Discovery by a proteomic approach of possible early biomarkers of drug-induced nephrotoxicity in medication-overuse headache

Medication-overuse headache (MOH) is a secondary cause of chronic daily headache, where headaches occur 15 or more days per month when the therapeutic agent is used excessively and on a regular basis for 3 or more months, and when headaches have developed or markedly worsened during the period of medication overuse [1]. Approximately 40% of all patients attending Headache Centers presents a chronic form of headache and 80% of these subjects excessively use symptomatic drugs which include analgesics, migraine-specific medications (such as triptans), opioids, or drugs combinations [2]. Although MOH has a prevalence of 1-2% in the general population, it represents a relevant health problem associated with significant long-term morbidity and disability [3]. MOH manifests as increased frequency and intensity of migraine attacks and as enhanced sensitivity to stimuli that elicit migraine episodes [4]. Moreover, there are evidences to support a role for genetic factors in the development of MOH [5]. It is now recognised that MOH has characteristic clinical features, a clear biological basis and causes considerable morbidity [6, 7]. A recent study concerning the epidemiology of drug-induced disorders has demonstrated that medication overuse could lead to nephrotoxicity and potential renal damage [8]. Particularly, drug-associated nephrotoxicity accounts for 18-27% of all acute kidney injury cases in US and medications can affect all aspects and every part of the kidney’s structure by different mechanisms of renal dysfunctions. The objective of this work was to extend and confirm the results from a previous study [9], in which we analyzed the urinary proteome of MOH patients overusing non-steroidal anti-inflammatory drugs (NSAIDs), triptans and mixtures, in comparison with healthy non-abusers individuals as controls, with the aim to identify significantly differences in excreted proteins induced by the excessive consumption of drugs and potentially related to nephrotoxicity. In the present study, other than to increase the number of patients and controls, we adopted further specialized proteomic techniques, namely two-dimensional gel electrophoresis (2-DE) coupled with mass spectrometry (MS), and the innovative and sensitive Surface-Enhanced Laser Desorption/Ionization Time-of-Flight mass spectrometry (SELDI-TOF-MS), with the main purpose to discover new proteomic profiles associated with MOH condition.

This work extends and completes a previous proteomic study concerning the overuse of antimigraine drugs in MOH patients and the possible risk to develop nephrotoxicity [9]. Urinary proteomics is among the most rapidly growing subdisciplines of proteomics applied to biomedical research and represent one of the key emerging technologies to discover new biomarkers for renal disease [15, 16]. In the present study we applied powerful proteomic methods for protein profiling and biomarker discovery, i.e. 2-DE together with MS and SELDI-TOF-MS. Urine analysis by 2-DE confirmed 6 proteins previously identified as differentially expressed and, in addition, revealed 15 new differential proteins in MOH patients respect to control subjects (Figure  1B, bold), in each range of MW (Figure  2 and Table  2). In particular, at high-MW we observed the over-expression of ALBU in NSAIDs and mixtures patients. This finding, even if it could be reasonably expected, has never been shown in migraineurs patients, which suffer from a neurological disorder in the absence of any tissue injury; particularly, these patients do not manifest any other disease or physical dysfunction that could be related to the observed changes in protein expression, but purely they abuse drugs from several years. From this point of view, the medication overuse can be considered the pivotal cause of the altered protein excretion. Traditionally, the occurrence of albuminuria has been held as a marker of altered glomerular permeability, and the use of ALBU excretion has been well established as a diagnostic and prognostic marker to evaluate the severity degree of glomerular damage in the progression of chronic kidney disease [17]. Moreover, in urine of MOH abusers we found also several ALBU fragments. This could be explained because during glomerular filtration (particularly in the course of acute renal injury) ALBU undergoes fragmentation through the activation of intrarenal proteases, producing modified forms and smaller peptides (<10 kDa) that are excreted in urine [18]. Moreover, in support of this hypothesis, we found also increased levels of some proteases, such as A1AT, which proteolytic activity has been observed in tubulo-interstitial damage [19], other than SAP and SPB3, that may act as hydrolases. Therefore, we might suppose the occurrence of a “proteolytic attack” against ALBU in urine of MOH patients, causing its extensive degradation. At medium-MW we found 2 proteins significantly increased in all MOH abusers, namely ACTB and ANXA1, and the over-expression of APOH only in NSAIDs group. ACTB is a component of the cytoskeleton and is a mediator of the internal cell motility, while ANXA1 is implicated in several processes, including apoptosis and specialized renal functions [20]. About APOH, despite its relatively high-MW, it is excreted in elevated quantity by patients with primary renal tubular disorders [21]. In general, high-MW proteins (such as ALBU) are used as markers of glomerular filtration and low-MW proteins (traditionally B2MG) are markers of tubular damage; both glomerular and tubular damages are followed by renal dysfunctions [22]. The increase in urinary excretion of APOH and low-MW proteins in subjects with plasma creatinine concentrations within the reference range and normal urinalysis, as MOH patients enrolled in our study, may be explained by an overflow mechanism. One possibility is that there may be competition from ALBU and other proteins for reabsorption by the proximal tubule cells. An alternative probable explanation is that it results from diminution in the ability of the proximal tubule cells to bind and incorporate these proteins into endocytic vesicles or to hydrolyze them within lysosome. In our study we detected significantly increased excretion of B2MG in both NSAIDs and triptans groups. This protein (as well as the majority of the low-MW ones), is readily and freely filtered through the glomerulus and normally almost completely reabsorbed and destroyed by proximal tubular cells, so that only 0.3% of the filtered protein is found daily in the urine. Impairment of B2MG tubular uptake results in a raised urinary excretion, so B2MG level in urine has been taken as marker of proximal tubular dysfunctions [23]. We revealed increased excretion of ALBU and B2MG in MOH abusers either by 2-DE analysis that with SELDI-TOF-MS (Figure  3 and Figure  4, respectively). This latter is a versatile high-throughput technique able to provide a rapid protein expression profile from a variety of complex biological materials [24]. Analyzing MOH patients and controls urine samples by SELDI-TOF, considerable differences in protein profiles emerged within groups, both in terms of quality (Figure  3 and Figure  4), and quantity (Table  3). In fact, as illustrated in Table  3, the total number of protein peaks detected was significantly higher in NSAIDs and mixtures patients. These results are in accordance with those obtained by the 2-DE spots count, showing a strict overlap of percentage increase values. Interestingly, total urinary proteins (together with B2MG and CYTC, also identified in our studies), have been described as nephrotoxicity markers to detect drug-induced kidney damage [25]. Another low-MW protein over-excreted in all MOH patients groups was the PTGDS, also known as prostaglandin-D₂ synthase or β-trace protein. Prostaglandins play a major role in renal functions and development. Elevated levels of serum PTGDS, then physiologically eliminated via the kidneys, have been found in patients with impaired renal function [26], while urinary PTGDS excretion was found significantly increased in patients who were given long-term administration of gentamicin [27]. Other low-MW proteins over-expressed were: PGBM (fragment), a proteoglycan present in the glomerular basement membrane that provides a charge-selective barrier for glomerular filtration [28]; NTF2, a transport protein; S10A8, a calcium-binding regulatory protein (like S10AB), that plays a role as pro-inflammatory mediator in acute and chronic inflammation, other than to be differentially expressed in various tumor types. The hormone-binding protein TTHY (increased only in NSAIDs abusers), is mainly synthesized in the liver and its defects are the causes of amyloidosis. Insights into disease revealed that amyloid renal deposits commonly occur, together with kidney damage and proteinuria [29]. Finally, we found an over-expression of FABP5. This is an intracellular fatty acid carrier protein predominantly expressed in human proximal tubules. Several clinical studies demonstrated that in kidney disease the expression of FABP was increased and urinary excretion was accelerated, reflecting the disease progression [30]. For these reasons FABP has been proposed as an excellent biomarker for predicting and monitoring deterioration of renal function and for early detection of kidney injury [31]. The mechanisms of nephrotoxicity can vary between agents, including indirect or direct effects, that may be localized to a particular anatomical site. For this reason, there is much interest in the discovery of new biomarkers, that might provide a sensitive and rapid tool for the diagnosis of possible acute kidney injury after drug exposure, and thus allow earlier intervention with remedial therapy. In recent years, a broad range of serum and urinary enzymes and proteins have been proposed as possible early biomarkers of drug-induced nephrotoxicity [32], also through the use of proteomic methods [33]. In the present study, utilizing a proteomic approach, we evidenced an over-expression of a definite pattern of urinary proteins largely associated with various renal dysfunctions, particularly in NSAIDs abusers; this finding supports our previous results. Moreover, NSAIDs and mixtures abusers showed a significantly elevated number of total proteins, with a mean increase of around 50% respect to controls.

To summarize, regarding the possibility of drug-induced nephrotoxicity, we confirmed no particular risk only associated with the use of triptans. On the other hand, it is important to emphasize that, so far, none of the MOH patients participating to this study has shown any clinical sign or symptoms of kidney impairment; furthermore, their serum creatinine levels are in the normal range. It is specifically for these reasons that the over-expressed proteins that we found in urine of these abusers might represent potential early candidate biomarkers of kidney injury. Based on the results obtained in this work, we point out that it is important to monitor the renal function of MOH patients, especially for NSAIDs and mixtures abusers, because they might be prone to gradually progress toward different renal syndromes. In this way, an earlier detection of possible acute kidney dysfunctions using a distinctive panel of protein biomarkers, each informing on the integrated aspects of glomerular, tubular and interstitial function, might provide an opportunity to minimize the potential risk to develop severe or persistent renal damages also in presence of a normal clinical pattern.