**1. Introduction**

More than 20 million Americans suffer, and ultimately die, from chronic kidney disease (CKD). Based on data from the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK), the yearly cost of dialysis treatment of patients with end stage renal disease (ESRD) is currently \$35 billion [1], and this number is predicted to rise as the US population ages and more people develop obesity, metabolic syndrome, and diabetes. CKD is associated with progressive renal fibrosis and inflammation, and currently there is no cure for the disease.

The most common primary illnesses which result in end stage renal disease (ESRD) are diabetes (~37%), hypertension (~24%), glomerulonephritis (~15%), cystic kidney diseases (~4.7%) and urologic diseases (2.5%) [1]. There were 111,000 new ESRD patients diagnosed in 2007 and out of a total of ~500,000 ESRD patients 368,500 people received dialysis treatment in the same year. Dialysis patients have poor quality of life due to high hospitalization rate (458/1000 patients in 2008), high morbidity and mortality (~20%) [1]. Presently, kidney transplant is the only option for these patients to have a close to normal life. According to the US Renal Data System 2010 [1] however, out of the ~85,000 patients awaiting transplant about 18,000 will receive kidney since the amount of available organs did not increase significantly above this number for several years.

Angiotensin converting enzyme inhibitors (ACEIs) and angiotensin receptor blockers (ARBs) are widely used to attenuate the development of cardiovascular diseases and support renal function in CKD patients. However, novel therapeutic targets are desperately needed to effectively treat CKD and slow down disease progression.

Currently, there are about 2,000 clinical trials worldwide addressing some aspects and/or co-morbidities of CKD [2]. These include treatment of anemia, hypertension, secondary hyperparathyroidism, depression and inflammation among others. So far increasing frequency and quality of dialysis did not show advantages in survival rate [2]. Similarly, treatments targeting hypercholesterolemia [3] and hyperhomocysteinemia [4] or the usage of statins [5] failed to increase significantly the survival of ESRD patients.

In recent years, we and others obtained exciting new data on the pathophysiological role of the disintegrin and metalloenzyme ADAMs in renal fibrosis and CKD. This chapter is dedicated to summerize these discoveries and discuss their significance and potential role in the future treatment of patients with renal diseases.

ADAM Proteases as Novel Therapeutic Targets in Chronic Kidney Disease 3

ADAMTSs are secreted proteins which anchor to extracellular matrix molecules through their thrombospondin-1 domain [12] and are involved in proteolytic cleavage of proteoglycans [13], and of the von Willebrand factor [14]. Both protein families can have

There are several ADAM and ADAMTS proteins which expression was shown in the human or murine kidney by various techniques. Histochemical analysis showed that ADAM9 was expressed in the nephron: both in the glomerulus and in tubular epithelial cells [15]. Expression of a short form of the enzyme lacking the cytoplasmic region was also reported in the kidney [16]. ADAM10 expression was first shown in chick kidney [17], in mouse kidney of mesenchymal origin [18] and later in humans in the distal tubule, in the connecting tubule, in the principal cells of the collecting duct and in the thick ascending limb of Henle [19]. ADAM11, which is known as a disintegrin metalloenzyme primarily expressed in the central and peripheral nervous system, was also expressed in the epithelial cells of the collecting duct at a low level [20]. Since ADAM11 is differentially expressed during development, it may have an important role in normal kidney morphogenesis. There is also data on the expression of ADAM13 mRNA in the developing mouse kidney [21]. ADAM17 is a disintegrin metalloenzyme which is ubiquitously expressed in almost all mammalian cells. It is present in the kidney [22] and its expression is upregulated in various renal diseases in humans [23]. The mRNA of ADAM19 was present in developing human kidney, and in the endothelial cells and in cell of the distal tubules of the adult kidney [23]. Expression of ADAM31, another proteolytically active disintegrin metalloenzyme was also identified in the epithelium of the convoluted tubuli [24]. High mRNA level of mouse ADAM33 was also shown in the kidney [25]. Since this protein is catalytically inactive, it

Of the ADAMTS proteins ADAMTS-1 is expressed at high levels in the adult mice kidney [26], and in situ hybridization showed high level of ADAMTS-1 in the epithelia of the developing kidney [27]. In the rat higher level of ADAMTS-1 was observed in the adult animals compared to newborns, and expression pattern of the protease was restricted to the renal medulla and the principal cells of the collecting ducts in the kidney [28]. ADAMTS-5 was observed in glomerular mesangial cells [29]. ADAMTS-9 [30] and ADAMTS-10 [18] are highly expressed in the developing and adult kidney, respectively, similarly to human ADAMTS-14, -15, -16 [31] with no known function at the present. ADAMTS-13 was shown in healthy human kidney samples and in kidneys of patients with thrombotic thrombocytopenic purpura by real-time PCR and immunohistochemistry. ADAMTS-13 was present in the glomeruli as well as in the tubuli [32]. Also, various transcripts of ADAM16

**2.1.1 ADAM and ADAMTS in kidney development - what we learned from knockout** 

There is very few data available on the role of ADAMs and ADAMTS enzymes in kidney development. There is evidence that expression pattern of ADAMTS-1 [27] and ADAM10 [35] and ADAM13 [21] changes in the kidney during development and that ADAMTS-9 is

significant contribution to CKD progression.

**2.1 Expression of ADAM enzymes in the normal kidney** 

may have a role in cell-cell interaction and communication.

were shown in the developing human and rat kidneys [33, 34].

**studies** 

### **2. Physiology of ADAMs and ADAMTS**

ADAMs (a disintegrin and metalloenzymes) and ADAMTS (ADAMs with thrombospondin-1-like domains) are membrane-bound multidomain proteins similar to snake venom metalloenzymes and disintegrins. Both groups have pro-, metalloenzyme-like, disintegrinlike and cysteine-rich domains, but compared to ADAMs ADAMTS do not possess cytoplasmic or transmembrane regions. Catalytically active ADAMs are Zn2+-dependent endopeptidases and are best known for their sheddase activity. They cleave epidermal growth factor ligands, cytokines and their receptors, adhesion molecules and the infamous amyloid precursor protein among others [6]. ADAMs participate in interreceptor crosstalk between G protein coupled receptors (like angiotensin receptors [7], bradykinin receptors [8] and serotonin receptors [9]) and members of the tyrosine kinase receptors (epidermal growth factors receptor, tumor necrosis factor receptor) by shedding membrane-bound proforms of tyrosine kinase ligands (Figure 1). ADAMs are indispensable for normal development, cell proliferation and growth however, at the same time, they can drive pathological cell division and inflammation and have major role in the development of several proliferative and inflammatory diseases [8]. Some of the ADAMs have mutation in their so-called hemopexin-domain (HEXXHXXGXXH) which is responsible for the Zn2+ binding of the protein. These ADAMs are catalytically inactive and may have a role in cellmatrix and cell-cell interactions rather than in proteolytic processes [11].

Fig. 1. ADAMs participate in inter-receptor crosstalk: triple membrane spanning signalling. AII: angiotensin-II, BK: bradykinin; GPCR: G protein-coupled receptor; mGF: membranebound growth factor, sGF: soluble growth factor; EGFR: epidermal growth factor receptor.

ADAMs (a disintegrin and metalloenzymes) and ADAMTS (ADAMs with thrombospondin-1-like domains) are membrane-bound multidomain proteins similar to snake venom metalloenzymes and disintegrins. Both groups have pro-, metalloenzyme-like, disintegrinlike and cysteine-rich domains, but compared to ADAMs ADAMTS do not possess cytoplasmic or transmembrane regions. Catalytically active ADAMs are Zn2+-dependent endopeptidases and are best known for their sheddase activity. They cleave epidermal growth factor ligands, cytokines and their receptors, adhesion molecules and the infamous amyloid precursor protein among others [6]. ADAMs participate in interreceptor crosstalk between G protein coupled receptors (like angiotensin receptors [7], bradykinin receptors [8] and serotonin receptors [9]) and members of the tyrosine kinase receptors (epidermal growth factors receptor, tumor necrosis factor receptor) by shedding membrane-bound proforms of tyrosine kinase ligands (Figure 1). ADAMs are indispensable for normal development, cell proliferation and growth however, at the same time, they can drive pathological cell division and inflammation and have major role in the development of several proliferative and inflammatory diseases [8]. Some of the ADAMs have mutation in their so-called hemopexin-domain (HEXXHXXGXXH) which is responsible for the Zn2+ binding of the protein. These ADAMs are catalytically inactive and may have a role in cell-

matrix and cell-cell interactions rather than in proteolytic processes [11].

Fig. 1. ADAMs participate in inter-receptor crosstalk: triple membrane spanning signalling. AII: angiotensin-II, BK: bradykinin; GPCR: G protein-coupled receptor; mGF: membranebound growth factor, sGF: soluble growth factor; EGFR: epidermal growth factor receptor.

**2. Physiology of ADAMs and ADAMTS** 

ADAMTSs are secreted proteins which anchor to extracellular matrix molecules through their thrombospondin-1 domain [12] and are involved in proteolytic cleavage of proteoglycans [13], and of the von Willebrand factor [14]. Both protein families can have significant contribution to CKD progression.

#### **2.1 Expression of ADAM enzymes in the normal kidney**

There are several ADAM and ADAMTS proteins which expression was shown in the human or murine kidney by various techniques. Histochemical analysis showed that ADAM9 was expressed in the nephron: both in the glomerulus and in tubular epithelial cells [15]. Expression of a short form of the enzyme lacking the cytoplasmic region was also reported in the kidney [16]. ADAM10 expression was first shown in chick kidney [17], in mouse kidney of mesenchymal origin [18] and later in humans in the distal tubule, in the connecting tubule, in the principal cells of the collecting duct and in the thick ascending limb of Henle [19]. ADAM11, which is known as a disintegrin metalloenzyme primarily expressed in the central and peripheral nervous system, was also expressed in the epithelial cells of the collecting duct at a low level [20]. Since ADAM11 is differentially expressed during development, it may have an important role in normal kidney morphogenesis. There is also data on the expression of ADAM13 mRNA in the developing mouse kidney [21]. ADAM17 is a disintegrin metalloenzyme which is ubiquitously expressed in almost all mammalian cells. It is present in the kidney [22] and its expression is upregulated in various renal diseases in humans [23]. The mRNA of ADAM19 was present in developing human kidney, and in the endothelial cells and in cell of the distal tubules of the adult kidney [23]. Expression of ADAM31, another proteolytically active disintegrin metalloenzyme was also identified in the epithelium of the convoluted tubuli [24]. High mRNA level of mouse ADAM33 was also shown in the kidney [25]. Since this protein is catalytically inactive, it may have a role in cell-cell interaction and communication.

Of the ADAMTS proteins ADAMTS-1 is expressed at high levels in the adult mice kidney [26], and in situ hybridization showed high level of ADAMTS-1 in the epithelia of the developing kidney [27]. In the rat higher level of ADAMTS-1 was observed in the adult animals compared to newborns, and expression pattern of the protease was restricted to the renal medulla and the principal cells of the collecting ducts in the kidney [28]. ADAMTS-5 was observed in glomerular mesangial cells [29]. ADAMTS-9 [30] and ADAMTS-10 [18] are highly expressed in the developing and adult kidney, respectively, similarly to human ADAMTS-14, -15, -16 [31] with no known function at the present. ADAMTS-13 was shown in healthy human kidney samples and in kidneys of patients with thrombotic thrombocytopenic purpura by real-time PCR and immunohistochemistry. ADAMTS-13 was present in the glomeruli as well as in the tubuli [32]. Also, various transcripts of ADAM16 were shown in the developing human and rat kidneys [33, 34].

#### **2.1.1 ADAM and ADAMTS in kidney development - what we learned from knockout studies**

There is very few data available on the role of ADAMs and ADAMTS enzymes in kidney development. There is evidence that expression pattern of ADAMTS-1 [27] and ADAM10 [35] and ADAM13 [21] changes in the kidney during development and that ADAMTS-9 is

ADAM Proteases as Novel Therapeutic Targets in Chronic Kidney Disease 5

Besides initiating inflammation, TNF has important pathophysiological role in insulin resistance (reviewed in [40]). After activation by ADAM17, the soluble homotrimer of TNF activates the TNF receptor and downstream signaling molecules. Activation of the MAP kinase pathway initiates serine phosphorylation of the insulin receptor substrate (IRS) intracellularly. Being phosphorylated on serine inhibits tyrosine phosphorylation of the IRS which results in insensitivity of the insulin receptor to extracellular insulin and contributes

the development of diabetes (Figure 3).

Fig. 3. Mechanism of TNF-induced insulin resistance

highly expressed in the mesenchyme of the developing kidney [30]. However, as of present, there is no detail about how knocking down ADAMs influence kidney development.

Targeted knockout of Adamts-1 in mice showed that the enzyme has an important role in kidney development. Deletion of exon 2 (encoding part of the metalloenzyme domain) resulted in lack of ADAMTS-1 protein in mice and high perinatal lethality of the animals due to kidney malfunction [36]. In these animals both the cortical and medullary areas were reduced with concomitant increase in the caliceal space. Another group found that lack of the whole metalloenzyme domain (deletion of exon 2-4) rendered ADAMTS-1 catalytically inactive which resulted in enlarged renal calices and fibrosis of the uteropelvic junction [37]. These animals also developed bilateral hydronephrosis and papillary atrophy shortly after birth [38]. Since normally there is a high level of ADAMTS-1 expressed in the epithelium of the collecting ducts and of the uteropelvic junction, and because the phenotype greatly resembles to symptoms of the human uteropelvic obstruction, these animals can be good models for this genetic disease.

These data also show that targeting strategies can greatly influence the evolving phenotypes.
