Urinary Proteomics—a Tool for Biomarker Discovery

Abstract

The strong need for the discovery of novel disease markers together with the development of high-throughput techniques that provide highly sensitive analysis of protein content in tissues and bodily fluids, using proteomics, has opened the completely new chapter in biomarker discovery. The detection of biomarkers based on urinary proteome analysis is rapidly advancing and may provide new tools to improve non-invasive diagnostics, prognostics, and therapy enhancement. As a tool for biomarker discovery, urinary proteomics is especially fruitful in the area of early diagnostics and differentiation of renal damage, and it possesses enormous potential for improving and expanding non-invasive cancer diagnostics. An abundance of urinary proteins could provide a wide variety of biomarkers for the diagnosis and follow-up of many systemic diseases as well. This article reviews the utility of urinary proteomics for biomarker discovery from the perspective of clinical application. Despite huge potential and prompt development of urinary proteomics, many challenges are still in front of us. Research effort and financial investment have to be oriented on providing strategies for exceeding current methodological and technical obstacles in a way to ensure the successful validation and implementation of newly discovered urinary biomarkers. The result is expected to be the development of new non-invasive tests and procedures able to guarantee higher efficiency of patient care and provide needed personalized medical approach.

Keywords:

• urine
• proteomics
• biomarkers
• kidney disease
• diagnostics

INTRODUCTION

From the constant challenges that modern medicine is facing emerged the desperate need for the identification of novel disease-specific biomarkers, which may find application in preventive screening and early diagnosis, and improve prognostic and therapeutic possibilities as well. Scientific excitement recently induced by the Human Genome Project has been swiftly replaced by the enthusiasm for proteomics, with the aim to find new biomarkers for human diseases. The new era in biomarker discovery has started together with the revolutionary technological achievements for detection, quantification, and identification of proteins. Increased knowledge about molecular mechanisms hand-in-hand with the application of proteomic techniques for high-throughput analysis of proteomes is providing identification of new specific protein biomarker panels for different diseases.^[1]

Proteomics represents systematic study of proteomes, which refers to the whole protein content of cell, tissue, or an organism, as well as of different bodily fluids. Under the influence of different factors, variations in gene expression, alternative splicing, and posttranslational modifications generate constant changes of proteome profiles, which depict the complexity of the specific proteome. Biomedical proteome research with the aim of biomarker discovery is mainly based on expression proteomics, analyzing quantity of certain proteins in different conditions. Functional proteomics interferes in interactions among proteins in order to discover molecular functions and signaling pathways of identified biomarkers, explain pathogenesis, and discover targets for novel therapeutic interventions.^[2]

POTENTIAL OF URINARY PROTEOMICS IN BIOMARKER DISCOVERY

Proteinuria, a cardinal symptom of renal disease, has always been considered a nonspecific source for diagnostic and prognostic information. Currently, new proteomic technologies provide greater capabilities in urinary protein investigation; thus, it is clear that protein content of urine could actually be a quite accurate and specific diagnostic and prognostic tool for different diseases. Recent proteomic analysis has identified more than 1500 proteins in urine of healthy individuals,^[3] and their quantitative and qualitative changes might have a significant biomarker role.

Urine represents a modified ultrafiltrate of plasma, with protein concentration approximately 1000-fold lower than plasma.^[4] However, urinary proteins/peptides have higher thermodynamic stability and are less complex and interactive compared with the plasma proteins,^[5] so that low protein concentration does not make it a less promising diagnostic specimen than plasma. Furthermore, noninvasive accessibility of urine and simplicity of sampling makes it very convenient diagnostic source. As a reflection of serum composition and kidney function, urine is a source of most plasma proteins, with increased proportions of low-molecular-weight protein and peptide components, and it is enriched with proteins released along the urinary tract.^[4] Changes of urinary proteome are reflecting disease-related changes of tissue or cellular proteomes. This gives urine enormous potential for tasks such as early detection of disease, classification of disease, choice of therapeutic agents, assessment of prognosis, and monitoring of a particular therapeutic regimen.^[6] In that way, urinary proteomics is especially fruitful in the area of early diagnostics and differentiation of renal damage, and it possesses enormous potential for improving non-invasive cancer diagnostics.^[7] Furthermore, systemic diseases associated with the generation of small circulating proteins and peptides as proteolytic fragments, which pass glomerular filter, could also manifest themselves by changes in urinary proteome. Thus, urine could be the source for biomarkers of different systemic diseases as well.

Soluble urinary proteins originate mainly from plasma by glomerular filtration, or they could be proteolytically cleaved membrane bound proteins, secreted by the thick ascending limb of Henle loop (Tamm-Horsfall protein or uromodulin). Solid phase elements consist of slough epithelial cells and casts, small membrane fragments from shedding of microvilli or by apoptosis, and urinary exosomes.^[6,8] Because of the different origins of this proteins, each of this protein fraction and sub-fraction contains potential biomarkers for relevant disorders. Urinary prefractionation by different speed centrifugation or ultrafiltration is an important step prior to proteome investigation, and could be the method for enriching markers for particular types of diseases.^[6]

TECHNIQUES USED IN URINARY PROTEOMICS

There are several different techniques for proteomic studies. Most of the approaches start with a separation step, followed by ionization and subsequent mass spectrometry (MS) analysis. Ionization of the compound of interest can be achieved using matrix-assisted laser desorption ionization (MALDI) or electro-spray ionization (ESI). Now, different MS analyzers are in use, and individual advantages of the different mass detectors are frequently combined. Tandem mass spectrometry (MS/MS) is used to obtain protein sequence information.^[9]

For pre-MS fractionation, different techniques provide very different types and quality of data. Two-dimensional gel electrophoresis (2-DE) is most commonly applied method, which allows separation and characterization of proteins according to their isoelectric point and molecular mass. The application of 2-DE to urine is a labor- and time-consuming low-throughput approach that requires a relatively large amount of proteins.^[10,11] Liquid chromatography coupled to mass spectrometry (LC-MS) offers sensitive urine proteome analysis. LC/MS seems to be suitable for urine analysis due to the small amount of protein required. CE-MS couples the high-resolution properties of capillary electrophoresis (CE) with the powerful identification ability of the electro-spray time-of-flight MS to profile and sequence urinary proteins.

Array technology, using specific antibodies immobilized onto specially treated surfaces in an array format, could provide rapid examination of urine samples.^[7] The surface-enhanced laser desorption/ionization (SELDI) technique effectively couples high-end MS with array formats. This method has the capacity to analyze multiple samples in a short time, and it is widely used in urinary proteomic analysis.

All of these high-throughput techniques have limitations in detecting proteins, as well as in protein identification and quantification, but they can be used as screening tools for biomarker discovery.

STRATEGY FOR APPLICATION OF URINARY BIOMARKERS

Biomarker discovery based on proteomic approaches is comparative analysis and requires quantification of proteins. High-throughput proteomic techniques facilitate whole-expression proteomics, and make possible to determine protein profiles between normal and disease states. However, targeted analysis, oriented on predefined subset of proteins, is useful whenever is possible.^[12] Once a presumptive biomarker or a set of biomarkers has been identified, its predictive capabilities have to be determined through validation studies on large patient cohorts in order to define sensitivity and specificity. The final step is the implementation of specific biomarker in clinical settings by development of a clinical assay incorporated on a single antibody array chip to allow personalized medicine to be carried out at costs that will allow everyone to benefit.^[6] The future will probably bring interesting opportunities for introducing proteomics in routine laboratory for urine diagnostics.

URINARY BIOMARKERS WITH POTENTIAL FOR DIAGNOSTIC APPLICATION

The discovery of new urinary biomarkers with the potential for early diagnostics is particularly useful for many kidney diseases that could lead to chronic renal failure, especially because current diagnostics is based on non-specific parameters and invasive procedures. Therefore, research interest has been aiming to discover specific urinary proteome patterns for many nephrourological diseases and cancers.

Glomerular Disease

The most common cause of glomerulonephritis worldwide is IgA nephropathy (IgAN). Although this immune complex-mediated disease has the potential for development of chronic renal failure, diagnosis still relies on kidney biopsy. Comparison of urinary polypeptide profile of IgAN with the profiles of normal controls and other glomerular diseases, using CE-MS, has brought a specific IgAN urinary polypeptide pattern. This biomarker panel also showed high sensitivity and specificity for prediction of IgAN.^[13] Additional research has led to establishment of 2-D urinary proteomic map for IgAN, with the total of 84 differentially expressed spots, representing 59 different proteins comparing to urine of healthy controls.^[14] These findings could be crucial for discovering a reliable non-invasive early diagnostic test for IgAN.

Membranous nephropathy, an idiopathic antibody-mediated autoimmune disease, is one of the most common causes of nephrotic syndrome in adults. In order to define biomarker candidates for human membranous nephropathy, serial analysis of urinary proteome profile of a rat model of passive Heymann nephritis, which resembles human membranous nephropathy, revealed a group of significantly differed proteins in the disease course.^[15] Early stage of the disease was characterized by markedly increased level of urine haptoglobin, which could be a biomarker for early membranous nephropathy.^[16] Similar serial urinary proteome profiling in murine models of induced focal segmental glomerulosclerosis (FSGS) identified a group of urinary proteins that showed characteristic patterns of dynamic changes along the disease course. Urinary proteins that are identified to appear before glomerular sclerotic changes could serve as early diagnostic biomarkers of FSGS.^[17]

Allograft Rejection and Dysfunction

Rejection constitutes the major impediment to the success of renal transplantation, and currently available methods often fail to detect rejection until late stages of progression. Discovery of biomarkers using urinary proteomic could improve early non-invasive diagnosis of allograft rejection and prognosis of renal transplants. Although substantial differences in concentration of several proteins have been found in the urine of patients who received a transplant comparing to healthy individuals, these differences are mostly due to immunosuppressant.^[18] Two different works succeeded to detect potential urinary biomarkers for allograft rejection in kidney-transplanted patients using SELDI-TOF MS. However, they pointed to completely different biomarkers for the same disorder.^[19,20] Other researchers discovered that in patients with rejection, there is a reduction in β-Defensin-1, while α-1-antichymotrypsin increased compared with clinically stable transplants.^[21] The differences in detected peaks for allograft rejection are probably a result of the complexity of urinary proteome and variation in the performance of applied techniques. Urinary polypeptide analysis in different types of chronic allograft dysfunction established a pattern for two different lesions associated with distinct graft outcomes (viz., pure interstitial fibrosis and tubular atrophy and chronic active antibody-mediated rejection), which constitutes a first step toward the design of a specific, noninvasive diagnostic tool for chronic allograft dysfunction.^[22]

Acute Kidney Injury

Acute kidney injury (AKI) is a common clinical problem associated with high morbidity and mortality. Diagnosis of AKI is based on unreliable and nonspecific indicators, such as serum creatinine, detection of casts, and fractional excretion of sodium. Nevertheless, early identification of AKI is an imperative, because it demands rapid and aggressive treatment. A number of candidate urinary proteins, mainly renal tubular proteins, have been evaluated as biomarkers of renal injury. It has been suggested that a “urine panel” consisting of interleukin-18, kidney injury molecule-1 (KIM), and neutrophil gelatinase-associated lipocalin may provide more reliable information.^[23] Recently, exosomal fetuin-A has been identified as a potential urinary biomarker of AKI.^[24]

Diagnostic Potential of Urinary Exosomes

Urinary exosomes are derived from all cell types that face the urinary space.^[25] Exosomes are small vesicles that originate as internal vesicles of multivesicular bodies and are excreted from the cell by fusion of the multivesicular body and plasma membrane. They contain membrane proteins and cytosolic proteins as well. Many cell types have been shown to secrete exosomes.^[6]

The proteomics analysis of urinary exosomes from normal human subjects using LC-MS identified a number of proteins involved in exosome biogenesis and also protein products known to be responsible for renal and systemic diseases, such as autosomal dominant polycystic kidney disease, Gitelman syndrome, Bartter syndrome, autosomal recessive syndrome of osteopetrosis with renal tubular acidosis, familial renal hypomagnesemia, and hypertension.^[25] The isolation of exosomes can provide very large enrichment of urinary proteins that are derived from renal tubule epithelial cells, and they may be the best source of biomarkers for renal tubulopathies. In addition, they could be useful in cancer proteomics studies, because erythroleukemia and other tumor cells also secrete exosomes.^[6]

Urinary Biomarkers of Renal and Urological Cancers

Renal cell cancer (RCC) is often detected incidentally, and at the time of presentation, is frequently in advanced stage. Currently, there are no routinely used tumor markers for RCC. The examination of potential clinical utility of SELDI profiling of urine samples to diagnose clear cell RCC has proved diagnostic potential of defined protein pattern, with sensitivity and specificity of about 82%. However, the model could not maintain required sensitivity and specificity due to factors that affect methodology standardization.^[26] Kininogen levels were found to be elevated in the urine of RCC patients, and the concentration of this protein fell after nephrectomy.^[27] This link between RCC and kininogen levels could help in defining RCC biomarkers. In analyzing the expression of human kidney injury molecule 1 (hKIM) in patients with RCC, its presence in tissue samples and urine is shown. Urinary levels of hKIM were significantly higher in patients with RCC compared to patients with prostate cancer and normal subjects, with significant decreasing or disappearance after nephrectomy.^[28] This suggests hKIM as a new possible biomarker for RCC. A very interesting field in cancer biomarker discovery is tissue proteomics with definition of RCC proteomic map^[29] and analysis of proteins from conditioned media of RCC cell lines.^[30] However, the potential of identified proteins as biomarkers in clinical settings has to be further explored.

Urothelial cancer is the second most frequent type of cancer in urogenital tract, with the highest recurrence rate of any cancer. Bladder cancers are mostly transitional cell carcinomas (TCC); squamous carcinomas and adenocarcinomas rarely occur. With the use of SELDI-TOF, multiple protein changes in urine of patients with TCC compared to healthy individuals were found. Among five potential biomarkers and seven protein clusters that were found to be specific for TCC patients, one that was identified belongs to the defensin family, a group of endogenous antibiotic peptides that are locally secreted in response to inflammation. The combination of these protein biomarkers and clusters provided sensitivity of 87% and specificity of 66% for detecting TCC, with much higher sensitivity for detecting low-grade TCC comparing to voided urine or bladder-washing cytology.^[31] An additional set of polypeptides associated with bladder cancer is demonstrated using protein-chip technology, and had 80% sensitivity and 90–97% specificity in the training set, as well as about 55% and 60% in the test set, with the possibility to segregate different stages of TCC.^[32] However, urinary biomarker panel of TCC, that was validated in the prospective study, has been discovered using CE-MS. Discovered panel was further refined using additional urine samples from healthy volunteers and patients with malignant and non-malignant genitourinary disease. The final pattern consisted of 22 polypeptide masses, and in masked assessment, it showed 100% sensitivity and specificity for TCC. One of the identified protein was fibrinopeptide A, known as biomarker of ovarian and gastric cancer.^[33] Performing 2-DE with subsequent MS, other research pointed to orosomucoid and zinc-α2-glycoprotein as potential tumor markers of TCC. Furthermore, it was noticed that the abundance of these markers in urine was increasing according to the stage of the tumors, and that both of proteins identified might be related to the development of bladder cancer.^[34] 2-DE analysis of tissue proteome identified calreticulin, γ-synuclein, and a soluble isoform of catechol O-methyltransferase as other potential TCC biomarkers.^[35,36] These proteins were subsequently identified in urine samples immunochemically, and later analysis of urine specimens supported the use of urinary calreticulin as a biomarker for bladder cancer.^[35] Recent research proved that biomarkers for bladder cancer could be in glycosylated protein fraction of urine, by detecting α-1B-glycoprotein, specific for TCC.^[37] Squamous cell carcinoma is a much less common type of bladder cancer, and potential biomarker for this type of cancer is identified as the calcium-binding protein psoriasin.^[38]

Research interests have made urinary proteomics an especially fruitful source for biomarkers of cancers. Despite these efforts, most of these studies are based on small patient cohorts, and the reported sensitivity and specificity of urine biomarkers are wide ranging. Larger validation studies are underway to assess the use of detected biomarkers in clinical settings.

Diagnostic Biomarkers for Diseases of Other Systems

Biomarkers for graft-versus-host disease after bone marrow transplantation have been defined using CE-MS-based urine proteomics^[39] and validated in a prospective multicenter study. Recent investigations identified urinary biomarkers predictive of acute pancreatitis,^[40] obstructive sleep apnea,^[41] pancreatic ductal adenocarcinoma,^[42] and non-small-cell lung cancer.^[43] It is likely that urine testing will be used in the future to screen for a wide range of systemic disorders, the majority of which will have no or limited renal involvement.

URINARY BIOMARKERS FOR PROGNOSTIC ASSESSMENT

Diabetic Nephropathy

In patients with diabetes mellitus, diabetic nephropathy (DN) is the most important cause of end-stage renal disease and a major health problem. Diagnosis and assessment of DN is based on the presence of microalbuminuria, which is neither a sensitive nor specific early marker of the disease. By using CE-MS, it was found that urinary polypeptide pattern of patients with Type 1^[44] and Type 2^[45] diabetes significantly differ from the normal controls, and there was a specific polypeptide pattern of “diabetic renal damage” in patients with overt proteinuria. Evaluation of the urinary proteome profile of the patients with DN and overt proteinuria, applying 2-D DIGE, identified one spot that was increased in the diabetic urine as α1-antitrypsin. This finding was also confirmed by an independent group using ELISA.^[46] Recently, SELDI proteome profiling of urine samples from DN patients discovered the SELDI peak that was exclusively present in the urine of patients with DN and was subsequently identified as UbA52, an ubiquitin ribosomal fusion protein. Additionally, it was found that the processed form of ubiquitin was selectively missing in urine of these patients. These findings suggest that UbA52 could serve as a potential diagnostic biomarker of DN, and a lack of processed form of ubiquitin may have prognostic implications.^[47] Additional research established several proteins that were found to have progressively increased expression related to the degree of proteinuria, whereas the other group of proteins had progressively declined levels in association with the degree of proteinuria.^[48] More recently, SELDI-TOF MS was performed to compare the baseline urine samples of Pima Indian patients with Type 2 diabetes who developed DN 10 years later to those who remained normoalbuminuria after 10 years of the sample collection. This study identified a molecular signature that could differentiate patients who subsequently developed DN from those who remained normoalbuminuria and provided 71% sensitivity and 76% specificity for prediction of diabetic nephropathy in an independent validation set.^[49] Obtained data emphasize the potential of urinary proteomics to predict DN in a long-term period.

Lupus Nephritis

In systemic lupus erythematosus, active glomerulonephritis is the major cause of renal failure and an important cause of morbidity and mortality. Disease activity in lupus nephritis (LN) may require multiple biopsies for guiding a treatment. It is expected that new urinary protein biomarkers should be able to determine class of LN, activity, and progression of the disease, besides providing non-invasiveness. Serial analysis of the urinary SELDI profiles revealed dynamic changes in levels of low-molecular weight proteins during the course of the LN flare.^[50] Other research compared the SELDI proteome profiles of the urine samples from pediatric patients with SLE and the controls that had juvenile idiopathic arthritis. It was possible to define the molecular signature of lupus nephritis, and intensities of these protein peaks were greater in urine samples of patients with lupus nephritis, compared to those without nephritis and no lupus controls. Some of these proteins could be used for the prediction of the disease activity.^[51] Urinary proteomic research in patients with LN revealed two proteins that distinguished active from inactive LN, with 92% sensitivity and specificity each, and multiple regression scores could predict both relapse and remission even earlier than traditional clinical markers.^[52] Identification of these proteins will allow one to devise specific assays to routinely monitor disease progression and alter immunosuppressive drug regimens accordingly. These proteins may also play a critical role in the pathogenesis of LN and could therefore provide targets for therapeutic intervention.

Obstructive Nephropathy

Congenital unilateral ureteropelvic junction (UPJ) obstruction is a frequently encountered pathology in newborns. Urinary proteomics has been applied to discover specific biomarker panel for the prognosis of the disease, which would be helpful in the assessment of the need for surgical intervention. Urinary biomarkers that were able to distinguish different levels of UPJ obstruction have been defined applying CE-MS. These protein biomarkers could predict with 97% accuracy and several months in advance the clinical outcome of UPJ obstruction in newborns. Moreover, a newly identified marker, proSAAS (proprotein convertase subtilisin/kexin type 1 inhibitor), generated a new hypothesis in the pathogenesis of UPJ obstruction.^[53]

Diagnostic Biomarkers for Diseases of Other Systems

Coronary Artery Disease

Proteomics analysis of urine could yield a panel of biomarker peptides that would be useful as additional tools for the diagnosis and monitoring of coronary artery disease (CAD). Urinary proteome analysis using CE-MS defined a characteristic CAD signature panel with high sensitivity and specificity. This peptide pattern significantly changed toward the healthy signature correlating with the therapeutic physical activity level. Identified polypeptides that showed to be upregulated were collagen type I or III fragments. These collagens are increasingly produced in thickened intima of atherosclerotic lesions and fractionated by increased level of collagenases, which proved to be an independent predictor of rapid lumen diameter reduction and fatal cardiovascular events and probably play a crucial role in plaque rupture.^[54] Therefore, non-invasive urinary proteomics can become a valuable addition to other biomarkers used in cardiovascular risk assessment in personalized medical approach.

Preeclampsia

Preeclampsia has been implicated in 20% of pregnancy-related maternal deaths and is the leading cause of mandated preterm delivery. It was reported that women with preeclampsia requiring mandated delivery exhibit a urinary proteomic signature characterized by nonrandom fragments of SERPINA1 (abundant serum protease inhibitor) and albumin. Proteomic fingerprint in urine precedes the onset of clinical symptoms, predicts the severity of preeclampsia and need for delivery better than other common methods, and better differentiates preeclampsia from other uncontrolled hypertensive disorders. Aside from the practicality of producing a noninvasive diagnostic and prognostic test, identifying biomarkers of preeclampsia could lead to an explanation of pathogenesis of this condition.^[55]

Biomarkers for Therapeutic Response Assessment

Efficacy of the treatment is determined by therapeutic response. Moreover, knowing the drug that the patient is not responding to could provide early initiation of individualized treatment. The aim of the clinical proteomics is to provide personal medical approach, tailored individually for every patient.

In order to prevent the progression of the disease, standard treatment for patients with IgAN is the use of angiotensin-converting enzyme inhibitors (ACEI) or angiotensin II receptor blockers. However, some patients do not respond to these drugs. Therapeutic response to ACEI in IgAN was evaluated by comparative urinary protein analysis. Markers with significantly different urinary levels between responders and non-responders were kininogen, inter-α-trypsin inhibitor heavy chain 4, and transthyretin. It is confirmed that very low levels of urine kininogen could serve as a marker for the prediction of the poor response to the ACEI therapy.^[56]

Minimal change disease is the most frequent cause of nephrotic syndrome in children. The prognosis of pediatric nephrotic syndrome, as well as the need for renal biopsy, correlates with the responsiveness to glucocorticoid therapy. Thus, several studies analyzed urinary proteome with the purpose to identify urinary biomarkers of steroid resistance. SELDI urinary proteome profiling in pediatric patients with steroid-sensitive (SSNS) and steroid-resistant nephrotic syndrome (SRNS) could reliably differentiate control from the nephrotic urine. What is more, it could distinguish SSNS from SRNS with 100% sensitivity and specificity.^[57] In another report, a similar approach was used, but disease stage or activity was also considered. Analysis revealed a group of SELDI peaks that could distinguish SRNS from the other groups, and one peak was subsequently identified as β2-microglobulin, thus proving to be a biomarker associated with steroid resistance.^[58]

POTENTIAL OF URINARY BIOMARKERS FOR SCREENING

Early diagnosis and treatment are essential for better clinical outcome in malignant diseases. Several urinary biomarkers that show high sensitivity and specificity may provide the basis for reliable screening tests in the near future. Prostate cancer (PC) ranks third among cancers in men, and there is an objective need for a more sensitive tumor marker than PSA. There is a report about the identification and validation of a panel of 12 novel biomarkers for prostate cancer that was discovered in first void urine using CE-MS. This urinary biomarker panel in combination with PSA could provide a more predictive diagnostic tool for prostate cancer screening.^[59]

Among gynecologic malignancies, ovarian cancer is associated with the highest death rate, largely because of its tendency to present at an advanced stage associated with poorer survival. Urinary proteomic analysis in ovarian cancer patients resulted in identification of glycosylated form of eosinophil-derived neurotoxin (EDN) and a cluster of COOH-terminal osteopontin that were elevated. Their combination resulted in 93% specificity and 72% sensitivity for early-stage ovarian cancer. These findings could be used for the development of potential noninvasive screening tests for early diagnosis of ovarian cancer.^[60]

SPECIFIC PROBLEMS AND LIMITATIONS OF URINARY PROTEOMICS

Regardless of the impressive potential of urinary proteomics, specific methodological problems impede the translation of new biomarkers to clinical settings. The shortcomings of current proteomic studies, such as modest number of subjects, absence of disease controls, small number of defined biomarkers, and diversity of analytical platforms, are making it impossible to merge all biomarkers into human urinary proteome database.^[61] Moreover, there is a lack of standardization in many proteomic techniques for future clinical application.

In order to obtain consistent and reliable data that could be comparable, protocols for urine sampling and handling, storage, shipment, enrichment, and quantification have to be standardized.^[62] As it is not possible to obviate significant changes in the urinary proteome due to age, gender, and diet, these factors should be matched between cases and controls in the study design. To achieve greater specificity, controls should include not only healthy subjects, but also patients with similar diseases, as it is possible that similar pathological processes are represented with undistinguished urinary proteomic profiles.^[63]

The application of a new biomarker depends on its predictive capabilities to determine specific clinical state. In many proteomic studies, the validation process of new biomarkers is lacking, or it has been conducted on small number of patients and did not show satisfying predictive power. For the precise determination of applicability of new biomarkers, there is an objective need for prospective validation studies on large patient cohorts, which will involve screening test in a relevant population to determine specificity and sensitivity of discovered urinary biomarkers.^[64]

The Human Kidney and Urine Proteome Project (HKUPP) has been designed with the goal of finding ways for analyzing urine specimens and standardizing methods and data so that large-scale studies can be properly compared. Further enhancement of networking activities has been initiated as EuroKUP (European Kidney and Urine Proteomics) by bringing together scientists and clinicians working in this field to promote expertise and collaborations.^[65] With united efforts of science and industry, it will be possible to surpass existing limitations, and we could expect the clinical application of some urinary biomarkers in the near future.

CONCLUSIONS

Over the past few years, proteomics has proved itself as an extremely promising method regarding protein analysis. Current research effort and financial investment in this field undoubtedly confirm the great expectations of answers that urinary proteome analysis might offer. However, the discovery of biomarkers in urine is still a huge challenge, even with the analytical tools that have been developed and steps made toward method standardization. Only a few of the potential candidates that are identified in these studies will turn out to be clinically useful biomarkers. Nevertheless, it is crucial for discovered biomarkers to find their way into proper clinical trials so that their efficacy as disease specific markers can be validated. If the necessary effort is not directed toward putting such discoveries into clinical practice, then proteomics may languish as a technology that creates large databases of minimal use.

Contemporary proteomics studies of urine are expected to provide the possibility of the development of new non-invasive tests and procedures, along with the potential advantage of lower costs and higher efficiency of patient care. Future clinical application of novel urinary biomarkers is expected to improve early detection and obviate the need for invasive diagnostics, particularly in kidney diseases and cancers. Moreover, urine biomarker testing could provide tools for prognostic assessment and facilitate monitoring the course of the disease and response to therapy, with the possibility to tailor the therapy accordingly. Finally, some of the discovered biomarkers could be the targets for novel therapeutic interventions. Adjustment of proteomic analytical methods to current clinical settings will enable the application of proteomics at the bedside of the patient, which would direct medical practice toward personalized medicine.

ACKNOWLEDGMENTS

This work was supported by grant number 145004, from the Ministry of Science and Technological Development of Serbia.