Ochratoxin A: 50 Years of Research

Since ochratoxin A (OTA) was discovered, it has been ubiquitous as a natural contaminant of moldy food and feed. The multiple toxic effects of OTA are a real threat for human beings and animal health. For example, OTA can cause porcine nephropathy but can also damage poultries. Humans exposed to OTA can develop (notably by inhalation in the development of acute renal failure within 24 h) a range of chronic disorders such as upper urothelial carcinoma. OTA plays the main role in the pathogenesis of some renal diseases including Balkan endemic nephropathy, kidney tumors occurring in certain endemic regions of the Balkan Peninsula, and chronic interstitial nephropathy occurring in Northern African countries and likely in other parts of the world. OTA leads to DNA adduct formation, which is known for its genotoxicity and carcinogenicity. The present article discusses how renal carcinogenicity and nephrotoxicity cause both oxidative stress and direct genotoxicity. Careful analyses of the data show that OTA carcinogenic effects are due to combined direct and indirect mechanisms (e.g., genotoxicity, oxidative stress, epigenetic factors). Altogether this provides strong evidence that OTA carcinogenicity can also occur in humans.


Introduction
Ochratoxin A (OTA) is one of the most important and deleterious mycotoxins [1,2]. OTA was isolated and chemically characterized in 1965 [3,4]. OTA was discovered in South Africa as a toxic metabolite of Aspergillus ochraceus in a corn meal that was intentionally inoculated with this microfungus [3]. Further research has shown that OTA is nephrotoxic, hepatotoxic, embryotoxic, teratogenic, neurotoxic, immunotoxic, genotoxic, and carcinogenic in many species with species and sex-related differences [5][6][7][8][9][10]. The International Agency for Research on Cancer classified OTA as a possible human carcinogen (group 2B) in 1993 based on a great amount of evidence of its carcinogenity discovered in several animal studies [11]. The susceptibility to cancer is species-and sex-dependent [8,9,[12][13][14][15]. Frequent exposure of animals or humans to OTA may cause a range of health problems. In particular, OTA could be a threat of cancer for humans. It will be shown further in this article that OTA acts as a nephrotoxin and an urothelial carcinogen as a result of both the oxidative stress and direct genotoxic mechanisms. Strikingly, chronic exposure to low OTA doses could be even more damaging than acute exposure to a high dose [16,17]. Humans are normally exposed to OTA-as they are to other mycotoxins-through several routes, dietary intake being the most prominent. Dermal contact or inhalation exposures are of a minor importance with respect to the general population [18], although, occasionally, these routes may also play a role [19,20].
During coffee roasting (at 225˝C), 2 1 -DC-OTA and 2 1 R-OTA, two products of thermal degradation of OTA, were identified [66]. Ochratoxin α amide, which is formed at high temperatures during coffee roasting, was discovered. This represents another product of thermal degradation of OTA [67].

OTA Analysis
Principal methods developed for OTA determination in biological materials are summarized in Table 4.
In fact, more sensitive analytical methods or new methods for determining OTA and ochratoxins in biological materials are being developed consecutively toward the sophisticated development of instrumentation and analytical techniques but also toward the improvement of laboratory analytical methods. The most used and traditional analytical techniques include thin-layer chromatography, HPLC, and ELISA. Therefore, in the present article, the analytical techniques are divided into traditional ones, and the others. Generally, all chemical methods for the analysis of OTA consist of several steps (extraction, clean-up, separation, detection, quantification, and confirmation of identity) [100]. Conventional sample extraction and clean-up are usually achieved by liquid extraction for OTA determination in kidneys of swine [101]. More recently, solid-phase extractions (SPE) notably for OTA determination in animal feed [102] and immunoaffinity columns (IAC) [103,104] (/homemade of IAC/; immunoaffinity cartridges commercially available) have become popular [105]. At present, different kinds of cartridges are commercially available for clean-up and pre-concentration, including IAC and molecular imprinted polymers (MIPs) cartridges, composed by anti-OTA antibodies and three-dimensional network specific for the target molecule. In this case, OTA passed through cartridges (e.g., Mycosep™ or Mycospin™) [106]. It is based on adsorption and the ion-exchange process [107]. The use of immunoaffinity chromatography in the clean-up step improves mycotoxin analysis and has a number of advantages: clean extracts, precision and accuracy, rapidity, and reduction of the use of dangerous solvents [82]. The main advantages of these columns are the specific binding of OTA onto the antibody and the near-complete removal of matrix interference [108]. Nevertheless, in the case of OTA, underestimation can be observed if extraction is done in an alkaline condition, because OTA is converted into open-ring OTA (OP-OA) and no longer recognized by antibodies [109][110][111][112].
The confirmation of OTA presence in biological materials is very important in order to guarantee quality of analytical results. Hult and Gatenbeck (1976) presented the OTA confirmation with carboxypeptidase A [70], as did Hunt et al. (1980) with boron trifuoride methanol [73] and Studer-Rohr et al. (1995) with diazomethane [113]. Quality assurance of analytical results (a laboratory accreditation, participation in proficiency testing, and the use of certified reference materials) according to the past norm EN 45001 (1989) [114] and the recent norm which is in force EN ISO/IEC 17025 (2005) [115] is very important for the purposes of OTA determination in biological materials.
Many analytical methods for the determination of OTA have been developed over time [100], and most of them involve the use of thin-layer chromatography (TLC) [68,69] and, predominantly, high-performance liquid chromatography (HPLC) with fluorescence detection (FLD) [72]. Subsequently, OTA is identified and detected by LC-MS [77], LC-MS/MS [83,84], aptamers [88,92,116], ELISA [76,117], and immunosensing methods [118]. However, the technique most commonly used is based on liquid chromatography (LC) coupled with a fluorometric detector for highly sensitive detection signal [106]. It is known that, due to natural OTA fluorescence, OTA is generally determined by chromatographic techniques [119,120].

Occurrence of OTA in Food and Feed
In 1969, Shotwell et al. [139] with colleagues from the U.S. Department of Agriculture (USDA) published the very first piece of information about the amount of OTA in a maize sample at levels from 110 to 150 ng/g. In 1970, Scott et al. [140] from Health Canada published data on OTA in moldy cereals, beans, and peanuts. OTA concentrations in wheat, oats, barley, and rye (62.0% positive samples) ranged from 30 to 27.000 ng/g [140,141]. The occurrence of OTA in pig kidney was first mentioned by Hald and Krogh in 1972 [142] and by Hunt et al. in 1979 [143]. Since that time, more than 90 kinds of foodstuffs of both plant and animal origins, including milk, have been found to contribute to the OTA dietary exposure [33].
As for foodstuffs of plant origin, OTA occurs in cereal products, olives, beans, beer, wine, coffee, cocoa products, raisins, figs, licorice, pulses, pumpkin seeds, and tea. In general, the average concentration of OTA is reported to range from 0.1 to 100 ng/g. OTA concentration in black pepper, cayenne pepper, caraway, cardamom, coriander, chili powder, curcuma, and dried red pepper ranges from 1 to 100 ng/g. Feedstuffs of plant origin-those made of wheat, oats, barley, rye, maize, rice, millet, sorghum, soybean, horse bean, peas, bean, broad bean, alfalfa, sunflower or pumpkin seeds, coconut, peanut cake, and hay/silage-also contain from 1 to 100 ng/g of OTA [144][145][146].
In foodstuffs of animal origin, e.g., in pork blood products, edible offal, pork meat, chicken meat and offal, and dry-cured ham, the levels of OTA range from 0.1 to 1 ng/g. The same amounts are measured in feedstuffs of animal origin, e.g., in pork kidney and liver, pork meat, chicken liver, and viscera, and in mechanically separated chicken used as ingredients in pet food for cats and dogs [144,145,147].
Tables 5-7 summarize the recent data related to OTA in foodstuffs obtained from the EU Rapid Alert System for Food and Feed (RASFF) [146]. The RASFF dealt with OTA in 175 cases in 2000-2015. Alert notifications are sent whenever a foodstuff presenting a serious health risk to humans is identified at the internal market and whenever the rapid action of the competent authorities is required. Border rejections concern food and feed consignments that have been tested and rejected at the external borders of the EU. Food that is only present in the notifying EU Member State is worth noting.

OTA Nephrotoxicity
OTA has been found to cause porcine and poultry nephropathy. OTA is implicated in the pathogenesis of some renal diseases including Balkan endemic nephropathy (BEN), kidney tumors occurring in certain endemic regions of the Balkan Peninsula [14,148], and chronic interstitial nephropathy (CIN) occurring in Tunisia [149,150] and other North African countries [151].
Kidney lesions have been observed on proximal tubules. The epithelial cells were damaged, for example, membrane integrity was lost, and the size and the density of the brush border were reduced. The chromatin was condensed, and the nuclear envelope disappeared. The histologic picture shows an enlargement of tubular membrane and an apparition of collagen fibers [152].
At the beginning, the BEN disease is characterized by a modification to epithelial cells without any change in the size of the organ. After chronic exposure, kidneys are reduced and interstitial fibrosis is the most important picture. At the end stage, impairment of renal function leads to enzymuria (e.g., gamma glutamyl transferase, alkaline phosphatase, lactate dehydrogenase) [153], polyuria accompanied with red tongue, thirst, and bitter taste [153]. Neither edema nor hypertension can be observed. Other symptoms such as headaches, lumbar pain, asthenia, and anemia (iron deficiency) were recorded. Several biochemical parameters changed including glycosuria, proteinuria (0.15-0.5 g/ 24 h), alkalinization of urine, elevated serum creatinine, and an increase in immunoglobulin M (IgM) and immunoglobulin E (IgE) [154,155].
Data on OTA nephrotoxicity are summarized in Table 8.

1972
Balkan endemic nephropathy (BEN) has been suggested to be the result of fungal poisoning. The role of OTA in mycotoxicosis-BEN in humans and porcine nephropathy. [156] 1972 In view of the similarities between BEN and OTA induced porcine nephropathy, it has been suggested that OTA may be involved in the etiology of BEN. [157] 1978 OTA is potentially nephrotoxic in all species tested with the exception of adult ruminants. [158] 1987 Findings of higher OTA levels in the serum of patients suffering from BEN, which is a subtype of tubulointerstitial nephritis, led to hypotheses about the association between the nephrotoxicity of OTA and the BEN and also the incidence of renal system tumors in the population of these Balkan regions. [159] 1991 Nephropathy is primarily related to the mobilization of intracellular calcium. [160] 1992 In terms of human pathologies, OTA is suspected to be the main etiological agent responsible for BEN and associated urinary tract tumors (UTT) in humans. [161] 1993 Experimental studies on the nephrotoxicity of OTA both in vitro and in vivo have shown that OTA disturbs the intracellular metabolic processes (with subsequent apoptosis of the renal cells), renal hemodynamics, and-significantly and perhaps preponderantly-the functions of the proximal tubules (even after subchronic exposition). OTA causes the decrease of glomerular filtration and tubular resorption and affects all parts of the nephron and kidneys in toto.

2002-2005
The kidney is the main target of OTA toxicity in all animal species tested. [14,172] 2002-2005 OTA has been also implicated in the etiology of BEN, a chronic degenerative kidney disease, in kidney tumors in humans in certain regions of the Balkan Peninsula, and in chronic interstitial nephropathy (CIN) in Tunisia and other North African countries. [14,148,150] 2005 Exposure to low OTA doses is responsible for nephrotoxicity; at nanomolar concentrations, OTA leads to specific changes of function and phenotype in renal cells. [173] 2007-2010 Very low OTA concentrations administered for a prolonged time (up to 14 days) influence the cellular fate (cellular hypertrophy) in human proximal tubule; furthermore, they act not only in the target organ, e.g., in the kidney, but also in as yet unsuspected cells, such as fibroblasts; the same damage will likely occur in chronic exposure. [174,175] 2013 Nephrotoxicity is a consequence of acute, sub-acute, and also chronic exposure to OTA. [9] 2014 OTA inhibits the nuclear factor, erythroid 2-like 2 (Nrf2) oxidative stress response pathway. Nrf2 overexpression confers a survival advantage and is often associated with cancer cell survival. [176] 2015 Dietary exposure to OTA represents a serious health issue including, e.g., human endemic nephropathies. [50]

OTA Carcinogenicity
Data on OTA carcinogenicity are summarized in Table 9. Table 9. OTA carcinogenicity and genotoxicity.

1978
OTA induces renal and hepatic tumors in mice. [177] 1984 OTA is carcinogenic for mice. [178] 1984 CIT increases OTA carcinogenicity. [179] 1987 OTA carcinogenicity to humans: OTA classified in Group 3 (not classifiable as to its carcinogenicity to humans). [180] 1989 Male rats are more susceptible to renal tumors than female rats (NTP study). [181] 1989 The genotoxicity of ochratoxin A is reviewed. [35,182] 1991 OTA-DNA adducts: For the first time, OTA-DNA adducts are found in the kidney, liver, and spleen of mice. [183] 1993 OTA is re-classified as a possibly carcinogenic to humans based on a great amount of evidence of carcinogenity in several animal studies of 2B in 1993. [11] 1993 OTA-DNA adducts: Other studies take place in mice and rat tissues after acute and subchronic exposure, and in urinary tract tumors (UTT) of Bulgarian subjects. [184][185][186] 1993-2009 OTA-DNA adducts are also detected in tissues of humans presumably exposed to OTA in several countries (Bulgaria, Serbia, Croatia, Germany, Belgium, France, Tunisia). [16,17,185,[187][188][189][190] 1998-2002 DNA adduction following chronic exposure (carcinogenic study) of rats to OTA first described; sex differences and dual mechanism-oxidative pathways and DNA adduction-are observed [12,13,191] 1998 OTA-DNA adducts are observed in mother and progeny of mice fed OTA nine months after birth male mice develop cancer. [192] 2000-2001 In vitro formation of dG-OTA adduct. [193,194]

2001-2002
Other studies with radiolabeled OTA were unable to detect any DNA binding of OTA, but explanation of this discrepancy is given in depth by Pfohl-Leszkowicz and Castegnaro in 2005 [195] [ 60,196] 2003 OTA-DNA adduct in pigs subchronically exposed to low doses of OTA. Relation with biotransformation. [197] 2002-2010 OTA may be involved in testicular cancer. [175,[198][199][200][201] 2003-2008 CIT increases genotoxicity of OTA and modifies the metabolism of rats exposed to low doses for three weeks. [202,203] 2004 Evidence for covalent DNA adduction by OTA following chronic exposure to OTA in rats (and subacute exposure in pigs). [190] 2004 Another research group, using the highly sensitive accelerator of the mass spectrometry technique, does not detect DNA adducts after the administration of 14 C-labeled OTA to rats. [204] 2004 In 2004, a review of the NTP experimental rat tumor data for OTA also places OTA in the category of "chemicals inducing renal tumors through direct interaction of the parent compound or metabolite with renal DNA" based on histopathological evidence.

2004-2010
The long-term OTA studies confirm the incidence of tumors in rats; in male rats, these tumors are related to OTA dose [205][206][207] 2004-2012 OTA is a direct genotoxic forming covalent DNA adducts in the kidney OTA can indeed react with DNA via a phenolic radical resulting in C8-deoxyguanosine adduct (synthetized and chemical identified by mass spectrum). [175,190,201,[207][208][209] 2006 Confirmation of OTA genotoxicity via measurement of comet in rat kidneys. [210] 2007 Chronic exposure to low OTA doses can be much more damaging than acute exposure to a high dose. [16] 2007 DNA diploidy in rat tumors is associated to genetic damage. [211] 2007 OTA induces an increase of mutation at two loci-hypoxantine-guanine phophoribosyl transferase (HPRT) and thymidine kinase (TK). [212] 2008 DNA adduct cannot be confirmed, but the explanation is given by Pfohl-Leszkowicz et al. (2009) [64] [213] 2008 Correlation between biotransformation of OTA and direct covalent binding on DNA. [214] 2009 It is found that the kidney DNA adduct pattern of BEN patients is similar to the kidney DNA adduct pattern of pigs living in the same farm and pigs co-exposed to OTA, fumonisins, and citrinin. [17] 2009 A different proposal of mechanism for OTA-mediated renal carcinogenesis and threshold model for its risk assessment. [215] 2009-2010 Identification by LC-MS/MS of these DNA adduct in rat tissues. [64,201] 2010 OTA is carcinogenic for poultry. [216] 2011 Induction of mutation only in medulla of rat kidney exposed to carcinogenic dose. [217] 2012 Relation structure activity studies clearly indicate that OTHQ (ochratoxin hydroxyquinone) is responsible of direct genotoxicity, whereas some others are cytotoxic. [65,209] 2012 OTA is activated to a species that is a directly genotoxic mutagen. OTHQ in presence of cysteine is also mutagenic. [218] A new approach to cancer represents miRNA. [219,220] 2013 The induction of miR-132 and miR-200c by OTA elevates reactive oxygen species (ROS) levels and profibrotic (profibrotic transforming growth factors β, TGFβ) expression. [221] 2014 OTA has the potential to initiate or support the development of fibrotic kidney diseases by involving post-transcriptional regulation mechanisms comprising miR-29b. OTA reduces the impact of miR-29b and thus enhances collagen protein expression. [222] 2014 A low dose of OTA induces micronuclei, and OTA delays the DNA repair kinetics. [223] 2014 OTA increases proliferating cell nuclear antigen after 13 weeks in kidney and kidney damages. Limited oxidative stress. [224] 2015 Dietary exposure to OTA represents a serious health issue, including urinary tract tumors in humans. [50] In 1976 and 1983, IARC first evaluated the carcinogenic risk that OTA poses toHuman. No report on cases of cancer or epidemiological studies were available at that time and, in the absence of adequate epidemiological data, no evaluation of the carcinogenicity of OTA with respect to Humans could have been made [225,226]. In 1987, the IARC reclassified OTA into Group 3 (not classifiable for its carcinogenicity to humans). Based on a great amount of evidence of OTA carcinogenicity revealed in new animal studies, it was again reclassified into Group 2B (possibly carcinogenic to humans) in 1993. At present, new information regarding genotoxicity of OTA (formation of OTA-DNA adducts), its role in oxidative stress, and the identification of epigenetic factors involved in OTA carcinogenesis-should they indeed provide strong evidence that OTA carcinogenicity is mediated by a mechanism that also occurs in humans-could lead to another reclassification of OTA. In the light of recently available data, it does not seem inappropriate to upgrade its carcinogenicity from Group 2B (possibly carcinogenic to humans) to at least Group 2A (probably carcinogenic to humans) [227] or, in our opinion, even to Group 1 (carcinogenic to humans).

OTA Biomarkers
Biomonitoring of OTA provides the best approach to assess the human exposure to OTA from any source and through any route [228]. The first studies reporting the presence of OTA in human blood were carried out in the Balkans in the 1970s [229]. The exposure of the human population to OTA and other ochratoxins represents a worldwide problem. Baldwin et al. (2011) reviewed biomarker researches for the most important mycotoxins and defined biomarkers [230]. Recently, a biomarker of exposure has been defined to be a biological measure which is correlated with the quantity of the xenobiotic ingested; resulting in the improved exposure classification in comparison with more traditional approaches [231]. OTA in milk (non-invasive sampling), OTA in blood serum (invasive sampling), OTA in urine (non-invasive sampling), and OTA in human kidneys (sampling post-mortem or after nephroctomia) are qualified as biomarker of exposure to OTA [232]. Soto et al. (2015) have recently used several biomarkers for evaluating the OTA exposure. The values of OTA detected in potential biomarkers of exposure for blood, breast milk, and urine ranged from 0.15 to 18.0, from 0.002 to 13.1, and from 0.013 to 0.2 ng/mL, respectively. The calculated EDI for OTA in plasma ranged from 0.15 to 26 ng/kg bw/day and has turned out to be higher than that obtained in urine (0.017 to 0.4 ng/kg bw/day). All these values have been correlated with the range of EDI for OTA calculated from food products: 0.0001-25.2 ng/kg bw/day [233].

OTA in Human Blood
In 1979, OTA determination in human whole blood and serum was developed [234]. In the past several decades, OTA has been detected in human blood samples on a worldwide scale. Scott (2005) has described OTA in blood serum as a uniquely useful biomarker of OTA exposure due to its high-affinity binding to serum albumin or to other small proteins, which should result in higher serum OTA levels and long persistence of OTA in blood serum [235]. OTA blood amounts will integrate exposure over longer periods [236]. The use of serum or plasma has been described as more suitable matrices in comparison to whole blood [105,237]. Generally, the determination of OTA in blood samples remains the basic method of how to monitor human exposure to OTA, which is ubiquitous in human blood serum/plasma and indicates continuous exposure to the toxin, originating mainly from food intake [235]. Table 10 describes some of the most notable findings of OTA in blood on a worldwide scale. Advantages arising from monitoring OTA in the blood of healthy persons consist mainly in relatively high OTA levels found compared with OTA determinations in urine [232]. OTA blood determination will integrate exposure over longer periods, while biomarker analysis in urine apparently better reflects day-to-day variations in the exposure of adults and infants [231,[236][237][238][239][240].   [302] Abbreviations: n+ (%): percentage of positive samples; *study included persons working at grain storage facilities; rural, urban (population).

OTA in Urine
Urine is a major excretion route for both OTA and OTα (5-chloro-8-hydroxy-3methyl-1-oxo-3,4-dihydroisochromene-7-carboxylic acid; formula, see Table 3) in humans [45]. OTA can be found in urine several days after OTA ingestion [8]. The elimination of OTA through human urine has been reported to be low (mean value between 20 and 80 ng/day) and independent of the dose ingested [237]. The OTA uptake has been described as dependent on the free OTA concentration, which is severely limited by the binding of OTA to serum albumin [8]. Thus, the relationship between OTA in urine and OTA intake remains a complex issue as in the case of OTA in blood.
The first study measuringing OTA in urine in Europe was conducted by Mac Donald et al. [270] in the UK. In this study, OTA was found in 46 urine samples (92%) collected over 24 h from 50 volunteers (healthy individuals from the UK). OTA concentrations ranged from <10 to 58 ng/L, and the mean value was about 21 ng/L. This study demonstrated a strong correlation between OTA concentrations in urine and its dietary intake. The second study in Europe was conducted in Bulgaria by   [303]. A total of 152 urine samples collected from patients with BEN (Balkan endemic nephropathy) or urinary tract tumors (UTT) and from the control families were analyzed, and OTA was detected in about 33% of the samples of urine (more often in endemic villages than in nonendemic ones) in the range 5-604 ng/L and in healthy people in the range 5-43 ng/L (LOQ = 5 ng/L). In Europe, another one-month follow-up study of OTA in urine samples after a 24-h collection of urine from the inhabitants with BEN in Bulgaria (from 16 healthy volunteers from two villages located in the Vratza district with a high risk area for BEN; 5 of Gorno Pestehne, 11 of Beli Izvor) was conducted by Petkova-Bocharova et al. (2003) [85]. 98% of samples were positive and contained OTA in the range 10-1910 ng/L. The OTA mean value in Gorno Pestehne was 50.8 ng/L, and in Beli Izvor it was 168.6 ng/L [85]. In a Czech study carried out in 2010, OTA was measured in a total of 236 samples of urine collected from healthy persons within a 24-h cycle (males/females, 45-60 years old, two samples per person from non-consecutive days, with at least a 14-day time difference). A total of 185 samples (78%) of these 236 samples were positive, with a limit of quantification (LOQ) of 2.0 ng/L, a mean of 7.32 ng/L, and a median of 4.47 ng/L [304,305]. These data signalize the real exposure of the given population group to OTA, with a higher percentage of positive urine samples in men (92%) than women (65%) [305].
OTA was usually determined in morning urine (not 24-h urine) in these countries (see also Table 11). However, in exposure studies, it is recommended that urine is collected over 24 h-representative of the excretion throughout a day [306].
The multibiomarker methods have been applied in several pilot studies to prove their applicability and to estimate mycotoxin exposure in the populations/individuals tested. The application of these methods resulted in advanced data on exposure patterns and revealed new findings on co-exposure to the mycotoxin combinations [307]. In addition, it must be mentioned that urinary excretion mainly reflects the recent mycotoxin intake, whereas measurements in plasma/serum are more likely to reflect the long-term exposure [307]. As a result of the advent of the latest generation of high-performance LC-MS/MS instruments, a clear trend toward the development and application of multianalyte methods in mycotoxin biomarker research can be observed [308].  [91] covered seven mycotoxins and several important conjugation and breakdown products (for a total of 18 analytes). In this study, OTA, OTα, and 4-OH OTA were measured in human volunteers [91]. However, none of the target metabolites of OTA such as OTα or 4-OH OTA were confirmed in another study performed with urine samples in Cameroon [309], but the data correlate with similar findings reported for a Korean population [89]. According to Munoz et al. (2010a) [276], interindividual variability in the detoxification of OTA in human urine may account for the observed variations in urinary OTα, and the possibility cannot be excluded that a low rate of OTA detoxification is a characteristic of some human populations [309]. The highest concentration of OTA reported so far in human urine was detected in Sierra Leone with a range of 70-148,000 ng/L, but no mean was reported [310]. Table 11 summarizes the OTA detection in human morning urine around the world. Last but not least, in dietary studies carried out in Serbia, in addition to OTA, several OTA derivatives have been detected in urine (and in blood). A clear difference between men and women has been observed [17]. Abbreviations: n: numbers of samples; n+ %: percentage of positive samples; * ng/g creatinine; ** GM: geometric mean; *** range in newborns consuming colostrums; **** range of samples collected between 4 and 6 months of infants' life; ***** mean in ng/kg.

OTA in Human Milk
As OTA is also excreted via human milk, breastfed children including babies are exposed to OTA as well [239,330]. Nevertheless, OTA amounts in milk are reported to be much lower than concentrations of OTA in blood (down to 10 times) [331]. In Italy, OTA was detected in milk from healthy women with varying daily diets in different geographical regions [332]. The relationship between OTA contamination of human milk and its dietary intake was examined [333], and it was confirmed that OTA occurrence in human milk was likely associated with maternal dietary habits. The strongest associations were observed with foodstuffs of plant origin and, to a lesser extent, with foodstuffs of animal origin [333]. Table 12 summarizes data on OTA presence in human milk worldwide. In some countries, e.g., Egypt, Turkey, and Sierra Leone, OTA milk concentrations were found to be more than 100-fold higher in comparison with Europe (see Table 12). It can be concluded that, despite the fact OTA concentrations in milk compared with blood are much lower, OTA contamination of human breast milk presents a potentially serious health hazard [354].

OTA in Human Kidneys
OTA presence in human tissues seems to be direct and definite proof of human exposure to OTA, although practicability of such measurements "in vivo" is obviously limited [355]. Taking OTA's nephrotoxicity in mind, in particular, there are not many studies available that have attempted to determine OTA in human kidneys. Several studies have been carried on the content of OTA in human kidneys, e.g., in Germany [356], in the Czech Republic in 30 samples of kidney (40% positive/detectable/samples; OTA ranged from 0.1 to 0.2 ng/g; mean 0.07 ng/g; results of OTA < 0.1 ng/g (LOQ) given as 1/2 limit of quantification = 0.05 ng/g) [357], and in Poland in 19 samples of kidney (78.9% positive/detectable; OTA ranged from 0.15 to 0.39 ng/g with mean 0.26 ng/g) [268]. Several human kidneys samples (60) obtained from patients suffering from kidney (or urinary bladder) cancer from Bulgaria (8 samples) [186], Serbia (10 samples), Croatia (16 samples), and France (18) [16,17] have been analyzed up to now. Not only was OTA detected but also OTA derivatives such as OTHQ, OTHQ-GSH, 4-OH OTA, and OTB. Interestingly, DNA adducts were detected, and the nature of the DNA was in relation to the OTA derivatives. In Croatia, the DNA adducts profile of a farmer was similar to the profile of the pigs and poultry from his farm. It has been observed that the exposure has been higher in rural regions, and co-exposure to CIT and/or FB has been systematic [16,17].

Regulation of OTA in Food and Feed
Due to its toxic properties, OTA is subject to legal regulation both on national and international levels. The toxicity of OTA became more or less evident by the end of the 1970s. A real debate on whether OTA in food and feed shall be regulated on a national or international level does not seem to predate the 1990s. This circumstance contrasts with the case of other mycotoxins, in particular, the aflatoxins (in the USA, the first limits for aflatoxins were established as early as the 1960s; soon after their discovery [358], the European Communities followed in the 1970s) [359].
For OTA, in 1991, van Egmond estimated that in 60 countries where some legal regulations with respect to mycotoxins existed, only 11 set limits on OTA (Brazil, Czechoslovakia, Denmark, France, Greece, Hungary, Israel, The Netherlands, Romania, Sweden, and the United Kingdom) [360]. In 2003, when a worldwide survey on legal regulation of mycotoxins was conducted by the FAO in cooperation with the Dutch Foreign Service, the number of countries with legal limits on OTA in food and feed grew to 37 (compared to more than 76 countries with legal limits for aflatoxins) [359]. No such large-scale survey has been reported ever since [361]. However, it may be assumed that the number of countries where OTA presence in food and feed is subject to legal regulation is not lower now than it was in 2003 (see Figure 5). This assumption can be based on two major arguments. Firstly, since 2003, research has provided new data on OTA's harmful effects to human and animal health. Secondly, due to the globalization of food and feed markets, discussion on how to tackle the health hazards linked to OTA (and other mycotoxins) has intensified on an international level and has had repercussions back on the national level. By way of example, China seems to have recently established limits on OTA in both food and feed [362]. derivatives such as OTHQ, OTHQ-GSH, 4-OH OTA, and OTB. Interestingly, DNA adducts were detected, and the nature of the DNA was in relation to the OTA derivatives. In Croatia, the DNA adducts profile of a farmer was similar to the profile of the pigs and poultry from his farm. It has been observed that the exposure has been higher in rural regions, and co-exposure to CIT and/or FB has been systematic [16,17].

Regulation of OTA in Food and Feed
Due to its toxic properties, OTA is subject to legal regulation both on national and international levels. The toxicity of OTA became more or less evident by the end of the 1970s. A real debate on whether OTA in food and feed shall be regulated on a national or international level does not seem to predate the 1990s. This circumstance contrasts with the case of other mycotoxins, in particular, the aflatoxins (in the USA, the first limits for aflatoxins were established as early as the 1960s; soon after their discovery [358], the European Communities followed in the 1970s) [359].
For OTA, in 1991, van Egmond estimated that in 60 countries where some legal regulations with respect to mycotoxins existed, only 11 set limits on OTA (Brazil, Czechoslovakia, Denmark, France, Greece, Hungary, Israel, The Netherlands, Romania, Sweden, and the United Kingdom) [360]. In 2003, when a worldwide survey on legal regulation of mycotoxins was conducted by the FAO in cooperation with the Dutch Foreign Service, the number of countries with legal limits on OTA in food and feed grew to 37 (compared to more than 76 countries with legal limits for aflatoxins) [359]. No such large-scale survey has been reported ever since [361]. However, it may be assumed that the number of countries where OTA presence in food and feed is subject to legal regulation is not lower now than it was in 2003 (see Figure 5). This assumption can be based on two major arguments. Firstly, since 2003, research has provided new data on OTA's harmful effects to human and animal health. Secondly, due to the globalization of food and feed markets, discussion on how to tackle the health hazards linked to OTA (and other mycotoxins) has intensified on an international level and has had repercussions back on the national level. By way of example, China seems to have recently established limits on OTA in both food and feed [362]. Membership of States in international or regional organizations may also contribute to adoption of legal regulations on OTA. For the time being, the binding maximum limits on OTA appear to exist only in the European Union (EU) (see infra). On the global level, debate on the feasibility of Membership of States in international or regional organizations may also contribute to adoption of legal regulations on OTA. For the time being, the binding maximum limits on OTA appear to exist only in the European Union (EU) (see infra). On the global level, debate on the feasibility of establishing the maximum limits on OTA has taken place at the Codex Alimentarius Commission (CAC), the joint intergovernmental body established by the FAO and WHO responsible for implementing the Joint FAO/WHO Food Standards Programme. After the Joint Food and Agricultural Organization (FAO)/World Health Organization (WHO) Expert Committee on Food Additives (JECFA), an expert body which provides scientific advice to the CAC repeatedly dealt with OTA in 1991OTA in , 1995OTA in , 2001, and 2007, the maximum limit of 5 µg/kg with respect to wheat, barley, and rye has been recently established under the Codex General Standard for Contaminants and Toxins in Food and Feed [363]. In addition, there are four codes of practice that aim at the prevention and reduction of OTA contamination in cereals [364], wine [365], coffee [366], and [367] adopted between 2007 and 2014 [368]. Although the Codex Alimentarius standards are not per se binding, their importance stems especially from the fact the World Trade Organization (WTO) considers the measures taken by its Member States in conformity with the Codex Alimentarius standards to be science-based, appropriate, and nondiscriminatory under the WTO Agreement on the Application of Sanitary and Phytosanitary Measures signed in 1994 and thus does not treat them as breaches of world trade rules.
As far as the existing limits on OTA are concerned, those of the EU are generally assessed to be the most comprehensive and detailed [359].
As for the limits on OTA in food, these were first established on the EU level by the Commission Regulation ( [370] setting maximum levels for certain contaminants in foodstuffs (see Table 13

Foodstuffs
Maximum levels (ng/g) Cereals (including rice and buckwheat) and derived cereal products 5 Raw cereal grains (including raw rice and buckwheat) 5 All products derived from cereals (including processed cereal products and cereal grains intended for direct human consumption) 3 Dried vine fruit (currants, raisins and sultanas) 10 Green and roasted coffee and coffee products, wine, beer, grape juice, cocoa and cocoa products, and spices -In the EU, the Regulation 1881/2006 remains in force today, although it has been amended nearly 26 times. As of February 2016, the Regulation No 1881/2006 sets the maximum limits on OTA not only in cereals (both in the unprocessed cereals and cereal products) but in a wide variety of other food commodities as well (see Table 14). These limits are legally binding on all 28 EU Member States, which are obliged to apply these rules in full.
Apart from setting binding limits on OTA in food, since 2002, the EU has also unified the methods of sampling and analysis for purposes of the official control of the levels of As for OTA in feed, however, up to now, only a non-binding recommendation exists with respect to cereal feed, and feed for pigs and poultry on the EU level (Commission Recommendation 2006/576/EC [373] of 17 August 2006 on the presence of deoxynivalenol, zearalenone, OTA, T-2 and HT-2, and fumonisins in products intended for animal feeding). For details, see Table 15.

Feed Guidance Value in mg/kg Relative to Feedstuffs with a Moisture Content of 12%
Feed materials *-Cereals and cereal products ** 0.25 Complementary and complete feedstuffs -Complementary and complete feedstuffs for pigs 0.05 -Complementary and complete feedstuffs for poultry 0.1 * Particular attention must be paid to cereals and cereals products fed directly to the animals that their use in a daily ration should not lead to the animal being exposed to a higher level of these mycotoxins than the corresponding levels of exposure where only the complete feedstuffs are used in a daily ration. ** The term "Cereals and cereal products" includes not only the feed materials listed under Heading 1, "Cereal grains, their products and by-products," of the non-exclusive list of main feed materials referred to in Part B of the Annex to Council Directive 96/25/EC of 29 April 1996 on the circulation and use of feed materials (OJ L 125, 23.5.1996, p. 35), but also other feed materials derived from cereals in particular cereal forages and roughages.
Toxins 2016, 8, 191 27 of 49 There are, however, approaches to legal regulation of OTA other than establishing and enforcing the binding maximum limits on OTA in food and feed commodities as in the EU. Most notably, no binding limits on OTA in food or feed exist in the USA. Even more strikingly, no advisory or regulatory action limits have been established by the US authorities. Instead, the US Food and Drug Administration (FDA), acting under the Federal Food, Drug and Cosmetic Act (FFDCA), has instead consistently relied on laying down good agricultural and manufacturing practices and on requiring the implementation of food safety plans in food industry undertakings [358]. In extension, the FDA monitors the compliance with these practices and the presence of OTA in domestic and imported foods (Food Compliance Programme No 7307.001 entitled "Mycotoxins in Domestic and Imported Foods"). An approach analogous to that of the USA has been adopted by a range of other countries such as Australia, Canada, and Japan [374].
For some authors, the US approach to regulating mycotoxins including OTA is clearly preferable because it is seen as an option that might "diffuse trade frictions, and at the same time help reduce economic losses from mycotoxin contamination and divergent standards" [375]. The truth is that the US approach seems to exert a non-negligible influence on the international level, e.g., within the CAC, which has, as mentioned above, adopted four codes of good practice with the aim of reducing the OTA occurrence in several food commodities that are commercially important.
To sum up, 50 years after the discovery of OTA, differences in how to legally regulate mycotoxins including OTA are still marked. However, even in an era when further liberalization of world trade is envisaged (e.g., a project of the Transatlantic Trade and Investment Partnership and the TTIP between the USA and the EU), due to economic and political controversies linked to the existing policies on mycotoxins, it cannot be expected that some harmonized approach to legally regulating mycotoxins including OTA will be easily established on a global level [375][376][377].

Conclusions
OTA is ubiquitously found all over the world in many foodstuffs and feedstuffs. OTA is recognized for its nephrotoxicity and, to date has been identified as one of the most potent renal carcinogens in rodents ever studied by the National Cancer Institute/National Toxicological Program (NCI/NTP) [181]. OTA is deleterious for the pig and poultry industries. For human beings, many authors consider it to be the main contributor in the pathogenesis of Balkan endemic nephropathy and some nephropathies in other parts of the world.
The development of effective strategies alleviating OTA-induced toxicity is very complex because the mechanism of action of OTA is still unclear. The toxic effect is the result of many effects such as the inhibition of protein synthesis, direct genotoxicity, and cell cycle arrest. Inhibition of OTA uptake and stimulation of OTA elimination of the body preventing OTA accumulation will be promising approaches [378].
Since its discovery in 1965, numerous studies have been performed with respect to OTA, which have permitted the establishment of different mechanisms for OTA nephrotoxicity and carcinogenicity (summarized in Figures 6 and 7). The mechanisms leading to OTA nephrotoxicity, its hepatotoxicity and immunotoxicity, can be linked to inhibition of protein synthesis, lipoperoxidation, and modulation of MAP kinase cascade (Figure 6), whereas its carcinogenicity arises after the metabolic activation of OTA in a way similar to pentachlorophenol derivatives (Figure 7).  OTA forms covalent DNA adducts through radical and benzoquinone intermediates. The OTHQ metabolite of OTA can undergo an autoxidative process to generate the quinone electrophile OTQ that reacts with DNA. In addition, the formation of OTQ or phenoxy and aryl radicals can lead to increased  OTA forms covalent DNA adducts through radical and benzoquinone intermediates. The OTHQ metabolite of OTA can undergo an autoxidative process to generate the quinone electrophile OTQ that reacts with DNA. In addition, the formation of OTQ or phenoxy and aryl radicals can lead to increased OTA forms covalent DNA adducts through radical and benzoquinone intermediates. The OTHQ metabolite of OTA can undergo an autoxidative process to generate the quinone electrophile OTQ that reacts with DNA. In addition, the formation of OTQ or phenoxy and aryl radicals can lead to increased ROS production that causes cytotoxicity. Radical species generate a C-bound C8-dG adduct, while benzetheno-type DNA adducts are expected from the quinone electrophile. The quinone-type adducts form faster in cells and stem from P450 activation of OTA. The C-bound C8-OTA adduct forms at a slower rate and is predicted to stem from reductive dehalogenation of OTA (via GSH and cyclooxygenase or lipoxygenase). The C5-Cl atom is critical for DNA adduction (genotoxicity) but not for cytotoxicity (OTB is cytotoxic but not genotoxic) (Figure 7). Several quinone derivatives have been isolated from blood and urine and also in human or animal tissues exposed to OTA. The OTB-dG adduct is consistently found by 32 P-postlabeling in kidney DNA from OTA-treated rats, pigs, and humans. These metabolites and this adduct could serve as biomarker for OTA exposure.
Increases in carcinogenicity and genotoxicity during co-exposure with citrinin (CIT), fumonisin (FB), or both can be explained by both factors. FB and CIT induce COX2, thus favoring the biotransformation of OTA into a genotoxic compound. Moreover, the quinone methide structure of CIT could easily explain the generation of DNA adduct. It may be capable of oxidizing OTA into the phenoxyl radical to promote C-C8 adduct formation. The new findings on OTA mutagenicity favor direct genotoxicity and rule out oxidative DNA damage as a contributor to the induction of deletion mutations or renal carcinogenesis. Therefore, further research should focus on co-exposure.
Altogether, OTA is a complete carcinogen, active since the earliest stage of life. Intake evaluation based on real analysis shows that the daily intake was three times greater than the virtual safety dose of 4 ng/kg bw/day-against carcinogenicity (intake per day 648 ng/60 kg adult) [379].
Maternal-fetal risk assessment of OTA during pregnancy was conducted using the benchmark dose approach for genotoxic carcinogens. Considering the sensitivity of a fetus, risk reduction is a high priority. It is essential to keep exposure to OTA as low as possible in women, notably during pregnancy [380].
Among the professional community, it is agreed that OTA is one of the five most agriculturally important mycotoxins; therefore, continued attention must be paid to research on ochratoxins and OTA in order to elucidate their metabolism, genotoxicity, and mechanism of action for renal carcinogenicity, with the ultimate aim of protecting public health and preventing economic losses. Dedicated to the memory of all researchers who substantially contributed to OTA research and helped to build general knowledge on OTA. Apologies to all the collegues whose important work on OTA is not highlighghted in this article.
Author Contributions: Frantisek Malir, Vladimir Ostry, Jan Malir and Jakub Toman reviewed the available data and wrote the core of the paper. Annie Pfohl-Leszkowicz performed a scientific supervision and manuscript revision. Jan Malir is responsible for a chapter on legal regulation of OTA and the correction of the English. All authors read and approved the final manuscript. The authors thank to Yann Grosse for all informations from the IARC, Lyon.

Conflicts of Interest:
The authors do not declare any conflict of interests.