J Nephrol DOI 10.1007/s40620-014-0044-5
INVITED REVIEW
Proteomics and diabetic nephropathy: what have we learned from a decade of clinical proteomics studies? Massimo Papale • Salvatore Di Paolo • Grazia Vocino • Maria Teresa Rocchetti Loreto Gesualdo
•
Received: 18 September 2013 / Accepted: 15 October 2013 Ó Italian Society of Nephrology 2014
Abstract Diabetic nephropathy (DN) has become the most frequent cause of chronic kidney disease worldwide due to the constant increase of the incidence of type 2 diabetes mellitus in developed and developing countries. The understanding of the pathophysiological mechanisms of human diseases through a large-scale characterization of the protein content of a biological sample is the key feature of the proteomics approach to the study of human disease. We discuss the main results of over 10 years of tissue and urine proteomics studies applied to DN in order to understand how far we have come and how far we still have to go before obtaining a full comprehension of the molecular mechanisms involved in the pathogenesis of DN and identifying reliable biomarkers for accurate management of patients. Keywords Diabetic nephropathy Proteome Proteomics Urine Tissues Biomarkers
M. Papale Core Facility of Proteomics and Mass Spectrometry, Department of Surgery and Medical Sciences, University of Foggia, Foggia, Italy S. Di Paolo Nephrology Unit, ‘‘Dimiccoli’’ Hospital, Barletta, Italy G. Vocino M. T. Rocchetti L. Gesualdo (&) Nephrology, Dialysis and Transplantation Unit, Department of Emergency and Organ Transplantation (D.E.T.O), University of Bari ‘‘Aldo Moro’’, Bari, Italy e-mail: [email protected]
The pathophysiology of diabetes and diabetic nephropathy Diabetic nephropathy (DN) is a severe complication of both type 2 (T2DM) and type 1 diabetes mellitus (T1DM). It has become the most common cause of chronic kidney disease (CKD) in developed countries [1] and the most frequent cause of end-stage renal disease (ESRD) worldwide given that DN [2] affects about 20-40 % of nephropathic patients undergoing dialysis. Historically, the incidence of DN in T1DM exceeds that recorded in T2DM, since in the latter group death from cardiovascular causes generally antedates the appearance of renal failure [3, 4]. However, the use of renin–angiotensin system inhibitors along with a strict glycemic control is improving the life expectancy of T2DM patients and slowing the rate of mortality due to cardiovascular causes [5]. As a result, it is expected that the incidence of DN in T2DM patients will significantly increase over the next few years [6]. Chronic hyperglycemia triggers specific modifications of electron transport proteins via advanced glycation end-products (AGEs). This alters normal mitochondrial metabolism and leads to an increased production of reactive oxygen species (ROS) [7], which in turn strongly contribute to chronic renal damage in T2DM. On the other hand, less than 50 % of T2DM patients finally develop DN, which likely highlights the role of genetic background and lifestyle modifications [8]. Interestingly, the increased oxidative stress induces long-lasting effects which persist even after glycemia normalization, and lead to a persistent activation of endothelial nitric oxide synthase and prostacyclin synthase, which, in turn, contributes to defective angiogenesis and persistent expression of pro-inflammatory genes [9]. The onset and progression of renal damage in T2DM proceeds through distinct and sequential phases: it has been
123
J Nephrol
demonstrated that the initial renal response to hyperglycemia is glomerular hyperfiltration and hyperperfusion, a process mediated by vasoactive hormones such as angiotensin II [10–13]. Even if these early hemodynamic changes are positively modulated by normalization of blood glucose levels, they probably trigger the incipient molecular events that, in some patients, induce the progression of the disease. Incipient DN is then characterized by albumin leakage from glomerular capillaries, overproduction and extracellular accumulation of mesangial cell matrix, thickening of the glomerular basement membrane and podocytes injury [14–16]. At the urinary level the earliest putative diagnostic sign of diabetic renal damage is microalbuminuria (urine albumin excretion 30–300 mg/24 h) which, unfortunately, does not always correlate with the complex histopathological picture of glomerular and tubular damage of T2DM [17, 18]; thus this parameter represents a predictor of cardiovascular disease rather than of renal damage progression [19]. The rapid development of new high-throughput approaches for the massive analysis of the protein content of biological samples and the possibility to perform fast and accurate in silico analysis of complex datasets, that was unthinkable until few years ago, is now making it possible to better characterize the key events occurring during the onset and progression of DN and to identify more accurate disease-specific biomarkers. In the following paragraphs we will discuss the main contributions of proteome analysis to the understanding of DN pathophysiology and to the identification of new sets of biomarkers potentially useful for achieving an early diagnosis and a better management of patients. The present review represents the fourth part of a series of papers [20–22] published by the Study Group on ‘‘OMICS’’ sciences of the Italian Society of Nephrology (http://www.sin-italy.org/Soci_ GdS/gruppi_omiche.asp), addressing the main contributions of the Omics sciences to key-relevant topics in renal pathophysiology.
Tissue proteomics and diabetic nephropathy The application of proteomic technologies to the study of morpho-structural and functional changes of the kidney represents the most focused approach to understanding the molecular events that initiate and propagate renal damage. However, renal biopsy is poorly applied to diabetic patients and the progression of renal damage is commonly gathered from clinical and/or laboratory findings, such as the measurement of urine albumin excretion rate. Further, the small amounts of renal tissue obtained from renal biopsy specimens make it difficult to perform a satisfactory proteomic analysis. As a consequence, most of the studies have been
123
carried out on animal models mimicking the renal damage occurring in humans with diabetes [23–26]. Tilton et al. [27] used two-dimensional gel electrophoresis (2DE) to study the renal cortex proteins whose expression was deregulated in the db/db (diabetic mice) model of T2DM. More than 50 % of the deregulated proteins were involved in mitochondrial metabolic pathways including lipid and fatty acid metabolism. Ingenuity pathways analysis (IPA) identified peroxisome proliferator activated receptor c (PPAR-c) as a central player in this network. Further, Zhang et al. [28] recognized in the same animal model a significant increase of ketogenesis and Barati et al. [29] reported a significant increase of antioxidant and glycation enzymes. Of note, the above studies analysed whole tissue samples, making it more difficult to associate the different expression of disease-related proteins to a specific renal sub-compartment. Proteomic analysis of isolated glomeruli allowed to identify a panel of 17 significantly and differentially expressed actin cytoskeleton-associated proteins [30] and to characterize SORBS2, an adapter protein presumably involved in the assembly of the actin stress fibres, as a central player in the pathogenesis of DN. Interestingly, the results of the above studies are poorly replicable in humans due to the difficulty of performing efficient isolation of the glomeruli from human biopsies. Further, most tissue samples available in humans are those routinely prepared for diagnostic purposes, i.e. they are formalin fixed and paraffin embedded (FFPE). Recent methodological improvements [31–33] have demonstrated that it is possible to extract intact and unmodified proteins from FFPE samples and further analyze them by mass spectrometry (MS) thus permitting the application of proteomic technologies to a vast array of formalin fixed wax embedded tissue samples stored in pathology department archives [34]. In order to perform compartment-specific analyses, glomeruli and tubuli need to be dissected from FFPE tissues by laser capture microdissection (LCM) [35] and directly analyzed by proteomic techniques. Proteomic analysis of LCM-isolated glomeruli revealed an increased and selective expression of nephronectin, a protein implicated in the assembly of extracellular matrix, in diabetic nephropathy [36]. Further, a similar study described, in the glomeruli of DN patients, the presence of complement factors such as C3 and the membrane attack complex (C5b-9) as well as a marked decrease of podocyte-associated proteins and antioxidant proteins [37]. Alternatively, FFPE tissues may be directly analyzed by imaging mass spectrometry (IMS) [38, 39] to provide both the identification of disease-related biomarkers and their localization in specific renal compartments. This approach may allow the molecular mapping of changes observed in diabetic kidney in both experimental models and human disease, and to identify the major molecular alterations linked to the development of diabetic nephropathy.
Renal tissue
Renal tissue (glomeruli)
Renal tissue (isolated glomeruli)
Renal tissue (microdissected glomeruli)
Renal tissue (microdissected glomeruli from biopsies)
Urine samples
Urine samples
Urine samples
Zhang et al. [28]
Barati et al. [29]
Nakatani et al. [30]
Nakatani et al. [36]
Satoskar et al. [37]
Zurbig et al. [60]
Dihazi et al. [63]
Papale et al. [66]
Urine samples
Renal cortex
Tilton et al. [27]
Jin et al. [67]
Sample type
References
Normoalbuminuric vs. microalbuminuric T2DM patients
Normoalbuminuric vs. microalbuminuric vs. biopsyproven DN patients
Normoalbuminuric vs. microalbuminuric vs. macroalbuminuric T2DM patients
Normoalbuminiric diabetic patients developing microalbuminuria over the time
FFPE tissue (human)
FFPE tissue (human)
OLETF rats (spontaneous type 2 diabetic) vs. LETO rats (in the early and proteinuric stages of diabetic nephropathy)
db/db mice
db/db mice
db/db mice
Source
iTRAQ labelling ? LC–MS/ MS; WB
SELDI TOF/MS; ELISA; WB
SELDI–TOF/MS; WB
CE-MS
LCM ? LC/MS/MS; IF
LCM ? LC/MS/MS; IHC
LC-ESI MS/MS; IHC; IF
2DE-MALDI-TOF MS; WB; IHC
2DE-MALDI-TOF MS; WB
2DE-MALDI-TOF MS; WB; IPA
Methods
Table 1 Key proteomic studies applied to tissue and urine in diabetic nephropathy
:Prostate stem cell antigen
:Alpha-1-acid glycoprotein 1
:Alpha-1-antitrypsin
Immunity, cell adhesion, protein metabolism, signal transduction
Tubular injury and ubiquitination
:b2-Microglobulin :Ubiquitin
Tubular injury and ubiquitination
Assembly and accumulation of mesangial extracellular matrix
Activation of complement and oxidative stress
Assembly and accumulation of mesangial extracellular matrix
Cytoskeleton rearrangement
Cellular redox regulation and detoxification
:b2-microglobulin :ubiquitin ribosomal fusion protein (UbA52)
:b-Microglobulin
;Uromodulin
:I a 1-Antitrypsin
;Collagen a1 chain
:Synaptopodin
:Fibrinogen a-chain
:C5b-9
:C3
;Actinin, alpha 4
:Laminin, beta 2
:Integrin, alpha 1
:Nephronectin
:Tubulin alpha 1c
:Actin-related protein 2/3 complex subunit 1 beta
:Alpha-actinin
:Sorbin and SH3 domain containing 2
:Glyoxalase I
:Superoxide dismutase 1
:Glutathione peroxidase 1
:Peroxiredoxin 1
;Protein kinase C inhibitor protein 1 :3-hydroxy-3-methylglutaryl-CoA synthase (HMGCS2)
Ketogenesis
Mitochondrial metabolism; lipid and fatty acid metabolism
:Peroxisome proliferator activated receptor c (PPAR-c) :Phosphoenolpyruvate carboxykinase 1 (Pepck)
Pathways
Key features
J Nephrol
123
123
WB western blotting, IPA ingenuity pathway analysis, IHC immunohistochemistry, IF immunofluorescence, FFPE formalin fixed and paraffin embedded, DN diabetic nephropathy, T2DM type 2 diabetes mellitus, 2DE-MALDI-TOF MS two-dimensional gel electrophoresis matrix-assisted laser desorption/ionization time-of-flight mass spectrometry, LC-ESI/MS/MS liquid chromatography-electrospray ionization/multi-stage mass spectrometry, LCM laser capture microdissection, CE capillary electrophoresis, SELDI–TOF/MS surface enhanced laser desorption/ionization time-of-flight mass spectrometry, ELISA enzyme-linked immunosorbent assay, iTRAQ isobaric tags for relative and absolute quantitation, SDS-PAGE sodium dodecyl sulphate-polyacrylamide gel electrophoresis
Intracellular signaling and transport; cell adhesion ;Major urinary protein 1
:Xaa-Pro dipeptidase
SDS-PAGE ? nLC-ESI–MS/ MS; WB Urine exosomes Raimondo et al. [71]
Zucker diabetic fatty (ZDF) rats
Sample type References
Table 1 continued
Source
Methods
Key features
Pathways
J Nephrol
Specifically, IMS applied to experimental models of diabetes, or to human biopsy samples collected in the very early stages of renal injury, would possibly allow the identification of the molecular markers heralding the progression of renal damage. Even if there is not any original study so far exploring this topic, a wide application of this approach is expected in the forthcoming years. On the whole, the technological and methodological improvements in tissue proteomics will without doubt contribute revealing new pathogenic mechanisms of renal damage in T2DM (Table 1).
Urine proteomics and diabetic nephropathy Urine has been recognized as one of the most appropriate biological fluids for biomarker discovery in kidney diseases [40–46]. Some peculiarities of urine could make it the gold standard for kidney biomarker discovery studies. They include the easy and non-invasive accessibility, the presence of both kidney-derived (about 70 %) and plasmaderived (about 30 %) proteins, and the lower complexity and increased stability of the urine proteome when compared to that of other biological fluids [47–49]. However, a number of published studies have pointed out the inherent limitations of this biological fluid (i.e. presence of salts and contaminants; high intra- and inter-subject variability; release, in nephropathic patients, of high amounts of serum proteins which makes the identification of the scarcely excreted biomarkers really burdensome) [43, 49–53]. Thus, the use of standardized protocols for the collection, management and analysis of urine is crucial to maximize the potentiality of this sample. Although the 2-DE based analysis of the urine proteome can provide useful information about the key changes of the urine proteome in DN patients, it is clear that the identification of new reliable biomarkers requires the setup of multicentre studies and the analysis of thousands of biological samples that can be possible only by means of massive screening by high throughput strategies such as liquid chromatography (LC) [54], capillary electrophoresis (CE) [55], and thin-layer chromatography [56] coupled to mass spectrometry (MS). In order to provide a qualified picture of the main developments in DN biomarker discovery we will limit the further discussion to the studies based on the use of high throughput strategies. CE allows the identification, in urine from type 1 diabetics, of 65 urine peptides, many of them being type I collagen fragments, making it possible to recognize the presence of DN with 97 % sensitivity and specificity [57]. The validation of this pattern in a multicentre independent cohort of type 2 diabetic patients provided the first evidence that the presence of collagen fragments may allow identification of DN in both type 1 and type 2 DM [58]. Interestingly,
J Nephrol
most of the 65 urine peptides were further included in a new CE-MS urinary pattern, composed of 273 urinary small peptides (called ‘‘classifier 273’’), that was highly specific and sensitive for CKD irrespective of the underlying pathology [59]. Recently, Zurbig et al. [60] evaluated the predictive value of classifier 273 on T1DM and T2DM normoalbuminuric patients and demonstrated that the classifier was able to reliably predict the progression to macroalbuminuria, long before the modification of albumin excretion rate. Surface-enhanced laser desorption ionization (SELDI)time-of-flight (TOF)-MS analysis is a further high throughput technology which has been extensively used for identifying urine biomarkers of DN. Adopting this proteomic technique, Otu et al. [61] detected a 12-peak signature predicting the development of diabetic nephropathy ten years prior to the increase of albumin to creatinine ratio, while Wu et al. [62] recently described a 4-peak pattern able to recognize T2DM patients with DN with 88 % sensitivity and 97 % specificity. Unfortunately, both these studies failed to identify any of the candidate biomarkers, thus limiting the possibility to test the predictive and diagnostic power of these models in independent testing sets. Two DN biomarkers (b2-microglobulin and ubiquitin ribosomal fusion protein) were instead purified and identified by Dihazi et al. [63] among more than 100 mass peaks, which were differentially excreted in the urine of patients with diabetic nephropathy. Indeed, all the above studies identified patients with diabetic nephropathy based upon their albumin excretion rate, without any investigation of the renal lesions underlying the clinical features. In fact, renal changes different from diabetic glomerulosclerosis can be detected in over 40 % of type 2 diabetics with proteinuria [18]. Conversely, typical renal structural changes of diabetic nephropathy may be found in diabetics with normoalbuminuria and renal insufficiency [64, 65]. Papale et al. [66] performed a SELDI analysis on the urine of type 2 diabetics comparing patients with biopsy proven Kimmestiel–Wilson lesions with those displaying non diabetic chronic renal changes (CKD). Their results suggested that the combination of b2-microglobulin and free ubiquitin might make it possible to set up a diagnostic model able to distinguish DN from CKD in both non-diabetic and diabetic patients. Furthermore, Jin et al. [67] used an LC/MS based analysis with isobaric tags for relative and absolute quantitation (iTRAQ) [68] to select and quantify differentially excreted urinary proteins in pooled urine samples of microalbuminuric versus normoalbuminuric diabetic patients. The bioinformatics management of the entire dataset (196 differentially expressed proteins) allowed to restrict the further validation to ten disease-correlated biomarkers. Further, the combination of three candidate biomarkers (alpha-1-antitrypsin, alpha-1-acid glycoprotein 1, and prostate stem cell antigen) produced a receiver operating characteristic (ROC) curve with the highest diagnostic power (about 92 %) for diabetic microalbuminuric patients.
It is worth noting that most urine proteomic studies limited the analysis to the soluble fraction, but diseasespecific biomarkers may be detected also in cells and microvesicles. In particular, urinary exosomes, 30–100 nm diameter vesicles, derived from the endosomal compartment and released via fusion of multivesicular bodies with the plasma membrane [69], are emerging as a new source of potential biomarkers [70]. These vesicles are released from every epithelial cell lining the nephron, including podocytes, and may deliver segment-specific biomarkers in the urine. The recent study by Raimondo et al. [71] emphasizes this concept: the authors reported, in Zucker diabetic fatty (ZDF) rats undergoing progressive worsening of renal function, significant changes in the urine exosomes’ proteome revealing how these microvesicles may provide detailed information about metabolic and immune processes correlated with the onset and progression of renal damage in T2DM. Taken together, the above results demonstrate that the proteomic analysis of the urinary exosomes, together with the analysis of the soluble urinary proteins, may fruitfully contribute to reveal the pathophysiological alterations occurring in DN progression, and to enlarge the panel of DN biomarker candidates (Table 1).
Conclusion and perspectives The proteomic approach to the study of diabetic nephropathy has made a decisive contribution to the understanding of DN pathophysiology and the identification of disease-related biomarkers. In the last 10 years, the advances have been much more than we may have perceived. In fact, there have been concomitant improvements in many fields crucial for the comprehension of the pathogenesis of diabetic renal damage: the definition of animal models mimicking the pathophysiology of type 2 diabetic renal damage in humans; the setup of new protocols for the direct proteomic analysis of FFPE tissues; the standardization of urine collection, management and storage; the development of new MS analyzers that may allow the detection and characterization even of barely expressed biomarkers; the development of sophisticated bioinformatics tools to manage complex datasets. All these aspects are contributing to establish an optimal workflow for the discovery and validation of reliable protein biomarkers in T2DM as well as in other complex diseases (Fig. 1). In the near future, we will probably be able to combine the power of proteomic analysis of human tissues and biological fluids with the bioinformatics management of complex datasets, in order to collect enough information to yield a reliable picture of the key events occurring during the onset and progression of renal damage in T2DM. This new way to manage proteomic datasets will probably favour the identification of reliable biomarkers by reducing the effect of confounding factors. Moreover, the
123
J Nephrol
Fig. 1 Workflow of the proteomic analysis applied to tissue and urine samples. Left panel Kidney sub-compartments (i.e. glomeruli, tubuli, etc.) can be isolated from whole biopsy specimens through laser capture microdissection (LCM) prior to performing proteomic analysis. Alternatively, frozen or formalin fixed and paraffin embedded (FFPE) tissues can be directly analysed by matrix-assisted laser desorption/ionization imaging mass spectrometry (MALDI-IMS). Right panel Ultracentrifugation of whole urine samples is used to separate soluble and vesicular fractions, then proteomic analysis may
allow the identification of compartment-specific biomarkers. The bioinformatic management of urine and tissue putative biomarkers is then applied to each dataset in order to select functionally correlated disease biomarkers [72]. Comparative analysis may finally allow to identify tissue-derived biomarkers also detectable in the urine samples. LCM glomeruli and MALDI-IMS figures are reproduced with permission from Kohda et al. [73] and Chaurand et al. [74], respectively
complex intracellular behaviour implies a constant interaction of proteins, genes, transcripts and metabolites that all together contribute to the observed phenotypes. Bioinformatics will have to develop more accurate tools to correlate specific proteomic datasets with the corresponding genomic, transcriptomic and metabolomic profiles in order to pursue a global characterization of the biological systems and identify a multilevel panel of molecular players contributing to the onset of pathological phenotypes.
National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda (2007) KDOQI Clinical Practice Guidelines and Clinical Practice Recommendations for Diabetes and Chronic Kidney Disease. Am J Kidney Dis 49:S1–S179 Craig KJ, Donovan K, Munnery M, Owens DR, Williams JD, Phillips AO (2003) Identification and management of diabetic nephropathy in diabetes clinic. Diabetes Care 26:1806–1811 Gross JL, de Azevedo MJ, Silverio SP, Canani LH, Caramori ML, Zelmanovitz T (2005) Diabetic nephropathy: diagnosis, prevention, and treatment. Diabetes Care 28:164–174 Ziyadeh FN, Sharma K (2003) Overview: combating diabetic nephropathy. J Am Soc Nephrol 14:1355–1357 Stewart JH, McCredie MR, Williams SM et al (2007) Trends in incidence of treated end-stage renal disease, overall and by primary renal disease, in persons aged 20–64 years in Europe, Canada and the Asia-Pacific region, 1998–2002. Nephrology (Carlton) 12:520–527 Rosca MG, Mustata TG, Kinter MT et al (2005) Glycation of mitochondrial proteins from diabetic rat kidney is associated with excess superoxide formation. Am J Physiol Renal Physiol 289:F420–F430
2.
3.
4.
5. Conflict of interest of interest.
The authors declare that there are no conflicts
6.
References 7. 1. U.S. Renal Data System (2012) USRDS 2012 Annual Data Report: Atlas of Chronic Kidney Disease and End-Stage Renal Disease in the United States. National Institutes of Health,
123
J Nephrol 8. Krolewski AS, Warram JH (1995) Genetic susceptibility to diabetic kidney disease: an update. J Diabetes Complications 9(4):277–2781 9. Giacco F, Brownlee M (2010) Oxidative stress and diabetic complications. Circ Res 107:1058–1070 10. Takenaka T, Inoue T, Ohno Y et al (2012) Elucidating mechanisms underlying altered renal autoregulation in diabetes. Am J Physiol Regul Integr Comp Physiol 303(5):R495–R504 11. Patinha D, Fasching A, Pinho D, Albino-Teixeira A, Morato M, Palm F (2013) Angiotensin II contributes to glomerular hyperfiltration in diabetic rats independently of adenosine type I receptors. Am J Physiol Renal Physiol 304(5):F614–F622 12. Sochett EB, Cherney DZ, Curtis JR, Dekker MG, Scholey JW, Miller JA (2006) Impact of renin angiotensin system modulation on the hyperfiltration state in type 1 diabetes. J Am Soc Nephrol 17(6):1703–1709 13. Mathis KM, Banks RO (1996) Role of nitric oxide and angiotensin II in diabetes mellitus-induced glomerular hyperfiltration. J Am Soc Nephrol 7(1):105–112 14. Kim HJ, Cho EH, Yoo JH et al (2007) Proteome analysis of serum from type 2 diabetics with nephropathy. J Proteome Res 6:735–743 15. Ziyadeh FN, Snipes ER, Watanabe M, Alvarez RJ, Goldfarb S, Haverty TP (1990) High glucose induces cell hypertrophy and stimulates collagen gene transcription in proximal tubule. Am J Physiol 259:F704–F714 16. Schordan S, Schordan E, Endlich N et al (2009) Alterations of the podocyte proteome in response to high glucose concentrations. Proteomics 9:4519–4528 17. Colantonio DA, Chan DW (2005) The clinical application of proteomics. Clin Chim Acta 357:151–158 18. Mazzucco G, Bertani T, Fortunato M et al (2002) Different patterns of renal damage in type 2 diabetes mellitus: a multicentric study on 393 biopsies. Am J Kidney Dis 39(4):713–720 19. Thongboonked V, Malasit P (2005) Renal and urinary proteomics: current applications and challenges. Proteomics 5:1033–1042 20. Bonomini M, Sirolli V, Magni F, Urbani A (2012) Proteomics and nephrology. J Nephrol 25(06):865–871 21. Magni F, Lalowski M, Mainini V et al (2013) Proteomics imaging and the kidney. J Nephrol 26(3):430–436 22. Santucci L, Candiano G, Bruschi M et al (2013) Urinary proteome in a snapshot: normal urine and glomerulonephritis. J Nephrol 26(4):610–616 23. Hummel KP, Dickie MM, Coleman DL (1966) Diabetes, a new mutation in the mouse. Science 153(3740):1127–1128 24. Chua SC, Chung WK, Wu-Peng XS et al (1996) Phenotypes of mouse diabetes and rat fatty due to mutations in the OB (leptin) receptor. Science 271(5251):994–996 25. Zhao HJ, Wang S, Cheng H et al (2006) Endothelial nitric oxide synthase deficiency produces accelerated nephropathy in diabetic mice. J Am Soc of Nephrol 17(10):2664–2669 26. Kurtz TW, Morris RC, Pershadsingh HA (1989) The Zucker fatty rat as a genetic model of obesity and hypertension. Hypertension 13:896–901 27. Tilton RG, Haidacher SJ, Lejeune WS et al (2007) Diabetes induced changes in the renal cortical proteome assessed with twodimensional gel electrophoresis and mass spectrometry. Proteomics 7:1729–1742 28. Zhang D, Yang H, Kong X et al (2011) Proteomics analysis reveals diabetic kidney as a ketogenic organ in type 2 diabetes. Am J Physiol Endocrinol Metab 300:E287–E295 29. Barati MT, Merchant ML, Kain AB et al (2007) Proteomic analysis defines altered cellular redox pathways and advanced glycation endproduct metabolism in glomeruli of db/db diabetic mice. Am J Physiol Renal Physiol 293:F1157–F1165
30. Nakatani S, Kakehashi A, Ishimura E et al (2011) Targeted proteomics of isolated glomeruli from the kidneys of diabetic rats: sorbin and SH3 domain containing 2 is a novel protein associated with diabetic nephropathy. Exp Diabetes Res 2011:979354 31. Reimel BA, Pan S, May DH et al (2009) Proteomics on fixed tissue specimens—a review. Curr Proteomics 6:63–69 32. Ralton LD, Murray GI (2011) The use of formalin fixed wax embedded tissue for proteomic analysis. J Clin Pathol 64:297–302 33. Matsuda KM, Chung JY, Hewitt SM (2010) Histo-proteomic profiling of formalin-fixed, paraffin-embedded tissue. Expert Rev Proteomics 7:227–337 34. Murray GI (2012) Has the proteome of formalin-fixed waxembedded tissue been unlocked? Nephrol Dial Transplant 27(9):3395–3398 35. Emmert-Buck MR, Bonner RF, Smith PD et al (1996) Laser capture microdissection. Science 274(5289):998–1001 36. Nakatani S, Wei M, Ishimura E et al (2012) Proteome analysis of laser microdissected glomeruli from formalin-fixed paraffinembedded kidneys of autopsies of diabetic patients: nephronectin is associated with the development of diabetic glomerulosclerosis. Nephrol Dial Transplant 27(5):1889–1897 37. Satoskar AA, Shapiro JP, Bott CN et al (2012) Characterization of glomerular diseases using proteomic analysis of laser capture microdissected glomeruli. Mod Pathol 25(5):709–721 38. Casadonte R, Caprioli RM (2011) Proteomic analysis of formalin-fixed paraffin-embedded tissue by MALDI imaging mass spectrometry. Nat Protoc 6:1695–1709 39. Seeley EH, Schwamborn K, Caprioli RM (2011) Imaging of intact tissue sections: moving beyond the microscope. J Biol Chem 286:25459–25466 40. Thongboonkerd V (2008) Urinary proteomics: towards biomarker discovery, diagnostics and prognostics. Mol BioSyst 4:810–815 41. Thongboonkerd V (2010) Current stauts of renal and urinary proteomics: ready for routine clinical application? Nephrol Dial Transplant 25:11–16 42. Bramham K, Mistry HD, Poston L et al (2009) The non-invasive biopsy-will urinary proteomics make the renal tissue biopsy redundant? QJM 102:523–538 43. Thongboonkerd V (2007) Practical points in urinary proteomics. J Proteome Res 6:3881–3890 44. Barratt J, Topham P (2007) Urine proteomics: the present and future of measuring urinary protein components in disease. CMAJ 177:361–368 45. Gonzales-Buitrago JM, Ferreira L, Lorenzo I (2007) Urinary proteomics. Clin Chim Acta 375:49–56 46. Decramer S, Gonzales de Peredo A, Breuil B et al (2008) Urine in clinical proteomics. Mol Cell Proteomics 7:1850–1862 47. Schaub S, Wilkins J, Weiler T et al (2004) Urine protein profiling with surface-enhanced laser-desorption/ionization time-of-flight mass spectrometry. Kidney Int 65:323–332 48. Theodorescu D, Wittke S, Ross MM et al (2006) Discovery and validation of new protein biomarkers for urothelial cancer: a prospective analysis. Lancet Oncol 7:230–240 49. Papale M, Pedicillo MC, Thatcher BJ et al (2007) Urine profiling by SELDI–TOF/MS: monitoring of the critical steps in sample collection, handling and analysis. J Chromatogr B Analyt Technol Biomed Life Sci 856(1–2):205–213 50. Yamamoto T, Langham RG, Ronco P, Knepper MA, Thongboonkerd V (2008) Towards standard protocols and guidelines for urine proteomics: a report on the Human Kidney and Urine Proteome Project (HKUPP) symposium and workshop, 6 October 2007, Seoul, Korea and 1 November 2007, San Francisco, CA, USA. Proteomics 8(11):2156–2159 51. Jackson DH, Banks RE (2010) Banking of clinical samples for proteomic biomarker studies: a consideration of logistical issues
123
J Nephrol
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
with a focus on pre-analytical variation. Proteomics Clin Appl 4(3):250–270 Calvano CD, Aresta A, Iacovone M et al (2010) Optimization of analytical and pre-analytical conditions for MALDI-TOF-MS human urine protein profiles. J Pharm Biomed Anal 51(4):907–914 Court M, Selevsek N, Matondo M et al (2011) Toward a standardized urine proteome analysis methodology. Proteomics 11(6):1160–1171 Yates JR, Ruse CI, Nakorchevsky A (2009) Proteomics by mass spectrometry: approaches, advances, and applications. Annu Rev Biomed Eng 11:49–79 Kolch W, Neussus C, Pelzing M et al (2005) Capillary electrophoresis-mass spectrometry as a power tool in clinical diagnosis and biomarker discovery. Mass Spectrom Rev 34:959–977 Wright GL Jr (2002) SELDI proteinchips MS: a platform for biomarker discovery and cancer diagnosis. Expert Rev Mol Diagn 2:549–563 Rossing K, Mischak H, Dakna M, PREDICTIONS Network et al (2008) Urinary proteomics in diabetes and CKD. J Am Soc Nephrol 19(7):1283–1290 Alkhalaf A, Zu¨rbig P, Bakker SJ, PREDICTIONS Group et al (2010) Multicentric validation of proteomic biomarkers in urine specific for diabetic nephropathy. PLoS ONE 5(10):e13421 Good DM, Zu¨rbig P, Argile´s A et al (2010) Naturally occurring human urinary peptides for use in diagnosis of chronic kidney disease. Mol Cell Proteomics 9:2424–2437 Zu¨rbig P, Jerums G, Hovind P et al (2012) Urinary proteomics for early diagnosis in diabetic nephropathy. Diabetes 61(12):3304–3313 Otu HH, Can H, Spentzos D, Nelson RG et al (2007) Prediction of diabetic nephropathy using urine proteomic profiling 10 years prior to development of nephropathy. Diabetes Care 30(3):638–643 Wu J, Chen YD, Yu JK, Shi XL, Gu W (2011) Analysis of urinary proteomic patterns for type 2 diabetic nephropathy by Protein Chip. Diabetes Res Clin Pract 91(2):213–219
123
63. Dihazi H, Mu¨ller GA, Lindner S et al (2007) Characterization of diabetic nephropathy by urinary proteomic analysis: identification of a processed ubiquitin form as a differentially excreted protein in diabetic nephropathy patients. Clin Chem 53(9):1636–1645 64. MacIsaac RJ, Tsalamandris C, Panagiotopoulos S, Smith TJ, McNeil KJ, Jerums G (2004) Nonalbuminuric renal insufficiency in type 2 diabetes. Diabetes Care 27:195–200 65. Ekinci EI, Jerums G, Skene A et al (2013) Renal structure in normoalbuminuric and albuminuric patients with type 2 diabetes and impaired renal function. Diabetes Care 36:3620–3626 66. Papale M, Di Paolo S, Magistroni R et al (2010) Urine proteome analysis may allow noninvasive differential diagnosis of diabetic nephropathy. Diabetes Care 33(11):2409–2415 67. Jin J, Ku YH, Kim Y et al (2012) Differential proteome profiling using iTRAQ in microalbuminuric and normoalbuminuric type 2 diabetic patients. Exp Diabetes Res 2012:168602 68. Shadforth IP, Dunkley TP, Lilley KS, Bessant C (2005) i-Tracker: for quantitative proteomics using iTRAQ. BMC Genom 20(6):145 69. Cocucci E, Racchetti G, Meldolesi J (2009) Shedding microvesicles: artefacts no more. Trends Cell Biol 19:43–51 70. Simpson RJ, Lim JW, Moritz RL, Mathivanan S (2009) Exosomes: proteomic insights and diagnostic potential. Expert Rev Proteomics. 6:267–283 71. Raimondo F, Corbetta S, Morosi L et al (2013) Urinary exosomes and diabetic nephropathy: a proteomic approach. Mol BioSyst 9(6):1139–1146 72. Papale M, Rocchetti MT, Gesualdo L (2013) Clinical proteomics: the potentiality of urine analysis for understanding diabetic nephropathy. Euro Med J Nephrol (1):32–39 73. Kohda Y et al (2000) Analysis of segmental renal gene expression by laser capture microdissection. Kidney Int 57:321–331 74. Chaurand P et al (2006) Molecular imaging of thin mammalian tissue sections by mass spectrometry. Curr Opin Biotechnol 17(4):431–436
INVITED REVIEW
Proteomics and diabetic nephropathy: what have we learned from a decade of clinical proteomics studies? Massimo Papale • Salvatore Di Paolo • Grazia Vocino • Maria Teresa Rocchetti Loreto Gesualdo
•
Received: 18 September 2013 / Accepted: 15 October 2013 Ó Italian Society of Nephrology 2014
Abstract Diabetic nephropathy (DN) has become the most frequent cause of chronic kidney disease worldwide due to the constant increase of the incidence of type 2 diabetes mellitus in developed and developing countries. The understanding of the pathophysiological mechanisms of human diseases through a large-scale characterization of the protein content of a biological sample is the key feature of the proteomics approach to the study of human disease. We discuss the main results of over 10 years of tissue and urine proteomics studies applied to DN in order to understand how far we have come and how far we still have to go before obtaining a full comprehension of the molecular mechanisms involved in the pathogenesis of DN and identifying reliable biomarkers for accurate management of patients. Keywords Diabetic nephropathy Proteome Proteomics Urine Tissues Biomarkers
M. Papale Core Facility of Proteomics and Mass Spectrometry, Department of Surgery and Medical Sciences, University of Foggia, Foggia, Italy S. Di Paolo Nephrology Unit, ‘‘Dimiccoli’’ Hospital, Barletta, Italy G. Vocino M. T. Rocchetti L. Gesualdo (&) Nephrology, Dialysis and Transplantation Unit, Department of Emergency and Organ Transplantation (D.E.T.O), University of Bari ‘‘Aldo Moro’’, Bari, Italy e-mail: [email protected]
The pathophysiology of diabetes and diabetic nephropathy Diabetic nephropathy (DN) is a severe complication of both type 2 (T2DM) and type 1 diabetes mellitus (T1DM). It has become the most common cause of chronic kidney disease (CKD) in developed countries [1] and the most frequent cause of end-stage renal disease (ESRD) worldwide given that DN [2] affects about 20-40 % of nephropathic patients undergoing dialysis. Historically, the incidence of DN in T1DM exceeds that recorded in T2DM, since in the latter group death from cardiovascular causes generally antedates the appearance of renal failure [3, 4]. However, the use of renin–angiotensin system inhibitors along with a strict glycemic control is improving the life expectancy of T2DM patients and slowing the rate of mortality due to cardiovascular causes [5]. As a result, it is expected that the incidence of DN in T2DM patients will significantly increase over the next few years [6]. Chronic hyperglycemia triggers specific modifications of electron transport proteins via advanced glycation end-products (AGEs). This alters normal mitochondrial metabolism and leads to an increased production of reactive oxygen species (ROS) [7], which in turn strongly contribute to chronic renal damage in T2DM. On the other hand, less than 50 % of T2DM patients finally develop DN, which likely highlights the role of genetic background and lifestyle modifications [8]. Interestingly, the increased oxidative stress induces long-lasting effects which persist even after glycemia normalization, and lead to a persistent activation of endothelial nitric oxide synthase and prostacyclin synthase, which, in turn, contributes to defective angiogenesis and persistent expression of pro-inflammatory genes [9]. The onset and progression of renal damage in T2DM proceeds through distinct and sequential phases: it has been
123
J Nephrol
demonstrated that the initial renal response to hyperglycemia is glomerular hyperfiltration and hyperperfusion, a process mediated by vasoactive hormones such as angiotensin II [10–13]. Even if these early hemodynamic changes are positively modulated by normalization of blood glucose levels, they probably trigger the incipient molecular events that, in some patients, induce the progression of the disease. Incipient DN is then characterized by albumin leakage from glomerular capillaries, overproduction and extracellular accumulation of mesangial cell matrix, thickening of the glomerular basement membrane and podocytes injury [14–16]. At the urinary level the earliest putative diagnostic sign of diabetic renal damage is microalbuminuria (urine albumin excretion 30–300 mg/24 h) which, unfortunately, does not always correlate with the complex histopathological picture of glomerular and tubular damage of T2DM [17, 18]; thus this parameter represents a predictor of cardiovascular disease rather than of renal damage progression [19]. The rapid development of new high-throughput approaches for the massive analysis of the protein content of biological samples and the possibility to perform fast and accurate in silico analysis of complex datasets, that was unthinkable until few years ago, is now making it possible to better characterize the key events occurring during the onset and progression of DN and to identify more accurate disease-specific biomarkers. In the following paragraphs we will discuss the main contributions of proteome analysis to the understanding of DN pathophysiology and to the identification of new sets of biomarkers potentially useful for achieving an early diagnosis and a better management of patients. The present review represents the fourth part of a series of papers [20–22] published by the Study Group on ‘‘OMICS’’ sciences of the Italian Society of Nephrology (http://www.sin-italy.org/Soci_ GdS/gruppi_omiche.asp), addressing the main contributions of the Omics sciences to key-relevant topics in renal pathophysiology.
Tissue proteomics and diabetic nephropathy The application of proteomic technologies to the study of morpho-structural and functional changes of the kidney represents the most focused approach to understanding the molecular events that initiate and propagate renal damage. However, renal biopsy is poorly applied to diabetic patients and the progression of renal damage is commonly gathered from clinical and/or laboratory findings, such as the measurement of urine albumin excretion rate. Further, the small amounts of renal tissue obtained from renal biopsy specimens make it difficult to perform a satisfactory proteomic analysis. As a consequence, most of the studies have been
123
carried out on animal models mimicking the renal damage occurring in humans with diabetes [23–26]. Tilton et al. [27] used two-dimensional gel electrophoresis (2DE) to study the renal cortex proteins whose expression was deregulated in the db/db (diabetic mice) model of T2DM. More than 50 % of the deregulated proteins were involved in mitochondrial metabolic pathways including lipid and fatty acid metabolism. Ingenuity pathways analysis (IPA) identified peroxisome proliferator activated receptor c (PPAR-c) as a central player in this network. Further, Zhang et al. [28] recognized in the same animal model a significant increase of ketogenesis and Barati et al. [29] reported a significant increase of antioxidant and glycation enzymes. Of note, the above studies analysed whole tissue samples, making it more difficult to associate the different expression of disease-related proteins to a specific renal sub-compartment. Proteomic analysis of isolated glomeruli allowed to identify a panel of 17 significantly and differentially expressed actin cytoskeleton-associated proteins [30] and to characterize SORBS2, an adapter protein presumably involved in the assembly of the actin stress fibres, as a central player in the pathogenesis of DN. Interestingly, the results of the above studies are poorly replicable in humans due to the difficulty of performing efficient isolation of the glomeruli from human biopsies. Further, most tissue samples available in humans are those routinely prepared for diagnostic purposes, i.e. they are formalin fixed and paraffin embedded (FFPE). Recent methodological improvements [31–33] have demonstrated that it is possible to extract intact and unmodified proteins from FFPE samples and further analyze them by mass spectrometry (MS) thus permitting the application of proteomic technologies to a vast array of formalin fixed wax embedded tissue samples stored in pathology department archives [34]. In order to perform compartment-specific analyses, glomeruli and tubuli need to be dissected from FFPE tissues by laser capture microdissection (LCM) [35] and directly analyzed by proteomic techniques. Proteomic analysis of LCM-isolated glomeruli revealed an increased and selective expression of nephronectin, a protein implicated in the assembly of extracellular matrix, in diabetic nephropathy [36]. Further, a similar study described, in the glomeruli of DN patients, the presence of complement factors such as C3 and the membrane attack complex (C5b-9) as well as a marked decrease of podocyte-associated proteins and antioxidant proteins [37]. Alternatively, FFPE tissues may be directly analyzed by imaging mass spectrometry (IMS) [38, 39] to provide both the identification of disease-related biomarkers and their localization in specific renal compartments. This approach may allow the molecular mapping of changes observed in diabetic kidney in both experimental models and human disease, and to identify the major molecular alterations linked to the development of diabetic nephropathy.
Renal tissue
Renal tissue (glomeruli)
Renal tissue (isolated glomeruli)
Renal tissue (microdissected glomeruli)
Renal tissue (microdissected glomeruli from biopsies)
Urine samples
Urine samples
Urine samples
Zhang et al. [28]
Barati et al. [29]
Nakatani et al. [30]
Nakatani et al. [36]
Satoskar et al. [37]
Zurbig et al. [60]
Dihazi et al. [63]
Papale et al. [66]
Urine samples
Renal cortex
Tilton et al. [27]
Jin et al. [67]
Sample type
References
Normoalbuminuric vs. microalbuminuric T2DM patients
Normoalbuminuric vs. microalbuminuric vs. biopsyproven DN patients
Normoalbuminuric vs. microalbuminuric vs. macroalbuminuric T2DM patients
Normoalbuminiric diabetic patients developing microalbuminuria over the time
FFPE tissue (human)
FFPE tissue (human)
OLETF rats (spontaneous type 2 diabetic) vs. LETO rats (in the early and proteinuric stages of diabetic nephropathy)
db/db mice
db/db mice
db/db mice
Source
iTRAQ labelling ? LC–MS/ MS; WB
SELDI TOF/MS; ELISA; WB
SELDI–TOF/MS; WB
CE-MS
LCM ? LC/MS/MS; IF
LCM ? LC/MS/MS; IHC
LC-ESI MS/MS; IHC; IF
2DE-MALDI-TOF MS; WB; IHC
2DE-MALDI-TOF MS; WB
2DE-MALDI-TOF MS; WB; IPA
Methods
Table 1 Key proteomic studies applied to tissue and urine in diabetic nephropathy
:Prostate stem cell antigen
:Alpha-1-acid glycoprotein 1
:Alpha-1-antitrypsin
Immunity, cell adhesion, protein metabolism, signal transduction
Tubular injury and ubiquitination
:b2-Microglobulin :Ubiquitin
Tubular injury and ubiquitination
Assembly and accumulation of mesangial extracellular matrix
Activation of complement and oxidative stress
Assembly and accumulation of mesangial extracellular matrix
Cytoskeleton rearrangement
Cellular redox regulation and detoxification
:b2-microglobulin :ubiquitin ribosomal fusion protein (UbA52)
:b-Microglobulin
;Uromodulin
:I a 1-Antitrypsin
;Collagen a1 chain
:Synaptopodin
:Fibrinogen a-chain
:C5b-9
:C3
;Actinin, alpha 4
:Laminin, beta 2
:Integrin, alpha 1
:Nephronectin
:Tubulin alpha 1c
:Actin-related protein 2/3 complex subunit 1 beta
:Alpha-actinin
:Sorbin and SH3 domain containing 2
:Glyoxalase I
:Superoxide dismutase 1
:Glutathione peroxidase 1
:Peroxiredoxin 1
;Protein kinase C inhibitor protein 1 :3-hydroxy-3-methylglutaryl-CoA synthase (HMGCS2)
Ketogenesis
Mitochondrial metabolism; lipid and fatty acid metabolism
:Peroxisome proliferator activated receptor c (PPAR-c) :Phosphoenolpyruvate carboxykinase 1 (Pepck)
Pathways
Key features
J Nephrol
123
123
WB western blotting, IPA ingenuity pathway analysis, IHC immunohistochemistry, IF immunofluorescence, FFPE formalin fixed and paraffin embedded, DN diabetic nephropathy, T2DM type 2 diabetes mellitus, 2DE-MALDI-TOF MS two-dimensional gel electrophoresis matrix-assisted laser desorption/ionization time-of-flight mass spectrometry, LC-ESI/MS/MS liquid chromatography-electrospray ionization/multi-stage mass spectrometry, LCM laser capture microdissection, CE capillary electrophoresis, SELDI–TOF/MS surface enhanced laser desorption/ionization time-of-flight mass spectrometry, ELISA enzyme-linked immunosorbent assay, iTRAQ isobaric tags for relative and absolute quantitation, SDS-PAGE sodium dodecyl sulphate-polyacrylamide gel electrophoresis
Intracellular signaling and transport; cell adhesion ;Major urinary protein 1
:Xaa-Pro dipeptidase
SDS-PAGE ? nLC-ESI–MS/ MS; WB Urine exosomes Raimondo et al. [71]
Zucker diabetic fatty (ZDF) rats
Sample type References
Table 1 continued
Source
Methods
Key features
Pathways
J Nephrol
Specifically, IMS applied to experimental models of diabetes, or to human biopsy samples collected in the very early stages of renal injury, would possibly allow the identification of the molecular markers heralding the progression of renal damage. Even if there is not any original study so far exploring this topic, a wide application of this approach is expected in the forthcoming years. On the whole, the technological and methodological improvements in tissue proteomics will without doubt contribute revealing new pathogenic mechanisms of renal damage in T2DM (Table 1).
Urine proteomics and diabetic nephropathy Urine has been recognized as one of the most appropriate biological fluids for biomarker discovery in kidney diseases [40–46]. Some peculiarities of urine could make it the gold standard for kidney biomarker discovery studies. They include the easy and non-invasive accessibility, the presence of both kidney-derived (about 70 %) and plasmaderived (about 30 %) proteins, and the lower complexity and increased stability of the urine proteome when compared to that of other biological fluids [47–49]. However, a number of published studies have pointed out the inherent limitations of this biological fluid (i.e. presence of salts and contaminants; high intra- and inter-subject variability; release, in nephropathic patients, of high amounts of serum proteins which makes the identification of the scarcely excreted biomarkers really burdensome) [43, 49–53]. Thus, the use of standardized protocols for the collection, management and analysis of urine is crucial to maximize the potentiality of this sample. Although the 2-DE based analysis of the urine proteome can provide useful information about the key changes of the urine proteome in DN patients, it is clear that the identification of new reliable biomarkers requires the setup of multicentre studies and the analysis of thousands of biological samples that can be possible only by means of massive screening by high throughput strategies such as liquid chromatography (LC) [54], capillary electrophoresis (CE) [55], and thin-layer chromatography [56] coupled to mass spectrometry (MS). In order to provide a qualified picture of the main developments in DN biomarker discovery we will limit the further discussion to the studies based on the use of high throughput strategies. CE allows the identification, in urine from type 1 diabetics, of 65 urine peptides, many of them being type I collagen fragments, making it possible to recognize the presence of DN with 97 % sensitivity and specificity [57]. The validation of this pattern in a multicentre independent cohort of type 2 diabetic patients provided the first evidence that the presence of collagen fragments may allow identification of DN in both type 1 and type 2 DM [58]. Interestingly,
J Nephrol
most of the 65 urine peptides were further included in a new CE-MS urinary pattern, composed of 273 urinary small peptides (called ‘‘classifier 273’’), that was highly specific and sensitive for CKD irrespective of the underlying pathology [59]. Recently, Zurbig et al. [60] evaluated the predictive value of classifier 273 on T1DM and T2DM normoalbuminuric patients and demonstrated that the classifier was able to reliably predict the progression to macroalbuminuria, long before the modification of albumin excretion rate. Surface-enhanced laser desorption ionization (SELDI)time-of-flight (TOF)-MS analysis is a further high throughput technology which has been extensively used for identifying urine biomarkers of DN. Adopting this proteomic technique, Otu et al. [61] detected a 12-peak signature predicting the development of diabetic nephropathy ten years prior to the increase of albumin to creatinine ratio, while Wu et al. [62] recently described a 4-peak pattern able to recognize T2DM patients with DN with 88 % sensitivity and 97 % specificity. Unfortunately, both these studies failed to identify any of the candidate biomarkers, thus limiting the possibility to test the predictive and diagnostic power of these models in independent testing sets. Two DN biomarkers (b2-microglobulin and ubiquitin ribosomal fusion protein) were instead purified and identified by Dihazi et al. [63] among more than 100 mass peaks, which were differentially excreted in the urine of patients with diabetic nephropathy. Indeed, all the above studies identified patients with diabetic nephropathy based upon their albumin excretion rate, without any investigation of the renal lesions underlying the clinical features. In fact, renal changes different from diabetic glomerulosclerosis can be detected in over 40 % of type 2 diabetics with proteinuria [18]. Conversely, typical renal structural changes of diabetic nephropathy may be found in diabetics with normoalbuminuria and renal insufficiency [64, 65]. Papale et al. [66] performed a SELDI analysis on the urine of type 2 diabetics comparing patients with biopsy proven Kimmestiel–Wilson lesions with those displaying non diabetic chronic renal changes (CKD). Their results suggested that the combination of b2-microglobulin and free ubiquitin might make it possible to set up a diagnostic model able to distinguish DN from CKD in both non-diabetic and diabetic patients. Furthermore, Jin et al. [67] used an LC/MS based analysis with isobaric tags for relative and absolute quantitation (iTRAQ) [68] to select and quantify differentially excreted urinary proteins in pooled urine samples of microalbuminuric versus normoalbuminuric diabetic patients. The bioinformatics management of the entire dataset (196 differentially expressed proteins) allowed to restrict the further validation to ten disease-correlated biomarkers. Further, the combination of three candidate biomarkers (alpha-1-antitrypsin, alpha-1-acid glycoprotein 1, and prostate stem cell antigen) produced a receiver operating characteristic (ROC) curve with the highest diagnostic power (about 92 %) for diabetic microalbuminuric patients.
It is worth noting that most urine proteomic studies limited the analysis to the soluble fraction, but diseasespecific biomarkers may be detected also in cells and microvesicles. In particular, urinary exosomes, 30–100 nm diameter vesicles, derived from the endosomal compartment and released via fusion of multivesicular bodies with the plasma membrane [69], are emerging as a new source of potential biomarkers [70]. These vesicles are released from every epithelial cell lining the nephron, including podocytes, and may deliver segment-specific biomarkers in the urine. The recent study by Raimondo et al. [71] emphasizes this concept: the authors reported, in Zucker diabetic fatty (ZDF) rats undergoing progressive worsening of renal function, significant changes in the urine exosomes’ proteome revealing how these microvesicles may provide detailed information about metabolic and immune processes correlated with the onset and progression of renal damage in T2DM. Taken together, the above results demonstrate that the proteomic analysis of the urinary exosomes, together with the analysis of the soluble urinary proteins, may fruitfully contribute to reveal the pathophysiological alterations occurring in DN progression, and to enlarge the panel of DN biomarker candidates (Table 1).
Conclusion and perspectives The proteomic approach to the study of diabetic nephropathy has made a decisive contribution to the understanding of DN pathophysiology and the identification of disease-related biomarkers. In the last 10 years, the advances have been much more than we may have perceived. In fact, there have been concomitant improvements in many fields crucial for the comprehension of the pathogenesis of diabetic renal damage: the definition of animal models mimicking the pathophysiology of type 2 diabetic renal damage in humans; the setup of new protocols for the direct proteomic analysis of FFPE tissues; the standardization of urine collection, management and storage; the development of new MS analyzers that may allow the detection and characterization even of barely expressed biomarkers; the development of sophisticated bioinformatics tools to manage complex datasets. All these aspects are contributing to establish an optimal workflow for the discovery and validation of reliable protein biomarkers in T2DM as well as in other complex diseases (Fig. 1). In the near future, we will probably be able to combine the power of proteomic analysis of human tissues and biological fluids with the bioinformatics management of complex datasets, in order to collect enough information to yield a reliable picture of the key events occurring during the onset and progression of renal damage in T2DM. This new way to manage proteomic datasets will probably favour the identification of reliable biomarkers by reducing the effect of confounding factors. Moreover, the
123
J Nephrol
Fig. 1 Workflow of the proteomic analysis applied to tissue and urine samples. Left panel Kidney sub-compartments (i.e. glomeruli, tubuli, etc.) can be isolated from whole biopsy specimens through laser capture microdissection (LCM) prior to performing proteomic analysis. Alternatively, frozen or formalin fixed and paraffin embedded (FFPE) tissues can be directly analysed by matrix-assisted laser desorption/ionization imaging mass spectrometry (MALDI-IMS). Right panel Ultracentrifugation of whole urine samples is used to separate soluble and vesicular fractions, then proteomic analysis may
allow the identification of compartment-specific biomarkers. The bioinformatic management of urine and tissue putative biomarkers is then applied to each dataset in order to select functionally correlated disease biomarkers [72]. Comparative analysis may finally allow to identify tissue-derived biomarkers also detectable in the urine samples. LCM glomeruli and MALDI-IMS figures are reproduced with permission from Kohda et al. [73] and Chaurand et al. [74], respectively
complex intracellular behaviour implies a constant interaction of proteins, genes, transcripts and metabolites that all together contribute to the observed phenotypes. Bioinformatics will have to develop more accurate tools to correlate specific proteomic datasets with the corresponding genomic, transcriptomic and metabolomic profiles in order to pursue a global characterization of the biological systems and identify a multilevel panel of molecular players contributing to the onset of pathological phenotypes.
National Institute of Diabetes and Digestive and Kidney Diseases, Bethesda (2007) KDOQI Clinical Practice Guidelines and Clinical Practice Recommendations for Diabetes and Chronic Kidney Disease. Am J Kidney Dis 49:S1–S179 Craig KJ, Donovan K, Munnery M, Owens DR, Williams JD, Phillips AO (2003) Identification and management of diabetic nephropathy in diabetes clinic. Diabetes Care 26:1806–1811 Gross JL, de Azevedo MJ, Silverio SP, Canani LH, Caramori ML, Zelmanovitz T (2005) Diabetic nephropathy: diagnosis, prevention, and treatment. Diabetes Care 28:164–174 Ziyadeh FN, Sharma K (2003) Overview: combating diabetic nephropathy. J Am Soc Nephrol 14:1355–1357 Stewart JH, McCredie MR, Williams SM et al (2007) Trends in incidence of treated end-stage renal disease, overall and by primary renal disease, in persons aged 20–64 years in Europe, Canada and the Asia-Pacific region, 1998–2002. Nephrology (Carlton) 12:520–527 Rosca MG, Mustata TG, Kinter MT et al (2005) Glycation of mitochondrial proteins from diabetic rat kidney is associated with excess superoxide formation. Am J Physiol Renal Physiol 289:F420–F430
2.
3.
4.
5. Conflict of interest of interest.
The authors declare that there are no conflicts
6.
References 7. 1. U.S. Renal Data System (2012) USRDS 2012 Annual Data Report: Atlas of Chronic Kidney Disease and End-Stage Renal Disease in the United States. National Institutes of Health,
123
J Nephrol 8. Krolewski AS, Warram JH (1995) Genetic susceptibility to diabetic kidney disease: an update. J Diabetes Complications 9(4):277–2781 9. Giacco F, Brownlee M (2010) Oxidative stress and diabetic complications. Circ Res 107:1058–1070 10. Takenaka T, Inoue T, Ohno Y et al (2012) Elucidating mechanisms underlying altered renal autoregulation in diabetes. Am J Physiol Regul Integr Comp Physiol 303(5):R495–R504 11. Patinha D, Fasching A, Pinho D, Albino-Teixeira A, Morato M, Palm F (2013) Angiotensin II contributes to glomerular hyperfiltration in diabetic rats independently of adenosine type I receptors. Am J Physiol Renal Physiol 304(5):F614–F622 12. Sochett EB, Cherney DZ, Curtis JR, Dekker MG, Scholey JW, Miller JA (2006) Impact of renin angiotensin system modulation on the hyperfiltration state in type 1 diabetes. J Am Soc Nephrol 17(6):1703–1709 13. Mathis KM, Banks RO (1996) Role of nitric oxide and angiotensin II in diabetes mellitus-induced glomerular hyperfiltration. J Am Soc Nephrol 7(1):105–112 14. Kim HJ, Cho EH, Yoo JH et al (2007) Proteome analysis of serum from type 2 diabetics with nephropathy. J Proteome Res 6:735–743 15. Ziyadeh FN, Snipes ER, Watanabe M, Alvarez RJ, Goldfarb S, Haverty TP (1990) High glucose induces cell hypertrophy and stimulates collagen gene transcription in proximal tubule. Am J Physiol 259:F704–F714 16. Schordan S, Schordan E, Endlich N et al (2009) Alterations of the podocyte proteome in response to high glucose concentrations. Proteomics 9:4519–4528 17. Colantonio DA, Chan DW (2005) The clinical application of proteomics. Clin Chim Acta 357:151–158 18. Mazzucco G, Bertani T, Fortunato M et al (2002) Different patterns of renal damage in type 2 diabetes mellitus: a multicentric study on 393 biopsies. Am J Kidney Dis 39(4):713–720 19. Thongboonked V, Malasit P (2005) Renal and urinary proteomics: current applications and challenges. Proteomics 5:1033–1042 20. Bonomini M, Sirolli V, Magni F, Urbani A (2012) Proteomics and nephrology. J Nephrol 25(06):865–871 21. Magni F, Lalowski M, Mainini V et al (2013) Proteomics imaging and the kidney. J Nephrol 26(3):430–436 22. Santucci L, Candiano G, Bruschi M et al (2013) Urinary proteome in a snapshot: normal urine and glomerulonephritis. J Nephrol 26(4):610–616 23. Hummel KP, Dickie MM, Coleman DL (1966) Diabetes, a new mutation in the mouse. Science 153(3740):1127–1128 24. Chua SC, Chung WK, Wu-Peng XS et al (1996) Phenotypes of mouse diabetes and rat fatty due to mutations in the OB (leptin) receptor. Science 271(5251):994–996 25. Zhao HJ, Wang S, Cheng H et al (2006) Endothelial nitric oxide synthase deficiency produces accelerated nephropathy in diabetic mice. J Am Soc of Nephrol 17(10):2664–2669 26. Kurtz TW, Morris RC, Pershadsingh HA (1989) The Zucker fatty rat as a genetic model of obesity and hypertension. Hypertension 13:896–901 27. Tilton RG, Haidacher SJ, Lejeune WS et al (2007) Diabetes induced changes in the renal cortical proteome assessed with twodimensional gel electrophoresis and mass spectrometry. Proteomics 7:1729–1742 28. Zhang D, Yang H, Kong X et al (2011) Proteomics analysis reveals diabetic kidney as a ketogenic organ in type 2 diabetes. Am J Physiol Endocrinol Metab 300:E287–E295 29. Barati MT, Merchant ML, Kain AB et al (2007) Proteomic analysis defines altered cellular redox pathways and advanced glycation endproduct metabolism in glomeruli of db/db diabetic mice. Am J Physiol Renal Physiol 293:F1157–F1165
30. Nakatani S, Kakehashi A, Ishimura E et al (2011) Targeted proteomics of isolated glomeruli from the kidneys of diabetic rats: sorbin and SH3 domain containing 2 is a novel protein associated with diabetic nephropathy. Exp Diabetes Res 2011:979354 31. Reimel BA, Pan S, May DH et al (2009) Proteomics on fixed tissue specimens—a review. Curr Proteomics 6:63–69 32. Ralton LD, Murray GI (2011) The use of formalin fixed wax embedded tissue for proteomic analysis. J Clin Pathol 64:297–302 33. Matsuda KM, Chung JY, Hewitt SM (2010) Histo-proteomic profiling of formalin-fixed, paraffin-embedded tissue. Expert Rev Proteomics 7:227–337 34. Murray GI (2012) Has the proteome of formalin-fixed waxembedded tissue been unlocked? Nephrol Dial Transplant 27(9):3395–3398 35. Emmert-Buck MR, Bonner RF, Smith PD et al (1996) Laser capture microdissection. Science 274(5289):998–1001 36. Nakatani S, Wei M, Ishimura E et al (2012) Proteome analysis of laser microdissected glomeruli from formalin-fixed paraffinembedded kidneys of autopsies of diabetic patients: nephronectin is associated with the development of diabetic glomerulosclerosis. Nephrol Dial Transplant 27(5):1889–1897 37. Satoskar AA, Shapiro JP, Bott CN et al (2012) Characterization of glomerular diseases using proteomic analysis of laser capture microdissected glomeruli. Mod Pathol 25(5):709–721 38. Casadonte R, Caprioli RM (2011) Proteomic analysis of formalin-fixed paraffin-embedded tissue by MALDI imaging mass spectrometry. Nat Protoc 6:1695–1709 39. Seeley EH, Schwamborn K, Caprioli RM (2011) Imaging of intact tissue sections: moving beyond the microscope. J Biol Chem 286:25459–25466 40. Thongboonkerd V (2008) Urinary proteomics: towards biomarker discovery, diagnostics and prognostics. Mol BioSyst 4:810–815 41. Thongboonkerd V (2010) Current stauts of renal and urinary proteomics: ready for routine clinical application? Nephrol Dial Transplant 25:11–16 42. Bramham K, Mistry HD, Poston L et al (2009) The non-invasive biopsy-will urinary proteomics make the renal tissue biopsy redundant? QJM 102:523–538 43. Thongboonkerd V (2007) Practical points in urinary proteomics. J Proteome Res 6:3881–3890 44. Barratt J, Topham P (2007) Urine proteomics: the present and future of measuring urinary protein components in disease. CMAJ 177:361–368 45. Gonzales-Buitrago JM, Ferreira L, Lorenzo I (2007) Urinary proteomics. Clin Chim Acta 375:49–56 46. Decramer S, Gonzales de Peredo A, Breuil B et al (2008) Urine in clinical proteomics. Mol Cell Proteomics 7:1850–1862 47. Schaub S, Wilkins J, Weiler T et al (2004) Urine protein profiling with surface-enhanced laser-desorption/ionization time-of-flight mass spectrometry. Kidney Int 65:323–332 48. Theodorescu D, Wittke S, Ross MM et al (2006) Discovery and validation of new protein biomarkers for urothelial cancer: a prospective analysis. Lancet Oncol 7:230–240 49. Papale M, Pedicillo MC, Thatcher BJ et al (2007) Urine profiling by SELDI–TOF/MS: monitoring of the critical steps in sample collection, handling and analysis. J Chromatogr B Analyt Technol Biomed Life Sci 856(1–2):205–213 50. Yamamoto T, Langham RG, Ronco P, Knepper MA, Thongboonkerd V (2008) Towards standard protocols and guidelines for urine proteomics: a report on the Human Kidney and Urine Proteome Project (HKUPP) symposium and workshop, 6 October 2007, Seoul, Korea and 1 November 2007, San Francisco, CA, USA. Proteomics 8(11):2156–2159 51. Jackson DH, Banks RE (2010) Banking of clinical samples for proteomic biomarker studies: a consideration of logistical issues
123
J Nephrol
52.
53.
54.
55.
56.
57.
58.
59.
60.
61.
62.
with a focus on pre-analytical variation. Proteomics Clin Appl 4(3):250–270 Calvano CD, Aresta A, Iacovone M et al (2010) Optimization of analytical and pre-analytical conditions for MALDI-TOF-MS human urine protein profiles. J Pharm Biomed Anal 51(4):907–914 Court M, Selevsek N, Matondo M et al (2011) Toward a standardized urine proteome analysis methodology. Proteomics 11(6):1160–1171 Yates JR, Ruse CI, Nakorchevsky A (2009) Proteomics by mass spectrometry: approaches, advances, and applications. Annu Rev Biomed Eng 11:49–79 Kolch W, Neussus C, Pelzing M et al (2005) Capillary electrophoresis-mass spectrometry as a power tool in clinical diagnosis and biomarker discovery. Mass Spectrom Rev 34:959–977 Wright GL Jr (2002) SELDI proteinchips MS: a platform for biomarker discovery and cancer diagnosis. Expert Rev Mol Diagn 2:549–563 Rossing K, Mischak H, Dakna M, PREDICTIONS Network et al (2008) Urinary proteomics in diabetes and CKD. J Am Soc Nephrol 19(7):1283–1290 Alkhalaf A, Zu¨rbig P, Bakker SJ, PREDICTIONS Group et al (2010) Multicentric validation of proteomic biomarkers in urine specific for diabetic nephropathy. PLoS ONE 5(10):e13421 Good DM, Zu¨rbig P, Argile´s A et al (2010) Naturally occurring human urinary peptides for use in diagnosis of chronic kidney disease. Mol Cell Proteomics 9:2424–2437 Zu¨rbig P, Jerums G, Hovind P et al (2012) Urinary proteomics for early diagnosis in diabetic nephropathy. Diabetes 61(12):3304–3313 Otu HH, Can H, Spentzos D, Nelson RG et al (2007) Prediction of diabetic nephropathy using urine proteomic profiling 10 years prior to development of nephropathy. Diabetes Care 30(3):638–643 Wu J, Chen YD, Yu JK, Shi XL, Gu W (2011) Analysis of urinary proteomic patterns for type 2 diabetic nephropathy by Protein Chip. Diabetes Res Clin Pract 91(2):213–219
123
63. Dihazi H, Mu¨ller GA, Lindner S et al (2007) Characterization of diabetic nephropathy by urinary proteomic analysis: identification of a processed ubiquitin form as a differentially excreted protein in diabetic nephropathy patients. Clin Chem 53(9):1636–1645 64. MacIsaac RJ, Tsalamandris C, Panagiotopoulos S, Smith TJ, McNeil KJ, Jerums G (2004) Nonalbuminuric renal insufficiency in type 2 diabetes. Diabetes Care 27:195–200 65. Ekinci EI, Jerums G, Skene A et al (2013) Renal structure in normoalbuminuric and albuminuric patients with type 2 diabetes and impaired renal function. Diabetes Care 36:3620–3626 66. Papale M, Di Paolo S, Magistroni R et al (2010) Urine proteome analysis may allow noninvasive differential diagnosis of diabetic nephropathy. Diabetes Care 33(11):2409–2415 67. Jin J, Ku YH, Kim Y et al (2012) Differential proteome profiling using iTRAQ in microalbuminuric and normoalbuminuric type 2 diabetic patients. Exp Diabetes Res 2012:168602 68. Shadforth IP, Dunkley TP, Lilley KS, Bessant C (2005) i-Tracker: for quantitative proteomics using iTRAQ. BMC Genom 20(6):145 69. Cocucci E, Racchetti G, Meldolesi J (2009) Shedding microvesicles: artefacts no more. Trends Cell Biol 19:43–51 70. Simpson RJ, Lim JW, Moritz RL, Mathivanan S (2009) Exosomes: proteomic insights and diagnostic potential. Expert Rev Proteomics. 6:267–283 71. Raimondo F, Corbetta S, Morosi L et al (2013) Urinary exosomes and diabetic nephropathy: a proteomic approach. Mol BioSyst 9(6):1139–1146 72. Papale M, Rocchetti MT, Gesualdo L (2013) Clinical proteomics: the potentiality of urine analysis for understanding diabetic nephropathy. Euro Med J Nephrol (1):32–39 73. Kohda Y et al (2000) Analysis of segmental renal gene expression by laser capture microdissection. Kidney Int 57:321–331 74. Chaurand P et al (2006) Molecular imaging of thin mammalian tissue sections by mass spectrometry. Curr Opin Biotechnol 17(4):431–436
