Urinary Proteomics and Drug Discovery in Chronic Kidney Disease: A New Perspective Marco Prunotto,*,† Gian Marco Ghiggeri,‡ Giovanni Candiano,‡ Pierre Lescuyer,§ Denis Hochstrasser,§ and Solange Moll| Pharma Research Development and Metabolic Disease, Hoffmann-La Roche Ltd., Basel, Switzerland, Nephrology Unit Laboratory, Giannina Gaslini Children’s Hospital, Genova, Italy, Clinical Proteomics Laboratory, Department of Genetics and Laboratory Medicine, Geneva University Hospitals, and Clinical Pathology, Geneva University Hospital, Geneva, Switzerland Received May 14, 2010
Chronic kidney disease (CKD) is becoming a worldwide public health problem. The identification of a specific set of early biomarkers for CKD is extremely relevant to progress in disease knowledge, improving diagnosis, treatment, or development, and monitoring efficacy of new drugs. As kidney fibrosis can be considered the common pathological way to end stage renal failure, independent of the initial renal insult, these biomarkers are therefore biomarkers of early tubulo-interstitial fibrosis. The availability of a specific set of biomarkers for CKD is the mandatory condition to create new dedicated drugs and validate them in clinics without waiting years for a functional response in patients. We suggest here specific cohorts of patients where this early signature of fibrosis may be simpler to be identified. Keywords: CKD • biomarker • drug discovery • urine • proteomics
Introduction Chronic kidney disease (CKD) is becoming a worldwide public health problem. At present, approximately 8-10% of individuals in western countries are affected by chronic progressive kidney failure,1–3 and diffusion of diabetes and metabolic syndrome within youngsters will even worsen the scenario in the next years to come. The identification of a specific set of early biomarkers for CKD is extremely relevant to progress in disease knowledge, improving diagnosis, treatment or development and monitoring efficacy of new drugs. Biomarkers and surrogate biomarkers are currently used in clinical medicine for disease diagnosis, as indicators of disease activity or to predict and monitor response to treatment. A biomarker is typically defined as a parameter “objectively measured and evaluated as an indicator of normal biologic processes, pathogenic processes, or pharmacologic responses to therapeutic intervention”.4 A surrogate biomarker is instead defined as “[...] a laboratory measurement or physical sign that is used in therapeutic trials as a substitute for a clinically meaningful end point that is a direct measure of how a patient feels, functions, or survives and is expected to predict the effect of the therapy.”5 Presence of antinuclear antibodies (ANA),6–8 antineutrophyl * To whom correspondence should be addressed. Marco Prunotto, Ph.D., Pharma Research Development and Metabolic Disease, Hoffmann-La Roche Ltd., Basel, Switzerland. Phone: 0041.79.812.64.88. E-mail: marco.prunotto@ roche.com. † Hoffmann-La Roche Ltd. ‡ Giannina Gaslini Children’s Hospital. § Department of Genetics and Laboratory Medicine, Geneva University Hospitals. | Clinical Pathology, Geneva University Hospital.
126 Journal of Proteome Research 2011, 10, 126–132 Published on Web 08/19/2010
cytoplasmic antibodies (ANCA)9 or antiglomerular basement membrane antibodies (anti-GBM)10 in autoimmune diseases are clinically accepted and validated biomarkers for disease identification. All these markers correlate with the disease process, helping clinicians for diagnosis and selection of course of treatment. Moreover, these markers allow for a simple assessment procedure. While urine proteomics for biomarkers identification seems promising, drug discovery in renal diseases is far from being considered urine-driven. Historical advances in last decades have focused only on potential mechanisms of diseases, increasing knowledge in immunological processes (steroids, immunosuppressors) or mechanism tailored to cells (e.g., antiCD20 antibodies, Bortezomib). Two notable exceptions are Cyclosporin A, which plays an effect on podocyte cytoskeleton, and blockers of the angiotensin system, which affect renal hemodynamics and also inhibit TGFβ. No therapeutical evolution considered urinary biomarkers and/or originated from urine research. Lack of controlled studies on urine composition in different diseases and technology problems determined this failure in the past, but we are now on the road to fill up this gap.
Urine Proteomics The search for such biomarkers in kidney diseases seems to be oriented naturally toward the analysis of urine. Urine is abundant and easily available, it demands noninvasive sampling procedures and simple storage conditions, and it can be collected over time for clinico-pharmacological surveillance. As with all organism bodily fluids, urine contains thousands 10.1021/pr100464q
2011 American Chemical Society
Urinary Proteomics and Drug Discovery in CKD of proteins and peptides that are the result of complex filtration/secretion/reabsorption processes along the whole nephron. Proteins expressed throughout the entire body, and exchanged into the blood compartment, can potentially be expressed in urine either as peptide constituents or whole proteins. Proteomic technologies hold the potential to revolutionize clinical care by providing tools for the discovery of these protein biomarkers for diagnosis, prediction of disease course, guiding therapeutic selection and monitoring response to new drug therapies. Considerable technological and methodological advances have recently been made that might allow the use of urine as a source of candidate biomarkers in proteomics studies. Normal, nonpathological, urine11–13 (see reviews 14 and 15) displays ∼1000 spots in 2D-page map, only a small part of which are characterized, the large majority still being completely unknown. Normal urine samples show a major limitation residing in highly expressed proteins overwhelming low represented ones. These highly abundant urine proteins, for example, albumin, R1-antitrypsin or Tamm Horsfall glycoprotein, interfere with proteomic analysis of low represented proteins, which may in fact reside crucial biomarkers. The situation is even worsened in the case of pathologies producing abundant albuminuria. Despite this major limitation, almost any known mass spectrometry technique has been used for the analysis of the urinary proteome: two-dimensional gel electrophoresis (2DE)-MS, LC-MS, SELDI-TOF and capillary electrophoresis (CE)-MS. Each specific technique give a different scenario: 2DE-MS16 allowing for detection of large molecules but losing low molecular weight (
