Nuclear magnetic resonance and surface-assisted laser desorption/ionization mass spectrometry-based metabolome profiling of urine samples from kidney cancer patients

Highlights

• Fifty human urine samples and similar number of controls were analyzed.

• NMR and MS were used for urine metabolic profiling.

• New potential biomarkers of kidney cancer were found.

Abstract

Kidney cancer is one of the most frequently diagnosed cancers of the urinary tract in the world. Despite significant advances in kidney cancer treatment, no urine specific biomarker is currently used to guide therapeutic interventions. In an effort to address this knowledge gap, metabolic profiling of urine samples from 50 patients with kidney cancer and 50 healthy volunteers was undertaken using high-resolution proton nuclear magnetic resonance spectroscopy (¹H NMR) and silver-109 nanoparticle enhanced steel target laser desorption/ionization mass spectrometry (¹⁰⁹AgNPET LDI MS). Twelve potential urine biomarkers of kidney cancer were identified and quantified using one-dimensional (1D) ¹H NMR metabolomics. Seven mass spectral features which differed significantly in abundance (p < 0.05) between kidney cancer patients and healthy volunteers were also detected using ¹⁰⁹AgNPET-based laser desorption/ionization mass spectrometry (LDI MS). This work provides a framework to expand biomarker discovery that could be used as useful diagnostic or prognostic of kidney cancer progression

Introduction

Kidney cancer is among the 15 most commonly occurring cancers worldwide in terms of incidence and mortality in both men and women. More than 400,000 new kidney cancer cases and nearly 180,000 deaths were recorded in 2018. Kidney cancer is not a single disease, as there exists a number of different types of kidney tumors which differ in histology, responses to therapy, and progression to different clinical outcomes. Kidney tumors can be benign, indolent, or malignant. Non-cancerous tumors include adenoma, oncocytoma and angiomyolipoma (AML). Renal cell carcinoma (RCC) is the most common and malignant type of kidney cancer accounting for approximately 90 % of all neoplasms arising from the kidney. There are three main types of RCC including clear cell (ccRCC), papillary RCC (pRCC) and chromophobe RCC (cRCC) that differ in their stage, grade, and cancer-specific survival. Subtypes of RCC such as angiomyolipoma (AML), collecting duct carcinoma (CDC), or simple renal cyst (SRC) are very rare [1].

Currently, kidney cancer diagnosis is based on abdominal ultrasound, magnetic resonance imaging, or computed tomography; however, more than 60 % of RCCs are diagnosed incidentally when patients are examined for other reasons. Kidney cancer is one of the few cancers whose occurrence is increasing every year. In most cases, RCC progresses asymptomatically, and is difficult to detect at an early stage due to the lack of characteristic symptoms such as classic triad of visible haematuria, flank pain and palpable abdominal mass symptoms [2].

Unfortunately, though great efforts have been dedicated in the past decades to identify characteristic small molecule indicators of kidney cancer, there are currently no reliable biomarkers available to guide more effective therapies, diagnosis, or disease prognosis. Therefore, further research and the development of new kidney cancer sensitive biomarkers are of great importance, not only to improve prognosis, early detection as well as to monitor treatment, but also to enhance our understanding of the molecular processes underlying kidney cancer, using preferably non-invasive methods [3].

Over the past decades, the use of metabolomics applications to cancer research has increased significantly. Analysis of metabolite profiles from non-invasive sources such as biofluids is a promising approach for the discovery of valuable biomarkers to enhance our abilities to predict cancer progression, screen cancer pathologies, and to assess the effectiveness of cancer treatments. Urine is a preferred source of biospecimens for metabolomics analysis of kidney cancer due to its close association with disease origin. Urine metabolomes provide biochemical fingerprints of systemic changes in organisms, and an avenue to identify and characterize potential biomarkers associated with cancer, including kidney cancer [4]. The most frequently used techniques for metabolomic analysis of kidney cancer have been liquid chromatography-coupled mass spectrometry (LC–MS) [5], gas chromatography-coupled mass spectrometry (GC–MS) [6], and ¹H nuclear magnetic resonance (NMR) spectroscopy [7]. Since kidney cancer is recognized as a metabolomic disease, a growing number of studies focusing on the metabolite profiling of tissues [8] and biofluids from patients, including plasma [9], serum [10], and urine [11] have been reported.

MS-based methods have been some of the most prominent approaches utilized in untargeted metabolomic analyses of urine samples. In 2006, Perround et al. employed metabolomics techniques to characterize the urine metabolome of a small group of patients (5 RCC and 5 control), and to identify potential biomarkers of kidney cancer [12]. A similarly small group of RCC patients was examined a year later by the Weiss group, who employed three different analytical techniques to broaden the detection and coverage of urinary metabolites [13]. In 2011, Kim et al. utilized MS-based metabolomics to evaluate the differential levels of compounds present in the urine of 29 kidney cancer patients and 33 control patients. They found that measuring the differential concentrations of the three metabolites gentisate, quinolinate, and 4-hydroxybenzoate could be used successfully to distinguish RCC patients from controls [14]. In 2012, Ganti et al. analyzed the urine metabolome of 29 RCC and 33 control patients using LCMS and GCMS, and found that most acylcarnitines were increased in the urine of cancer patients, and that the concentrations of these compounds were dependent on both cancer status and kidney cancer grade [11]. In 2016, Monteiro et al. performed a metabolomic profiling analysis of urine from 42 RCC patients and 49 controls using NMR spectroscopy, and found 32 metabolites whose significantly altered levels between the two groups [15].

NMR-based metabolomic studies of urine samples from RCC patients are rarely reported in literature [16,17]. To our knowledge, there are no current studies that have combined both NMR and LDI MS approaches to conduct comprehensive analyses of the urine metabolome of patients with kidney cancer.

This work is the first to report on the metabolomics-based profiling of urine samples from patients with kidney cancer (n = 50) and controls (n = 50), using two orthogonal analytical methods: high resolution ¹H NMR and laser desorption/ionization mass spectrometry with 109-silver nanoparticle-enhanced steel target (¹⁰⁹AgNPET LDI MS) [18]. Results from this study have identified interesting small molecule candidate biomarkers, which may be useful to discriminate kidney cancer patients from healthy controls based on differential urine metabolome profiles.