM) in liver injury arises from increased synthesis by activated hepatic stellate cells and other hepatic brogenic cell types, as well as from bone marrow and circulating brocytes (Guo & Friedman, 2007).

Fibrocytes participation in liver fibrosis is a growing field of research and has been assessed in different models. In a murine model of bile duct ligation-induced liver brosis, investigators found bone marrow derived collagen-expressing GFP+ cells in the liver of chimeric mice (Kisseleva et al., 2006). The majority of these bone marrow derived cells coexpressed collagen-GFP+ and CD45+, suggesting that collagen-producing brocytes were recruited from the bone marrow to the damaged liver (Kisseleva et al., 2006). Later, fibrocyte migration in response to liver injury was investigated using bone marrow (BM) from chimeric mice expressing luciferase (Col-Luc-wt) or green fluorescent protein (Col-GFP-wt) under control of the α1(I) collagen promoter and enhancer, respectively. Migration of CD45+Col I+ fibrocytes was regulated by chemokine receptors CCR2 and CCR1. In addition to CCR2 and CCR1, egress of BM CD45+Col I+ cells was regulated by TGF-β and liposaccharide in vitro and in vivo. Interestingly, development of liver fibrosis was also increased in aged mice and correlated with high numbers of liver fibrocytes (Kisseleva et al., 2011). However, it is unknown what proportion of tissue myobroblasts/brocytes are derived from bone marrow or circulating brocytes, whether myobroblasts of these origins transition through a stellate cell phenotype, and what happens to activated myobroblasts from various sources when liver injury resolves (Guo & Friedman, 2007).

#### **5.5 Cardiovascular disease**

332 Advances in Hematopoietic Stem Cell Research

upregulation of procollagen type I compared with wild-type mice. Pharmacologic inhibition of angiotensin II type 1 receptor (AT1R) with valsartan reduced the degree of renal brosis and the number of brocytes in both the kidney and the bone marrow. In isolated human brocytes, inhibition of AT2R signalling increased the angiotensin II-stimulated expression of type I collagen, whereas inhibition of AT1R decreased collagen synthesis. These results suggest that AT1R/AT2R signalling may contribute to the pathogenesis of renal brosis by at least two fibrocytes-related mechanisms: by regulating the number of brocytes in the bone

More recently, the presence of fibrocytes was investigated by immunohistochemistry in kidney biopsy specimens from 100 patients with chronic disease; in addition 6 patients with thin basement membrane disease were studied as a disease control. In patients with chronic kidney disease, the infiltration of fibrocytes was observed mainly in the interstitium and their numbers were higher than that in patients with thin basement membrane disease. Moreover, there was an inverse correlation between the number of interstitial fibrocytes and kidney function at the time of biopsy (Sakai et al., 2010). These results suggest that fibrocytes may be involved in the pathogenesis of human chronic kidney disease though the

CD34+spindle-shaped cells have also been detected in tubulointerstitial lesions in patients with renal interstitial fibrosis. Although in this work the complete phenotype corresponding to fibrocytes was not documented, it is possible that the described CD34+ cells were actually

Hepatic brogenesis represents a wound-healing response of liver to a variety of insults. The net accumulation of extracellular matrix (ECM) in liver injury arises from increased synthesis by activated hepatic stellate cells and other hepatic brogenic cell types, as well as

Fibrocytes participation in liver fibrosis is a growing field of research and has been assessed in different models. In a murine model of bile duct ligation-induced liver brosis, investigators found bone marrow derived collagen-expressing GFP+ cells in the liver of chimeric mice (Kisseleva et al., 2006). The majority of these bone marrow derived cells coexpressed collagen-GFP+ and CD45+, suggesting that collagen-producing brocytes were recruited from the bone marrow to the damaged liver (Kisseleva et al., 2006). Later, fibrocyte migration in response to liver injury was investigated using bone marrow (BM) from chimeric mice expressing luciferase (Col-Luc-wt) or green fluorescent protein (Col-GFP-wt) under control of the α1(I) collagen promoter and enhancer, respectively. Migration of CD45+Col I+ fibrocytes was regulated by chemokine receptors CCR2 and CCR1. In addition to CCR2 and CCR1, egress of BM CD45+Col I+ cells was regulated by TGF-β and liposaccharide in vitro and in vivo. Interestingly, development of liver fibrosis was also increased in aged mice and correlated with high numbers of liver fibrocytes (Kisseleva et al., 2011). However, it is unknown what proportion of tissue myobroblasts/brocytes are derived from bone marrow or circulating brocytes, whether myobroblasts of these origins transition through a stellate cell phenotype, and what happens to activated myobroblasts

marrow, and by activation of these cells in the tissues (Sakai et al. 2008).

mechanisms involved in their participation are yet to be studied.

from bone marrow and circulating brocytes (Guo & Friedman, 2007).

from various sources when liver injury resolves (Guo & Friedman, 2007).

fibrocytes (Okona et al., 2003).

**5.4 Liver fibrosis** 

Deposition and remodeling of connective tissue in the heart plays a critical role in cardiac repair and response to injury. Fibrosis also occurs on a reactive basis around coronary vessels (perivascular fibrosis) and in the interstitial space (Haudek et al., 2006). It is generally considered that both reactive and reparative fibrosis may contribute to adverse remodeling. A number of studies have supported the contribution of bone marrow progenitor cells or fibrocytes to remodeling in diverse areas of the cardiovascular system where fibrotic response seems to be a common feature.

In a mouse model of brotic ischemia/reperfusion cardiomyopathy (I/RC) it was observed a prolonged elevation of MCP-1, and concomitantly a population of small spindle-shaped broblasts with a distinct phenotype appeared in the sites of lesion. These cells were highly proliferative and expressed collagen I and α-smooth muscle actin as well as CD34, and CD45; these cells represented 3% of all non myocyte live cells. Haudek and coworkers confirmed the bone marrow origin of these cells creating a chimeric mice expressing lacZ; I/RC injury resulted in a large population of spindle-shaped broblasts containing lacZ. Interestingly, the administration of SAP in vivo markedly reduced the number of proliferative spindle-shaped broblasts and completely prevented I/RCinduced brosis and global ventricular dysfunction (Haudek et al., 2006). Similar results were reported later, in a model induced by Ang-II. Ang-II infusion resulted in the appearance of bone marrow-derived CD34+/CD45+ broblasts that expressed collagen type I and the cardiac broblast marker DDR2 while local broblasts were CD34−/CD45−. Genetic deletion of MCP-1 (MCP-1-deficient mice) prevented the Ang-IIinduced cardiac brosis and the appearance of CD34+/CD45+ broblasts. Interestingly, Ang-II-treated hearts showed induction of types I and III collagens, TGF-β1, and TNF mRNA expression. Apparently the differentiation of a CD34+/CD45+ broblast precursor population in the heart is induced by Ang-II and mediated by MCP-1 (Haudek et al., 2010).

Neointimal hyperplasia is a common feature of various cardiovascular diseases such as atherosclerosis, postangioplasty restenosis and transplant arteriopathy. Neointima usually consists of smooth muscle cells and deposited extracellular matrix. In an in vivo ovine model of carotid artery synthetic patch graft, circulating leukocytes were shown to express collagen and α-SMA. Importantly, these cells also expressed markers unique to fibrocytes (CD34, CD45, vimentin; Varcoe et al., 2006), suggesting an association between intimal hyperplasia and fibrocyte migration. In other work performed in a rat model of transplant vasculopathy, accelerated transplant vasculopathy was associated with increased levels of host-endothelial chimerism and increased neointimal smooth muscle cell proliferation; moreover, accelerated transplant vasculopathy was associated with increased frequency of circulating CD45+vimentin+ brocytes (Onauta et al., 2009).

CD34+ brocyte-like cells are detectable in normal mitral valves. In cases of myxomatous degeneration CD34+ brocytes make up the majority of mitral valve stromal cells (Barth et al., 2005). Since major factors in the development of myxomatous valve degeneration are the MMP-9 and collagen I and III, which are secreted by CD34+ brocytes, they propose that these cells might be involved in the pathogenesis of myxomatous mitral valve (Barth et al., 2005).

Hematopoietic Derived Fibrocytes: Emerging Effector Cells in Fibrotic Disorders 335

Fibrocytes have been also identied in the skin of patients with cutaneous brosing diseases, such as nephrogenic systemic brosis. Nephrogenic systemic brosis (NSF) is a recently described cutaneous brosing disorder that exhibits pathologic similarities with scleroderma but occurs exclusively in patients with renal insufficiency who have received gadolinium containing magnetic resonance contrast agents. The onset of the disease varies from days to several months following exposure to gadolinium-based contrast. It is a debilitating disease characterized by the development of discolored plaques on the skin of the extremities and trunk. Over time, contractures develop and complete loss of range of motion can occur (Cowper & Bucala, 2003; Cowper et al., 2008). Skin biopsies from patients with this disease have revealed an important accumulation of CD34, pro-Col-I+ brocytes in the dermis with abundant connective tissue matrix production; it is noteworthy that in vitro studies revealed that gadolinium may decrease the ability of endogenous mediators, such as SAP and IL-12, to inhibit brocyte outgrowth (Vakil et al., 2009). The reason for why brocytes are present in high numbers and are such a prominent feature of the dermatopathology of NSF remains unclear, but may be due to the acute and abrupt

The study of fibrocytes and their participation in the pathogenesis of chronic inflammation and fibroproliferative diseases presents both important challenges and opportunities for researchers. To advance this eld, detailed molecular characterization of these cells and establishment of dened experimental strategies in animals and humans will be necessary to catalyze progress in this area of investigation. Recent studies and emerging concepts have significantly improved our understanding of the participation of fibrocytes in health and disease and so have opened the door to new hypotheses and approaches aimed at

One of the main therapeutic targets, suggested since the initial works on fibrocyte biology research, was the serum amyloid P (SAP), a member of the pentraxin family of proteins. In this context, it was first demonstrated that SAP could inhibit the differentiation of monocytes into fibrocytes (Pilling et al., 2003). SAP binds to Fcγ receptors through which apparently mediates its anti-fibrotic activities affecting peripheral blood monocyte differentiation and activation states (Lu et al., 2008). In a rat model of bleomycin-induced lung injury it was shown that purified rat SAP could suppress development of lung fibrosis which correlated with reduced fibrocyte numbers within the lung tissue (Pilling et al., 2007). More recently, SAP ability to reduce fibrosis was tested in models of renal and lung fibrosis where this therapeutic potential was confirmed. In both models, the mechanisms through which SAP exerts its antifibrotic effect seemed to be independent of monocyte to fibrocyte differentiation (Casraño et al., 2009; Murray 2010). Further analysis of this molecule and its potential as antifibrogenic therapy is needed to identify all the mechanisms involved in its

Several chemokines are abundantly expressed in experimental models of fibrosis and in the human disease (Agostini & Gurrieri 2006). Regarding fibrocytes, several studies have focused on the role of chemokines in recruiting these cells to the injured lung. In human IPF, the

**5.7 Nephrogenic systemic fibrosis** 

development of skin brosis (Bucala, 2008).

therapeutic targets and strategies.

**6. Opportunities for research and therapeutic targets** 

effect as well as the feasibility of its use in human disease.

#### **5.6 Skin disease**

Fibrocytes are thought to play a role in skin repair by several mechanisms such as the secretion of ECM, antigen presentation, cytokine production, angiogenesis, and wound closure (Metz, 2003). After the original work by Bucala, several groups examined the participation of fibrocytes in the wound healing process. Mori and coworkers examined the phenotype of fibrocytes and myofibroblasts present in the wounded skin of BALB/c mice and observed that during wound healing, between 4 and 7 days post-wounding, more than 50% of the cells present at the site of injury were CD13+/collagen I+ fibrocytes that could be isolated at an early stage of the healing process from digested fragments of wounded tissue by fluorescence-activated cell sorting (Mori et al., 2005). Fibrocytes have been identified in postburn hypertrophic scar tissue but were absent from normal skin, moreover, the number of fibrocytes was higher in hypertrophic than in mature scar tissue (Yang et al., 2005). It is noteworthy that over time the expression of CD34 on brocytes present in these wounds decreases, whereas the expression of proline-4-hydroxylase (an enzyme involved in collagen synthesis) increases in both hypertrophic or keloid scars (Aiba and Tagami, 1997). This finding has been corroborated by other authors (Abe et al., 2001; Phillips et al., 2004) and it's an important feature to be considered for the analysis of these cells in organ fibrosis. In other words, it seems that fibrocytes, once in the tissues, progressively lose their typical markers and can be difficult to identify.

Also, the participation of fibrocytes in wound healing of human skin has been postulated as a useful marker for wound age determination in the legal pathology area. In an interesting study (Ishida et al., 2009) a double-color immunofluorescence analysis was carried out using anti-CD45 and anti-collagen type I antibodies to examine the time-dependent appearance of fibrocytes in 53 human skin wounds with different wound ages. Fibrocytes were initially observed in wounds aged 4 days, and their number increased in lesions proportionally with advances in wound age. These findings imply that human skin wounds containing fibrocytes are at least 4 days old. Moreover, a fibrocyte number of over 10 indicates a wound age between 9 and 14 days. Fibrocytes numbers, evaluated with these markers (CD45+/Col I+) showed a marked decrease from day 17 to 21 which was the longest time of evaluation, exposing the need to use other parameters to confirm the wound ages since fibrocytes numbers in day 4 were similar to numbers in day 17-21.

Yang and his group reported high percentages of brocytes present in the cultures of peripheral blood mononuclear cells obtained from burn patients compared with controls (89.7 +/- 7.9% versus 69.9 +/- 14.7%, p < 0.001) and this percentages were consistently higher in patients with more than 30% extent of burn; moreover, they found a positive correlation between the levels of serum TGF-1 and the percentage of fibrocytes developed in the cultures of PBMC derived from these patients (Yang et al., 2002). Interestingly, it has been postulated that the principal role of fibrocytes in burn injury as well as in hypertrophic scars is the regulation of the function of local fibroblasts. Thus, dermal broblasts treated with conditioned medium obtained from burn patient brocytes, but not by those derived from normal subjects, showed an increase in cell proliferation and migration (Wang et al., 2007). Furthemore, it has been suggested that fibrocytes can be reprogrammed by changes in the culture media, and that this reprogrammed fibrocytes have the ability to increase cell proliferation and MMP-1 expression in dermal broblasts (Medina, A & Ghahar, A. 2010). These findings have opened a new research line worthy of follow up.

#### **5.7 Nephrogenic systemic fibrosis**

334 Advances in Hematopoietic Stem Cell Research

Fibrocytes are thought to play a role in skin repair by several mechanisms such as the secretion of ECM, antigen presentation, cytokine production, angiogenesis, and wound closure (Metz, 2003). After the original work by Bucala, several groups examined the participation of fibrocytes in the wound healing process. Mori and coworkers examined the phenotype of fibrocytes and myofibroblasts present in the wounded skin of BALB/c mice and observed that during wound healing, between 4 and 7 days post-wounding, more than 50% of the cells present at the site of injury were CD13+/collagen I+ fibrocytes that could be isolated at an early stage of the healing process from digested fragments of wounded tissue by fluorescence-activated cell sorting (Mori et al., 2005). Fibrocytes have been identified in postburn hypertrophic scar tissue but were absent from normal skin, moreover, the number of fibrocytes was higher in hypertrophic than in mature scar tissue (Yang et al., 2005). It is noteworthy that over time the expression of CD34 on brocytes present in these wounds decreases, whereas the expression of proline-4-hydroxylase (an enzyme involved in collagen synthesis) increases in both hypertrophic or keloid scars (Aiba and Tagami, 1997). This finding has been corroborated by other authors (Abe et al., 2001; Phillips et al., 2004) and it's an important feature to be considered for the analysis of these cells in organ fibrosis. In other words, it seems that fibrocytes, once in the tissues, progressively lose their typical markers

Also, the participation of fibrocytes in wound healing of human skin has been postulated as a useful marker for wound age determination in the legal pathology area. In an interesting study (Ishida et al., 2009) a double-color immunofluorescence analysis was carried out using anti-CD45 and anti-collagen type I antibodies to examine the time-dependent appearance of fibrocytes in 53 human skin wounds with different wound ages. Fibrocytes were initially observed in wounds aged 4 days, and their number increased in lesions proportionally with advances in wound age. These findings imply that human skin wounds containing fibrocytes are at least 4 days old. Moreover, a fibrocyte number of over 10 indicates a wound age between 9 and 14 days. Fibrocytes numbers, evaluated with these markers (CD45+/Col I+) showed a marked decrease from day 17 to 21 which was the longest time of evaluation, exposing the need to use other parameters to confirm the wound ages since fibrocytes

Yang and his group reported high percentages of brocytes present in the cultures of peripheral blood mononuclear cells obtained from burn patients compared with controls (89.7 +/- 7.9% versus 69.9 +/- 14.7%, p < 0.001) and this percentages were consistently higher in patients with more than 30% extent of burn; moreover, they found a positive correlation between the levels of serum TGF-1 and the percentage of fibrocytes developed in the cultures of PBMC derived from these patients (Yang et al., 2002). Interestingly, it has been postulated that the principal role of fibrocytes in burn injury as well as in hypertrophic scars is the regulation of the function of local fibroblasts. Thus, dermal broblasts treated with conditioned medium obtained from burn patient brocytes, but not by those derived from normal subjects, showed an increase in cell proliferation and migration (Wang et al., 2007). Furthemore, it has been suggested that fibrocytes can be reprogrammed by changes in the culture media, and that this reprogrammed fibrocytes have the ability to increase cell proliferation and MMP-1 expression in dermal broblasts (Medina, A & Ghahar, A. 2010).

**5.6 Skin disease** 

and can be difficult to identify.

numbers in day 4 were similar to numbers in day 17-21.

These findings have opened a new research line worthy of follow up.

Fibrocytes have been also identied in the skin of patients with cutaneous brosing diseases, such as nephrogenic systemic brosis. Nephrogenic systemic brosis (NSF) is a recently described cutaneous brosing disorder that exhibits pathologic similarities with scleroderma but occurs exclusively in patients with renal insufficiency who have received gadolinium containing magnetic resonance contrast agents. The onset of the disease varies from days to several months following exposure to gadolinium-based contrast. It is a debilitating disease characterized by the development of discolored plaques on the skin of the extremities and trunk. Over time, contractures develop and complete loss of range of motion can occur (Cowper & Bucala, 2003; Cowper et al., 2008). Skin biopsies from patients with this disease have revealed an important accumulation of CD34, pro-Col-I+ brocytes in the dermis with abundant connective tissue matrix production; it is noteworthy that in vitro studies revealed that gadolinium may decrease the ability of endogenous mediators, such as SAP and IL-12, to inhibit brocyte outgrowth (Vakil et al., 2009). The reason for why brocytes are present in high numbers and are such a prominent feature of the dermatopathology of NSF remains unclear, but may be due to the acute and abrupt development of skin brosis (Bucala, 2008).
