Sequential Analysis of Trace Elements in a Micro Volume Urine Sample Using Inductively Coupled Plasma Mass Spectrometry

In this work, we describe a simple, fast, cheap, accurate and high-throughput method for the determination of Al; V; Cr; Mn; Fe; Co; Cu; Zn; As; Se; Mo; Cd; Sn; Ba; TI; Pb in a micro volume of urine by using inductively coupled mass spectrometry (ICP-MS) equipped with an octupole-based collision cell. The samples were directly analyzed after a simple acidification with 3% nitric acid. Validation of this method was performed by using certified urine reference material. He and H were used as collision gas for reducing polyatomic interference for most of the measured elements. Finally, we partook in an external quality assurance conducted by ISNTAND e.V. 2. Results show that our high-throughput method is ideal for detecting elements in newborns and infants because of its simplicity, speed, accuracy and low sample volume.


Introduction
Urine is one of the oldest and most basic human specimens used to evaluate the presence, severity and cause of diseases within the kidney and urinary tract [1]. Several diseases can be traced back to an imbalance of trace elements [2][3][4][5]. Since the early 1990s, trace elements have been attracting attention in a wide range of scientific areas related to human health such as clinical analysis, nutritional diagnosis and environmental analysis [6][7][8][9][10]. These elements may be either essential or non-essential. Essential elements (manganese (Mn), cobalt (Co), copper (Cu), zinc (Zn), molybdenum (Mo)) are those that are required for normal physiological processes in human beings [11]. Many elements are part of protein complexes that are required for enzymatic activities and may also play structural roles in tissue or cell membranes. Non-essential trace elements (aluminum (Al), vanadium (V), arsenic (As), cadmium (Cd), tin (Sn), lead (Pb)) are considered toxic and are not essential for physiological processes [9,11]. Moreover, essential trace elements can cause toxicity if they reach excessive levels in the human body. Chelation therapy is commonly used by physicians to counteract excessive and harmful levels of non-essential trace elements. Recently, several researchers have claimed that chelation therapy can also help in treating many other conditions, including heart disease [12,13], autism [14], Alzheimer's disease [15], and diabetes [12]. Chelation therapy is used to remove metals (e.g., Al, V, chromium (Cr), Mn, iron (Fe), Co, Cu, Zn, As, selenium (Se), Mo, Cd, Sn, barium (Ba), Tl, Pb) from the body by the formation of a chelate complex. Chelators  [32] In the past, atomic absorption spectrometry (AAS) was used for biomonitoring the levels of trace elements in various laboratories. Nowadays, many laboratories have moved towards methods based on inductively coupled plasma coupled with mass spectrometry (ICP-MS). ICP-MS has become the method of choice for analyzing trace elements in urine and blood due to its many advantages, including sequential multielement measurements, and lower cost-and time-consuming analysis [8,9]. The typical amount of sample for trace element analysis in urine has varied amongst authors and number of trace elements. Currently, the sample volume of urine reported for the sequential analysis of trace elements is usually between 200 µL [37] and 2000 µL [38]. However, the amount of urine is often a limiting factor in clinical testing laboratories, especially when diagnosing newborns or infants. Therefore, well validated methods using a micro volume of sample with minimal hands-on time are crucial for such applications. In this study, we develop and validate a simple dilute and shoot method for the sequential determination of 16 trace elementsnamely, Al; V; Cr; Mn; Fe; Co; Cu; Zn; As; Se; Mo; Cd; Sn; Ba; TI; and Pb-that are relevant in the monitoring of chelation therapy using a micro volume of human urine. We performed our experiments on an ICP-MS 7900 (Agilent Technologies, Tokyo, Japan) equipped with an octupole-based collision cell. As a further sign of quality, it is necessary to validate a method and participate in an external quality assessment in order to increase credibility with health agencies. Reducing the turnaround time and sample volume in routine analysis is also one of the most challenging parts in method development. Such traits are highly in demand due to the fact that they are moderately prized and timesaving in sample preparation as well as analysis time.

Reagents
For analytical purity suprapure nitric acid (Merck, Kenilworth, NJ, USA, Ref. 1.00441) and high-purity (>18 MΩ cm −1 ) deionized distilled water was used in all experiments. All solutions were prepared in high-purity water obtained from a Milli-Q Water-station (Milli-Q ® HX-7040-SD). All consumable materials such as tubes, gloves and vessels were purchased from Carl Roth GmbH and Co KG (Karlsruhe, Germany) laboratories. Plastic bottles and tubes were cleaned by soaking in 3% nitric acid overnight, rinsing in high-purity water, and then air-drying before use.

Samples
Commercially available control and calibration sample material produced for quality assurance purposes-manufactured by Recipe ® (ClinCheck, Munich, Germany) Chemicals and Instruments GmbH and Seronorm ® Trace Elements Urine Level 1 and 2 (Sero AS, Billingstad, Norway)-was used for method development, validation, and application. Therefore, no ethics approval was necessary.

Calibration and Sample Preparation
A three-point calibration using a blank consisting of 3% nitric acid and Urine Level 1 (REF 210605, Sero AS, Billingstad, Norway) and 2 (REF 210705, Sero AS, Billingstad, Norway) was prepared daily for the comparative analysis of trace element concentrations of Al, V, Cr, Mn, Fe, Co, Cu, Zn, As, Se, Mo, Cd, Sn, Ba, TI and Pb in urine. For assessment of linearity, slope, limit of detection (LOD) and limit of quantification (LOQ), a fivepoint calibration was used. For method validation reference materials, ClinCheck ® Trace Elements Urine L-1 (REF 8847, Recipe ® , Munich, Germany) and ClinCheck ® Trace Elements Urine L-2 (REF 8848, Recipe ® , Munich, Germany) were analyzed. A serial dilution in 3% nitric acid (made from suprapur nitric acid 65%, Merck, Darmstadt, Germany) of the reference samples was prepared in decreasing amounts. For sample preparation, only 100 µL of the corresponding urine sample was acidified with 3% nitric acid filled up to 3 mL in a 15 mL polypropylene tube. After that, the samples were homogenized on a vortex for 5 s and were then ready for measurement. Devices used for sample preparation were pipettes (volumes 50-1000 mL) and a dispenser with adjustable volumes from 1-5 mL (Eppendorf, Hamburg, Germany). This is an easy, quick, and accurate dilute and shoot sample preparation method for preparing many samples in a short time. A similar amount of Rh and Ir were added as an internal standard to all samples throughout the measurement.

Data Validation and Statistics
Statistical methods were validated according to the ICH Guidelines and implemented on three different days. All of the analytes presented in this work are in the acceptance range specified by the manufacturer, whilst most analytes are within clinical reference ranges, as shown in Table 4. A validation of the methods was carried out to determine performance and comprised of precision in series (CV in %), accuracy, limits of detection (LOD) and limits of quantification (LOQ). Precision was assessed using the coefficient of variation and accuracy by analyte recovery determination.

External Quality Assessment
The accuracy and the suitability of the method for routine use was tested by participation in an external quality assessment conducted by INSTAND e.V. (Düsseldorf, Germany) according to the Guidelines of the Society for the Promotion of Quality Assurance in Medical Laboratories. The freeze-dried samples were then transported to our laboratory at room temperature. The samples were ready for preparation after reconstituting them according to the INSTAND e.V. guidelines.

Results and Discussion
The aim of our study was to develop a fast, simple, cheap and well-validated method for the sequential detection of trace elements in urine by using a micro amount of sample which makes this method suitable for routine use in the biomonitoring of newborns and infants.

Matrix Interference
We developed simple and cost-effective dilute and shoot sample preparation procedures for evaluating and comparing sixteen important essential and non-essential trace elements. Urine samples could be analyzed directly in the simple dilution described above. The direct introduction of a highly concentrated organic matrix in the plasma often results in matrix interferences and spectral interferences from polyatomic ions. Spectral interferences have been reported for several elements such as 40 [39,40]. Al, V, Co, Cr, Cu, Zn, As, and Cd were determined using an octupole-based collision cell with helium and hydrogen as the collision gas in order to avoid interferences of molecules [39].

Sample Preparation and Digestion
Sample preparation is recognized to be the largest source of errors and one of the most critical analysis points [41]. Therefore, it is essential to eliminate any disturbing matrix influence to avoid undesirable effects on sample preparation or analysis. Variations in matrix effects of urine depends on various factors including diet, renal function or hydration status [42]. Nevertheless, urine is a body fluid with a comparably low amount of sample matrix, requiring little sample preparation (e.g., simple dilution with nitric acid [41]). Whilst research groups have reported simple dilution procedures [43,44], several used significantly more sample material. Nitric acid (≤5%) is commonly used for metal dissolution and stabilization for ICP-MS urine analysis [45]. Another advantage of adding acid is that it reduces the pH to less than 2 of the sample. It also stabilizes the sample and prevents losses through wall absorptions in the tube and the growth of bacteria [46].

Validation
The described procedure was thoroughly validated in accordance with the International Conference for Harmonization (ICH) guidelines to verify its fit for the intended purpose [47]. Several industry committees, regulatory agencies, and individual researchers have published reviews about validation strategies, quality assurance, and regulatory purposes [48], wherein most are related to the pharma and chemical industry. Over the years, method validation guidelines have been offered by numerous international organizations such as the American Society for Testing and Material (ASTM), the European Analytical Chemistry Group (EURACHEM), and the European Committee for Normalization (CEN) [49]. Consequently, several validation guidelines with different scopes have been issued describing the validation parameters to be studied, the way to determine each one, and their acceptance criteria. The ICH guideline is our preference, especially as this guideline has been developed for harmonization purposes. Our method was validated regarding linearity and slope, method LODs and LOQs, repeatability and accuracy and precision.

Linearity and Slope
The linearity was evaluated using correlation coefficients (R2). Regression is considered to be linear if R2 ≥ 0.995 [50]. R2 values were higher than or equal to 0.98 for all elements of this work. The linearity was calculated based on five analyte concentrations. Mean values of intensity from three repetitions were taken for each calibration point. Table 5 shows the linearity, slope and the validation range of our method. A linear regression can be assumed for all analyzed elements except for Al, Co, Zn, Se and Pb-where the R2 is slightly lower than the recommended value [50].

Limit of Detection (LOD) and Limit of Quantification (LOQ)
The limit of detection (LOD) and limit of quantification (LOQ) were calculated as a signal-to-noise ratio of 3 and 10, respectively. The LOD is the minimum concentration of the analyte that can be detected but not quantitated as an exact value. The LOQ is defined as the minimum concentration of the analyte that can be quantitatively determined with suitable precision and accuracy. LOD and LOQ were calculated according to Armbruster and Pry (2008) [51]. Table 6 shows the LOD and LOQ for all elements measured in this study.

Repeatability, Accuracy and Precision
Spectral interferences, as described above, play an important role in trace element analysis using ICP-MS. He and H collision or reaction gas, respectively, were chosen in order to avoid relevant spectral interferences, as shown in Table 3. According to Batista et al. [52], the use of standard reference material is recommended when validating a method. The element composition of the urine samples and the intra-and inter-day analysis of the samples are shown in Table 7. The values shown in Table 7 are the average of triplicate preparations within a day analysis (n = 3). The final average concentration is an average concentration of the average values of each day.   Table 7 demonstrates the determination of the elements Al, V, Cr, Mn, Fe, Co, Cu, Zn, As, Se, Mo, Cd, Sn, Ba, TI and Pb with an accuracy ranging from 90.97% to 150.73%. The values (except V, Co and Se as they were close to the LOQ) were in accordance with the methods' performance criteria recommended by INMETRO (2011) [53] in which the expected accuracy must be between 80-120%. The result was consistent with similar methods in the literature. Freire et al. [54] reported a recovery of 78-111% for trace elements extracted from freeze-dried urine standard reference material using a sample preparation method consisting of acidification with nitric acid and Triton X-100. In another acid digestion procedure for the determination of trace element contents in urine, recoveries of 87-116% were reported in urine reference material, respectively [55]. Moreover, trace elements in urine have been extracted and digested using a microwave system. For example, Yang Z et al., reported recoveries of between 81-107% for Al, V, Cr, Mn, Co, Cu, Cd, Sn, Ba, TI, and Pb as determined by ICP-MS [56]. Soubhia et al. [57] obtained Cr, Mn, Cu, As, Mo, Cd, Tl, Pb accuracy and repeatability results similar to this work. Repeatabilities for the intra-and inter-day values vary significantly amongst working groups and methods used. Marco and Hernandez-Caraballo reported repeatability values between 10-25% for V, Cr, Mn, Fe, Co, Cd, Zn, As and Se [58]. Minich et al. defined a smaller range of 3-9% for their method validation [59]. Nevertheless, we showed inter-day and intra-day repeatabilities (Cv) for all elements below 13% with our method. Regarding the results of the internal quality control, the measured values are in good agreement with the declared values of the control material from Seronorm TM (Sero AS, Billingstad, Norway), as indicated in Table 4.

External Quality Assurance
Well-validated and checked methods are essential if used to determine an abnormal amount of trace elements in the human body. Therefore, participation in a recognized external quality assurance has several benefits, including information on the relative performance of different methods, and knowledge about one's own ability to perform tests and report results accurately. Doing so also gains the confidence of clinicians and patients [60]. We partook in an external quality assurance using our method given the extensive benefits-including timesaving, low material costs, using only a micro volume of sample, and good accuracy and repeatability values-of our new method in comparison to other published methods, as shown in Table 8. A batch of ten analytes was tested during the INSTAND interlaboratory comparative program of the Institute for Standardization and Documentation in Clinical Laboratories. Laboratories partake in external quality assessment or proficiency testing schemes in order to evaluate the quality of their results and to fulfil accreditation requirements. Moreover, in some jurisdictions, it is mandated by law to participate in external quality assessments to guarantee a high standard of analysis and credibility with health authorities [61,62]. External quality assurance assessment scheme results for ten analytes are listed in Table 8. The measured and recommended concentrations are in good agreement with the interlaboratory comparative program of the Institute for Standardization and Documentation in Clinical Laboratories. This satisfactory correspondence was also found for extremely low concentrations (Cd value 1.02 µg L −1 ) as well as very high concentrations (Zn value 0.43 mg L −1 ) in urine. The results from the external quality assurance program demonstrate that the described method is valid for the quantification of the elements in urine, requiring a simple one-step dilute and shot method using ICP-MS.

Conclusions
The main goal of the work described above was to provide a simple, fast, inexpensive, sensitive, reliable and non-time-consuming method for the determination of essential and non-essential trace elements in urine using a micro volume of sample. To assess this, a suite of sixteen (Al, V, Cr, Mn, Fe, Co, Cu, Zn, As, Se, Mo, Cd, Sn, Ba, TI and Pb) trace elements-which play an important role in the context of chelation therapy and heavy metal discharge-were selected. After all conditions had been established, measurements became very efficient. After the subsequent participation in an external quality assessment, we can conclude that our method is suitable for detecting a lack or an overload of the mentioned trace elements. Moreover, its advantage to other reported methods lies in the speed, robustness and simplicity of the former. This method is also suitable for routine analysis and diagnosis in infants and newborns since it uses only micro amounts of urine.

Data Availability Statement:
The data presented in this study are available in this article.