Search

Barrett’s esophagus (BE) is a common premalignant condition of the distal esophagus with a high incidence among Caucasian males. It is an important risk factor for the development of esophageal adenocarcinoma (EAC). EACs arise in BE following a multistep sequence through low- and high-grade dysplasia, finally culminating in an invasive adenocarcinoma. Because of this increased risk of developing an EAC and poor prognosis of EAC patients, BE patients are offered endoscopic surveillance in order to detect early signs of EAC development. However, recent studies have shown that these surveillance programs are not cost-effective and not able to reduce EAC-related mortality. This thesis contains several studies with the aims: 1) to enhance our understanding of BE development and neoplastic progression towards EAC, 2) to contribute to developing a risk stratification index that may eventually replace the endoscopic surveillance program, 3) to improve current treatment modalities, and 4) to predict the prognosis of EAC patients. In order to fulfill these goals, we performed studies to discover genetic and epigenetic changes that contribute to the development and/or neoplastic progression of BE, and EAC progression, and investigated the function of these changes using in vitro models. The dissertation launches with a review article that provides an overview of traditional and novel hypotheses on the development of BE. Also, the role of activated developmental signaling pathways are discussed. In this thesis it is further demonstrated that next-generation sequencing (NGS) can be utilized to create microRNA (miR) profiles and genomic landscapes of endoscopically obtained mucosal biopsies consisting of neoplasia. In chapter 3, using NGS several miRs that may play a role in the development of BE are identified. In addition, it is shown that miR-203 is down-regulated in all columnar epithelial lesions of the esophagus and may initiate BE through activation of Hedgehog signaling and attenuation of intestinal differentiation markers. The objective of the study presented in chapter 4 was to uncover genomic alterations during BE pathogenesis and neoplastic progression of BE. The data suggested that genomic alterations in the metaplasia – carcinoma sequence of BE occur much earlier than the histological changes of frank dysplasia. Furthermore, it was demonstrated that ARID1A inactivation represents a driver event in a subset of the EAC patients and enhances cell growth and invasion of EAC cells. The mutations identified in this chapter may be utilized to predict neoplastic progression of BE. In chapter 6 and 7 it is shown that genetic and epigenetic markers can also function as predictors of therapy sensitivity and patient outcome, respectively. We determined that miR-223 is up-regulated during neoplastic progression of BE and modulates the sensitivity to DNA-damaging agents by targeting PARP1. The results of this study suggest that EAC patients with high miR-223 levels in their tumors might benefit from current chemotherapies. In chapter 7 it is demonstrated that MUC16 presence bears on the prognosis of digestive adenocarcinoma patients.

Like all types of cancer, pancreatic ductal adenocarcinoma (PDAC), the most common pancreatic malignancy, is a disease of the genes and the genetic alterations that are involved in the development of PDAC have been under investigation for many years. The research described in this thesis focuses on these genetic alterations and their clinical relevance. In order to provide an overview of the current insights in pancreatic carcinogenesis, the review articles in Chapter 3 and Chapter 4 are included. In Chapter 5, multivariate analysis was conducted investigating the effects on survival of both clinical and histological features of PDAC. The study showed that lymph node ratio

Chronic kidney disease is an important cause of morbidity and mortality, and its prevalence is increasing. The histopathological hallmark of chronic kidney disease is renal fibrosis, a process of extracellular matrix formation and atrophy of functional renal tissue resulting from sustained injury. The pathophysiology of this process is complex, with many mediators and mechanisms involved in remodeling of extracellular matrix. In this thesis, we focus on the role of the antifibrotic bone morphogenetic proteins (BMPs, particularly BMP-6) and the profibrotic connective tissue growth factor (CTGF) in renal fibrogenesis, via experimental and human studies. Data are presented on the interplay between BMP-expression levels, oxidative stress and inflammation in a mouse model of obstructive nephropathy. The impact of BMP-6 knockdown on renal fibrosis was studied by implementing unilateral ureteral obstruction (UUO) in BMP-6 null and wild-type mice. There was increased inflammation in obstructed kidneys of BMP-6 null compared to wild-type animals, with upregulation of genes and proteins involved in matrix formation. There was also increased tubulointerstitial damage and an altered, profibrotic growth factor profile. Exploring the mechanisms by which BMP-6 deficiency exerts its harmful effects, we noticed increased levels of circulating myofibroblast progenitor cells and increased renal iron deposition with increased oxidative stress responsive gene expression. To study the effects of genetic CTGF attenuation on fibrosis in the context of chronic and severe renal injury, transgenic CTGF+/- and wild-type CTGF+/+ mice were subjected to models of long-term diabetes (6 months), UUO and aristolochic acid nephropathy. All three models resulted in a severe renal phenotype with generation of tubulointerstitial damage. In the diabetes model, also significant increases in albuminuria, mesangial matrix volume and glomerular basement membrane thickness were present. However, the experiments failed to show differences in renal phenotype between CTGF+/- and wild-type mice. This possibly relates to the context of pronounced kidney damage in this study, or the relatively mild effect of genetic CTGF lowering. It is conceivable that reducing CTGF expression to subnormal levels is needed to prevent harmful effects in severe renal damage. The SNP -945GC in the promoter of the human CTGF gene was investigated with respect to plasma levels of CTGF and to outcome in a large population of diabetes type 1 patients with or without nephropathy. It was confirmed that plasma CTGF levels were higher in patients with diabetic nephropathy as compared to normoalbuminuric diabetic patients, but there was no relationship between genotype and nephropathy, renal and cardiovascular outcome, or mortality. Furthermore, urinary CTGF (uCTGF) levels were evaluated in kidney transplant patients. At 3 and 12 months after transplantation, there was a correlation between the low molecular weight proteins α1 and β2-microglobulin and uCTGF, reflecting tubular dysfunction. However, there was no predictive effect of uCTGF on long-term renal function. The data from this study underscore the complexity of interpreting biomarkers for renal fibrosis, especially in complicated settings such as kidney transplantation.

Intestinal cancer research has focussed on understanding the process of tumour formation for which hereditary disorders have proven to be valuable models. Although much is known about the genetic events that occur during the adenoma-carcinoma sequence, little is known about the genetic events that take place before the visible histopathological changes, a process called pre-tumour progression. This is mainly because many mutations are compatible with a histologically normal phenotype which allows adenomas, in general, to appear relatively late in life after the age of 50. Expanding knowledge of (pre) tumour progression in Familial Adenomatous Polyposis (FAP), Juvenile Polyposis (JPS) and Peutz-Jeghers (PJS) syndrome patients allows an increased understanding of cancer development as a whole and to potentially predict an individuals’ susceptibility to tumour development. Our research provides evidence for the presence of pre-tumour progression in FAP patients; FAP patients reveal an increase in methylation patterns diversity which can be used as a read-out for enhanced stem cell longevity. Since intestinal cancer seems to be a disease of the stem cell, altered stem cell dynamics is essential in cancer development. As such, an increase in survival allows stem cells to have more time to accumulate the mutations required to initiate tumour formation. The increase in stem cell longevity seems to be caused by an imbalance in stem cell division due to the APC germline mutation carried by FAP patients. Interestingly, similar results are found in PJS patients; PJS also reveals enhanced stem cell survival, possibly due to the LKB1 germline mutation which they carry. JPS patients carry a SMAD4 or a BMPR1A germline mutation; we have been able to provide evidence that different genotypes result in different phenotypes. Polyps with a predominant stromal compartment are found in BMPR1Amutation carriers, while a more epithelial polyp variant is found in SMAD4 mutation carriers. Interestingly, when determining methylation pattern diversity to be able to establish the presence or absence of pre-tumour progression, BMPR1A mutation carriers reveal enhanced stem cell longevity and an expansion of the stem cell pool while in SMAD4 mutation carriers, hyperproliferation is seen. Both result in tumourigenesis but along different routes. Interestingly, literature has shown SMAD4 mutations to play a central role in the adenoma-carcinoma sequence suggesting that JPS SMAD4 mutation carriers develop tumours through the conventional adenoma-carcinoma sequence. However, by conducting APC, SMAD4 and K-ras mutation analysis and p53, SMAD4 and β-catenin immunohistochemistry, we have been able to contradict this theory. In conclusion, our research has provided evidence that multiple pathways lead to neoplasia in polyposis syndrome patients where aberrant stem cell dynamics seems to play an essential role.

Response to renal injury is dependent on growth factors that determine how resident cells act, which cells are attracted to the site of injury, and how these resident cells, together with infiltrating cells and their surrounding matrix act together. These mechanisms are not confined to the kidney and also apply to other tissues, including the vasculature. Chronic injury can induce an inappropriate response to injury, where the growth factors are out of balance, and excessive scar formation occurs instead of appropriate regeneration and repair. Diabetes is an important cause of chronic tissue injury, leading to both macrovascular disease (e.g. stroke) and microvascular disease (e.g. diabetic nephropathy). Bone morphogenetic proteins (BMPs) are on the ‘good’ side of the growth factor balance. In the kidney, BMPs are important for maintenance of functional epithelium and for counteracting profibrotic responses. Connective tissue growth factor (CTGF), on the other side of the balance, promotes fibrosis. This thesis describes BMP signalling activity and CTGF involvement in tissue response to injury in renal fibrosis, diabetic nephropathy and during atherosclerotic plaque remodelling. Using the BRE:gfp reporter mouse, we directly visualized the distribution of transcriptional activity downstream of BMP signalling in healthy kidneys and in two distinct models of kidney disease. We conclude that BMP transcriptional activity is much more restricted than previously suggested, and its response to injury varies according to cell type and nephron segment. In glomeruli of human diabetic kidneys, BMP signalling activity was decreased together with an increase of CTGF. These findings suggest that overexpression of CTGF might be an important determinant of the loss of BMP signalling activity in human diabetic nephropathy. To better understand the meaning of urinary CTGF as a biomarker in diabetes we evaluated which factors contribute to elevated uCTGF in this disease. We investigated the renal handling of CTGF in diabetic mice, and the relation between urinary CTGF levels and tubular dysfunction in type 1 diabetic patients. Our results suggest that the majority of uCTGF derives from local production, and tubular damage is an important determinant of uCTGF. Urinary CTGF may therefore be a promising biomarker, integratedly reflecting local intrarenal fibrotic processes and tubular functional status. We next evaluated whether genetic lowering of CTGF in a long-term diabetes model protects against diabetic nephropathy. Unexpectedly, we found that development of an advanced renal phenotype in long-term streptozotocin-induced diabetes, as evidenced by albuminuria, mesangial matrix increase and tubulointerstitial damage, was not prevented by a 50% reduced CTGF expression. CTGF is known to play a role not only in (diabetic) renal injury, but also in the development of atherosclerosis. In a cross-sectional study we obtained evidence that following stroke CTGF may contribute to development of a more stable plaque phenotype. In conclusion, data in this thesis provide a better understanding of the localization and activity of BMP and CTGF in chronic tissue injury such as diabetic nephropathy and atherosclerosis. These findings could help guide the development of new therapies to halt chronic kidney disease and stabilize atherosclerotic plaques.

Patiënten met inflammatoire darmziekte (IBD) hebben een verhoogd risico op het ontstaan van colorectaal carcinoom (CRC; 20-jaars risico 8%). Wij vonden dat het risico op CRC bij patiënten die naast IBD ook bekend zijn met de galwegaandoening primaire scleroserende cholangitis (PSC) nog veel hoger is (20-jaars risico: 31%). Uit een vergelijking van CRCs van PSC-IBD patiënten met die van IBD patiëntenen viel op dat het aandeel van rechtszijdige tumoren in de PSC-IBD groep beduidend hoger was dan in de IBD groep (67% vs. 36%). Daarnaast hebben wij op verschillende manieren gekeken naar de moleculaire achtergrond van sporadische en IBD-(PSC)-gerelateerde CRCs. Microsatelliet instabiliteit werd niet aangetoond bij PSC-IBD-gerelateerde tumoren ondanks de hogere frequentie rechtszijdige tumoren. Ook was er geen verschil in voorkomen van Braf- en Kras-genmutaties en in expressie van verschillende carcinogenese-gerelateerde eiwitten, met uitzondering van Cycline D1, dat minder tot expressie kwam in IBD-CRC. Daarnaast was er meer verlies van SMAD4-expressie in darmtumoren bij IBD. Hypermetylatie kwam vaker voor in sporadische CRC dan bij IBD-CRC. Tenslotte bleek de Wnt-pathway geactiveerd te zijn in normaal colonslijmvlies van IBD-patiënten met elders in het colon CRC, wat duidt op een vroege rol van de Wnt-pathway. Dit past ook bij het zogenaamde ‘field effect’, waarbij normaal weefsel reeds genveranderingen laat zien die er op duiden dat het colon diffuus een verhoogde kans heeft op ontstaan van CRC. Dit proefschrift laat zien dat de verschillen in moleculaire achtergrond tussen sporadische en IBD-tumoren in een laat (invasief) stadium niet meer duidelijk aanwezig zijn. Deze verschillen zijn mogelijk wel aanwezig in een vroeger stadium en kunnen dan geﮤuceerd worden door exogene factoren zoals chronische darmontsteking (IBD) en toxische galzouten (PSC-IBD). Meer kennis over deze factoren kan bijdragen aan een meer individueel gerichte surveillance en wellicht zelfs preventie van CRC bij (PSC-) IBD

Ductal adenocarcinoma of the pancreas is a very aggressive disease with a high mortality rate. Pancreatic carcinoma is the fourth leading cause of cancer-related death in Western countries, despite the fact this cancer accounts for only about 3% of all malignant tumors. Most pancreatic cancers (approximately 80%) are diagnosed at a late, incurable stage. The poor prognosis and late presentation of pancreatic cancer patients underscore the importance of early detection, which is the sine qua non for the fight against pancreatic cancer. The possibility for early detection of pancreatic cancer may be realized by increasing our knowledge of the histology and molecular genetics of precursor lesions of pancreatic cancer in conjunction with the development of sensitive and specific screening tests (both invasive and non-invasive) to detect early pancreatic cancer. In this thesis we examined the potential use of MicroRNAs (miRNAs) as a biomarker for the early detection of pancreatic cancer and investigated several signaling pathways to explore their role in pancreatic tumorigenesis and the development of its precursor lesions. We found out that aberrant miRNA expression is an early event in the multistage progression of pancreatic cancer, and specifically miR-155 warrants further evaluation as a biomarker for pancreatic cancer in clinical samples. Furthermore, we looked at the Notch and Axl pathway in the setting of pancreatic cancer. Both pathways play a significant role in pancreatic carcinogenesis. Inhibition of the Notch or Axl pathway, in a panel of pancreatic cancer cell lines, resulted in significant reduction in cell viability, anchorage independent growth, as well as attenuation of migratory and invasive properties. These reports support and emphasize an important function of Notch and Axl in pancreatic cancer and both signaling pathways might be a promising therapeutic strategy in this carcinoma. Finally, we have characterized DNA Damage Response markers (DDR) in PanIN lesions to determine whether or not these are involved in early pancreatic neoplasia. Activation of the ATM-Chk2 checkpoint pathways was commonly observed in PanIN lesions, which suggest that ATM and Chk2 activation at early stage of pancreatic tumorigenesis suppresses tumor progression. Loss of p53 function appears to be a critical determinant for bypassing this checkpoint and the subsequent progression to invasive adenocarcinoma. In summary, the above studies in this thesis describe potential clinically useful biomarkers for pancreatic carcinogenesis. Moreover, our increased knowledge of the Notch and Axl signaling pathway in pancreatic tumorigenesis may provide the basis for developing more sensitive screening strategies and the identification of new drug targets enabling directed drug design. All this will enable us to develop better tools for primary and secondary prevention of pancreatic cancer, as well as improve existing tools for early diagnosis.

Gastric cancer is a result from the combination of environmental factors and an accumulation of specific genetic alterations, and affects mainly the older population. It is known that genetic factors play a more important role in early onset gastric cancers than in conventional gastric cancer patients as they have experienced less exposure to environmental factors. It is postulated that EOGC is a key tool in the explanation of the role of genetic changes in gastric carcinogenesis. The specific pathogenetic changes underlying gastric carcinogenesis remain largely unknown. The aim of this thesis is to uncover genetic alterations relevant to gastric carcinogenesis and to elucidate the role of the genes and proteins involved in the progress of cancer in different sub-types of gastric cancers. Additionally, we try to find the best approach for early diagnostics of gastric stump cancers. Gastric cancer is considered a multistep process and multifactor process. An overview of the involved factors as well as important genes in gastric carcinogenesis is provided in chapters 1 and 2. In these chapters the background of gastric cancer is extensively discussed. Additionally in these chapters the concept of early onset gastric cancer is introduced together with the reasoning behind the study of this subtype of gastric cancer. Chapter 3 focuses on a comparison between molecular alterations occurring in primary gastric cancers, corresponding xenografts and gastric carcinoma cell lines. Chapter 4 takes a step towards a better understanding of the progression of gastric carcinogenesis in early onset and conventional gastric cancers. Using Western blot, Q-PCR and sequencing analysis, we examined the correlation between two critical molecules in the development and progression of gastric cancer: COX-2 and E-cadherin. The relationship between these two molecules was examined in vitro in a panel of cell lines and in vivo using tissue microarrays. In chapter 5 we examine the COX-2 -765 G/C promoter polymorphism in different subtypes of gastric cancer. Real time PCR, sequencing analysis and immunohistochemical staining on tissue microarrays were used to find differences between EOGC, conventional gastric cancer and gastric stump cancer. Chapter 6 takes a look into the IL-1B -31 T/C polymorphism distribution in EOGC, conventional and gastric stump cancers using the same techniques as described in chapter 5. Chapter 7 reviews the literature on gastric stump cancer and its precursor lesions and focuses on the reliability of early diagnosis. Finally, chapters 8 and 9 provide brief summaries and a discussion of the thesis (in English and in Dutch respectively).

Hereditary breast cancer runs in families where several family members in different generations are affected. Most of these breast cancers are caused by mutations in the high penetrance genes BRCA1 and BRCA2 which account for about 5% of all breast cancers. However, mutations in BRCA1 and BRCA2 may be missed, and small families and sometimes insufficient medical records may cause uninformative family history. This thesis aimed to investigate the immunoprofile of hereditary breast cancers caused by germline mutations in the BRCA1 or BRCA2 genes in order to identify patients and families eligible for mutation screening. In addition, such studies on hereditary precursors could help to better understand hereditary breast carcinogenesis and identify new therapeutic targets. We started with evaluating a panel of clinicopathological variables to classify breast cancers as hereditary or sporadic using a multivariate approach. Most of the investigated markers were expressed in line with earlier data. However, we were the first to describe the high expression of EGFR in BRCA1 and BRCA2 related breast cancers. In addition, the observed high cyclin A expression is another useful proliferation marker in BRCA1 related breast cancers, and has been herewith established for the first time. Although our data were based on only five cases of BRCA2-related breast cancers, the cyclin D1 expression in these tumours was remarkably high. As such, most invasive breast carcinomas can be classified as sporadic or BRCA1 related with a high degree of certainty using a decision tree based on age, Ki67 and EGFR. Furthermore, we are the first to report that in BRCA1 and BRCA2 mutation carriers the immunoprofile of ductal carcinoma in situ (DCIS), a premalignant lesion, parallels the accompanying invasive cancer lesion, underlining that DCIS is probably a direct precursor lesion in these patients. In addition, crucial carcinogenetic events leading to these phenotypes apparently occur at an early stage in hereditary predisposed patients, most likely even before the stage of DCIS. We describe the novel finding of loss of expression of FANCD2, a Fanconi anaemia gene previously implicated in breast cancer in mice, in hereditary and sporadic invasive breast cancer, indicating that somatic inactivating (epi)genetic events in FANCD2 may occur in both sporadic and hereditary breast carcinogenesis. Our results do however not indicate that somatic (epi)genetic changes in FANCD2 are a frequent secondary carcinogenetic event in BRCA1 germline mutated patients. The expression of HIF-1?, the key regulator of the hypoxia response, in hereditary breast cancer is for the first time reported in comparison with sporadic breast cancer. 90% of BRCA1 germline mutation related breast cancers showed overexpression of HIF-1?, a percentage significantly higher than in sporadic controls. The overexpression of HIF-1? in BRCA1 related breast cancers is a frequent event and is likely hypoxia regulated, suggesting that the HIF-1? pathway plays a role in the BRCA-1 carcinogenesis and progression. All these findings will lead to improved discrimination of the BRCA1 and BRCA2 related breast cancer. Furthermore, the new biomarkers may serve as therapeutic targets for BRCA1 and BRCA2 related breast cancers.

Colorectal cancer (CRC) is an important cause death. In the Netherlands, approximately 10.000 patients are diagnosed with CRC each year. Rare hereditary gastrointestinal polyposis syndromes predisposing to CRC, including familial adenomatous polyposis (FAP), juvenile polyposis (JPS) and Peutz-Jeghers syndrome (PJS), have provided a model for the investigation of cancer initiation and progression in the general population. Many insights in the molecular genetic basis of CRC have emerged from studying these syndromes. The first part of this thesis is about duodenal polyposis in FAP and chemoprevention with NSAIDs. With improved management of colorectal disease and increased life expectancy of FAP patients, duodenal polyps and adenocarcinoma have become major health problems in these patients. Surveillance, endoscopic and surgical treatment and the value of NSAID chemoprevention for duodenal polyposis are reviewed. Results of NSAIDs on regression or prevention of duodenal adenomas in FAP appear disappointing. Expression of the main target of NSAIDs, cyclooxygenase-2 (COX-2) was studied and compared between colorectal and duodenal adenomas from patients with FAP. Levels of COX-2 appeared to be significantly higher in the duodenum than in the colorectum of FAP patients, which may explain the poorer response of duodenal adenomas to COX-2 inhibition, suggesting that higher dosages of COX-2 inhibitors are needed for treatment of duodenal adenomas. The second part of this thesis focuses on clinical and molecular aspects of JPS. We performed person-year analysis to define the magnitude of risk for gastrointestinal cancer in JPS. A relative risk of colorectal cancer of 34 and a cumulative life-time risk of 38.7% in JPS was found. JPS is caused by germline mutation of the SMAD4 or BMPR1A gene, but a germline mutation is found in only 30-40% of patients. Likely, alternative ways of inactivation of these genes exist. Therefore, we used MLPA to investigate whether germline deletion of these genes may cause JPS. By combining sequencing and MLPA, a germline defect in SMAD4 or BMPR1A was found in almost 50% of the patients. Nine mutations (33.3%) were detected by direct sequencing, whereas four germline hemizygous large genomic deletion (14.8%) were found using MLPA. Prevention of polyp and neoplasia development in juvenile polyposis using chemoprevention may be useful in the treatment of these patients. We measured expression of COX-2 in juvenile polyps to investigate the potential value of chemoprevention by COX-2 inhibition. Increased levels of COX-2 were found in juvenile polyps from JPS patients, suggesting that chemoprevention using COX-2 inhibition can be beneficial in JPS. Lastly, we studied molecular aspects of juvenile polyps from patients with a SMAD4 germline defect. Focal reduction or loss of SMAD4 expression was present in the epithelium of about half of the juvenile polyps from SMAD4 mutation carriers. LOH or somatic mutation of the wild type SMAD4 allele was found in cases with aberrant SMAD4 protein expression. These findings do not support a “landscaper” mechanism for SMAD4, although stromal haploinsufficiency or loss of SMAD4 may still have a role in polyp initiation and epithelial loss may only occur at a later stage in polyp progression. Surprisingly, loss of epithelial SMAD4 immunostaining did not correlate with dysplasia in polyps, suggesting that loss of the wild type allele is not obligatory for dysplastic change in juvenile polyps, not fitting with the typical tumor-suppressor/“gatekeeper” role for SMAD4, as previously suggested.

Diabetes mellitus is a severe and rapidly growing problem in health care, accounting for approximately 150 million patients worldwide. Patients with diabetes are at increased risk to develop diabetic nephropathy, which is currently the most important cause of end-stage renal disease in large parts of the world. The natural course of diabetic nephropathy is unpredictable and the pathogenesis of progression is not completely understood. Recently, connective tissue growth factor (CTGF) and bone morphogenetic proteins (BMPs) have emerged as key players in renal disease. The studies presented in this thesis aimed to elucidate the role of CTGF and BMPs, and their possible interaction in diabetic nephropathy. Both urinary CTGF excretion and plasma CTGF were increased in patients with diabetic nephropathy and were correlated with clinical markers of renal disease. The association of elevated CTGF levels with diabetic nephropathy proved to be at least as strong as that of the established risk factors hypertension and hyperglycemia. Next, in a prospective study, baseline plasma CTGF was identified as an independent predictor of end-stage renal disease and mortality in patients with diabetic nephropathy. Moreover, in diabetic patients with nephrotic range albuminuria, plasma CTGF was the only parameter predicting end-stage renal disease. These observations suggest that CTGF in urine and plasma might find clinical application as a biomarker. To study CTGF involvement in the pathogenesis of diabetic nephropathy, we investigated development of diabetic nephropathy in the context of high and low levels of CTGF. In transgenic CTGF+/- mice, which express 50% lower CTGF levels than wild-type mice, structural and functional changes of diabetic nephropathy were significantly attenuated. This was exemplified by less severe albuminuria, absence of glomerular basement membrane thickening, and preserved matrix metalloproteinase activity in diabetic CTGF+/- mice. Interestingly, we observed that CTGF inhibited the signaling activity and target gene expression of BMP-7, both in diabetic mice as well as in cultured renal cells. Thus, overexpression of CTGF plays a role in the pathogenesis of diabetic nephropathy, and this appears to involve inhibition of the antifibrotic activity of BMP-7. Another novel aspect in the pathogenesis of renal fibrosis where we identified involvement of BMPs is the contribution of circulating bone-marrow derived myofibroblast progenitor cells (MFPC). We showed that the number of MFPC is increased in patients with diabetes as compared to matched healthy control subjects, and that diabetes-derived MFPC were deficient in BMP-6 expression. We demonstrated that BMP-6, like BMP-7, could antagonize profibrotic activity of transforming growth factor-beta in renal cells. The direct impact of BMP-6 on MFPC and renal fibrosis was revealed in BMP-6 deficient mice, which developed more severe renal fibrosis after unilateral ureteral obstruction, in association with higher numbers of circulating MFPC. In conclusion, the studies in this thesis have identified CTGF and BMPs as determinants of renal fibrosis, and in particular diabetic nephropathy. We suggest that restoration of growth factor balance might constitute a novel strategy for treatment of diabetic nephropathy.

Gastrointestinal polyposis syndromes are characterized by the development of gastrointestinal polyps and an increased risk for neoplastic transformation. Peutz-Jeghers syndrome is a polyposis syndrome characterized by the development of gastrointestinal hamartomas; the molecular and histopathological origin of these polyps is unclear and hence the pre-malignant potential, if any, of these polyps remains unresolved. Current follow-up protocols mandate removal of polyps to decrease risk of intussusception as well as to decrease risk of neoplastic transformation. The current thesis investigates the premalignant potential of these polyps and proposes that in fact the polyps are an epiphenomenon to the malignant condition. Polyp removal therefore does not necessarily reduce cancer risk. Furthermore we show that abnormalities can be demonstrated in normal (id est unaffected) epithelia in Peutz-Jeghers syndrome that may suggest protracted clonal evolution at the level of the intestinal stem cell niche. Altered clonal evolution in the crypt would jointly explain the development of the benign gastrointestinal hamartomas as well as an increased risk for neoplastic transformation in these patients. Lastly, we show that the linchpin component of the Wnt signalling pathway, beta-catenin, appears on the plasma membrane upon Wnt treatment in Wnt-responsive cells. This suggests a previously unappreciated step in a signalling pathway frequently deregulated in familial and sporadic forms of colorectal cancer.