Analytical Biochemistry 395 (2009) 224–230

Contents lists available at ScienceDirect

Analytical Biochemistry journal homepage: www.elsevier.com/locate/yabio

Multiple screening of urolithic organic acids with copper nanoparticle-plated electrode: Potential assessment of urolithic risks Chung-Wei Yang a, Jyh-Myng Zen b, Yu-Lin Kao c, Cheng-Teng Hsu b, Tung-Ching Chung a, Chao-Chin Chang d, Chi-Chung Chou a,* a

Veterinary Medical Teaching Hospital and Department of Veterinary Medicine, College of Veterinary Medicine, National Chung Hsing University, Taichung 402, Taiwan Department of Chemistry, National Chung Hsing University, Taichung 402, Taiwan Department of Urology, Chung-Shan Medical University Hospital, Taichung 402, Taiwan d Institute of Microbiology and Public Health, College of Veterinary Medicine, National Chung Hsing University, Taichung 402, Taiwan b c

a r t i c l e

i n f o

Article history: Received 29 May 2009 Available online 21 August 2009 Keywords: Urolithiasis Creatinine Organic acids Electrochemical Nanoparticle-plated electrode

a b s t r a c t There is yet to be a reliable prediction of urolithiasis. To facilitate early diagnosis, a simple and rapid high performance liquid chromatography method with electrochemical detection using disposable coppernanoparticle-plated electrodes (Cun-SPE) was developed for multiple detection of creatinine and 4 urolithic organic acids. A total of 206 normal and urolithic human and canine urines and urolith samples were collected for direct analysis of creatinine, cystine, uric acid, oxalic acid, and citric acid without sample cleanup and derivatization processes. Urinary organic acids were separated in 11 min and were devoid of ascorbic acid interference. The detection limits (S/N > 3) were at the nanomolar level with linear dynamic ranges spanning 2–3 orders of magnitude. Recoveries in urine ranged from 99.5% for creatinine to 86.5% for citric acid. The analytical variations (RSD) were less than 6.2% in phosphate buffer and 7.7% in urine. Important differences in organic acid levels/profiles between animal species and among normal and urolithic urines/urolith were unveiled and corresponded well (70–90%) with the urolithic risk in a retrospective assessment. The simplicity and reproducibility of this method using disposable Cun-SPE has made routine urine analysis possible and can be of great clinical and diagnostic potential in the screening of urolithiasis and abnormal states related to excess secretion of organic acids and amino acids in humans and animals. Ó 2009 Elsevier Inc. All rights reserved.

Urolithiasis refers to the causes and effects of the presence of uroliths (stones) or excessive amounts of crystal anywhere in the urinary tract [1,2]. This condition affects 1–20% of the general population, with an increasing lifetime incidence of 2–5% in Asia, 8– 15% in Europe and North America, and 20% in Middle East [3]. The prevalence of human urolithiasis in Taiwan is 9.6%, accounting for 15–24% of human patients in urological departments [4]. In veterinary medicine, the prevalence of canine and feline urolithiasis is 1.2–2% [1,5]. Currently, there are few effective medical options to treat urolithiasis other than urine pH modification and surgical intervention to remove the urolith, but the reoccurrence of urolithiasis is as high as 80% in humans and 20–50% in canines [1,2,5]. In order to develop rational and effective approaches to prevent and/ or treat urolithiasis and its reoccurrence, early recognition of the types of uroliths that are in formation or have high potential of formation is strongly desirable such that appropriate practice could * Corresponding author. Address: College of Veterinary Medicine, National Chung-Hsing University, 250-1 Kuo-Kuang Road, Taichung, Taiwan. Fax: +886 4 22862073. E-mail address: [email protected] (C.-C. Chou). 0003-2697/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.ab.2009.08.020

be initiated to create a favorable microenvironment for dissolution of uroliths [6–8]. Techniques commonly used for the diagnosis of urolith involve the use of X-ray and ultrasonography, but they do not differentiate among different urolith types [9]. Urine sediment examination could provide some useful information but it is considered nonquantitative and the sensitivity is usually unsatisfactory. Thus, a fast, inexpensive, noninvasive, and routinely applicable method to identify major urinary metabolites comprising the urolith is very valuable. The major organic acids in human urolith are uric acid (UA)1 and oxalic acids (OA) [3,4,6], while in canine, OA and struvite (ammonium magnesium phosphate) are the predominate types [1,5], accounting for more than 80% of the prevalence in both species. Cystine (Cys) stone is a distant second to OA and UA [3,5]. Other important urinary metabolites involved in the formation and/or dissociation of urolith include stone promoters (glycolate, glycerate) and inhibitors (citric acid, CA) [8,10–12]. Since a high concentration

1 Abbreviations used: CA, citric acid; Crt, creatinine; Cys, cystine; OA, oxalic acid; PBS, phosphate-buffered solution; UA, uric acid.

Urinary Organic Acid Screening for Urolithic Risks / C.-W. Yang et al. / Anal. Biochem. 395 (2009) 224–230

of urinary organic acids consisting major uroliths is one of the most important predisposing factors for urolith formation [13–16], various analytical methods have been developed for the detection of major and minor urinary organic acids. These methods include enzymatic method [17], gas chromatography (GC) [10,18], capillary electrophoresis (CE) [19–21], and high-performance liquid chromatography with UV (HPLC-UV) [22–24] or electrochemical (HPLCEC) [13,25,26] detectors. However, most of these published methods target primarily single or double organic acids of UA [10,13,21– 23,27], OA [10,17,28], and CA [10,25,29]. To our knowledge, despite that multiple screening of more than 3 urinary organic acids is not uncommon [20,25,28,29], there were very few analytical methods that focus on simultaneous determination of UA, OA, and their inhibitor CA [30] in single urine sample. In addition, creatinine (Crt) detection was not normally included in the published method, which makes accurate quantitation of urinary metabolites less attainable. The Recommendations from the National Institute of Health (NIH) Consensus Conference on the Evaluation of Stone Formers [10] suggested that measurement of Crt, UA, OA, and CA all be included in the assessment of patients with urinary stone disease. Therefore, in order to facilitate risk assessment of urolithic potential by multiple screening of urinary metabolites, electrochemical detection following liquid chromatography was optimized in view of the fact that Crt is electroactive and many urolithic organic acids do not possess any pronounced chromophoric and/or fluorophoric groups for UV-visible and fluorescence detectors [26]. The current work reports on the development and use of a simple, rapid, and sensitive electrochemical-based analytical method devoid of sample pretreatment for routine determination of specific urinary organic acids and Crt. The utilization of the optimized method to uncover important differences between normal and urolithic urine for potential clinical screening of subjects with ongoing or higher risk of urolithiasis was demonstrated.

225

(at 2 mM) from 0.6 to 0.6 V were performed to decide the optimal working potential. Current–time (i–t) plots were then carried out for all 5 compounds using the decided fixed potential of 0.05 V to examine the voltammetric responses. Chromatographic separations were performed using a silica-based HPLC column (Hamilton RCX-10 anion-exchange column 250 mm  4.1 mm i.d., Alltech, Reno, NA, USA) with a mobile phase of 0.1 M, phosphate buffer. Chronoamperometric signals for quantification were analyzed by CHI721B electrochemical analyzer with CHI system software (CH Instruments, Inc., Austin, TX, USA). Sample collection and preparation A total of 206 normal and urolithic urine samples, including human (60 normal and 35 urolithic) and canine (35 normal and 50 urolithic) origin, were collected for this study. In addition, 26 urine stones (6 human and 20 canines) were also extracted to analyze for target organic acid concentrations. Human urine samples were collected from the first morning catch of the outpatients at the Department of Urology, Chung-Shan Medical University Hospital, Taichung, Taiwan. Canine samples were collected at random times for a period of 1 year from the animals visiting Veterinary Medical Teaching Hospital at the Department of Veterinary Medicine, National Chung-Hsing University, Taichung, Taiwan. Immediately before analysis, urine samples were filtered through 0.22-lm (pore size) Millipore filters (Millex GV 13 mm, Billerica, MA, USA) and 20 ll was injected. Uroliths were dried for 24 h at 37 °C in an oven and crushed in an agate mortar until a fine powder appeared. One hundred milligrams of the grounded powder was inundated with 1 ml of 0.1 M PBS at 3 pHs (pH 6, 7.5, and 8) and placed on a vortex overnight for organic acid extraction. The solution was then centrifuged at 3000 rpm for 10 min and the supernatant was filtered by passing through a 0.22-lm Millipore syringe filter before analyzed by HPLC.

Materials and methods Urinalysis Chemicals and reagents Creatinine, UA, OA, CA, and Cys were purchased from Sigma (St. Louis, MO, USA). The mobile phase consisted of 0.1 M phosphatebuffered solution (PBS, 0.866 g Na2HPO4 and 0.468 g NaH2PO4 in 1 L HPLC grade water, pH 7.5). All compounds used in this work were ACS-certified reagent grade. Standard stock solutions (250 mM for Crt, CA, and OA; 100 mM for Cys; 10 mM for UA) were prepared by dissolving appropriate amounts of Crt and each organic acid in distilled deionized water. The stock solutions were stored in brown centrifuge vials and kept at 80 °C. Fresh working solutions were prepared by appropriate dilution of the stock solutions in pH 7.5 PBS before use. Instrumentation and HPLC procedure The HPLC system consists of a Hitachi L-6200 intelligent pump drive, a Rheodyne Model 7125 sample injector (20 ll sample loop) with an interconnecting Teflon tube and a specially designed electrochemical flow cell [26] incorporating a three-electrode system consisting of a working electrode (Cun-SPE, copper nanoparticleplated screen-printed electrode, geometric area = 0.2 cm2), an Ag/ AgCl reference electrode, and a platinum auxiliary electrode (geometric area = 0.07 cm2). The disposable Cun-SPE electrode was purchased from Zensor R&D (Taichung, Taiwan) or prepared as reported earlier [26,30]. The electrode was first washed by deionized water followed by equilibrating in a carrier solution (pH 7.5 PBS) at 0.05 V until the current become constant. Before the flow injection analysis, cyclic voltammograms (CV) of each organic acid

All urine samples were subjected to regular urine reagent strips tests (URS-10, Teco Diagnostics, Anaheim, CA, USA) including the estimation of specific gravity. A urine precipitant exam for observation of urocrystallizations was also performed. Briefly, 10 ml of urine sample was concentrated 10-fold by centrifugation and one drop (ca. 20 ll) of the precipitate was observed under light microscope at 200- and 400-fold magnification. Data analysis Descriptive statistics were generated for the statistical comparisons of organic acid concentrations and ratios in normal and urolithic urines and in urolith extracts for each group. The differences in average organic acid concentrations were assessed using t tests (Sigmaplot 8.0, Systat Software Inc., Richmond, CA, USA). Multiple comparisons among different organic acid levels and ratios were conducted within each species using a computerized program (SAS 6.12 for Windows; SAS Institute, Cary, NC, USA). For all comparisons, differences were considered significant when P < 0.05. Results and discussion Urolithiasis is a common disorder in human and companion animals with high recurrence rates. The etiology for urolithiasis remained debatable but believed to be multifactorial mostly in association with dietary and water intake as well as infections in the lower urinary tract. Diagnosis of urolithiasis heavily relied on imaging techniques but reliable results cannot be obtained before

226

Urinary Organic Acid Screening for Urolithic Risks / C.-W. Yang et al. / Anal. Biochem. 395 (2009) 224–230

the size of the stone becomes significant. Since a high concentration of urolithic organic acids is one of the most important predisposing factors for urolith formation [13–16], detection of urolithic components in the urine becomes valuable in the identification of subjects with higher risk potential and could be used to monitor the progress of the disorder or effectiveness of treatments for prevention of reoccurrence. Ideally, multiple determinations of major urolithic organic acids are favorable due to the complex nature of urolith formation. However, despite that multiple screening of organic acids as bases for screening of inborn disease or metabolic disorders is not uncommon [20,29,31,32], analytical targets studying urolithiasis have been mainly focusing on the most common single or double organic acids (UA and/or OA with or without concurrent analysis of Crt) [22,23]. There were not very many methods aiming at multiple (more than 4) detection of major urolithic metabolites (see review in Ref. [10]) except for some studies on minor urinary organic acids due to their possible roles in the initiation of stone formation [23,29,31,32]. Recently, Hsu et al. reported an EC method with the ability to detect 5 urinary metabolites; however, the sensitivity was only in the micromolar range and the peak resolution was unsatisfactory. In addition, the use of the method was minimally tested on clinical samples. Here, the EC method was optimized to achieve much improved resolution and sensitivity (nM) using novel copper nanoparticle-coated electrodes and successfully applied to real human and canine urine and urolith samples. HPLC-EC using CunSPE Current–time (i–t) plots for Crt, Cys, OA, UA, and CA in PBS (pH 7.5) using CunSPE are shown in Fig. 1. With duplicated additions of each compound, a constant increase in anodic current was observed, indicating steady electrochemical responses between the tested compounds and the electrode. Similar electrochemical behavior was observed for all 5 test compounds when sequential voltages ( 0.6 to 0.6 V) were applied and a working potential of 0.05 V was chosen based on the highest responding current produced (Fig. 1 inset and [30]). An anodic current increases as the cathodic current decreases as depicted in the inset of Fig. 1. The underlying mechanism for organic acid detection by CunSPE could be explained by the formation of cyclic four-member complex intermediate (Cu(II)-o-quinolate) or Cu(amino acid)2 as seen with the detection of o-diphenols and a-amino acids, respectively, in a neutral or alkaline medium [26,33]. Based on the CV results, an

optimal operating potential of 0.05 V versus Ag/AgCl was selected for all five investigated compounds. Anion-exchange chromatography was employed for separation since the CunSPE showed strong electrochemical activity for the oxidation of the targeted compounds especially in a neutral PBS medium. Various flow rates and mobile phase pHs and ionic strengths were investigated. Optimum conditions were found using a 0.1 M pH 7.5 PBS at a 1 ml/min flow rate. Under these optimum chromatographic conditions, a standard mixture of these 5 urolithic compounds could be well resolved in 11 min (Fig. 2) with at least equivalent or improved resolution and/or superior sensitivity than previous reports [17,19,22,30] (also see review in Ref. [10]). Peak identification was confirmed by retention time and spiking authentic standard solutions into the sample matrices. It is important to note that under the same experimental conditions, the electrochemical response was selective to these interested organic acids while acetic acid, lactic acid, meleic acid, and succinic acid, which may all appear in urine, showed no response to the current system [30]. Another common organic acid that usually presented as interference to analysis (especially for enzymatic methods) of the urine is ascorbic acid (vitamin C). While it is not directly related to urolith formation and thus not the intended target of this study, we found that it was a noninterfering, electroactive compound to CunSPE [34], and could be detected in both human and canine real samples (Fig. 3, arrow). This was confirmed by analyzing spiked ascorbic acid standard in a urine sample containing Crt and oxalic acid (peak f in the inset of Fig. 2). Although there are many other organic acids (such as fumaric acid, methylmalonic, etc.) also present in the urine with diagnostic implications to other disorders (e.g., acidemia) [32], they are present at much lower concentrations and not directly associated with urolith formation. Therefore, the high selectivity of the CunSPE to polycarboxylic organic acids is advantageous in simplifying the complicated organic metabolite profiles normally presented in human and canine urines. Aside from polycarboxylic acids, essential amino acids that are not known to associate with urolith formation could also be found in the urine. Since they are mostly also electroactive to CunSPE [36], their detection in real samples is possible. An interfering peak near cystine was observed in the real urine samples, which could be attributed to amino acids structurally and electrochemically similar to cysteine, such as glycine [36], but further identification is warranted.

0.0 Citric acid

50 mM

-1.0

50 mM

d

Current (µA)

10 µA

-2.0

b

20 mM

c

0.8 0.6 0.4 0.2 0.0 -0.2 -0.4 -0.6 -0.8

Potential / V

-3.0

20 mM

e

f 1µA

-4.0

20 mM

a

-5.0 Oxalic acid

-6.0

Creatinine

Citric acid

Uric acid

0

Cystine

200

400

600

800

Time (sec) 0

100

200

300

400

500

600

Time (sec) Fig. 1. Current–time plot of cystine, creatinine, oxalic acid, uric acid, and citric acid using the CunSPE in pH 7.5 0.1 M PBS. The inset depicts representative cyclic voltammogram of citric acid, with solid line indicating background and dashed line representing 2 mM citric acid. Scan rate, 10 mV/s, potential of 0.05 V vs Ag/AgCl.

Fig. 2. Chromatogram of a standard mixture containing (a) 800 lM cystine, (b) 10 mM creatinine, (c) 5 mM oxalic acid, (d) 400 lM uric acid, and (e) 500 lM citric acid. The inset depicts the detection of spiked 0.1 mM ascorbic acid standard (peak f) in a urine sample containing creatinine and oxalic acid. Mobile phase, pH 7.5 0.1 M PBS; flow rate, 1 ml/min; sample loop, 20 ll; amperometric conditions, CunSPE at a potential of 0.05 V vs Ag/AgCl.

227

Urinary Organic Acid Screening for Urolithic Risks / C.-W. Yang et al. / Anal. Biochem. 395 (2009) 224–230

mostly to the many advantageous characters exhibited by nanosizing the copper particles to obtain greater specific surface area, greater density of surface active centers, higher surface activity, higher catalytic efficiency, and stronger adsorbing ability [35]. The nano-sized copper material also renders the electrode less sensitive to large proteins [36], further decreasing possible background interferences in biological samples, which contribute both to the improved sensitivity and to better resolution of peaks (selectivity). To test for the accuracy of this method, normal urine sample was spiked at 2 concentrations with the 5 compounds and percentage recoveries were calculated. A summary of the results is shown in Table 2. All urolithic compounds had excellent percentage recoveries signifying the high accuracy of this procedure and the potential for further clinical studies. The precision of the detection system was also examined at multiple normal concentrations and the RSDs were below 6.2% in PBS (Table 1) and 7.7% in urine (Table 2). For urolith extractions, neutral pH (7.5) gave rise to the best average extraction efficiency in OA and UA and was therefore chosen. Based on these results, the developed method was very useful for a direct and fast analysis of organic acids in original biological fluids (urine and urolith extractions) without derivatization and complicated pretreatment of the samples, which are both desirable features for LC-based analytical methods. Organic solvents such as acetonitrile and methanol were not required in this method, which was eco-friendly and advantageous [22].

Human

A

B

Canine

A

e B

Analysis of human and canine urines

a d

c b 0

200

400

600

800

Time (sec) Fig. 3. Differential chromatographic profiles of organic acids between urine (solid line) and urolith extractions (dotted line) in the same human and canine subject. Urolith is subjected to organic acid extraction as described under Materials and methods. Samples A and B represent characteristic chromatograms rich in distinctive organic acids. Peak identification: (a) cystine, (b) creatinine, (c) oxalic acid, (d) uric acid, and (e) citric acid. The possible presence of ascorbic acid is pointed out by arrows.

The quantitative responses of these compounds, including retention time, limit of detection (LOD, S/N > 3), limit of quantitation (LOQ, S/N > 10), and linear working range, are summarized in Table 1. The LOD from 10 nM for UA to 250 nM for Crt, the LOQ from 0.1 lM for UA to 1.25 lM for Crt, and correlation coefficients of at least 0.992 were obtained. The detection limit of this EC method was well below the physiological levels of these endogenous (usually at high lM range) and exogenous compounds in urine so the current method should be an effective tool for the study of urolithiasis in which all these compounds appear at elevated concentrations. The improved sensitivity using CunSPE was another important feature of the current method and could be attributed

The separation and detection of urolith-associated organic acids in urine and urolith is important to the understanding of urolithiasis. Urolithiasis should not be viewed conceptually as a single disease with a single cause but rather as a sequel of familial, congenital, or acquired pathophysiological multiple interacting abnormalities that, in combination, progressively increase the risk of precipitation of excretory organic acids in urine to form stones. Therefore, multiple urine profiles could be generated with different etiologies or in different urolithiatic stages. The developed method proved to be a powerful tool for identifying these differences in both human and canine urine samples. Representative urinary organic acid and urolith extraction profiles identified with the current method are seen in Fig. 3. The results suggested that urinary (solid lines) and urolithic (dotted line) organic acid profiles could be very different, as shown in the chromatograms exhibiting distinctive major peak patterns (Fig. 3). There was no detectable Crt in the urolith. Hypotheses regarding the initiation and formation process of urolithiasis could be partially examined based on the differential profiling of organic acids in a urine sample and its matching urolith sample. The differential profiles indicated that the presence of UA in urine did not necessarily suggest the presence of UA in urolith (peak d of both human B and canine B in Fig. 3) and vice versa. This result was further supported by the observation that urocrystallization results from the urinalysis did not correlate well with the levels of organic acids or the existence of urolith in this study (data not shown). Nevertheless, statistical

Table 1 Quantitative responses of creatinine and urolithic organic acids (n = 10). RT (s) Cystine Creatinine Oxalic acid Uric acid Citric acid

173.5 241.7 369.6 489.7 601.8

LOD (nM) 100 250 250 10 250

LOQ (lM) 1.00 1.25 1.25 0.10 1.25

RSD (%) 6.2 2.5 2.9 5.3 3.4

RT, retention time; LOD, limit of detection (S/N > 3); LOQ, limit of quantitation (S/N > 10).

Linear equation 6

3

Y = 1.13  10 [c] + 2.61  10 Y = 4.05  10 7[c] + 3.62  10 5 Y = 2.19  10 6[c] + 2.70  10 Y = 1.11  10 7[c] + 6.06  10 3 Y = 8.99  10 7[c] + 3.34  10 4

3

Linear range (lM)

R2

0.10–250 0.25–250 0.25–250 0.05–100 0.25–250

0.999 0.992 0.999 0.998 0.993

228

Urinary Organic Acid Screening for Urolithic Risks / C.-W. Yang et al. / Anal. Biochem. 395 (2009) 224–230

Table 2 Reproducibility and recovery of creatinine and urolithic organic acids in urine.

Cystine Creatinine Oxalic acid Uric acid Citric acid

Original (lM)

Spike Low

High

Low

High

Low

High

265 832 18 423 1

5 5 5 5 2.5

250 250 250 250 250

238 760 22 372 3

482 1076 232 629 244

88.0 90.8 86.6 86.8 86.5

93.6 99.5 94.0 93.4 97.0

After spike

Recovery (%)

RSD(%)

5.4 6.4 4.0 6.5 7.7

All concentrations are in lM (n = 10).

Urolithic human urine Normal human urine

Concentration (µM)

Concentration (µM)

1000

10500

7000

750

c 500

d

d

d

Concentration (µM)

300

c

c d

200 100 0

c

d

c

0

Creatinine

Oxalic acid

Uric acid

500 Urolithic canine urine n = 50 Normal canine urine n = 35 Canine urolith extraction n = 20

400

a

a

300 a b

200 c

c

250

100

d

d

c

d

0

Oxalic acid

0

Creatinine

Oxalic acid

2500

Uric acid

Citric acid

Urolithic human urine n = 35 Normal human urine n = 60 Human urolith extraction n = 6 a

2000

Concentration (µM)

5000

400

0

3500

1500 a b b c

b c c

500

500

10000

n = 35 n = 60

b

1000

b

a

14000

B

15000

B 17500

a Urolithic canine urine n = 50 Normal canine urine n = 35

Concentration (µM)

A

A 20000 Concentration (µM)

analysis indicated that OA, UA, and CA (only in human) concentrations were significantly higher in urolithic than in normal urine samples (Figs. 4B and 5B). These results were in agreement with the general concept that although crystalurine and/or high UA or OA levels are important predisposing factors to urolithiasis, the determining factors are multiple and a threshold level for the organic acid to reach might be necessary before they can assume a significant role (see below). Other cofactor(s) in addition to higher urinary levels of organic acid might be essential facilitating urolith formation. Therefore, we further examined the contribution from cations (Na+, Ca2+, Mg2+) and the balance of the major organic acids by calculating their respective organic acid ratios. The statistic results indicated that significantly (at P < 0.01 level) higher Na+  OA and Na+  UA in human, and higher Na+  OA, Mg2+  OA, Ca2+  OA, and Mg2+  UA in canine, were found in urolithic urines (data not shown). In contrast, while the averaged ratios of UA/OA

d

c d

0

Oxalic acid

Uric acid

Citric acid

Fig. 4. Average concentrations of creatinine, oxalic acid, uric acid, and citric acid in normal and urolithic human urines and in uroliths are shown in panel A. Creatininecorrected organic acid concentrations are shown in panel B. Columns with the same letter and bar were not statistically different at P < 0.05 level.

Uric acid

Fig. 5. Average concentrations of creatinine, oxalic acid, and uric acid in normal and urolithic canine urines and in uroliths are shown in panel A. Creatinine-corrected organic acid concentrations are shown in panel B. Columns with the same letter and bar were not statistically different at P < 0.05 level.

and UA/CA were higher in human (but not canine) urolithic urines, it was not statistically significant (data not shown) duo to the relatively large ranges of organic acid concentrations. These two supplemental data suggested that the combination of urinary cations and anions and/or their ratios are also potential parameters useful in further differentiating normal and urolithic urines. Further studies are warranted before a more conclusive reference could be deducted. Species difference in the organic acid profiles and peak ratios were also unveiled between human and canine urines using the developed method. The average Crt-corrected OA and UA concentrations were both higher in human (Figs. 4 and 5). We have examined limited numbers (3 urolithic and 13 normal) of feline urine samples and the results were similar to those of canines (data not shown). The observed species difference might be the result of compositional differences of urolith types between human and

229

Urinary Organic Acid Screening for Urolithic Risks / C.-W. Yang et al. / Anal. Biochem. 395 (2009) 224–230

2500

Non-citric urolithic urine Citric urolithic urine Non-citric normal Citric normal urine

Concentration(µM)

2000

n = 16 n = 19 n = 43 n = 17 a

1000 b

500

c

b

b

c d

0

oxalic acid

Non-corrected n = 32 Crt-corrected n = 32 Specific-gravity-corrected

2000

a

n = 15

1500

b

b 1000

500

0

c

Oxalic acid

c

c

c

Uric acid

c

c

Citric acid

Fig. 7. Comparison of urinary oxalic acid, uric acid, and citric acid concentrations in human before and after creatinine or specific gravity correction. Columns with the same letter are not statistically different at P < 0.05 level.

advanced copper nanomaterial as electrode with a nanomolar range limit of detection, good resolution, and the ability for simultaneous detection of four urolith-associated important organic acids as well as Crt. No sample pretreatment was necessary. Significant differences among urinary organic acid profiles and species were discovered such that potential parameters for identification of patients with higher risk of current or developing urolithic disorder could be derived. These differences also provided information not previously revealed that is important to the understanding of the pathological squeal of urolithiasis in different species. Although mechanistic elucidation was not the main purpose of the study, the current results in general supported the idea that urolith formation is promoted when certain urinary organic acid levels are elevated, or when various organic acids become imbalanced [14–16]. The findings signify a possible clinical value of this method in the prediction and/or diagnosis of patients with potentially higher risk of having urolithic disorders. Retrospective analysis of the current data suggested that by using a threshold of 300 lM OA in human urine, 70% of the urolith patient could be correctly predicted while using 250 lM OA or 150 lM UA as thresholds, 90% and 80% of the urolithic canine could be predicted, respectively. We have implanted this HPLC-EC system in the veterinary diagnostic laboratory as a routine assessment of the propensity of urolith formation so that appropriate treatment or dietary modification could be initiated to ameliorate the high levels of urinary organic acids. It should be noted that for these results to be used as a radical diagnostic/predictive tool for urolithiasis, subsequent studies to increase the sample sizes and to include information regarding the urolith types are warranted. Nevertheless, the potential of using this method to identify subjects with higher urolith potential cannot be overlooked. Acknowledgments

a

1500

2500

Concentration (µM)

animals. Human stones are usually of a simple type in nature in which only one organic acid is predominant (mostly UA, occasionally OA). On the other hand, canine or feline stones are usually a mixed type where multiple organic acids could all be present in different ratios [1,3,5,6]. Another noticeable species difference was CA where it was only detectable in human samples (Fig. 3, Human A, peak e). Further analysis revealed that only in the presence of CA, concentrations of UA and OA are significantly higher in urolithic urine (Fig. 6), suggesting that the higher OA or UA was likely a result of improved solubility of acidic urolith by CA [11]. These results further highlighted the beneficial use of the developed method to detect dynamic changes of organic acids and to monitor the treatment effects of the alkalizing agent citrate because of the capability to detect all 3 compounds. The ability to detect Crt concurrently with organic acids in the same urine sample was another strength of this EC method. Traditional analysis of Crt is either time consuming or prone to assay interferences [37]. Urinary Crt level is known to be a sensitive and specific index for evaluating glomerular filtration rate and thus has been a reliable indicator as a urine dilution factor. The concurrent determination of Crt concentration allowed the correction of organic acid concentrations to better reflect the dilution status of the urine and the clinically relevant concentration of organic acids in relation to the hydration state of the animal. The Crt correction was made by dividing the original concentration of each organic acid by its own Crt concentration (i.e., the Crt concentration in the same urine sample) and then multiplied by the average Crt concentration in the same group (i.e., urolithic or normal). As a result, different organic acid levels were obtained before and after Crt correction (Figs. 4 and 5, panel A vs B). Most importantly, significant differences between normal and urolithic samples were only evident with the corrected data. To further demonstrate the desirable use of Crt, the specific gravity (SG), another parameter commonly used to evaluate the concentrating ability of kidney, was comparably used to correct OA, UA, and CA concentrations in human samples. Organic acid concentrations in urine sample were divided by the SG of the same sample and then multiplied by the averaged SG of samples in the same group (i.e., urolithic or normal). The results clearly indicated the superiority of Crt over SG as SGcorrected data showed no statistical difference compared to the noncorrected (Fig. 7). To our knowledge, this is the first report validating a HPLC-EC method for the analysis of urolithic organic acids in both canine and human urines and uroliths. The method employed the more

The authors thank Dr. Chih-Huan Yang, Dr. Wei-Ming Lee, and Dr. Shih-Chieh Chang for providing canine urine and urolith samples. This work was supported by National Science Council, Taiwan (NSC93-2313-B-005-075). References

uric acid

Fig. 6. Creatinine-corrected oxalic acid and uric acid concentrations in normal and urolithic human urines with or without the presence of citric acid. Columns with the same letter and bar were not statistically different at P < 0.05 level.

[1] C.A. Osborne, J.P. Lulich, J.M. Kruger, L.K. Ulrich, L.A. Koehler, Analysis of 451,891 canine uroliths, feline uroliths, and feline urethral plugs from 1981 to 2007: perspectives from the Minnesota Urolith Center, Vet. Clin. North Am. Small Anim. Pract. 39 (2009) 183–197. [2] G. Bihl, A. Meyers, Recurrent renal stone disease-advances in pathogenesis and clinical management, Lancet 358 (2001) 651–656.

230

Urinary Organic Acid Screening for Urolithic Risks / C.-W. Yang et al. / Anal. Biochem. 395 (2009) 224–230

[3] H.G. Tiselius, Epidemiology and medical management of stone disease, Br. J. Urol. Int. 91 (2003) 758–767. [4] Y.H. Lee, W.C. Huang, J.Y. Tsai, C.M. Lu, W.C. Chen, M.H. Lee, H.S. Hsu, J.K. Huang, L.S. Chang, Epidemiological studies on the prevalence of upper urinary calculi in Taiwan, Urologia 68 (2002) 172–177. [5] L.A. Koehler, C.A. Osborne, M.T. Buettner, J.P. Lulich, R. Behnke, Canine uroliths: frequently asked questions and their answers, Vet. Clin. North Am. Small Anim. Pract. 39 (2009) 161–181. [6] M.S. Pearle, Prevention of nephrolithiasis, Curr. Opin. Nephrol. Hypertens. 10 (2001) 203–209. [7] Y. Ogawa, H. Yonou, S. Hokama, S. Hokam, M. Oda, M. Morozumi, K. Sugaya, Urinary saturation and risk factors for calcium oxalate stone disease based in spot and 24-h urine specimens, Front. Biosci. 8 (2003) 167–176. [8] C.Y.C. Pak, Citrate and renal calculi; new insights and future directions, Am. J. Kidney Dis. 17 (1991) 420–425. [9] G. Meissner, B. Oehme, J. Strackeljan, A. Kuhr, J.T. Kocher, A method for the validation of a new calculus detection system, J. Clin. Periodontol. 32 (2005) 659–664. [10] C. Barbas, A. García, L. Saavedra, M. Muros, C.L. Mollerup, Urinary analysis of nephrolithiasis markers, J. Chromatogr. B 781 (2002) 443–455. [11] K. Chow, J. Dixon, S. Gilpin, J.P. Kavanagh, P.N. Rao, Citrate inhibits growth of residual fragments in an in vitro model of calcium oxalate renal stones, Kidney Int. 65 (2004) 1724–1730. [12] S. Srinivasan, P. Kalaiselvi, R. Sakthivel, V. Pragasam, V. Muthu, P. Varalakshmi, Uric acid: an abettor or protector in calcium oxalate urolithiasis, biochemical study in stone formers, Clin. Chim. Acta 353 (2005) 45–51. [13] J.S.N. Dutt, M.F. Cardosi, C. Livingstone, J. Davis, Diagnostic implications of uric acid in eletroanalytical measurements, Electroanalysis 14 (2005) 1233–1243. [14] W.G. Robertson, J.S. Jones, M.A. Heaton, A.E. Stevenson, P.J. Markwell, Predicting the crystallization potential of urine from cats and dogs with respect to calcium oxalate and magnesium ammonium phosphate (struvite), J. Nutr. 132 (2002) 1637S–1641S. [15] M. Daudon, C. Hennequin, G. Boujelben, B. Lacour, P. Jungers, Serial crystalluria determination and the risk of recurrence in calcium stone formers, Kidney Int. 67 (2005) 1934–1943. [16] J.H. Parks, M. Coward, F.L. Coe, Correspondence between stone composition and urine supersaturation in nephrolithiasis, Kidney Int. 51 (1997) 894–900. [17] G.L. Mingen, M.M. Madappally, Rapid enzymatic determination of urinary oxalate, Clin. Chem. 35 (1989) 2330–2333. [18] A. Liu, M.M. Kushnir, W.L. Roberts, M. Pasquali, Solid phase extraction procedure for urinary organic acid analysis by gas chromatography mass spectrometry, J. Chromatogr. B 806 (2004) 283–287. [19] A. Garcia, M. Muros, C. Barbas, Measurement of nephrolithiasis urinary markers by capillary electrophoresis, J. Chromatogr. B 755 (2001) 287–295. [20] E.A. Clark, J.C. Fanguy, C.S. Henry, High-throughput multi-analyte screening for renal disease using capillary electrophoresis, J. Pharm. Biomed. Anal. 25 (2001) 795–801. [21] Y. Guan, T. Wu, J. Ye, Determination of uric acid and p-aminohippuric acid in human saliva and urine using capillary electrophoresis with electrochemical

[22]

[23]

[24]

[25]

[26] [27]

[28]

[29]

[30]

[31]

[32]

[33] [34]

[35] [36]

[37]

detection potential application in fast diagnosis of renal disease, J. Chromatogr. B 821 (2005) 229–234. J.F. Jen, S.L. Hsiao, K.H. Liu, Simultaneous determination of uric acid and creatinine in urine by an eco-friendly solvent-free high performance liquid chromatographic method, Talanta 58 (2001) 711–717. Y.D. Yang, Simultaneous determination of creatine, uric acid, creatinine and hippuric acid in urine by high performance liquid chromatography, Biomed. Chromatogr. 12 (1998) 47–49. K. Kusmierek, R. Glowacki, E. Bald, Analysis of urine for cysteine, cysteinylglycine, and homocysteine by high-performance liquid chromatography, Anal. Bioanal. Chem. 385 (2006) 855–860. A. Kotain, Y. Miyaguchi, E. Tomita, K. Takamura, F. Kusu, Determination of organic acids by high-performance liquid chromatography with electrochemical detection during wine brewing, J. Agric. Food Chem. 52 (2004) 1440–1444. J.M. Zen, C.T. Hsu, A.S. Kumar, H.J. Lyuu, K.Y. Lin, Amino acid analysis using disposable copper nanoparticle plated electrodes, Analyst 129 (2004) 841–845. Y. Zhao, T. Ye, H. Liu, Y. Kou, M. Li, Y. Shao, Z. Zhu, Q. Zhuang, A novel electrochemical biosensor for the detection of uric acid and adenine, Front. Biosci. 11 (2006) 2976–2982. N. Blau, A. Matasovic, A. Lukasiewicz-Wedlechowicz, C.W. Heizmann, E. Leumann, Simultaneous determination of oxalate, glycolate, citrate, and sulfate from dried urine filter paper spots in a pediatric population, Clin. Chem. 44 (1998) 1554–1556. C. Barbas, A. García, L. Miguel, C. Simó, Evaluation of filter paper collection of urine samples for detection and measurement of organic acidurias by capillary electrophoresis, J. Chromatogr. B 780 (2002) 73–82. C.T. Hsu, H.J. Lyuu, D.C. Eric, T.H. Yang, J.M. Zen, Profiling clinically important metabolites in human urine by an electrochemical system containing disposable electrodes, Sens. Actuators B B113 (2006) 22–28. W.C. Yang, Y.Q. Dai, A.M. Yu, H.Y. Chen, Simultaneous determination of polycarboxylic acid by capillary electrophoresis with a copper electrode, J. Chromatogr. A 867 (2000) 261–269. T. Hagen, M.S. Korson, M. Sakamoto, J.E. Evans, A GC/MS/MS screening method for multiple organic acidemias from urine specimens, Clin. Chim. Acta 283 (1999) 77–88. J.M. Zen, H.H. Chung, A.S. Kumar, Selective detection of o-diphenols on copperplated screen-printed electrodes, Anal. Chem. 74 (2002) 1202–1206. Z. Dursun, G. Nisßli, Voltammetric behavior of copper(I)oxide modified carbon paste electrode in the presence of cysteine and ascorbic acid, Talanta 63 (2004) 873–878. C.B. Catherine, S.G.C. Adam, Fictionalization of magnetic nanoparticles for applications in biomedicine, J. Phys. D 13 (2003) R198–R206. C.C. Chou, S.P. Lin, K.M. Lee, C.T. Hsu, T.W. Vickroy, J.M. Zen, Fast differentiation of meats from fifteen animal species by liquid chromatography with electrochemical detection using copper nanoparticle plated electrodes, J. Chromatogr. B 846 (2007) 230–239. K. Spencer, Analytical reviews in clinical biochemistry: the estimation of creatinine, Ann. Clin. Biochem. 23 (1986) 1–25.